Supportive Studies

SUPPORTIVE STUDIES

A.C.O.G. and others have confused the public in regards to bio-identical HRT (BHRT). Understanding the conflict of interest with Big Pharma is one variable. Not being up to date on the literature is another variable.

ESTRADIOL LEVELS AND SERUM LIPIDS

Frank Z. Stanczyk et al. American Journal of OB/GYN, Volume 159, No. 6 A Randomized Comparison of Non Oral Estradiol Delivery in Post-Menopausal Women.

• Pellets produce more reproducible estrogen levels faster and remain steady longer than patches. Estradiol levels with patches were less than half of pellet patient levels and fluctuated widely.

SIGNIFICANT INCREASE IN HDL

Maturitas 1984: 5: 177-84, Linda Cardozo

• HDL increased at 12 weeks with pellets (sub-cutaneous hormone implants: bio-identical estradiol and testosterone). • 24 weeks with patches • Hot flashes were resolved in 100% of patients. • Depression resolved in 99% • Loss of libido resolved in 92%

76% REDUCTION IN HOT FLASHES

Lobo RA. Fertil Steril 2009; 92:1025 and Lindsey R. Fertil Steril 2009; 92:1045

• Using CEE and Bazedoxifene demonstrated 76% success in hot flash reduction This is but one example of the inferiority of synthetic HRT vs Bio-identical hormone pellets.

DECREASED TOTAL CHOLESTEROL, SERUM TRIGLYCERIDES, AND INCREASED HDL.

Susan Davis, et al Menopause Volume 7, No. 6, pp.395-401

• Using subcutaneous bio-identical hormone pellets This again is superior to oral HRT where triglycerides are increased and there is a null effect on HDL.

BONE DENSITY

Studd, J WW, et al (1990) Am Journal OB/GYN 163, 1474-1479

• Testosterone and estradiol pellets increased BMD 8.3% per year vs 1-2% for oral HRT.

Morris Notelovitz et al Obstetrics & Gynecology, Volume 70, No. 5, Nov 1987 Metabolic & Hormonal Effects of 17-B-Estadiol Implants

• Two year study: marked increase in bone density with no adverse effects were noted in the coagulation inhibition and fibrinolysis assays in the pellet patients • Multiple others showing similar results Testosterone that stimulates the osteoblast working with the estradiol inhibiting the osteoclast.

THE BRAIN

Intl. Journal. Alzheimer’s Disease; 2012: 1-18

• HRT and particularly ERT plays an efficacious role in preventing neurodegenerative conditions. • 17B Estradiol reduced risk for Alzheimer’s disease. • Minimizes cognitive decline in otherwise healthy women • Estradiol (E2) protects against B-amyloid induced degeneration. Progestins may actually dampen this affect. • Compared to E2 users vs Non-users E2: For avg. 15 years had increased cerebral blood flow

CEREBRAL METABOLIC ACTIVITY

Psychoneuroendocrinology 2010; 36:502-513, Silverman et al looked at • 17B Estradiol vs CEE vs CEE plus progestin on 17-B Estradiol performed the best on verbal memory (an early warning of Alzheimer’s disease) by more than 3 standard deviations.

Pike,CJ: Frontiers in Neuroendocrine 30(2009):239

• Both Estrogen and Testosterone have neuro-protective roles. Women with lower E2 levels have an even greater risk of AD. • There is overwhelming evidence that E and T help decrease apoptosis. Protective effect of both hormones decreases beta amyloid deposition.

THE HEART

Zmuda J., Amer. J. Card. 1996; 77:1244-1247, Zmuda J., Atherosclerosis. 1997; 130:199-202, and Collins P., Circulation. 1999; 100: 1690-1696,

• Multiple benefits including men given aromatize-able testosterone • Increase blood flow to the coronary arteries (even in patients with C.A.D.) • Decrease plaque in the coronary arteries • Decrease inflammation in the coronary arteries Coupled with the lipid data above is impressive and safe way to reduce the number one killer of men and women in America.

THE BREAST

Dimitrakakis and Bondy. Breast Cancer Research 2009; 11:212

• Protected by the use of estradiol and testosterone • Clinical and non-human primate studies suggest androgens inhibit mammary epithelial proliferation and breast growth. Estrogen, particularly in oral form, stimulates SHBG and reduces free testosterone. Testosterone is being used worldwide as a treatment for breast cancer.

Glaser R. Menopause: The Journal of the North American Menopause Society Vol. 21, No. 6 Rebecca Glaser M.D., a renowned breast cancer surgeon: Testosterone pellets reduce significantly tumor volume in an active breast cancer patient. Patient achieved physiologic testosterone blood levels.

DR. GARY DONOVITZ, MD

• Over the past 6 years followed nearly 10,000 patients on sub-cutaneous hormone pellet therapy • A total of 9 breast cancers - zero mortality • 30 breast cancers per year in W.H.I. in the placebo arm following nearly the same number of patients

THE CONSENSUS:

Testosterone in bio-identical form is certainly protective to the breast. There have been NO CONTRADICTORY STUDIES in the world literature.

Evaluation of bazedoxifene/conjugated estrogens for the treatment of menopausal symptoms and effects on metabolic parameters and overall safety proﬁle

Rogerio A. Lobo, M.D.,a JoAnn V. Pinkerton, M.D.,b Margery L. S. Gass, M.D.,c Maxine H. Dorin, M.D.,d Sheila Ronkin, M.D.,e James H. Pickar, M.D.,e and Ginger Constantine, M.D.e,* a Columbia University Medical Center, New York, New York; b University of Virginia Health System, Charlottesville, Virginia; c University of Cincinnati College of Medicine, Cincinnati, Ohio; d University of New Mexico Medical School, Albuquerque, New Mexico; and e Wyeth Research, Collegeville, Pennsylvania

Objective: To evaluate the effects of a tissue-selective estrogen complex (TSEC) composed of bazedoxifene/conjugated estrogens (BZA/CE) on menopausal symptoms, metabolic parameters, and overall safety. Design: Multicenter, double-blind, placebo- and active-controlled phase 3 trial (Selective estrogens, Menopause, And Response to Therapy [SMART]-1). Setting: Outpatient clinical. Patient(s): Healthy, postmenopausal women (n ¼ 3,397) age 40 to 75 with an intact uterus. Intervention(s): Single tablets of BZA (10, 20, or 40 mg), each with CE (0.625 or 0.45 mg); raloxifene 60 mg; or placebo taken daily for 2 years. Main Outcome Measure(s): Hot flushes, breast pain, vaginal atrophy, metabolic parameters, and adverse events. Result(s): BZA (20 mg)/CE (0.625 or 0.45 mg) significantly reduced the frequency and severity of hot flushes and improved measures of vaginal atrophy compared with placebo. At week 12, the daily number of hot flushes decreased by 51.7% to 85.7% with all BZA/CE doses vs. 17.1% for placebo. BZA/CE improved lipid parameters and homocysteine levels, did not significantly change carbohydrate metabolism, and had only minor effects on some coagulation parameters. The incidences of breast pain and adverse events were similar between BZA/CE and placebo. Conclusion: The TSEC composed of BZA (20 mg)/CE (0.625 or 0.45 mg) is an effective and safe treatment for menopausal symptoms. (Fertil Steril 2009;92:1025–38. 2009 by American Society for Reproductive Medicine.) Key Words: Bazedoxifene, breast pain, coagulation, conjugated estrogens, hot flushes, metabolism, vaginal atrophy

An increasing number of women in the United States are postmen- cluded that it is an appropriate option for such symptomatic women opausal. According to the U.S. Census Bureau, an estimated 60 mil- (2). Another significant concern for postmenopausal women is the lion women will be over 45 years of age by the year 2010 (1). risk of developing osteoporosis, with an estimated 8 million women Because of the decreasing levels of estrogen associated with meno- affected and another 34 million women at risk (3). pause, many women might experience bothersome symptoms, with Selective estrogen receptor modulators (SERMs) have been the hallmark feature being vasomotor symptoms or hot flushes. developed to treat postmenopausal osteoporosis. As a class, SERMs Recent analyses of the risks and benefits of hormone therapy con- generally provide estrogen agonistic effects on bone and the liver, but may be agonistic or antagonistic for the uterus and vagina, and Received October 20, 2008; revised February 11, 2009; accepted March are usually antagonistic for the breast and brain, thus potentially 17, 2009; published online July 26, 2009. inducing or worsening hot flushes in some women. An indole-based * Clinical Investigators and sites listed in appendix. Supported by Wyeth Research, Collegeville, Pennsylvania. SERM, bazedoxifene (BZA), has been shown to be effective for the R.A.L. is a clinical trial investigator for Wyeth and is a consultant for Bayer, prevention and treatment of osteoporosis (4, 5) and to provide antag- Schering, Plough-Organon, and Wyeth. J.V.P. is a consultant for onistic effects on the breast and uterus (6–9). Merck, Duramed, Novo Nordisk, and Wyeth, is a board member for In light of continued efforts to provide women with additional Boehringer Ingelheim, National Women’s Health Resource, and therapeutic options for the treatment of menopausal symptoms, con- National American Menopause Society, is a multicenter trial PI for Sol- sideration has been given to the pairing of BZAwith estrogens in the vay. M.L.S.G. received research grants from Boehringer-Ingelheim, form of a tissue-selective estrogen complex (TSEC). The partnering Proctor & Gamble, and Wyeth and is a consultant for Proctor & Gamble, Wyeth, Eli Lilly, Palatin Technologies, Roche, Sciele, and Upsher-Smith of BZA with conjugated estrogens might offer advantages over the Laboratories. M.H.D. is a speaker for Wyeth. S.R. is an employee of use of progestins in women with an intact uterus receiving hormone Wyeth Research. J.H.P. is an employee of Wyeth Research. G.C. is therapy, including an improved safety and tolerability profile. The an employee of Wyeth Research. BZA/conjugated estrogens (CEs) combination has the potential to Clinical trial registration information is available at http:// not only reduce vasomotor symptoms and prevent osteoporosis, www.ClinicalTrials.gov, NCT00675688. but also to minimize or antagonize stimulatory effects on the breast Reprint requests: Rogerio A. Lobo, M.D., Department of Obstetrics and Gynecology, Columbia University Medical Center, 622 W. 168th and uterus, which may reduce breast tenderness (10) and decrease St., PH-16, New York, NY 10032 (TEL: 646-756-8282; FAX: the occurrence of vaginal bleeding. The results of a preliminary 646-756-8280; E-mail: [email protected]). phase 2 study (11) evaluating a TSEC at varying doses of BZA

0015-0282/09/$36.00 Fertility and Sterility Vol. 92, No. 3, September 2009 1025 doi:10.1016/j.fertnstert.2009.03.113 Copyright ª2009 American Society for Reproductive Medicine, Published by Elsevier Inc. (5–20 mg) with CEs (0.3 or 0.625 mg) indicated that doses of CE glucose levels. Coagulation factors (prothrombin time, partial thromboplas- higher than 0.3 mg were necessary to consistently reduce the occur- tin time, fibrinogen, antithrombin III activity, protein C activity, protein S ac- rence of hot flushes when combined with BZA at doses that pre- tivity, plasminogen activity, plasminogen activator inhibitor-1 [PAI-1] vented estrogen-induced endometrial proliferation. activity, PAI-1 antigen, and D–dimer) and lipid parameters (total cholesterol, This article reports findings from the Selective estrogens, Meno- high-density lipoprotein [HDL] cholesterol, low–density lipoprotein [LDL] cholesterol, very-low-density lipoprotein [VLDL] cholesterol, triglycerides, pause, And Response to Therapy-1 (SMART-1) trial, a 2-year, ran- VLDL triglycerides, HDL2 cholesterol, HDL3 cholesterol, apoliprotein A1, domized, double-blind, placebo- and active-controlled clinical trial apolipoprotein B, and lipoprotein [a]) were assessed in the metabolic sub- evaluating the efficacy and safety of varying doses of BZA/CE in study. Safety and metabolic data for month 24 (study end) are presented here. postmenopausal women. This report describes the effects of BZA/ CE on menopausal symptoms (hot flushes, vaginal atrophy), meta- Statistical Analyses bolic parameters, as well as its overall safety profile. The effects of BZA/CE on endometrium, bone mineral density (BMD), and Analyses of any data of interest included only those subjects who received at uterine bleeding are reported separately in this journal issue (12–14). least one dose of the study drug, in addition to criteria specific for the parameter of interest. Hot flush data were analyzed for the efficacy evaluable 1 population or those subjects who reported at least seven moderate or severe hot MATERIALS AND METHODS flushes per day or 50 per week during screening (n ¼ 216). Complete data Study Design were summarized for the baseline week, for each week during weeks 1–12 Healthy women (40–75 years of age) who were postmenopausal for at least 1 (using a last observation carried forward approach), and for every four- year were eligible for participation in this 2-year, double-blind, randomized, week period thereafter. The mean daily number of hot flushes was calculated multicenter, placebo- and active–controlled phase 3 trial conducted at 94 using only moderate and severe hot flushes, whereas the mean daily severity study sites worldwide. All subjects were required to have an intact uterus of hot flushes was calculated using all three intensities (1 ¼ mild; 2 ¼ moder- and acceptable endometrial biopsy results at screening. ate; and 3 ¼ severe). Pairwise comparisons vs. placebo were made using an A subset of women was enrolled in the Osteoporosis Prevention I Substudy analysis of covariance (ANCOVA). or the Osteoporosis Prevention II and Metabolic Substudy. The two osteopo- Vaginal epithelial maturation data were analyzed for the vulvar/vaginal at- rosis substudies were designed to assess BMD changes in later and earlier rophy population, or for those subjects with no more than 5% superficial cells postmenopausal women. Subjects in the Osteoporosis Prevention I Substudy at screening and those who had a baseline and at least one on-therapy assess- were >5 years postmenopause, with a BMD T-score between 1 and 2.5 at ment (n ¼ 1,867). Vaginal atrophy was measured by the change from baseline the lumbar spine or total hip and at least one additional risk factor for oste- in the percentage of superficial, intermediate, and parabasal cells at each time oporosis (family history of osteoporosis, early menopause, current history of point using an ANCOVA, with treatment and center as factors and baseline smoking, past history of excessive alcohol use, diet low in calcium, inactive value as covariate. The proportion of each cell type was analyzed using a non- lifestyle, thin and/or small frame, Caucasian or Asian). Subjects in the Oste- parametric one-way Kruskal-Wallis test for between-group comparisons (vs. oporosis Prevention II and Metabolic Substudy were R1 year but %5 years placebo) and a signed-rank test for within-group comparisons. postmenopause with at least 1 risk factor for osteoporosis. The incidence of sexual activity and dyspareunia was summarized for seven consecutive baseline days before treatment initiation and for sequential 28-day periods during the study. Breast pain data were analyzed for subjects with diary Treatments data for at least 5 of the 7 days during screening and at least 20 days for the Subjects were randomly assigned through a computerized randomization/en- relevant 4-week interval. Among-group differences in the incidence of breast rollment interactive voice recognition system to one of eight treatment regi- pain during each time period (weeks 1–4, 5–8, and 9–12) were evaluated using mens, including six BZA/CE doses (BZA [10, 20, or 40 mg] each with CE Fisher’s exact test. The mean change from baseline in the percentage of days [0.45 or 0.625 mg]), raloxifene 60 mg, or placebo. Subjects were required with breast pain in a given 4-week interval was evaluated using ANCOVA, to take one capsule orally at approximately the same time each day and main- and pairwise comparisons versus placebo were made using the t test. tain a consistent daily intake of dietary and supplemental calcium and vita- Data from the metabolic substudy were analyzed in those subjects who had min D (total daily calcium intake, 1,000–1,600 mg; vitamin D, 200–400 IU). baseline and on–therapy values for the parameter of interest. For lipid param- Use of the following concomitant medications was permitted: acetamino- eters, differences in the mean percent change from baseline at each time point phen; inhaled steroids (maximum daily intake, 1,000 mg); dermal steroids; between each BZA/CE treatment group and placebo were evaluated using an intra-articular injections (maximum of three injections during the treatment analysis of variance (ANOVA) model. For all other parameters (e.g., coagu- period); oral corticoids at standard therapeutic doses for periods of up to 10 lation, carbohydrate), the mean absolute change from baseline at each time days; up to two antihypertensive medications; and vitamin/mineral supple- point was assessed; within- and between-group comparisons (vs. placebo) ments if they had been taken continuously for R12 weeks before the study. were made using the ANOVA model. Prohibited therapy included estrogen-, progestin-, androgen-, or SERM-con- Among-group differences in the incidence of AEs were evaluated using c2 taining medications other than the study drug. Also prohibited was the con- analysis (overall P value), and Fisher’s exact test was used to compare the tinued use of medications that could affect bone metabolism (osteoporosis incidence of AEs between each BZA/CE group and the placebo group. For substudies) or prescription lipid–lowering agents and anticoagulants other each laboratory parameter, the adjusted mean change from baseline as well than aspirin (metabolic substudy). as the number and percentage of subjects with potentially clinically important (PCI) values were summarized. Within- and among-group differences Assessments in the mean change from baseline for all laboratory tests were evaluated using ANCOVA. Pairwise comparisons for the incidence of PCI values were made Subjects were instructed to record in daily diaries information on hot flushes, using Fisher’s exact test. sexual activity/dyspareunia, and breast pain. Vaginal atrophy was measured by vaginal smears at months 6, 12, 18, and 24, which quantified the degree of maturation of the vaginal epithelium by the proportion of parabasal, interme- RESULTS diate, and superficial cells obtained in the sample. Subjects Safety assessments included adverse event (AE) reporting and clinical laboratory evaluations (e.g., hematology, blood chemistry), which were per- A total of 3,397 subjects were randomly assigned to a treatment formed at the screening visit and at months 3, 6, 12, 18, and 24. Reports of group and received at least 1 dose of the study drug (Fig. 1). Subject AEs were summarized using terms from the Medical Dictionary for Regula- demographics and baseline characteristics were similar across all tory Activities. In the metabolic substudy, fasting serum samples were col- treatment groups (Table 1). The rates of study discontinuation lected at randomization and at months 6, 12, and 18 to evaluate insulin and were not significantly different among treatment groups (range,

1026 Lobo et al. Effects of BZA/CE on menopausal symptoms Vol. 92, No. 3, September 2009 29.8–35.7%; Fig.1). The most frequent reason for discontinuation than placebo in increasing the mean proportion of superficial cells was AEs, followed by subject request unrelated to the study. A sig- from baseline to most time points (P < 0.001; Fig. 3A). Further- nificantly higher percentage of subjects in the placebo group with- more, all four BZA/CE doses containing BZA (10 or 20 mg)/CE drew from the study because of unsatisfactory response compared (0.625 or 0.45 mg) were significantly more effective than placebo with those in any other treatment group (P ¼ 0.002). in increasing the mean proportion of intermediate cells and decreasing the proportion of parabasal cells from baseline to all time points Hot Flushes (P < 0.001; Fig. 3B, C). There was a dose–related attenuation of the Examination of the mean daily number of moderate and severe hot beneficial estrogenic effect on vaginal atrophy with increasing doses flushes demonstrated that all doses of BZA/CE provided signifi- of BZA, which was most noted with BZA (40 mg)/CE (0.625 or cantly better relief of hot flushes than placebo at most time points 0.45 mg). However, with BZA doses of 10 and 20 mg, the effects (P < 0.01; Fig. 2A). At week 12, the adjusted mean change from on vaginal endpoints were substantially improved compared with baseline in the average daily number of hot flushes for the BZA/ raloxifene or placebo. CE treatment groups ranged from 5.53 to 8.98 (51.7% to 85.7%) compared with 2.45 (17.1%) and 5.29 (44.1%) Sexual Activity, Dyspareunia, and Breast Pain for the placebo and raloxifene treatment groups, respectively. Treat- At baseline, sexual activity was reported by 34–43% of the partici- ment with BZA (20 mg)/CE (0.625 or 0.45 mg) was significantly pants and dyspareunia was reported by 16–26%. There were no sig- more effective than placebo at every weekly time point from weeks nificant among-group differences in the incidence of sexual activity 6 to 12. Improvements in the frequency and severity of hot flushes throughout the study. Compared with subjects who received placebo observed with BZA (10 or 20 mg)/CE (0.625 or 0.45 mg) were or raloxifene, subjects treated with BZA (10 mg)/CE (0.625 mg) had sustained through the second year of therapy (data not shown). a lower incidence of dyspareunia at weeks 5–8 (P < 0.05). With Although the daily number of hot flushes reported with BZA BZA (10 mg)/CE (0.625 mg) and BZA (20 mg)/CE (0.625 or (40 mg)/CE (0.625 mg) or BZA (40 mg)/CE (0.45 mg) was also sig- 0.45 mg) there was a significantly lower incidence of dyspareunia nificantly reduced compared with placebo at week 12, this decrease during weeks 9–12 (P < 0.05). was not as great as that noted with BZA (10 or 20 mg)/CE (0.625 or For most of the weeks analyzed, BZA (10 and 20 mg)/CE (0.625 0.45 mg) at most time points (Fig. 2). Bazedoxifene/CE groups dem- or 0.45 mg) had significantly less dyspareunia than with raloxifene onstrated significant decreases in hot flush number and severity P < 0.05. Breast pain occurred with similar frequency for subjects in compared with raloxifene; significant differences in number and/ the BZA/CE, raloxifene, and placebo groups, and there were no sig- or severity were seen as early as week 2 for BZA (10 mg)/CE nificant differences in the incidence of breast pain among the groups (0.625 or 0.45 mg) and at week 6 for BZA (20 mg)/CE (0.625 or for any 28-day interval. 0.45 mg), and continued through week 12. Metabolic Parameters Vaginal Atrophy The adjusted mean percent changes from baseline in LDL and HDL Treatment with BZA (10 mg)/CE (0.625 mg or 0.45 mg) and BZA cholesterol are presented in Figure 4. Reductions in LDL cholesterol (20 mg)/CE (0.625 or 0.45 mg) was significantly more effective for all BZA/CE doses (range,5.7% to 10.9%) were significantly

FIGURE 1

Disposition of study subjects. Following screening, a total of 3,544 subjects were randomly assigned to 1 of 8 treatment groups, and 3,397 subjects took at least one dose of the study drug. The diagram outlines reasons for study discontinuation for each treatment group. Note: BZA ¼ bazedoxifene; CE ¼ conjugated estrogens. aOverall P < 0.01; c2 analysis.

Lobo. Effects of BZA/CE on menopausal symptoms. Fertil Steril 2009.

Fertility and Sterility 1027 TABLE 1 Subject demographic and baseline characteristics.

CE (0.625 mg) CE (0.45 mg)

Characteristic BZA (10 mg) BZA (20 mg) BZA (40 mg) BZA (10 mg) BZA (20 mg) BZA (40 mg) Raloxifene (60 mg) Placebo

Age, y N 430 414 417 430 433 423 423 427 Mean (SD) 56.35 (5.60) 56.29 (5.98) 56.68 (5.81) 56.84 (5.73) 56.22 (5.80) 56.31 (5.76) 56.54 (5.72) 56.48 (6.04) Ethnic origin White 354 (82.33) 343 (82.85) 341 (81.77) 346 (80.47) 351 (81.06) 327 (77.30) 341 (80.61) 340 (79.63) Black 58 (13.49) 53 (12.80) 55 (13.19) 57 (13.26) 54 (12.47) 66 (15.60) 57 (13.48) 66 (15.46) Hispanic 13 (3.02) 13 (3.14) 16 (3.84) 15 (3.49) 20 (4.62) 18 (4.26) 14 (3.31) 15 (3.51) Other 5 (1.16) 5 (1.20) 5 (1.20) 12 (2.79) 8 (1.84) 12 (2.83) 11 (2.59) 6 (1.40) BMI, kg/m2 N 430 412 417 430 433 423 423 426 Mean (SD) 25.74 (3.44) 25.87 (3.55) 25.66 (3.18) 25.83 (3.36) 25.97 (3.45) 25.57 (3.46) 25.92 (3.28) 25.94 (3.54) Years since last menstrual period N 430 414 417 429 433 423 423 427 Mean (SD) 7.80 (5.65) 8.10 (5.71) 8.29 (5.71) 7.94 (5.60) 8.11 (5.70) 7.9 (5.80) 8.33 (5.83) 8.36 (5.78)

Note: BMI¼ body mass index.

Lobo. Effects of BZA/CE on menopausal symptoms. Fertil Steril 2009.

greater compared with placebo (range, 0.1 to 2.2%) at all time all time points (P < 0.05), with the exception of BZA (10 mg)/CE points (P < 0.01). Increases in HDL cholesterol for all BZA/CE (0.625 mg) at months 18 and 24. Decreases in plasma homocysteine doses (range, 7.0–13.5%) were significantly greater compared with BZA (40 mg)/CE (0.625 mg) and BZA (10 mg)/CE (0.45 mg) with placebo (range, 1.3% to 5.4%) at all time points (P < 0.05), were significantly greater than that observed with placebo at all time and significantly greater compared with raloxifene (range, 3.1– points (P < 0.05). 6.6%) at most time points (P < 0.05). There was no apparent None of the BZA/CE doses had any effect on partial thromboplas- dose-related attenuation of HDL cholesterol levels with BZA, and tin time, prothrombin time, or serum concentrations of D-dimer. the observed increases were sustained throughout 2 years of therapy. Effects of BZA/CE on fibrinolysis (assessed by plasminogen activ- Changes in other lipid parameters are provided in Table 2. Total ity, PAI-1 activity, and serum levels of PAI-1 antigen) were similar to cholesterol decreased from baseline for all BZA/CE treatment those with known beneficial effects of estrogen. All six doses of groups at all time points (range, 0.8 to 3.7%). At month 24, BZA/CE were associated with small decreases from baseline in the increase from baseline in triglycerides was higher for all BZA/ mean PAI-1 activity (range, 1.3 to 3.1 IU/mL at month 24; P CE doses (range, 12.0–25.1%) compared with placebo (6.1%) or < 0.05 for BZA [10 and 20 mg]/CE [0.625 mg] and BZA raloxifene (6.9%). [40 mg]/CE [0.45 mg]) and PAI-1 antigen levels (range, 0.92 to Increases from baseline in HDL2 cholesterol noted for all BZA/ 6.6 mg/day at month 24; P < 0.05 for BZA [20 mg]/CE [0.625 CE treatment groups (range, 17.9–33.4% at month 24) were signif- mg]), and small increases from baseline in mean plasminogen activ- icantly greater compared with placebo at all time points and ralox- ity (range, 0.07–0.11 mg/day at month 24; P < 0.001 for all BZA/CE ifene at some time points (P < 0.05). There was a minor attenuation doses). There were small decreases from baseline in mean levels of the increase seen in HDL2 cholesterol levels as the dose of BZA of the procoagulation factor fibrinogen for all BZA/CE treatment within the TSEC increased. groups (range, 0.3 to 0.5 mg/day), which were significantly Significant increases from baseline in apolipoprotein A1 reported different from those observed in the placebo group (P < 0.001). with BZA/CE (range, 9.0–11.2% at month 24) were not attenuated There was no significant change with raloxifene. (Note: Overall by increasing the BZA dose and were greater than that observed for these coagulation parameters, changes in the raloxifene with raloxifene or placebo (Table 2). Decreases from baseline in group were similar to those seen with BZA/CE with no statistically apolipoprotein B levels were observed with BZA (20 or 40 mg)/ significant difference between raloxifene 60 mg and either CE (0.625 or 0.45 mg) at all time points (range, 0.7 to 2.0% at BZA/CE dose). month 24), whereas increases were noted with placebo. Reductions There was no significant change from baseline in protein C activ- in lipoprotein (a) levels noted for all BZA/CE treatment groups were ity for any BZA/CE doses relative to placebo. However, some significantly greater compared with the placebo group at months 12 increases in protein C activity observed with BZA/CE were and 24 (P < 0.05), and significantly greater compared with the significantly different from the decreases observed with raloxifene raloxifene group at month 12 (P % 0.05). at some time points (P < 0.05). With the exception of BZA (20 There were no significant changes in fasting glucose, fasting mg)/CE (0.45 mg) at month 6, all doses of BZA/CE were associated insulin, or C-reactive protein levels relative to baseline or placebo with small but significantly greater changes in protein S activity at any time point with BZA/CE. Decreases from baseline in plasma (range, 0.1 to 0.1 mg/day) compared with placebo at all time homocysteine were significant for the BZA/CE treatment groups at points (P < 0.05). At month 24, minor increases in protein S activity

1028 Lobo et al. Effects of BZA/CE on menopausal symptoms Vol. 92, No. 3, September 2009 FIGURE 2

Mean changes from baseline in (A) the daily number and (B) the severity of hot flushes in each treatment group. (A) Mean change in the daily number of moderate or severe hot flushes for weeks 1–12. P < 0.05 for all BZA/CE doses compared with placebo for weeks 5–12. P < 0.05 for BZA (10 mg)/CE (0.625 mg) at all time points and for BZA (10 mg)/CE (0.45 mg) at all time points except Week 1. (B) Adjusted mean change in total hot flush severity (mild, moderate, and severe) from baseline at weeks 4 and 12. Note: BZA ¼ bazedoxifene; CE ¼ conjugated estrogens. aP < 0.001 vs. placebo. A Daily Number of Hot Flushes

BZA 10 mg/CE 0.625 mg BZA 10 mg/CE 0.45 mg Raloxifene 60 mg

BZA 20 mg/CE 0.625 mg BZA 20 mg/CE 0.45 mg Placebo 2 BZA 40 mg/CE 0.625 mg BZA 40 mg/CE 0.45 mg

-2

-4

-6 Mean change from baseline -8

-10 12345678 9 10 11 12 Weeks

Daily Severity of Hot Fluhes B 0.1

-0.4

-0.9 a Week 4 a Week 12 a a a -1.4 Mean change from baseline

a -1.9 BZA BZA BZA BZA BZA BZA Raloxifene Placebo 10 mg 20 mg 40 mg 10 mg 20 mg 40 mg 60 mg

CE 0.625 mg CE 0.45 mg

Lobo. Effects of BZA/CE on menopausal symptoms. Fertil Steril 2009.

Fertility and Sterility 1029 FIGURE 3

Adjusted mean percent changes from baseline in the proportion of superﬁcial, intermediate, and parabasal cells at month 24. Vaginal atrophy was assessed by vaginal smears, which were obtained from those who took at least one dose of the study drug, had a baseline and at least one on-therapy value, and had no more than 5% superﬁcial cells at screening. Note: BZA ¼ bazedoxifene; CE ¼ conjugated estrogens. aP < 0.001 vs. placebo. bP < 0.01 vs. placebo.

A) Superficial B) Intermediate

10 50

40 8 a a a a 30 6 b 20 a 4 10 a 2 0 0 -10

-2 -20 Mean percent change from baseline -4 Mean percent change from baseline -30 BZA BZA BZA BZA BZA BZA Raloxifene Placebo BZA BZA BZA BZA BZA BZA Raloxifene Placebo 10 mg 20 mg 40 mg 10 mg 20 mg 40 mg 60 mg 10 mg 20 mg 40 mg 10 mg 20 mg 40 mg 60 mg CE 0.625 mg CE 0.45 mg CE 0.625 mg CE 0.45 mg

C) Parabasal 20

-10 a -20 a -30

-40 a a -50

Mean percent change from baseline -60 BZA BZA BZA BZA BZA BZA Raloxifene Placebo 10 mg 20 mg 40 mg 10 mg 20 mg 40 mg 60 mg CE 0.625 mg CE 0.45 mg

Lobo. Effects of BZA/CE on menopausal symptoms. Fertil Steril 2009. were noted in all BZA/CE treatment groups, whereas decreases were the study, which were not thought to be study related. These deaths observed at earlier time points. Changes in antithrombin III activity were caused by bronchoaspiration, intracerebral hemorrhage (range, 0.1 to 0.3 mg/day) were also small but significantly secondary to metastatic lung cancer, chronic obstructive airway greater compared with placebo for all BZA/CE doses (P < 0.05), disease, unknown causes, and accidental injury (two subjects). with the exception of BZA (20 mg)/CE (0.625 mg) at month 6. Changes from baseline in antithrombin III activity were similar for subjects who received raloxifene or BZA/CE. There were no appre- Venous Thromboembolic Events ciable dose-related effects of BZA/CE on anticoagulation factors. Overall, the incidence of venous thromboembolic events (VTEs) was similar for subjects treated with BZA/CE or placebo (0.76 vs. 1.56 per 1,000 women-years, respectively; relative risk, 0.48; 95% Adverse Events confidence interval [CI], 0.05–4.66). Two subjects in the BZA/CE Overall, the incidence of AEs and serious AEs was similar among treatment groups and one subject in the placebo group reported treatment groups (Table 3). There were no significant differences deep vein thrombosis. Pulmonary embolism was reported in one in the incidence of treatment-emergent AEs among groups (range, subject who received BZA (40 mg)/CE (0.625 mg). There were no 90–94%; Table 3). The majority of treatment-emergent AEs, which reports of retinal vein thrombosis. Similarly, the incidence of super- were generally mild or moderate in severity, were not considered ficial thromboses or phlebitis was low across all treatment groups related to the study drug. There were no significant among-group (<1%), with no statistically significant among-group differences; differences in the incidence of these AEs. There were 6 deaths in importantly, none of these cases were classified as VTEs.

1030 Lobo et al. Effects of BZA/CE on menopausal symptoms Vol. 92, No. 3, September 2009 FIGURE 4 of PCI increases in cholesterol levels was 8.1% in the placebo group and 5.7% in the raloxifene group, and it did not exceed 6.5% in any Adjusted mean percent changes from baseline in (A) LDL BZA treatment group. However, significantly higher percentages of cholesterol and (B) HDL cholesterol levels in each treatment group subjects in the BZA/CE treatment groups experienced PCI increases (metabolic substudy) at month 24. The adjusted mean percent in triglyceride levels compared with the placebo group (P < 0.05). changes from baseline in levels of LDL cholesterol and HDL A total of five subjects treated with BZA/CE had triglyceride cholesterol were quantified for each treatment group. Note: BZA ¼ values that were considered clinically important. Analysis of other bazedoxifene; CE ¼ conjugated estrogens; HDL ¼ high-density laboratory safety data for blood chemistry and hematology a lipoprotein; LDL ¼ low-density lipoprotein. P < 0.001 vs. placebo. parameters demonstrated no trends of concern. Results of liver func- b c P < 0.01 vs. placebo. P < 0.05 vs. raloxifene. tion tests also indicated no values of clinical importance. A) LDL Cholesterol 3 Discussion 1 The TSEC was designed to provide tissue-selective activities of -1 a SERM with the proven benefits of estrogen therapy (ET). The -3 SMART-1 trial evaluated the efficacy and safety of the first TSEC change -5 composed of BZA/CE in postmenopausal women over a 2-year period. Findings from this study demonstrated favorable effects of -7 a a BZA/CE on the relief of menopausal symptoms. Specifically, -9 a BZA (20 mg)/CE (0.625 or 0.45 mg) were significantly and clini- b b -11 b cally more effective than placebo in reducing the incidence of hot Adjusted mean % Adjusted -13 flushes, and BZA (20 mg)/CE (0.625 or 0.45 mg) also significantly reduced the severity of hot flushes compared with placebo. The -15 BZA BZA BZA BZA BZA BZA Raloxifene Placebo BZA/CE groups decreased the daily number of hot flushes by 10 mg 20 mg 40 mg 10 mg 20 mg 40 mg 60 mg 51.7–85.7% compared with only 17.1% with placebo. Whereas CE 0.625 mg CE 0.45 mg raloxifene has been shown to increase hot flushes in this population, in our study, raloxifene was associated with a numeric reduction in B) HDL Cholesterol flush frequency, but not severity; however, BZA (10 and 20 mg)/CE 13 groups BZA/CE groups reduced both hot flush frequency and sever- b, c b, c b, c 11 b, c b, c ity significantly better than raloxifene. Furthermore, BZA (20 mg)/ b CE (0.625 or 0.45 mg) was significantly more effective than placebo 9 in improving vaginal atrophy with significant increases in superficial change 7 and intermediate cells and reductions in parabasal cells. Accord- 5 ingly, BZA (20 mg)/CE (0.625 or 0.45 mg) significantly reduced the incidence of dyspareunia relative to placebo. Breast pain is 3 known to be increased in women taking estrogen/progestin therapy 1 (EPT) (10). In this study, none of the BZA/CE doses increased the

Adjusted mean % Adjusted incidence of breast pain compared with placebo or raloxifene. -1 The menopausal transition is known to confer unfavorable -3 changes in lipid and carbohydrate metabolism (15, 16), which BZA BZA BZA BZA BZA BZA Raloxifene Placebo 10 mg 20 mg 40 mg 10 mg 20 mg 40 mg 60 mg may be associated with an increased risk of cardiovascular disease CE 0.625 mg CE 0.45 mg in women (17, 18). In this study, treatment with all BZA/CE regimens was associated with decreases in total cholesterol. Impor- Lobo. Effects of BZA/CE on menopausal symptoms. Fertil Steril 2009. tantly, increases in LDL and decreases in HDL cholesterol are known risk factors for cardiovascular disease in women (19, 20). All BZA/CE doses were associated with marked decreases in Cardiovascular AEs LDL and increases in HDL cholesterol throughout the 2-year study period. Greater improvements in HDL cholesterol, HDL choles- The cardiovascular AEs of interest included myocardial infarction, 2 terol, and apolipoprotein A1 were observed with administration of coronary artery disease, and coronary artery insufficiency. The BZA/CE compared with raloxifene. Such favorable effects on lipid incidence of cardiovascular AEs was low (<1%) across all treatment parameters have previously been observed in studies that random- groups, with no significant differences among groups. Compared ized women to receive ET or EPT for up to 3 years (21–23). with subjects who received placebo, the relative risk of experiencing A minor attenuating effect on HDL cholesterol levels was a myocardial infarction with BZA/CE was 0.48 (95% CI, 0.05– noted with increasing dose of BZA. However, the reductions in 4.66), or an incidence of 0.76 vs. 1.56 per 1,000 woman-years. LDL cholesterol and apolipoprotein B appeared to be even For coronary artery disease and coronary artery insufficiency, greater with increasing dose of BZA, particularly when paired the relative risk with BZA/CE vs. placebo was 1.29 (95% CI, with CE (0.45 mg). Decreases in PAI-1 activity and PAI-1 anti- 0.16–10.34), or an incidence of 2.02 vs. 1.56 per 1,000 women–years. gen levels were also observed with BZA/CE treatment. Improve- ments in lipoprotein (a) observed with BZA/CE treatment in this Clinical Laboratory Evaluations study are consistent with those observed with EPT in the The percentage of subjects with PCI increases in cholesterol levels at Women’s Health, Osteoporosis, Progestin, Estrogen (HOPE) trial any time point was similar across all treatment groups. The incidence (23) and the Heart and Estrogen/progestin Replacement Study

Fertility and Sterility 1031 1032 ooe al. et Lobo

fet fBAC nmnpua symptoms menopausal on BZA/CE of Effects TABLE 2 Adjusted mean changes from baseline in selected lipid and coagulation parameters at month 24 (metabolic substudy).

CE (0.625 mg) CE (0.45 mg)

Parameters BZA (10 mg) BZA (20 mg) BZA (40 mg) BZA (10 mg) BZA (20 mg) BZA (40 mg) Raloxifene (60 mg) Placebo

Mean (SE) percent change Lipid parameters Total cholesterol 2.3 (1.3) 2.6 (1.4) 2.3 (1.4) 2.3 (1.4) 3.7 (1.3)a 3.4 (1.3) 2.6 (1.4) 0.2 (1.4) Triglycerides 18.8 (4.9) 25.1 (5.1)b,c 23.6 (5.0)a,d 13.3 (5.3) 23.1 (4.7)b,d 12.0 (4.9) 6.9 (5.2) 6.1 (5.0) c,e d,e b c,e d,e e HDL2-C 33.4 (4.9) 28.4 (5.1) 17.9 (5.0) 30.1 (5.3) 28.2 (4.7) 21.0 (4.9) 10.7 (5.2) 4.3 (5.0) Apo A1 11.2 (1.3)c,e 11.1 (1.4)c,e 10.9 (1.4)c,e 9.4 (1.4)e 11.1 (1.3)c,e 9.0 (1.3)e 6.0 (1.4) 1.2 (1.4) Apo B 1.4 (1.8) 2.0 (1.9)a 0.7 (1.9) 0.7 (2.0) 1.9 (1.7)a 1.9 (1.8)a 0.2 (1.9) 4.2 (1.9) Lp(a) 21.4 (3.0)e 21.2 (3.1)b 19.4 (3.1)b 18.7 (3.2)b 19.0 (2.8)b 16.9 (3.0)a 14.9 (3.2) 7.7 (3.1) Mean (SE) change, mg/day Coagulation factors Fibrinogen 0.38 (0.07)e 0.50 (0.07)e 0.40 (0.07)e 0.39 (0.07)e 0.44 (0.06)e 0.48 (0.07)e 0.36 (0.07) 0.01 (0.07) Protein C activity 0.06 (0.02)d 0.04 (0.02)d 0.02 (0.02) 0.04 (0.02)d 0.03 (0.01) 0.01 (0.02) 0.01 (0.02) 0.03 (0.02) Protein S activity 0.02 (0.02)e,c 0.07 (0.02)b 0.07 (0.02)b 0.07 (0.02)b 0.09 (0.02)b 0.08 (0.02)b 0.11 (0.02) 0.16 (0.02) Antithrombin III activity 0.29 (0.02)b 0.28 (0.02)b 0.28 (0.02)b 0.27 (0.02)a 0.27 (0.01)a 0.28 (0.02)b 0.26 (0.02) 0.22 (0.02)

Note: Apo ¼ apolipoprotein; HDL2-C ¼ high-density lipoprotein 2 cholesterol; Lp(a) ¼ lipoprotein (a). a P < 0.05 vs. placebo. b P < 0.01 vs. placebo. c P < 0.01 vs. raloxifene. d P < 0.05 vs. raloxifene. e P < 0.001 vs. placebo.

Lobo. Effects of BZA/CE on menopausal symptoms. Fertil Steril 2009. o.9,N.3 etme 2009 September 3, No. 92, Vol. TABLE 3 Summary of safety proﬁle and incidence (R10% in any treatment group) of treatment-emergent AEs.

CE (0.625 mg) CE (0.45 mg)

BZA BZA BZA BZA BZA BZA Raloxifene (10 mg) (20 mg) (40 mg) (10 mg) (20 mg) (40 mg) (60 mg) Placebo n [ 430 n [ 414 n [ 417 n [ 430 n [ 433 n [ 423 n [ 423 n [ 427

Any AE 403 (93.7) 382 (92.3) 384 (92.1) 400 (93.0) 401 (92.6) 381 (90.1) 391 (92.4) 392 (91.8) Any serious AE 32 (7.24) 23 (5.6) 24 (5.8) 35 (8.1) 26 (6.0) 26 (6.1) 32 (7.6) 34 (8.0) Any TEAE 403 (93.7) 382 (92.3) 384 (92.1) 400 (93.0) 401 (92.6) 381 (90.1) 391 (92.4) 392 (91.8) Body as a whole Abdominal pain 56 (13.0) 39 (9.4) 38 (9.1) 41 (9.5) 54 (12.5) 38 (90.1) 36 (8.5) 32 (7.5) Inﬂuenza 74 (17.2) 78 (18.8) 72 (17.3) 80 (18.6) 97 (22.4) 89 (21.0) 91 (21.5) 90 (21.1) Headache 141 (32.8) 129 (31.2) 112 (26.9) 141 (32.8) 135 (31.2) 119 (28.1) 126 (29.8) 117 (27.4) Infections and 278 (64.7) 252 (60.9) 243 (58.3) 261 (60.7) 276 (63.7) 238 (56.3) 245 (57.9) 254 (59.5) infestations Digestive system Diarrhea 28 (6.5) 28 (6.8) 30 (7.2) 35 (8.1) 44 (10.2) 30 (7.1) 36 (8.5) 26 (6.1) Abdominal pain upper 43 (10.0) 50 (12.1) 51 (12.2) 54 (12.6) 51 (11.8) 47 (11.1) 42 (9.9) 31 (7.3) Nausea 30 (7.0) 30 (7.2) 31 (7.4) 32 (7.4) 46 (10.6) 29 (6.9) 27 (6.4) 23 (5.4) Musculoskeletal system Arthralgia 98 (22.8) 110 (26.6) 97 (23.3) 104 (24.2) 101 (23.3) 119 (28.1) 122 (28.8) 112 (26.2) Back pain 109 (25.3) 107 (25.8) 87 (20.9) 110 (25.6) 106 (24.5) 82 (19.4) 89 (21.0) 86 (20.1) Muscle spasmsa 44 (10.2) 30 (7.2) 45 (10.8) 40 (9.3) 47 (10.9) 41 (9.7) 35 (8.3) 22 (5.2) Myalgia 58 (13.5) 61 (14.7) 55 (13.2) 56 (13.0) 53 (12.2) 51 (12.1) 60 (14.2) 58 (13.6) Pain in extremity 53 (12.3) 60 (14.5) 61 (14.6) 73 (17.0) 70 (16.2) 58 (13.7) 59 (13.9) 63 (14.8) Nervous system Insomnia 35 (8.1) 23 (5.6) 33 (7.9) 33 (7.7) 38 (8.8) 38 (9.0) 46 (10.9) 48 (11.2) Respiratory system Nasopharyngitis 69 (16.0) 77 (18.6) 65 (15.6) 68 (15.8) 79 (18.2) 56 (13.2) 60 (14.2) 66 (15.5) Sinusitis 37 (8.6) 43 (10.4) 36 (8.6) 38 (8.8) 27 (6.2) 21 (5.0) 36 (8.5) 41 (9.6) Upper respiratory 54 (12.6) 42 (10.1) 36 (8.6) 47 (10.9) 52 (12.0) 47 (11.1) 55 (13.0) 47 (11.0) infection Pharyngolaryngeal pain 43 (10.0) 40 (9.7) 37 (8.9) 33 (7.7) 48 (11.1) 36 (8.5) 46 (10.9) 37 (8.7) Urogenital system Urinary tract infection 36 (8.4) 42 (10.1) 43 (10.3) 40 (9.3) 40 (9.2) 32 (7.6) 35 (8.3) 35 (8.2)

Note: TEAE ¼ treatment-emergent adverse event. a P < 0.05 overall.

Lobo. Effects of BZA/CE on menopausal symptoms. Fertil Steril 2009.

(HERS) (24). This beneficial effect on lipoprotein (a) observed observation in a larger population of subjects will be able to provide with BZA/CE in the present study was greater than the response definitive risk information regarding possible adverse effects of with placebo or raloxifene at month 12 and was further enhanced therapy, as this study was not powered to detect small differences at month 24. in these cardiovascular safety endpoints. A favorable decrease in fibrinogen activity was noted with all In this study, analysis of most clinical laboratory determinations BZA/CE doses, a finding supported by that reported with ET and (e.g., hematology, blood chemistry, liver function) revealed no clin- EPT in the HOPE trial (23). Also consistent with findings of the ically important differences among treatment groups and no trends HOPE trial were the changes in protein S activity and antithrombin of concern. In that the pairing of BZA and CE alleviates the need III activity noted with BZA/CE vs. placebo. No significant changes for a progestin, it is of interest to determine whether BZA attenuates in carbohydrate metabolism or serum concentrations of D-dimer any of the beneficial estrogenic effects of CE on metabolism. were observed with any BZA/CE regimens throughout the study pe- Among the parameters assessed in this study, the higher BZA riod. Because CE alone is known to reduce levels of fasting insulin, dose (40 mg) was found to decrease, somewhat, the beneficial effect future studies will help determine whether BZA might affect this of CE on hot flushes and vaginal atrophy. Apart from a relatively beneficial estrogen response. It is also important to note that any minor attenuation noted for HDL2 cholesterol, no other attenuating beneficial effects of BZA/CE on surrogate markers might not predict effects of BZA on clinical laboratory determinations were observed. clinical events. For instance, triglycerides were unaffected by BZA dose, and some Overall, BZA/CE was generally well tolerated and demonstrated estrogenic effects were enhanced with increasing BZA dose, includ- a safety profile similar to that of placebo. The incidence of AEs, ing decreases in LDL cholesterol. serious AEs, and study discontinuations owing to AEs was similar One possible limitation of this study in evaluating the relief of across all treatment groups. Treatment with BZA/CE doses was vasomotor symptoms and vaginal atrophy is the wide range in not associated with an increased risk of VTEs or cardiovascular ages of the subject population (45–70 years of age), because the AEs; however, it is important to note that a longer period of occurrence of menopausal symptoms is typically highest in the early

Fertility and Sterility 1033 years of menopause. However, an important objective of the and vaginal atrophy, without increasing the incidence of breast pain SMART-1 trial was to assess the efficacy of BZA/CE for the preven- or overall AEs, including VTEs and cardiovascular AEs. Favorable tion of postmenopausal osteoporosis, requiring a postmenopausal effects on the lipid profile and minimal changes in coagulation and population at sufficient risk for osteoporosis and enrolling women carbohydrate parameters were consistent with well known estrogen of increasing age. Nevertheless, BZA/CE was associated with effec- effects. Based on these findings, BZA/CE demonstrated a favorable tive relief of menopausal symptoms in postmenopausal women of benefit-risk profile and represents a promising new menopausal ther- varying ages. apy with improved tolerability. In conclusion, the SMART-1 trial showed that the administration of a TSEC that partners BZA and CE was effective in treating symp- Acknowledgment: The authors acknowledge the contributions of the other toms associated with menopause, particularly vasomotor symptoms investigators of the study (see Appendix).

REFERENCES

1. Population Division U.S. Census Bureau. Table 1: ceptor modulator, in postmenopausal women. Obstet 17. Lobo RA. Lipids, clotting factors, and diabetes: en- Annual estimates of the population by sex and five- Gynecol 2005;105:1397–404. dogenous risk factors for cardiovascular disease. year age groups for the United States: April 1, 2000 9. Miller CP, Harris HA, Komm BS. Bazedoxifene ace- Am J Obstet Gynecol 1988;158:1584–91. to July 1, 2007. NC-EST2007–01. May 2008. tate. Drugs Future 2002;27:117–21. 18. Mosca L, Grundy SM, Judelson D, King K, 2. Lobo RA, Belisle S, Creasman WT, Frankel NR, 10. Boudes P, Ronkin S, Korner P, Baracat E, Limacher M, Oparil S, et al. Guide to preventive cardi- Goodman NE, Hall JE, et al. Should symptomatic Constantine G. Effects of bazeoxifene (TSE-424), ology for women. AHA/ACC Scientific Statement Con- menopausal women be offered hormone therapy? a novel tissue selective estrogen receptor modulator sensus Panel statement. Circulation 1999;99:2480–4. Med Gen Med 2006;8:1. (SERM), on the incidence of breast pain. Osteoporos 19. Manolio TA, Pearson TA, Wenger NK, Barrett- 3. National Osteoporosis Foundation. Fast Facts. Avail- Int 2003;14:S14. Connor E, Payne GH, Harlan WR. Cholesterol and able at http://www.nof.org/osteoporosis/diseasefacts. 11. Van Duren D, Ronkin S, Pickar J, Constantine G. heart disease in older persons and women. Review of htm. Accessed January 29, 2007. Bazedoxifene combined with conjugated estrogens: an NHLBI workshop. Ann Epidemiol 1992;2:161–76. 4. Miller PD, Chines AA, Christiansen C, Hoeck HC, a novel alternative to traditional home therapies. Fer- 20. Bass KM, Newschaffer CJ, Klag MJ, Bush TL. Plasma Kendler DL, Lewiecki EM, et al. Effects of bazedox- til Steril 2006;86:S88–9. lipoprotein levels as predictors of cardiovascular ifene on BMD and bone turnover in postmenopausal 12. Pickar JH, Yeh I-T, Bachmann G, Speroff L. Endome- death in women. Arch Intern Med 1993;153:2209–16. women: 2-yr results of a randomized, double-blind, trial effects of a tissue selective estrogen complex 21. Lobo RA, Pickar JH, Wild RA, Walsh B, Hirvonen E. placebo-, and active-controlled study. J Bone Miner containing bazedoxifene/conjugated estrogens as Metabolic impact of adding medroxyprogesterone Res 2008;23:525–35. a menopausal therapy. Fertil Steril 2009;92:1018–24. acetate to conjugated estrogen therapy in postmeno- 5. Silverman SL, Christiansen C, Genant HK, 13. Lindsay R, Gallagher JC, Kagan R, Pickar JH, pausal women. The Menopause Study Group. Obstet Vukicevic S, Zanchetta JR, De Villiers TJ, et al. Constantine G. Efficacy of tissue-selective estrogen Gynecol 1994;84:987–95. Efficacy of bazedoxifene in reducing new vertebral complex (TSEC) of bazedoxifene/conjugated estro- 22. Writing Group for the PEPI Trial. Effects of estrogen fracture risk in postmenopausal women with osteopo- gens (BZA/CE) for osteoporosis prevention in at- or estrogen/progestin regimens on heart disease risk rosis: results from a 3-year, randomized, placebo- and risk postmenopausal women. Fertil Steril 2009;92: factors in postmenopausal women. The Postmeno- active-controlled clinical trial. J Bone Miner Res 1045–52. pausal Estrogen/Progestin Interventions (PEPI) Trial. 2008;23:1923–34. 14. Archer DF, Lewis V, Carr BR, Olivier S, Pickar JH. J Am Med Assoc 1995;273:199–208. 6. Gruber C, Gruber D. Bazedoxifene (Wyeth). Curr Bazedoxifene/conjugated estrogens (BZA/CE): inci- 23. Lobo RA, Bush T, Carr BR, Pickar JH. Effects of Opin Investig Drugs 2004;5:1086–93. dence of uterine bleeding in postmenopausal women. lower doses of conjugated equine estrogens and me- 7. Komm BS, Kharode YP, Bodine PV, Harris HA, Fertil Steril 2009;92:1039–44. droxyprogesterone acetate on plasma lipids and lipo- Miller CP, Lyttle CR. Bazedoxifene acetate: 15. Walton C, Godsland IF, Proudler AJ, Wynn V, proteins, coagulation factors, and carbohydrate a selective estrogen receptor modulator with im- Stevenson JC. The effects of the menopause on insu- metabolism. Fertil Steril 2001;76:13–24. proved selectivity. Endocrinology 2005;146: lin sensitivity, secretion and elimination in non-obese, 24. Shlipak MG, Simon JA, Vittinghoff E, Lin F, Barrett- 3999–4008. healthy women. Eur J Clin Invest 1993;23:466–73. Connor E, Knopp RH, et al. Estrogen and progestin, 8. Ronkin S, Northington R, Baracat E, Nunes MG, 16. Spencer CP, Godsland IF, Stevenson JC. Is there lipoprotein(a), and the risk of recurrent coronary heart Archer DF, Constantine G, et al. Endometrial effects a menopausal metabolic syndrome? Gynecol Endo- disease events after menopause. J Am Med Assoc of bazedoxifene acetate, a novel selective estrogen re- crinol 1997;11:341–55. 2000;283:1845–52.

1034 Lobo et al. Effects of BZA/CE on menopausal symptoms Vol. 92, No. 3, September 2009 APPENDIX James Dockery, M.D. Drug Research & Analysis Corporation. Montgomery, Alabama. INVESTIGATOR LIST, SMART-1 STUDY Mark Akin, M.D. Costante Donati Sarti, M.D. CEDRA Clinical Research. Azienda Ospedaliera Perugina Policlinico Monteluce. Austin, Texas. Perugia, Italy. David Archer, M.D. Maxine Dorin, M.D. Eastern Virginia Medical School. University of New Mexico Hospital. Norfolk, Virginia. Albuquerque, New Mexico. Gloria Bachmann, M.D. Kyrin Dunston, M.D. Robert Wood Johnson Medical School. Womancare Obstetrics & Gynecology, PC. University of Medicine and Dentistry of New Jersey. Savannah, Georgia. New Brunswick, New Jersey. Mildred Farmer, M.D. Edmund Baracat, M.D. Meridian Research. Hospital do rim e Hipertenassao Fundacao. St. Petersburg, Florida. Sao Paulo, Brazil. Jan Faska, M.D. Thomas Benninger, M.D. NZOZ Medical. Bluegrass Clinical Research, Inc. Katowice, Poland. Louisville, Kentucky. Murray Berger, M.D. Robert Feldman, M.D. Radiant Research–Las Vegas. Miami Research Associates, Inc. Las Vegas, Nevada. Miami, Florida. Richard Beyerlein, M.D. Carol Feltheim, M.D. Paciﬁc Women’s Center, LLC. College Park Family Care Center. Eugene, Oregon. Overland Park, Kansas. Verdayne Brandenburg, M.D. Susan Floyd, M.D. Sioux Valley Clinic. Primary Physicians Research, Inc. Sioux Falls, South Dakota. Wexford, Pennsylvania. Kirk Brody, M.D. Sandra Force-Obrowki, M.D. ClinSearch. Investigative Clinical Research. Chattanooga, Tennessee. Rancho Cucamonga, California. Michael Burnette, M.D. Hendrik Robert Franke, M.D. Tampa Medical Group, P.A. Medisch Spectrum Twente Hospital Group. Tampa, Florida. Enschede, the Netherlands. David G. Chafﬁn, Jr., M.D. J. Christopher Gallagher, M.D. Marshall University Medical Center. Creighton University Medical School. Huntington, West Virgina. Omaha, Nebraska. Bryan Cowan, M.D. Nicola Garcea, M.D. University of Mississippi Medical Center. Azienda Ospedaliera S. Giovanni Addolorata. Jackson, Mississippi. Rome, Italy. Uel Crosby, M.D. Margery Gass, M.D. U.T. Southwestern Medical Center. Reproductive Medicine Research. Dallas, Texas. Holmes Hospital. Cincinnati, Ohio. Florenzo De Cicco Nardone, M.D. Universita Cattolica del Sacro Cuore Istituto di David Gearhart, M.D. Clinica Ostetrica e Ginecologica. New Ballas OB-GYN. Rome, Italy. Creve Coeur, Missouri. Domenico DeAloyisio, M.D. Andrea R. Genazzani, M.D. Universita degli Studi di Bologna. Universita degli Studi di Pisa. Bologne, Italy. Pisa, Italy. Janet Dietrich, M.D. Catherine Gerrish, M.D. Montana Health Research Institute. Brown Clinic, PLLP. Billings, Montana. Watertown, South Dakota. Peter Dietze, Jr., M.D. Robert Greene, M.D. Women’s Health Care at Frost Street. Specialty Care for Women. San Diego, California. Redding, California.

Fertility and Sterility 1035 Richard Godt, M.D. Scott Kleber, M.D. ClinPhase Research, LLC. Laureate Clinical Research Group. Upland, California. Atlanta, Georgia. Charles Goldsmith, M.D. Michael Kleerekoper, M.D. Clinical Research Institute of South Florida. University Women’s Care. Aventura, Florida. Southﬁeld, Michigan. Philippe Koninckx, M.D. Misericordia Guinot, M.D. University Hospital Gasthuisberg. Hospital de la Sta. Creu i San Pau. Leuven, Belgium. Barcelona, Spain. Gary Kraus, M.D. Virginia Hall, M.D. East Coast Clinical Research. Milton S. Hershey Medical Center. Salisbury, Massachusetts. Hershey, Pennsylvania. Samuel Lederman, M.D. L. Clay Harrell, M.D. Radiant Research, Inc. Metrolina Medical Research. West Palm Beach, Florida. Charlotte, North Carolina. Vivian Lewis, M.D. Bryan Hecht, M.D. University of Rochester. Mercy Medical Center. Medical Center. Canton, Ohio. Rochester, New York. Jorma Heikkinen, M.D. Gary Lipscomb, M.D. Deaconess Institute. UT Medical Group. Osteoporosis Clinic. Memphis, Tennessee. Oulu, Finland. Rogerio Lobo, M.D. Lester Ho, M.D. Columbia University. The Medical Group of Northern Nevada. New York, New York. Reno, Nevada. Steven London, M.D. Hans-Olav Hoivik, M.D. Center for Fertility & Reproductive Health. Hedmark Medisinske Senter AS. Shreveport, Louisiana. Harmar, Norway. Jose Lopez-Cintron, M.D. Mary Holm, M.D. Coastal Clinical Research. Odyssey Research Services. Orange City, Florida. Fargo, North Dakota. Mel Lucas, D.O. Terrence Horrigan, M.D. Patterson Medical Clinic. Medical College of Ohio. PrimeCare Research Associates. Toledo, Ohio. Florissant, Missouri. Joseph Hume, M.D. James Lyle, M.D. University of Kansas Medical Center. Alabama Clinical Therapeutics. Kansas City, Kansas. Birmingham, Alabama. William M. Johnson, III, M.D. Raymond Malamet, M.D. OB/GYN Associates of Alabama. The Osteoporosis and Clinical Trials Center. Birmingham, Alabama. Hagerstown, Maryland. Risa Kagan, M.D. Abe Marcadis, M.D. Foundation for Osteoporosis Research and Education. Comprehensive NeuroScience, Inc. Oakland, California. Boynton Beach, Florida. Bruce Kahn, M.D. Phyllis Marx, M.D. Scripps Clinic. Radiant Research–Chicago. La Jolla, California. Chicago, Illinois. Bruce Kessel, M.D. Diane Merritt, M.D. Queen Emma Outpatient Center. Washington University. Honolulu, Hawaii. St. Louis, Missouri. Douglas Kiel, M.D. Paul Miller, M.D. Beth Israel Deaconess Medical Center. GHS–Center for Women’s Medicine. Boston, Massachusetts. Greenville, South Carolina. Ellen Kim, M.D. Sam Miller, M.D. Albuquerque Clinical Trials, Inc. SAM Clinical Research Center. Albuquerque, New Mexico. San Antonio, Texas.

1036 Lobo et al. Effects of BZA/CE on menopausal symptoms Vol. 92, No. 3, September 2009 Valerie Montgomery-Rice, M.D. Rebecca Ryder, M.D. Meharry Medical College. Mid-Atlantic Women’s Care, PLC. Nashville General Hospital Clinical Research Center. Chesapeake, Virginia. Nashville, Tennessee. Joseph Sanﬁlippo, M.D. Arnold Moses, M.D. Magee Women’s Hospital. SUNY Upstate Medical University. Pittsburgh, Pennsylvania. Institute for Human Performance. Allan Sawyer, M.D. Syracuse, New York. HOPE Research Institute, LLC. Ken Muse, M.D. Phoenix, Arizona. University of Kentucky. Sherwyn Schwartz, M.D. Lexington, Kentucky. Diabetes & Glandular Disease Research Associates, P.A. Manubai Nagamani, M.D. San Antonio, Texas. University of Texas Medical Branch. Eric Sheldon, M.D. Galveston, Texas. Miami Research Associates, Inc. Robert Nordland, M.D. Miami, Florida. Western OB/GYN. Marek Sienkiewicz, M.D. Ridgeview Research. Skandynawskie Centrum Medyczne. Chaska, Minnesota. Wroclaw, Poland. Dale Osterling, M.D. Wilma Smit, M.D. The Florida Wellcare Alliance, LC. Gemini Hospital. Inverness, Florida. Den Helder, the Netherlands. Santiago Palacios, M.D. Robert Smith, M.D. Instituto Palacios de Salud y Medicina de la Mujer. Suncoast Clinical Research, Inc. Madrid, Spain. New Port Richey, Florida. Robert Parker, M.D. Rodney Smith, M.D. Lyndhurst Gynecologic Associates. Arizona Wellness Center for Women. Winston-Salem, North Carolina. Phoenix, Arizona. Tomasz Pertynski, Professor. William B. Smith, M.D. Klinika Ginekologii I Chorob Menopauzy. New Orleans Center for Clinical Research. Lodz, Poland. New Orleans, Louisiana. JoAnn Pinkerton, M.D. Erik Snorre Ofjord, M.D. University of Virginia Health System. Center for Clinical Trials. Northridge Midlife Health Center. Paradis (Bergen), Norway. Charlottesville, Virginia. Marek Spaczynski, Professor. Larry Popeil, M.D. Ginekologii I Poloznictwa Akademii. Magnolia Research Group, Inc. Poznan, Poland. Ocala, Florida. Ronald Spencer, M.D. Bruno Pornel, M.D. Renstar Medical Research. Brussels Menopause Center (BMC). Ocala, Florida. Brussels, Belgium. Elizabeth Puscheck, M.D. Leon Speroff, M.D. University Women’s Care, Inc. Oregon Health & Science University. Southﬁeld, Michigan. Portland, Oregon. Harvey Resnick, M.D. Daniel Spratt, M.D. R/D Clinical Research, Inc. OB/GYN Associates. Lake Jackson, Texas. Portland, Maine. Melvin Robinson, M.D. Dale A. Sundwall, M.D. Radiant Research, Inc. Salt Lake Women’s Center, PC. Pinellas Park, Florida. Sandy, Utah. Daniel Rowe, M.D. Nancy Teaff, M.D. Palm Beach Research Center. REACH. West Palm Beach, Florida. Charlotte, North Carolina. John Rubino, M.D. Pawel Teter, M.D. Triangle Medical Research Associates. Krajowe Centrum Osteoporozy. Raleigh, North Carolina. Warszawa, Poland.

Fertility and Sterility 1037 Janusz Tomaszewski, M.D. Lynn Westphal, M.D., Ph.D. Private Clinic of Obstetrics and Gynaecology. Stanford University Medical Center. Bialystok, Poland. Stanford, California. Daniel Tomlinson, M.D. Robert Wild, M.D. Medford Women’s Clinic, LLP. University of Oklahoma. Medford, Oregon. Health Sciences Center. Oklahoma City, Oklahoma. Suzanne Trupin, M.D. Women’s Health Practice. R. Stan Williams, M.D. Champaign, Illinois. University of Florida. Women’s Health at Magnolia Park. Marjo Tuppurainen, M.D. Gainesville, Florida. Laakariasema Cantti. Kuopio, Finland. Kathryn Witzeman, M.D. Denver Health Center. Wulf Utian, M.D., Ph.D., D.Sc.(Med) Women’s Care Clinic. Rapid Medical Research, Inc. Denver, Clorado. Cleveland, Ohio. Grattan Woodson, III, M.D. Dyonne van Duren, M.D. Atlanta Research Center. Menox. Decatur, Georgia. Nijmegen, the Netherlands. Mary Yankaskas, M.D. Richard Wasnich, M.D. Clinical Physiology Associates. Radiant Research–Honolulu. Clinical Study Center. Honolulu, Hawaii. Ft. Myers, Florida.

1038 Lobo et al. Effects of BZA/CE on menopausal symptoms Vol. 92, No. 3, September 2009 Menopause: The Journal of The North American Menopause Society Vol. 7, No. 6, pp. 395–401 © 2000 The North American Menopause Society ࠗϱ Text printed on acid-free paper.

Effects of estradiol with and without testosterone on body composition and relationships with lipids in postmenopausal women

Susan R. Davis, MD, PhD,1 Karen Z. Walker, PhD,2 and Boyd J. G. Strauss, MD3

ABSTRACT Objective: The cardioprotective effects of postmenopausal estrogen replacement therapy are mediated by several mechanisms, including favorable effects on lipids and lipoproteins. The extent to which the latter reflects modification of body fat distribution by sex steroids is not known. Hence, we investigated the relationships between changes in lipids and measures of body composition in postmenopausal women who were administered estrogen therapy with and without testosterone. Design: We randomized 33 postmenopausal women to treatment with either estradiol 50 mg (E) alone or estradiol 50 mg plus testosterone 50 mg implants (E&T) administered every 3 months for 2 years in conjunction with cyclic oral progestins for women with an intact uterus. Results: Both therapies were associated with sustained reductions in total cholesterol and low- density lipoprotein (LDL) cholesterol. In women who received E but not E&T, hip (p < 0.001) and abdominal circumferences (p < 0.05) and fat mass:fat-free mass (FM:FFM) ratio over the abdomen (p < 0.05) declined. E&T but not E resulted in increased FFM (p < 0.001) and a reduced FM:FFM ratio (p < 0.05). For E but not E&T, the decrease in LDL cholesterol was significantly related to changes in total and compartmental body fat and to change in the FM:FFM ratio (p < 0.05). Conclusion: Estrogen replacement has effects on body fat distribution in postmenopausal women that are associated with improved lipid parameters. Addition of parenteral testosterone does not negate the favorable effects of estrogen on LDL cholesterol levels but may attenuate the reduction in centralized body fat achieved with E implants. Key Words: Testosterone – Estradiol – Menopause – Body composition – Blood lipids.

oronary heart disease (CHD), the leading sequence of menopause, total cholesterol, triacylglyc- cause of death in women in industrialized erol (TAG), and low-density lipoprotein (LDL) choles- countries, generally affects women later in terol increase and high-density lipoprotein (HDL) Clife than men. The loss of ovarian estrogen cholesterol and its subfraction HDL-2 decrease.1 Body production at menopause is associated with metabolic composition and fat distribution also change in the changes that adversely influence cardiovascular dis- postmenopausal years such that women tend to lose ease risk. Probably the most studied of these is the ef- lean body mass,2,3 increase total body fat,3–5 and de- fect of estrogen deprivation on lipids and lipoproteins. velop a more centralized (android) pattern of body fat With increasing age in women and possibly as a con- distribution.2,5,6 The more android pattern of fat distribution is associated with higher risk of CHD.7,8 It is Received December 28, 1999; revised and accepted April 3, 2000. likely that these changes in body composition and fat From the 1Department of Epidemiology and Preventive Medicine, distribution are related to the development of a more Monash University, Prahran, Australia; 2Centre for Population Health adverse lipid profile in the postmenopausal years. and Nutrition, Monash University, Clayton, Australia; and 3Body Com- Estrogen replacement therapy (ERT), with and with- position Laboratory, Monash Medical Centre, Melbourne, Australia. out progestin, improves lipid profiles in both normoli- Address reprint requests to Dr. Susan R. Davis, Director of Research, 9 The Jean Hailes Foundation, 173 Carinish Road, Clayton, Victoria 3168, pemic and hypercholesterolemic postmenopausal 10,11 Australia. women, and these effects do not seem to be ad-

Menopause, Vol. 7, No. 6, 2000 395 ESTRADIOL AND BODY COMPOSITION versely influenced by the use of parenteral testoster- cyclical medroxyprogesterone acetate (Provera, Up- one.9 There are, however, few prospective data pertain- john Pty. Ltd, Rydalmere, NSW, Australia) 5–10 mg or ing to the effects of ERT on body composition and fat norethisterone (Primolut N, Schering Pty. Ltd., Alex- distribution or to the relationships between such andria, NSW, Australia) 2.5 mg orally for 12 days per changes and the effects of ERT on plasma lipids and month. All investigations were performed at entry into lipoproteins and the effects of concomitant testosterone the study and then at six monthly intervals for 2 years. therapy on these relationships. All investigators and research assistants involved in The administration of androgen replacement body composition measurements and data analysis therapy, usually in the form of testosterone, is used for were blinded as to each patients therapy. the restoration of libido in symptomatic women9,12 and Women were weighed without shoes and while for the prevention of bone loss after menopause.9,13 wearing light clothing or underwear. Weight was mea- With the increasing availability of testosterone and sured to the nearest 0.1 kg on a digital scale, and body other androgen supplements for women, the inclusion height was measured using a wall-mounted stadiom- of androgen therapy in postmenopausal hormone regi- eter. Body mass index was calculated as weight (kg) mens is becoming more widespread. However, the ef- divided by height (m) squared. Measurements of body fects of adding testosterone to estrogen therapy on body circumferences and skinfold thicknesses were under- composition have not been reported. We report the re- taken by a single skilled individual using standard pro- lationships between changes in body composition and cedures.14 The World Health Organization abdominal fat distribution and lipids and lipoproteins after long- circumference was taken as the greatest circumference term administration of estradiol alone or estradiol plus between the lowest rib and the top of the pelvis. Skin- testosterone. fold thicknesses were measured at the triceps, biceps, and subscapular and suprailiac sites with Harpenden MATERIALS AND METHODS calipers (Holtain Ltd., Crymych, UK). Body composi- Subjects tion was measured by dual-energy X-ray absorptiom- 9 etry (DXA) after a whole-body scan taken on a DPX-L As reported previously, 34 postmenopausal women scanner (DPX-L; Lunar Corporation, Madison, WI), (>12 months of amenorrhea and had serum follicle- which was standardized daily against a calibration stimulating hormone levels > 15 IU/L) who attended block. Total body fat mass (FM), the sum of fatty ele- the menopause clinic of Monash Medical Centre, Mel- ments in all fat tissue,15 was derived according to com- bourne, Australia, volunteered for this study, which puter algorithms supplied by the manufacturer (DPX-L was approved by the Human Research and Ethics Ad- software version 3.4, Lunar Corporation), and free-fat visory Committee of Monash Medical Centre, Mel- mass (FFM) was taken as total body tissue minus FM. bourne. All subjects gave their written informed con- In addition, an abdominal region of interest was de- sent. None of these women had previously been treated fined manually by delineating a superior border at the with androgens or had received hormone implants, al- level of the top of the L2 vertebra, an inferior border at though some had received oral ERT. the bottom of the L4 vertebra, and vertical borders drawn through the intersection of the superior border Methods with the left and right costal margins. The ratio of ab- Women were randomized independently to single dominal FM to FFM was then determined after compo- blind treatment with either estradiol 50-mg implants sitional analysis of this region of interest. alone (E) or estradiol 50-mg plus testosterone 50-mg Accuracy of total body fat by DXA has been as- implants (E&T), both donated for the study by Organon sessed by comparison to underwater hydrodensitom- Australia Ltd.9 Implants were administered every 3 etry in 12 healthy adult volunteers. The correlation was months for a period of 2 years. E or T implants were not r = 0.895 (p < 0.0001) with a between-method bias of + inserted if a preceding blood test indicated that serum 4.8% (range = 2–9%). Precision of the DXA was as- estradiol or testosterone levels exceeded 500 pmol/L or sessed by 10 repeated measures on one healthy volun- 4 nmol/L, respectively. During the study, 13 estradiol teer. The coefficient of variability (CV) was 1% for implants were withheld from seven women who re- percentage of fat, 0.6% for FFM, and 2.1% for FM. ceived E&T treatment, and 7 estradiol implants were Precision of the sum of the thicknesses of four skinfolds withheld from four women who received E alone. Thir- was assessed by eight repeated measures in three teen testosterone implants were similarly withheld.9 healthy volunteers by two trained technicians. The CV Women with an intact uterus were treated with either varied from 5.9% to 6.3%, depending on the technician.

396 Menopause, Vol. 7, No. 6, 2000 S. R. DAVIS ET AL.

For accuracy of the sum of the four skinfold thicknesses, the correlation with DXA was r = 0.921 (p < 0.00001). Serum estradiol and testosterone were measured by radioimmunoassay.9 Total cholesterol, TG, and HDL cholesterol were measured by automated standard methods,9 and LDL cholesterol was calculated according to Friedwald.16 Statistical analyses The data comprised repeated measurements on each individual at baseline and then every 6 months for 2 years. Results are expressed as the mean ± standard de- viation (SD). The baseline data were tested for treatment differences by two sample t tests. For 6-, 12-, 18-, and 24-month data, parameters were analyzed by mul- tivariate analysis of covariance (MANCOVA). Data were also compared at baseline and after 2 years in each group by Student’s paired t test. Relations between variables were established by Pearson’s correlation coefficient. Analyses were performed using Microsoft Excel 5.0 for the Macintosh. The level of significance was taken as 0.05. RESULTS Thirty-two women completed the study: 17 in the E group and 15 in the E&T group. One woman discontinued for personal reasons early after commencement, and the other discontinued after 12 months because of weight gain. At baseline, after randomization, the E group did not differ from the E&T group in smoking or alcohol habits, hysterectomy or ovariectomy status, estrogen or testosterone levels, or levels of blood lipids. Body mass index also did not differ between the two FIG. 1. Serum levels of estradiol (A) and testosterone (B) at baseline 2 (closed histogram) and at 2 years (open histogram) in 32 postmenopausal groups (24.6 ± 3.1 and 24.6 ± 3.3 kg/m for E and E&T women who received estradiol implants or estradiol implants plus tes- therapy, respectively). The mean age of the 17 women tosterone. Vertical lines indicate the standard error of the mean. *, sig- in the E group, however, was significantly lower than nificant change from baseline (p < 0.001). that of the 15 women in the E&T group (51.3 ± 5.7 years and 57.0 ± 5.2 years, respectively, p < 0.01). The effects of therapy on bone mineral density have All body composition variables were analyzed using been previously reported.9 Here we report the detailed age as a covariate; no significant effect of age was analysis of change in body composition and fat distri- demonstrated. bution and the relationships of these changes and Changes in the hormonal status of women before and change in lipoprotein lipids. after the study intervention are shown in Fig. 1. At Over 2 years, mean body weight decreased slightly baseline, the women who were receiving E and the in the E group (from 65.1 ± 9.2 kg to 64.1 ± 8.9 kg) and women who were receiving E&T did not differ signifi- increased slightly in the E&T group (from 63.4 ± 7.8 kg cantly in their levels of estradiol or testosterone. After 2 to 64.6 ± 9.7 kg), but these changes were not statisti- years of therapy, serum estradiol levels were signifi- cally significant. Similarly, total body FM as deter- cantly higher than at baseline in both treatment groups mined by DXA did not change significantly throughout (p < 0.001), whereas, as expected, serum testosterone the study (Fig. 2). Although there was a modest decline remained unchanged in the E group but increased with in FM with E&T therapy over 2 years (37.4 ± 7.1 kg to E&T treatment (p < 0.001). 35.9 ± 7.7 kg), this change did not achieve statistical

Menopause, Vol. 7, No. 6, 2000 397 ESTRADIOL AND BODY COMPOSITION

lesterol, however, was only significant in women who were given E alone (p < 0.01). Relationships between changes in LDL cholesterol and change in body fat were also examined. With E alone, there was a positive and significant relationship between the change in LDL cholesterol over 2 years and the change in total body fat (r = 0.494, p< 0.05) (Table 3). Change in LDL cholesterol was also significantly related to change in the FM:FFM ratio over the abdomen (r = 0.642, p < 0.01), to change in the total body FM:FFM ratio (r = 0.519, p < 0.05), and to the change in hip circumference (r = 0.670, p < 0.01). Simi- larly, the change in the ratio of LDL cholesterol to HDL cholesterol over 2 years was significantly related to change in total body FM, to change in the total body FM:FFM ratio, to change in the FM:FFM ratio over FIG. 2. Total body fat determined by DXA in 32 postmenopausal the abdomen, and to the change in hip circumference women who received estradiol implants (᭿) or estradiol implants plus (all p < 0.05) (Table 3). Analysis of comparable data testosterone (᭺) over a 2-year period. Vertical lines indicate the standard in women who received treatment with E&T indi- error of the mean. cated that none of the relationships was of statistical significance. significance (Table 1). In the E&T group but not in the E group, levels of FFM increased significantly over the DISCUSSION 2-year period (24.8 ± 5.9 kg to 27.9 ± 5.9 kg, p < 0.01), Women who are commencing ERT are often anxious whereas the FM:FFM ratio declined (p < 0.05). The that their treatment will exacerbate weight gain. This MANCOVA of body variables gave a significant treat- study confirms that estradiol implants, which result in ment effect (␹2 ≈ 17.26, 9 df, p < 0.05). relatively high circulating levels of estradiol, do not Change in the deposition of abdominal fat was also significantly increase body weight or total body fat in examined. In the group that was given estradiol im- postmenopausal women over a 2-year period. More- plants alone, there was a significant decline in the over, the addition of testosterone did not adversely af- FM:FFM ratio measured in a region drawn directly fect body weight or total body fat. over the abdomen (p < 0.05). This reflected both a de- Our results with E alone are in agreement with many crease in fat over the abdomen from 1.51 ± 0.57 to 1.43 previous studies. Although weight gain with ERT has ± 0.59 kg and a concurrent increase in FFM in this been reported,17,18 most studies indicate that ERT ei- area, from 3.12 ± 0.55 to 3.30 ± 0.52 kg, although ther decreases or has no influence on body weight.5,19–22 these changes in themselves did not reach statistical Similarly, despite one report that ERT increased the significance. percentage of body fat,17 the present study is consistent Changes in anthropometric measures of girth over with other findings that total body FM by DXA is un- the 2 years of the study are given in Table 2. Women affected by ERT20,23 and with a recent twin study that who received E implants alone showed significant de- indicated that in monozygotic twins discordant for ERT creases in hip and abdominal circumferences (p < 0.01 use, the twin that received the hormone therapy had the and p < 0.05, respectively). There was also a trend to- lower FM.24 ward a decrease in the umbilical circumference (p < After menopause, women typically experience a 0.08). These changes were not observed in women who loss of FFM,2,3 a decline that is not alleviated by were given E&T. ERT.17,20,23 In this study, FFM remained stable over 2 Over 2 years, both total cholesterol and LDL choles- years of ERT alone, but treatment with E&T signifi- terol declined in both treatment groups (Fig. 3). LDL cantly increased total FFM (p < 0.01) while decreasing cholesterol fell from 4.0 ± 0.89 mmol/L to 3.3 ± 0.94 the FM:FFM ratio (p < 0.05). This change has the po- mmol/L (p < 0.01) in women who were given E im- tential to influence the lipid profile; a study of 426 plants and from 4.1 ± 0.77 mmol/L to 3.4 ± 0.91 women in the Virgilio Menopausal Health Project in- mmol/L (p < 0.01) in women who were given E&T. dicated that levels of both total cholesterol and LDL The decline in the ratio of LDL cholesterol to HDL cho- cholesterol are inversely related to FFM.25

398 Menopause, Vol. 7, No. 6, 2000 S. R. DAVIS ET AL.

TABLE 1. Change in body composition in postmenopausal women who received hormonal replacement therapy with estradiol (n = 17) or with estradiol plus testosterone (n = 15) implants for a period of 2 years

Estradiol Estradiol plus testosterone

Parameter Baseline value After 2 y Baseline value After 2 y

Total body fat mass (kg) 35.7 ± 6.6 35.1 ± 6.6 37.4 ± 7.1 35.9 ± 8.0 Total body FFM (kg) 28.0 ± 6.5 28.2 ± 5.8 24.8 ± 5.9 27.9 ± 5.9a FM:FFM ratio 1.34 ± 0.41 1.31 ± 0.42 1.62 ± 0.58 1.38 ± 0.52b Fat mass over abdomen (kg) 1.51 ± 0.57 1.43 ± 0.59 1.60 ± 0.55 1.64 ± 0.65 FFM over abdomen (kg) 3.12 ± 0.55 3.30 ± 0.52 3.17 ± 0.45 3.31 ± 0.38 FM:FFM over the abdomen 0.48 ± 0.16 0.42 ± 0.14b 0.51 ± 0.18 0.51 ± 0.21 Data are the mean ± SD. aSignificant change over 2 y, p < 0.01. bSignificant change over 2 y, p < 0.05.

TABLE 2. Change in anthropometric measures of girth in postmenopausal women who received hormonal replacement therapy with estradiol (n = 17) or with estradiol plus testosterone (n = 15) implants for a period of 2 years

Treatment: estrogen plus Treatment: estrogen testosterone

Parameter Baseline value After 2 y Baseline value After 2 y

WHO abdominal circumference (cm) 92.8 ± 10.9 88.4 ± 11.8a 93.1 ± 11.5 89.9 ± 9.3 Umbilical circumference (cm) 89.7 ± 11.1 86.7 ± 10.8b 89.6 ± 11.1 88.3 ± 9.4 Hip circumference (cm) 100.4 ± 8.3 95.5 ± 9.4c 101.2 ± 6.0 97.9 ± 6.2 Data are the mean ± SD. aSignificant change over 2 y, p < 0.05. bTrend towards a change over 2 y, p < 0.08. cSignificant change over 2 y, p < 0.01.

TABLE 3. Relationships between change in LDL-cholesterol or in change in the ratio of LDL-cholesterol to HDL-cholesterol with change in body fat or hip circumference in postmenopausal women treated with estradiol (n = 17) or with estradiol plus testosterone (n = 15) implants for a period of2y

Change in Change in ratio of LDL: LDL-cholesterol HDL-cholesterol

E E&T E E&Tr value

Change in total body fat 0.49a 0.34 0.49a Change in the total body FM:FFM ratio 0.52a 0.33 0.49a FIG. 3. Changes in the lipid profile (total cholesterol, HDL cholesterol, Change in FM:FFM LDL cholesterol, and the ratio of LDL to HDL cholesterol) in 32 women over the abdomen 0.64b 0.41 0.52a 0.31 who received estradiol implants (E) or estradiol implants plus testoster- Change in hip one (E&T). ᮀ, E at 0 years; , E at 2 years; ᭿, E&T at 0 years; , E&T circumference 0.67b −0.04 0.56a 0.13 at 2 years. Vertical bars indicate the standard error of the mean. *Sig- Data given are r values. nificant different between 0 and 2 years, p < 0.05. aSignificant relationship, p < 0.05. bSignificant relationship, p < 0.01.

Although E and E&T treatment seem similar in their FM:FFM ratio in this region remained unchanged in effects on body weight and total body fat, they differed women who received E&T. There were, however, only in their effect on tissue composition in the abdomen. 15 women in the E&T group; this caused a lack of sta- Although the women who were treated with E implants tistical power. Nevertheless, our sample size calcula- alone exhibited a mean decrease in the FM:FFM ratio tions indicate that when observed differences in body in a region drawn over the abdomen (p < 0.05), the composition failed to reach significance in the E&T

Menopause, Vol. 7, No. 6, 2000 399 ESTRADIOL AND BODY COMPOSITION group, this was due to a lack of biologically relevant Endogenous androgen excess in postmenopausal difference because the sample size that was required to women is clearly associated with increased cardiovas- show significance was unrealistically high. cular risk, because of perturbations in lipid and carbo- Anthropometric measures corroborate our DXA hydrate metabolism, and a more android weight distri- findings. E therapy alone was associated with signifi- bution.26,28 In this study, undertaken with women of cant decreases in both hip and abdominal circumfer- normal body weight, it was found that although testos- ences (p < 0.01, p < 0.05, respectively). This was not terone therapy offset the reduction of centralized body seen for women who were treated with E&T. The an- fat evident after estrogen alone, it nevertheless still pre- thropometric and DXA data from women who were served the favorable effects of estrogen on the lipid pro- given E implants are consistent with previous studies file. Therefore, undesirable effects are extremely un- that showed by DXA4,5,23,26 or anthropom- likely and uncommon with testosterone replacement etry6,18,21,22,26,27 that ERT either prevents the increase therapy, with the caveat that circulating androgen lev- of central body fat after menopause or has a neutral ef- els are maintained close to or within the normal female fect, but the difference in the response of the women reproductive range and that patients are closely clini- who were given E&T suggests that the addition of T cally monitored.9 Additional studies are needed to es- may attenuate the favorable effect of reducing post- tablish whether testosterone therapy is inappropriate menopausal centralized fat accumulation with E alone. for more obese postmenopausal women, in whom an The clinical implication of this observation is that the adverse effect on blood lipids might become apparent. positive relationships between changes in body fat distribution and LDL cholesterol and the LDL cholester- Acknowledgment: We thank Elizabeth King for her clinical ol:HDL cholesterol ratio seen with E alone were not aid; Dr. Elizabeth Farrell for allowing us to conduct the study in the Menopause Clinic; and Nick Balazs, Director of Clinical seen with E&T. Differences in circulating lipoprotein- Biochemistry, Monash Medical Centre, for the lipid and hor- lipid concentrations have previously been reported to mone measurements. The study was supported by a grant from be associated with variations in the regional distribu- Organon Australia Ltd. tion of body fat.28 Over the 2-year period of our study, beneficial lipid changes were observed in women who REFERENCES were treated with E. This outcome is consistent with 10,11,29 1. Prelevic GM, Jacobs HS. Menopause and post-menopause. Bail- other studies and supports the hypothesis that the lieres Clin Endocrinol Metab 1997;11:311–40. beneficial effects on lipid parameters observed with 2. Aloia JF, Vaswani A, Russo L, Sheehan M, Flaster E. The influence postmenopausal estrogen replacement are the result of of menopause and hormonal replacement therapy on body cell mass and body fat mass. Am J Obstet Gynecol 1995;172:896–900. direct effects of estrogen on lipid metabolism com- 3. Poehlman ET, Toth MJ, Gardner AW. Changes in energy balance bined with favorable effects on central fat deposition. It and body composition at menopause: a controlled longitudinal is of interest that although the addition of testosterone study. Ann Intern Med 1995;123:673–5. 4. Ley CJ, Lees B, Stevenson JC. Sex- and menopause-associated to the estrogen therapy resulted in less pronounced changes in body-fat distribution. Am J Clin Nutr 1992;55:950–4. change in central body fat, it was nevertheless associ- 5. Gambacciani M, Ciaponi M, Cappagli B, et al. Body weight, body ated with an improved lipid profile (Fig. 3). Whether fat distribution, and hormonal replacement therapy in early post- the increase in FFM with testosterone acted as a meta- menopausal women. J Clin Endocrinol Metab 1997;82:414–7. 6. Bjorkelund C, Lissner L, Andersson S, Lapidus L, Bengtsson C. bolic counterbalance is not known. Reproductive history in relation to relative weight and fat distribu- Testosterone levels decline with age,30 and bioavail- tion. Int J Obes Relat Metab Disord 1996;20:213–9. able testosterone may decrease further in postmeno- 7. Lapidus L, Bengtsson C, Larson B, Pennert K, Rybo E, Sjostrom L. 31 Distribution of adipose tissue and risk of cardiovascular disease and pausal women who take oral estrogen. A case there- death: a 12 year follow up of participants in the population study of fore can be argued for concurrent androgen and women on Gothenberg, Sweden. Br Med J (Clin Res Ed) 1984;289: estrogen replacement after menopause.31 In particular, 1257–61. 8. Freedman DS, Williamson DF, Croft JB, Ballew C, Byers T. Rela- the addition of testosterone can significantly increase tion of body fat distribution to ischemic heart disease. The National 9,13 bone mineral density, and it also markedly im- Health and Nutrition Examination Survey I (NHANES I) Epide- proves measures of sexuality.9 In addition, as we miological Follow-up Study. Am J Clin Nutr 1995;142:53–63. showed here, the addition of testosterone to ERT re- 9. Davis SR, McCloud P, Strauss BJG, Burger H. Testosterone en- 2,3 hances estradiol’s effects on postmenopausal bone density and verses the decline of FFM seen after menopause, but sexuality. Maturitas 1995;21:227–36. in considering the use of testosterone therapy, these ad- 10. Darling GM, Johns JA, McCloud PI, Davis SR. Estrogen and pro- vantages need to be balanced against possible long- gestin compared with simvastatin for hypocholesterolemia in postmenopausal women. N Engl J Med 1997;337:595–601. term adverse effects on deposition of centralized body 11. Davidson MH, Testolin LM, Maki KC, von Duvillard S, Drennan fat. KB. A comparison of estrogen replacement, pravastatin and com-

400 Menopause, Vol. 7, No. 6, 2000 S. R. DAVIS ET AL.

bined treatment for the management of hypercholesterolemia in 22. Espeland MA, Stefanick ML, Kritz-Silverstein D, et al. Effect of postmenopausal women. Arch Intern Med 1997;157:1186–92. postmenopausal hormone therapy on body weight and waist and hip 12. Sherwin BB. Sex hormones and psychological functioning in post- girths. Postmenopausal Estrogen-Progestin Interventions Study In- menopausal women. Exp Gerontol 1994:29:423–30. vestigators. J Clin Endocrinol Metab 1997;82:1549–56. 13. Watts NB, Notelovitz M, Timmons MC, Addison WA, Wiita B, 23. Haarbo J, Marslew U, Gotfredsen A, Christiansen C. Postmeno- Downey LJ. Comparison of oral estrogens and estrogens plus an- pausal hormone replacement therapy prevents central distribution drogen on bone mineral density, menopausal symptoms, and lipid- of body fat after menopause. Metabolism 1991;40:1323–6. lipoprotein profiles in surgical menopause. Obstet Gynecol 1995; 24. Samaras K, Kelly PJ, Spector TD, Chiano MN, Campbell LV. To- 85:529–37. bacco smoking and oestrogen replacement are associated with 14. Lohman TG, Roche AF, Martorell R. Anthropometric standardiza- lower total and central fat in monozygotic twins. Int J Obes Relat tion reference manual. Champaign, IL: Human Kinetic Books, Metab Disord 1998;22:149–56. 1988. 25. Pasquali R, Casmirri F, Pascal G, et al. Influence of menopause on 15. Svendsen OL, Haarbo J, Hassager C, Christiansen C. Accuracy of blood cholesterol levels in women: the role of body composition, fat measurements of body composition by dual-energy x-ray absorpti- distribution and hormonal milieu. J Int Med 1997;241:195–203. ometry in vivo. Am J Clin Nutr 1993;57:605–8. 26. Heiss CJ, Sanborn CF, Nichols DL, Bonnick SL, Alford BB. Asso- 16. Friedwald WT, Levy RJ, Frederickson DS. Estimation of the con- ciations of body fat distribution, circulating sex hormones, and bone centration of low-density lipoprotein cholesterol in plasma without density in postmenopausal women. J Clin Endocrinol Metab 1995; use of the preparative ultracentrifuge. Clin Chem 1972;18:499–509. 80:1591–6. 17. Aloia JF, Vaswani A, Russo L, Sheehan M, Flaster E. The influence 27. Troisi RJ, Wolf AM, Mason JE, Klingler KM, Colditz GA. Relation of menopause and hormonal replacement therapy on body cell mass of body fat distribution to reproductive factors in pre- and post- and body fat mass. Am J Obstet Gynecol 1995;172:896–900. menopausal women. Obes Res 1995;3:143–51. 18. Reubinoff BE, Wurtman J, Rojansky N, et al. Effects of hormone 28. Despres J-P, Moorjani S, Lupien PJ, et al. Regional distribution of replacement therapy on weight, body composition, fat distribution, body fat, plasma lipoproteins, and cardiovascular disease. Arterio- and food intake in early postmenopausal women: a prospective sclerosis 1990;10:497–511. study. Fertil Steril 1995;64:963–8. 29. Pickar JH, Thorneycroft I, Whitehead M. Effects of hormone re- 19. Jensen J, Christiansen C, Rodbro P. Oestrogen-progestogen replacement therapy on the endometrium and lipid parameters: a replacement therapy changes body composition in early post- view of randomized clinical trials, 1985 to 1995. Am J Obstet Gy- menopausal women. Maturitas 1986;8:209–16. necol 1998;178:1087–99. 20. Hassager C, Christiansen C. Estrogen/gestagen therapy changes 30. Turcato E, Zamboni M, De Pergola G, et al. Interrelationships be- soft tissue body composition in postmenopausal women. Metabo- tween weight loss, body fat distribution and sex hormones in pre- lism 1989;38:662–5. and postmenopausal obese women. J Intern Med 1997;241:363–72. 21. Kritz-Silverstein D, Barrett-Connor E. Long-term postmenopausal 31. Casson PR, Elkind-Hirsch KE, Buster JE, Hornsby PJ, Carson SA, hormone use, obesity, and fat distribution in older women. JAMA Snabes MC. Effect of postmenopausal estrogen replacement on cir- 1996;275:46–9. culating androgens. Obstet Gynecol 1997;90: 995–8.

Menopause, Vol. 7, No. 6, 2000 401 The relationship between plasma estradiol and the increase in bone density in postmenopausal women after treatment willi subcutaneous hormone implants

J. Studd," M. Savv.. ,. N. Waston," T. Garnett: I. Fogelman,~ and D. Cooper" London, England

Twenty-three postmenopausal women with a median of 2 years past menopause (range, 1 to 12 years) and a median age of 52 years (range, 39 to 62 years) were recruited to participate In a longitudinal study designed to Investigate the factors that influence the Increase In bone density with subcutaneous estradiol and testosterone Implants. All women received 75 mg of estradiol with 100 mg testosterone subcutaneously. Bone density was measurad at the spine and hlp by dual-photon absorptiometry before therapy and after 1 year of subcutaneous hormonal therapy. The mean pratreatment bone density at the lumber vertebrae and neck of the femur was 0.84 grams of hydroxyapatite per squara centlmer (SO, 0.11) and 0.73 grams of hydroxyapatite per squara centimeter (SO, 0.10), respectively. The bone density at both sites Increased with values of 0.91 grams of hydroxyapatite per squara centimeter (SO, 0.11) and 0.75 grams of hydroxyapatite per square centimeter (SO, 0.11), respectively. These values represent an Increase of 8.3% (p < 0.0001) at the spine and 2.8% (p < 0.01) at the neck of the femur. The plasma estradiol level Incressed from a median of 80.5 pmollL to 453 pmol /L (p < 0.001). The percentage Incraaae of vertebral bone density was not related to age, number of years past the menopause, pretraatment bone density, or serum testosterone levels, but a slgnlficent correlation was found between the percentage Increase In bone density at the spine and the serum estradiol level (p <0.02, r '" 0.45). (AM J OSSTET GVNECOL 1990;163:1474-9.)

Key words: Osteoporosis, menopause, estradiol

After Albright'S· observation that estrogen therapy of estrogen necessary to achieve an increase in boDe can reduce urinary calcium excretion in postmeno density is not known. pausal women a number of prospective studies con The percutaneous route of administration avoKli the firmed the values of estrogen replacement therapy in enterohepatic circulation and is associated with p-liysi the prevention of postmenopausal osteoporosis.'''' Ep ologic plasma levels of estradiol and estrone. 'f.tiis lil in idemiologic studiest'·· also showed a reduction in contrast to the low levels of estradiol and high levCIs of the incidence of osteoporotic fractures with such estrone found after oral therapy with both conjugl!ted therapy. equine estrogens or estradiol valerate. 8.10 Subcutaneous Most studies used oral estrogen therapy, which al implants of estradiol and testosterone are effective in though effective in the suppression of climacteric symp the alleviation of climacteric symptoms," and a cross toms usually results in plasma estradiol and follicle sectional study" showed an apparent superiority ofequine estrogen' the optimal dose and route with this route when compared with oral estrOgeII therapy. We present the results of a prospective study of es From tJUJ Dulwich HospillJl Mmopawe Clinic, Dulwich Hospital and tradiol and testosterone implants on the bone densitY Fertility and Endocrine CnUre, Lister HoSPiliJl," tItt Dtpartment of and plasma hormone levels in postmenopausal women Nuclear Medicine, Guy's HOSPiliJl,b and tItt Compuur Unit, King's Collep School of Medicine and Dentistry.' over 1 year of therapy. Rectiutd for publication March 26, 1990; revised July 16, 1990; accepted July 2J, 1990. Patlente and methode Repri1" nquats.' John Studd, MD, Dulwich HosPillJl, East Dulwich Grove, London, England SE 22 8PT. A total of 23 postmenopausal women with a med@ll 611124109 age of 52 yean (range, !l9 to 62) were recruited intD

1474

-_.,;,a\~_ II. Vo!Ullle 163 Relationship between plasma estradiol and bone density 1475 NUJllber 5. Part I

Table L Mean values of bone density at lumbar spine and neck of femur, body weight, blood pressure, ease in and plasma hormone values before therapy and after 1 year of estradiol and testosterone implants : with = Trelltmmt I Pretreatment I 1 Y tllr IIfttT /rell/men/

Mean bone density at 0.84 (0.11) 0.91 (0.11)* spine gHa{cml (SO) Mean bone density at 0.73 (0.10) 0.75 (O.II)t femur gHa/cml (SO) FSH (lUl L) 71 (28-100) 12 (1-62)* median (range) Estradiol (pmoIlL) 80.5 (30-580) 453 (204-883)* median (range) Testosterone (nmoIlL) 0.6 (0.3-1.8) 0.9 (0.4-2.4)1 median (range) Mean weight in kg (SO) 67.3 (7.7) 66.7 (7.0) Mean blood presure 120 (37)173 (12) 125 (21)175 (13) (mm Hg) (SO)

*~ '" 0.0001, Student's paired / test. t~ <: om, Student's paired / test. *~ < 0.00 1, Wilcoxon matched-pair signed-rank test. I~ < 0.01, Wilcoxon matched-pair signed-rank test.

this prospective study. The median number of years tween the changes in bone density and the serum es ~. the menopause was 2 (range, 1 to 12). Menopausal tradiol levels, serum testosterone levels, age, number status was defined as amenorrhea > 1 year's duration of years past the menopause, and pretreatment bone with a ICrum FSH level >20 IU IL. density. All women received hormone replacement therapy by subcutaneous implants of 75 mg estradiol and 100 Results mg testosterone. A further implant of the same dose Of the 23 patients studied 22 patients showed an crease in bone of. estradiol and testosterone was given after 6 months increase in the bone density after 1 year of therapy and again at 1 year. (Table I). The mean bone density at the lumbar spine ltion avoiClilthe Assays of FSH, estradiol, and testosterone were per before therapy was 0.84 gHA/an' (SD 0.11), which ted with pbysi formed before treatment and at 1 year. The bone den increased to 0.91 gHA/cm' (SD, 0.11) after 1 year rone. This is in sity was estimated in the spine of L2-4 and in the (P < 0.0001). The value for the proximal femur was d high levels of neck of the right femur with a Novo 22A BMC-LAB 0.73 gHA/an' (SD, 0.10), which increased to 0.75 oth conjugated with gadolinium 153 as the source of radiation. The gHA/cml (SD, 0.11, P < 0.01). The mean increase in ~ Subcutaneous absorptiometer was standardized with a solution ra the lumbar spine was 0.07 gHa/cm' (95% confidence CIiologillaUy equivalent to hydroxyapatite, and the re interval, 0.06 to 0.08 gHa/cm'). The mean increase at sults were expressed as grams of hydroxyapatite per the femur was 0.02 gHa/cm' (95% confidence interval, 2 unit projected area· of bone in square centimeters 0.01 to 0.04 gHa/cm ). These values represent a mean l (gHa/an ). The precision of the machine with the use increase in the bone density of 8.3% at the spine and of, llie phantom Was 0.69% and short-term precision 2.8% at the neck of the femur after 1 year. for nOl'mal su~jects was 2.04%. Measurements were The median serum FSH level before therapy was made before ihse~on of the hormone implant and at 71 lUlL (range. 28 to 100 lUlL) with a median se 12 months after therapy within 1 week of the insertion rum estradiol level of 80.5 pmol/L (range, 30 to 580 of the third implant. pmol/L) and a serum testosterone level of 0.6 nmol/L ve study of< C5' Statistical analysis. The bone density values before' (range, 0.3 to 1.8). After 1 year of therapy the serum e bone density an~ after t~eatmer:lt approximated to a normal Clistri FSH level significantly reduced to 12 lUlL (range, 1 JpausaiwomeP lJutibn and the increase was therefore ~~yzed with to 62; p < 0.001), the serum estradiolleve1 increased the paired Student t test. The 'tiormon~ parameters to 453 pmol/L (range, 204 to 883; P < 0.001), and the coula not ?e regarded as normally distributed and the serum testosterone level increased to 0.9 nmol/L ~ges were analyzed with 'the Wilcoxon matched (range, 0.4 to 2.4; p < 0.01). with a medilm ~ signed~ratik test. The Spearman correlation co A significant correlation was found between the per recruited into ef1icient ',Yas calculated to express the relationship be- centage increase of vertebral bone density and the 1476 Studd at al. NoVl:nlliei; AmJObstet~

V. Change in 15 vertebral bone density

- 5 200 300 400 1100 SOD 700 SOD 1100 Serum estradiol pmolll after one year of Implant therapy

Fig. 1. Correlation between percentage increase in venebral bone density with plasma estradiol levels after I year of therapy with estradiol and testosterone implants.

15 - " Change In vertebral bone density 10

· 11 OB 07 DB 09 11

Initial bane denalty gHA/cm2

Fig. 2. Effect of pretreatment bone density on percentage increase in bone density with estradiol and testosterone implants.

plasma estradiol levels achieved after I year of ther Christiansen et aI. IS reported that the use of a com apy(Fig. I, r = 0.45; p < 0.02). There was no signifi bined preparation of oral estradiol, estriol, and nor· cant correlation between the increase in bone density ethindrone within 3 years of menopause resulted in an arid the initial bone density (Fig. 2), age (Fig. 3), the increase in bone density of 3% after 3 years of therapy. number of years past the menopause (Fig. 4), and the Lindsay et aI. \I (1984) found an increase of 2% to 4 serum testosterone level (Fig. 5). over 5 years when mestranol 25 ).Lg was prescribeCI within 3 years of menopause but this dose only maiJI,. Comment tained bone density if it was started more than 3 years These data show an increase of 8.3% in the bone after menopause. More recently Munk-Jensen et ail density at the spine and 2.8% at the hip after 1 year of reported an increase of6.4% in the vertebral bOnede:n therapy with subcutaneous estradiol and testosterone sity of women treated within 2 years of menopawe wiJ!I implants. Evidence from our cross-sectional study with continuous oral estradiol and norethindrone for 1 year. the use of the same method of bone densitometry re None of these authors reported the estradiollevels l ~ porting the vertebral bone density of 1.02 gHal cmt and tained with the therapy. In our study we were atile: to proximal femur values of 0.80 gHa/cm2 after 8 years show an increase in bone mass even 10 years afttr of such therapy'2 strongly suggest that this increase is menopause. This increase was the same as founa in the not transient and will be maintained over the years. younger postmenopausal woman and was also inlle:- Relationship between plasma estradiol and bone density 1477 ~el6! \IIII\iCr 5. Pan I

% Change In 16 ~ vertebral bone density 10

j t

·5 36 40 45 150 515 60 615 70 Age levels Fig. 5. Effect of age on percentage increase in bone density with estradiol and testosterone implants.

% Change In 115 vertebral bone density 10 I

- 5 o 2 4 8 S 10 12 14 16 lB 20 22 Menopausal age

adiol Fig. 4. Effect of number of yean past menopause on percentage increase in bone density with estradiol and testo~te' .ne implants.

the use of a cOlii' eendent of the initial bone density. The correlation fects of testosterone on the skeleton may partly explain estriol. and nor· between the percentage increase in bone density and these results and this component of the therapy is sub use resulted in an ibe serum estradiol levels supportS the hypothesis that ject to further investigation . . yean of thenlPY' illegreater effectiveness of this mode of therapy may The increase in bone density was greater in the spine ease of 2% to 4 lie a result of the higher serum estradiol levels achieved. than that at the neck of the femur because the more g was prescribed Oral estrogen therapy is associated with lower estradiol active trabecular bone predominates in the vertebra. I. I dose only main· belS and a less good skeletal response. I~ There are several models that show such substantial nore than !I years l'lie importance of the testosterone component of reversal of bone loss. particularly in the vertebra. Trea nk-Jensen et aLII ihe treatment is unknown but there was no correlation sure et a1. IV showed in a cross-sectional study that the rtebral bone den' III diU study between plasma testosterone levels and the bone loss of anorexia nervosa is reversed when men f menopause with lIIcrease in bone density. Barlow et a1. IO could not show struation returns. Greenspan"" showed that the osteo ldrone for 1 yeat· that the addition of testosterone to estrogen therapy porosis of hyperprolactinemic men with hypogondism stradiol levels 011- l!II}lJoved bone density. However there is some evi recovers when the hyperprolactinemia is treated. Matta f we were able to ~e that adrenal androgens may have a role in main et a1. (1988)21 reported that the 5.9% loss of bone den n 10 years after taining the bone mass in postmenopausal women in the sity after the use of buserelin for 6 months is reversed le as found imdle SlUdy of postmenopausal women with treated Addi- when it is discontinued. These conditions are all char d was aIiQ inde- 1C!Il1s disease. 17 It is thus possible that the anabolic ef- acterized by hypogonadism and the studies indicate 1478 Studd et a!. November lllgo Am J Obslet GytIecoJ

15 '!(, Change In + vertebral bone density 10

- 15 0 0.15 1.5 2 25 3 Serum testosterone nmolll after one year of Implant thll.fapy

Fig. 5. Effect of serum testosterone after I year of therapy on percentage increase in bone density at spine.

that bone can be replaced in both men and women We gratefully acknowledge the assistance of DerecJ( when plasma sex hormones return to normal physio Lowe, medical statistician, King's College Hospital, with logic levels. the statistical analysis. The current recommended oral dosage of estrogens will suppress menopausal symptoms and prevent fur REFERENCES ther bone loss but the serum estradiol and FSH levels I. Albright F. Smith PH. Richardson AM. Postmenopausal achieved often remain in the postmenopausal range.'2 osteoporosis; its clinical features.JAMA 1941;1l6:246S- The dosage of orally administered estrogen may be 74. limited by gastrointestinal symptoms and the liver im 2. Lindsay R. Hart OM. MacLean A. et al. Pathogenesis and of postmenopausal osteoporosis. In: Cooke ID, ea. The pact on estrogen metabolism. It is probable that the role of estrogen! progestogen management of the meno, safest way to achieve serum estradiol levels adequate to pause. M. Lancaster. 1978:9-25. produce significant increases in bone mass is by the !I. Natchigall LE. NatchigaU RH. Natchigall RD. Beclunao E. Estrogen replacement therapy: a 10-year prospective percutaneous route with either transdermal patches or study in the relationship to osteoporosis. Obstet GynecoI subcutaneous implants. 1979;53:277-81. The mechanism by which estrogen deficiency causes 4. Christiansen C. Christiensen MS. McNair p. Hagen C. Stocklund K. Transbol I. Prevention of early postmeoo loss of bone remains unclear but the original Albright pausal bone loss: controlled 2-year srudy in !l15 nonnal hypothesis of an association with a generalized loss of females. Eur J Clin Invest 1980;10:27!1-9. collagen including the collagenous matrix of the bone 5. Ross RK. Panganini-Hill A. Mack TM. Reduction in £rae. rures and other effects of estrogen replacement theral!l' and the collagen of the skin is relevant. The substantial in human population. In: Christiansen C. ed. Proc:eediIIp increase in vertebral spine density shown in this study of the International Symposium on Osteoporosis 1984; can be compared with the 30% increase in skin collagen 1:289-97. 6. Keil OP. Felson DT. AndersonJ1. Wilson DWF. Moskowitz and a 25% increase in skin thickness that occurs in MA. Hip fractures and the use of estrogens in post. postmenopausal women who receive percutaneous es menopausal women. N EnglJ Med 1987;317:1169-74: tradiol therapy." Only serial histomorphometric stud 7. Chetkowski RJ. Meldrum DR. Steingold KA. et aL BIO logic effects of transdermal estradiol. N Engl J Med ies of bone biopsy specimens in these patients will reveal 1986;!I 14: 1615-20. whether the estrogen-promoted changes in skin colla 8. Lindsay R. Hart DM. Clark HD. The minimum effecdYe gen are reproduced in the collagenous matrix of post dose of estrogen for prevention of postmenopau5il !Jc!pe loss. Obstet Gynecol 1984;6!1:759-63. . menopausal women. 9. Thorn M. Collins WP. Studd JWW. Hormonal profiles III It is reassuring from this study that estradiol and post-menopausal women after therapy with subcutaneous testosterone implants may not only prevent osteopo implants. Br J Obstet Gynaecol 1981:88:426-S!I. 10. Powers MS. Schenkel L. Darey PEt Good WR. Balesi;"' rosis but will also be valuable in the older woman who Je. Place VA. Pharmacokinetics and pharmacod}'l!~ might believe that it is too late to commence estrogen of transdermal forms of 17P.estradiol: a comparison WI'" therapy. This therapy is also appropriate for the youn conventional oral estrogens used for hormone ~p'lice ment therapy. AM J OBSTEl' GYNECOL 1985; 152: 109!P111§; ger woman with premature ovarian failure who has 11. Cardozo lO, Gibb OMF. Studd JWW. Tuck SM, 'F6: already suffered substantial bone los5. MH. Cooper DJ. The use of hormone implants for November 19Do Volume 1611 RelationshIp between plasma estradIol and bone densIty 1479 I J Obslel Gfl1tco1 NUJIIbcr 5, Pan 1

maeteric symptoms. AM J Ossn:r GVNECOL 1984:148: an effect of adrenal androgens on bone mass. Br Med J !I!I6·7. 1986:294:798-800. 12. SaVYas M. Studd JWW. Fogelman I. Dooley M, Montgo' 18. Smith R. Recent advances in the metabolism and physi· mery J. Murby B. Skeletal effects of oral estrogen com ology of bone. In: Baker PF. ed. Recent advances in phys pared with subcutaneous estrogen and testosterone in iology. vol 10. Edinburgh: Churchill Livingstone. 1984: postmenopausal women. Br Med J 1988;297:!l81·8. 817-48. 15. Christiansen C, Christiansen MS, Transbol I. Bone 19. TreasureJ. Russell GFM, Fogelman I. Murby B. Revers· mass in postmenopausal women after withdrawal of es ible bone loss in anorexia nervosa. Br Med J 1987; trogen/ gestagen replacement therapy. Lancet 1981: 1: 295:474-5. 459·61. 20. Greenspan SL. Neer RM, Ridgway EL. Klibanski A. Os H. Undsay R, Hart DM. Aitken JM. MacDonald EB, An teoporosis in men with hyperprolactinaemic hypogonad derson JB. Clare AC. Long-term prevention of post ism. Ann Intern Med 1986;104:777·82. menopausal osteoporosis by estrogen. Lancet 1976; 1: 21. Matta WHo Shaw RW. Hesp R. Evans R. Reversible tra 10!l8·40. becular bone density loss following GnRH analaogue, 15. Munk-Jensen N. Pors Nichen S. Obe1 EB. Bonne Eriksen Buserelin in premenopausal women. Clin Endocrinol P. Reversal of postmenopausal vertebral bone loss byes 1988:29:45-51. trogen and progestogen: a double-blind placebo con 22. Studd JWW. Anderson HM. Montgomery JC. Hormone troUed study. Br Med J 1988;296: 1150-2. treatment selection of patients- kind and duration of 16. Barlow EH, Abdalla HI. Roberts ADG. AI Azzawi F. Leg treatment. In: Greenblatt RB, ed. A modern approach to pte I. Hart DM. Long-term hormone implanllherapy the menopause. Berlin: de Gruyter, 1986;182·98. sily hormonal and clinical effects. Obstet Gyneco1 1986: 2!1. Brineat M. Kabalan S. Studd JWW. et al. A study of the 67:!l21-5. relationship of skin collagen content, skin thickness and 17. Devogelaer JP. Crabbe J. Nagant De Deuxehainsnes C. bone mass in the post menopausal woman. Obstet Gyneeol Bone mineral density in Addison's disease: evidence for 1987:840·5. ,tance of Dereck ge Hospital, with

il. Posunenop'!!1IiIl IA 1941:116:2~~

I. Pathogenesis!Jld ::ooke lD. ed. iFhe ment of tlie menD-

igall RD. Becbm,an 10-yearpr~ ,sis. Obstet Gyneall

cNair P. Hagen <;, of early pos!!!IeIIO .udy in 81l! oonml 78-9. I. Reduction in frae eplacement· dierapy n C, ed. PTOteedinP Osteoporosis 1984;

IOn DWF. MoskoWil . estrogens in P,lII!' 1987:!l17:1169-74- gold KA. et all BIG: iot N Engl J Met!

e minimum effedJ'C oSlmenopau,.!(bollC '. 1:'-111 Hormonal pro ...... ly with subCiJ~ 1:88:426-!l!1. . Good WR.B~ 1 pharmac~}'l.I~ Dl:acom~n~ Dr hormone ,rep! 06- .1985;152:1~1JdI 'W. Tuck SM, i& ,one implants for

Hindawi Publishing Corporation International Journal of Alzheimer’s Disease Volume 2012, Article ID 258454, 18 pages doi:10.1155/2012/258454

Review Article Hormone Replacement Therapy and Risk for Neurodegenerative Diseases

Richelin V. Dye,1 Karen J. Miller,1 Elyse J. Singer,2 and Andrew J. Levine2

1 Longevity Center, Department of Psychiatry and Biobehavioral Sciences, David Geﬀen School of Medicine at the University of California, Los Angeles, CA 90025, USA 2 National Neurological AIDS Bank, Department of Neurology, David Geﬀen School of Medicine at the University of California, Los Angeles, CA 90025, USA

Correspondence should be addressed to Andrew J. Levine, [email protected]

Received 29 November 2011; Revised 17 January 2012; Accepted 18 January 2012

Academic Editor: K. S. Jagannatha Rao

Copyright © 2012 Richelin V. Dye et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Over the past two decades, there has been a significant amount of research investigating the risks and benefits of hormone replacement therapy (HRT) with regards to neurodegenerative disease. Here, we review basic science studies, randomized clinical trials, and epidemiological studies, and discuss the putative neuroprotective effects of HRT in the context of Alzheimer’s disease, Parkinson’s disease, frontotemporal dementia, and HIV-associated neurocognitive disorder. Findings to date suggest a reduced risk of Alzheimer’s disease and improved cognitive functioning of postmenopausal women who use 17β-estradiol. With regards to Parkinson’s disease, there is consistent evidence from basic science studies for a neuroprotective effect of 17β-estradiol; however, results of clinical and epidemiological studies are inconclusive at this time, and there is a paucity of research examining the association between HRT and Parkinson’s-related neurocognitive impairment. Even less understood are the effects of HRT on risk for frontotemporal dementia and HIV-associated neurocognitive disorder. Limits to the existing research are discussed, along with proposed future directions for the investigation of HRT and neurodegenerative diseases.

1. Introduction abilities in healthy postmenopausal women, improvement of Parkinson’s symptoms, and variably altering risk of Hormone replacement therapy (HRT), defined here as use neurodegenerative disease. of various types of estrogen alone or in conjunction with progestins (synthetic or exogenous progestogen), has long been studied as a possible prophylactic against Alzheimer’s 2. Alzheimer’s Disease disease. While the association between HRT and Alzheimer’s disease has been explored through several observational Alzheimer’s disease (AD) represents the most common and randomized clinical trials to date, the relationship neurodegenerative disease, accounting for more than 50% between HRT and other neurodegenerative diseases has of all dementia types [1]. Within the United States alone, received relatively little attention. In this review, we explore national prevalence estimates indicate that AD affects 2.4 the body of research on HRT as a prophylactic against million individuals aged 70 and older [1, 2]. With increasing various neurodegenerative conditions, including Alzheimer’s age, AD progressively affects more individuals, affecting 2.5% disease, Parkinson’s disease, frontotemporal dementia, and of those aged 71–79, 18% of those aged 80–89, and 30% of HIV-associated neurocognitive disorder. In reviewing obser- those aged 90 and older [1, 2]. vational studies, randomized clinical trials, and basic sci- Cognitive decline in AD is characterized by insidious ence studies, we find evidence that some forms of HRT onset and gradual progression over a course of several are neuroprotective, resulting in preservation of cognitive years [3–5]. Clinical research has identified subtle losses 2 International Journal of Alzheimer’s Disease of cognitive functioning that precede AD. Studies have resources. An additional criticism is the lack of controls in consistently shown that a deficit in episodic or verbal the studies; for instance, all observational studies described memory, specifically the ability to encode novel information, above involved varied ERT regimens among all participants is an early symptom of AD and often presents several rather than a uniform ERT regimen. Thus, the findings of the years before a formal diagnosis of AD [3, 6–9]. Such observational studies present with several limitations. observations have led to the identification of a preclinical Although all of the studies examined above have included stage of AD that represents the transition between normal women who underwent natural menopause, recent observa- aging and AD. Specifically, mild cognitive impairment (MCI) tional studies have examined the differences between women represents the mild neurocognitive decline that occurs in the who underwent natural versus surgical menopause [15]. In presence of relatively intact day-to-day functioning [4, 5]. one, women who had surgical menopause demonstrated an Although there are several subtypes of MCI, the subtypes increased long-term risk for cognitive impairment compared that are at increased risk for the development of AD involve to women with natural menopause. In another paper predominant memory impairment. It has been estimated based on the same data, the same researchers reported a that approximately 10–15% of those diagnosed with MCI linear trend, with increased risk seen with younger age at with predominant memory impairment convert to AD per oophorectomy (bilateral or unilateral) [16]. These findings year [4, 5]. The identification of MCI as a possible prodrome suggest that earlier age of surgical menopause increased risk to AD, as well as the recent development classifier algorithms of cognitive impairment and that estrogen deficiency may that assess later risk for AD based on a variety of clinical initiate risk for neurological diseases such as AD. Notably, factors [10], leaves open the potential for initiating therapies, the researchers also found increased risk of depression and including HRT, that may prevent progression to AD in those cardiovascular disease among women with history of bilat- at risk. eral oopherectomy, suggesting that the relationship between surgical menopause and cognitive impairment may be 2.1. Estrogen and Risk for AD—Observational Studies. Inci- multifactorial [17, 38]. dence rates indicate the risk of AD among women is double that of men after the age of 80, even after controlling 2.2. Randomized Clinical Trials of HRT in Healthy and At- for protective factors such as education [11]. The higher Risk Women. While observational studies generally support incidence rate of AD among women have led to explorations a neuroprotective role for ERT against AD, the results of on the association between estrogen deficiency and AD. randomized clinical trials (RCTs) have been equivocal. To Observational studies have examined both HRT and date, the largest study has been the Women’s Health Initiative estrogen replacement therapy (ERT), or estrogen alone, in Memory Study (WHIMS), an ancillary study of the Women’s relation to incidence of AD (see Table 1). For instance, in a Health Initiative (WHI), a prospective study that enrolled sample of 514 women enrolled in the Baltimore Longitudinal 7479 postmenopausal women [39, 40]. A total of 4532 Study of Aging, Kawas et al. found that ERT was associated women with natural menopause (intact uterus) were ran- with significantly reduced risk for AD [12]. Although the domized into a trial comparing conjugated equine estrogen duration of use ranged between 1–15 years, the data did not (CEE) + medroxyprogesterone (MPA) versus placebo [40]. show a significant effect for duration of ERT. In addition, However, the trial was discontinued before completion due no effect was observed for age of menopause. In another to unexpected health risks. Despite the early termination, observational study reported by Tang et al., ERT was also data revealed that women who received CEE + MPA associated with significantly reduced risk for AD in a sample demonstrated greater cognitive decline compared to the of 1124 women enrolled in the Manhattan Study of Aging placebo group [40]. Additional analyses revealed that risk [13]. Here, however, an inverse relationship was observed for dementia was doubled for women who received CEE + for duration of use and risk for AD, with the lowest risk MPA compared to the placebo group [39]. Taken together, noted for women taking estrogen for longer than one year. data from the WHIMS demonstrated a higher incidence Other observational studies have provided moderate support of dementia and greater cognitive decline among hormone for decreased AD risk with ERT and the importance of users relative to placebo groups. duration of use. Using retrospective data on a sample of Although the WHIMS has been considered one of the 355 women, Paganini-Hill and Henderson found that ERT largest and longest randomized studies examining HRT and was associated with moderately reduced risk for development cognitive deficits, generalizability of the findings is affected of AD [14]. An inverse relationship was seen for duration by several limitations. First, external validity of the WHIMS of ERT and risk for all-cause dementia (AD as well as findings has come into question, as the participants in the other causes of dementia), with those on ERT for seven or treatment group were at high risk for cardiovascular and more years having the lowest risk for AD. While findings cerebrovascular disease; thus the higher rates of dementia from these observational studies suggest that ERT may may have been attributed to vascular disease. Second, in their reduce risk of AD, given the nature of observational studies analyses of the WHIMS data, the researchers included all the findings may be affected by several biases. Specifically, dementia types into an “all-cause” dementia that included the women who decided to take ERT for several years AD, vascular dementia, dementia due to Parkinson’s disease, may have been healthier to begin with; they may have and frontotemporal dementia, thus limiting the interpreta- also been more proactive in seeking early postmenopausal tion of results. Third, a methodological limitation included treatment due to higher education and/or availability of the unavailability of baseline cognitive measures prior to International Journal of Alzheimer’s Disease 3

Table 1: Observational studies of ERT and risk for dementia. Study (reference) Sample description Overall findings 355 postmenopausal women (165 users; ERT (not specified) for 1–7 years was associated with reduced Paganini-Hill and 190 nonusers) with a mean age of 86.5 risk for AD (OR: 0.67, CI 95% 0.38–1.17) compared to Henderson [14] years at death; retrospective data from the nonusers. Risk for AD decreased with longer duration of use. Leisure World, Laguna Hills cohort 1124 healthy postmenopausal women After controlling for age, education, and ethnicity, ERT (156 users; 968 nonusers), with a mean (majority used CEE) for 6–8 years was associated with lower Tang et al. [13] age of 74.2, enrolled in the Manhattan risk for AD (OR 0.50, 95% CI, 0.25–0.90) compared to Study of Aging nonusers. Risk for AD decreased with longer duration of use. 514 healthy postmenopausal women After controlling for education, ERT (not specified) for 1–10 (230-users; 242-non-users), with a mean years was associated with lower risk for AD (OR: 0.46, 95% Kawas et al. [12] age of 65.5, enrolled in the Baltimore CI, 0.21–0.99) compared to non-users. No effect was Longitudinal Study of Aging observed for duration of use. Women who underwent oophorectomy (unilateral or 813 women with unilateral bilateral) before onset of menopause were at increased risk Rocca et al., oophorectomy, 676 women with bilateral for cognitive impairment or dementia (OR: 1.46, 95% CI, [15–17] oophorectomy, and 1,472 women who 1.13–1.90) compared to women who did not undergo did not undergo oophorectomy. oophorectomy. Risk increased with younger age at oophorectomy. treatment; thus, participants may have already been cog- combination therapies that include progestin may actually nitively impaired prior to beginning HRT. Still another dampen the beneficial effects of estrogen. criticism has been the age of the participants; participants Since the discontinuation of the WHIMS trials, the case were age 65 or older, at least a decade past the average age for ERT in reducing the risk for AD and improving the cogni- of menopause. Together, these limitations have called into tive functioning of postmenopausal women has continued to question the validity of the WHIMS findings, suggesting that gain at least modest support through further RCTs. Indeed, the WHIMS may not be the best model for understanding results from several RCTs published in the past few years have the effect of HRT on Alzheimer’s disease. demonstrated support that E2 formulations are associated Another limitation in the generalizability of the WHIMS with a reduced amount of decline in verbal memory among involves the type of HRT that was used. Specifically, it has healthy postmenopausal women when compared to controls. been pointed out that CEE does not contain the hormone The benefits of these treatments have been observed in trials 17β-estradiol, [41] the estrogen compound that has been with durations ranging from three months to two years (see shown in basic science studies to be neuroprotective [42– Table 2)[18, 22–24]. In contrast, at least one study has found 44]. In addition, the greater rates of dementia seen among no benefit on verbal memory associated with E2 compared participants of the CEE + MPA study trial of the WHIMS to placebo [20]; however, it was noted that the women in suggest that simultaneous use of MPA may present additional that study used E2 for only two months. Thus, it is possible risk [45]. Indeed, consistent with WHIMS findings, a recent that the effects of E2 on verbal memory may be evident only randomized-controlled study by Maki et al. found that after three months or more. In a separate study, Joffeetal. women receiving CEE + MPA for four months demonstrated found that E2 was not associated with an improvement in mild declines in verbal memory compared to women verbal memory scores but rather decreased likelihood for receiving placebo [21]. Additionally, a recent comparison of errors during the memory tasks [19]. Specifically, women several different HRT types has provided some insight into on E2 demonstrated less perseverative errors during recall which treatment provides the most cognitive benefit. Using tasks compared to women on placebo. These women, as a functional neuroimaging as an outcome measure, Silverman group, were also less likely to demonstrate an interference et al. compared the cerebral metabolic activity associated effect when retaining previously learned information. Thus, with three hormone regimens over the course of one year: although E2 was not found to enhance verbal memory 17β-estradiol (E2), CEE, and CEE + progestin [24]. Results scores per se, the authors concluded that E2 enhanced verbal revealed that the E2 group performed significantly better information processing by decreasing the forgetfulness of a on verbal memory than the CEE group. This group also response already given [19]. demonstrated higher metabolism in the receptive language and auditory association areas. Additionally, the CEE + 2.3. Neuropathological and Neurophysiological Studies of HRT: progestin group demonstrated lower metabolism in areas Relevance to AD. While results of recent RCTs show modest associated with long-term memory storage (i.e., mesial and support for a beneficial effect, evidence from histopatholog- inferior lateral temporal regions) compared to the CEE ical and neurophysiological studies has provided stronger group. Taken together, these findings suggest that E2-based support for estrogen’s neuroprotective effects, particularly therapies may provide the most beneficial neuroprotective for the neurodegenerative disease process thought to under- effect. In addition, the Silverman et al. study suggests that lie AD [46–48]. Neuroimaging and autopsy results have 4 International Journal of Alzheimer’s Disease

Table 2: Randomized clinical trials of HRT and verbal memory.

Study Hormone Sample Outcome Age Overall findings (reference) treatment used size measure Followup study of women randomized 5, 10 and 15 years earlier to HRT or placebo during clinical trials. Logistic regression showed that for women who received HRT E2 2 mg + varied Bagger et al., Cognitive for 2-3 years, the relative risk for cognitive progestins versus 261 54.1 [18] screening task impairment was significant decreased by placebo for 2 years 64% compared to the never users. Long-term/current users of HT also demonstrated a decreased risk of 66% compared to the never users. Women on E2 had fewer perseverative E2 0.5 mg versus errors during verbal recall when Verbal memory; Joffeetal.[19] placebo for 12 52 40–60 placebo-treated women. Women on E2 also Functional MRI months showed greater retention of new information without interference. Women on estrogen therapy did not show Estradiol 2 mg 53.26 LeBlanc et al., higher cognitive performance on verbal versus placebo for 32 (treatment) Verbal memory [20] memory tasks compared to women on 2months 52.08 (placebo) placebo. (CEE) + medroxyprogesterone Modest negative effects on verbal memory 51.9 (treatment) Maki et al., [21] acetate (MPA) 158 Verbal memory (short- and long-term recall) were found in 52.4 (placebo) versus placebo for the HRT versus placebo group. 4months All women were administered the antimuscarinic drug scopaline (SCOP) or E2 2 mg versus placebo. E2 pretreatment significantly Dumas et al. 50-62 (younger) placebo for 3 22 Verbal memory decreased the anticholinergic drug-induced [22] 70–81 (older) months impairments on verbal memory task for the younger group only compared to the older group. Women on E2 who scored at or above Tierney et al. E2 1 mg versus 142 61–87 Verbal memory average showed less decline in delay verbal [23] placebo for 2 years memory compared to women on placebo. Women on E2 had significantly higher 17β-estradiol (E2) verbal memory than CEE and showed versus conjugated higher metabolism in Wernicke’s and Silverman et al. Verbal memory; equine estrogen 53 50–65 auditory association. E2 was also associated [24] FDG-PET (CEE) versus CEE with higher metabolism in mesial and +Pfor1year inferior lateral temporal regions and inferior frontal cortex compared to PE.

indicated that β-amyloid and tau proteins are involved in the regimens (CEE or CEE + MPA) [24]. However, observational structural changes that lead to AD pathology, particularly in studies and RCTs have also demonstrated support for varied the hippocampus and other medial temporal regions, as well ERT regimens. Compared to nonusers, long-term ERT (E2 as the parietal and frontal cortical regions [49]. Evidence has or CEE for an average of 15 years) has been associated with shown that estrogen (particularly E2) provides protection increased cerebral blood flow to the hippocampus and left against β-amyloid-induced damage and tau-related changes superior temporal gyrus at a two-year followup [51]. Further, [50]. Observational and RCT studies that also utilized compared to placebo, a four-month trial of ethinylestradiol neuroimaging outcomes have also been supportive of the and progestin was associated with increased activation in benefits of 17β-estradiol, particularly in the brain regions brain regions associated with the left middle/superior frontal that show preclinical abnormalities in individuals who are at cortex, and left inferior parietal cortex during verbal memory risk for AD. For instance, as mentioned earlier, E2 has been encoding tasks on functional magnetic resonance imaging associated with higher metabolism in language processing [52]. Finally, in another study, long-term users of ERT (E2 and auditory association areas compared to other HRT or CEE for an average of 18 years) demonstrated higher International Journal of Alzheimer’s Disease 5 density of muscarinic receptors in the hippocampus and in most observational studies, of an inverse relationship prefrontal cortex than individuals who had never used ERT, between length of HRT treatment and risk for AD. suggesting that one of the neuroprotective effects of E2 or Other investigators have hypothesized that there may other ERT regimens could also include the maintenance of be a “critical period” for postmenopausal women during the cholinergic system in the hippocampus and frontal cortex which 17β-estradiol selectively provides a beneficial effect for [48]. younger as opposed to older women with an intact uterus A recently proposed explanation may explain the incon- [41, 55]. This hypothesis has also received support from at sistent results of the aforementioned observational studies least one RCT. For example, LeBlanc et al. randomized 22 and RCTs. Known as the “healthy-cell bias” [53], the postmenopausal women to receive either E2 or placebo for hypothesis is that E2 may selectively benefit healthy neurons. 3 months [20]. At the end of the trial, the antimuscarinic In the context of human studies, based on the findings from drug scopolamine (SCOP) was administered before a verbal task to initiate anticholinergic-induced memory impair- observational studies and RCTs, this hypothesis predicts ment. Results showed that E2 pretreatment significantly that E2 can be protective if initiated before or during decreased the anticholinergic-induced impairment on the times of neuronal stress, but harmful if given after the verbal memory task for the younger group (age 50–62); cells have progressed toward degeneration. In their study, however, the benefit of E2 was not observed in the older Chen et al. administered E2 to rat hippocampal neurons group (age 70–81). Interestingly, the beneficial effects of exposed to β-amyloid, using varied doses and dose schedules E2 were only observed during the anticholinergic challenge (acute versus continuous versus intermittent). Data indi- with SCOP and not during the placebo challenge. LeBlanc cated that neurodegeneration was prevented when E2 was et al. concluded that younger women benefit from E2 more administered before or during β-amyloid exposure, and a than older women, and that the benefits of E2 in younger continuous dose was found to demonstrate the strongest women may be observed only when the cholinergic system effects. In contrast, exposure to higher doses of E2 actually is temporarily disrupted. Consistent with this finding is worsened neuronal death when β-amyloid was present. that younger women have a higher density of muscarinic Additionally, E2 administered after β-amyloid exposure receptors than older women, and thus may be more sensitive exacerbated neuronal death. It was concluded that the best to cholinergic changes [48]. Thus, it is plausible that the E2 dosing was pretreatment and continuous exposure to women in the aforementioned WHIMS may have been past prevent degeneration. Consistent with the “healthy-cell bias” the “critical period” for the beneficial effects of E2. hypothesis, Dumas et al. demonstrated a selective benefit of 17β-estradiol toward cognitively intact women [22]. A group of 142 postmenopausal women (age range: 61–87) 2.4. Summary—AD. Taken together, the findings from stud- were randomized to receive E2 (n = 70) or placebo (n = ies employing a variety of methods demonstrate that some 72) for two years. Verbal memory was assessed at baseline forms of ERT are neuroprotective, resulting in preserva- and at 1-year and 2-year followup. Results revealed that tion of cognitive abilities and reduced risk of AD. While women who received E2 and who performed at or above some studies have affirmed that young and healthy post- average on verbal recall at baseline demonstrated higher menopausal women may benefit the most from estrogen scores at the 1-year and 2-year followup compared to the exposure, other studies have suggested that older and healthy placebo group. In contrast, women who received E2 and women with intact verbal memory can also benefit from performed below average on verbal recall at baseline showed estrogen. The consistent findings from the observational no significant difference compared to the placebo group. studies reviewed above seem to be that ERT (most commonly Dumas et al. concluded that these findings provided support CEE), with a minimal duration of at least one year, is benefi- to the healthy cell bias hypothesis, as they considered it cial in reducing risk for AD among healthy postmenopausal improbable that women with a normal score or better women. Although benefits have been observed among varied had significant neurodegenerative changes [22]. Notably, the regimens (CEE, CEE + P, E2) [48, 51, 52]; the most women who benefitted from estrogen exposure were age beneficial estrogen formulation seems to be E2 unopposed 70 (average) and approximately 20 years postmenopause, by progestin [24, 50]. Randomized clinical trials in healthy, suggesting that older women who have intact verbal memory postmenopausal women have suggested that E2 has been can benefit from a new regimen of ERT late in life, as most beneficial in reducing cognitive decline, particularly long as they have not demonstrated memory impairment. verbal memory, which is the predominant symptom of Interestingly, basic science research has supported the biased early AD [18, 22–24]. Additionally, both observational and neuroprotective effect of E2 toward healthy individuals; in RCT studies utilizing neuroimaging outcomes have been fact, the presence of apoplipoprotein E4 (APOE4) genotype supportive of the benefits of E2, particularly in the brain has been found to reduce the neuroprotective role of E2 regions that show preclinical abnormalities in individuals in an animal model [54]. Thus, an alternative explanation whoareatriskforAD[21, 24, 51, 52]. for the findings of Dumas et al. could be that the women who demonstrated lower than average recall at baseline may 3. Parkinson’s Disease have had the APOE4 genotype; in turn, they may have not experienced the neuroprotective effects of ERT. The healthy Parkinson’s disease (PD) is the second most common cell bias hypothesis also helps explain the finding, reported neurodegenerative disorder after AD, with an estimated 6 International Journal of Alzheimer’s Disease prevalence of 0.3% in the general population. Risk increases authors emphasized a nonsignificant trend of improvement with age, with a prevalence of 1% in those over 60, and on the total and motor scores of the Unified PD Rating Scale. 4% in those 80 years and older [56]. Many, but not all, It is conceivable that the null findings were due to the small studies have reported higher risk for PD and younger age sample size; however, the existing literature on ERT and PD of onset in males [57–66]. This observation, along with symptoms remains equivocal at this time. the fact that the neuropathological process underlying PD commonly begins before menopause, suggests that estrogen 3.2. HRT and Observational Studies of PD Risk. Epidemi- may play a modulatory role. In addition, estrogen has a ff direct modulatory affects on dopaminergic functioning [67]. ological studies of the protective e ects of HRT against Together, these observations suggest a potential protective PD have been mixed as well (Table 3). The relationship effect of estrogen against PD, or ameliorative impact on between lifetime reproductive events and PD was examined symptoms. by Martignoni et al. Comparing a large sample of women diagnosed with PD to healthy controls, they found that 3.1. Estrogen and PD Symptoms. A variety of studies have the duration of reproductive life was similar between the addressed the impact of estrogen on PD. Perhaps the most two groups [28]. Time and mode of menopause onset indirect are observational studies of PD symptoms during were also similar between the groups; however, women the menstrual cycle. Early studies in the 1980s reported with PD reported less access to HRT. In addition, the PD thatsomefemalepatientswithPDhadfluctuationsin group overall reported more premenstrual symptoms, fewer motor symptoms that paralleled presumed fluctuations in deliveries and abortions, and less use of contraception, endogenous estrogen levels [68, 69], with presumably lower indicating a relationship between PD and reproductive levels of estrogen associated with greater motor symptoms. events. Benedetti et al. reported a case-control study in which However, more recent studies have shown mixed results. women with PD had an earlier reported age of menopause, Kompoliti et al. did not find significant correlation between a higher frequency of hysterectomies, and lower occurrence endogenous hormone levels and motor examination in of HRT [27]. Further, Currie et al. found that ERT in the “off” state (a state of decreased mobility as a result postmenopausal women was associated with a significantly of nonresponsiveness to medication) among female PD reduced risk of developing PD [29], and Ragonese et al. patients examined at various times during their menstrual found that factors reducing estrogen stimulation during life cycle [70]. were associated with development of PD [30]. Specifically, A small number of prospective studies of ERT and PD was significantly associated with shorter fertile life PD have also been reported, with mixed results (Table 4). lengths (<36 years) and a longer cumulative length of Strijks et al. did not find a significant dopaminergic effect pregnancies (>30 months). This group later reported a in their 8-week placebo-controlled, randomized, double- significant correlation between age of PD onset and both blind trial pilot study of E2 in 12 postmenopausal female age at menopause and fertile life duration [32]. Despite patients under the age of 80 [35]. However, an 8-week these findings, others have found contrary results. Popat et double-blind, parallel-group, prospective study using Pre- al. found that the association of postmenopausal HRT and marin (CEE) versus placebo in PD patients with motor PD risk depended on the type of menopause [31]. Among fluctuations showed a statistically significant improvement women with history of hysterectomy (with or without an in “off” times (i.e., when dopamine agonist medications have oopherectomy), ERT use was associated with a 2.6-fold diminished efficacy) among the estrogen treated group [36]. increased risk for PD, and a trend for additional risk was Further, another double-blind, placebo-controlled crossover noted for increasing duration of estrogen use. Conversely, study of high-dose transdermal E2 in 8 postmenopausal among women with natural menopause, no increased risk women with mild-to-moderate PD demonstrated a slight of PD was observed with HRT (ERT alone or in conjunction anti-Parkinsonian effect without significantly worsening with progestin). Contrary to the findings of Benedetti et al., dyskinesias [34]. earlier age of menopause was associated with reduced risk Although the overall symptomatic effect of ERT on PD of PD. Further, Simon et al. recently reported results of a remains unclear, these early studies raised the possibility 22-year prospective study of 244 participants in the Nurses’ that some forms of estrogen may mitigate the symptoms of Health Study who developed PD [33]. Among their sample, PD. Despite this early optimism, a more recent multicenter, risk of PD was not significantly associated with reproductive randomized, double-blind, placebo-controlled, pilot trial of factors or HRT use. However, they did find that use of HRT CEE in postmenopausal women with PD experiencing motor may modify the associations of smoking and caffeine with fluctuations did not find any benefit of ERT in ameliorating PD risk; specifically, the inverse relationship between caffeine symptoms [37]. In that study, 23 women received either use and risk of PD was observed only in non-HRT users. 0.625 mg/day of CEE or matching placebo for 8 weeks. Further, whereas the researchers also reported an inverse None of the outcome measures, including changes from relationship between pack-years of smoking and risk of PD baseline to study completion in Unified PD Rating Scale for both HRT users and nonusers, risk was reduced more scores, “on” time (i.e., duration that dopamine agonist in the latter group. As such, HRT use appeared to attenuate medication is effective), dyskinesia ratings, and results the observed beneficial effects of caffeine use and tobacco from neuropsychological testing, were significantly different smoking. Of note, this study did not separately analyze the between the placebo and treatment groups, although the data based on type of HRT. International Journal of Alzheimer’s Disease 7

Table 3: Case-control and epidemiological studies of HRT and Parkinson’s disease.

Study Sample description Overall findings (reference) 87 women with Parkinson’s disease ERT reduced risk of dementia among the PD-only sample (OR = without dementia (PDND), 80 women Marder et al. 0.22, 95% CI: 0.05–1.0), and also when PDD patients were with Parkinson’s disease with dementia [25] compared to healthy controls (OR = 0.24, 95% CI: 0.07–0.78). (PDD), and 989 nondemented healthy ERT did not affect the risk of PD. women. Data from 10,145 elderly women with PD available via the Systematic Assessment in Independent of age, estrogen users had better cognitive Fernandez and Geriatric drug use via Epidemiology functioning and were more independent with regards to activities Lapane [26] (SAGE) database. Included 195 women of daily living. More estrogen users were depressed and likely to be with PD who received estrogen and 9950 taking antidepressant medications. who did not receive estrogen. The PD group had undergone hysterectomy (with or without unilateral oophorectomy) more than the control group (OR = 3.36; 95% CI: 1.05–10.77). The PD group had more frequent Benedetti et al. 72 women with PD and 72 healthy occurrence of early menopause (< or = 46 years) (OR = 2.18; 95% [27] women. CI: 0.88–5.39). The PD group used ERT for at least 6 months after menopause less frequently than the control group (14%; OR = 0.47; 95% CI = 0.12–1.85). The PD group did not have earlier menopause than the control group. Duration of reproductive life was similar between women with PD and those without PD. Women with PD reported less access to Martignoni et 150 women with idiopathic PD and 300 HRT. The PD group also reported more premenstrual symptoms, al. [28] healthy women, all postmenopausal. fewer deliveries and abortions, and less use of contraception, indicating a relationship between PD and reproductive events 50% of women in the control group took ERT, as compared to 25% of women in the PD group. Women who had taken 68 women with PD and 72 healthy postmenopausal ERT were less likely to develop PD than those Currie et al. [29] women, all postmenopausal. who had not (odds ratio, 0.40; 95% CI: 0.19–0.84). Among women with PD, postmenopausal ERT was not associated with age of onset. PD was significantly associated with a fertile life length of less than 36 years (OR 2.07; 95% CI: 1.00 to 4.30). PD was also associated Ragonese et al. 131 women with idiopathic PD and 131 with a cumulative pregnancy length of longer than 30 months (OR [30] healthy women. 2.19; 95% CI: 1.22 to 3.91). There was an inverse association between PD and surgical menopause (OR 0.30; 95% CI: 0.13 to 0.77). Among women with history of hysterectomy (with or without an oopherectomy), ERT use was associated with a 2.6-fold increased risk for PD, and a trend for additional risk was noted for increasing 178 women with PD and 189 healthy Popat et al. [31] duration of estrogen use. Among women with natural menopause, women. no increased risk of PD was observed with HRT (ERT alone or in conjunction with progestin). Earlier age of menopause was associated with reduced risk of PD. A significant correlation was found between age at PD onset and Ragonese et al. 145 women with PD. age at menopause, and also between age at PD onset and fertile life [32] duration. Women who underwent either unilateral or bilateral 1,252 women with unilateral and 1,075 Rocca et al. oophorectomy had an increased risk of parkinsonism compared to women with bilateral ophorectomy, and [16, 17] referent women (HR 1.68; 95% CI: 1.06–2.67). This risk increased 2,368 referent women. with younger age at oophorectomy. Risk of PD was not significantly associated with reproductive 22-year prospective study of 244 women factors or HRT. The association of smoking and caffeine with PD Simon et al. [33] with PD enrolled in the Nurses’ Health risk was modified by HRT, however. Based on a very small sample Study. (4), women using progestin only hormones had increased risk for PD. 8 International Journal of Alzheimer’s Disease

Table 4: RCTs of ERT and Parkinson’s disease. Study Sample Hormone treatment used Outcome measure Overall findings (reference) Size All but one participant had High-dose transdermal E2. levodopa-induced dyskinesia at start of Cross-over design with 2 study. After 10 days of E2 treatment a Therapeutic threshold for Blanchet [34] weeks on E2, 2 week 8 significant reduction was observed in the levodopa. washout, and 2 weeks on anti-parkinsonian threshold dose of placebo intravenous levodopa without significantly worsening dyskinesias Motor score from the Unified 17β-estradiol (E2) versus Parkinson’s Disease Rating Scale No differences in outcome measures Strijks et al. [35] 12 placebo for 8 weeks (UPDRS); patient report of between E2 and placebo. subjective changes. UPDRS, timed tapping score, CEEversusplacebofor8 “On” and “off” times, and motor score on Tsang et al. [36] 40 Hamilton Depression Scale, weeks the UPDRS improved with estrogen. patient self-report. Primary outcome was ability to The Parkinson complete the trial. Other Study Group The estrogen group showed a trend for CEEversusPlacebofor8 outcome measures included Poetry I 23 improvement on the total and motor weeks adverse events, UPDRS, “on” Investigators UPDRS scores. time, dyskinesia ratings, and [37] neuropsychological functioning

In one of the largest observational studies to date, Rocca andcolleaguesinvestigatedriskofPDbothwithandwithout et al. examined 1,252 women with unilateral oophorec- dementia among a sample of 1156 women. They reported tomy, 1,075 women with bilateral oophorectomy, and 2,368 that ERT protected against development of PD-associated controls for development of PD. Data for the partici- dementia, but not against PD itself [25]. Similarly, Fernandez pants were collected until death or the termination of and Lapane found that estrogen use was associated with the study using direct or proxy interviews, neurologic better cognitive functioning and greater independence in examinations, medical records, and/or death certificates. activities of daily living among a large sample of elderly The authors found that women who underwent either women living in nursing homes [26]. They also noted that unilateral or bilateral oopherectomy before the onset of estrogen users were more depressed and likely to be on an natural menopause, thereby decreasing endogenous estrogen antidepressant as compared to nonusers. One-year death levels, had an increased risk of parkinsonism compared with rates were comparable between estrogen users and nonusers. referent women. Further, risk increased with younger age at oophorectomy. The findings were similar regardless of 3.4. Mechanisms of Estrogen Action in PD. While epidemi- unilateral or bilateral oopherectomy. Importantly, while the ologic, observational, and experimental studies of ERT and authors reported a trend, the surgical menopause group was PD have produced equivocal results, the biological mecha- not at increased risk for PD. nisms for a beneficial effect of estrogen upon dopaminergic Although these studies might appear to provide conflict- functioning are less so. There are two general mechanisms ing results, complex factors are at play. The indication for of action through which estrogen might influence PD: HRT (posthysterectomy, posthysterectomy + oopherectomy, symptomatic and neuroprotective. Estrogen receptors have natural menopause), the specific type of HRT (CEE, E2, been located in the nuclei of nigral dopaminergic (DA) estrogen/progestin combinations), and other variables may neurons, including estrogen receptor alpha (ERα)andbeta combine in ways yet unknown to increase or decrease PD (ERβ)[73, 74], suggesting that estrogen might therefore risk. Clearly, further study is necessary. directly influence DA functioning. ERα has also been found in midbrain glial cells [75], and ERβ in striatal medium spiny 3.3. Studies of HRT and Dementia due to PD. PD is also asso- neurons [74]. Novel surface membrane estrogen receptors ciated with cognitive decline, with anywhere between 24– have also been described [76, 77]. Perhaps related to these, 31% becoming demented [71]. PD dementia is considered a administration of exogenous conjugated estrogens results in subcortical dementia, with associated deficits ranging from an increase in binding of the DA transporter ligand TRODAT simple motor ability to higher-order cognitive functions in otherwise healthy postmenopausal women [78]. It has also [72]. Despite the high incidence of neurocognitive dysfunc- been shown that, in the absence of nigral neuroprotection, tion in PD, the relationship between HRT and dementia in central E2 synthesis limits striatal DA loss caused by 6- those with Parkinson’s disease has received considerably less OHDA in male rodents, implicating a modulatory effect attention. Only two case-control studies were found. Marder on DA function [79]. These studies provide evidence that International Journal of Alzheimer’s Disease 9 estrogens may upregulate the nigrostriatal pathway, either and animal models of PD, results of clinical and epidemio- pre- or postsynaptically, by an effect on nuclear or surface logical studies are inconclusive at this time. Recent findings membrane estrogen receptors. of estradiol’s modulation of alpha-synuclein indicate a Estrogen’s neuroprotective actions have been well specific mechanism through which the hormone may reduce established. In PD, there are animal models that are risk for PD and/or mitigate symptoms. Longer clinical trials exquisitely specific for nigral cell death, of which the 6-hy- with specific estrogen compounds (i.e., 17β-estradiol), as droxydopamine (6-OHDA) and MPTP/MPP+ models are well as biological markers of disease progress (e.g., neu- perhaps the best known [80, 81]. There is ample evidence roimaging), will be more likely to definitively determine if that both endogenous and exogenous estrogen ameliorate ERT is protective against PD or if it can mitigate the disease. DA depletion in the MPTP/MPP+ model [75, 82–91]. There With specific regards to PD-associated dementia, only two is similar evidence that estrogen is neuroprotective in the case-control studies were located, both suggesting that ERT 6-OHDA animal model [79, 92–96], a methamphetamine reduces risk of cognitive impairment in women with PD. model [97–100], and a wide range of other relevant animal models [101–103]. The exact mechanisms of neuroprotection,however,arenotclear.Studieshaveshownarolefor 4. HIV-Associated Neurocognitive binding of estrogen to the nuclear estrogen receptor [104], Disorder (HAND) the ERα subtype, [105]ERα with a glial contribution, [75] ERα +ERβ [106], and ER-independent mechanisms [88]. Internationally, an estimated 33 million individuals have This has implications for potential therapeutic agents, as HIV/AIDS, [110] and in many areas women comprise the some estrogen analogues lack activity at one or both nuclear majority of those infected [111]. Aggressive intervention receptors; while others, such as the “inactive” enantiomer E2, with a regimen of multiple antiretroviral drugs (combined may have no ER binding activity at all. E2 has been shown in antiretroviral therapy, or cART) has successfully increased the MPTP model to have neuroprotective properties [101], lifespan and attenuated some of the most dire neurological and has been investigated as a possible neuroprotective agent effects of HIV infection. However, cART cannot eradicate [107]. HIV, and it has attenuated, not eliminated, the most com- It is important, however, to recognize the imperfect mon neurological complication of HIV, or HIV-associated nature of these preclinical models. First, while PD is a neurocognitive disorder (HAND) [112]. In this section, we chronic, slowly progressive disorder, the aforementioned discuss what is known about estrogen and HAND from animal models use agents that cause acute toxicity. Second, observational studies in humans, studies in animal models, despite the wide use of these models over the past two and in vitro studies. No relevant human clinical trials of decades and the demonstration in preclinical models that estrogen for HAND have been published. many agents are neuroprotective against 6-OHDA, MPTP, or HAND is a constellation of cognitive impairments caused both, none of these agents have proven neuroprotective in by HIV infection [112]. Because of the lack of diagnostic human subjects with PD. There may be a simple explanation biomarkers, HAND remains largely a clinical diagnosis, for this. We now know that neurodegeneration in most cases made when an HIV+ individual experiences neurocognitive of familial PD is due to impaired ubiquitin-proteosomal decline, sometimes with concomitant deficits in day-to-day function and alpha-synuclein protein aggregation [108]. functioning, and only after other conditions that might cause Although the relationship between these abnormalities and this decline have been ruled out. The severity of HAND those replicated by the 6-OHDA and MPTP models are ranges between mild neurocognitive impairment with no complex, it appears likely that any agent that will be impact on day-to-day functioning to a debilitating HIV- neuroprotective in humans with idiopathic PD will need associated dementia [112]. While the incidence of new cases to act to reduce alpha-synuclein aggregation. This can of HAD has declined dramatically [113, 114], the prevalence occur either by reducing its synthesis, reducing protein of milder forms of HAND has actually increased along aggregation, enhancing its elimination, or reducing the with the longevity of the cART-treated HIV+ population toxic effects of excessive alpha-synuclein. Only recently [113]. This phenomenon has been variously ascribed to has evidence been found that estrogen has the ability to several explanations, including the presence of irreparable act on alpha-synuclein in a beneficial manner. Hirohata CNS damage pre-cART [115], the failure of many cART et al. found a variety of sex hormones, including estriol, regimens to adequately penetrate and treat the CNS [116], estradiol, estrone, androstenedione, and testosterone to exert persistent low levels of HIV despite treatment [117], and to significant antiaggregation and fibril-destabilizing effects on persistent CNS inflammation [118], among others. The latter alpha-synuclein in vitro. Estradiol was especially effective is particularly relevant to the putative therapeutic benefit of [109]. Further, Marwarha et al. showed that activation of estrogen, as it appears that cART does not always reduce and ERβ, in conjunction with inhibition of LXRβ,mayreduce in some cases may increase, the CNS inflammation [119] that progression of PD by slowing α-synuclein accumulation. is associated with HAND [120]. Estrogen has significant anti- inflammatory and neuroprotective properties [121–123]and can potentially counteract inflammation in the HIV+ brain, 3.5. Summary—PD. While in vitro and non-human in vivo as discussed in more detail below. experiments have consistently demonstrated evidence for There are several other important reasons for investi- estradiol’s neuroprotective activity in dopaminergic neurons gating the use of estrogen as an adjunctive treatment in 10 International Journal of Alzheimer’s Disease

HIV and HAND. First, estrogen and other gonadal steroids the levels of neurotrophins, decreasing apoptotic factors, and have significant effects on the course and presentation of antioxidant properties [138]. Zemlyak et al. reported two HIV disease itself. For example, women are at increased risk different beneficial effects of estrogen in the amelioration for acquiring HIV compared to men, and this vulnerability of gp120-induced toxicity: a major effect of attenuating the may be affected by gonadal hormones [124]. Further, in neurotoxicity of factors released by gp120-treated microglial a macaque model of HIV infection, progestogen-based cultures, and a minor effect of enhancing the ability hormonal contraceptives increased the risk of acquiring of neuronal cultures to survive exposure to neurotoxic simian immunodeficiency virus (SIV), increased disease factors [122]. Another neurotoxic HIV protein, tat, the progression, and increased genital shedding of SIV; whereas nuclear trans-activating protein, is essential in promoting treatment with estrogen lowered risk of acquiring SIV [125]. the transcription and replication of HIV. tat can act both Results of natural history studies suggest a gender role in directly to harm neurons [139], and indirectly by stimulating disease progression, possibly due to hormonal differences. macrophages, microglia, and astrocytes to synthesize harm- For example, women have lower HIV RNA viral loads at ful substances such as proinflammatory cytokines [140], and seroconversion compared to men [126], and when adjusted by increasing free radicals and oxidative stress [141]. In cell for CD4+ count, women have lower viral loads throughout culture, 17β-estradiol suppressed tat-activated transcription the course of their infection [127]. While one study found of HIV in astrocytes [142]. 17β-estradiol also attenuated a lower risk of clinical progression to AIDS among HIV+ the tat-induced release of pro-inflammatory mediators in women versus HIV+ men treated with cART [128], others endothelial cells [143], prevented oxidative stress and cell have found no differences in clinical outcome by gender death associated with combined gp120 and tat neurotoxicity [129]. A possible explanation for such gender disparity, in vitro [144], and prevented gp120/tat-induced loss of should it turn out to be valid, is estrogen, which decreases dopamine transporter function [144]. HIV replication in peripheral blood mononuclear cells [130]. These observations have led to the proposal that serum However, all such studies must be interpreted with caution estradiol levels be maintained in HIV+ women as a possible because of the reported gender differences between HIV+ neuroprotective agent against HAND [145]. Despite this, men and women in socioeconomic status, risk behavior, there is little clinical information about estrogen and HAND substance abuse, and access to care [131], which also affect in HIV+ women. A single retrospective study from the pre- progression to AIDS [132, 133]. With regards to HAND, cART era, of 84 older (age 40+ years) HIV+ women, reported whether women develop HAND at the same rate as men that hormone replacement therapy (HRT) was associated or if there are different clinical manifestations of HAND in with a significantly decreased risk of mortality [146]. Of men and women remains a controversial topic. In part, this interest, there were six women in the cohort who were is because so few studies had sufficient numbers of females to diagnosed with HIV-associated dementia, none of whom evaluate. A sub-study of the Women’s Interagency HIV Study reported taking HRT. This study has been interpreted by is beginning to address this problem [134]. some to indicate a neuroprotective effect of HRT; however, There is neurobiological reason to expect a reduction this was not a prospective study that examined cognition in of HIV-related neuropathological changes with ERT. Firstly, an organized or standardized fashion. However, based on this microglia are the resident immune cells of the CNS, and last report and on the neuroprotective role of estrogen in these cells play an important role in driving inflammation other inflammatory and degenerative conditions, the role of in many neurodegenerative diseases, thus representing an estrogen and other hormones in HAND has become an area important target for therapy [135]. In HIV infection, of growing interest among basic scientists. microglia can be infected and/or activated; they are major No studies of HAND or neurocognitive functioning sources of complete HIV virions, individual neurotoxic viral in HIV+ persons have considered hormonal status or use proteins, proinflammatory substances, and other potential of exogenous hormones. The preponderance of evidence mechanisms that drive neurotoxicity, neuroinflammation, to date indicates that HIV+ men and women develop oxidative stress, and neurodegeneration. Microglia express neurocognitive impairment at a similar rate, when issues endogenous estrogen receptors [136], and treatment with such as access to care, education, and substance abuse history estrogen is anti-inflammatory provided it is administered are similar. While some have reported a higher occurrence of early in the course of an insult [121, 123]. Secondly, estro- HIV-associated dementia among women [147], others have gen’s anti-inflammatory effects may directly counteract the not found this [148, 149]. More recently, Martin et al. studied neuroinflammation caused by HIV proteins. HIV-infected a large well-matched group of adult male and females, strati- cells can generate both replication-competent virions and fied by HIV status, all with a history of substance dependence excess viral proteins, which are shed or secreted into [150]. Participants were abstinent at the time of testing. the extracellular space. The HIV coat protein, gp120,is Whereas the performance of HIV+ men did not differ from the binding protein for viral entry [137] and acts as an HIV-negative counterparts of measures of motor skill and indirect neurotoxin via its effects on microglia, macrophages, probabilistic learning, the HIV+ women performed worse and astrocytes, initiating a cascade of events that damage than their seronegative counterparts, suggesting that women neurons. Estrogen has been reported to have a broad anti- mightbemorevulnerabletotheeffects of HIV. However, due inflammatory effect on microglia [121]. Estrogen reduces to the absence of a nonsubstance-dependent control group, the neuroinflammatory responses to gp120 and exerts neu- they could not exclude the possibility that the observed roprotective effects on gp120-exposed neurons, by raising differences were due to gender-related differences in the International Journal of Alzheimer’s Disease 11 cognitive effects of addiction. Another study reported no and E2 is a compelling reason to further study the influence gender difference in rate of neurocognitive decline over time of ERT on risk for FTD. [151]; and still another found that while rates of impairment were similar between men and women, there were some 6. Summary and Conclusions differences in the neurocognitive profiles [148]. Whether this is related to estrogen or other gonadal hormones remains to In summarizing the evidence discussed above, HRT, in be determined. particular ERT, appears to play an efficacious role in treating and preventing several neurodegenerative conditions. Figure 1 depicts putative neurobiological and neurobehaivo- 4.1. Summary—HIV/HAND. HAND shares many features ral sequelae resulting from 17β-estradiol use, based on with other neurodegenerative diseases, including microglial studies reviewed in this paper. The case for a neuroprotective activation and neuroinflammation. Preliminary studies in role of HRT and AD is supported by research from epidemi- animal and in vitro models indicate that, like many other ological and RCT studies, which have shown that estrogen, neurodegenerative diseases, the effect of HIV on the brain specifically E2 (17β-estradiol), can reduce the risk for AD may be blunted by treatment with 17β-estradiol, and possi- and minimize cognitive decline in otherwise healthy women, bly other gonadotrophic hormones. This would have to be particularly verbal memory. Based on basic science research, balanced against the risks of adding estrogen to the regimens the mechanisms for this neuroprotection may involve E2’s of HIV+ patients, both male and female. However, there is a protection against β-amyloid-induced degeneration and may pressing need to determine if HRT may benefit patients with even include the maintenance of the cholinergic system in AIDS who remain at risk for HAND even when treated with the hippocampus and frontal cortex. In addition, at least one HAART. study has demonstrated that the presence of progestins in combination therapies may actually dampen the beneficial 5. Frontotemporal Dementia effects of estrogen [24]. Similarly, in vitro and non-human in vivo experiments Frontotemporal dementia, or FTD, is the most common have demonstrated E2’s neuroprotective effects in dopamin- form of a group of related neurodegenerative diseases that ergic neurons and animal models of PD. In addition, E2’s primarily affect the frontal and/or temporal lobes. The modulation of alpha-synuclein indicates a specific mecha- others include semantic dementia and progressive nonfluent nism through which the hormone may reduce risk for PD aphasia. Collectively, these have been called frontotemporal and/or mitigate symptoms. To date, results of clinical and lobar degenerative diseases [152], and they are believed epidemiological studies of ERT alleviating motor symptoms to account for an estimated 20% of dementia cases with in PD patients have been mixed and warrant further presenile onset [153]. investigation. The effects of HRT on the neurocognitive Only one study to date has addressed the relationship symptoms of PD have received little attention, with the two between HRT and FTD. Levine and Hewett reviewed the case-control studies to date indicating that ERT reduces risk medical files of all women seen at an Alzheimer’s disease of cognitive impairment in women with PD. center (ADC) in Central California and found that 70% of Preliminary studies in animal and in vitro mod- women diagnosed with FTD had been taking HRT (exact els indicate that treatment with E2, and possibly other regimen unspecified) when evaluated, as compared to an gonadotrophic hormones, may reduce the effect of HIV on estimated 24% of the surrounding population [154]. While the brain. To date, much research on the neuroprotec- one easy interpretation would be that women exhibiting tive effects for HIV neurodegenerative changes has been cognitive impairment would have been more likely to be conducted on animal models and has yet to extend to placed on HRT before coming to the ADC, only 20% humans. Nonetheless, preliminary research has suggested of women diagnosed with AD at the same center had that development of HAND may be alleviated by HRT been taking HRT, so it is therefore unlikely that HRT was pretreatment. Conversely, and contrary to the findings from administered as a result of preclinical cognitive problems. other neurodegenerative diseases, there is some evidence that The women diagnosed with FTD were also similar in age to E2 may actually augment risk for FTD via its action on tau. women entering the center with AD (average age of symptom Additional research is needed to further delineate the onset was 65, average age of initial evaluation was 70). While molecular mechanisms through which E2 and other estro- poor diagnostic accuracy and estrogen’s beneficial effects on gens act to delay or prevent neuropathological progres- mood were cited as possible reasons for the findings, a more sion, or possibly cause progression in the case of FTD. compelling reason offered was a marked upregulation of Large-scale observational studies that accurately document tau in response to E2 administration, as evidenced in vitro HRT regimen and control for factors such as depression, [155]. The neuropathological correlates of many FTD cases education, and medical comorbidities (e.g., vascular risk appear to be tau-related, and in some cases directly linked factors) will also help to elucidate the role of ERT in to mutations in the tau gene [156]. In such cases, E2 may the neurodegenerative disease etiology. While observational increase risk of FTD by increasing production of mutated studies and RCTs examining ERT and AD have demonstrated forms of tau. However, while the role of tau in FTD has been long-term beneficial effects of varied ERT regimens (E2 or well established, it is now known that it does not account CEE), future studies may include long-term followup (5– for all forms of FTD [157]. Still the relationship between tau 10 years) of E2-based therapies alone on cognitive measures 12 International Journal of Alzheimer’s Disease

Lowered risk of Reduction of Alzheimer’s disease parkinsonian symptoms and, in healthy post and possible reduced menopausal women, benefit to cognitive risk of Parkinson’s functions mediated by disease and Parkinson’s hippocampus and dementia. frontal lobes Higher cerebral blood Augmentation of nigro- flow, maintenance of striatal dopamine cholinergic system, functioning. Slowed or and protection reduced alpha- against β-amyloid synuclein aggregation induced damage 17β- estradiol Anti-inflammatory effect on microglia and neurons, Increased production increase of neurotrophins, of τ protein decrease of apoptotic factors, suppresses HIV Increased risk of transcription in astrocytes frontotemporal Lowered risk of Alzheimer’s dementia. This may be disease. In healthy post specific to those with τ menopausal women, gene mutations and benefit to cognitive τ or pathology. functions mediated by hippocampus and frontal lobes

Figure 1: Putative mechanisms through which 17β-estradiol exerts neuroprotective and neuro-adverse eﬀects. In the context of Alzheimer’s disease, Parkinson’s disease, and HIV, 17β-estradiol appears to be neuroprotective. However, frontotemporal dementia is often the result of mutated tau protein and/or tau-related pathology. Because 17β-estradiol increases production of tau, it may accelerate risk for some forms of frontotemporal dementia.

and neuroimaging outcomes, as such would provide helpful information is insufficient at this time to support a clear link information on the duration of the benefits of E2 following between HRT and increased risk for breast cancer, at least discontinuation. one study from the WHI has reported an increased risk of Notably, possible medical risks should be considered breast cancer among users of estrogen plus progestin with in study of HRT and neurocognitive functioning [45]. For new onset of breast tenderness. This is an issue that requires instance, breast cancer is often a substantial concern that continued investigation. is linked with HRT. In fact, it is claimed that combined In clinical settings, the financial cost will need to be HRT with estrogen plus progestin is a cause for breast considered when recommending E2-based therapies for pre- cancer. However, while followup analysis approximately vention of AD or other neurodegenerative diseases. Patients three years after termination of the WHI study demonstrated and their physicians will have to determine whether the an increased risk for “all-cause cancer” for participants in potential cognitive benefit associated with E2 will outweigh the CEE + MPA trial compared to the placebo group [158], the financial cost, as well as the above-mentioned medical the risk for breast cancer and other types of cancer did not risks. differ between groups. Similarly, recent retrospective analyses of the WHI data found insufficient evidence that estrogen References plus progestin increased risk of breast cancer [159]. Another study using the WHI data found that among women in the [1]B.L.Plassman,K.M.Langa,G.G.Fisheretal.,“Prevalence CEE + MPA trial, increased breast cancer risk was especially of dementia in the United States: the aging, demographics, pronounced among women with breast tenderness [160]. In and memory study,” Neuroepidemiology,vol.29,no.1-2,pp. fact, new onset of breast tenderness after HRT initiation was 125–132, 2007. [2] R. Brookmeyer, D. A. Evans, L. Hebert et al., “National associated with increased breast cancer risk among women estimates of the prevalence of Alzheimer’s disease in the assigned to the CEE + MPA trial, but not among women United States,” Alzheimer’s and Dementia,vol.7,no.1,pp. assigned to CEE-alone. In contrast, an additional followup 61–73, 2011. analyses after the termination of the WHI data demonstrated [3] B. J. Small, L. Fratiglioni, M. Viitanen, B. Winblad, and that participants in the CEE-alone trial did not demonstrate L. Bachman,¨ “The course of cognitive impairment in pre- increased risk for breast cancer [161]. Although the available clinical Alzheimer disease: three- and 6-year follow-up of a International Journal of Alzheimer’s Disease 13

population-based sample,” Archives of Neurology, vol. 57, no. and recently postmenopausal women,” Menopause, vol. 13, 6, pp. 839–844, 2000. no. 3, pp. 411–422, 2006. [4] M. Grundman, R. C. Petersen, S. H. Ferris et al., “Mild [20] E. S. LeBlanc, M. B. Neiss, P. E. Carello, M. H. Samuels, cognitive impairment can be distinguished from Alzheimer and J. S. Janowsky, “Hot flashes and estrogen therapy disease and normal aging for clinical trials,” Archives of do not influence cognition in early menopausal women,” Neurology, vol. 61, no. 1, pp. 59–66, 2004. Menopause, vol. 14, no. 2, pp. 191–202, 2007. [5] A. Levey, J. Lah, F. Goldstein, K. Steenland, and D. Bliwise, [21] P. M. Maki, M. J. Gast, A. J. Vieweg, S. W. Burriss, and K. “Mild cognitive impairment: an opportunity to identify pa- Yaffe, “Hormone therapy in menopausal women with cog- tients at high risk for progression to Alzheimer’s disease,” nitive complaints: a randomized, double-blind trial,” Neurol- Clinical Therapeutics, vol. 28, no. 7, pp. 991–1001, 2006. ogy, vol. 69, no. 13, pp. 1322–1330, 2007. [6]D.Blacker,H.Lee,A.Muzikanskyetal.,“Neuropsychological [22] J. Dumas, C. Hancur-Bucci, M. Naylor, C. Sites, and P. measures in normal individuals that predict subsequent Newhouse, “Estradiol interacts with the cholinergic system cognitive decline,” Archives of Neurology,vol.64,no.6,pp. to affectverbalmemoryinpostmenopausalwomen:evidence 862–871, 2007. for the critical period hypothesis,” Hormones and Behavior, [7] J. Saxton, O. L. Lopez, G. Ratcliff et al., “Preclinical Alzheimer vol. 53, no. 1, pp. 159–169, 2008. disease: neuropsychological test performance 1.5 to 8 years [23] M. C. Tierney, P. Oh, R. Moineddin et al., “A random- prior to onset,” Neurology, vol. 63, no. 12, pp. 2341–2347, ized double-blind trial of the effects of hormone therapy 2004. on delayed verbal recall in older women,” Psychoneuroen- [8] M. H. Tabert, J. J. Manly, X. Liu et al., “Neuropsychological docrinology, vol. 34, no. 7, pp. 1065–1074, 2009. prediction of conversion to alzheimer disease in patients with [24] D. H. S. Silverman, C. L. Geist, H. A. Kenna et al., “Dif- mild cognitive impairment,” Archives of General Psychiatry, ferences in regional brain metabolism associated with spe- vol. 63, no. 8, pp. 916–924, 2006. cific formulations of hormone therapy in postmenopausal [9]M.C.Tierney,C.Yao,A.Kiss,andI.McDowell,“Neuropsy- women at risk for AD,” Psychoneuroendocrinology, vol. 36, pp. chological tests accurately predict incident Alzheimer disease 502–513, 2010. after 5 and 10 years,” Neurology, vol. 64, no. 11, pp. 1853– [25] K. Marder, M. X. Tang, B. Alfaro et al., “Postmenopausal 1859, 2005. estrogen use and Parkinson’s disease with and without [10] O. Kohannim, X. Hua, D. P. Hibar et al., “Boosting power for dementia,” Neurology, vol. 50, no. 4, pp. 1141–1143, 1998. clinical trials using classifiers based on multiple biomarkers,” [26] H. H. Fernandez and K. L. Lapane, “Estrogen use among Neurobiology of Aging, vol. 31, no. 8, pp. 1429–1442, 2010. nursing home residents with a diagnosis of Parkinson’s [11] P. P. Zandi, M. C. Carlson, B. L. Plassman et al., “Hormone disease,” Movement Disorders, vol. 15, no. 6, pp. 1119–1124, replacement therapy and incidence of Alzheimer disease in 2000. older women: the Cache County Study,” JAMA, vol. 288, no. [27] M. D. Benedetti, D. M. Maraganore, J. H. Bower et al., 17, pp. 2123–2129, 2002. “Hysterectomy, menopause, and estrogen use preceding [12] C. Kawas, S. Resnick, A. Morrison et al., “A prospective study Parkinson’s disease: an exploratory case-control study,” of estrogen replacement therapy and the risk of developing Movement Disorders, vol. 16, no. 5, pp. 830–837, 2001. Alzheimer’s disease: the Baltimore Longitudinal Study of [28] E. Martignoni, R. E. Nappi, A. Citterio et al., “Reproductive Aging,” Neurology, vol. 48, no. 6, pp. 1517–1521, 1997. life milestone in women with Parkinson’s disease,” Functional [13] M. X. Tang, D. Jacobs, Y. Stern et al., “Effectofoestrogen Neurology, vol. 18, no. 4, pp. 211–217, 2003. during menopause on risk and age at onset of Alzheimer’s [29] L. J. Currie, M. B. Harrison, J. M. Trugman, J. P. Bennett, and disease,” The Lancet, vol. 348, no. 9025, pp. 429–432, 1996. G. F. Wooten, “Postmenopausal estrogen use affects risk for [14] A. Paganini-Hill and V. W. Henderson, “Estrogen deficiency Parkinson disease,” Archives of Neurology, vol. 61, no. 6, pp. and risk of Alzheimer’s disease in women,” American Journal 886–888, 2004. of Epidemiology, vol. 140, no. 3, pp. 256–261, 1994. [30] P. Ragonese, M. D’Amelio, G. Salemi et al., “Risk of Parkin- [15] W. A. Rocca, J. H. Bower, D. M. Maraganore et al., “Increased son disease in women: effect of reproductive characteristics,” risk of cognitive impairment or dementia in women who Neurology, vol. 62, no. 11, pp. 2010–2014, 2004. underwent oophorectomy before menopause,” Neurology, [31] R. A. Popat, S. K. Van Den Eeden, C. M. Tanner et al., “Effect vol. 69, no. 11, pp. 1074–1083, 2007. of reproductive factors and postmenopausal hormone use on [16] W. A. Rocca, B. R. Grossardt, and D. M. Maraganore, “The the risk of Parkinson disease,” Neurology, vol. 65, no. 3, pp. long-term effects of oophorectomy on cognitive and motor 383–390, 2005. aging are age dependent,” Neurodegenerative Diseases, vol. 5, [32] P.Ragonese, M. D’Amelio, G. Callari, G. Salemi, L. Morgante, no. 3-4, pp. 257–260, 2008. and G. Savettieri, “Age at menopause predicts age at onset of [17] W. A. Rocca, B. R. Grossardt, Y. E. Geda et al., “Long-term Parkinson’s disease,” Movement Disorders, vol. 21, no. 12, pp. risk of depressive and anxiety symptoms after early bilateral 2211–2214, 2006. oophorectomy,” Menopause, vol. 15, no. 6, pp. 1050–1059, [33] K. C. Simon, H. Chen, X. Gao, M. A. Schwarszchild, and A. 2008. Ascherio, “Reproductive factors, exogenous estrogen use, and [18] Y. Z. Bagger, L. B. Tanko,P.Alexandersen,G.Qin,andC.´ risk of Parkinson’s disease,” Movement Disorders, vol. 24, no. Christiansen, “Early postmenopausal hormone therapy may 9, pp. 1359–1365, 2009. prevent cognitive impairment later in life,” Menopause, vol. [34]P.J.Blanchet,J.Fang,K.Hyland,L.A.Arnold,M.M.Moura- 12, no. 1, pp. 12–17, 2005. dian, and T. N. Chase, “Short-term effects of high-dose 17β- [19] H. Joffe, J. E. Hall, S. Gruber et al., “Estrogen ther- estradiol in postmenopausal PD patients: a crossover study,” apy selectively enhances prefrontal cognitive processes: a Neurology, vol. 53, no. 1, pp. 91–95, 1999. randomized, double-blind, placebo-controlled study with [35] E. Strijks, J. A. M. Kremer, and M. W. I. M. Horstink, functional magnetic resonance imaging in perimenopausal “Effects of female sex steroids on Parkinson’s disease in 14 International Journal of Alzheimer’s Disease

postmenopausal women,” Clinical Neuropharmacology, vol. and cognition,” Neurobiology of Aging, vol. 21, no. 2, pp. 373– 22, no. 2, pp. 93–97, 1999. 383, 2000. [36] K. L. Tsang, S. L. Ho, and S. K. Lo, “Estrogen improves [52]C.C.Persad,J.K.Zubieta,T.Love,H.Wang,A.Tkaczyk, motor disability in parkinsonian postmenopausal women and Y. R. Smith, “Enhanced neuroactivation during verbal with motor fluctuations,” Neurology, vol. 54, no. 12, pp. memory processing in postmenopausal women receiving 2292–2298, 2000. short-term hormone therapy,” Fertility and Sterility, vol. 92, [37] The Parkinson Study Group Poetry I, “A randomized pilot no. 1, pp. 197–204, 2009. trial of estrogen replacement therapy in post-menopausal [53] S. Chen, J. Nilsen, and R. D. Brinton, “Dose and temporal women with Parkinson’s disease,” Parkinsonism & Related pattern of estrogen exposure determines neuroprotective Disorders, vol. 17, no. 10, pp. 757–760, 2011. outcome in hippocampal neurons: therapeutic implications,” [38] C. M. Rivera, B. R. Grossardt, D. J. Rhodes et al., “Increased Endocrinology, vol. 147, no. 11, pp. 5303–5313, 2006. cardiovascular mortality after early bilateral oophorectomy,” [54]C.M.Brown,E.Choi,Q.Xu,M.P.Vitek,andC.A.Colton, Menopause, vol. 16, no. 1, pp. 15–23, 2009. “The APOE4 genotype alters the response of microglia and [39] S. A. Shumaker, C. Legault, S. R. Rapp et al., “Estrogen plus macrophages to 17β-estradiol,” Neurobiology of Aging, vol. progestin and the incidence of dementia and mild cognitive 29, no. 12, pp. 1783–1794, 2008. impairment in postmenopausal women: the Women’s Health [55] M. C. Craig and D. G. M. Murphy, “Alzheimer’s disease in Initiative Memory Study: A randomized-controlled trial,” women,” Best Practice and Research, vol. 23, no. 1, pp. 53–61, JAMA, vol. 289, no. 20, pp. 2651–2662, 2003. 2009. ff [40] S. R. Rapp, M. A. Espeland, S. A. Shumaker et al., “E ect [56] L. M. de Lau and M. M. Breteler, “Epidemiology of Par- of estrogen plus progestin on global cognitive function kinson’s disease,” The Lancet Neurology, vol. 5, no. 6, pp. 525– in postmenopausal women: the Women’s Health Initiative 535, 2006. Memory Study: a randomized controlled trial,” JAMA, vol. 289, no. 20, pp. 2663–2672, 2003. [57] M. Baldereschi, A. Di Carlo, W. A. Rocca et al., “Parkinson’s disease and parkinsonism in a longitudinal study: two-fold [41] B. B. Sherwin, “Estrogen and memory in women: how can higher incidence in men,” Neurology, vol. 55, no. 9, pp. 1358– we reconcile the findings?” Hormones and Behavior, vol. 47, 1363, 2000. no. 3, pp. 371–375, 2005. [58] G. Alves, E. B. Forsaa, K. F. Pedersen, M. Dreetz Gjerstad, and [42] L. D. McCullough and P. D. Hurn, “Estrogen and ischemic J. P. Larsen, “Epidemiology of Parkinson’s disease,” Journal of neuroprotection: an integrated view,” Trends in Endocrinol- Neurology, vol. 255, supplement 5, pp. 18–32, 2008. ogy and Metabolism, vol. 14, no. 5, pp. 228–235, 2003. [43] R. D. Brinton, “Estrogen-induced plasticity from cells to [59] G. Alves, B. Muller,¨ K. Herlofson et al., “Incidence of circuits: predictions for cognitive function,” Trends in Phar- Parkinson’s disease in Norway: the Norwegian ParkWest macological Sciences, vol. 30, no. 4, pp. 212–222, 2009. study,” Journal of Neurology, Neurosurgery and Psychiatry, vol. 80, no. 8, pp. 851–857, 2009. [44] M. L. Voytko, G. P.Tinkler, C. Browne, and J. R. Tobin, “Neu- roprotective effects of estrogen therapy for cognitive and [60] J. F. Kurtzke and I. D. Goldberg, “Parkinsonism death rates neurobiological profiles of monkey models of menopause,” byrace,sex,andgeography,”Neurology, vol. 38, no. 10, pp. American Journal of Primatology, vol. 71, no. 9, pp. 794–801, 1558–1561, 1988. 2009. [61] K. Marder, M. X. Tang, H. Mejia et al., “Risk of Parkinson’s [45] P. M. Maki and E. Sundermann, “Hormone therapy and disease among first-degree relatives: a community- based cognitive function,” Human Reproduction Update, vol. 15, no. study,” Neurology, vol. 47, no. 1, pp. 155–160, 1996. 6, pp. 667–681, 2009. [62] K. P. Ebmeier, S. A. Calder, J. R. Crawford, L. Stewart, [46] R. B. Gibbs, “Estrogen therapy and cognition: a review of the R. H. B. Cochrane, and J. A. O. Besson, “Dementia in cholinergic hypothesis,” Endocrine Reviews, vol. 31, no. 2, pp. idiopathic Parkinson’s disease: prevalence and relationship 224–253, 2010. with symptoms and signs of Parkinsonism,” Psychological [47] W. Wharton, C. E. Gleason, K. R. Lorenze et al., “Poten- Medicine, vol. 21, no. 1, pp. 69–76, 1991. tial role of estrogen in the pathobiology and prevention [63]G.F.Wooten,L.J.Currie,V.E.Bovbjerg,J.K.Lee,andJ. of Alzheimer’s disease,” American Journal of Translational Patrie, “Are men at greater risk for Parkinson’s disease than Research, vol. 1, no. 2, pp. 131–147, 2009. women?” Journal of Neurology, Neurosurgery and Psychiatry, [48] R. Norbury, M. J. Travis, K. Erlandsson, W. Waddington, P. vol. 75, no. 4, pp. 637–639, 2004. J. Ell, and D. G. M. Murphy, “Estrogen Therapy and brain [64] M. C. de Rijk, M. M. B. Breteler, G. A. Graveland et al., muscarinic receptor density in healthy females: a SPET “Prevalence of Parkinson’s disease in the elderly: the Rotter- study,” Hormones and Behavior, vol. 51, no. 2, pp. 249–257, dam study,” Neurology, vol. 45, no. 12, pp. 2143–2146, 1995. 2007. [65] M. C. de Rijk, L. J. Launer, K. Berger et al., “Prevalence [49] G. W. Small, V. Kepe, L. M. Ercoli et al., “PET of brain of Parkinson’s disease in Europe: a collaborative study amyloid and tau in mild cognitive impairment,” The New of population-based cohorts,” Neurology, vol. 54, no. 11, England Journal of Medicine, vol. 355, no. 25, pp. 2652–2663, supplement 5, pp. S21–S23, 2000. 2006. [66]C.A.Haaxma,B.R.Bloem,G.F.Bormetal.,“Gender [50] H. Xu, R. Wang, Y. W. Zhang, and X. Zhang, “Estrogen, differences in Parkinson’s disease,” JournalofNeurology, β-amyloid metabolism/trafficking, and Alzheimer’s disease,” Neurosurgery and Psychiatry, vol. 78, no. 8, pp. 819–824, Annals of the New York Academy of Sciences, vol. 1089, pp. 2007. 324–342, 2006. [67] T. Di Paolo, “Modulation of brain dopamine transmission by [51] P. M. Maki and S. M. Resnick, “Longitudinal effects of sex steroids,” Reviews in the Neurosciences,vol.5,no.1,pp. estrogen replacement therapy on PET cerebral blood flow 27–41, 1994. International Journal of Alzheimer’s Disease 15

[68] R. Sandyk, “Estrogens and the pathophysiology of Parkin- [83] K. A. Disshon and D. E. Dluzen, “Estrogen as a neuromod- son’s disease,” International Journal of Neuroscience, vol. 45, ulator of MPTP-induced neurotoxicity: effects upon striatal no. 1-2, pp. 119–122, 1989. dopamine release,” Brain Research, vol. 764, no. 1-2, pp. 9– [69] N. P. Quinn and C. D. Marsden, “Menstrual-related fluctua- 16, 1997. tions in Parkinson’s disease,” Movement Disorders, vol. 1, no. [84] H. Sawada, M. Ibi, T. Kihara et al., “Estradiol protects 1, pp. 85–87, 1986. dopaminergic neurons in a MPP+Parkinson’s disease [70]K.Kompoliti,C.L.Comella,J.A.Jaglin,S.Leurgans,R. model,” Neuropharmacology, vol. 42, no. 8, pp. 1056–1064, Raman, and C. G. Goetz, “Menstrual-related changes in 2002. motoric function in women with Parkinson’s disease,” Neu- [85] T. Obata, “Environmental estrogen-like chemicals and hy- rology, vol. 55, no. 10, pp. 1572–1574, 2000. droxyl radicals induced by MPTP in the striatum: a review,” [71] D. Aarsland, J. Zaccai, and C. Brayne, “A systematic review Neurochemical Research, vol. 27, no. 5, pp. 423–431, 2002. of prevalence studies of dementia in Parkinson’s disease,” [86] M. Grandbois, M. Morissette, S. Callier, and T. Di Paolo, Movement Disorders, vol. 20, no. 10, pp. 1255–1263, 2005. “Ovarian steroids and raloxifene prevent MPTP-induced [72] D. P. Salmon and J. V. Filoteo, “Neuropsychology of cortical dopamine depletion in mice,” NeuroReport,vol.11,no.2,pp. versus subcortical dementia syndromes,” Seminars in Neurol- 343–346, 2000. ogy, vol. 27, no. 1, pp. 7–21, 2007. [87]M.C.Morale,P.A.Serra,F.L’Episcopoetal.,“Estrogen,neu- [73] A. Quesada, H. E. Romeo, and P. Micevych, “Distribution roinflammation and neuroprotection in Parkinson’s disease: and localization patterns of estrogen receptor-β and insulin- glia dictates resistance versus vulnerability to neurodegener- like growth factor-1 receptors in neurons and glial cells of the ation,” Neuroscience, vol. 138, no. 3, pp. 869–878, 2006. female rat substantia nigra: localization of ERβ and IGF-1R [88] P. J. Shughrue, “Estrogen attenuates the MPTP-induced loss in substantia nigra,” Journal of Comparative Neurology, vol. of dopamine neurons from the mouse SNc despite a lack of 503, no. 1, pp. 198–208, 2007. estrogen receptors (ERα and ERβ),” Experimental Neurology, [74] A. Stroppolo, C. Tian, B. Guinea et al., “17β-estradiol vol. 190, no. 2, pp. 468–477, 2004. promotes striatal medium size spiny neuronal maturation in [89] S. Al Sweidi, M. G. Sanchez,´ M. Bourque, M. Morissette, D. vitro,” Neuroendocrinology, vol. 79, no. 5, pp. 259–267, 2004. Dluzen, and T. di Paolo, “Oestrogen receptors and signalling [75] M. Bains, J. C. Cousins, and J. L. Roberts, “Neuroprotection pathways: implications for neuroprotective effects of sex by estrogen against MPP+-induced dopamine neuron death steroids in Parkinson’s disease,” Journal of Neuroendocrinol- is mediated by ERα in primary cultures of mouse mesen- ogy, vol. 24, no. 1, pp. 48–61, 2012. cephalon,” Experimental Neurology, vol. 204, no. 2, pp. 767– [90] S. Al-Sweidi, M. Morissette, M. Bourque, and T. Di Paolo, 776, 2007. “Estrogen receptors and gonadal steroids in vulnerability [76] O. K. Rønnekleiv, A. Malyala, and M. J. Kelly, “Membrane- and protection of dopamine neurons in a mouse model of initiated signaling of estrogen in the brain,” Seminars in Parkinson’s disease,” Neuropharmacology,vol.61,no.4,pp. Reproductive Medicine, vol. 25, no. 3, pp. 165–177, 2007. 583–591, 2011. [77] T.A.Roepke,O.K.Ronnekleiv,andM.J.Kelly,“Physiological [91] W. Tripanichkul, O. Gerdprasert, and E. O. Jaroensup- consequences of membrane-initiated estrogen signaling in paperch, “Estrogen reduces BDNF level, but maintains the brain,” Frontiers in Bioscience, vol. 16, no. 4, pp. 1560– dopaminergic cell density in the striatum of MPTP mouse 1573, 2011. model,” International Journal of Neuroscience, vol. 120, no. 7, [78] S. A. Gardiner, M. F. Morrison, P. D. Mozley et al., “Pilot pp. 489–495, 2010. study on the effect of estrogen replacement therapy on [92] A. M. Baraka, A. A. Korish, G. A. Soliman, and H. Kamal, brain dopamine transporter availability in healthy, post- “The possible role of estrogen and selective estrogen receptor menopausal women,” American Journal of Geriatric Psychi- modulators in a rat model of Parkinson’s disease,” Life atry, vol. 12, no. 6, pp. 621–630, 2004. Sciences, vol. 88, no. 19-20, pp. 879–885, 2011. [79] S. McArthur, H. E. Murray, A. Dhankot, D. T. Dexter, [93] D. Dluzen, “Estrogen decreases corpus striatal neurotoxicity and G. E. Gillies, “Striatal susceptibility to a dopaminergic in response to 6- hydroxydopamine,” Brain Research, vol. 767, neurotoxin is independent of sex hormone effects on cell no. 2, pp. 340–344, 1997. survival and DAT expression but is exacerbated by central [94] A. Quesada and P. E. Micevych, “Estrogen interacts with the aromatase inhibition,” Journal of Neurochemistry, vol. 100, IGF-1 system to protect nigrostriatal dopamine and maintain no. 3, pp. 678–692, 2007. motoric behavior after 6-hydroxdopamine lesions,” Journal of [80] J. Bove,´ J. Serrats, G. Mengod, R. Cortes,´ E. Tolosa, and Neuroscience Research, vol. 75, no. 1, pp. 107–116, 2004. C. Marin, “Neuroprotection induced by the adenosine A2A [95] H. E. Murray, A. V. Pillai, S. R. Mcarthur et al., “Dose- and antagonist CSC in the 6-OHDA rat model of parkinsonism: sex-dependent effects of the neurotoxin 6-hydroxydopamine effect on the activity of striatal output pathways,” Experimen- on the nigrostriatal dopaminergic pathway of adult rats: tal Brain Research, vol. 165, no. 3, pp. 362–374, 2005. differential actions of estrogen in males and females,” [81] J. Bove,´ J. Serrats, G. Mengod, R. Cortes,´ E. Aguilar, and C. Neuroscience, vol. 116, no. 1, pp. 213–222, 2003. Marin, “Reversion of levodopa-induced motor fluctuations [96] M. F. Cordellini, G. Piazzetta, K. C. Pinto et al., “Effect of by the A2A antagonist CSC is associated with an increase different doses of estrogen on the nigrostriatal dopaminergic in striatal preprodynorphin mRNA expression in 6-OHDA- system in two 6-hydroxydopamine-induced lesion models of lesioned rats,” Synapse, vol. 59, no. 7, pp. 435–444, 2006. Parkinson’s disease,” Neurochemical Research, vol. 36, no. 6, [82] R. S. Kenchappa, L. Diwakar, J. Annepu, and V. Ravin- pp. 955–961, 2011. dranath, “Estrogen and neuroprotection: higher constitutive [97] T. M. Gajjar, L. I. Anderson, and D. E. Dluzen, “Acute effects expression of glutaredoxin in female mice offers protection of estrogen upon methamphetamine induced neurotoxicity against MPTP-mediated neurodegeneration,” The FASEB of the nigrostriatal dopaminergic system,” Journal of Neural Journal, vol. 18, no. 10, pp. 1102–1104, 2004. Transmission, vol. 110, no. 11, pp. 1215–1224, 2003. 16 International Journal of Alzheimer’s Disease

[98] H. Ohtani, M. Nomoto, and T. Douchi, “Chronic estrogen potent antiretroviral therapy: charter Study,” Neurology, vol. treatment replaces striatal dopaminergic function in ovariec- 75, no. 23, pp. 2087–2096, 2010. tomized rats,” Brain Research, vol. 900, no. 2, pp. 163–168, [114] L. Garvey, “HIV-associated central nervous system diseases in 2001. the recent combination antiretroviral therapy era,” European [99] G. C. Wagner, T. L. Tekirian, and C. T. Cheo, “Sexual Journal of Neurology, vol. 18, no. 3, pp. 527–534, 2011. differences in sensitivity to methamphetamine toxicity,” [115] V. Tozzi, P. Balestra, R. Bellagamba et al., “Persistence of Journal of Neural Transmission—General Section, vol. 93, no. neuropsychologic deficits despite long-term highly active 1, pp. 67–70, 1993. antiretroviral therapy in patients with HIV-related neurocog- [100] Y.-L. Yu and G. C. Wagner, “Influence of gonadal hormones nitive impairment: prevalence and risk factors,” Journal of on sexual differences in sensitivity to methamphetamine- Acquired Immune Deficiency Syndromes,vol.45,no.2,pp. induced neurotoxicity,” Journal of Neural Transmission— 174–182, 2007. Parkinson’s Disease and Dementia Section,vol.8,no.3,pp. [116] H. E. Wynn, R. C. Brundage, and C. V. Fletcher, “Clinical 215–221, 1994. implications of CNS penetration of antiretroviral drugs,” [101] H. Sawada, M. Ibi, T. Kihara, M. Urushitani, A. Akaike, and CNS Drugs, vol. 16, no. 9, pp. 595–609, 2002. S. Shimohama, “Estradiol protects mesencephalic dopamin- [117] K. J. Kallianpur, G. R. Kirk, N. Sailasuta et al., “Regional ergic neurons from oxidative stress-induced neuronal death,” cortical thinning associated with detectable levels of HIV Journal of Neuroscience Research, vol. 54, no. 5, pp. 707–719, DNA,” Cerebral Cortex. In press. 1998. [118] J. Harezlak, S. Buchthal, M. Taylor et al., “Persistence of HIV- [102] H. Sawada and S. Shimohama, “Neuroprotective effects of associated cognitive impairment, inflammation, and neu- estradiol in mesencephalic dopaminergic neurons,” Neuro- ronal injury in era of highly active antiretroviral treatment,” science and Biobehavioral Reviews, vol. 24, no. 1, pp. 143–147, AIDS, vol. 25, no. 5, pp. 625–633, 2011. 2000. [119] I. C. Anthony, S. N. Ramage, F. W. Carnie, P. Simmonds, [103] E. Biewenga, L. Cabell, and T. Audesirk, “Estradiol and and J. E. Bell, “Influence of HAART on HIV-related CNS raloxifene protect cultured SN4741 neurons against oxidative disease and neuroinflammation,” Journal of Neuropathology stress,” Neuroscience Letters, vol. 373, no. 3, pp. 179–183, and Experimental Neurology, vol. 64, no. 6, pp. 529–536, 2005. 2005. [104] A. D. Ramirez, X. Liu, and F. S. Menniti, “Repeated estradiol [120] T. Wang, J. A. Rumbaugh, and A. Nath, “Viruses and the treatment prevents MPTP-induced dopamine depletion in brain: from inflammation to dementia,” Clinical Science, vol. male mice,” Neuroendocrinology, vol. 77, no. 4, pp. 223–231, 110, no. 4, pp. 393–407, 2006. 2003. [121] E. Vegeto, S. Belcredito, S. Etteri et al., “Estrogen receptor- [105] M. D’Astous, P. Mendez, M. Morissette, L. M. Garcia-Segura, α mediates the brain antiinflammatory activity of estradiol,” and T. Di Paolo, “Implication of the phosphatidylinositol- Proceedings of the National Academy of Sciences of the United 3 kinase/protein kinase b signaling pathway in the neuro- States of America, vol. 100, no. 16, pp. 9614–9619, 2003. protective effect of estradiol in the striatum of 1-methyl-4- [122] I. Zemlyak, S. Brooke, and R. Sapolsky, “Estrogenic protec- phenyl-1,2,3,6-tetrahydropyridine mice,” Molecular Pharma- tion against gp120 neurotoxicity: role of microglia,” Brain cology, vol. 69, no. 4, pp. 1492–1498, 2006. Research, vol. 1046, no. 1-2, pp. 130–136, 2005. [106] X. Liu, X. L. Fan, Y. Zhao et al., “Estrogen provides neuropro- [123] E. Vegeto, V. Benedusi, and A. Maggi, “Estrogen anti- tection against activated microglia-induced dopaminergic inflammatory activity in brain: a therapeutic opportunity neuronal injury through both estrogen receptor-α and for menopause and neurodegenerative diseases,” Frontiers in estrogen receptor-β in microglia,” Journal of Neuroscience Neuroendocrinology, vol. 29, no. 4, pp. 507–519, 2008. Research, vol. 81, no. 5, pp. 653–665, 2005. [124] C. Kaushic, K. L. Roth, V. Anipindi, and F. Xiu, “Increased [107] J. A. Dykens, W. H. Moos, and N. Howell, “Development prevalence of sexually transmitted viral infections in women: of 17α-estradiol as a neuroprotective therapeutic agent: the role of female sex hormones in regulating susceptibility rationale and results from a phase I clinical study,” Annals and immune responses,” Journal of Reproductive Immunol- of the New York Academy of Sciences, vol. 1052, pp. 116–135, ogy, vol. 88, no. 2, pp. 204–209, 2011. 2005. [125] Z. Hel, E. Stringer, and J. Mestecky, “Sex steroid hormones, [108] K. S. P. McNaught and C. W. Olanow, “Protein aggregation hormonal contraception, and the immunobiology of human in the pathogenesis of familial and sporadic Parkinson’s immunodeficiency virus-1 infection,” Endocrine Reviews, vol. disease,” Neurobiology of Aging, vol. 27, no. 4, pp. 530–545, 31, no. 1, pp. 79–97, 2010. 2006. [126] R. A. Coutinho, “Some aspects of the natural history of HIV [109] M. Hirohata, K. Ono, A. Morinaga, T. Ikeda, and M. Yam- infection,” Tropical Medicine and International Health, vol. 5, ada, “Anti-aggregation and fibril-destabilizing effects of sex no. 7, pp. A22–A25, 2000. hormones on α-synuclein fibrils in vitro,” Experimental [127] H. Farzadegan, D. R. Hoover, J. Astemborski et al., “Sex Neurology, vol. 217, no. 2, pp. 434–439, 2009. differencesinHIV-1viralloadandprogressiontoAIDS,”The [110] (UNAIDS) JUNPoHA. UNAIDS Report on the Global AIDS Lancet, vol. 352, no. 9139, pp. 1510–1514, 1998. Epidemic, 2010. [128] E. Nicastri, C. Angeletti, L. Palmisano et al., “Gender differ- [111] T. C. Quinn and J. Overbaugh, “HIV/AIDS in women: an ences in clinical progression of HIV-1-infected individuals expanding epidemic,” Science, vol. 308, no. 5728, pp. 1582– during long-term highly active antiretroviral therapy,” AIDS, 1583, 2005. vol. 19, no. 6, pp. 577–583, 2005. [112] A. Antinori, G. Arendt, J. T. Becker et al., “Updated research [129] P. G. A. Cornelisse, V. Montessori, B. Yip et al., “The impact nosology for HIV-associated neurocognitive disorders,” Neu- of zidovudine on dementia-free survival in a population of rology, vol. 69, no. 18, pp. 1789–1799, 2007. HIV-positive men and women on antiretroviral therapy,” [113] R. K. Heaton, D. B. Clifford, D. R. Franklin Jr. et al., “HIV- International Journal of STD and AIDS, vol. 11, no. 1, pp. 52– associated neurocognitive disorders persist in the era of 56, 2000. International Journal of Alzheimer’s Disease 17

[130] S. N. Asin, A. M. Heimberg, S. K. Eszterhas, C. Rollenhagen, [145] M. E. Wilson, F. O. Dimayuga, J. L. Reed et al., “Immune and A. L. Howell, “Estradiol and progesterone regulate HIV modulation by estrogens: role in CNS HIV-1 infection,” type 1 replication in peripheral blood cells,” AIDS Research Endocrine, vol. 29, no. 2, pp. 289–297, 2006. and Human Retroviruses, vol. 24, no. 5, pp. 701–716, 2008. [146] R. A. Clark and R. Bessinger, “Clinical manifestations [131] A. J. Davidson, S. L. Bertram, D. C. Lezotte et al., “Com- and predictors of survival in older women infected with parison of health status, socioeconomic characteristics, and HIV,” Journal of Acquired Immune Deficiency Syndromes and knowledge and use of HIV-related resources between HIV- Human Retrovirology, vol. 15, no. 5, pp. 341–345, 1997. infected women and men,” Medical Care, vol. 36, no. 12, pp. [147] A. Chiesi, S. Vella, L. G. Dally et al., “Epidemiology of AIDS 1676–1684, 1998. dementia complex in Europe,” Journal of Acquired Immune [132] M. Cohen, “Natural history of HIV infection in women,” Deficiency Syndromes and Human Retrovirology, vol. 11, no. Obstetrics and Gynecology Clinics of North America, vol. 24, 1, pp. 39–44, 1996. no. 4, pp. 743–758, 1997. [148] J. M. Fa´ılde-Garrido, M. R. Alvarez, and M. A. Simon-L´ opez,´ [133] S. L. Melnick, R. Sherer, T. A. Louis et al., “Survival and “Neuropsychological impairment and gender differences in disease progression according to gender of patients with HIV HIV-1 infection,” Psychiatry and Clinical Neurosciences, vol. infection: the Terry Beirn Community Programs for Clinical 62, no. 5, pp. 494–502, 2008. Research on AIDS,” JAMA, vol. 272, no. 24, pp. 1915–1921, [149] O. L. Lopez, J. Wess, J. Sanchez, M. A. Dew, and J. T. 1994. Becker, “Neurological characteristics of HIV-infected men [134] S. E. Barkan, S. L. Melnick, S. Preston-Martin et al., “The and women seeking primary medical care,” European Journal Women’s Interagency HIV study,” Epidemiology, vol. 9, no. of Neurology, vol. 6, no. 2, pp. 205–209, 1999. 2, pp. 117–125, 1998. [150] E. Martin, R. Gonzalez, J. Vassileva, and P. Maki, “HIV+ men ff [135] S. T. Dheen, C. Kaur, and E. A. Ling, “Microglial activation and women show di erent performance patterns on proce- and its implications in the brain diseases,” Current Medicinal dural learning tasks,” Journal of Clinical and Experimental Chemistry, vol. 14, no. 11, pp. 1189–1197, 2007. Neuropsychology, vol. 33, no. 1, pp. 112–120, 2011. [151] K. R. Robertson, C. Kapoor, W. T. Robertson, S. Fiscus, S. [136] G. Mor, J. Nilsen, T. Horvath et al., “Estrogen and microglia: ff a regulatory system that affects the brain,” Journal of Ford, and C. D. Hall, “No gender di erences in the progres- Neurobiology, vol. 40, no. 4, pp. 484–496, 1999. sion of nervous system disease in HIV infection,” Journal of Acquired Immune Deficiency Syndromes,vol.36,no.3,pp. [137] D. E. Brenneman, S. K. McCune, R. F. Mervis, and J. M. Hill, 817–822, 2004. “gp120 as an etiologic agent for NeuroAIDS: neurotoxicity [152] D. Neary, J. S. Snowden, L. Gustafson et al., “Frontotemporal and model systems,” Advances in Neuroimmunology, vol. 4, lobar degeneration: a consensus on clinical diagnostic crite- no. 3, pp. 157–165, 1994. ria,” Neurology, vol. 51, no. 6, pp. 1546–1554, 1998. [138] S. M. Brooke, J. R. McLaughlin, K. M. Cortopassi, and R. [153] D. Neary, J. S. Snowden, and D. M. A. Mann, “Classification M. Sapolsky, “Effect of GP120 on glutathione peroxidase and description of frontotemporal dementias,” Annals of the activity in cortical cultures and the interaction with steroid New York Academy of Sciences, vol. 920, pp. 46–51, 2000. hormones,” Journal of Neurochemistry, vol. 81, no. 2, pp. 277– [154] A. J. Levine and L. Hewett, “Estrogen replacement therapy 284, 2002. and frontotemporal dementia,” Maturitas,vol.45,no.2,pp. [139] I. I. Kruman, A. Nath, and M. P. Mattson, “HIV-1 protein tat 83–88, 2003. induces apoptosis of hippocampal neurons by a mechanism [155] A. Ferreira and A. Caceres, “Estrogen-enhanced neurite involving caspase activation, calcium overload, and oxidative growth: evidence for a selective induction of tau and stable stress,” Experimental Neurology, vol. 154, no. 2, pp. 276–288, microtubules,” Journal of Neuroscience, vol. 11, no. 2, pp. 1998. 392–400, 1991. [140] A. Nath, K. Conant, P. Chen, C. Scott, and E. O. Major, [156] T. M. See, A. K. Lamarre, S. E. Lee, and B. L. Miller, “Genetic “Transient exposure to HIV-1 Tat protein results in cytokine causes of frontotemporal degeneration,” Journal of Geriatric production in macrophages and astrocytes: a hit and run Psychiatry and Neurology, vol. 23, no. 4, pp. 260–268, 2010. phenomenon,” The Journal of Biological Chemistry, vol. 274, [157] H. Seelaar, J. D. Rohrer, Y. A. L. Pijnenburg, N. C. Fox, no. 24, pp. 17098–17102, 1999. and J. C. Van Swieten, “Clinical, genetic and pathological [141] L. Minghetti, S. Visentin, M. Patrizio, L. Franchini, M. A. heterogeneity of frontotemporal dementia: a review,” Journal Ajmone-Cat, and G. Levi, “Multiple actions of the human of Neurology, Neurosurgery and Psychiatry,vol.82,no.5,pp. immunodeficiency virus type-1 tat protein on microglial cell 476–486, 2011. functions,” Neurochemical Research, vol. 29, no. 5, pp. 965– [158] G. Heiss, R. Wallace, G. L. Anderson et al., “Health risks and 978, 2004. benefits 3 years after stopping randomized treatment with [142] A. J. Bruce-Keller, S. W. Barger, N. I. Moss, J. T. Pham, J. estrogen and progestin,” JAMA, vol. 299, no. 9, pp. 1036– N. Keller, and A. Nath, “Pro-inflammatory and pro-oxidant 1045, 2008. properties of the HIV protein Tat in a microglial cell line: [159] S. Shapiro, R. D. T. Farmer, A. O. Mueck, H. Seaman, and attenuation by 17β-estradiol,” Journal of Neurochemistry, vol. J. C. Stevenson, “Does hormone replacement therapy cause 78, no. 6, pp. 1315–1324, 2001. breast cancer? An application of causal principles to three [143] Y. W. Lee, S. Y. Eum, A. Nath, and M. Toborek, “Estrogen- studies. Part 2. The Women’s Health Initiative: estrogen plus mediated protection against HIV Tat protein-induced progestogen,” Journal of Family Planning and Reproductive inflammatory pathways in human vascular endothelial cells,” Health Care, vol. 37, no. 3, pp. 165–172, 2011. Cardiovascular Research, vol. 63, no. 1, pp. 139–148, 2004. [160] C. J. Crandall, A. K. Aragaki, J. A. Cauley et al., “Breast ten- [144] D. R. Wallace, S. Dodson, A. Nath, and R. M. Booze, derness and breast cancer risk in the estrogen plus progestin “Estrogen attenuates gp120- and tat1-72-induced oxidative and estrogen-alone women’s health initiative clinical trials,” stress and prevents loss of dopamine transporter function,” Breast Cancer Research and Treatment, vol. 132, no. 1, pp. Synapse, vol. 59, no. 1, pp. 51–60, 2006. 275–285. 18 International Journal of Alzheimer’s Disease

[161] S. Shapiro, R. D. T. Farmer, A. O. Mueck, H. Seaman, and J. C. Stevenson, “Does hormone replacement therapy cause breast cancer? An application of causal principles to three studies: part 3. The women’s health initiative: unopposed estrogen,” Journal of Family Planning and Reproductive Health Care, vol. 37, no. 4, pp. 225–230, 2011. Copyright of International Journal of Alzheimer's Disease is the property of Hindawi Publishing Corporation and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. Psychoneuroendocrinology (2011) 36, 502—513

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/psyneuen

Differences in regional brain metabolism associated with speciﬁc formulations of hormone therapy in postmenopausal women at risk for AD

Daniel H.S. Silverman a,*, Cheri L. Geist a, Heather A. Kenna b, Katherine Williams b, Tonita Wroolie b, Bevin Powers b, John Brooks c, Natalie L. Rasgon b a UCLA David Geffen School of Medicine, Department of Molecular and Medical Pharmacology, Ahmanson Biological Imaging Clinic, CHS AR-144, University of California, Los Angeles School of Medicine, Los Angeles, CA 90095-6942, USA b Stanford University School of Medicine, Department of Psychiatry and Behavioral Sciences, Stanford, CA 94305-5723, USA c UCLA Semel Institute for Neuroscience and Human Behavior, Los Angeles, CA 90095-6942, USA

Received 22 September 2009; received in revised form 30 June 2010; accepted 3 August 2010

KEYWORDS Summary Differential cerebral metabolic effects of various hormone therapy formulations, and PET; their associations with cognitive status, remain to be established. The principal aim of the current Estrogen formulations; studywastoassessrelationshipsbetweenregionalcerebralmetabolismandestrogen-basedhormone Postmenopausal; therapies. Postmenopausal women (n = 53) at elevated risk for Alzheimer’s disease (AD) were on Hormone therapy; estrogen-containing hormone therapy for at least one year prior to enrollment in a prospective, AD; randomized clinical trial. Subjects underwent an FDG-PET scan, along with neuropsychological, Verbal memory medical, and demographic assessments at time of enrollment, to be repeated one year following randomization to hormone therapy continuation versus discontinuation, and results from analyses of the baseline assessments are reported here. Across all subjects, years of endogenous estrogen exposure correlated most closely with metabolism in right superior frontal gyrus ( p < 0.0005). Women taking 17b-estradiol (E) performed three standard deviations higher in verbal memory than women taking conjugated equine estrogen (CEE), and their verbal memory performance positively correlated with metabolism in Wernicke’s ( p = 0.003) and auditory association ( p =0.002)areas. Women taking progesterone-plus-estrogen had lower metabolism than women taking unopposed estrogen within the mesial and inferior lateral temporal regions ( p < 0.0005) and the inferior frontal cortex, contralateral to Broca’s area ( p < 0.0005). In conclusion, particular areas of relatively preservedmetabolismwereseeninwomenwithmoreyearsofendogenousestrogenexposure,aswell as in women taking estradiol-based formulations or estrogen therapies unopposed by progesterone, together suggesting regionally speciﬁc neuroprotective estrogenic effects. # 2010 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +1 310 825 4257; fax: +1 310 206 4899. E-mail address: [email protected] (D.H.S. Silverman).

1. Introduction (MPA) specifically diminished verbal memory performance (Resnick et al., 2006; Maki et al., 2007). More recently it Sex hormones exert a wide variety of effects upon brain was found that unopposed CEE initiated in postmenopausal development, aging, and function. Though estrogen influ- women aged 65 years and older did not diminish verbal mem- ences have been the focus of particularly intensive investiga- ory performance, though it was associated with somewhat tion in this regard for decades, the nature of their impact lower spatial rotational ability (Resnick et al., 2009a). Rela- upon the mature human brain remains a subject of substan- tive metabolic effects of CEE vs. E also remain controversial tial controversy. Answers to even some of the most basic (Maki and Resnick, 2001). questions–—such as, whether the net influence of exogenous Evidence from functional brain imaging studies, on the estrogen exposure on neurologic function is more harmful other hand, has been more consistent. Neuroimaging studies than beneficial–—have been elusive. This in part reflects the in healthy aging women have demonstrated enhanced func- complexity of estrogen-related effects, such that obtaining tion of medial temporal structures, including the hippocam- meaningful answers may require asking more specific ques- pus, amygdala, and entorhinal cortex among estrogen users tions: for example, whether the effects of estrogen hormone vs. non-users (Maki and Resnick, 2001). Studies of specific therapy (HT) exposure are influenced by the duration of neurochemical systems in living human brain suggest that endogenous estrogen exposure (i.e. length of reproductive postmenopausal hormone therapy positively modulates both life), whether the type of estrogen formulation (17b-estra- cortical serotonin binding (Moses et al., 2000) and choliner- diol (E) compared to conjugated equine estrogens (CEE)) gic receptor density (Smith et al., 2001). Moreover, in influences brain function, and whether the use of concurrent postmenopausal women at genetic risk for AD, HT is asso- progesterone has an effect on brain function or a moderating ciated with attenuated metabolic decline in cortical regions effect on the influence of estrogen on brain function. especially affected by AD pathology, such as posterior cingulate, superior temporal, and lateral temporal cortical regions Experimentally addressing these more focused questions (Rasgon et al., 2001, 2005, 2008a,b). calls for prospective investigation of subpopulations charac- The present study is part of an ongoing larger prospective terized by greater degrees of homogeneity, and who are randomized longitudinal clinical trial, in which postmeno- studied by more sophisticated neurological tools, than has pausal women at increased risk for eventual development of typified previous clinical research in this field. dementia and on HT at time of study enrollment have been Support for neuroprotective roles of estrogen comes both randomly assigned to continue or discontinue HT for at least 2 from human epidemiologic studies (see below), and from years. Cognitive status and regional cerebral metabolism are experimental models of brain function in vitro and in vivo. assessed at baseline and 2 years after randomization. An Estrogen has long been known to influence several neurotrans- interim analysis of the data analyzed from the first 25 sub- mitter systems, including those that are cholinergic (Luine, jects to complete that trial has thus far indicated that women 1985; Dominguez et al., 2004; Kompoliti et al., 2004; Bora randomized to continue HT: (1) experience less significant et al., 2005; Bartholomeusz et al., 2008; Ping et al., 2008), decline of their posterior cortical metabolism (which is most serotonergic (Kendall et al., 1981; Halbreich et al., 1995; marked in right inferior parietal cortex in women who dis- Rubinow et al., 1998; Archer, 1999; Lasiuk and Hegadoren, continue HT), and (2) preserve anterior cortical metabolism 2007), adrenergic (Sar and Stumpf, 1981; Ungar et al., 1993; relative to overall brain metabolism in the medial prefrontal Wang et al., 2006), or dopaminergic (Roy et al., 1990). Hor- area, which is not seen in their demographically matched mones may act through inducing temporary changes in neuro- counterparts who discontinue HT (Rasgon et al., 2008a,b). In nal microstructure, such as dendritic spine formation (Woolley the study reported here, we describe the regional cerebral et al., 1990), affecting neurotransmitters and receptors (McE- metabolism for the entire cohort of subjects for whom PET wen, 1981; Arnold and Breedlove, 1985; Meusburger and Keast, was performed at time of enrollment, for a relatively homo- 2001), altering cell membranes (McEwen et al., 1991)and geneous group of postmenopausal women on HT. We speci- modifying cerebral glucose metabolism and blood flow (Namba fically examine the relationships between brain metabolism and Sokoloff, 1984; Nehlig et al., 1985; Bishop and Simpkins, and three hormonal factors potentially affecting geriatric 1995; Eberling et al., 2000). cognitive decline: length of prior endogenous estrogen expo- Evidence of effects of HT on human brain aging and sure, whether the HT regimen includes a progesterone com- cognition currently is mixed. Observational studies support pound, and type of estrogen formulation being taken at time a decreased risk of clinically diagnosed AD for women on HT of enrollment. (Maki, 2006). Surgical menopause has been associated with increased risk of cognitive impairment dementia later in life (Rocca et al., 2007). Surgically menopausal HTusers have also 2. Methods been reported to have higher verbal memory performance compared to non-users in clinical trials with randomized, The study in its entirety was approved by the Stanford placebo-controlled design (Phillips and Sherwin, 1992), as University Human Research Protection Program and the Insti- well as cross-sectional studies (Nappi et al., 1999; Verghese tutional Review Board at the University of California, Los et al., 2000). On the other hand, the large randomized Angeles (UCLA). Cognitively normal postmenopausal women control trial (RCT), Women’s Health Initiative Memory Study ages 50—65 at risk for AD and receiving estrogen-containing (WHIMS), has reported a nearly doubled risk for all-cause hormone therapy (HT) for at least one year were enrolled. All dementia for women on HTcompared to women not receiving subjects were receiving either estrogen therapy opposed or HT (Shumaker et al., 2004) and estrogen plus progestin unopposed by progesterone through any route of adminis- therapy in the form of CEE and medroxyprogesterone acetate tration (i.e. oral, transdermal, vaginal), and no changes in 504 D.H.S. Silverman et al. their HT regimen were implemented by the investigators Hormone levels were assayed at baseline. Follicle-stimu- prior to randomization or thereafter. Assessment of endo- lating hormone (FSH) was measured by immunoassay. Estra- crine reproductive markers included gathering information diol was measured using the enzyme-linked immunosorbent on age at menarche and menopause, parity, use of hormonal assay (Bio-Quant Inc., San Diego, CA). Limits of detection contraception during reproductive years, duration of peri- were 0.10 mIU/mL for FSH, and the minimum detectable menopausal transition in relation to time of start of HT use, concentration of the estradiol assay was 2 standard devia- and type of menopausal symptoms. All subjects had a screen- tions plus mean of a zero standard, which was estimated to be ing and baseline visit. The screening visit included psychia- 0.4 pg/mL. tric, physical, and neurological examination, and laboratory blood measures to determine eligibility for the study. During 2.1. PET scan acquisition and analysis the physical and neurological examination, subjects were screened for Parkinson disease using the motor examination Participants were required to fast 4—6 h for the PET imaging (items 18—31) of the Unified Parkinson’s Disease Rating Scale studies. The [F-18] fluorodeoxyglucose (FDG) PET method (Fahn et al., 1987). After the screening visit, subjects was used for the determination of patterns of regional underwent positron emission tomography (PET) scan and cerebral metabolism. An intravenous line was placed 10— neuropsychological testing. 15 min prior to injection of 370 MBq FDG. Uptake of FDG The study required the following inclusion criteria: will- proceeded while subjects were supine with low ambient ingness to sign human subject consent prior to enrollment lighting and sound, eyes and ears unoccluded. Scans were into the study; willingness to be randomized to continuation/ performed 40 min post FDG injection using a CTI/Siemens discontinuation of HT, women ages 50—65 years of age; 1 (Siemens Corp, Hoffman Estates, Il) HR+ tomograph (63 year post complete cessation of menses; 1 year current HT image planes). use; 8 years of education, and adequate visual and auditory We analyzed PET data by both a standardized volume of acuity to allow neuropsychological testing. In addition, all interest method as well as the statistical parametric mapping subjects were required to be at elevated risk for eventual method developed by Friston et al. (1995a,b). Briefly, images development of dementia, as defined by one or more of the from all subjects were co-registered and reoriented into a following risk factors: personal history of mood disorder; standardized coordinate system using the SPM2 software personal history of hypothyroidism; family history of AD; package courteously provided by the members of the image documentation of the apolipoprotein (APOE) allele e4, con- analysis team at the Wellcome Department of Cognitive ferring increased risk for AD. Neurology, Functional Imaging Laboratory (London, UK). Data Because cognitive decline may be caused by a wide variety were spatially smoothed, and normalized to mean global of conditions having different cerebral metabolic signatures, activity as previously described (Silverman et al., 2007), we excluded subjects with impairment from numerous causes with the exception that a 12 mm (full-width half-maximum) (e.g., vascular disease, etc.), to enrich for those at increased smoothing filter was applied to images prior to statistical risk specifically for AD. Volunteers with a history of TIAs, analysis. The set of pooled data were then assessed with the carotid bruits, or lacunes on MRI scan were excluded. Other t-statistic on a voxel-by-voxel basis, to identify the profile of exclusion criteria included evidence of current depression as voxels that significantly covaried with parameters character- determined by a score of 8 on the 17-item Hamilton Depres- izing each subject. sion Rating Scale (Hamilton, 1960), history of drug or alcohol In addition, relative quantification of regional brain activ- abuse, contraindication for MRI scan (e.g., metal in body, ity was performed using software originally developed at claustrophobia), history of mental illness (excluding mood UCLA dedicated to the visual display and quantitative ana- disorders), or significant cognitive impairment, as evidenced lysis of brain PET data, which has been cleared for those by impairment in daily functions and/or MMSE < 24 (Folstein purposes by the U.S. Food and Drug Administration and is et al., 1975), history of myocardial infarction within the commercially available as the clinical package NeuroQTM previous year or unstable cardiac disease, significant cere- (Syntermed Inc., Atlanta). The software implements an algo- brovascular disease, as evidenced by neurological examina- rithm for automatically measuring, after correction for tis- tion, uncontrolled hypertension (systolic BP > 170 mmHg or sue-based attenuation, the number of radioactive events diastolic BP > 100 mmHg), history of significant liver disease, emitted by a positron source (gamma-ray lines of coinci- clinically significant pulmonary disease, diabetes, or cancer. dence) per second detected by the PET scanner, emanating Subjects were excluded if they already had possible or from pixel locations assigned by a computerized reconstruc- probable AD (McKhann et al., 1984) or any other dementia tion algorithm as falling within each standardized region of (e.g., vascular, Lewy body, frontotemporal), or evidence of interest (sROI). Mean pixel activity values were calculated neurologic or other physical illness that could be expected to within each of 240 sROI’s defined throughout those transaxial imminently produce cognitive deterioration. Subjects were planes across the field of view in which brain tissue was also excluded if they used drugs with potential to significantly represented, following the transformation of each PET scan affect psychometric test results, including centrally active to a template space by a method previously described by Tai beta-blockers, narcotics, clonidine, anti-Parkinsonian med- et al. (1997). The sROI’s were then automatically grouped ications, antipsychotics, benzodiazepines, systemic corticos- into 47 clusters of regions falling within structurally defined teroids, medications with significant cholinergic or boundaries corresponding to distinct neuroanatomical (e.g. anticholinergic effects, anticonvulsants, warfarin, or spora- left inferior parietal lobule) or functional (e.g., Broca’s area) dic use of phytoestrogen-containing products, which may standardized volumes of interest (sVOI’s), and the mean produce estrogen-like agonist and antagonist effects (Polk- activities for each of these volumes calculated. Finally, all owski and Mazurek, 2000; Vincent and Fitzpatrick, 2000). mean activity values were automatically normalized to the Differences in regional brain metabolism 505 mean pixel activity measured throughout that brain scan, or Color Trail Making Test (Color Trails 1 & 2), Delis Kaplan to the mean activity of an individual user-specified reference Executive Function System (DKEFS), Rey-Osterrieth Complex sVOI within that scan. Figure Test (RCFT), Wechsler Adult Intelligence Scale-3rd For the purpose of guiding interpretations of statistical Edition (WAIS-III) — Digit Span, Symbol Coding, and Letter analyses, a priori hypotheses were established for six specific Number Sequencing subtests, and the Wechsler Memory cortical volumes bilaterally, based on their known association Scale-III (WMS) — Logical Memory I & II subtests. The Wechsler with physiologic memory processing and/or pathologic invol- Abbreviated Scale of Intelligence was used to characterize vement in neurodegenerative disorders: left and right medial intelligence (IQ). Individual test scores were z-transformed temporal sVOI’s, including amygdala and hippocampus/para- and parceled into the cognitive domains of attention, verbal hippocampal areas, were chosen for their established role in memory, visual memory, word finding and category fluency, ‘‘new learning’’ (Squire et al., 1992; Stern et al., 1996; Tulving and executive functioning. Lastly, the Memory Function et al., 1996; Gabrieli et al., 1997). Two sVOI’s of the dorso- Questionnaire (MFQ) was administered to assess subjective lateral prefrontal cortex (DLPFC) were chosen for known memory functioning. involvement in both encoding and working memory/retrieval. The posterior cingulate cortex was chosen for its role in 2.3. Reproductive endocrine variables encoding (Shallice et al., 1994) and retrieval (Tulving et al., 1996; Fletcher et al., 1998). Finally, the parietotem- Information was obtained from all subjects and supported by poral and inferior lateral temporal cortical VOI’s were assessed documentation from their primary health care providers on due to their importance in language, semantic memory and the duration of endogenous and exogenous exposure to 54,55 related memory deficits in AD (Martin and Fedio, 1983; reproductive hormones. Duration of endogenous exposure Wiggs et al., 1999). The selected volumes have also been was calculated by subtracting age at menarche from the implicated as biomarkers of impending or actual cognitive age at menopause, duration and type of hormone therapy decline in functional imaging studies. Consistent patterns of was recorded as well as type of menopause, parity and past hypometabolism and hypoperfusion in the parietal, temporal, use of steroidal contraception. and posterior cingulate cortices (Small et al., 1989, 1995; Smith et al., 1992; Reiman et al., 1996; Ibań˜ez et al., 3. Results 1998) have been observed in asymptomatic persons at increased genetic risk for AD, as well as in patients with AD. Results were reported in terms of locations of the most Eighty-one subjects were initially recruited; baseline data significant effects (regionally, and/or in x,y,z Talairach-style from 53 subjects were included in the final analysis; the millimeter coordinates). The probability of finding by chance other 28 subjects had incomplete data sets caused by sche- an effect in any volume containing a voxel of maximal dule conflicts, technically limited MRI or PET imaging, or due significance by statistical parametric mapping, or in any sVOI, to a variety of reasons not noted during initial screening was assessed after a Bonferroni-type adjustment for multiple (hysterectomy at a young age, discontinued HT prior to comparisons, with a correspondingly less harsh correction randomization, TIA, claustrophobia emerging during ima- required for the 12 volumes (6 left, 6 right) specified by a ging, severe mood episode with menopause transition, con- priori hypotheses, and effects at each region of the brain genital ovarian agenesis, progesterone-only therapy). were considered significant if that adjusted probability was Demographic and personal characteristics of these 53 sub- less than 0.05. For large areas, corrected values based upon jects are presented in Table 1. All subjects had MMSE and the number of contiguous voxels achieving a pre-specified cognitive performance scores within the normal range for level of significance were additionally provided. For sVOI persons of their same age and educational level. There was a analyses, based on the number of regions, effects in the history of depression in 41 subjects and 26 were currently sVOI’s specified a priori were considered significant for pre- taking antidepressants. adjusted p 0.004, while effects in other sVOI’s were considered significant for pre-adjusted p 0.001; sVOI analyses 3.1. Endogenous estrogen exposure serving specifically to support another pre-established result were considered corroborative for p < 0.05. Analyses were As a potential factor related to cognitive status in postme- also controlled for examination of multiple effects by using nopausal women, duration of pre-menopausal endogenous analysis of variance (ANOVA) methods with all three main estrogen exposure was explored by statistical parametric hormonal effects entered into the statistical model (see mapping for associations with regional cerebral metabolism. below). The total duration of endogenous estrogen exposure (i.e. age at menopause minus age at menarche) positively corre- 2.2. Neuropsychological assessment lated most closely with metabolism in the posterior part of the right superior frontal gyrus (t = 5.64, p < 0.0005 at voxel An extensive neuropsychological evaluation was conducted of peak significance: 22,32,58), remaining significant after at the baseline and final visits by the study neuropsycholo- full correction for multiple comparisons ( p = 0.005), as well gist. Tests in the neuropsychological battery were chosen as after adjustment for variance in age and education based on prior studies that indicated their ability to predict (t = 5.19, p < 0.0005; p = 0.024 after multiple comparison cognitive decline (Ha¨nninen et al., 1995) and sensitivity to correction). The sVOI analyses corroborated SPM analyses in predict cognitive change (Small et al., 1995). The battery the right superior frontal gyrus (r = 0.27, p = 0.047). Again included the Auditory Consonant Trigrams (ACT), Benton adjusting for age and education, years of endogenous estro- Visual Retention Test (BVRT), Boston Naming Test (BNT), gen exposure also positively correlated with activity in the 506 D.H.S. Silverman et al.

Table 1 Study sample demographics and clinical characteristics (n = 53). Mean SD Range Age 57.8 4.8 49, 69 Years of education 16.3 2.0 12, 20 Length of HT 9.7 5.7 1, 22.5 Age at menarche 13.0 1.6 9, 16 Age at menopause 47.2 5.7 29, 58 Endogenous estrogen exposure 34.2 5.4 17, 44 BMI 25.5 3.6 19, 37 Estradiol (pg/mL) 44.6 32.9 9, 185 FSH (mIU/mL) 60.7 23.7 22.7, 112.2

n Natural or surgical menopause Natural = 33, surgical = 20 APOE type E2/E3 = 3, E2/E4 = 2, E3/3 = 30, E3/4 = 16, E4/4 = 2 Family history of AD (FamH Â AD) FamH Â AD+ = 25, FamHÂ ADÀ =28 Type of HT E + P = 19, E + MPA = 5, CEE + MPA = 9, CEE + P = 2, Unopposed E = 11, Unopposed CEE = 7 Vasomotor symptoms Present = 38, not present = 15, All times are expressed in years. HT: hormone therapy, E: estradiol, CEE: conjugated equine estrogen, P: progesterone, MPA: medroxyprogesterone acetate, AD: Alzheimer’s Disease. right peri-insular area at the axial level of the anterior p < 0.0005), area in the vicinity of the subgenual anterior commissure (48,16,0; t = 5.29, p < 0.0005, p = 0.017 after cingulate (À4,44,À14; t =4.25,p < 0.0005), and an area of multiple-comparison correction); an area of similar size and the left superior lateral temporal cortex (À48,À24,10; signiﬁcance was seen in the contralateral hemisphere but t = 4.58, p < 0.0005) (Fig. 1). The sVOI analyses also corro- centered slightly more superiorly and anteriorly, in the left borated the correlation with metabolism in the left superior inferior frontal gyrus overlapping with Broca’s area lateral temporal cortex (r =0.28, p = 0.037). Not surpris- (À48,20,14; t = 5.28, p < 0.0005, p = 0.017 after multiple- ingly, age of menopause, which was highly correlated with comparison correction). Endogenous estrogen exposure also the duration of endogenous estrogen exposure (r = 0.92, tended to positively correlate with metabolism in the left p < 0.00001), also positively correlated with metabolism [(Fig._1)TD$IG]parahippocampal gyrus (À28,À30,À20; t = 4.05, in these regions.

Fig. 1 Metabolic correlation with years of endogenous estrogen exposure. All voxels positively correlating with endogenous estrogen exposure at p < 0.001 are shown in color. Years of endogenous estrogen exposure correlated with metabolism in areas within the right superior frontal gyrus (red arrow), right peri-insular region (dark blue), left inferior frontal gyrus (light blue), left parahippocampal gyrus (not visible from these surface rendering views), left subgenual anterior cingulate (yellow arrow), left superior lateral temporal gyrus (green arrow) ( p < 0.0005). (For interpretation of the references to colour in this ﬁgure legend, the reader is referred to the web version of the article.) Differences[(Fig._2)TD$IG] in regional brain metabolism 507

Fig. 2 Specific areas of greater metabolism in E-users compared to CEE-users. Metabolism in the (left) Wernicke’s area (yellow arrow) and right superior temporal gyrus (red arrow) was greater in E-users compared to CEE-users ( p < 0.0005). Voxels with significance p < 0.005 are shown above in color. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article.)

3.2. Hormone replacement therapy formulations tions were seen (t = 2.95, p = 0.002; t = 2.50, p = 0.008; t = 2.58, p = 0.007, respectively). The correlations tended 3.2.1. 17b-Estradiol and conjugated equine estrogen to be somewhat stronger among E-users with respect to formulations metabolism in both the Wernicke’s (t = 3.66, p < 0.0005; Among 53 subjects, 35 subjects were taking formulations of t = 3.48, p < 0.001 after adjusting for age, education, and 17b-estradiol (E) and 18 subjects were taking conjugated years of endogenous estrogen exposure) and auditory asso- equine estrogen (CEE) at time of recruitment. Groups were ciation (t = 3.30, p = 0.001; t = 2.94, p = 0.003 after adjust- similar with respect to years of education (av Æ SD; 17 Æ 2.0 ing for age, education, and years of endogenous estrogen vs. 16 Æ 1.9), age at menarche (13 Æ 1.6, 13 Æ 1.7), age at exposure) areas, while CEE-users alone demonstrated the menopause (48 Æ 5.2, 46 Æ 6.7), age at the time of study most significant correlation with metabolism of the superior recruitment (57 Æ 4.5, 58 Æ 5.3), and length of HT (9 Æ 5.7, frontal region (t = 4.37, p < 0.0005 after adjusting for age, 11 Æ 5.7). Approximately 2/3 of the subjects in the E and CEE education, and years of endogenous estrogen exposure). groups had opposed therapy (24 of 35, and 11 of 18, respectively). The proportion of subjects taking E and CEE were also similar for the natural and surgical menopause groups, con- 3.2.2. Estrogen—progesterone versus unopposed stituting 65% E for surgical group (13 of 20), and 67% E for the estrogen replacement therapies women who underwent natural menopause (22 of 33). Among the 53 subjects taking estrogen-based hormone replacement therapy, 18 women were taking unopposed estrogen 3.2.1.1. Group differences in cerebral metabolism. Ana- and 35 were taking progesterone-plus-estrogen replacement lyses by statistical parametric mapping directly comparing therapies, of whom 14 were taking MPA. Groups were similar E-users to CEE-users showed that levels of metabolism in in years of education (av Æ SD; 16 Æ 1.5 vs. 16 Æ 2.2), age at parietotemporal cortex in the vicinity of Wernicke’s area of menarche (13 Æ 1.8, 13 Æ 1.5), age at recruitment (58 Æ 5.6, the left cerebral hemisphere, and the superior temporal 58 Æ 4.4), and length of HT (10 Æ 5.7, 9 Æ 5.7) respectively. gyrus in the auditory association area of the right cerebral They differed in age at menopause (43 Æ 7.0, 49 Æ 3.7), hemisphere, were significantly greater in E-users (t = 4.41 (with the younger average age of the unopposed estrogen and 4.67, respectively; p < 0.0005) (Fig. 2). group primarily reflecting prematurely induced menopause secondary to surgical removal of uterus and/or ovaries in 3.2.1.2. Neuropsychologic performance and cerebral meta- these subjects), and thus in mean number of years of endo- bolic correlates. In examining associations with neuropsy- genous estrogen exposure (31 Æ 6.4, 36 Æ 3.9). A similar chologic performance, we found that E-users attained scores proportion were taking E versus CEE: 69% were on estradiol, for the verbal memory cognitive domain that were 3 SD (24 of 35) in the progesterone-plus-estrogen group, and 61% higher than those of CEE-users ( p = 0.007). We therefore were on estradiol in the unopposed group (11 of 18). examined associations of verbal memory performance with Analyses by statistical parametric mapping showed that cerebral metabolism. postmenopausal women taking progesterone-plus-estrogen Across all subjects, statistical parametric mapping ana- had lower metabolism than postmenopausal women taking lyses indicated that verbal memory performance positively unopposed estrogen within the right mesial temporal region correlated most strongly with the magnitude of cortical (t =4.51,p < 0.0005 at voxel of peak significance: 38,20,À36) metabolism in the right superior frontal gyrus (t = 3.60, and left inferior lateral temporal cortex (t = 4.64, p < 0.0005 p < 0.0005 at voxels of peak significance: 10,32,54), in the at voxel of peak significance: À48,22,À34); the voxel of peak (left) Wernicke’s area (t = 2.86, p = 0.003 at voxel of peak significance within the right mesial temporal region was in the significance: À50,À80,34) and with metabolism of the right largest cluster (3085 contiguous voxels at p < 0.01, pcor- auditory association area (t = 3.00, p = 0.002 at voxel of peak rected = 0.002) (Fig. 3). After adjusting for years of endogenous significance: 70,À36,14). When adjusting for age, education, estrogen exposure, the right and left temporal relative hypo- and years of endogenous estrogen exposure, similar correla- metabolism in progesterone-plus-estrogen users remained 508[(Fig._3)TD$IG] D.H.S. Silverman et al.

(rMT) t =2.61,p = 0.01, and a trend towards lower metabolism in the left inferior lateral temporal cortex (liLT) t = 1.72, p = 0.09 (which became more significant after adjusting for endogenous estrogen exposure; see below.) Only 2 of 35 progesterone-plus-estrogen users had rMT metabolism levels greater than mean rMT metabolism of unopposed estrogen users and, conversely, only 5 of 18 unopposed estrogen users had rMT metabolism levels lower than mean rMT metabolism of progesterone-plus-estrogen users (Fig. 4). Analyses by statistical parametric mapping also demonstrated that women taking progesterone-plus-estrogen had lower metabolism than women taking unopposed estrogen in the right inferior frontal cortex (t = 3.65, p < 0.0005 at voxel of peak significance: 24,26,À14). The significance of this difference again persisted when statistical correction for the disparate number of years of endogenous estrogen exposure was taken into account (t = 3.69, p < 0.0005) and occurred in the largest regional cluster (1747 contiguous voxels at p < 0.005, pcorrected = 0.005) (Fig. 5). The sVOI analyses corroborated SPM analyses, with progesterone- plus-estrogen users having lower mean metabolism than unopposed estrogen users in the right posterior inferior frontal gyrus (rpIFG), contralateral to Broca’s area in the left hemisphere (av Æ SD; 1.14 Æ 0.03 vs. 1.16 Æ 0.03, t = 2.29, p = 0.026, a difference that became more significant following statistical correction for number of years of endogenous estrogen exposure: F = 9.39, p = 0.004). Only 3 of 18 Fig. 3 Posterior temporal areas of decreased metabolism in unopposed estrogen users demonstrated levels of metabo- progesterone-plus-estrogen users compared to unopposed estro- lism in this region that were lower than the mean metabolism gen users. SPM analyses demonstrated right mesial temporal of progesterone-plus-estrogen users and, conversely, only (t = 4.51, p < 0.0005) and left inferior lateral relative hypome- five of the 35 progesterone-plus-estrogen users had meta- tabolism (t = 4.64, p < 0.0005) in progesterone-plus-estrogen bolic levels that were greater than the mean metabolism of users compared to unopposed estrogen users. Voxels with signif- unopposed estrogen users. icance p < 0.01 are shown above in color. The crosshairs inter- Potential interactions of opposed versus unopposed estro- sect at voxel of peak significance (38,20,À36). gen with the other major variables discussed above (years of endogenous estrogen exposure and E vs. CEE) relative to significant (t = 4.01, p < 0.0005; t = 4.42, p < 0.0005, respec- rates of metabolism in the standardized VOI’s were examined tively). The sVOI analyses corroborated SPM analyses, demon- with ANOVA. Taking into account endogenous estrogen expo- strating that progesterone-plus-estrogen users had lower sure, sVOI analyses corroborated significantly lower metabo- mean metabolism in corresponding temporal regions than lism in right mesial temporal, and left inferior lateral [(Fig._4)TD$IG]unopposed estrogen users in the right mesial temporal cortex temporal cortex, in women taking opposed estrogen regi-

Fig. 4 Lower right mesial temporal metabolism in progesterone-plus-estrogen users compared to unopposed estrogen users. In the individual-subject plot above, 35 postmenopausal women on progesterone-plus-estrogen therapy are represented by blue diamonds and the average right mesial temporal metabolism, assessed by sVOI analysis, is shown in light blue (0.797). 18 postmenopausal women on unopposed estrogen therapy are represented by red squares and the average right mesial temporal metabolism is shown in orange (0.815). Progesterone-plus-estrogen users had a lower mean rate of metabolism than unopposed estrogen users ( p = 0.012), a difference that became more signiﬁcant after correcting for years of endogenous estrogen exposure ( p = 0.007). (For interpretation of the references to colour in this ﬁgure legend, the reader is referred to the web version of the article.) Differences[(Fig._5)TD$IG] in regional brain metabolism 509

Fig. 5 Frontal areas of decreased metabolism in progesterone-plus-estrogen users compared to unopposed estrogen users. SPM analyses with correction show right inferior frontal relative hypometabolism in progesterone-plus-estrogen users (n = 35) compared to unopposed estrogen users (n = 18) (top row, sagittal view) (t = 3.69, p < 0.0005). All voxels shown in color have p < 0.005. In the second row of the figure, metabolism in the right posterior inferior frontal gyral (rpIFG) area of a 61 year old postmenopausal woman taking progesterone-plus-estrogen (left) is compared to metabolism of a 62 year old postmenopausal woman taking unopposed estrogen (right). mens (F = 7.788, p = 0.007; F = 4.199, p = 0.046, respec- CEE-users. Postmenopausal women taking estrogen-plus-pro- tively). Simultaneously taking into account years of endo- gesterone users had lower metabolism than postmenopausal genous estrogen exposure and E vs CEE use, sVOI analyses women taking unopposed estrogen in the right mesial tem- again corroborated significantly lower metabolism in these poral region, left inferior lateral temporal cortex, and right regions (F = 7.624, p = 0.008; F = 4.122, p = 0.048). inferior frontal cortex. Overall, particular areas of relatively Looking at the other direction, women taking unopposed preserved metabolism were seen in women with more years estrogen did not have any areas demonstrating significantly of endogenous estrogen exposure, as well as in women taking lower metabolism after correction, than women taking pro- estradiol-based formulations or estrogen therapies unop- gesterone-plus-estrogen formulations. posed by progesterone, together suggesting regionally specific neuroprotective estrogenic effects. The results 4. Discussion presented here provide support for estrogen having a neuroprotective role in brain regions affected during aging. Relative regional cerebral hypometabolism of 53 postmeno- Several of the regions in the current study have also been pausal women on HT selected to be at heightened risk of implicated in a number of previous studies on hormonal eventually developing dementia was associated with a pre- influences involving neuroimaging. Relevant to our identifi- sumptive historical risk factor for cognitive decline: shorter cation of hormone related activity in superior and inferior endogenous estrogen exposure. Years of endogenous estro- frontal gyri, similar or overlapping regions of frontal cortical gen exposure positively correlated most robustly with meta- involvement have been identified as positively correlating bolism in the right superior frontal gyrus, and also in right with HT: orbital gyrus of the frontal cortex (Eberling et al., peri-insular area, left inferior frontal gyrus, left parahippo- 2004), right superior frontal gyrus (Ottowitz et al., 2008), campus, subgenual anterior cingulate, and left superior lat- left middle/superior frontal cortex (Persad et al., 2009) and eral temporal cortex. Additionally, in the present analysis E- left medial frontal cortex (Maki and Resnick, 2000; Persad et users had greater metabolism than CEE-users in the Wer- al., 2009). Initiation of E/HT has also been shown to result in nicke’s area and the auditory association area. Verbal mem- more effective activation of the prefrontal cortex during fMRI ory was better in postmenopausal women taking E than CEE, visual working memory tasks, suggesting functional plasticity and cognitive differences were associated with metabolic in these frontocortical working memory systems among post- differences in these receptive language and auditory asso- menopausal women that can be altered by E/HT (Smith et ciation areas for E-users and in right superior frontal gyrus for al., 2006). In a recent double-blind, placebo-controlled 510 D.H.S. Silverman et al. study, the inferior frontal region was also an area that observational studies and clinical trials have emphasized lack increased during fMRI tests of verbal and spatial working of benefits or likely harm of hormonal therapies in older memory in E/HT users, suggesting improved executive func- postmenopausal women and/or insufficient evidence for tion (Joffe et al., 2006). In addition to increases in inferior benefit in younger women (Genazzani et al., 2007; Bar- frontal activation, Joffe et al. (2006) also found parietal rett-Connor and Laughlin, 2009) of particular note, results activation (bordering Brodman’s area 40, near the Wernicke’s from WHIMS indicated significant increases in risk for prob- region) in E/HT users (40—60 yrs) during a verbal recall task. able dementia in the combination HTsubtrial alone and in the This finding is pertinent to our observation of hormonal combined analysis of combination HTand CEE-alone subtrials effects in Wernicke’s area and its correlation with verbal (Shumaker et al., 2003, 2004) and poorer performance on the memory in all subjects, especially E-users. In a recent, 3MS (Rapp et al., 2003; Espeland et al., 2004), after initia- double-blind placebo-controlled study, postmenopausal tion of HT in postmenopausal women aged 65 and older. On women (56—60 yrs) randomized to hormone therapy had the other hand, in the Multi-Institutional Research in Alzhei- increased activation in the frontal cortex and left inferior mer Genetic Epidemiology case—control study, the protec- parietal cortex (Brodman’s Area 40) during memory encoding tive association of HT was modified by age and was seen (Persad et al., 2009). These findings suggest that estrogen among younger, but not older, postmenopausal women (Hen- may provide cognitive benefits and protect language-related derson et al., 2005). brain areas when given to younger women shortly after or It remains an open question, whether the neuroprotective during the menopausal transition. effects of estrogen documented in preclinical studies can be We also observed hormonal effects in the mesial temporal replicated in randomized controlled trials performed with structures including hippocampal and parahippocampal the most suitable patient group under appropriate condi- regions. Functional neuroimaging studies utilizing PET and tions. On a path to accomplish these goals, variables to be fMRI in healthy aging women, have consistently reported examined as our data suggest, include both the formulations greater metabolic activity in mesial temporal structures, of estrogen and progestogens (progesterone versus MPA) including the hippocampus, amygdala, and entorhinal cor- employed in human studies and clinical trials, dosage and tex, among users of E/HT vs. non-users (Resnick et al., 1998; levels of exposure, timing of hormone therapy exposure, Shaywitz et al., 1999; Eberling et al., 2000; Maki and Resnick, duration of hormone therapy, as well as substrate differences 2000, 2001; Rasgon et al., 2001; Gleason et al., 2006). (i.e., neuronal health status and risk for or presence of Regional brain volumes were recently assessed in structural degenerative disease). One clinical trial which takes these MRI images of the brains of women who had been previously variables in the account is on its way (KEEPS) (Miller et al., enrolled in WHIMS (Resnick et al., 2009b). Overall, mean 2009), but results are not yet being made available. Other hippocampal volumes tended to be smaller among women limitations of the current work that future studies will be randomized to hormone therapy, by an average of À0.10 cm3 able to address, include assessing the potential effect of ( p = 0.05); in the group with at least 2 cm3 ischemic burden, testosterone in hormone replacement regimens (present in the effect was more pronounced, with hippocampal volumes the HT of 5 of the 53 subjects in the study, too small a smaller in the HTarm by an average of À0.16 cm3 ( p = 0.005). subgroup to separately assess) and any differences in effects Hence, those data are consistent with the WHIMS findings of of MPA vs. micronized progesterone that may exist. detrimental cognitive effects associated with opposed CEE Further along these lines, since many demographic factors therapy. The fact that both types of HTwere associated with and other types of patient characteristics may potentially increase in hippocampal neuronal firing could be due to affect cognitive performance and corresponding patterns of predominant effects of the estrogen component, rather than cerebral metabolism, greater homogeneity among subjects the presence of the progesterone. In addition, Ottowitz et can be expected to produce greater statistical power, in al. found that estradiol increased connectivity between the analysis of cognitive and metabolic data from a subject pool right hippocampus and right prefrontal cortex suggesting of given size. In the present investigation, all of our subjects that estradiol may enhance verbal memory performance are postmenopausal women with well-documented pharma- by means of recruiting a bilateral cooperation between cologic status with respect to hormonal (and other) thera- prefrontal and hippocampal systems during verbal memory pies, all are middle age (between 50 and 65 years old), and tasks (Ottowitz et al., 2008). initiated hormone therapy perimenopausally. In addition, the Overall, basic science analyses using both in vitro and in detailed information obtained by self-report and recorded vivo model systems have also indicated that estrogen–—typi- from the medical records with respect to estrogen status–— cally 17b-estradiol but also in some instances conjugated both historical (ages at menarche and menopause, etc.) and equine estrogens–—protectneurons against insults associated current (specifics regarding types of estrogen used, and with AD (Brinton, 2005). Moreover, these same estrogens in whether they were progesterone-containing or unopposed, the same model systems can activate biochemical, genomic, etc.)–—allows for subgroup and correlational analyses of cellular and behavioral mechanisms of memory (Singh et al., cerebral metabolism that is generally not feasible in analysis 1994; Toran-Allerand, 2000; Frye et al., 2007). An important of brain PET data obtained in other investigations. These aspect of these studies, and of virtually all of the basic advantages are purchased at the cost of an inevitable corre- science in vitro and in vivo analyses, is that neurons were sponding disadvantage: an inherent limitation is that the healthy prior to estrogen exposure and prior to exposure to findings cannot be assumed to be generalizable beyond the neurodegenerative insults or lesions (Brinton, 2005). In particular kind of populations comprising our subject pool. In human studies, beyond the neuroimaging results discussed the current study, evidence is presented for relative pre- above, data on the effects of estrogen or postmenopausal servation of metabolism in specific brain regions in middle cognition have been consistent. Recent reviews of human age women at increased risk for future development of AD Differences in regional brain metabolism 511 being associated with greater endogenous estrogen expo- References sure, and unopposed or estradiol-based HT formulations initiated in the perimenopausal period. While these findings Archer, J.S., 1999. Relationship between estrogen, serotonin, and suggest a potential neuroprotective role for estrogen under depression. Menopause 6 (1), 71—78. these conditions, future studies including other subpopula- Arnold, A.P., Breedlove, S.M., 1985. Organizational and activational tions will be needed to examine the degree of generaliz- effects of sex steroids on brain and behavior: a reanalysis. Horm. ability of the findings reported here. Behav. 19 (4), 469—498. Barrett-Connor, E., Laughlin, G.A., 2009. Endogenous and exogenous estrogen, cognitive function, and dementia in postmenopausal Role of funding source women: evidence from epidemiologic studies and clinical trials. Semin. Reprod. Med. 27 (3), 275—282. Funding for this study was provided by a grant from the Bartholomeusz, C.F., Wesnes, K.A., Kulkarni, J., Vitetta, L., Croft, National Institute on Aging (R01 AG22008) and supported in R.J., Nathan, P.J., 2008. Estradiol treatment and its interaction part by grant M01 RR-00070 from the National Center for with the cholinergic system: effects on cognitive function in Research Resources, National Institutes of Health. The grant healthy young women. Horm. Behav. 54 (5), 684—693. did not have any further role in study design, collection, Bishop, J., Simpkins, J.W., 1995. Estradiol enhances brain glucose analysis, interpretation of the data, writing of the report, uptake in ovariectomized rats. Brain. Res. Bull. 36 (3), 315— and in the decision to submit the paper for publication. 320. Bora, S.H., Liu, Z., Kecojevic, A., Merchenthaler, I., Koliatsos, V.E., 2005. Direct, complex effects of estrogens on basal forebrain Contributors cholinergic neurons. Exp. Neurol. 194 (2), 506—522. Brinton, R.D., 2005. Investigative models for determining hormone Natalie Rasgon, Daniel Silverman, Heather Kenna, and Tonita therapy-induced outcomes in brain: evidence in support of a Wroolie designed the study; Natalie Rasgon, Heather Kenna, healthy cell bias of estrogen action. Ann. N. Y. Acad. Sci. Tonita Wroolie, and Bevin Powers acquired the data; Natalie 1052, 57—74. Dominguez, R., Jalali, C., de Lacalle, S., 2004. Morphological Rasgon, Heather Kenna, Bevin Powers, Tonita Wroolie, Cheri effects of estrogen on cholinergic neurons in vitro involves Geist, Daniel Silverman, and John Brooks analyzed and inter- activation of extracellular signal-regulated kinases. J. Neu- preted the data; Natalie Rasgon, Heather Kenna, Katherine rosci. 24 (4), 982—990. Williams, Cheri Geist, Daniel Silverman, and John Brooks Eberling, J.L., Reed, B.R., Coleman, J.E., Jagust, W.J., 2000. Effect critically revised the manuscript for intellectual content; of estrogen on cerebral glucose metabolism in postmenopausal Natalie Rasgon and Daniel Silverman obtained funding; Nata- women. Neurology 55 (6), 875—877. lie Rasgon, Heather Kenna, Katherine Williams, Daniel Silver- Eberling, J.L., Wu, C., Tong-Turnbeaugh, R., Jagust, W.J., 2004. man, and Tonita Wroolie provided supervision; Natalie Estrogen- and tamoxifen-associated effects on brain structure Rasgon, Heather Kenna, Katherine Williams, Bevin Powers, and function. Neuroimage 21 (1), 364—371. and John Brooks provided administrative, technical, or mate- Espeland, M.A., Rapp, S.R., Shumaker, S.A., Brunner, R., Manson, J.E., Sherwin, B.B., Hsia, J., Margolis, K.L., Hogan, P.E.,Wallace, rial support. Cheri Geist wrote the first draft of the manu- R., Dailey, M., Freeman, R., Hays, J., 2004. Conjugated equine script. All authors contributed to and have approved the final estrogens and global cognitive function in postmenopausal wom- manuscript. en: Women’s Health Initiative Memory Study. JAMA 291 (24), 2959—2968. Conflict of interest Fahn, S., Elton, R.L., 1987. The UPRDS development committee. In: Fahn, S., Marsden, C.D., Calne, D., Goldstein, M. (Eds.), Recent Developments in Parkinson’s Disease. Macmillan, New Jersey, 2, Authors Daniel H.S. Silverman, MD PhD, Cheri L. Geist, BS, pp. 153—163. Heather A Kenna, MA, Katherine Williams, MD, Tonita Wroo- Fletcher, P.C.,Shallice, T., Frith, C.D., Frackowiak, R.S., Dolan, R.J., lie, PhD, Bevin Powers, BA, and John Brooks, MD, PhD have no 1998. The functional roles of prefrontal cortex in episodic mem- disclosures or potential conflict of interest including any ory. II. Retrieval. Brain 121 (PT7), 1249—1256. financial, personal, or other relationships with other people Folstein, M.F., Folstein, S.E., McHugh, P.R., 1975. ‘‘Mini-mental or organizations within three years of beginning the work state’’. A practical method for grading the cognitive state of submitted that could inappropriately influence, or be per- patients for the clinician. J. Psychiatr. Res. 12 (3), 189—198. ceived to influence, their work. Friston, K., Ashburner, J., Frith, C., et al., 1995a. Spatial registra- Natalie Rasgon, MD, PhD, would like to disclose that she is tion and normalization of images. Hum. Brain Mapp. 2, 165—189. a consultant for Wyeth. She certifies that all of her affiliations Friston, K., Holmes, A., Worsley, K., et al., 1995b. Statistical with or financial involvement with any organization or entity parametric maps in functional imaging: a general linear approach. Hum. Brain Mapp. 2, 189—210. with a financial interest in or financial conflict with the Frye, C.A., Duffy, C.K., Walf, A.A., 2007. Estrogens and progestins subject matter or materials discussed in manuscript are enhance spatial learning of intact and ovariectomized rats in the completely disclosed above. object placement task. Neurobiol. Learn. Mem. 88 (2), 208—216. Acknowledgements Gabrieli, J.D., Brewer, J.B., Desmond, J.E., Glover, G.H., 1997. Separate neural bases of two fundamental memory processes in the human medial temporal lobe. Science 276 (5310), 264— This study was funded by a grant from the National Institute 266. on Aging (R01 AG22008) and supported in part by grant M01 Genazzani, A.R., Pluchino, N., Luisi, S., Luisi, M., 2007. Estrogen, RR-00070 from the National Center for Research Resources, cognition and female ageing. Hum. Reprod. Update 13 (2), National Institutes of Health. 175—187. 512 D.H.S. Silverman et al.

Gleason, C., Schmitz, T., Hess, T., et al., 2006. Hormone effects on Meusburger, S.M., Keast, J.R., 2001. Testosterone and nerve growth fMRI and cognitive measures of encoding: importance of hormone factor have distinct but interacting effects on structure and preparation. Neurology 67 (11), 2039—2041. neurotransmitter expression of adult pelvic ganglion cells in vitro. Halbreich, U., Rojansky, N., Palter, S., Tworek, H., Hissin, P., Wang, Neuroscience 108 (2), 331—340. K., 1995. Estrogen augments serotonergic activity in postmeno- Miller, V.M., Black, D.M., Brinton, E.A., Budoff, M.J., Cedars, M.I., pausal women. Biol. Psychiatr. 37 (7), 434—441. Hodis, H.N., Lobo, R.A., Manson, J.E., Merriam, G.R., Naftolin, F., Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neu- Santoro, N., Taylor, H.S., Harman, S.M., 2009. Using basic science rosurg. Psychiatr. 23, 56—62. to design a clinical trial: Baseline characteristics of women Ha¨nninen, T., Hallikainen, M., Koivisto, K., Helkala, E.L., Reinikai- enrolled in the Kronos Early Estrogen Prevention Study (KEEPS). nen, K.J., Soininen, H., Mykka¨nen, L., Laakso, M., Pyo¨ra¨la¨, K., J. Cardiovasc. Transl. Res. 2 (3), 228—239. Riekkinen Sr., P.J., 1995. A follow-up study of age-associated Moses, E.L., Drevets, W.C., Smith, G., Mathis, C.A., Kalro, B.N., memory impairment: neuropsychological predictors of dementia. Butters, M.A., Leondires, M.P., Greer, P.J., Lopresti, B., Loucks, J. Am. Geriatr. Soc. 43 (9), 1007—1015. T.L., Berga, S.L., 2000. Effects of estradiol and progesterone Henderson, V.W., Benke, K.S., Green, R.C., Cupples, L.A., Farrer, administration on human serotonin 2A receptor binding: a PET L.A., 2005. MIRAGE Study Group. Postmenopausal hormone ther- study. Biol. Psychiatr. 48 (8), 854—860. apy and Alzheimer’s disease risk: interaction with age. Am. J. Namba, H., Sokoloff, L., 1984. Acute administration of high doses of Psychiatr. 162 (4), 683—690. estrogen increases glucose utilization throughout brain. Brain Ibań˜ez, V., Pietrini, P., Alexander, G.E., Furey, M.L., Teichberg, D., Res. 291 (2), 391—394. Rajapakse, J.C., Rapoport, S.I., Schapiro, M.B., Horwitz, B., Nappi, R.E., Sinforiani, E., Mauri, M., Bono, G., Polatti, F., Nappi, G., 1998. Regional glucose metabolic abnormalities are not the 1999. Memory functioning at menopause: impact of age in ovari- result of atrophy in Alzheimer’s disease. Neurology 50 (6), ectomized women. Gynecol. Obstet. Invest. 47 (1), 29—36. 1585—1593. Nehlig, A., Porrino, L.G., Crane, A.M., Sokoloff, L., 1985. Local Joffe, H., Hall, J., Gruber, S., et al., 2006. Estrogen therapy cerebral glucose utilization in normal female rats: variations selectively enhances prefrontal cognitive processes: a random- during the estrous cycle and comparison with males. J. Cereb. ized, double-blind, placebo-controlled study with functional Blood Flow Metab. 5 (3), 393—400. magnetic resonance imaging in perimenopausal and recently Ottowitz, W.E., Siedlecki, K.L., Lindquist, M.A., Dougherty, D.D., postmenopausal women. Menopause 13 (3), 411—422. Fischman, A.J., Hall, J.E., 2008. Evaluation of prefrontal-hip- Kendall, D.A., Stancel, G.M., Enna, S.J., 1981. Imipramine: effect of pocampal effective connectivity following 24 h of estrogen ovarian steroids on modifications in serotonin receptor binding. infusion: an FDG-PET study. Psychoneuroendocrinology 33 Science 211 (4487), 1183—1185. (10), 1419—1425. Kompoliti, K., Chu, Y., Polish, A., Roberts, J., McKay, H., Mufson, Persad, C.C., Zubieta, J.K., Love, T., Wang, H., Tkaczyk, A., Smith, E.J., Leurgans, S., Morrison, J.H., Kordower, J.H., 2004. Effects Y.R., 2009. Enhanced neuroactivation during verbal memory of estrogen replacement therapy on cholinergic basal forebrain processing in postmenopausal women receiving short-term hor- neurons and cortical cholinergic innervation in young and aged mone therapy. Menopause 92 (1), 197—204. ovariectomized rhesus monkeys. J. Comp. Neurol. 472 (2), 193— Phillips, S.M., Sherwin, B.B., 1992. Effects of estrogen on memory 207. function in surgically menopausal women. Psychoneuroendocri- Lasiuk, G.C., Hegadoren, K.M., 2007. The effect of estradiol on nology 17 (5), 485—495. central serotonergic systems and its relationship to mood in Ping, S.E., Trieu, J., Wlodek, M.E., Barrett, G.L., 2008. Effects of women. Biol. Res. Nurs. 9 (2), 147—160. estrogen on basal forebrain cholinergic neurons and spatial learn- Luine, V.N., 1985. Estradiol increases choline acetyltransferase ac- ing. J. Neurosci. Res. 86 (7), 1588—1598. tivity in specific basal forebrain nuclei and projection areas of Polkowski, K., Mazurek, A.P.,2000. Biological properties of genistein. female rats. Exp. Neurol. 89 (2), 484—490. A review of in vitro and in vivo data. Acta. Pol. Pharm. 57 (2), Maki, P.M., Gast, M.J., Vieweg, A.J., Burriss, S.W., Yaffe, K., 2007. 135—155. Hormone therapy in menopausal women with cognitive com- Rapp, S.R., Espeland, M.A., Shumaker, S.A., Henderson, V.W., Brun- plaints: a randomized double-blind trial. Neurology 69 (13), ner, R.L., Manson, J.E., Gass, M.L., Stefanick, M.L., Lane, D.S., 1322—1330. Hays, J., Johnson, K.C., Coker, L.H., Dailey, M., Bowen, D., 2003. Maki, P.M., Resnick, S.M., 2000. Longitudinal effects of estrogen Effect of estrogen plus progestin on global cognitive function in replacement therapy on PET cerebral blood flow and cognition. postmenopausal women: The Women’s Health Initiative Memory Neurobiol. Aging. 21 (2), 373—383. Study: a randomized controlled trial. JAMA 289 (20), 2663—2672. Maki, P.M., Resnick, S.M., 2001. Effects of estrogen on patterns of Rasgon, N.L., Geist, C.L., Kenna, H., Powers, B., Wroolie, T.E., brain activity at rest and during cognitive activity: a review of Silverman, D.H., 2008a. Decline of inferior parietal brain metab- neuroimaging studies. Neuroimage 14 (4), 789—801. olism in postmenopausal women who discontinue hormone ther- Maki, P.M.,2006. Hormone therapy and cognitive function: is there a apy. J. Nucl. Med. Meeting Abstracts 49, 226. critical period for benefit? Neuroscience 138 (3), 1027—1030. Rasgon, N.L., Kenna, H.A., Geist, C., Small, G., Silverman, D., 2008b. Martin, A., Fedio, P., 1983. Word production and comprehension in Cerebral metabolic patterns in untreated postmenopausal Alzheimer’s disease: the breakdown of semantic knowledge. women with major depressive disorder. Psychiatr. Res. 164 (1), Brain Lang. 19 (1), 124—141. 77—80. McEwen, B.S., Coirini, H., Danielsson, A., Frankfurt, M., Gould, E., Rasgon, N.L., Silverman, D., Siddarth, P., Miller, K., Ercoli, L.M., Mendelson, S., Schumacher, M., Segarra, A., Woolley, C., 1991. Elman, S., Lavretsky, H., Huang, S.C., Phelps, M.E., Small, G.W., Steroid and thyroid hormones modulate a changing brain. J. 2005. Estrogen use and brain metabolic change in postmenopaus- Steroid Biochem. Mol. Biol. 40 (1—3), 1—14. al women. Neurobiol. Aging 26 (2), 229—235. McEwen, B.S., 1981. Neural estrogen receptors in the life cycle of the Rasgon, N.L., Small, G.W., Siddarth, P., Miller, K., Ercoli, L.M., albino rat. Exp. Brain Res. Suppl. 3, 61—79. Bookheimer, S.Y.,Lavretsky, H., Huang, S.C., Barrio, J.R., Phelps, McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., M.E., 2001. Estrogen use and brain metabolic change in older Stadlan, E.M., 1984. Clinical diagnosis of Alzheimer’s disease: adults: A preliminary report. Psychiatr. Res. 107 (1), 11—18. report of the NINCDS-ADRDA Work Group under the auspices of Reiman, E.M., Caselli, R.J., Yun, L.S., Chen, K., Bandy, D., Min- Department of Health and Human Services Task Force on Alzhei- oshima, S., Thibodeau, S.N., Osborne, D., 1996. Preclinical evi- mer’s Disease. Neurology 34 (7), 939—944. dence of Alzheimer’s disease in persons homozygous for the Differences in regional brain metabolism 513

epsilon 4 allele for apolipoprotein E. N. Engl. J. Med. 334 (12), Small, G.W., Kuhl, D.E., Riege, W.H., Fujikawa, D.G., Ashford, J.W., 752—758. Metter, E.J., Mazziotta, J.C., 1989. Cerebral glucose metabolic Resnick, S., Maki, P., Golski, S., Kraut, M., Zonderman, A., 1998. patterns in Alzheimer’s disease. Effect of gender and age at Effects of estrogen replacement therapy on PET cerebral blood dementia onset. Arch. Gen. Psychiatr. 46 (6), 527—532. flow and neuropsychological performance. Horm. Behav. 34 (2), Small, G.W., La Rue, A., Komo, S., Kaplan, A., Mandelkern, M.A., 171—182. 1995. Predictors of cognitive change in middle-aged and older Resnick, S.M., Espeland, M.A., An, Y., Maki, P.M., Coker, L.H., adults with memory loss. Am. J. Psychiatr. 152 (12), 1757—1764. Jackson, R., Stefanick, M.L., Wallace, R., Rapp, S.R., 2009a. Smith, G.S., de Leon, M.J., George, A.E., Kluger, A., Volkow, N.D., Women’s Health Initiative Study of Cognitive Aging Investigators. McRae, T., Golomb, J., Ferris, S.H., Reisberg, B., Ciaravino, J., Effects of conjugated equine estrogens on cognition and affect in et al., 1992. Topography of cross-sectional and longitudinal glu- postmenopausal women with prior hysterectomy. J Clin. Endo- cose metabolic deficits in Alzheimer’s disease. Pathophysiologic crinol. Metab. 94 (11), 4152—4161. implications. Arch. Neurol. 49 (11), 1142—1150. Resnick, S.M., Espeland, M.A., Jaramillo, S.A., Hirsch, C., Stefanick, Smith, Y., Love, T., Persad, C., Tkaczyk, A., Nichols, T., Zubieta, J., M.L., Murray, A.M., Ockene, J., Davatzikos, C., 2009b. Postmen- 2006. Impact of combined estradiol and norethindrone therapy on opausal hormone therapy and regional brain volumes: the WHIMS- visuospatial working memory assessed by functional magnetic MRI study. Neurology 72 (2), 135—142. resonance imaging. J. Clin. Endocrinol. Metab. 91 (11), 4476— Resnick, S.M., Maki, P.M., Rapp, S.R., Espeland, M.A., Brunner, R., 4481. Coker, L.H., Granek, I.A., Hogan, P., Ockene, J.K., Shumaker, Smith, Y.R.,Minoshima, S., Kuhl, D.E., Zubieta, J.K., 2001. Effects of S.A., 2006. Women’s health initiative study of cognitive aging long-term hormone therapy on cholinergic synaptic concentra- investigators. Effects of combination estrogen plus progestin tions in healthy postmenopausal women. J. Clin. Endocrinol. hormone treatment on cognition and affect. J. Clin. Endocrinol. Metab. 86 (2), 679—684. Metab. 91 (5), 1802—1810. Squire, L.R., Ojemann, J.G., Miezin, F.M., Petersen, S.E., Videen, Rocca, W.A., Bower,J.H., Maraganore, D.M., Ahlskog, J.E., Grossardt, T.O., Raichle, M.E., 1992. Activation of the hippocampus in B.R., de Andrade, M., Melton, L.J., 2007. Increased risk of normal humans: a function anatomical study of memory. Proc. cognitive impairment or dementia in women who underwent Natl. Acad. Sci. U. S. A. 89 (5), 1837—1841. oophorectomy before menopause. Neurology 69 (11), 1074—1083. Stern, C.D., Corkin, S., Gonzalez, R.G., Guimaraes, A.R., Baker,J.R., Roy, E.J., Buyer, D.R., Licari, V.A., 1990. Estradiol in the striatum: Jennings, P.J., Carr, C.A., Sugiura, R.M., Vedantham, V., Rosen, effects on behavior and dopamine receptors but no evidence for B.R., 1996. The hippocampal formation participates in novel membrane steroid receptors. Brain Res. Bull. 25 (2), 221—227. picture encoding: evidence from functional magnetic resonance Rubinow, D.R., Schmidt, P.J., Roca, C.A., 1998. Estogen-serotonin imaging. Proc. Natl. Acad. Sci. U. S. A. 93 (16), 8660— interactions: implications for affective regulation. Biol. Psy- 8665. chiatr. 44 (9), 839—850. Tai, Y.C., Lin, K.P., Hoh, C.K., Huang, S.C., Hoffman, E.J., 1997. Sar, M., Stumpf, W.E., 1981. Central noradrenergic neurons concen- Utilization of 3-D elastic transformation in registration of chest X- trate 3H-oestradiol. Nature 289 (5797), 500—502. ray CT and whole body PET. IEEE Trans. Nucl. Sci. 44, 1606— Shallice, T., Fletcher, P., Frith, C.D., Grasby, P., Frackowiak, R.S., 1612. Dolan, R.J., 1994. Brain regions associated with acquisition and Toran-Allerand, C.D., 2000. Novel sites and mechanisms of oestrogen retrieval of verbal episodic memory. Nature 368 (6472), 633—635. action in the brain. Novartis Fond. Symp. 230, 56—69 discussion Shaywitz, S., Shaywitz, B., Pugh, K., et al., 1999. Effect of estrogen 69—73. on brain activation patterns in postmenopausal women during Tulving,E., Markowitsch, H.J., Craik, F.E.,Habib, R., Houle, S., 1996. working memory tasks. JAMA 281 (13), 1197—1202. Novelty and familiarity activations in PET studies of memory Shumaker, S.A., Legault, C., Kuller, L., Rapp, S.R., Thal, L., Lane, encoding and retrieval. Cereb. Cortex 6 (1), 71—79. D.S., Fillit, H., Stefanisk, M.L., Hendrix, S.L., Lewis, C.E., Masaki, Ungar, S., Makman, M.H., Morris, S.A., Etgen, A.M., 1993. Estrogen K., Coker, L.H., 2004. Women’s Health Initiative Memory Study. uncouples beta-adrenergic receptor from the stimulatory guanine Conjugated equine estrogens and incidence of probably dementia nucleotide-binding protein in female rat hypothyalamus. Endo- and mild cognitive impairment in postmenopausal women: crinology 133 (6), 2818—2826. Women’s Health Initiative Memory Study. JAMA 291 (24), Verghese, J., Kuslansky, G., Katz, M.J., Sliwinski, M., Crystal, H.A., 2947—2958. Bushke, H., Lipton, R.B., 2000. Cognitive performance in surgi- Shumaker, S.A., Legault, C., Thal, L., Wallace, R.B., Ockene, J.K., cally menopausal women on estrogen. Neurology 55 (6), 872— Hendrix, S.L., Jones 3rd, B.N., Assaf, A.R., Jackson, R.D., Morley 874. Kotchen, J., Wassertheil-Smoller,S., Wactawski-Wende, J., 2003. Vincent, A., Fitzpatrick, L.A., 2000. Soy isoflavones: are they useful Estrogen plus progestin and the incidence of dementia and mild in menopause. Mayo Clin. Proc. 75 (11), 1174—1184. cognitive impairment in postmenopausal women: The Women’s Wang, G., Drake, C.T., Rozenblit, M., Zhou, P., Alves, S.E., Herrick, Health Initiative Memory Study: a randomized controlled trial. S.P., Hayashi, S., Warrier, S., Iadecola, C., Milner, T.A., 2006. JAMA 289 (20), 2651—2662. Evidence that estrogen directly and indirectly modulates C1 Silverman, D.H., Dy, C.J., Castellon, S.A., Lai, J., Pio, B.S., Abraham, adrenergic bulbospinal neurons in the rostral ventrolateral me- L., Waddell, K., Petersen, L., Phelps, M.E., Ganz, P.A., 2007. dulla. Brain Res. 1094 (1), 163—178. Altered frontocortical, cerebellar, and basal ganglia activity in Wiggs, C.L., Weisberg, J., Martin, A., 1999. Neural correlates of adjuvant-treated breast cancer survivors 5—10 years after che- semantic and episodic memory retrieval. Neuropsychologia 37 motherapy. Breast Cancer Res. Treat. 103 (3), 303—311. (1), 103—118. Singh, M., Meyer, E.M., Millard, W.J., Simkins, J.W., 1994. Ovarian Woolley, C.S., Gould, E., Frankfurt, M., McEwen, B.S., 1990. Natu- steroid deprivation results in a reversible learning impairment rally occurring fluctuation in dendritic spine density on adult and compromised cholinergic function in female Sprague-Dawley hippocampal pyramidal neurons. J. Neurosci. 10 (12), 4035— rats. Brain Res. 644 (2), 305—312. 4039. Frontiers in Neuroendocrinology 30 (2009) 239–258

Contents lists available at ScienceDirect

Frontiers in Neuroendocrinology

journal homepage: www.elsevier.com/locate/yfrne

Review Protective actions of sex steroid hormones in Alzheimer’s disease

Christian J. Pike a,*, Jenna C. Carroll a,b, Emily R. Rosario a, Anna M. Barron a a Davis School of Gerontology, University of Southern California, Los Angeles, CA 90089, USA b Neuroscience Graduate Program, University of Southern California, Los Angeles, CA 90089, USA article info abstract

Article history: Risk for Alzheimer’s disease (AD) is associated with age-related loss of sex steroid hormones in both Available online 7 May 2009 women and men. In post-menopausal women, the precipitous depletion of estrogens and progestogens is hypothesized to increase susceptibility to AD pathogenesis, a concept largely supported by epidemio- Keywords: logical evidence but refuted by some clinical findings. Experimental evidence suggests that estrogens Estrogen have numerous neuroprotective actions relevant to prevention of AD, in particular promotion of neuron Testosterone viability and reduction of b-amyloid accumulation, a critical factor in the initiation and progression of AD. Progesterone Recent findings suggest neural responsiveness to estrogen can diminish with age, reducing neuroprotec- Alzheimer’s disease tive actions of estrogen and, consequently, potentially limiting the utility of hormone therapies in aged b-Amyloid Neuroprotection women. In addition, estrogen neuroprotective actions are also modulated by progestogens. Specifically, continuous progestogen exposure is associated with inhibition of estrogen actions whereas cyclic delivery of progestogens may enhance neural benefits of estrogen. In recent years, emerging literature has begun to elucidate a parallel relationship of sex steroid hormones and AD risk in men. Normal age-related testosterone loss in men is associated with increased risk to several diseases including AD. Like estrogen, testosterone has been established as an endogenous neuroprotective factor that not only increases neuronal resilience against AD-related insults, but also reduces b-amyloid accumulation. Androgen neuroprotective effects are mediated both directly by activation of androgen pathways and indirectly by aromatization to estradiol and initiation of protective estrogen signaling mechanisms. The successful use of hormone therapies in aging men and women to delay, prevent, and or treat AD will require additional research to optimize key parameters of hormone therapy and may benefit from the continuing development of selective estrogen and androgen receptor modulators. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction protein tau that forms lesions called neurofibrillary tangles and neuropil threads, glial activation which is associated with inflam- As expertly described in the accompanying review articles, sex matory responses, and both synaptic and neuronal loss steroid hormones are potent regulators of neuron survival in mul- [3,37,139,140,196]. Although the mechanisms of AD pathogenesis tiple CNS regions and across a variety of circumstances ranging remain to be fully resolved, the leading hypothesis posits that from normal development to neural injury. A compelling, and as the disease is initiated and driven by prolonged elevation of Ab lev- yet largely unrealized, promise of sex steroid hormones is the els [140].Ab is a proteolytic byproduct of the metabolism of amy- translation of their neuroprotective properties into efficacious loid precursor protein, a widely expressed protein with numerous strategies for the treatment and or prevention of age-related neu- functions ranging from axonal transport to gene transcription rodegenerative disorders such as Alzheimer’s disease (AD). Despite [336]. As a consequence of amyloid precursor protein expression, this unfulfilled therapeutic potential, abundant experimental, epi- Ab is normally found as a soluble protein at low levels in fluids demiological and clinical evidence suggest that neural actions and tissues throughout the body. In theory, alterations in either androgens, estrogens, and perhaps even progestogens can reduce the production or clearance of Ab that sway Ab homeostasis to- the risk for AD. wards increased neural levels will promote the development of AD is an age-related neurological disease that is the leading AD [140]. The accumulation of Ab encourages its abnormal assem- cause of dementia. Neuropathologically, AD is characterized by bly into oligomeric species that exhibit an altered structural con- brain region-specific deposition of b-amyloid protein (Ab) which formation and can induce a range of neurodegenerative effects creates senile plaques, hyperphosphorylation of the cytoskeletal [136]. Consequently, enormous effort has been expended on iden- tifying factors that regulate Ab accumulation and or affect its neu- * Corresponding author. Fax: +1 213 740 4787. rodegenerative properties. One such class of factors is sex steroid E-mail address: [email protected] (C.J. Pike). hormones.

In this review, we will discuss the neuroprotective properties of 17b-estradiol (E2) [212] are reported to be lower in women with sex steroid hormones as they relate to AD pathogenesis, focusing AD in comparison to age-matched controls. If the depletion of largely on their effects on Ab accumulation and its associated neu- ovarian hormones at menopause contributes to women’s increased rodegeneration. Estrogens are the most thoroughly studied steroid risk of AD, then one would predict that estrogen-based hormone hormones in terms of AD. We will cover the epidemiological and therapy (HT) would be effective in the prevention and or treatment clinical evidence that suggests depletion of ovarian hormones at of AD. This critical issue remains unresolved with persuasive argu- menopause increases the risk of AD in post-menopausal women, ments both for and against the use of HT for AD. An early study of a danger some studies suggest may be mitigated by estrogen-based this issue found that AD risk was lower in women who used HT rel- hormone therapy (HT). Consistent with a protective role against ative to nonusers and this risk decreased significantly as both dose AD, experimental studies demonstrate that estrogens not only and duration of HT use increased [247]. Similarly, findings from reduce neuron loss induced by AD-related insults but also act to several other case control and prospective studies suggest that reduce Ab levels. However, in the Women’s Health Initiative post-menopausal women with a history of HT use were at reduced (WHI) trial, the most exhaustive clinical evaluation of HT thus risk of AD [52,151,184,248,329,351,382]. Further, a meta-analysis far, HT was associated with increased rather than decreased risk of studies found that HT was associated with decreased risk of cog- of AD. Analysis of experimental work reveals several key limita- nitive dysfunction [158,197,241,305]. Collectively, these studies tions of estrogen’s neuroprotective actions that may contribute to suggest a potential protective role of estrogen against the develop- this clinical observation, including loss of neural estrogen respon- ment of AD. siveness with age and interactions with progestogens that may Despite indications of benefits, the potential protective of HT limit estrogen neuroprotection. A more recent and still emerging against AD remains controversial. Arguing against a protective role, literature suggests a parallel relationship in men with their several studies found that HT use was not associated with reduced primary sex steroid hormone, testosterone. That is, normal age-re- risk of AD (reviewed in [146]) or failed to yield significant cognitive lated testosterone is associated with increased risk of AD in aging benefits [7,22,34,40,119,152,228,266,349,369]. One possible expla- men. Like estrogens, androgens also exert neuroprotective proper- nation for this discrepancy is suggested by findings from the Cache ties relevant to AD, including promotion of survival in neurons County Study, which demonstrated that the association between challenged with AD-related insults and reduction of Ab levels. HT use and reduced risk of AD was strengthened in long-term HT Finally, we consider future directions in this field, emphasizing users [382]. Interpretations of these findings include the concept the clinical potential of sex steroid hormones in prevention rather that HT may have a largely preventive role against AD and or the than treatment of AD and the emerging promise of selective estro- hypothesis that early initiation of HT is essential as women who gen receptor and androgen receptor modulators. took HT for longer periods likely began treatment nearer the time of menopause. The notion that estrogen-based HT can effectively reduce the 2. Menopause, hormone therapy, and Alzheimer’s disease risk of AD and improve age-related deficits in cognition has been challenged by findings from the Women’s Health Initiative Mem- Converging lines of evidence indicate a potentially important role ory Study (WHIMS). WHIMS was a randomized, multi-center, dou- of estrogens in regulating AD pathogenesis. Preliminary clues sug- ble-blind, placebo-controlled study of 4500 women between 65 gesting this possibility stemmed from reports of sex differences in and 79 yrs of age that evaluated effects of HT consisting of conju- AD risk, with women showing higher prevalence and incidence. gated equine estrogen (CEE) alone or CEE with the progestin Although sex differences in AD are difficult to interpret due to gender medroxyprogesterone acetate (MPA). This study reported that nei- differences in life expectancy, many studies of various cohorts ther CEE nor CEE + MPA significantly improved cognition versus indicate that women are at greater risk of AD [9,17,39,93,101, placebo in women showing cognitive decline associated with nor- 137,180,221,281,290]. Further, there is some evidence that AD path- mal aging [91,273] or dementia [312,311]. In the CEE alone arm, ogenesis may be more severe in women as indicated by sex differ- there was no significant difference in dementia incidence between ences in cognitive deficits and neuropathology [21,49,66,150], HT or placebo groups although there was a non-significant trend although other studies indicate men have higher levels of tau pathol- towards increased risk in the HT group [91,312]. In the CEE + MPA ogy [293,294]. Further, there is a stronger association between the arm, they found that women receiving HT had a higher risk of apolipoprotein E e4 allele and sporadic AD in women compared to probable dementia [311]. HT use was also associated with in- men [66,84,165] and the e4 allele has been shown to be associated creased incidence of stroke and breast cancer, suggesting that the with greater hippocampal atrophy and memory impairments in wo- risks of HT may outweigh its benefits. Further, the important dif- men compared to men [98]. When considered together, these epide- ferences between the CEE alone arm and the CEE + MPA arm raise miological and neuropathological studies indicate sex differences in several issues about the inclusion of a progestin component. AD, suggesting that women may be more vulnerable to AD than men. Although the WHIMS findings raise serious concerns over HT Several transgenic mouse models exhibit sex differences in use, many challenge the interpretation that these data dismiss AD-like neuropathology that appears to parallel that observed in the potential efficacy of HT in reducing the risk of AD human AD cases. For example, at both 15 and 19 mo of age, female [75,118,149,209,265,276]. A variety of issues have been identified Tg2576 mice display a higher plaque load burden and higher levels that may have affected HT outcomes in the WHIMS compared to of both soluble and insoluble Ab40 and Ab42 than age-matched the many observational studies such as differences in methodolog- males [51]. Similarly, female APPswexPS1 transgenic mice have ical techniques, outcome measures, hormone exposures, meno- higher Ab load burden and plaque number than age-matched pausal symptoms, and the timing of hormone use [149].In males [350]. The same pattern of greater Ab deposition in female particular, neural sensitivity to sex steroid hormones may diminish versus male mice is also observed in the 3xTg-AD triple transgenic during the menopause transition, resulting in a critical window in mouse [156]. These studies in transgenic mouse models of AD sug- which to initiate HT in order to realize benefits (reviewed in [75]). gest that the female brain may be more vulnerable to AD Because HT was initiated many years after menopause in the pathogenesis. WHIMS study, the study’s design may have inadvertently focused The increased risk of AD in women is presumed to be associated on an age group in which estrogen is minimally active in brain with the precipitous loss of estrogens and progesterone at meno- and thus was unlikely to detect potential cognitive benefits. In pause. Consistent with this position, plasma levels of the estrogen addition, as suggested by prior epidemiological and clinical find- C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 241

ings, estrogen may be most effective in preventing rather than [388] and recently it has been shown that E2 can also increase treating AD. In this case, the relatively advanced age of WHIMS Bcl-2 through Akt-dependent CREB activation [381]. subjects would also be biased against positive outcomes. Addi- Interestingly, estrogen dependent regulation of the Bcl-2 family tional issues include HT formulation, route of administration, and of proteins has also been implicated in neuroprotection against treatment regime (e.g., cyclic versus continuous hormone deliv- excitotoxicity, a form of neuronal injury implicated in AD neurode- ery). In order to unravel this conundrum, a greater understanding generation (reviewed in [29,280]). Evidence suggests that in AD, Ab of the neuroprotective actions of estrogens and progestogens are toxicity and glutamate excitotoxicity may cooperatively activate needed, as well as their limitations in the context of aging. pathways leading neuronal death. Glutamate-induced excitotoxic In the following sections, we examine the neuroprotective ef- injury is potentiated by Ab. In cell culture paradigms, the combina- fects of estrogen, focusing on its abilities to increase neuronal resil- tion of sub-lethal concentrations of glutamate combined with ience against AD insults and to antagonize AD pathogenesis by sub-lethal levels Ab yields robust neuronal loss [189,215]. Such reducing Ab accumulation. Importantly, we also discuss experi- degenerative interactions are predicted to occur in AD because mental evidence that addresses how these protective actions of the AD brain exhibits both Ab accumulation and evidence of gluta- estrogen are affected by the concerns raised by WHIMS. mate injury. Whether upstream and or downstream pathways of Ab and glutamate action are responsible for their synergistic toxic effects is unclear. Glutamate excitotoxicity leads to calcium dys- 3. Estrogen neuroprotection and Alzheimer’s disease insults regulation and oxidative stress, and excitotoxic neuron death is mediated in part by apoptosis (reviewed in [64]). An established neural action of estrogens that may contribute to Several studies have demonstrated that estrogen reduces excito- a protective role against AD is promotion of neuron viability. Estro- toxic neuronal death induced by glutamate agonists in cell culture gen is neuroprotective against a variety of insults in several cell [43,236,275,317,315,319]. For example, Dorsa and colleagues found culture and rodent paradigms of injury and neurodegenerative dis- that estrogen inhibited neuronal death in murine cortical cultures ease. Of particular interest is the ability of estrogen to protect following excitotoxic insult, an effect that could be pharmacologi- against neuronal loss induced by Ab, which is thought to be pri- cally blocked with the ER antagonist, tamoxifen [315]. Further, Brin- mary neurodegenerative agent in AD. Reports from several groups ton and collegues found estrogen to promote intracellular Ca2+ demonstrate that estrogen can protect cultured neurons and neu- accumulation in neuronal cultures treated glutamate at physiologi- ral cell lines from Ab mediated toxicity [26,124,132,135,222,259]. cal doses, while inhibiting intracellular Ca2+ accumulation following Estrogen may potentially protect against Ab-induced neurotox- treatment with excitotoxic glutamate doses [236]. Similar observa- icity at several steps in the degenerative process. The leading the- tions of estrogen neuroprotection following excitotoxic challenge ory of Ab toxicity posits a pathologic assembly of Ab involving have been reported by several groups [124,275,354]. adoption of a b-sheet conformation, resulting in a change in pro- Estrogen has also been shown to regulate the extent of excito- tein structure that is associated with a toxic gain of function (re- toxic injury in rodent models [15,14,55,286]. For example, admin- viewed in [347]). Consistent with this working hypothesis of Ab istration of exogenous E2 to ovariectomized (OVX) rats has been neurotoxicity, our prior work has shown that Ab is toxic only in reported to protect against kainate-induced neuronal loss [14]. an assembled state [263,261]. Assembled Ab in the form of soluble Additionally, depletion of endogenous estrogen levels may in- oligomers and insoluble fibrils can induce neuronal death crease susceptibility to excitotoxicity, with pronounced kainate-in- [72,348,373] degeneration of neurites [169,262] and synaptic dys- duced neural loss observed in intact rats during proestrus or ruption [145,182,288,348] leading to impaired learning and mem- following OVX [15]. Similarly, we observe neuroprotection against ory [61,198]. Interaction of Ab assemblies with neurons initiates a kainate lesion following administration of estrogen to OVX rats cascade of upstream signaling mechanisms associated with cell [55,286]. death, including calcium dysregulation [216,356], oxidative stress Interestingly, estrogen neuroprotection against excitotoxic in- [25,123,214], and activation of pro-inflammatory pathways pro- jury shares mechanistic similarities with protection against Ab-in- moting chronic gliosis [89,264]. Most evidence suggests that the duced apoptosis. That is, estrogen regulation of the Bcl-2 family is plethora of upstream signaling cascades elicited by Ab ultimately implicated in protective actions against glutamate-related injury mediate neurotoxicity by downstream activation of neuronal [232,234,316,388]. Brinton and colleagues found that estrogen apoptosis pathways [71,73]. In particular, we find that Ab-induced mediated neuroprotection against glutamate excitotoxicity by pro- neuronal apoptosis involves activation of JNK signaling and conse- moting mitochondrial Ca2+ sequestration, and this was associated quent dysregulation of the Bcl-2 family of apoptosis-related pro- with increased expression of Bcl-2 [232]. Estrogen dependent mod- teins [374]. ulation Bcl-2 following excitotoxicity may be mediated by rapid, Since neuronal apoptosis is an important downstream mediator non-genomic ER-dependent signaling mechanisms [387,388]. of Ab neurotoxicity, regulation of apoptosis is predicted to be a key Estrogen activates an ER-dependent Src/ERK/CREB signaling path- mechanism of estrogen protection from Ab. Consistent with this way that leads to upreguation of Bcl-2 [364]. In contrast, others hypothesis, estrogen has been implicated in the regulation of Bcl- suggest that estrogen may regulate Bcl-2 family expression 2 family members in neurons [86,110,191,234,254,259,316,320]. through a direct genomic mechanism. Supporting this we have de- The Bcl-2 family includes both proteins that promote cell survival scribed an estrogen responsive element (ERE) on the Bcl-x gene

(e.g., Bcl-2, Bcl-xL, and Bcl-w) and others that antagonize it (e.g., [259], while others describe EREs on Bcl-2 [83,256]. These path- Bax, Bad, Bak, Bik, Bid, BNIP3, and Bim) (reviewed in [12,70]). We ways are summarized in Fig. 1. have found that physiological levels of E2 inhibit neuronal apopto- In addition to regulation of the Bcl-2 family of proteins, estrogen sis at least in part by increasing expression of anti-apoptotic Bcl-xL has also been implicated in neuroprotection against many other [259,320] and Bcl-w [375] while down-regulating expression of prominent features of the AD-neurodegenerative cascade including the pro-apoptotic Bim [375]. The observed effects of E2 on Ab inﬂammation and oxidative stress. Many studies indicate that Ab mediated apoptosis were found to be mediated ER-dependent may be contribute to AD-related oxidative stress [25,123,214] and mechanisms, since the anti-apoptotic effects of E2 were blocked deposition of Ab may activate microglia and promote inﬂammation by pre-treatment with an ER antagonist [259, 375]. Supporting [291,338]. Abundant evidence indicates that estrogen is a potent this, it has been previously demonstrated that both ERa and ERb inhibitor of oxidative damage [340] and hydrogen peroxide medi- are crucial in regulating Bcl-2 expression and neuronal survival ated neuronal death [27,26,124]. Estrogen mediates these antioxi- 242 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258

Fig. 1. Estrogens activate neuroprotective pathways that may attenuate Alzheimer’s disease. Estrogens including 17b-estradiol (E2) reduce neuronal apoptosis by (i) non- genomic signaling cascades, including activation of PI3 K, protein kinase C (PKC) and Src/ERK pathways, and (ii) genomic pathways utilizing CREB response elements (CRE) and estrogen response elements (ERE) on members of the Bcl-2 family of genes including Bcl-2, Bcl-x, Bcl-w, and Bim. Similarly, estrogens decrease levels of the AD-related protein Ab by (i) non-genomic signaling that promotes non-amyloidogenic processing of the Ab precursor protein (APP), and perhaps (ii) classic genomic mechanisms that may involve ERE, CRE, and or other steroid response elements (SRE) on the Ab-catabolizing enzymes neprilysin (NEP) and insulin degrading enzyme (IDE).

dant effects through ER-independent mechanisms, acting as a free affected in AD such as the hippocampus, frontal cortex, and amyg- radical scavenger owing to its phenolic structure [279,278,324, dala [144,310,309]. Recently, changes in the subcellular distribu- 325]. However, these neuroprotective effects are only elicited at tion of ERs in hippocampal neurons have been implicated in AD supraphysiological estrogen dosages [27,202], thereby limiting the pathogenesis [205]. Specifically, the shift of ERa from the nucleus clinical relevance of these neuroprotective actions of estrogens. In to the cytoplasm may decrease the development of AD pathology contrast, the anti-inflammatory actions of estrogen are mediated in humans [155] and transgenic mice [181]. In addition, recent by ER-dependent mechanisms and may provide preventative or studies have demonstrated that ERa levels in the frontal cortex therapeutic benefit to a range of neurodegenerative diseases where correlated with mini-mental state examination scores in women neuroinflammation is a major degenerative process. Pluripotent ef- with end-stage AD [185] and that specific allele differences in fects of estrogen have been described on glial function and activa- ERa are correlated with an increased risk for AD in women with tion. Estrogen may suppress reactive gliotic responses associated Down syndrome [296]. Also, ERb immunoreactivity is reportedly with traumatic injury and neurodegeneration [109,111], while also increased in the hippocampus compared to age-matched controls promoting the neuroprotective properties of astrocytes by increas- [292]. Taken together, these studies suggest that the expression ing arborization and promoting synaptogenesis [80,323]. Interest- of ERa/b in the AD brain may play an integral role in the neuropro- ingly, estrogen has been found to attenuate microglial activation tective actions of E2. These effects are discussed in several recent only when estrogen treatment was given to the cultures prior to reviews [38,240]. inflammatory insult, indicating estrogen does not have the capacity Both ER subtypes ERa and ERb are implicated in mediating to modulate inflammatory reactions once microglial activation has estrogen neuroprotection, although their relative contributions been initiated [341]. This suggests the anti-inflammatory benefits have been incompletely defined. Selective expression of ERa versus of estrogen may be limited to prevention of AD rather than ERb in neural cell lines has suggested a more important role of ERa treatment. in mediating neuroprotection in some studies [187] but significant Another potentially significant neuroprotective action of estro- contributions from both ERa and ERb in others [97,203,218]. Cell gen that is highly relevant to AD and related neurodegenerative culture studies utilizing selective ERa (propylpyrazole triol, PPT), disorders is inhibition of pathological tau hyperphosphorylation. and ERb agonists (diarylpropionitrile; DPN) to study estrogen neu- Both estrogen and progesterone can modulate activities of kinases roprotection typically report similar levels of protection from both and phosphatases involved in regulating levels of tau phosphoryla- agonists, although some evidence suggests greater activity of the tion. Specifically, E2 and P4 regulate tau phosphorylation through ERa agonist PPT [28]. Our studies in primary neuron culture indi- the glycogen synthase kinase-3b (GSK-3b) pathway [8,122]. Estro- cate comparable levels of neuroprotection against Ab from E2, gen can reduce GSK-3b activity [122], and progesterone can de- PPT, and DPN, suggesting potential contributions from both ERa crease expression of both tau and GSK-3b. Estrogen can also and ERb in estrogen neuroprotection [67]. However, our data also lower tau hyperphosphorylation through the wnt signaling path- suggested potential differences between ER subtypes in terms of way and a gene called dickkopf-1 [386] as well as through the pro- protective mechanisms, with PPT but not DPN inducing PKC- tein kinase A pathway [204]. These results demonstrate some of dependent neuroprotection [67]. Similarly, Brinton and colleagues the first insights into the mechanism behind estrogen neuroprotec- find that PPT and DPN closely mimic E2 protection from glutamate tion in tau-related disorders. excitotoxicity, including activation of ERK signaling and upregula- The described neuroprotective actions of estrogen against AD- tion of Bcl-2 expression [387,388]. They also reported differences related insults are largely mediated by activation of estrogen between the agonists in which DPN effects showed greater calcium receptors (ER). It has been well established that both ER subtypes dependence [387]. Collectively, available evidence suggests that are widely distributed in the brain, including in brain regions both ERa and ERb likely contribute to neuroprotection against C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 243

AD-related insults but that each may mediate protection by prefer- tion to promoting the non-amyloidogenic APP processing pathway, entially activate different signaling pathways. estrogen has also been implicated in the modulation of APP levels through the regulation of alternative splicing [333] and APP overexpression post-injury [307], thereby altering substrate for APP 4. Estrogen regulation of b-amyloid accumulation processing and subsequent Ab production. Importantly, subsequent studies demonstrated that estrogen In addition to increasing neuronal resistance to AD-related in- also functions as a regulator of Ab in animal models. For example, sults, estrogen may also protect against AD by preventing the the depletion of endogenous estrogen by OVX in guinea pigs in- key initiator of AD pathogenesis, accumulation of Ab. Steady state creased the levels of soluble Ab in brain, an effect partially reversed levels of Ab are inﬂuenced by opposing pathways of Ab production by E2 treatment [258]. This estrogen action may be sex-dependent and Ab clearance, both of which appear to be regulated by estro- since androgen but not E2 treatment reduced elevated Ab levels gen. Estrogen regulation of Ab was ﬁrst suggested by cell culture resulting from orchiectomy (ORX) of adult male rats [271]. A sim- experiments focused on Ab production. Early studies demonstrated ilar pattern of E2 regulation of Ab has been observed in several that estrogen modulates processing of amyloid precursor protein transgenic mouse models of AD. That is, OVX is associated with in-

(APP), the transmembrane parent protein of Ab [106]. creased Ab and E2 treatment with reduced Ab levels in Tg2576 The majority of APP is metabolized by two competing path- [380,390], APPswe [199], Tg2576xPS1 [368,390], and 3xTg-AD ways, the amyloidogenic and non-amyloidogenic pathways. In [53,54] mice. The mechanism by which estrogen regulates Ab the amyloidogenic pathway, thought to occur following endocyto- in vivo has yet to be elucidated. In the APPswe transgenic mouse sis of cell-surface APP, APP is ﬁrst cleaved by b-secretase (BACE) to model of AD, E2 treatment was associated with increased sAPPa, liberate b-APPs. The C-terminal fragment (C99/b-CTF) is left indicating increased a-secretase APP processing [199]. However, embedded in the membrane and is cleaved by the c-secretase in wild type guinea pigs, experimental manipulation of estrogen enzyme liberating the Ab40/Ab42 peptides. It is thought that status was not associated with corresponding changes in sAPPa another fragment is also released termed the APP intracellular levels [258]. Increased levels and activity of BACE observed in aro- domain (AICD), which can translocate to the nucleus and activate matase knock-out mice also suggests a potential role for estrogen gene transcription. In the non-amyloidogenic pathway, which is in the regulation of secretase expression and/or activity [380]. Fur- the predominant pathway, APP is cleaved within the Ab domain ther work will be needed to elucidate whether estrogen regulation by a-secretase to liberate a neuroprotective, secreted form of APP of Ab in animals involves APP processing and, if so, to deﬁne the (a-APPs). A C-terminal fragment (C83/aCTF) is left embedded in relevant upstream signaling components (e.g., MAPK, PKC). the membrane for further cleavage into non-amyloidogenic frag- Curiously, estrogen levels are associated with Ab accumulation ments [297,343]. only in some but not all transgenic mouse models of AD

Estrogen appears to regulate Ab levels at least in part by pro- [120,133,148,380]. In the Tg2576, PDAPP, and APPswexPS1 mouse moting the non-amyloidogenic cleavage of APP, precluding pro- models, OVX was not associated with increased Ab levels and E2 duction of the Ab peptide. In the human kidney 293 cell line, E2 treatment did not reduce Ab levels. Discrepancies in the effects has been shown to reduce the level of Ab peptide in a concentra- of estrogen on Ab levels across transgenic mouse models may re- tion-dependent manner [59]. Further, some preliminary results flect several differences, ranging from molecular design of the suggest that in a clinical setting, short-term E2 treatment is able transgenic lines to variability in methodological parameters such to reduce plasma levels of Ab in post-menopausal women naïve as the timing and dosing of hormone manipulations. One poten- to HT [20], although the significance of plasma Ab in terms of both tially important methodological difference across the studies is AD pathogenesis and diagnostic value remains controversial. their varying techniques for Ab quantification, each of which pref- How estrogen regulates APP processing is not clear, although erentially measures different pools of Ab ranging from soluble multiple pathways have been implicated. First, most data suggest monomeric Ab to oligomeric and deposited forms. Whether estro- that estrogen increases the a-secretase pathway of APP processing gen differentially regulates these various Ab pools is currently un- [171,213,366,385]. There is evidence that estrogen can promote known. In our laboratory, we found that estrogen decreases Ab the a-secretase pathway via activation of extracellular-regulated accumulation in the 3xTg-AD mouse model as assessed by the kinases 1 and 2 (ERK1 and ERK2) signaling [213], a well-estab- immunoreactive load method [53,54], which detects relatively lished estrogen signaling pathway [319,335,353]. The action of insoluble intra- and extra-cellular Ab. Future studies will be estrogen on the ERK components of mitogen-activated protein ki- needed to determine exactly which Ab pool(s) estrogen is capable nase (MAPK) signaling pathway and APP generation are rapid of regulating and whether this contributes to observed differences and may be ER independent [213]. Estrogen may also regulate in estrogen actions across models. APP processing through protein kinase C (PKC)-dependent path- Discrepancies between studies on the role of estrogen as a reg- ways. PKC signaling is a strong activator of non-amyloidogenic ulator of Ab accumulation also may indicate differences in brain APP processing [74,190,242]. Further, estrogen is a significant acti- levels of estrogen. Recent work suggests that OVX has limitations vator of PKC in both neuron culture [69,68] and in brain as a strategy to fully deplete brain estrogens. For example, in the [11,268,299]. Consistent with this possibility, a recent cell culture estrogen-responsive element-luciferase mouse model, which was study has shown that estrogen activation of a-secretase APP pro- engineered to express the non-mammalian luciferase protein in re- cessing is blocked by PKC inhibitors [385]. However, not all studies sponse to classical ER activation, OVX resulted in relatively high demonstrate such straightforward results. For example, an in vitro brain estrogen activity in comparison to other body regions [65]. study demonstrated that E2 is capable of increasing the production In evaluating the role of brain estrogens in Ab regulation, Yue of sAPPa but not reducing the release of Ab in cortical neurons et al. [380] found that in the APP23 transgenic mouse model OVX over-expressing APPswe [344]. alone was not sufficient to remove all brain estrogen and did not Some evidence also suggests that estrogen may promote non- result in elevated Ab levels. However, they reported elevated Ab amyloidogenic APP processing by altering APP trafficking. Specifi- levels after preventing E2 formation in brain by crossing the cally, Greenfield et al. reported that estrogen promoted the secre- APP23 mice with aromatase knock-out mice [380]. Thus, brain lev- tion of APP containing vesicles from the primary site of els of sex steroid hormones, which are affected not only by gonadal amyloidogenic APP processing, the trans golgi nucleus, thereby hormone production but also by de novo steroid hormone synthesis decreasing available APP substrate for Ab formation [134]. In addi- in brain (i.e., neurosteroidogenesis), may be the critical factor in 244 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 regulation of brain Ab accumulation. This hypothesized impor- exposure. In general, results from gap studies indicate diminished tance of brain hormone levels is supported by the finding that estrogen responses following many months of hormone absence. brain levels of E2 are lower in female AD patients in comparison For example, E2 and E2 plus progesterone replacement was associ- to age-matched control cases [380], a finding we have recently ated with improved spatial memory on the delayed match task replicated [287]. In addition, it has recently been shown that when administered within 3 months post-OVX, but not when hor- long-term OVX in female mice significantly lowers E2 levels in mone treatment was delayed 10 months post-OVX [115]. Further, the hippocampus while increasing serum levels of Ab [105]. Thus, E2 given immediately after OVX in rats improved spatial memory estrogen regulation of Ab may depend primarily upon brain estro- performance on radial arm maze while E2 given after 5 months gen levels, which may be depend on both ovarian and brain steroid of OVX did not [78]. Recently, it was reported that E2 given imme- production. diately after OVX but not after a 5 months delay was able to in- In addition to modulating the production of Ab, estrogen may crease hippocampal ChAT protein levels in middle-aged OVX also promote Ab clearance. One mechanism of Ab clearance is wild type rats [35]. Although this topic requires further work to through microglial degradation [282]. Estrogen has been shown elucidate the key factors and underlying mechanism(s), the data to promote microglial phagocytosis [47,267] and E2 treatment in- generated thus far suggest that in laboratory animals, the brain creases microglial internalization of Ab in microglia of both murine can show reduced hormone responsiveness after an extended per- [143] and human origin [200]. Estrogen treatment has also been iod of hormone depletion. found to reduce Ab accumulation in rats following intracerebro- Another approach to evaluate the role of aging in neural estro- ventricular Ab injection [143]. Correspondingly, increased Ab bur- gen responsiveness has been to simply compare the effects of den and impaired microglial Ab clearance has been reported in an estrogen in young adult versus aging female rodents. This type of estrogen deficient transgenic mouse model of AD [380]. Estrogen study is critical in determining the efficacy of female sex steroid has also been implicated in the regulation of levels of two major hormones before and after the onset of reproductive senescence, Ab degrading enzymes, insulin degrading enzyme and neprilysin. the result of normal reproductive aging in female rodents (re- Both of these enzymes are significant regulators of Ab levels and viewed in [58]). While reproductive aging in rodents does not mi- their regulation and activities are implicated in AD pathogenesis mic menopause, rodents do experience a pattern of estrus cycle [331,337]. A few recent studies suggest that estrogen depletion irregularity followed by reproductive senescence that is similar by OVX can decrease neprilysin activity in female rat brain, an ef- in some respects with the perimenopause period in women [95]. fect reversed by E2 replacement [163]. Thus, estrogens may influ- Age-related reproductive changes in female rodents typically be- ence Ab clearance through regulation of neprilysin, although this come apparent between 9 and 11 months of age, depending upon pathway may involve an androgen responsive element on the nep- species and strain. For example, the age-related estrus cycle rilysin gene [365]. Estrogen pathways of Ab-lowering are illus- changes have been well characterized in the C57Bl6 mouse strain trated in Fig. 1. Given the significance of Ab accumulation to AD [95]. Like women, these female mice demonstrate a large range pathogenesis, future studies must clearly define the role of estro- of variability in cycle irregularity and hormone levels during mid- gen in regulating both Ab production and clearance pathways dle and old age. These mice experience cycle cessation between 11 and how they are affected by differences in E2 brain levels. and 16 mo of age during which a substantial proportion enter a period of persistent vaginal cornification lasting 2–4 months. After this variable period, all mice enter an irreversible final stage of per-

5. Aging effects on estrogen responsiveness manent diestrus characterized by low E2 and progesterone levels and elevated luteinizing hormone levels [114], ovarian follicle Whether the described neuroprotective effects of estrogen depletion, and loss of reproductive capability [126]. prove to have therapeutic relevance to age-related neural diseases In estrogen neuroprotection studies, middle-aged female including AD will depend in part on the brain’s responsiveness to rodents undergoing reproductive senescence show diminished ef- estrogen with advancing age. One of the primary criticisms of fects in some studies but retained protection in others. First,

WHIMS and other clinical studies of estrogen-based HT is that several reports suggest that the neural effects of OVX and E2 treat- the intervention may have been initiated beyond a critical window ment are diminished in aging female rodents. Studies by Sohrabji of opportunity [118,265,276]. This notion refers to the possibility and colleagues suggest that reproductive senescence in middle- that the aging brain age may lose responsiveness to sex steroid aged female rats reduces protective estrogen actions. For example, hormones after an extensive period of low hormone levels, such in assessing estrogen regulation of neurotrophin expression, they as occurs following menopause. According to this argument, HT found that E2 treatment in OVX young adult female rats (age 3 may exert estrogenic effects only if begun near the time of meno- mo) increased levels of BDNF, trkA, and trkB in olfactory bulb pause. Since most participants in the WHIMS study were many and diagonal band of Broca, whereas E2 treatment of middle-aged years beyond menopause, the critical window hypothesis could ex- (17 months) OVX reproductively senescent, female rats showed plain in part the absence of beneficial neural actions of HT. Consis- either no increase or decreased expression of neurotrophins tent with this position, several clinical studies have noted that HT [175]. Similar studies by this research group found that, in compar- is associated with positive neural effects in women showing men- ison to young adult female rats, reproductively senescent female opause symptoms (e.g., flushing), suggesting retained estrogen rats show several alterations in estrogen-mediated effects includ- responsiveness [154,153,157,306,305]. Although this issue re- ing cytokine and growth factor responses following injury mains to be fully evaluated, results from the Multi-Institutional Re- [176,237] and blood–brain-barrier permeability [18]. In another search on Alzheimer Genetic Epidemiology study show that HT paradigm, Finch and colleagues reported differences between was associated with reduced AD risk only in post-menopausal wo- young adult (3 mo) and middle-aged (18 mo) female rats on estro- men that initiated treatment at a relatively younger age [154]. gen regulation of compensatory neuronal sprouting following The critical window notion of an age-related loss in brain entorhinal cortex lesion. In comparison to young rats, older rats responsiveness to estrogen is supported by studies in animal mod- no longer exhibited an OVX-induced decrease in sprouting and els. One experimental approach to address this issue is the ‘‘gap showed differences in regulation of GFAP mRNA [321]. However, paradigm” in which OVX animals are treated with E2 replacement some neural actions of estrogen appear to remain relatively robust after short versus long periods of time, a design that assesses the during aging. For example, E2 treatment in middle-aged OVX effects of prolonged hormone deprivation on subsequent hormone rats enhanced performance on hippocampal-dependent spatial C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 245 memory tasks [78] as well as altered the hippocampal expression the extent to which observed age-related changes in estrogen of several genes [1]. responsiveness may be delayed, prevented and or reversed. This In neural injury paradigms, there is evidence of both retained phenomenon has important implications for the future of HT in and altered estrogen neuroprotection. In reproductively senescent post-menopausal women. Ongoing clinical and animal studies female rats, both E2 and progesterone reduced infarct size in a promise to shed new insight into this issue in the next few years. model of stroke injury [5]. Wise and colleagues reported similar neuroprotective effects of E2 treatment in OVX young (3–4 mo) and middle-aged (9–12 mo) rats in the middle cerebral artery 6. Progesterone interactions with estrogen in regulation of occlusion (MCAO) model of stroke [85,360]. In our laboratory, we Alzheimer’s disease have found that reproductively senescent female rats show evidence of altered but not completely diminished estrogen neuro- Estrogen actions must also be considered with respect to the protection. In young adult (3 mo) female rats, an excitotoxic second major class of ovarian hormones, progestogens. Progester- lesion to hippocampus induced by systemic application of the glu- one has long been recognized as a regulator of estrogen, particu- tamate agonist kainate is significantly worsened in animals OVX larly in the female reproductive system (reviewed in [130]), for 2 weeks prior to the lesion. In a similar paradigm, we find that where it often antagonizes estrogen action. Clinically, progesto-

E2 treatment, initiated at the time of OVX, signiﬁcantly increases gens are typically a key element of HT that are thought to minimize the number of viable hippocampal neurons following kainate le- deleterious effects of estrogen. Perhaps most importantly, in exper- sion [286]. However, in reproductively senescent (14 mo) rats, imental paradigms progesterone can inhibit human endometrial

OVX was not associated with further cell loss, although E2 replace- cancer cell growth [77,79] and in clinical studies progestogens ment still resulted in a signiﬁcant increase in neuron survival [55]. are associated with reduced risk of endometrial cancer [129,257].

In a model of spinal cord injury, the effects of E2 treatment in In clinical studies of AD and dementia, most results indicate young (2 mo) versus middle-aged (12 mo) sham-OVX and OVX fe- similar outcomes with both estrogens alone (ie., CEE) and estro- male rats were investigated. Treatment with E2 protected against gens in combination with progestogens (ie., CEE + MPA). In the several indices of injury in both young and aging OVX rats, how- WHIMS trial, a comparison between the CEE alone and CEE + MPA ever in ovary-intact rats E2 neuroprotection was lost in middle- arms of the study raised the question that the clinical efficacy of HT aged rats [60]. may be dependent upon the hormone constituents within. Both Interestingly, emerging data suggest that patterns of reproduc- the CEE and CEE + MPA arms failed to demonstrate a protective ef- tive aging in female rats may contribute to altered estrogen fect and both actually increased the risk of dementia compared to responsiveness with age. A recent study compared estrogen pro- women receiving placebo [311]. Notably, the CEE alone arm had a tection using the MCAO stroke model in middle-aged female rats negative impact on global cognitive function, however this nega- stratified by their stage of reproductive aging: reproductively tive impact was worsened when pooled with the data from the senescent rats that had entered a persistent acyclic state, and rats CEE + MPA arm [91]. Interestingly, short-term HT treatment in wo- with normal but lengthened cycles. In comparison to the middle- men with existing AD was associated with some benefits on psy- aged cycling rats, the reproductively senescent rats exhibited lar- chiatric symptoms in the CEE group but not in the CEE + MPA ger lesions and no longer showed reduced lesion size following group [162]. These and related issues suggest that the inclusion

E2 treatment [298]. Our recent data also suggest that responsive- of a progestogens may influence the effects of CEE alone on cogni- ness to estrogen protection vary according to status of reproductive outcomes and risk of dementia in post-menopausal women tive aging. In the kainate lesion model, we observed that E2 was [75]. most protective in rats showing an acyclic state of persistent vag- Despite the common use of the progestogen MPA in current HT inal cornification compared to rats with present, albeit irregular paradigms, researchers have only relatively recently begun to cycles [55]. Thus, although additional studies are necessary to de- investigate the effects of progesterone on the CNS in regards to fine the relationships, it is reasonable to hypothesize that patterns aging and neurodegenerative diseases. Although compelling exper- of reproductive aging affect estrogen neuroprotection. Such find- imental evidence indicates numerous protective actions of estro- ings are consistent with the ‘‘critical window” hypothesis of HT gen and progesterone when delivered independently [44,295], and have important implications for the future of clinical use of comparatively less is known about interactions between estrogen estrogen-based therapies. and progesterone when they are administered together. Interest- Why the brain shows is less responsive to estrogen with age is ingly, accumulating observations indicate that progesterone often unclear, but age-related decreases in ER expression as well as E2 antagonizes rather than synergizes with estrogen-mediated neuro- binding to ERs in aged rat brain have been reported [58,289,361]. protective actions. This incomplete understanding highlights the Estrogen actions also show age-related changes in other estrogen need to determine the interactive effects of these hormones in responsive tissues, including uterus, bone and heart. For example, the brain. the uterus becomes less responsive to estrogen with increasing Findings from an increasing body of research have begun to age, showing smaller OVX-induced decreases in uterine weight provide insight into how beneficial neural actions of estrogen [367] and uterotrophic effects of estrogen only when treated soon are affected by interactions with progesterone. Findings from after OVX [78]. Further, while most studies demonstrate that the both in vitro and in vivo paradigms suggest that progesterone trophic effects of E2 on bone extend through middle-age, some treatment can antagonize estrogen neuroprotection. For example, studies suggest that as aging progresses, this effect is altered from long-term progesterone treatment in aged (23–24 mo) OVX fe- an ERa/b-mediated effect to only an ERa mediated one [168]. male rats blocked estrogen upregulation of brain derived neuro- Interestingly, increased ERa predominance has also been sug- trophic factor, nerve growth factor, and neurotrophin 3 [32].Ina gested to underlie age changes in neural estrogen responsiveness similar paradigm of hormone treatment in middle-aged OVX [19]. rats, estrogen-induced improvement in spatial memory perfor- Taken together, available experimental research suggests that mance was blocked by co-administration of progesterone [33]. although estrogen neuroprotection is often observed in aging rats, More specific to neuron viability though are recent studies from it can also be significantly diminished. Still uncertain is how the our laboratory demonstrating that progesterone blocks estrogen many AD-related facets of estrogen neuroprotection may be im- neuroprotection from excitotoxic injury in female rats. In both pacted by aging. Additional studies are necessary to determine young adult (3 mo) [286] and middle-aged (14 mo) [55] OVX 246 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258

rats, continuous E2 treatment reduced kainate-induced neuron Despite evidence of antagonistic neural effects of progesterone loss in hippocampus CA2/3 whereas continuous progesterone on estrogen neuroprotective actions, progestogens are still deemed did not significantly affect neuron viability. Importantly, when a necessary component of HT in women with a uterus. Therefore, progesterone was included with E2, estrogen neuroprotection HT may need to be optimized to maximize the benefits and mini- was no longer observed [55,286]. Similarly, Brinton and col- mize the unwanted consequences associated with estrogen-pro- leagues recently reported that E2 and progesterone administered gestogen interactions. One possible strategy is the use of cyclic independently to wild type rats enhanced several markers of hormone delivery rather than the continuous, combined treatment brain mitochondrial function but, when replaced in combination, that is currently common to HT. Initial clinical evaluation of cyclic the hormones showed attenuated rather than enhanced re- progestogen exposure has been completed and more is underway. sponses [167]. Several clinical HT trials have incorporated a comparison between While the mechanisms behind this antagonistic property remain continuous versus cyclic progestogen in post-menopausal women unknown, our laboratory has begun to investigate the possibility on osteoporosis and cognitive function. For example, two com- that progesterone modulates estrogen function in part by regulat- pleted studies have both demonstrated that long-term HT with a ing ER expression. We have observed in primary neuron culture cyclic progestogen dose was able to increase bone mineral density that low physiological concentrations of progesterone rapidily in post-menopausal women [239,56]. Similarly, a continuous downregulate mRNA levels of both ERa and ERb [174]. Functionally, estrogen plus cyclic progestogen paradigm is currently employed this progesterone-mediated decrease in ER expression was associ- in the KEEPS (Kronos Early Estrogen Prevention Study) Cognitive ated with inhibition of both ERE-mediated transcriptional activity and Affective Study, a randomized, placebo-controlled, double- and estrogen neuroprotection against apoptosis [174]. Although blind study investigating the effects of HT in post-menopausal wo- progesterone treatment by itself did not significantly affect apopto- men who are within 36 months of their final menstrual period sis in this paradigm, abundant evidence demonstrates that proges- [357]. This and similar new trials promise to provide important in- terone can activate several neuroprotective cell signaling pathways, sight on the efficacy of cyclic hormone delivery and the hypothe- including Akt [318] and ERK [233,232] signaling and upregulation sized importance of a critical window of hormone intervention. of the anti-apoptotic protein Bcl-2 [232]. Further, as reviewed in Experimental studies in animal models lend support for the use an accompanying article, progesterone can significantly protect of cyclic rather than continuous progesterone to optimize estrogen neurons against numerous insults [44,295]. Thus, independently neuroprotection. For example, Gibbs have demonstrated that estrogen and progesterone can exert protective actions, but in com- short-term treatment with E2 and progesterone can improve cho- bination they can inhibit each other and thus fail to protect. Perhaps linergic function [116], with maximal benefit resulting from cyclic not unexpectedly, the interactive neuroprotective effects of the two administration of estradiol and progesterone and the least benefit female sex steroid hormones are not quite so straightforward. Be- from continuous combined treatment [115]. In our laboratory, we sides antagonistic effects, progesterone can also synergistically have begun investigating the potential utility of cyclic progester- interact with estrogen to promote beneficial neural effects includ- one against AD neuropathology by comparing cyclic versus contin- ing increased spine density [100,313]. In OVX female rats, acute uous progesterone in the presence and absence of continuous E2 in progesterone treatment (2–6 h) was observed to augment the OVX 3xTg-AD mice. Our results show that whereas continuous estrogen-induced increase in spine density, whereas prolonged progesterone largely inhibits estrogen protection from AD-related progesterone treatment (18 h) blocked the estrogen effect [362]. neuropathology, cyclic progesterone appears to significantly in- Thus, a key factor in understanding estrogen neuroprotection and crease estrogen protection against Ab accumulation, tau phosphor- perhaps its relevance to HT in post-menopausal women is elucida- ylation, and working memory deficits (unpublished observations). tion of the interactions between estrogen and progesterone. These exciting new findings support the hypothesis that estrogen– A similar progesterone regulatory relationship also appears to progesterone interactions can yield additive neuroprotection and affect estrogen protection from indices of AD-like neuropathology. that cyclic hormone delivery may be a critical parameter. In recent studies, our laboratory has begun to investigate how progesterone interacts with estrogen in regulating Ab accumulation in the 3xTg-AD transgenic mouse model of AD. As discussed above, 7. Age-related androgen depletion and Alzheimer’s disease we found that 3 months following OVX of young adult female 3xTg-AD mice, there were robust increases in Ab accumulation In parallel to the relationships between age-related estrogen and tau phosphorylation and impaired performance in spontane- loss in women and increased AD risk, testosterone is depleted as ous alternation behavior in comparison to sham OVX 3xTg-AD a normal consequence of aging in men and is linked with elevated mice [54]. Continuous E2 treatment during the 3 month OVX peri- risk of AD. As discussed above, a significant biological event in wood largely prevented the worsening of AD-like neuropathology and men that contributes to the role of aging in AD is menopause and behavioral performance, however continuous progesterone by it- the resultant loss of the sex steroid hormones estrogen and proges- self did not affect OVX-induced changes in either Ab accumulation terone. Although men do not experience menopause per se (i.e., a or behavior. Consistent with an antagonistic role of progesterone, cessation of reproductive ability, nearly complete loss of sex ste- we observed that co-treatment with progesterone antagonized roid hormones), men do experience a somewhat similar process the beneficial effect of estrogen in lowering Ab accumulation termed androgen deficiency in aging males (ADAM). ADAM refers [54]. However, estrogen and progesterone co-treatment did show to normal, age-related depletion of testosterone and the corre- a beneficial effect in reducing levels of tau hyperphosphorylation, sponding constellation of symptoms that reflect dysfunction and suggesting positive estrogen-progesterone interactions. Consistent vulnerability to disease in androgen-responsive tissues including with a beneficial role of progesterone, Frye and colleagues report brain [24,50,96,125,179,183,217,224,302]. The decline in testoster- that long-term progesterone treatment following OVX in another one levels begins in the 3rd decade and continues at an annual rate transgenic mouse model of AD was associated with some cognitive of 0.2–1% for total testosterone and 2–3% for bioavailable testoster- benefits [104]. Taken together, these studies provide the first infor- one [94,131,227]. Unlike menopause, aging men do not experience mation to date on the interactive effects of estrogen and progester- comparable levels of andropause. That is, although all men exhibit one on AD-like neuropathology and demonstrated the potential for significant age-related testosterone loss typically beginning in the both positive and negative outcomes in terms of protection from third decade of life, men vary in the extent of testosterone loss and AD-related neuropathology. the corresponding severity of clinical manifestations [226,326].It C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 247 is estimated that 30–70% of men aged 70 and older are hypogonad- terone therapy in men for protective roles against age-related al, resulting in at least 5 million aging men in the US suffering the cognitive decline and development of dementia. Androgen thera- consequences of andropause and only a small minority of those re- pies have been approved and used for the treatment of hypogonad- ceive hormone treatment [142,225]. ADAM is associated with in- ism in men and are typically associated with reduced fat mass and creased risk of sarcopenia, osteoporosis, falls, frailty, and all improved muscle mass and bone density as well as increased cause mortality. mood, libido, and overall quality of life [31,327,332]. However, The brain is a highly androgen responsive tissue where andro- clinical evaluation of beneficial cognitive effects of testosterone gens induce several beneficial actions. For example, androgens therapy have been mixed. For example, a study of hypogonadal have been shown to improve mood and promote select aspects and eugonadal men found that testosterone increased verbal flu- of cognition, including spatial abilities [127,172] and verbal fluency [4]. Others report improvements in spatial cognition and ency [4]. Men with a higher free testosterone index have been working memory following testosterone treatment [62,173,172]. found to perform better on visual and verbal memory and exhib- In contrast, some studies did not find significant changes in cogni- ited better long-term memory [23], while those with a low free tion following testosterone therapy [30,90,141,210,339]. testosterone index can show decreased visual memory, visuomotor A few studies have evaluated the effects of testosterone therapy scanning, verbal memory, and visuospatial processing [219]. in subjects with AD. In a small clinical study of men recently diag- ADAM has been associated with impaired cognitive performance nosed with AD, testosterone treatment for up to 1 year contributed in some but not all studies [141,219]. to improvement in both overall cognitive ability and visual spatial One recently established consequence of ADAM is an increased skills [328]. In contrast, other studies have not reported significant risk for the development of AD. Several [161,160,159,274,352] but benefits of testosterone therapy in men with mild cognitive not all [255] studies have identified a relationship between low cir- impairment and AD [63,206]. As with the observed inconsistencies culating levels of testosterone and a clinical diagnosis of AD. In in the literature of HT use in women, there are likely several factors these studies, the relationship between testosterone and AD ap- that contribute to the observed differences between testosterone pears to be strongest when circulating levels of free rather than to- studies, including cognitive domains, treatment type and duration, tal testosterone are examined, and when mean ages are under and the age and other characteristics of the subjects. 80 years [161,160,159,249]. The relationship between testosterone and AD may be influenced by the presence of at least one apolipoprotein e4 allele, a genetic risk factor for AD [322]. Specifically, 8. Testosterone neuroprotection and Alzheimer’s disease insults men with at least one e4 allele had lower levels of testosterone than men without an e4 allele [160]. Animal studies support a link One beneficial action of androgens that is hypothesized to con- between apolipoprotein E, testosterone, and AD [269,270]. tribute to a role in reducing risk of AD is neuroprotection. Andro- Although the majority of studies have identified a relationship gens are established promoters of neuron viability during neural between low testosterone and increased AD risk in men, most were development as well as in adult brain following mechanical injury unable to determine whether low testosterone contributes to the and disease-related toxicity. One target of androgen neuroprotec- disease process or is merely a result of it. However, two completion is motorneurons following axotomy [178]. In this paradigm, mentary studies suggest low testosterone occurs prior to or in testosterone treatment accelerates the rate of nerve regeneration the early stages of AD pathogenesis, and thus likely acts a risk fac- and attenuates neuron loss [164,178,177,193,194,195,330, tor. The first study compared clinical diagnosis of dementia with 377,379]. Similarly, following facial nerve crush in male hamsters, blood levels of testosterone in the prospective Baltimore Longitudi- testosterone increased the rate of axonal growth and functional nal Study on Aging [220]. Male subjects were followed for recovery [193,192]. These effects are true not only for testosterone 4–37 years with a mean of 19 years per subject and were diag- but also its potent androgen metabolite dihydrotestosterone (DHT) nosed as clinically normal at the time of their first testosterone [378,377,379]. In addition, the anti-androgen flutamide was able to measurement. Subjects that eventually received a clinical diagno- block testosterone’s neuroprotective effects on motor neurons sis of AD showed lower circulating levels of free testosterone. [195], corroborating the role of androgen versus estrogen path- Interestingly, in those men with AD, testosterone levels were re- ways. The mechanism behind this neuroprotective action appears duced at check-ups 5–10 years prior to diagnosis [220], suggesting to be through androgen regulation of trophic factors [378,379]. Re- androgen loss occurred well before clinical manifestations of the cently, studies have found that in addition to long-term treatment, disease. Consistent with this study are findings from a study by short-term testosterone, DHT, and estrogen treatments are protec- our laboratory in which we linked low brain levels of testosterone tive and suggest a more direct mechanism of hormone action with increased risk of AD in men [284]. First, using human post- [164]. mortem brain tissue from neuropathologically normal men, we In addition to neuroprotective effects on motor neurons, andro- found that levels of testosterone but not E2 show a significant gens have also been found to promote neuron survival in brain re- age-related decline. When we examined changes in brain levels gions vulnerable to neurodegenerative diseases such as of hormones across cases stratified by neuropathological status, Alzheimer’s disease. These areas include the hippocampus and cor- we found significantly decreased brain levels of testosterone in tical regions, which are both affected in AD and rich in androgen AD cases as compared with neuropathologically normal cases even receptors [314]. In a study by Garcia Segura and colleagues, acute after controlling for the age-related hormone loss [284]. We also testosterone treatment attenuated neuron loss in the hilus of the measured brain levels of androgens in cases with mild neuropatho- dentate gyrus following excitotoxic lesion in ORX male mice [16]. logical changes, consistent with the earliest stages of AD. These Interestingly, acute treatment of E2 was also protective while cases also exhibited low testosterone levels [284], again indicating DHT treatment did not protect against neuron loss in androgen de- that testosterone loss occurs prior to robust pathology and thus pleted mice. Furthermore, the protective effect of testosterone was may contribute to the development of AD. Taken together, these blocked by an aromatase inhibitor, suggesting that in this model of results suggest age-related testosterone depletion in men is a risk acute hormone treatment, estrogen is responsible for testosterone factor for AD. neuroprotection [16]. A study from our laboratory investigated the Unlike the numerous clinical studies that have evaluated the effect of long-term hormone replacement on neuronal death in- efficacy of HT use in treating and or preventing AD in post-meno- duced by excitotoxic lesion. In this study, depletion of endogenous pausal women, comparatively few studies have examined testos- androgens as a consequence of ORX resulted in increased hippo- 248 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 campal neuron loss following kainate lesion in comparison with gen pathways requires membrane AR, intracellular AR, or both is sham ORX rats. However, in ORX rats treated for two weeks with unclear. We observed that testosterone conjugated to albumin – DHT, this increase in cell loss was blocked [272]. In contrast to which is designed to prevent cell entry and thus permit only the results of Garcia Segura and colleagues, we found that rats trea- activation of membrane receptors – was ineffective in activating ted with E2 did not exhibit decreases cell loss following kainate le- protective MAPK/ERK androgen signaling [229]. Similarly, Gatson sion. This suggests that androgen regulation of neuroprotection in and colleagues reported that while DHT phosphorylates both ERK this paradigm is a result of androgen rather than estrogen path- and Akt, albumin-conjugated DHT resulted in dose-dependent sup- ways. Behavioral responses to seizures were measured and no dif- pression of ERK signaling in glioma cells expressing AR [113]. Path- ferences were observed between groups suggesting that the ways of androgen neuroprotection are summarized in Fig. 2. observed androgen neuroprotection was not a result of decreased In contrast to these observations of androgen neuroprotection, seizure severity [272]. Although we did not observe an androgen- there appear to be circumstances in which androgens fail to protect mediated effect on seizure severity, other studies have found a sig- against neural injury and can even exacerbate insults. For example, nificant effect of androgens on seizure severity. In studies by Frye Dluzen and colleagues found that E2 but not testosterone reduces and colleagues, testosterone decreased neuron loss by inhibiting methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) nigrostriatal seizure activity [102,103]. Subsequent studies found that protec- dopaminergic toxicity [82]. Further, testosterone has been found tion was due to DHT metabolism to 5a-androstane-3a,17a-diol, to exacerbate neuron loss following middle cerebral artery occlusion an androgen that acts on GABAA receptors. GABA activation re- (MCAO) [372,371]. In this model removal of testosterone 6 h prior to duces excitatory signaling, which in turn attenuates seizure activ- MCAO reduces lesion volume by as much as half [372,371]. It is not ity thereby minimizing lesion severity [87,88,277]. clear whether the absence of androgen neuroprotection in these par- While in vivo models provide valuable insight into androgen adigms reflects roles of brain region, insult, and or other factors. One neuroprotection, neuron culture models of toxicity have proven important issue in at least some studies may be androgen concentra- valuable in defining the underlying molecular mechanisms. Cell tion. That is, at physiological low nanomolar levels testosterone can culture models of neural injury have demonstrated testosterone protect against excitoxicity in cultured neurons, but worsens cell protection against serum deprivation [45,138],Ab toxicity death when present at supraphysiological micromolar levels [244]. [229,260,384], and oxidative damage [2]. Testosterone neuropro- Similarly, micromolar but not nanomolar concentrations of testos- tection against serum deprivation-induced apoptosis requires acti- terone were associated with increased calcium signaling and neuro- vation of an androgen receptor (AR) dependent mechanism [138]. nal apoptosis [92,112]. Thus, although there is ample evidence of Specifically, the anti-androgen flutamide attenuated protection androgen neuroprotection against AD-related insults, androgen ef- while an aromatase inhibitor had no effect on neuron viability fects on neuron viability may depend on a number of factors, includ- [138]. Consistent with this androgen-mediated mechanism of ing brain region, type of insult, and hormone concentration. androgen neuroprotection is an early study from our laboratory, In addition to direct neuroprotection, androgens also protect which found that testosterone neuroprotection against toxicity in- against another form of neuropathology directly relevant to AD, duced by extracellular Ab results from DHT not E2 [260]. DHT treat- hyperphosphorylation of tau. Abnormal, excessive phosphoryla- ment in this paradigm was equally as protective as testosterone, tion of the cytoskeletal protein tau in the form of neuropil threads but use of an anti-estrogen droloxifene failed to block protection, and neurofibrillary tangles is a defining neuropathological charac- suggesting androgen pathways are responsible for neuroprotection teristic of AD and several other neurodegenerative disorders [260]. Further, we observe androgen neuroprotection in PC12 cells [166,355,358]. Although relatively little research has examined transfected with AR but not in either untransfected PC12 cells or the relationship between androgens and tau hyperphosphoryla- those transfected with empty vector [229]. tion, available evidence does indicate that androgens are protective There are several potential mediators of androgen neuroprotec- against this pathology. Papasozomenos and colleagues have exam- tion downstream of AR. Some evidence suggests androgen neuro- ined the effects of testosterone and E2 on tau hyperphosphoryla- protection may be mediated through attenuation of oxidative tion using a model of heat shock-induced phosphorylation. In stress [2]. Another potential mechanism involves a classic genomic this paradigm, testosterone but not E2 prevents tau hyperphosph- mechanism, increased expression heat shock proteins. Androgen orylation [252,250,251]. In addition to affecting phosphorylation of attenuation of Ab induced toxicity was associated with elevated tau, recent evidence suggests that androgens also regulate tau levels of heat shock protein 70, which is known to participate in cleavage [253]. Specifically, testosterone prevented calpain-medi- protective responses against cellular stress and neurodegeneration ated tau cleavage preventing the generation of the 17-kDa tau frag- [384]. Recent work from our lab identified a non-genomic, AR- ment [253]. dependent mechanism involving activation of a mitogen-activated protein kinase (MAPK)/extracellular signal regulated kinase (ERK) pathway [229]. We observed that physiological levels of testoster- 9. Testosterone regulation of b-amyloid accumulation one and DHT rapidly and transiently activated MAPK/ERK signaling in cultured hippocampal neurons. Downstream of ERK, we found In addition to classic neuroprotective actions, androgens may that androgens activated p90 kDa ribosomal S6 kinase (Rsk), which also protect the brain from AD by regulating accumulation of Ab. in turn phosphorylates the pro-apoptotic protein Bad. Phosphory- Initial work suggesting a relationship between androgens and Ab lation of Bad results in its inactivation of Bad, thereby tilting the came from a small study evaluating men treated with anti-andro- balance of apoptosis towards increased cell viability. Pharmacolog- gen therapies for prostate cancer. Gandy and colleagues found that ical inhibition at any step of this pathway prevents both phosphor- within several weeks following initiation of anti-androgen therapy ylation of Bad and androgen neuroprotection [229]. Confirming the (consisting of leuprolide and flutamide), circulating levels of tes- non-genomic nature of this pathway, we found that in this neuron tosterone and E2 were largely depleted whereas plasma levels of culture paradigm the classic anti-androgens flutamide and cypro- Ab were significantly elevated [107]. Martins and colleagues simi- terone acetate mimicked neuroprotection against apoptotic insults larly reported an association between androgen depletion and ele- afforded by testosterone and DHT even though they also blocked vated Ab levels in males receiving anti-androgen therapy for the classic genomic actions of androgens [231]. Further, protective treatment of prostate cancer [6] and in older males suffering from actions of flutamide and cyproterone acetate were only observed memory loss or dementia [117]. To what extent these observations in AR-containing cell lines. Whether activation of protective andro- reflect effects of androgen versus estrogen pathways or perhaps re- C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 249

Fig. 2. Androgens activate neuroprotective pathways that may attenuate Alzheimer’s disease. First, testosterone (T) is aromatized in brain to 17b-estradiol (E2), which activates estrogen-mediated neuroprotective pathways (summarized in Fig. 1). Second, testosterone and its metabolite dihydrotestosterone (DHT) activate AR-dependent protective pathways. T and DHT reduce neuronal apoptosis by a non-genomic signaling cascade involving activation of MAPK/ERK, followed by activating phosphorylation (p) of Rsk, and inactivating phosphorylation of the pro-apoptotic protein Bad. Also, androgens decrease levels of the AD-related protein Ab by a classic genomic mechanism involving activated AR interaction with androgen response elements (ARE) on the neprilysin gene, which results in increased expression of this Ab-catabolizing enzyme.

flect associated changes in gonadotropins remains incompletely reduced Ab [128]. Unclear was whether these findings involved di- resolved [13,283]. rect androgen pathways, indirect activation of estrogen pathways, Consistent with the observations in aging men, experimental or both. In a subsequent culture study, testosterone was also ob- work in rodents also indicates that androgens function as endoge- served to promote proteolysis of APP by a-secretase, however this nous negative regulators of Ab. Early studies in our laboratory effect was blocked in the presence of aromatase inhibitors suggest- demonstrated that androgens but not estrogens reduce brain levels ing estrogen dependence [121]. In our study of male rats in which of soluble Ab in male rats [271]. Specifically, we found that ORX-in- androgens were found to regulate brain levels of soluble Ab, we did duced depletion of endogenous androgens was associated with a not observe detectable differences in either full-length APP or modest but significant increase in soluble Ab from hemi-brain ly- APPa across androgen treatment groups [271], perhaps indicating sates. Further, DHT treatment in ORX rats resulted in a significant that alterations in APP processing are not the only mechanism by reduction in brain Ab levels, although E2 treatment had no effect which androgens affect Ab levels. [271]. We observed similar findings on the relationship between In addition to regulating Ab generation via effects on APP prote- androgen levels and Ab accumulation in the 3xTg-AD mouse model olysis, androgens may also decrease Ab levels by promoting endog- of AD. Male 3xTg-AD mice were depleted of androgens by ORX at enous clearance pathways. Recent evidence from our group age 3 mo, a time prior to the development of significant AD-like demonstrates that androgens reduce Ab levels as a consequence pathology [238]. Mice were exposed immediately to either DHT of regulating expression of the Ab-catabolizing enzyme neprilysin or vehicle, treatments that were maintained continuously for the [376], a critical enzyme in homeostasis of brain Ab [170]. In neural next 4 months. After the treatment period, we observed significant cell cultures, we found that androgens robustly increase protein intracellular accumulation of presumably insoluble Ab in gonadally levels of neprilysin. A classic, AR-dependent genomic mechanism intact, sham ORX 3xTg-AD mice (age 7 mo) in the CA1 hippocam- is implicated since (i) the neprilysin gene contains androgen re- pus, subiculum, and amygdala [285]. In comparison to the sham sponse elements [304], (ii) neprilysin regulation was blocked by ORX animals, the ORX mice showed significantly higher levels of the anti-androgens flutamide and cyproterone acetate, and (iii) Ab in all three brain regions as well as significantly poorer behav- androgens only increased neprilysin in AR-containing cells [376]. ioral performance on a spontaneous alternation task [285]. The ele- Over-expression of human APP in cultured cells resulted in elevated pathology and exacerbated behavioral impairments in the vated levels of soluble Ab that were reduced by testosterone and ORX group were blocked in ORX mice treated with DHT, suggesting DHT. Further, we found that pharmacological inhibition of either a preventive effect of androgens in regulation of AD-related pathol- AR or neprilysin blocked the ability of androgens to reduce Ab lev- ogy. Confirming our findings in animal models, a recent study re- els, suggesting that the Ab-lowering actions of androgens is medi- ported that levels of Ab in plasma and cerebrospinal fluid were ated by AR-dependent regulation of neprilysin expression [376]. significantly elevated following ORX in guinea pigs, an effect pre- Importantly, these observations were replicated in male rats. We vented by testosterone replacement 1 week following ORX [346]. found that ORX-induced androgen depletion resulted in decreased There are likely several mechanisms that contribute to andro- levels of NEP and elevated Ab, and DHT replacement restored NEP gen regulation of Ab (Fig. 2). One obvious possibility is that aroma- and Ab to levels observed in sham GDX animals [376]. tase-mediated conversion of testosterone to E2 in brain allows activation of the several estrogen pathways of Ab regulation discussed above. In fact, cell culture studies are consistent with indi- 10. Emerging strategies of hormone-related therapies in rect testosterone activation of estrogen-mediated regulation of APP Alzheimer’s disease processing. The first study to examine testosterone regulation of APP and Ab in culture found that prolonged testosterone treatment As reviewed in this article, there are numerous neuroprotective was associated with increased a-secretase cleavage of APP and actions of estrogens and androgens that have direct relevance to 250 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258

AD pathogenesis and compelling potential to prevent and possibly lect androgen-responsive tissues of interest, including brain, treat the disease. However, the promise of estrogen-based and muscle, and bone. androgen based HTs in reducing AD risk have yet to be realized. Several strategies of SARM design are currently been pursued As findings and directions from clinical and basic science research [46,108,359]. One strategy is to develop novel steroidal com- become increasingly integrated, it is anticipated that critical pounds that are not substrates for the enzyme 5a-reductase or parameters affecting HT efficacy will be optimized, including age yield reduced metabolites with minimal androgencity. Testoster- of HT initiation as well as the formulation, regimen, and delivery one is converted to DHT by the actions of 5a-reductase, an enzyme of HT. As ongoing research continues to address these crucial and localized in specific target tissues such as prostate. Prostate growth immediate concerns, an emerging area of investigation is the depends largely on the actions of DHT rather than T because DHT development of natural and synthetic hormone mimetics that will exhibits 10-fold greater net potency, which reflects both a higher preferentially activate estrogen and androgen neuroprotective binding affinity for AR and a slower dissociation rate from AR mechanisms while minimizing deleterious consequences in other [359]. SARMs that are not 5a-reductase substrates or form DHT- tissues. like derivatives with weak androgenic activity have low androgen Estrogen compounds that show tissue-selective agonist actions action in prostate [108,188,201,211,245,301,342]. A promising are termed selective estrogen receptor modulators (SERMs). SARM in this category is 7a-methyl-19-nortestosterone, com- Currently the most studied and clinically relevant SERMs are monly called MENT [223,301,342], which was developed by the tamoxifen and raloxifene, synthetic compounds that exhibit tis- Population Council and is currently in clinical trials as an androgen sue-dependent ER agonist and antagonist actions. As a potent therapy for hypogonadal men [10,345]. MENT shows low androgen antagonist of estrogen action in breast tissue, tamoxifen is best activity in prostate but is more potent than T in other peripheral recognized as an antiestrogen used to treat breast cancer although androgen-responsive tissues including bone [76,301,342] and it can exert agonist ER effects on bone and lipids [48,300]. Interest- muscle [76,301,342]. The effects of MENT on neural function are ingly, low concentrations of tamoxifen can protect cultured neu- virtually unknown, but it has been shown to mimic the ability of rons from toxicity due to Ab and glutamate [243], suggesting the testosterone to induce sexual behavior in castrated rats [223] potential for a protective role against AD. However, tamoxifen and is a robust regulator of the hypothalamic–pituitary–gonadal has also been observed to block E2-mediated protection in cultured axis, suggesting neural efficacy. neurons [57,383]. Potential benefits of tamoxifen use in post-men- Another SARM design strategy is the development of non-ste- opausal women for prevention or treatment of AD has not been roidal synthetic AR ligands. One such SARM is flutamide, which well studied. Some studies indicate increased risk of cognitive def- mimics the abilities of T and DHT to increase hippocampal spine icits in tamoxifen users [246,308], whereas another study sug- density in both male and female rats [207,208]. In cultured neu- gested tamoxifen may reduce AD risk [41]. Like tamoxifen, rons, we have found that flutamide antagonizes classic genomic raloxifene antagonizes estrogen actions in breast but has agonist actions of T and DHT but mimics rather than blocks the nonge- actions on bone [48,81]. In brain, raloxifene mimics some but not nomic neuroprotective actions of these androgens [230]. Of par- all protective estrogen actions. Raloxifene increases choline acetyl- ticular interest are novel compounds that bind AR but have transferase in hippocampus of OVX rats [363] and in cultured altered interactions with AR binding pocket side chains that neurons it increases neurite outgrowth [235] and can reduce Ab underlie tissue specificity. Recent research has determined that toxicity [243]. Conversely, E2 but not raloxifene was effective in although synthetic SARMs must closely mimic the rigid backbone attenuating Ab-induced inflammatory reaction in OVX rats [334]. core structure of the AR ligand binding domain, there can be In post-menopausal women, raloxifene use has been linked with extensive variation in how they interact with amino side chains reduced risk of cognitive impairment and development of AD in the binding pocket [359]. Thus, SARMs are predicted to exert [370]. tissue-specific effects dependent in part upon how they interact Currently, effort is being focused on next generation SERMs that with amino acid sidechains in the AR binding pocket exhibit more robust and specific neuroprotective actions [42,303]. [36,46,201]. Recent research indicates success in developing For example, Brinton and colleagues recently developed a synthetic SARMs with desired tissue specificity [245]. As with SERMs, the SERM with both estrogenic and antioxidant potential that protects development of brain-specific SARMs that exert neuroprotective cultured neurons from cell death [389]. Evaluation of such com- actions associated with reduction of AD pathogenesis is a key pounds in animal models of AD is only beginning. Towards this topic of ongoing research. end, our laboratory has begun to investigate two particular SERMs, propylpyrazole triol (PPT) and diarylpropionitrile (DPN), which Acknowledgments show relative specificity for ERa and ERb, respectively. Although we observe that both compounds protect cultured neurons from This work was supported by NIH Grants AG026572 and Ab toxicity [67], PPT but not DPN treatments effectively mimicked AG23739. JCC was supported by NIH Grant F31NS059174. E2 in reducing Ab accumulation and improving behavioral deficits in OVX 3xTg-AD mice [53]. References In parallel to the ongoing research on SERMs, there has been a recent growth in the interest and development of tissue-spe- [1] K.K. Aenlle, A. Kumar, L. Cui, T.C. Jackson, T.C. Foster, Estrogen effects on cific selective androgen receptor modulators (SARMs) cognition and hippocampal transcription in middle-aged mice, Neurobiol. [108,147,186]. Given the prevalence of ADAM and its widespread Aging (2007). [2] E. Ahlbom, G.S. Prins, S. Ceccatelli, Testosterone protects cerebellar granule deleterious consequences, androgen based therapy is of consider- cells from oxidative stress-induced cell death through a receptor mediated able interest. However, prostate cancer, which is the second mechanism, Brain Res. 892 (2001) 255–262. leading cancer among aging men in terms of both prevalence [3] H. Akiyama, S. Barger, S. Barnum, B. Bradt, J. Bauer, G.M. Cole, N.R. Cooper, P. Eikelenboom, M. Emmerling, B.L. Fiebich, Inflammation and Alzheimer’s and cause of death [99], is androgen-dependent and typically disease, Neurobiol. Aging 21 (2000) 383–421. treated by androgen deprivation therapy. Consequently, the use [4] G.M. Alexander, R.S. Swerdloff, C. Wang, T. Davidson, V. McDonald, B. Steiner, of testosterone therapy with its potential to increase risk and M. Hines, Androgen-behavior correlations in hypogonadal men and or progression of prostate tumorigenesis has been controversial eugonadal men. II. Cognitive abilities, Horm. Behav. 33 (1998) 85–94. [5] N.J. Alkayed, S.J. Murphy, R.J. Traystman, P.D. Hurn, V.M. Miller, [183] and promoted the development of SARMs that lack signif- Neuroprotective effects of female gonadal steroids in reproductively icant androgen action in prostate but exert agonist effects in se- senescent female rats, Stroke 31 (2000) 161–168. C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 251

[6] O.P. Almeida, L. Flicker, Testosterone and dementia: too much ado about too [32] H.A. Bimonte-Nelson, M.E. Nelson, A.C. Granholm, Progesterone counteracts little data, J. Br. Menopause Soc. 9 (2003) 107–110. estrogen-induced increases in neurotrophins in the aged female rat brain, [7] O.P. Almeida, N.T. Lautenschlager, S. Vasikaran, P. Leedman, A. Gelavis, L. Neuroreport 15 (2004) 2659–2663. Flicker, A 20-week randomized controlled trial of estradiol replacement [33] H.A. Bimonte-Nelson, K.R. Francis, C.D. Umphlet, A.C. Granholm, Progesterone therapy for women aged 70 years and older: effect on mood, cognition and reverses the spatial memory enhancements initiated by tonic and cyclic quality of life, Neurobiol. Aging 27 (2006) 141–149. oestrogen therapy in middle-aged ovariectomized female rats, Eur. J. [8] M. Alvarez-De-La-Rosa, I. Silva, J. Nilsen, M.M. Perez, L.M. Garcia-Segura, J. Neurosci. 24 (2006) 229–242. Avila, F. Naftolin, Estradiol prevents neural Tau hyperphosphorylation [34] E.F. Binder, K.B. Schechtman, S.J. Birge, D.B. Williams, W.M. Kohrt, Effects of characteristic of Alzheimer’s disease, Ann. NY Acad. Sci. 1052 (2005) 210–224. hormone replacement therapy on cognitive performance in elderly women, [9] K. Andersen, L.J. Launer, M.E. Dewey, L. Letenneur, A. Ott, J.R.M. Copeland, J.-F. Maturitas 38 (2001) 137–146. Dartigues, P. Kragh-Sorensen, M. Baldereschi, C. Brayne, A. Lobo, J.M. [35] J. Bohacek, A.M. Bearl, J.M. Daniel, Long-term ovarian hormone deprivation Martinez-Lage, T. Stijnen, A. Hofman, Gender differences in the incidence of alters the ability of subsequent oestradiol replacement to regulate choline AD and vascular dementia: the EURODEM studies, Neurology 53 (1999) acetyltransferase protein levels in the hippocampus and prefrontal cortex of 1992–1997. middle-aged rats, J. Neuroendocrinol. 20 (2008) 1023–1027. [10] R.A. Anderson, C.W. Martin, A.W. Kung, D. Everington, T.C. Pun, K.C. Tan, J. [36] C.E. Bohl, D.D. Miller, J. Chen, C.E. Bell, J.T. Dalton, Structural basis for Bancroft, K. Sundaram, A.J. Moo-Young, D.T. Baird, 7a-Methyl-19- accommodation of nonsteroidal ligands in the androgen receptor, J. Biol. nortestosterone maintains sexual behavior and mood in hypogonadal men, Chem. 280 (2005) 37747–37754. J. Clin. Endocrinol. Metab. 84 (1999) 3556–3562. [37] H. Braak, E. Braak, Alzheimer’s disease: striatal amyloid deposits and [11] M.A. Ansonoff, A.M. Etgen, Estradiol elevates protein kinase C catalytic neurofibrillary changes, J. Neuropathol. Exp. Neurol. 49 (1990) 215– activity in the preoptic area of female rats, Endocrinology 139 (1998) 3050– 224. 3056. [38] D.W. Brann, K. Dhandapani, C. Wakade, V.B. Mahesh, M.M. Khan, [12] B. Antonsson, J.C. Martinou, The Bcl-2 protein family, Exp. Cell. Res. 256 Neurotrophic and neuroprotective actions of estrogen: basic mechanisms (2000) 50–57. and clinical implications, Steroids 72 (2007) 381–405. [13] C.S. Atwood, S.V. Meethal, T. Liu, A.C. Wilson, M. Gallego, M.A. Smith, R.L. [39] C. Brayne, C. Gill, F.A. Huppert, C. Barkley, E. Gehlhaar, D.M. Girling, D.W. Bowen, Dysregulation of the hypothalamic–pituitary–gonadal axis with O’Connor, E.S. Paykel, Incidence of clinically diagnosed subtypes of dementia menopause and andropause promotes neurodegenerative senescence, J. in an elderly population. Cambridge Project for Later Life, Br. J. Psychiatry 167 Neuropathol. Exp. Neurol. 64 (2005) 93–103. (1995) 255–262. [14] I. Azcoitia, A. Sierra, L.M. Garcia-Segura, Estradiol prevents kainic acid- [40] D.E. Brenner, W.A. Kukull, A. Stergachis, G. van Belle, J.D. Bowen, W.C. induced neuronal loss in the rat dentate gyrus, Neuroreport 9 (1998) 3075– McCormick, L. Teri, E.B. Larson, Postmenopausal estrogen replacement 3079. therapy and the risk of Alzheimer’s disease: a population-based case- [15] I. Azcoitia, A. Sierra, L.M. Garcia_Segura, Neuroprotective effects of estradiol control study, Am. J. Epidemiol. 140 (1994) 262–267. in the adult rat hippocampus: interaction with insulin-like growth factor-I [41] B. Breuer, R. Anderson, The relationship of tamoxifen with dementia, signalling., J. Neurosci. Res. 58 (1999) 815–822. depression, and dependence in activities of daily living in elderly nursing [16] I. Azcoitia, A. Sierra, S. Veiga, S. Honda, N. Harada, L.M. Garcia-Segura, Brain home residents, Women Health 31 (2000) 71–85. aromatase is neuroprotective, J. Neurobiol. 47 (2001) 318–329. [42] R.D. Brinton, Requirements of a brain selective estrogen: advances and [17] D.L. Bachman, P.A. Wolf, R. Linn, J.E. Knoefel, J. Cobb, A. Belanger, R.B. remaining challenges for developing a NeuroSERM, J. Alzheimers Dis. 6 D’Agostino, L.R. White, Prevalence of dementia and probable senile dementia (2004) S27–35. of the Alzheimer type in the Framingham study, Neurology 42 (1992) 115– [43] R.D. Brinton, S. Chen, M. Montoya, D. Hsieh, J. Minaya, The estrogen 119. replacement therapy of the Women’s Health Initiative promotes the [18] S. Bake, F. Sohrabji, 17Beta-estradiol differentially regulates blood–brain cellular mechanisms of memory and neuronal survival in neurons barrier permeability in young and aging female rats, Endocrinology 145 vulnerable to Alzheimer’s disease, Maturitas 34 (Suppl. 2) (2000) S35–52. (2004) 5471–5475. [44] R.D. Brinton, R.F. Thompson, M.R. Foy, M. Baudry, J. Wang, C.E. Finch, T.E. [19] S. Bake, L. Ma, F. Sohrabji, Estrogen receptor-alpha overexpression suppresses Morgan, C.J. Pike, W.J. Mack, F.Z. Stanczyk, J. Nilsen, Progesterone 17beta-estradiol-mediated vascular endothelial growth factor expression and receptors: form and function in brain, Front. Neuroendocrinol. 29 (2008) activation of survival kinases, Endocrinology 149 (2008) 3881–3889. 313–339. [20] L.D. Baker, K. Sambamurti, S. Craft, M. Cherrier, M.A. Raskind, F.Z. Stanczyk, [45] B.P. Brooks, D.E. Merry, H.L. Paulson, A.P. Lieberman, D.L. Kolson, K.H. S.R. Plymate, S. Asthana, 17Beta-estradiol reduces plasma Abeta40 for HRT- Fischbeck, A cell culture model for androgen effects in motor neurons, J. naive postmenopausal women with Alzheimer disease: a preliminary study, Neurochem. 70 (1998) 1054–1060. Am J. Geriat. Psychiatry 11 (2003) 239–244. [46] T.R. Brown, Nonsteroidal selective androgen receptors modulators (SARMs): [21] L.L. Barnes, R.S. Wilson, J.L. Bienias, J.A. Schneider, D.A. Evans, D.A. Bennett, designer androgens with flexible structures provide clinical promise, Sex differences in the clinical manifestations of Alzheimer disease pathology, Endocrinology 145 (2004) 5417–5419. Arch. Gen. Psychiatry 62 (2005) 685–691. [47] A.J. Bruce-Keller, J.L. Keeling, J.N. Keller, F.F. Huang, S. Camondola, M.P. [22] E. Barret-Connor, D. Kritz-Silverstein, Estrogen replacement therapy and Mattson, Antiinflammatory effects of estrogen on microglial activation, cognitive function in older women, J. Am. Med. Assoc. 269 (1993) 2637–2641. Endocrinology 141 (2000) 3646–3656. [23] E. Barrett-Connor, D. Goodman-Gruen, B. Patay, Endogenous sex hormones [48] H.U. Bryant, W.H. Dere, Selective estrogen receptor modulators: an and cognitive function in older men, J. Clin. Endocrinol. Metab. 84 (1999) alternative to hormone replacement therapy, Proc. Soc. Exp. Biol. Med. 217 3681–3685. (1998) 45–52. [24] R.N. Baumgartner, D.L. Waters, D. Gallagher, J.E. Morley, P.J. Garry, Predictors [49] J.G. Buckwalter, E. Sobel, M.E. Dunn, M.M. Diz, V.W. Henderson, Gender of skeletal muscle mass in elderly men and women, Mech. Ageing Dev. 107 differences on a brief measure of cognitive functioning in Alzheimer’s disease, (1999) 123–136. Arch. Neurol. 50 (1993) 757–760. [25] C. Behl, J.B. Davis, R. Lesley, D. Schubert, Hydrogen peroxide mediates [50] H. Burger, C.E. de Laet, P.L. van Daele, A.E. Weel, J.C. Witteman, A. Hofman, amyloid b protein toxicity, Cell 77 (1994) 817–827. H.A. Pols, Risk factors for increased bone loss in an elderly population: the [26] C. Behl, M. Widmann, T. Trapp, F. Holsboer, 17-Beta estradiol protects Rotterdam study, Am. J. Epidemiol. 147 (1998) 871–879. neurons from oxidative stress-induced cell death in vitro, Biochem. Biophys. [51] M.J. Callahan, W.J. Lipinski, F. Bian, R.A. Durham, A. Pack, L.C. Walker, Res. Commun. 216 (1995) 473–482. Augmented senile plaque load in aged female beta-amyloid precursor [27] C. Behl, T. Trapp, T. Skutella, F. Holsboer, Protection against oxidative stress- protein-transgenic mice, Am. J. Pathol. 158 (2001) 1173–1177. induced neuronal cell death – a novel role for RU486, Eur. J. Neurosci. 9 [52] M.C. Carlson, P.P. Zandi, B.L. Plassman, J.T. Tschanz, K.A. Welsh-Bohmer, D.C. (1997) 912–920. Steffens, L.A. Bastian, K.M. Mehta, J.C. Breitner, Hormone replacement therapy [28] S. Benvenuti, P. Luciani, G.B. Vannelli, S. Gelmini, E. Franceschi, M. Serio, A. and reduced cognitive decline in older women: the Cache County Study, Peri, Estrogen and selective estrogen receptor modulators exert Neurology 57 (2001) 2210–2216. neuroprotective effects and stimulate the expression of selective [53] J.C. Carroll, C.J. Pike, Selective estrogen receptor modulators differentially Alzheimer’s disease indicator-1, a recently discovered antiapoptotic gene, in regulate Alzheimer-like changes in female 3xTg-AD mice, Endocrinology 149 human neuroblast long-term cell cultures, J. Clin. Endocrinol. Metab. 90 (2008) 2607–2611. (2005) 1775–1782. [54] J.C. Carroll, E.R. Rosario, L. Chang, F.Z. Stanczyk, S. Oddo, F.M. LaFerla, C.J. Pike, [29] I. Bezprozvanny, M.P. Mattson, Neuronal calcium mishandling and the Progesterone and estrogen regulate Alzheimer-like neuropathology in female pathogenesis of Alzheimer’s disease, Trends Neurosci. 31 (2008) 454–463. 3xTg-AD mice, J. Neurosci. 27 (2007) 13357–13365. [30] S. Bhasin, L. Woodhouse, R. Casaburi, A.B. Singh, D. Bhasin, N. Berman, X. [55] J.C. Carroll, E.R. Rosario, C.J. Pike, Progesterone blocks estrogen Chen, K.E. Yarasheski, L. Magliano, C. Dzekov, J. Dzekov, R. Bross, J. Phillips, I. neuroprotection from kainate in middle-aged female rats, Neurosci. Lett. Sinha-Hikim, R. Shen, T.W. Storer, Testosterone dose-response relationships 445 (2008) 229–232. in healthy young men, Am. J. Physiol. Endocrinol. Metab. 281 (2001) E1172– [56] C. Castelo-Branco, F. Figueras, A. Sanjuan, F. Pons, J.J. Vicente, J.A. Vanrell, 1181. Long-term postmenopausal hormone replacement therapy effects on bone [31] S. Bhasin, G.R. Cunningham, F.J. Hayes, A.M. Matsumoto, P.J. Snyder, R.S. mass: differences between surgical and spontaneous patients, Eur. J. Obstet. Swerdloff, V.M. Montori, Testosterone therapy in adult men with androgen Gynecol. Reprod. Biol. 83 (1999) 207–211. deficiency syndromes: an endocrine society clinical practice guideline, J. Clin. [57] H.S. Chae, J.H. Bach, M.W. Lee, H.S. Kim, Y.S. Kim, K.Y. Kim, K.Y. Choo, S.H. Endocrinol. Metab. 91 (2006) 1995–2010. Choi, C.H. Park, S.H. Lee, Y.H. Suh, S.S. Kim, W.B. Lee, Estrogen attenuates cell death induced by carboxy-terminal fragment of amyloid precursor protein in 252 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258

PC12 through a receptor-dependent pathway, J. Neurosci. Res. 65 (2001) 403– gene expression by 17beta-estradiol in breast cancer cells, J. Biol. Chem. 274 407. (1999) 32099–32107. [58] T.R. Chakraborty, A.C. Gore, Aging-related changes in ovarian hormones, their [84] R. Duara, W.W. Barker, R. Lopez-Alberola, D.A. Loewenstein, L.B. Grau, D. receptors, and neuroendocrine function, Exp. Biol. Med. (Maywood) 229 Gilchrist, S. Sevush, S. St George-Hyslop, Alzheimer’s disease: interaction of (2004) 977–987. apolipoprotein E genotype, family history of dementia, gender, education, [59] D. Chang, J. Kwan, P.S. Timiras, Estrogens influence growth, maturation, and ethnicity, and age of onset, Neurology 46 (1996) 1575–1579. amyloid beta-peptide production in neuroblastoma cells and in a beta-APP [85] D.B. Dubal, P.M. Wise, Neuroprotective effects of estradiol in middle-aged transfected kidney 293 cell line, Adv. Exp. Med. Biol. 429 (1997) 261–271. female rats, Endocrinology 142 (2001) 43–48. [60] P. Chaovipoch, K.A. Jelks, L.M. Gerhold, E.J. West, S. Chongthammakun, C.L. [86] D.B. Dubal, P.J. Shughrue, M.E. Wilson, I. Merchenthaler, P.M. Wise, Estradiol Floyd, 17Beta-estradiol is protective in spinal cord injury in post- and pre- modulates Bcl-2 in cerebral ischemia: a potential role for estrogen receptors, menopausal rats, J. Neurotrauma. 23 (2006) 830–852. J. Neurosci. 19 (1999) 6385–6393. [61] I.H. Cheng, K. Scearce-Levie, J. Legleiter, J.J. Palop, H. Gerstein, N. Bien-Ly, J. [87] K.L. Edinger, C.A. Frye, Testosterone’s analgesic, anxiolytic, and Puolivali, S. Lesne, K.H. Ashe, P.J. Muchowski, L. Mucke, Accelerating amyloid- cognitive-enhancing effects may be due in part to actions of its beta fibrillization reduces oligomer levels and functional deficits in Alzheimer 5alpha-reduced metabolites in the hippocampus, Behav. Neurosci. 118 disease mouse models, J. Biol. Chem. 282 (2007) 23818–23828. (2004) 1352–1364. [62] M.M. Cherrier, S. Asthana, S. Plymate, L. Baker, A.M. Matsumoto, E. Peskind, [88] K.L. Edinger, C.A. Frye, Testosterone’s anti-anxiety and analgesic effects may M.A. Raskind, K. Brodkin, W. Bremner, A. Petrova, S. LaTendresse, S. Craft, be due in part to actions of its 5alpha-reduced metabolites in the Testosterone supplementation improves spatial and verbal memory in hippocampus, Psychoneuroendocrinology 30 (2005) 418–430. healthy older men, Neurology 57 (2001) 80–88. [89] J. El Khoury, S.E. Hickman, C.A. Thomas, L. Cao, S.C. Silverstein, J.D. Loike, [63] M.M. Cherrier, A.M. Matsumoto, J.K. Amory, S. Asthana, W. Bremner, E.R. Scavenger receptor-mediated adhesion of microglia to beta-amyloid fibrils, Peskind, M.A. Raskind, S. Craft, Testosterone improves spatial memory in men Nature 382 (1996) 716–719. with Alzheimer disease and mild cognitive impairment, Neurology 64 (2005) [90] M.H. Emmelot-Vonk, H.J. Verhaar, H.R. Nakhai Pour, A. Aleman, T.M. Lock, J.L. 2063–2068. Bosch, D.E. Grobbee, Y.T. van der Schouw, Effect of testosterone [64] D.W. Choi, Ionic dependence of glutamate neurotoxicity, J. Neurosci. 7 (1987) supplementation on functional mobility, cognition, and other parameters in 369–379. older men: a randomized controlled trial, Jama 299 (2008) 39–52. [65] P. Ciana, G. Di Luccio, S. Belcredito, G. Pollio, E. Vegeto, L. Tatangelo, C. [91] M.A. Espeland, S.R. Rapp, S.A. Shumaker, R. Brunner, J.E. Manson, B.B. Sherwin, Tiveron, A. Maggi, Engineering of a mouse for the in vivo profiling of estrogen J. Hsia, K.L. Margolis, P.E. Hogan, R. Wallace, M. Dailey, R. Freeman, J. Hays, receptor activity, Mol. Endocrinol. 15 (2001) 1104–1113. Conjugated equine estrogens and global cognitive function in [66] E.H. Corder, E. Ghebremedhin, M.G. Taylor, D.R. Thal, T.G. Ohm, H. Braak, The postmenopausal women: Women’s Health Initiative Memory Study, Jama biphasic relationship between regional brain senile plaque and 291 (2004) 2959–2968. neurofibrillary tangle distributions: modification by age, sex, and APOE [92] M. Estrada, P. Uhlen, B.E. Ehrlich, Ca2+ oscillations induced by testosterone polymorphism, Ann. NY Acad. Sci. 1019 (2004) 24–28. enhance neurite outgrowth, J. Cell Sci. 119 (2006) 733–743. [67] M. Cordey, C.J. Pike, Neuroprotective properties of selective estrogen receptor [93] L.A. Farrer, L.A. Cupples, J.L. Haines, B. Hyman, W.A. Kukull, R. Mayeux, R.H. agonists in cultured neurons, Brain Res. 1045 (2005) 217–223. Myers, M.A. Pericak-Vance, N. Risch, C.M. van Duijn, Effects of age, sex, and [68] M. Cordey, C.J. Pike, Conventional protein kinase C isoforms mediate ethnicity on the association between apolipoprotein E genotype and neuroprotection induced by phorbol ester and estrogen, J. Neurochem. 96 Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta (2006) 204–217. Analysis Consortium, Jama 278 (1997) 1349–1356. [69] M. Cordey, U. Gundimeda, R. Gopalakrishna, C.J. Pike, Estrogen activates [94] H.A. Feldman, C. Longcope, C.A. Derby, C.B. Johannes, A.B. Araujo, A.D. protein kinase C in neurons: role in neuroprotection, J. Neurochem. 84 (2003) Coviello, W.J. Bremner, J.B. McKinlay, Age trends in the level of serum 1340–1348. testosterone and other hormones in middle-aged men: longitudinal results [70] S. Cory, J.M. Adams, The Bcl2 family: regulators of the cellular life-or-death from the Massachusetts male aging study, J. Clin. Endocrinol. Metab. 87 switch, Nat. Rev. Cancer 2 (2002) 647–656. (2002) 589–598. [71] C.W. Cotman, A.J. Anderson, A potential role for apoptosis in [95] L.S. Felicio, J.F. Nelson, C.E. Finch, Longitudinal studies of estrous cyclicity in neurodegeneration and Alzheimer’s disease, Mol. Neurobiol. 10 (1995) 19– aging C57BL/6J mice: II. Cessation of cyclicity and the duration of persistent 45. vaginal cornification, Biol. Reprod. 31 (1984) 446–453. [72] C.W. Cotman, C.J. Pike, A. Copani, Beta-Amyloid neurotoxicity: a discussion of [96] A.A. Ferrando, M. Sheffield-Moore, C.W. Yeckel, C. Gilkison, J. Jiang, A. in vitro findings, Neurobiol. Aging 13 (1992) 587–590. Achacosa, S.A. Lieberman, K. Tipton, R.R. Wolfe, R.J. Urban, Testosterone [73] C.W. Cotman, D.H. Cribbs, C.J. Pike, K.J. Ivins, Cell death in Alzheimer’s death. administration to older men improves muscle function: molecular and When Cells Die: A Comprehensive Evaluation of Apoptosis and Programmed physiological mechanisms, Am. J. Physiol. Endocrinol. Metab. 282 (2002) Cell Death, 1998, pp. 385–409. E601–607. [74] C.M. Coughlan, K.C. Breen, Factors influencing the processing and function of [97] J.L. Fitzpatrick, A.L. Mize, C.B. Wade, J.A. Harris, R.A. Shapiro, D.M. Dorsa, the amyloid beta precursor protein — a potential therapeutic target in Estrogen-mediated neuroprotection against beta-amyloid toxicity requires Alzheimer’s disease?, Pharmacol Ther. 86 (2000) 111–144. expression of estrogen receptor alpha or beta and activation of the MAPK [75] M.C. Craig, P.M. Maki, D.G. Murphy, The Women’s Health Initiative memory pathway, J. Neurochem. 82 (2002) 674–682. study: findings and implications for treatment, Lancet Neurol. 4 (2005) 190– [98] A. Fleisher, M. Grundman, C.R. Jack Jr., R.C. Petersen, C. Taylor, H.T. Kim, 194. D.H. Schiller, V. Bagwell, D. Sencakova, M.F. Weiner, C. DeCarli, S.T. [76] D.E. Cummings, N. Kumar, C.W. Bardin, K. Sundaram, W.J. Bremner, Prostate- DeKosky, C.H. van Dyck, L.J. Thal, Sex, apolipoprotein E epsilon 4 status, sparing effects in primates of the potent androgen 7alpha-methyl-19- and hippocampal volume in mild cognitive impairment, Arch. Neurol. 62 nortestosterone: a potential alternative to testosterone for androgen (2005) 953–957. replacement and male contraception, J. Clin. Endocrinol. Metab. 83 (1998) [99] N. Fleshner, A.R. Zlotta, Prostate cancer prevention: past, present, and future, 4212–4219. Cancer 110 (2007) 1889–1899. [77] D. Dai, D.M. Wolf, E.S. Litman, M.J. White, K.K. Leslie, Progesterone inhibits [100] M. Frankfurt, E. Gould, C.S. Woolley, B.S. McEwen, Gonadal steroids modify human endometrial cancer cell growth and invasiveness: down-regulation of dendritic spine density in ventromedial hypothalamic neurons: a Golgi study cellular adhesion molecules through progesterone B receptors, Cancer Res. 62 in the adult rat, Neuroendocrinology 51 (1990) 530–535. (2002) 881–886. [101] L. Fratiglioni, M. Viitanen, E. von Strauss, V. Tontodonati, A. Herlitz, B. [78] J.M. Daniel, J.L. Hulst, J.L. Berbling, Estradiol replacement enhances working Winblad, Very old women at highest risk of dementia and Alzheimer’s memory in middle-aged rats when initiated immediately after ovariectomy disease: incidence data from the Kungsholmen Project, Stockholm, Neurology but not after a long-term period of ovarian hormone deprivation, 48 (1997) 132–138. Endocrinology 147 (2006) 607–614. [102] C.A. Frye, T.A. Reed, Androgenic neurosteroids: anti-seizure effects in an [79] S. Davies, D. Dai, D.M. Wolf, K.K. Leslie, Immunomodulatory and animal model of epilepsy, Psychoneuroendocrinology 23 (1998) 385–399. transcriptional effects of progesterone through progesterone A and B [103] C.A. Frye, A.M. Seliga, Testosterone increases analgesia, anxiolysis, and receptors in Hec50co poorly differentiated endometrial cancer cells, J. Soc. cognitive performance of male rats, Cogn. Affect. Behav. Neurosci. 1 (2001) Gynecol. Investig. 11 (2004) 494–499. 371–381. [80] S. Del Cerro, J. Garcia-Estrada, L.M. Garcia-Segura, Neuroactive steroids [104] C.A. Frye, A.A. Walf, Effects of progesterone administration and regulate astroglia morphology in hippocampal cultures from adult rats, Glia APPswe+PSEN1Deltae9 mutation for cognitive performance of mid-aged 14 (1995) 65–71. mice, Neurobiol. Learn. Mem. 89 (2008) 17–26. [81] P.D. Delmas, N.H. Bjarnason, B.H. Mitlak, A.C. Ravoux, A.S. Shah, W.J. Huster, [105] E. Fukuzaki, K. Takuma, Y. Funatsu, Y. Himeno, Y. Kitahara, B. Gu, H. M. Draper, C. Christiansen, Effects of raloxifene on bone mineral density, Mizoguchi, D. Ibi, K. Koike, M. Inoue, S.D. Yan, K. Yamada, Ovariectomy serum cholesterol concentrations, and uterine endometrium in increases neuronal amyloid-beta binding alcohol dehydrogenase level in the postmenopausal women, New Engl. J. Med. 337 (1997) 1641–1647. mouse hippocampus, Neurochem. Int. 52 (2008) 1358–1364. [82] D.E. Dluzen, Effects of testosterone upon MPTP-induced neurotoxicity of the [106] S. Gandy, S. Petanceska, Regulation of alzheimer beta-amyloid precursor nigrostriatal dopaminergic system of C57/B1 mice, Brain Res. 715 (1996) trafficking and metabolism, Adv. Exp. Med. Biol. 487 (2001) 85–100. 113–118. [107] S. Gandy, O.P. Almeida, J. Fonte, D. Lim, A. Waterrus, N. Spry, L. Flicker, R.N. [83] L. Dong, W. Wang, F. Wang, M. Stoner, J.C. Reed, M. Harigai, I. Samudio, M.P. Martins, Chemical andropause and amyloid-beta peptide, J. Am. Med. Assoc. Kladde, C. Vyhlidal, S. Safe, Mechanisms of transcriptional activation of Bcl-2 285 (2001) 2195–2196. C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 253

[108] W. Gao, P.J. Reiser, C.C. Coss, M.A. Phelps, J.D. Kearbey, D.D. Miller, J.T. Dalton, utilization in human SK-N-SH neuroblastoma cells, Brain Res. 778 (1997) Selective androgen receptor modulator treatment improves muscle strength 158–165. and body composition and prevents bone loss in orchidectomized rats, [136] C. Haass, D.J. Selkoe, Soluble protein oligomers in neurodegeneration: lessons Endocrinology 146 (2005) 4887–4897. from the Alzheimer’s amyloid beta-peptide, Nat. Rev. Mol. Cell Biol. 8 (2007) [109] J. Garcia-Estrada, J.A. Del Rio, S. Luquin, E. Soriano, L.M. Garcia-Segura, 101–112. Gonadal hormones down-regulate reactive gliosis and astrocyte proliferation [137] O. Hagnell, L. Ojesjo, B. Rorsman, Incidence of dementia in the Lundby study, after a penetrating brain injury, Brain Res. 628 (1993) 271–278. Neuroepidemiology 11 (Suppl. 1) (1992) 61–66. [110] L.M. Garcia-Segura, P. Cardona-Gomez, F. Naftolin, J.A. Chowen, Estradiol [138] J. Hammond, Q. Le, C. Goodyer, M. Gelfand, M. Trifiro, A. LeBlanc, upregulates Bcl-2 expression in adult brain neurons, Neuroreport 9 (1998) Testosterone-mediated neuroprotection through the androgen receptor in 593–597. human primary neurons, J. Neurochem. 77 (2001) 1319–1326. [111] L.M. Garcia-Segura, A. Wozniak, I. Azcoitia, J.R. Rodriguez, R.E. Hutchison, J.B. [139] J. Hardy, The Alzheimer family of diseases: many etiologies, one Hutchison, Aromatase expression by astrocytes after brain injury: pathogenesis?, Proc Natl. Acad. Sci. USA 94 (1997) 2095–2097. implications for local estrogen formation in brain repair, Neuroscience 89 [140] J.A. Hardy, G.A. Higgins, Alzheimer’s disease: the amyloid cascade hypothesis, (1999) 567–578. Science 256 (1992) 184–185. [112] J.W. Gatson, M. Singh, Activation of a membrane-associated androgen [141] M.T. Haren, G.A. Wittert, I.M. Chapman, P. Coates, J.E. Morley, Effect of oral receptor promotes cell death in primary cortical astrocytes, Endocrinology testosterone undecanoate on visuospatial cognition, mood and quality of life 148 (2007) 2458–2464. in elderly men with low-normal gonadal status, Maturitas 50 (2005) 124– [113] J.W. Gatson, P. Kaur, M. Singh, Dihydrotestosterone differentially modulates 133. the mitogen-activated protein kinase and the phosphoinositide 3-kinase/Akt [142] M.T. Haren, M.J. Kim, S.H. Tariq, G.A. Wittert, J.E. Morley, Andropause: a pathways through the nuclear and novel membrane androgen receptor in C6 quality-of-life issue in older males, Med. Clin. North Am. 90 (2006) 1005– cells, Endocrinology 147 (2006) 2028–2034. 1023. [114] D.M. Gee, K. Flurkey, C.E. Finch, Aging and the regulation of luteinizing [143] M.E. Harris-White, T. Chu, S.A. Miller, M. Simmons, B. Teter, D. Nash, G.M. hormone in C57BL/6J mice: impaired elevations after ovariectomy and Cole, S.A. Frautschy, Estrogen (E2) and glucocorticoid (Gc) effects on spontaneous elevations at advanced ages, Biol. Reprod. 28 (1983) 598–607. microglia and Ab clearance in vitro and in vivo, Neurochem. Int. 39 (2001) [115] R.B. Gibbs, Effects of gonadal hormone replacement on measures of basal 435–448. forebrain cholinergic function, Neuroscience 101 (2000) 931–938. [144] S.A. Hart, J.D. Patton, C.S. Woolley, Quantitative analysis of ERalpha and GAD [116] R.B. Gibbs, Oestrogen and the cholinergic hypothesis: implications for colocalization in the hippocampus of the adult female rat, J. Comp. Neurol. oestrogen replacement therapy in postmenopausal women, Novartis Found 440 (2001) 144–155. Symp 230 (2000) 94–107; discussion 107–111. [145] D.M. Hartley, D.M. Walsh, C.P. Ye, T. Diehl, S. Vasquez, P.M. Vassilev, D.B. [117] M.J. Gillett, R.N. Martins, R.M. Clarnette, S.A. Chubb, D.G. Bruce, B.B. Yeap, Teplow, D.J. Selkoe, Protofibrillar intermediates of amyloid beta-protein Relationship between testosterone, sex hormone binding globulin and induce acute electrophysiological changes and progressive neurotoxicity in plasma amyloid beta peptide 40 in older men with subjective memory loss cortical neurons, J. Neurosci. 19 (1999) 8876–8884. or dementia, J. Alzheimers Dis. 5 (2003) 267–269. [146] S.G. Haskell, E.D. Richardson, R.I. Horwitz, The effect of estrogen replacement [118] C.E. Gleason, C.M. Carlsson, S. Johnson, C. Atwood, S. Asthana, Clinical therapy on cognitive function in women: a critical review of the literature, J. pharmacology and differential cognitive efficacy of estrogen preparations, Clin. Epidemiol. 50 (1997) 1249–1264. Ann. NY Acad. Sci. 1052 (2005) 93–115. [147] B. He, J.T. Minges, L.W. Lee, E.M. Wilson, The FXXLF motif mediates androgen [119] J.A. Goebel, S.J. Birge, S.C. Price, J.M. Hanson, D.G. Fishel, Estrogen receptor-specific interactions with coregulators, J. Biol. Chem. 277 (2002) replacement therapy and postural stability in the elderly, Am. J. Otolaryng. 10226–10235. 16 (1995) 470–474. [148] T. Heikkinen, G. Kalesnykas, A. Rissanen, T. Tapiola, S. Iivonen, J. Wang, J. [120] M.S. Golub, S.L. Germann, M. Mercer, M.N. Gordon, D.G. Morgan, L.P. Mayer, Chaudhuri, H. Tanila, R. Miettinen, J. Puolivali, Estrogen treatment improves P.B. Hoyer, Behavioral consequences of ovarian atrophy and estrogen spatial learning in APP + PS1 mice but does not affect beta amyloid replacement in the APPswe mouse, Neurobiol. Aging 29 (2007) 1512–1523. accumulation and plaque formation, Exp. Neurol. 187 (2004) 105–117. [121] S. Goodenough, S. Engert, C. Behl, Testosterone stimulates rapid secretory [149] V.W. Henderson, Estrogen-containing hormone therapy and Alzheimer’s amyloid precursor protein release from rat hypothalamic cells via the disease risk: understanding discrepant inferences from observational and activation of the mitogen-activated protein kinase pathway, Neurosci. Lett. experimental research, Neuroscience 138 (2006) 1031–1039. 296 (2000) 49–52. [150] V. Henderson, J. Buckwalter, Cognitive deficits of men and women with [122] S. Goodenough, D. Schleusner, C. Pietrzik, T. Skutella, C. Behl, Glycogen Alzheimer’s disease, Neurology 44 (1994) 90–96. synthase kinase 3beta links neuroprotection by 17beta-estradiol to key [151] V.W. Henderson, A. Paganini-Hill, C.K. Emanuel, M.E. Dunn, J.G. Buckwalter, Alzheimer processes, Neuroscience 132 (2005) 581–589. Estrogen replacement therapy in older women. Comparisons between [123] Y. Goodman, M.P. Mattson, Secreted forms of b-amyloid precursor protein Alzheimer’s disease cases and nondemented control subjects, Arch. Neurol. protect hippocampal neurons against amyloid b-peptide-induced oxidative 51 (1994) 896–900. injury, Exp. Neurol. 128 (1994) 1–12. [152] V.W. Henderson, A. Paganini_Hill, B.L. Miller, R.J. Elble, P.F. Reyes, D. Shoupe, [124] Y. Goodman, A.J. Bruce, B. Cheng, M.P. Mattson, Estrogens attenuate and C.A. McCleary, R.A. Klein, A.M. Hake, M.R. Farlow, Estrogen for Alzheimer’s corticosterone exacerbates excitotoxicity, oxidative injury, and amyloid beta- disease in women: randomized, double-blind, placebo-controlled trial, peptide toxicity in hippocampal neurons, J. Neurochem. 66 (1996) 1836– Neurology 54 (2000) 295–301. 1844. [153] V.W. Henderson, J.R. Guthrie, E.C. Dudley, H.G. Burger, L. Dennerstein, [125] L. Gooren, Androgen deficiency in the aging male: benefits and risks of Estrogen exposures and memory at midlife: a population-based study of androgen supplementation, J. Steroid. Biochem. Mol. Biol. 85 (2003) 349–355. women, Neurology 60 (2003) 1369–1371. [126] R.G. Gosden, S.C. Laing, L.S. Felicio, J.F. Nelson, C.E. Finch, Imminent oocyte [154] V.W. Henderson, K.S. Benke, R.C. Green, L.A. Cupples, L.A. Farrer, exhaustion and reduced follicular recruitment mark the transition to Postmenopausal hormone therapy and Alzheimer’s disease risk: interaction acyclicity in aging C57BL/6J mice, Biol. Reprod. 28 (1983) 255–260. with age, J. Neurol. Neurosurg. Psychiatry 76 (2005) 103–105. [127] C. Gouchie, D. Kimura, The relationship between testosterone levels and [155] A. Hestiantoro, D.F. Swaab, Changes in estrogen receptor-alpha and -beta in cognitive ability patterns, Psychoneuroendocrinology 16 (1991) 323–334. the infundibular nucleus of the human hypothalamus are related to the [128] G.K. Gouras, H. Xu, R.S. Gross, J.P. Greenfield, B. Hai, R. Wang, P. Greengard, occurrence of Alzheimer’s disease neuropathology, J. Clin. Endocrinol. Metab. Testosterone reduces neuronal secretion of Alzheimer’s beta-amyloid 89 (2004) 1912–1925. peptides, PNAS 97 (2000) 1202–1205. [156] C. Hirata-Fukae, H.F. Li, H.S. Hoe, A.J. Gray, S.S. Minami, K. Hamada, T. Niikura, [129] D. Grady, T. Gebretsadik, K. Kerlikowske, V. Ernster, D. Petitti, Hormone F. Hua, H. Tsukagoshi-Nagai, Y. Horikoshi-Sakuraba, M. Mughal, G.W. Rebeck, replacement therapy and endometrial cancer risk: a meta-analysis, Obstet. F.M. LaFerla, M.P. Mattson, N. Iwata, T.C. Saido, W.L. Klein, K.E. Duff, P.S. Gynecol. 85 (1995) 304–313. Aisen, Y. Matsuoka, Females exhibit more extensive amyloid, but not tau, [130] J.D. Graham, C.L. Clarke, Physiological action of progesterone in target tissues, pathology in an Alzheimer transgenic model, Brain Res. 1216 (2008) 92– Endocr. Rev. 18 (1997) 502–519. 103. [131] A. Gray, H. Feldman, J. McKinlay, C. Longcope, Age, disease, and changing sex [157] M.A. Hlatky, D. Boothroyd, E. Vittinghoff, P. Sharp, M.A. Whooley, Quality-of- hormone levels in middle-aged men: results of the Massachusetts male aging life and depressive symptoms in postmenopausal women after receiving study, J. Clin. Endocrinol. Metab. 73 (1991) 1016–1025. hormone therapy: results from the Heart and Estrogen/Progestin [132] P.S. Green, K.E. Gridley, J.W. Simpkins, Estradiol protects against beta- Replacement Study (HERS) trial, Jama 287 (2002) 591–597. amyloid (25-35)-induced toxicity in SK-N-SH human neuroblastoma cells, [158] E. Hogervorst, M. Boshuisen, W. Riedel, C. Willeken, J. Jolles, 1998 Curt P. Neurosci. Lett. 218 (1996) 165–168. Richter Award. The effect of hormone replacement therapy on cognitive [133] P.S. Green, K. Bales, S. Paul, G.J. Bu, Estrogen therapy fails to alter amyloid function in elderly women, Psychoneuroendocrinology 24 (1999) 43–68. deposition in the PDAPP model of Alzheimer’s disease, Endocrinology 146 [159] E. Hogervorst, J. Williams, M. Biudge, L. Barnetson, M. Combrinck, A.D. Smith, (2005) 2774–2781. Serum total testosterone is lower in men with Alzheimer’s disease, [134] J.P. Greenfield, L.W. Leung, D. Cai, K. Kaasik, R.S. Gross, E. Rodriguez_Boulan, Neuroendocrinol. Lett. 22 (2001) 163–168. P. Greengard, H. Xu, Estrogen lowers Alzheimer beta-amyloid generation by [160] E. Hogervorst, D.J. Lehmann, D.R. Warden, J. McBroom, A.D. Smith, stimulating trans-Golgi network vesicle biogenesis, J. Biol. Chem. 277 (2002) Apolipoprotein E epsilon4 and testosterone interact in the risk of 12128–12136. Alzheimer’s disease in men, Int. J. Geriatr. Psych. 17 (2002) 938–940. [135] K.E. Gridley, P.S. Green, J.W. Simpkins, Low concentrations of estradiol reduce [161] E. Hogervorst, M. Combrinck, A.D. Smith, Testosterone and gonadotropin b-amyloid (25-35)-induced toxicity, lipid peroxidation and glucose levels in men with dementia, Neuroendocrinol. Lett. 24 (2003) 203–208. 254 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258

[162] H. Honjo, K. Iwasa, M. Kawata, S. Fushiki, T. Hosoda, H. Tatsumi, N. Oida, M. [188] J. Kim, D. Wu, D.J. Hwang, D.D. Miller, J.T. Dalton, The para substituent of S-3- Mihara, Y. Hirasugi, H. Yamamoto, N. Kikuchi, J. Kitawaki, Progestins and (phenoxy)-2-hydroxy-2-methyl-N-(4-nitro-3-trifluoromethyl-phenyl)-prop estrogens and Alzheimer’s disease, J. Steroid. Biochem. Mol. Biol. 93 (2005) ionamides is a major structural determinant of in vivo disposition and 305–308. activity of selective androgen receptor modulators, J. Pharmacol. Exp. Ther. [163] J. Huang, H. Guan, R.M. Booze, C.B. Eckman, L.B. Hersh, Estrogen regulates 315 (2005) 230–239. neprilysin activity in rat brain, Neurosci. Lett. 367 (2004) 85–87. [189] J.Y. Koh, L.L. Yang, C.W. Cotman, Beta-amyloid protein increases the [164] C.B. Huppenbauer, L. Tanzer, L.L. DonCarlos, K.J. Jones, Gonadal steroid vulnerability of cultured cortical neurons to excitotoxic damage, Brain Res. attenuation of developing hamster facial motoneuron loss by axotomy: equal 533 (1990) 315–320. efficacy of testosterone, dihydrotestosterone, and 17-beta estradiol, J. [190] E. Koo, Phorbol esters affect multiple steps in beta-amyloid precursor protein Neurosci. 25 (2005) 4004–4013. trafficking and amyloid beta-protein production, Mol. Med. 3 (1997) 204– [165] B.T. Hyman, T. Gomez-Isla, M. Briggs, H. Chung, S. Nichols, F. Kohout, R. 211. Wallace, Apolipoprotein E and cognitive change in an elderly population, [191] C.L. Koski, S. Hila, G.E. Hoffman, Regulation of cytokine-induced neuron death Ann. Neurol. 40 (1996) 55–66. by ovarian hormones: involvement of antiapoptotic protein expression and [166] K. Iqbal, C. Alonso Adel, S. Chen, M.O. Chohan, E. El-Akkad, C.X. Gong, S. c-JUN N-terminal kinase-mediated proapoptotic signaling, Endocrinology Khatoon, B. Li, F. Liu, A. Rahman, H. Tanimukai, I. Grundke-Iqbal, Tau 145 (2004) 95–103. pathology in Alzheimer disease and other tauopathies, Biochim. Biophys. [192] K.A. Kujawa, N.B. Kinderman, K.J. Jones, Testosterone-induced acceleration of Acta 1739 (2005) 198–210. recovery from facial paralysis following crush axotomy of the facial nerve in [167] R.W. Irwin, J. Yao, R.T. Hamilton, E. Cadenas, R.D. Brinton, J. Nilsen, male hamsters, Exp. Neurol. 105 (1989) 80–85. Progesterone and estrogen regulate oxidative metabolism in brain [193] K.A. Kujawa, E. Emeric, K.J. Jones, Testosterone differentially regulates the mitochondria, Endocrinology 149 (2008) 3167–3175. regenerative properties of injured hamster facial motoneurons, J. Neurosci. [168] U. Islander, M.C. Erlandsson, B. Hasseus, C.A. Jonsson, C. Ohlsson, J.A. 11 (1991) 3898–3906. Gustafsson, U. Dahlgren, H. Carlsten, Influence of oestrogen receptor alpha [194] K.A. Kujawa, J.M. Jacob, K.J. Jones, Testosterone regulation of the regenerative and beta on the immune system in aged female mice, Immunology 110 properties of injured rat sciatic motor neurons, J. Neurosci. Res. 35 (1993) (2003) 149–157. 268–273. [169] K.J. Ivins, E.T. Bui, C.W. Cotman, Beta-amyloid induces local neurite [195] K.A. Kujawa, L. Tanzer, K.J. Jones, Inhibition of the accelerative effects of degeneration in cultured hippocampal neurons: evidence for neuritic testosterone on hamster facial nerve regeneration by the antiandrogen apoptosis, Neurobiol. Dis. 5 (1998) 365–378. flutamide, Exp. Neurol. 133 (1995) 138–143. [170] N. Iwata, S. Tsubuki, Y. Takaki, K. Watanabe, M. Sekiguchi, E. Hosoki, M. [196] F.M. LaFerla, S. Oddo, Alzheimer’s disease: Abeta, tau and synaptic Kawashima-Morishima, H.J. Lee, E. Hama, Y. Sekine-Aizawa, T.C. Saido, dysfunction, Trends Mol. Med. 11 (2005) 170–176. Identification of the major Abeta1-42-degrading catabolic pathway in brain [197] E.S. LeBlanc, J. Janowsky, B.K.S. Chan, H.D. Nelson, Hormone replacement parenchyma: suppression leads to biochemical and pathological deposition, therapy and cognition: systematic review and meta-analysis, J. Am. Med. Nat. Med. 6 (2000) 143–150. Assoc. 285 (2001) 1489. [171] A.B. Jaffe, C.D. Toran_Allerand, P. Greengard, S.E. Gandy, Estrogen regulates [198] S. Lesne, M.T. Koh, L. Kotilinek, R. Kayed, C.G. Glabe, A. Yang, M. Gallagher, metabolism of Alzheimer amyloid beta precursor protein, J. Biol. Chem. 269 K.H. Ashe, A specific amyloid-beta protein assembly in the brain impairs (1994) 13065–13068. memory, Nature 440 (2006) 352–357. [172] J.S. Janowsky, S.K. Oviatt, E.S. Orwoll, Testosterone influences spatial [199] J.A. Levin-Allerhand, C.E. Lominska, J. Wang, J.D. Smith, 17Alpha-estradiol and cognition in older men, Behav. Neurosci. 108 (1994) 325–332. 17beta-estradiol treatments are effective in lowering cerebral amyloid-beta [173] J.S. Janowsky, B. Chavez, E. Orwoll, Sex steroids modify working memory, J. levels in AbetaPPSWE transgenic mice, J. Alzheimers Dis. 4 (2002) 449– Cogn. Neurosci. 12 (2000) 407–414. 457. [174] A. Jayaraman, C.J. Pike, Progesterone attenuates oestrogen neuroprotection [200] R. Li, Y. Shen, L.B. Yang, L.F. Lue, C. Finch, J. Rogers, Estrogen enhances uptake via downregulation of oestrogen receptor expression in cultured neurones, J. of amyloid beta-protein by microglia derived from the human cortex, J. Neuroendocrinol. 21 (2009) 77–81. Neurochem. 75 (2000) 1447–1454. [175] M.K. Jezierski, F. Sohrabji, Neurotrophin expression in the reproductively [201] J.J. Li, J.C. Sutton, A. Nirschl, Y. Zou, H. Wang, C. Sun, Z. Pi, R. Johnson, S.R. senescent forebrain is refractory to estrogen stimulation, Neurobiol. Aging 22 Krystek Jr., R. Seethala, R. Golla, P.G. Sleph, B.C. Beehler, G.J. Grover, A. Fura, (2001) 309–319. V.P. Vyas, C.Y. Li, J.Z. Gougoutas, M.A. Galella, R. Zahler, J. Ostrowski, L.G. [176] A.B. Johnson, S. Bake, D.K. Lewis, F. Sohrabji, Temporal expression of IL-1beta Hamann, Discovery of potent and muscle selective androgen receptor protein and mRNA in the brain after systemic LPS injection is affected by age modulators through Scaffold modifications, J. Med. Chem. 50 (2007) 3015– and estrogen, J. Neuroimmunol. 174 (2006) 82–91. 3025. [177] K.J. Jones, S. Coers, P.D. Storer, L. Tanzer, N.B. Kinderman, Androgenic [202] J.G. Liehr, D. Roy, Free radical generation by redox cycling of estrogens, Free regulation of the central glia response following nerve damage, J. Neurobiol. Rad. Biol. Med. 8 (1990) 415–423. 40 (1999) 560–573. [203] N. Linford, C. Wade, D. Dorsa, The rapid effects of estrogen are implicated in [178] K.J. Jones, T.J. Brown, M. Damaser, Neuroprotective effects of gonadal steroids estrogen-mediated neuroprotection, J. Neurocytol. 29 (2000) 367– on regenerating peripheral motoneurons, Brain Res. Brain Res. Rev. 37 (2001) 374. 372–382. [204] X.A. Liu, L.Q. Zhu, Q. Zhang, H.R. Shi, S.H. Wang, Q. Wang, J.Z. Wang, Estradiol [179] R.D. Jones, P.J. Pugh, J. Hall, K.S. Channer, T.H. Jones, Altered circulating attenuates tau hyperphosphorylation induced by upregulation of protein hormone levels, endothelial function and vascular reactivity in the testicular kinase-A, Neurochem. Res. 33 (2008) 1811–1820. feminised mouse, Eur. J. Endocrinol./Eur. Fed. Endocr. Soc. 148 (2003) 111– [205] Y.P. Lu, M. Zeng, D.F. Swaab, R. Ravid, J.N. Zhou, Colocalization and alteration 120. of estrogen receptor-alpha and -beta in the hippocampus in Alzheimer’s [180] A.F. Jorm, A.E. Korten, A.S. Henderson, The prevalence of dementia: a disease, Hum. Pathol. 35 (2004) 275–280. quantitative integration of the literature, Acta Psychiatr. Scand. 76 (1987) [206] P.H. Lu, D.A. Masterman, R. Mulnard, C. Cotman, B. Miller, K. Yaffe, E. Reback, 465–479. V. Porter, R. Swerdloff, J.L. Cummings, Effects of testosterone on cognition and [181] G. Kalesnykas, U. Roschier, J. Puolivali, J. Wang, R. Miettinen, The effect of mood in male patients with mild Alzheimer disease and healthy elderly men, aging on the subcellular distribution of estrogen receptor-alpha in the Arch. Neurol. 63 (2006) 177–185. cholinergic neurons of transgenic and wild-type mice, Eur. J. Neurosci. 21 [207] N.J. MacLusky, T. Hajszan, C. Leranth, Effects of dehydroepiandrosterone and (2005) 1437–1442. flutamide on hippocampal CA1 spine synapse density in male and female [182] F. Kamenetz, T. Tomita, H. Hsieh, G. Seabrook, D. Borchelt, T. Iwatsubo, S. rats: implications for the role of androgens in maintenance of hippocampal Sisodia, R. Malinow, APP processing and synaptic function, Neuron 37 (2003) structure, Endocrinology 145 (2004) 4154–4161. 925–937. [208] N.J. MacLusky, T. Hajszan, J. Prange-Kiel, C. Leranth, Androgen modulation of [183] J.M. Kaufman, A. Vermeulen, The decline of androgen levels in elderly men hippocampal synaptic plasticity, Neuroscience 138 (2006) 957–965. and its clinical and therapeutic implications, Endocr. Rev. (2005). [209] P. Maki, Hormone therapy and risk for dementia: where do we go from here?, [184] C. Kawas, S. Resnick, A. Morrison, R. Brookmeyer, M. Corrada, A. Gynecol Endocrinol. 19 (2003) 354. Zonderman, C. Bacal, D. Donnell Lingle, E. Metter, A prospective study of [210] P.M. Maki, M. Ernst, E.D. London, K.L. Mordecai, P. Perschler, S.C. Durso, J. estrogen replacement therapy and the risk of developing Alzheimer’s Brandt, A. Dobs, S.M. Resnick, Intramuscular testosterone treatment in disease: the Baltimore longitudinal study of aging, Neurology 48 (1997) elderly men: evidence of memory decline and altered brain function, J. 1517–1521. Clin. Endocrinol. Metab. 92 (2007) 4107–4114. [185] J.F. Kelly, J.L. Bienias, A. Shah, K.A. Meeke, J.A. Schneider, E. Soriano, D.A. [211] M.C. Manfredi, Y. Bi, A.A. Nirschl, J.C. Sutton, R. Seethala, R. Golla, B.C. Beehler, Bennett, Levels of estrogen receptors alpha and beta in frontal cortex of P.G. Sleph, G.J. Grover, J. Ostrowski, L.G. Hamann, Synthesis and SAR of patients with Alzheimer’s disease: relationship to Mini-Mental State tetrahydropyrrolo[1,2-b][1,2,5]thiadiazol-2(3H)-one 1,1-dioxide analogues Examination scores, Curr. Alzheimer Res. 5 (2008) 45–51. as highly potent selective androgen receptor modulators, Bioorg. Med. [186] E.J. Kilbourne, W.J. Moore, L.P. Freedman, S. Nagpal, Selective androgen Chem. Lett. 17 (2007) 4487–4490. receptor modulators for frailty and osteoporosis, Curr. Opin. Investig. Drugs 8 [212] J.J. Manly, C.A. Merchant, D.M. Jacobs, S. Small, K. Bell, M. Ferin, R. Mayeux, (2007) 821–829. Endogenous estrogen levels and Alzheimer’s disease among postmenopausal [187] H. Kim, O.Y. Bang, M.W. Jung, S.D. Ha, H.S. Hong, K. Huh, S.U. Kim, I. Mook- women, Neurology 54 (2000) 833–837. Jung, Neuroprotective effects of estrogen against beta-amyloid toxicity are [213] D. Manthey, S. Heck, S. Engert, C. Behl, Estrogen induces a rapid secretion of mediated by estrogen receptors in cultured neuronal cells, Neurosci. Lett. 302 amyloid b precursor protein via the mitogen-activated protein kinase (2001) 58–62. pathway, Eur. J. Biochem. 268 (2001) 4285–4291. C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 255

[214] R. Mark, K. Hensley, D. Butterﬁeld, M. Mattson, Amyloid beta-peptide impairs [241] T. Ohkura, K. Isse, K. Akazawa, M. Hamamoto, Y. Yaoi, N. Hagino, Evaluation of ion-motive ATPase activities: evidence for a role in loss of neuronal Ca2+ estrogen treatment in female patients with dementia of the Alzheimer type, homeostasis and cell death, J. Neurosci. 15 (1995) 6239–6249. Endocr. J. 41 (1994) 361–371. [215] M.P. Mattson, M.A. Lovell, K. Furukawa, W.R. Markesbery, Neurotrophic [242] A. Olariu, K. Yamada, T. Nabeshima, Amyloid pathology and protein kinase C factors attenuate glutamate-induced accumulation of peroxides, elevation of (PKC): possible therapeutics effects of PKC activators, J. Pharmacol. Sci. 97 intracellular Ca2+ concentration, and neurotoxicity and increase antioxidant (2005) 1–5. enzyme activities in hippocampal neurons, J. Neurochem. 65 (1995) 1740– [243] K. O’Neill, S. Chen, R.D. Brinton, Impact of the selective estrogen receptor 1751. modulator, raloxifene, on neuronal survival and outgrowth following toxic [216] M.P. Mattson, J. Partin, J.G. Begley, Amyloid beta-peptide induces apoptosis- insults associated with aging and Alzheimer’s disease, Exp. Neurol. 185 related events in synapses and dendrites, Brain Res. 807 (1998) 167– (2004) 63–80. 176. [244] R. Orlando, A. Caruso, G. Molinaro, M. Motolese, F. Matrisciano, G. Togna, D. [217] D.E. Meier, E.S. Orwoll, E.J. Keenan, R.M. Fagerstrom, Marked decline in Melchiorri, F. Nicoletti, V. Bruno, Nanomolar concentrations of anabolic- trabecular bone mineral content in healthy men with age: lack of association androgenic steroids amplify excitotoxic neuronal death in mixed mouse with sex steroid levels, J. Am. Geriatr. Soc. 35 (1987) 189–197. cortical cultures, Brain Res. 1165 (2007) 21–29. [218] A.L. Mize, L.J. Young, R.H. Alper, Uncoupling of 5-HT1A receptors in the brain [245] J. Ostrowski, J.E. Kuhns, J.A. Lupisella, M.C. Manfredi, B.C. Beehler, S.R. Krystek by estrogens: regional variations in antagonism by ICI 182,780, Jr., Y. Bi, C. Sun, R. Seethala, R. Golla, P.G. Sleph, A. Fura, Y. An, K.F. Kish, J.S. Neuropharmacology 44 (2003) 584–591. Sack, K.A. Mookhtiar, G.J. Grover, L.G. Hamann, Pharmacological and x-ray [219] S.D. Moffat, A.B. Zonderman, E.J. Metter, M.R. Blackman, S.M. Harman, S.M. structural characterization of a novel selective androgen receptor modulator: Resnick, Longitudinal assessment of serum free testosterone concentration potent hyperanabolic stimulation of skeletal muscle with hypostimulation of predicts memory performance and cognitive status in elderly men, J. Clin. prostate in rats, Endocrinology 148 (2007) 4–12. Endocrnol. Metab. 87 (2002) 5001–5007. [246] A. Paganini-Hill, L.J. Clark, Preliminary assessment of cognitive function in [220] S.D. Moffat, A.B. Zonderman, E.J. Metter, C. Kawas, M.R. Blackman, S.M. breast cancer patients treated with tamoxifen, Breast Cancer Res. Treat 64 Harman, S.M. Resnick, Free testosterone and risk for Alzheimer disease in (2000) 165–176. older men, Neurology 62 (2004) 188–193. [247] A. Paganini-Hill, V.W. Henderson, Estrogen deﬁciency and risk of Alzheimer’s [221] P.K. Molsa, R.J. Marttila, U.K. Rinne, Epidemiology of dementia in a Finnish disease in women, Am. J. Epidemiol. 140 (1994) 256–261. population, Acta Neurol. Scand. 65 (1982) 541–552. [248] A. Paganini-Hill, V.W. Henderson, Estrogen replacement therapy and risk of [222] I. Mook-Jung, I. Joo, S. Sohn, H.J. Kwon, K. Huh, M.W. Jung, Estrogen blocks Alzheimer disease, Arch. Intern. Med. 156 (1996) 2213–2217. neurotoxic effects of beta-amyloid (1-42) and induces neurite extension on [249] A.M. Paoletti, S. Congia, S. Lello, D. Tedde, M. Orru, M. Pistis, M. Pilloni, P. B103 cells, Neurosci. Lett. 235 (1997) 101–104. Zedda, A. Loddo, G.B. Melis, Low androgenization index in elderly women and [223] G. Morali, A.E. Lemus, R. Munguia, M. Arteaga, G. Perez-Palacios, K. Sundaram, elderly men with Alzheimer’s disease, Neurology 62 (2004) 301–303. N. Kumar, C.W. Bardin, Induction of male sexual behavior in the rat by 7 [250] S.C. Papasozomenos, The heat shock-induced hyperphosphorylation of tau is alpha-methyl-19-nortestosterone, an androgen that does not undergo 5 estrogen-independent and prevented by androgens: implications for alpha-reduction, Biol. Reprod. 49 (1993) 577–581. Alzheimer disease, Proc. Natl. Acad. Sci. USA 94 (1997) 6612–6617. [224] J.E. Morley, Androgens and aging, Maturitas 38 (2001) 61–71. discussion 71– [251] S.C. Papasozomenos, T. Papasozomenos, Androgens prevent the heat shock- 63. induced hyperphosphorylation but not dephosphorylation of t in female [225] J.E. Morley, H.M. Perry, Androgen treatment of male hypogonadism in older rats. implications for Alzheimer’s disease, J. Alzheimers Dis. 1 (1999) 147– males, J. Steroid Biochem. Mol. Biol. 85 (2003) 367–373. 153. [226] J.E. Morley, F.E. Kaiser, H.M. Perry, P. Patrick, P.M. Morley, P.M. Stauber, B. [252] S. Papasozomenos, A. Shanavas, Testosterone prevents the heat shock- Vellas, R.N. Baumgartner, P.J. Garry, Longitudinal changes in testosterone, induced overactivation of glycogen synthase kinase-3 beta but not of cyclin- luteinizing hormone, and follicle-stimulating hormone in healthy older men, dependent kinase 5 and c-Jun NH2-terminal kinase and concomitantly Metabol.: Clin. Exp. 46 (1997) 410–413. abolishes hyperphosphorylation of tau: implications for Alzheimer’s disease, [227] M. Muller, I. den Tonkelaar, J.H. Thijssen, D.E. Grobbee, Y.T. van der Schouw, Proc. Natl. Acad. Sci. USA 99 (2002) 1140–1145. Endogenous sex hormones in men aged 40–80 years, Eur. J. Endocrinol. 149 [253] S.Y. Park, C. Tournell, R.C. Sinjoanu, A. Ferreira, Caspase-3- and calpain- (2003) 583–589. mediated tau cleavage are differentially prevented by estrogen and [228] R.A. Mulnard, C.W. Cotman, C. Kawas, C.H. van Dyck, M. Sano, R. Doody, E. testosterone in beta-amyloid-treated hippocampal neurons, Neuroscience Koss, E. Pfeiffer, S. Jin, A. Gamst, M. Grundman, R. Thomas, L.J. Thal, Estrogen 144 (2007) 119–127. replacement therapy for treatment of mild to moderate Alzheimer disease: a [254] C. Patrone, S. Andersson, L. Korhonen, D. Lindholm, Estrogen receptor- randomized controlled trial, J. Am. Med. Assoc. 283 (2000) 1007. dependent regulation of sensory neuron survival in developing dorsal root [229] T.V.V. Nguyen, M. Yao, C.J. Pike, Androgens activate mitogen-activated ganglion, Proc. Natl. Acad. Sci. USA 96 (1999) 10905–10910. protein kinase signaling: role in neuroprotection, J. Neurochem. 94 (2005) [255] C. Pennanen, M. Kivipelto, S. Tuomainen, P. Hartikainen, T. Hanninen, M.P. 1639–1651. Laakso, M. Hallikainen, M. Vanhanen, A. Nissinen, E.L. Helkala, P. Vainio, R. [230] S.T. Nguyen, R. Prakash, C.J. Anderson, M. Frydenberg, J.M. Haynes, Sex Vanninen, K. Partanen, H. Soininen, Hippocampus and entorhinal cortex in steroids modulate alpha 1-adrenoceptor-stimulated Ca2+ elevation in human mild cognitive impairment and early AD, Neurobiol. Aging 25 (2004) 303– cultured prostatic stromal cells, Prostate 67 (2007) 74–82. 310. [231] T.V. Nguyen, M. Yao, C.J. Pike, Flutamide and cyproterone acetate exert [256] B. Perillo, A. Sasso, C. Abbondanza, G. Palumbo, 17beta-estradiol inhibits agonist effects: induction of androgen receptor-dependent neuroprotection, apoptosis in MCF-7 cells, inducing bcl-2 expression via two estrogen- Endocrinology 148 (2007) 2936–2943. responsive elements present in the coding sequence, Mol. Cell Biol. 20 [232] J. Nilsen, R.D. Brinton, Impact of progestins on estradiol potentiation of the (2000) 2890–2901. glutamate calcium response, Neuroreport 13 (2002) 825–830. [257] I. Persson, H.O. Adami, L. Bergkvist, A. Lindgren, B. Pettersson, R. Hoover, C. [233] J. Nilsen, R.D. Brinton, Divergent impact of progesterone and Schairer, Risk of endometrial cancer after treatment with oestrogens alone or medroxyprogesterone acetate (Provera) on nuclear mitogen-activated in conjunction with progestogens: results of a prospective study, Bmj 298 protein kinase signaling, Proc. Natl. Acad. Sci. USA 100 (2003) 10506–10511. (1989) 147–151. [234] J. Nilsen, R. Diaz_Brinton, Mechanism of estrogen-mediated neuroprotection: [258] S.S. Petanceska, V. Nagy, D. Frail, S. Gandy, Ovariectomy and 17beta-estradiol regulation of mitochondrial calcium and Bcl-2 expression, Proc. Natl. Acad. modulate the levels of Alzheimer’s amyloid beta peptides in brain, Exp. Sci. USA 100 (2003) 2842–2847. Gerontol. 35 (2000) 1317–1325. [235] J. Nilsen, G. Mor, F. Naftolin, Raloxifene induces neurite outgrowth in [259] C.J. Pike, Estrogen modulates neuronal Bcl-xL expression and beta-amyloid- estrogen receptor positive PC12 cells, Menopause 5 (1998) 211–216. induced apoptosis: relevance to Alzheimer’s disease, J. Neurochem. 72 (1999) [236] J. Nilsen, S. Chen, R.D. Brinton, Dual action of estrogen on glutamate-induced 1552–1563. calcium signaling: mechanisms requiring interaction between estrogen [260] C.J. Pike, Testosterone attenuates beta-amyloid toxicity in cultured receptors and src/mitogen activated protein kinase pathway, Brain Res. 930 hippocampal neurons, Brain Res. 919 (2001) 160–165. (2002) 216–234. [261] C.J. Pike, A.J. Walencewicz, C.G. Glabe, C.W. Cotman, Aggregation-related [237] V.L. Nordell, M.M. Scarborough, A.K. Buchanan, F. Sohrabji, Differential effects toxicity of synthetic beta-amyloid protein in hippocampal cultures, Eur. J. of estrogen in the injured forebrain of young adult and reproductive Pharmacol. 207 (1991) 367–368. senescent animals, Neurobiol. Aging 24 (2003) 733–743. [262] C.J. Pike, B.J. Cummings, C.W. Cotman, beta-Amyloid induces neuritic [238] S. Oddo, A. Caccamo, M. Kitazawa, B.P. Tseng, F.M. LaFerla, Amyloid dystrophy in vitro: similarities with Alzheimer pathology, Neuroreport 3 deposition precedes tangle formation in a triple transgenic model of (1992) 769–772. Alzheimer’s disease, Neurobiol. Aging 24 (2003) 1063–1070. [263] C.J. Pike, D. Burdick, A.J. Walencewicz, C.G. Glabe, C.W. Cotman, [239] Effects of Estrogen or Estrogen/Progestin Regimens on Heart Disease Risk Neurodegeneration induced by beta-amyloid peptides in vitro: the role of Factors in Postmenopausal Women. The Postmenopausal Estrogen/Progestin peptide assembly state, J. Neurosci. 13 (1993) 1676–1687. Interventions (PEPI) Trial. The Writing Group for the PEPI Trial, Jama 273 [264] C.J. Pike, B.J. Cummings, R. Monzavi, C.W. Cotman, Beta-amyloid-induced (1995) 199–208. changes in cultured astrocytes parallel reactive astrocytosis associated [240] M. Ogiue-Ikeda, N. Tanabe, H. Mukai, Y. Hojo, G. Murakami, T. Tsurugizawa, with senile plaques in Alzheimer’s disease, Neuroscience 63 (1994) 517– N. Takata, T. Kimoto, S. Kawato, Rapid modulation of synaptic plasticity by 531. estrogens as well as endocrine disrupters in hippocampal neurons, Brain Res. [265] J.V. Pinkerton, V.W. Henderson, Estrogen and cognition, with a focus on Rev. 57 (2008) 363–375. Alzheimer’s disease, Semin. Reprod. Med. 23 (2005) 172–179. 256 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258

[266] P. Polo-Kantola, R. Portin, O. Polo, H. Helenius, K. Irjala, R. Erkkola, The effect [297] D.J. Selkoe, T. Yamazaki, M. Citron, M.B. Podlisny, E.H. Koo, D.B. Teplow, C. of short-term estrogen replacement therapy on cognition: a randomized, Haass, The role of APP processing and trafficking pathways in the formation double-blind, cross-over trial in postmenopausal women, Obstet. Gynecol. 91 of amyloid beta-protein, Ann. NY Acad. Sci. 777 (1996) 57–64. (1998) 459–466. [298] A. Selvamani, F. Sohrabji, Reproductive age modulates the impact of focal [267] D.V. Pow, V.H. Perry, J.F. Morris, S. Gordon, Microglia in the neurohypophysis ischemia on the forebrain as well as the effects of estrogen treatment in associate with and endocytose terminal portions of neurosecretory neurons, female rats, Neurobiol. Aging (2008). Neuroscience 33 (1989) 567–578. [299] G. Sétáló, M. Singh, I.S. Nethrapalli, C.D. Toran-Allerand, Protein kinase C [268] J. Qiu, M.A. Bosch, S.C. Tobias, D.K. Grandy, T.S. Scanlan, O.K. Ronnekleiv, M.J. activity is necessary for estrogen-induced Erk phosphorylation in neocortical Kelly, Rapid signaling of estrogen in hypothalamic neurons involves a novel explants, J. Neurochem. Res. 30 (2005) 779–790. G-protein-coupled estrogen receptor that activates protein kinase C, J. [300] Y. Shang, M. Brown, Molecular determinants for the tissue specificity of Neurosci. 23 (2003) 9529–9540. SERMs, Science 295 (2002) 2465–2468. [269] J. Raber, Androgens, apoE, and Alzheimer’s disease, Sci. Aging Knowledge [301] T.C. Shao, H.L. Li, S. Kasper, R. Matusik, M. Ittmann, G.R. Cunningham, Environ. 2004 (2004) re2. Comparison of the growth-promoting effects of testosterone and 7-alpha- [270] J. Raber, AR, apoE, and cognitive function, Horm. Behav. 53 (2008) 706–715. methyl-19-nor-testosterone (MENT) on the prostate and levator ani muscle [271] M. Ramsden, A.C. Nyborg, M.P. Murphy, L. Chang, F.Z. Stanczyk, T.E. Golde, C.J. of LPB-tag transgenic mice, Prostate 66 (2006) 369–376. Pike, Androgens modulate beta-amyloid levels in male rat brain, J. [302] M. Sheffield-Moore, R.J. Urban, An overview of the endocrinology of skeletal Neurochem. 87 (2003) 1052–1055. muscle, Trends Endocrinol. Metab. 15 (2004) 110–115. [272] M. Ramsden, T.M. Shin, C.J. Pike, Androgens modulate neuronal vulnerability [303] L.C. Sheldahl, L.K. Marriott, D.M. Bryant, R.A. Shapiro, D.M. Dorsa, to kainate lesion, Neuroscience 122 (2003) 573–578. Neuroprotective effects of estrogen and selective estrogen receptor [273] S.R. Rapp, M.A. Espeland, S.A. Shumaker, V.W. Henderson, R.L. Brunner, J.E. modulators begin at the plasma membrane, Minerva Endocrinol. 32 (2007) Manson, M.L. Gass, M.L. Stefanick, D.S. Lane, J. Hays, K.C. Johnson, L.H. Coker, 87–94. M. Dailey, D. Bowen, Effect of estrogen plus progestin on global cognitive [304] R. Shen, M. Sumitomo, J. Dai, D.O. Hardy, D. Navarro, B. Usmani, C.N. function in postmenopausal women: the Women’s Health Initiative Memory Papandreou, L.B. Hersh, M.A. Shipp, L.P. Freedman, D.M. Nanus, Identification Study: a randomized controlled trial, Jama: J. Am. Med. Assoc. 289 (2003) and characterization of two androgen response regions in the human neutral 2663–2672. endopeptidase gene, Mol. Cell Endocrinol. 170 (2000) 131–142. [274] S. Rasmuson, B. Nasman, K. Carlstrom, T. Olsson, Increased levels of [305] B.B. Sherwin, Estrogen and/or androgen replacement therapy and cognitive adrenocortical and gonadal hormones in mild to moderate Alzheimer’s functioning in surgically menopausal women, Psychoneuroendocrinology 13 disease, Dement. Geriatr. Cogn. Disord. 13 (2002) 74–79. (1988) 345–357. [275] R.F. Regan, Y. Guo, Estrogens attenuate neuronal injury due to hemoglobin, [306] B.B. Sherwin, Estrogen and cognitive functioning in women, Endocr. Rev. 24 chemical hypoxia, and excitatory amino acids in murine cortical cultures, (2003) 133–151. Brain Res. 764 (1997) 133–140. [307] J. Shi, K.S. Panickar, S.-H. Yang, O. Rabbani, A.L. Day, J.W. Simpkins, Estrogen [276] S.M. Resnick, V.W. Henderson, Hormone therapy and risk of Alzheimer attenuates over-expression of b-amyloid precursor protein messager RNA in disease: a critical time, Jama 288 (2002) 2170–2172. an animal model of focal ischemia, Brain Res. 810 (1998) 87–92. [277] M.E. Rhodes, C.A. Frye, Androgens in the hippocampus can alter, and be [308] V. Shilling, V. Jenkins, L. Fallowfield, T. Howell, The effects of hormone altered by, ictal activity, Pharmacol. Biochem. Behav. 78 (2004) 483–493. therapy on cognition in breast cancer, J. Steroid Biochem. Mol. Biol. 86 (2003) [278] C.A. Rice-Evans, N.J. Miller, G. Paganga, Structure-antioxidant activity 405–412. relationships of flavonoids and phenolic acids, Free Rad. Biol. Med. 20 [309] P.J. Shughrue, I. Merchenthaler, Estrogen is more than just a ‘‘Sex Hormone”: (1996) 933–956. novel sites for estrogen action in the hippocampus and cerebral cortex, Front. [279] C. Rice-Evans, N. Miller, G. Paganga, Antioxidant properties of phenolic Neuroendocrinol. 21 (2000) 95–101. compounds, Trends Plant Sci. 2 (1997) 152–159. [310] P.J. Shughrue, M.V. Lane, I. Merchenthaler, Comparative distribution of [280] P. Riederer, S. Hoyer, From benefit to damage. Glutamate and advanced estrogen receptor-alpha and -beta mRNA in the rat central nervous system, J. glycation end products in Alzheimer brain, J. Neural. Transm. 113 (2006) Comp. Neurol. 388 (1997) 507–525. 1671–1677. [311] S.A. Shumaker, C. Legault, S.R. Rapp, L. Thal, R.B. Wallace, J.K. Ockene, S.L. [281] W.A. Rocca, L.A. Amaducci, B.S. Schoenberg, Epidemiology of clinically Hendrix, B.N. Jones, A.R. Assaf, R.D. Jackson, J.M. Kotchen, S. diagnosed Alzheimer’s disease, Ann. Neurol. 19 (1986) 415–424. Wassertheil_Smoller, J. Wactawski_Wende, Estrogen plus progestin and the [282] J. Rogers, R. Strohmeyer, C. Kovelowski, R.G. Li, Microglia and inflammatory incidence of dementia and mild cognitive impairment in postmenopausal mechanisms in the clearance of amyloid beta peptide, Glia 40 (2002) 260– women: the Women’s Health Initiative Memory Study: a randomized 269. controlled trial, Jama: J. Am. Med. Assoc. 289 (2003) 2651–2662. [283] E.R. Rosario, C.J. Pike, Androgen regulation of beta-amyloid protein and the [312] S.A. Shumaker, C. Legault, L. Kuller, S.R. Rapp, L. Thal, D.S. Lane, H. Fillit, M.L. risk of Alzheimer’s disease, Brain Res. Rev. 57 (2008) 444–453. Stefanick, S.L. Hendrix, C.E. Lewis, K. Masaki, L.H. Coker, Conjugated equine [284] E.R. Rosario, L. Chang, F.Z. Stanczyk, C.J. Pike, Age-related testosterone estrogens and incidence of probable dementia and mild cognitive depletion and the development of Alzheimer disease, Jama 292 (2004) 1431– impairment in postmenopausal women: Women’s Health Initiative 1432. Memory Study, Jama 291 (2004) 2947–2958. [285] E.R. Rosario, J.C. Carroll, S. Oddo, F.M. LaFerla, C.J. Pike, Androgens regulate [313] I. Silva, L.E. Mello, E. Freymuller, M.A. Haidar, E.C. Baracat, Estrogen, the development of neuropathology in a triple transgenic mouse model of progestogen and tamoxifen increase synaptic density of the hippocampus Alzheimer’s disease, J. Neurosci. 26 (2006) 13384–13389. of ovariectomized rats, Neurosci. Lett. 291 (2000) 183–186. [286] E.R. Rosario, M. Ramsden, C.J. Pike, Progestins inhibit the neuroprotective [314] R.B. Simerly, C. Chang, M. Muramatsu, L.W. Swanson, Distribution of effects of estrogen in rat hippocampus, Brain Res. 1099 (2006) 206–210. androgen and estrogen receptor mRNA-containing cells in the rat brain: an [287] E.R. Rosario, L. Chang, E.H. Head, F. Z. Stanczyk, C. J. Pike, Brain levels of sex in situ hybridization study, J. Comp. Neurol. 294 (1990) 76–95. steroid hormones in men and women during normal aging and in [315] C.A. Singer, K.L. Rogers, T.M. Strickland, D.M. Dorsa, Estrogen protects Alzheimer’s disease, Neurobiol. Aging, in press. primary cortical neurons from glutamate toxicity, Neurosci. Lett. 212 [288] M.J. Rowan, I. Klyubin, Q. Wang, N.W. Hu, R. Anwyl, Synaptic memory (1996) 13–16. mechanisms: Alzheimer’s disease amyloid beta-peptide-induced [316] C.A. Singer, K.L. Rogers, D.M. Dorsa, Modulation of Bcl-2 expression: a dysfunction, Biochem. Soc. Trans. 35 (2007) 1219–1223. potential component of estrogen protection in NT2 neurons, Neuroreport 9 [289] B.S. Rubin, T.O. Fox, R.S. Bridges, Estrogen binding in nuclear and cytosolic (1998) 2565–2568. extracts from brain and pituitary of middle-aged female rats, Brain Res. 383 [317] C.A. Singer, X.A. Figueroa_Masot, R.H. Batchelor, D.M. Dorsa, The mitogen- (1986) 60–67. activated protein kinase pathway mediates estrogen neuroprotection after [290] A. Ruitenberg, A. Ott, J.C. van Swieten, A. Hofman, M.M.B. Breteler, Incidence glutamate toxicity in primary cortical neurons, J. Neurosci.: Official J. Soc. of dementia: does gender make a difference?, Neurobiol Aging 22 (2001) Neurosci. 19 (1999) 2455–2463. 575–580. [318] M. Singh, Ovarian hormones elicit phosphorylation of Akt and extracellular- [291] M. Sastre, J. Walter, S.M. Gentleman, Interactions between APP secretases and signal regulated kinase in explants of the cerebral cortex, Endocrine 14 inflammatory mediators, J. Neuroinflammation 5 (2008) 25. (2001) 407–415. [292] E. Savaskan, G. Olivieri, F. Meier, R. Ravid, F. Muller-Spahn, Hippocampal [319] M. Singh, G. Setalo Jr., X. Guan, M. Warren, C.D. Toran-Allerand, Estrogen- estrogen beta-receptor immunoreactivity is increased in Alzheimer’s disease, induced activation of mitogen-activated protein kinase in cerebral cortical Brain Res. 908 (2001) 113–119. explants: convergence of estrogen and neurotrophin signaling pathways, J. [293] C. Schultz, H. Braak, E. Braak, A sex difference in neurodegeneration of the Neurosci. 19 (1999) 1179–1188. human hypothalamus, Neurosci. Lett. 212 (1996) 103–106. [320] S.E. Stoltzner, N.C. Berchtold, C.W. Cotman, C.J. Pike, Estrogen regulates Bcl-x [294] C. Schultz, E. Ghebremedhin, E. Braak, H. Braak, Sex-dependent cytoskeletal expression in rat hippocampus, Neuroreport 12 (2001) 2797–2800. changes of the human hypothalamus develop independently of Alzheimer’s [321] D.J. Stone, I. Rozovsky, T.E. Morgan, C.P. Anderson, L.M. Lopez, J. Shick, C.E. disease, Exp. Neurol. 160 (1999) 186–193. Finch, Effects of age on gene expression during estrogen-induced synaptic [295] M. Schumacher, R. Sitruk-Ware, A.F. De Nicola, Progesterone and progestins: sprouting in the female rat, Exp. Neurol. 165 (2000) 46–57. neuroprotection and myelin repair, Curr. Opin. Pharmacol. 8 (2008) 740–746. [322] W.J. Strittmatter, A.D. Roses, Apolipoprotein E and Alzheimer disease, Proc. [296] N. Schupf, J.H. Lee, M. Wei, D. Pang, C. Chace, R. Cheng, W.B. Zigman, B. Tycko, Natl. Acad. Sci. USA 92 (1995) 4725–4727. W. Silverman, Estrogen receptor-alpha variants increase risk of Alzheimer’s [323] R.G. Struble, B.P. Nathan, C. Cady, X. Cheng, M. McAsey, Estradiol regulation of disease in women with Down syndrome, Dement. Geriatr. Cogn. Disord. 25 astroglia and apolipoprotein E: an important role in neuronal regeneration, (2008) 476–482. Exp. Gerontol. 42 (2007) 54–63. C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258 257

[324] M.T. Subbiah, B. Kessel, M. Agrawal, R. Rajan, W. Abplanalp, Z. Rymaszewski, [353] J.J. Watters, J.S. Campbell, M.J. Cunningham, E.G. Krebs, D.M. Dorsa, Rapid Antioxidant potential of specific estrogens on lipid peroxidation, J. Clin. membrane effects of steroids in neuroblastoma cells: effects of estrogen on Endocrinol. Metab. 77 (1993) 1095–1097. mitogen activated protein kinase signalling cascade and c-fos immediate [325] K. Sugioka, Y. Shimosegawa, M. Nakano, Estrogens as natural antioxidants of early gene transcription, Endocrinology 138 (1997) 4030–4033. membrane phospholipid peroxidation, FEBS Lett. 210 (1987) 37–39. [354] C.E. Weaver Jr., M. Park-Chung, T.T. Gibbs, D.H. Farb, 17beta-Estradiol [326] R.S. Swerdloff, C. Wang, Androgen deficiency and aging in men, Western J. protects against NMDA-induced excitotoxicity by direct inhibition of Med. 159 (1993) 579–585. NMDA receptors, Brain Res. 761 (1997) 338–341. [327] R.S. Tan, J.W. Culberson, An integrative review on current evidence of [355] C.L. Weaver, M. Espinoza, Y. Kress, P. Davies, Conformational change as one of testosterone replacement therapy for the andropause, Maturitas 45 (2003) the earliest alterations of tau in Alzheimer’s disease, Neurobiol. Aging 21 15–27. (2000) 719–727. [328] R.S. Tan, S.J. Pu, A pilot study on the effects of testosterone in hypogonadal [356] J.H. Weiss, C.J. Pike, C.W. Cotman, Ca2+ channel blockers attenuate beta- aging male patients with Alzheimer’s disease, Aging Male 6 (2003) 13. amyloid peptide toxicity to cortical neurons in culture, J. Neurochem. 62 [329] M.X. Tang, D. Jacobs, Y. Stern, K. Marder, P.R. Schofield, B. Gurland, H. (1994) 372–375. Andrews, R. Mayeux, Effect of oestrogen during menopause on risk and age at [357] W. Wharton, E. Brinton, M. Cedars, R.A. Lobo, J. Manson, G.R. Merriam, V.M. onset of Alzheimer’s disease, The Lancet 348 (1996) 429. Miller, N. Santoro, H.S. Taylor, S.M. Harman, S. Asthana, Baseline cognitive [330] L. Tanzer, K.J. Jones, Gonadal steroid regulation of hamster facial nerve and demographic characteristics of women enrolled in the Kronos Early regeneration: effects of dihydrotestosterone and estradiol, Exp. Neurol. 146 Estrogen Prevention Study (KEEPS) Alzheimer’s & Dementia, J. Alzheimer’s (1997) 258–264. Assoc. 4 (2008) T792. [331] R.E. Tanzi, R.D. Moir, S.L. Wagner, Clearance of Alzheimer’s A[beta] Peptide: [358] K.C. Wilhelmsen, L.N. Clark, B.L. Miller, D.H. Geschwind, Tau mutations in the many roads to perdition, Neuron 43 (2004) 605–608. frontotemporal dementia, Dement. Geriatr. Cogn. Disord. 10 (Suppl. 1) (1999) [332] S.H. Tariq, M.T. Haren, M.J. Kim, J.E. Morley, Andropause: is the emperor 88–92. wearing any clothes?, Rev Endocr. Metab. Disord. 6 (2005) 77–84. [359] E.M. Wilson, Muscle-bound? A tissue-selective nonsteroidal androgen [333] M.K. Thakur, S.T. Mani, Estradiol regulates APP mRNA alternative splicing in receptor modulator, Endocrinology 148 (2007) 1–3. the mice brain cortex, Neurosci. Lett. 381 (2005) 154–157. [360] P.M. Wise, Estrogen therapy: does it help or hurt the adult and aging brain? [334] T. Thomas, M. Bryant, L. Clark, A. Garces, J. Rhodin, Estrogen and raloxifene Insights derived from animal models, Neuroscience 138 (2006) 831–835. activities on amyloid – induced inflammatory reaction, Microvasc. Res. 61 [361] P.M. Wise, B. Parsons, Nuclear estradiol and cytosol progestin receptor (2001) 28–39. concentrations in the brain and the pituitary gland and sexual behavior in [335] C.D. Toran-Allerand, M. Singh, G. Sétáló, Novel mechanisms of estrogen action ovariectomized estradiol-treated middle-aged rats, Endocrinology 115 in the brain: new players in an old story, Front. Neuroendocrinol. 20 (1999) (1984) 810–816. 97–121. [362] C.S. Woolley, Bruce S. McEwen, Roles of estradiol and progesterone in [336] P.R. Turner, K. O’Connor, W.P. Tate, W.C. Abraham, Roles of amyloid precursor regulation of hippocampal dendritic spine density during the estrous cycle in protein and its fragments in regulating neural activity, plasticity and the rat, J. Comp. Neurol. 336 (1993) 293–306. memory, Prog. Neurobiol. 70 (2003) 1–32. [363] X. Wu, M.A. Glinn, N.L. Ostrowski, Y. Su, B. Ni, H.W. Cole, H.U. Bryant, S.M. [337] E.R.L.C. Vardy, A.J. Catto, N.M. Hooper, Proteolytic mechanisms in amyloid- Paul, Raloxifene and estradiol benzoate both fully restore hippocampal [beta] metabolism: therapeutic implications for Alzheimer’s disease, Trends choline acetyltransferase activity in ovariectomized rats, Brain Res. 847 Mol. Med. 11 (2005) 464–472. (1999) 98–104. [338] S. Vasto, G. Candore, F. Listi, C.R. Balistreri, G. Colonna-Romano, M. Malavolta, [364] T.W. Wu, J.M. Wang, S. Chen, R.D. Brinton, 17Beta-estradiol induced D. Lio, D. Nuzzo, E. Mocchegiani, D. Di Bona, C. Caruso, Inflammation, genes Ca2+ influx via L-type calcium channels activates the Src/ERK/cyclic- and zinc in Alzheimer’s disease, Brain Res. Rev. 58 (2008) 96–105. AMP response element binding protein signal pathway and BCL-2 [339] C. Vaughan, F.C. Goldstein, J.L. Tenover, Exogenous testosterone alone or with expression in rat hippocampal neurons: a potential initiation finasteride does not improve measurements of cognition in healthy older mechanism for estrogen-induced neuroprotection, Neuroscience 135 men with low serum testosterone, J. Androl. 28 (2007) 875–882. (2005) 59–72. [340] H. Vedder, N. Anthes, G. Stumm, C. Wurz, C. Behl, J.-C. Krieg, Estrogen [365] Z.-M. Xiao, L. Sun, Y.-M. Liu, J.-J. Zhang, J. Huang, Estrogen regulation of the hormones reduce lipid peroxidation in cells and tissues of the central nervous neprilysin gene through a hormone-responsive element, J. Mol. Neurosci., in system, J. Neurochem. 72 (1999) 2531–2538. press. [341] E. Vegeto, C. Bonincontro, G. Pollio, A. Sala, S. Viappiani, F. Nardi, A. Brusadelli, [366] H. Xu, G.K. Gouras, J.P. Greenfield, B. Vincent, J. Naslund, L. Mazzarelli, G. B. Viviani, P. Ciana, A. Maggi, Estrogen prevents the lipopolysaccharide- Fried, J.N. Jovanovic, M. Seeger, N.R. Relkin, F. Liao, F. Checler, J.D. Buxbaum, induced inflammatory response in microglia, J. Neurosci. 21 (2001) 1809– B.T. Chait, G. Thinakaran, S.S. Sisodia, R. Wang, P. Greengard, S. Gandy, 1818. Estrogen reduces neuronal generation of Alzheimer beta-amyloid peptides, [342] K. Venken, S. Boonen, E. Van Herck, L. Vandenput, N. Kumar, R. Sitruk-Ware, Nat. Med. 4 (1998) 447–451. K. Sundaram, R. Bouillon, D. Vanderschueren, Bone and muscle protective [367] Y. Xu, S.J. Armstrong, I.A. Arenas, D.J. Pehowich, S.T. Davidge, potential of the prostate-sparing synthetic androgen 7alpha-methyl-19- Cardioprotection by chronic estrogen or superoxide dismutase mimetic nortestosterone: evidence from the aged orchidectomized male rat model, treatment in the aged female rat, Am. J. Physiol. Heart Circ. Physiol. 287 Bone 36 (2005) 663–670. (2004) H165–H171. [343] G. Verdile, S. Fuller, C.S. Atwood, S.M. Laws, S.E. Gandy, R.N. Martins, The role [368] H. Xu, R. Wang, Y.-W. Zhang, X. Zhang, Estrogen, -amyloid metabolism/ of beta amyloid in Alzheimer’s disease: still a cause of everything or the only trafficking, and Alzheimer’s disease, Ann. NY Acad. Sci. 1089 (2006) 324–342. one who got caught?, Pharmacol Res. 50 (2004) 397–409. [369] K. Yaffe, K. Krueger, S. Sarkar, D. Grady, E. Barrett-Connor, D.A. Cox, T. [344] B. Vincent, J.D. Smith, Effect of estradiol on neuronal Swedish-mutated beta- Nickelsen, The multiple outcomes of raloxifene evaluation investigators, amyloid precursor protein metabolism: reversal by astrocytic cells, Biochem. cognitive function in postmenopausal women treated with raloxifene, New Biophys/Res. Commun. 271 (2000) 82–85. Engl. J. Med. 344 (2001) 1207–1213. [345] S. von Eckardstein, G. Noe, V. Brache, E. Nieschlag, H. Croxatto, F. Alvarez, A. [370] K. Yaffe, K. Krueger, S.R. Cummings, T. Blackwell, V.W. Henderson, S. Sarkar, Moo-Young, I. Sivin, N. Kumar, M. Small, K. Sundaram, A clinical trial of 7 K. Ensrud, D. Grady, Effect of raloxifene on prevention of dementia and alpha-methyl-19-nortestosterone implants for possible use as a long-acting cognitive impairment in older women: the Multiple Outcomes of Raloxifene contraceptive for men, J. Clin Endocrinol. Metab. 88 (2003) 5232–5239. Evaluation (MORE) randomized trial, Am. J. Psychiatry 162 (2005) 683– [346] E.J. Wahjoepramono, L.K. Wijaya, K. Taddei, G. Martins, M. Howard, K.d. 690. Ruyck, K. Bates, S.S. Dhaliwald, G. Verdile, M. Carruthers, R.N. Martins, [371] S.H. Yang, E. Perez, J. Cutright, R. Liu, Z. He, A.L. Day, J.W. Simpkins, Distinct effects of testosterone on plasma and cerebrospinal fluid amyloid-b Testosterone increases neurotoxicity of glutamate in vitro and ischemia- levels, J. Alzheimer’s Dis. 129 (2008) 129–137. reperfusion injury in an animal model, J. Appl. Physiol. 92 (2002) 195–201. [347] D. Walsh, D. Selkoe, A beta oligomers – a decade of discovery, J. Neurochem. [372] S.H. Yang, R. Liu, Y. Wen, E. Perez, J. Cutright, A.M. Brun-Zinkernagel, M. 101 (2007) 1172–1184. Singh, A.L. Day, J.W. Simpkins, Neuroendocrine mechanism for tolerance to [348] D.M. Walsh, I. Klyubin, J.V. Fadeeva, M.J. Rowan, D.J. Selkoe, Amyloid-beta cerebral ischemia-reperfusion injury in male rats, J. Neurobiol. 62 (2005) oligomers: their production, toxicity and therapeutic inhibition, Biochem. 341–351. Soc. Trans. 30 (2002) 552–557. [373] B.A. Yankner, T. Lu, Amyloid {beta}-Protein Toxicity and the Pathogenesis of [349] P.N. Wang, S.Q. Liao, R.S. Liu, C.Y. Liu, H.T. Chao, S.R. Lu, H.Y. Yu, S.J. Wang, H.C. Alzheimer Disease, J. Biol. Chem. 284 (2009) 4755–4759. Liu, Effects of estrogen on cognition, mood, and cerebral blood flow in AD: a [374] M. Yao, T.V. Nguyen, C.J. Pike, Beta-amyloid-induced neuronal apoptosis controlled study, Neurology 54 (2000) 2061–2066. involves c-Jun N-terminal kinase-dependent downregulation of Bcl-w, J. [350] J. Wang, H. Tanila, J. Puolivali, I. Kadish, T. van Groen, Gender differences in Neurosci. 25 (2005) 1149–1158. the amount and deposition of amyloidbeta in APPswe and PS1 double [375] M. Yao, T.V. Nguyen, C.J. Pike, Estrogen regulates Bcl-w and Bim expression: transgenic mice, Neurobiol. Dis. 14 (2003) 318–327. role in protection against beta-amyloid peptide-induced neuronal death, J. [351] S.C. Waring, W.A. Rocca, R.C. Petersen, P.C. O_Brien, E.G. Tangalos, E. Kokmen, Neurosci. 27 (2007) 1422–1433. Postmenopausal estrogen replacement therapy and risk of AD: a population- [376] M. Yao, T.V. Nguyen, E.R. Rosario, M. Ramsden, C.J. Pike, Androgens regulate based study, Neurology 52 (1999) 965–970. neprilysin expression: role in reducing beta-amyloid levels, J. Neurochem. [352] T. Watanabe, S. Koba, M. Kawamura, M. Itokawa, T. Idei, Y. Nakagawa, T. Iguchi, (2008). T. Katagiri, Small dense low-density lipoprotein and carotid atherosclerosis in [377] W.H. Yu, Effect of testosterone on the regeneration of the hypoglossal nerve relation to vascular dementia, Metabolism 53 (2004) 476–482. in rats, Exp. Neurol. 77 (1982) 129–141. 258 C.J. Pike et al. / Frontiers in Neuroendocrinology 30 (2009) 239–258

[378] W.H. Yu, Administration of testosterone attenuates neuronal loss following amyloid beta(1-42) toxicity through heat shock protein 70, J. Neurosci. 24 axotomy in the brain-stem motor nuclei of female rats, J. Neurosci. 9 (1989) (2004) 5315–5321. 3908–3914. [385] S. Zhang, Y. Huang, Y.C. Zhu, T. Yao, Estrogen stimulates release of secreted [379] W.H. Yu, R. Srinivasan, Effect of testosterone and 5 alpha- amyloid precursor protein from primary rat cortical neurons via protein dihydrotestosterone on regeneration of the hypoglossal nerve in rats, Exp. kinase C pathway, Acta Pharmacol. Sin. 26 (2005) 171–176. Neurol. 71 (1981) 431–435. [386] Q.G. Zhang, R. Wang, M. Khan, V. Mahesh, D.W. Brann, Role of Dickkopf-1, an [380] X. Yue, M. Lu, T. Lancaster, P. Cao, S.-I. Honda, M. Staufenbiel, N. Harada, Z. antagonist of the Wnt/beta-catenin signaling pathway, in estrogen-induced Zhong, Y. Shen, R. Li, Brain estrogen deﬁciency accelerates Ab plaque neuroprotection and attenuation of tau phosphorylation, J. Neurosci. 28 formation in an Alzheimer’s disease animal model, Proc. Natl. Acad. Sci. (2008) 8430–8441. 102 (2005) 19198–19203. [387] L. Zhao, R.D. Brinton, Estrogen receptor alpha and beta differentially regulate [381] T.Y. Yune, H.G. Park, J.Y. Lee, T.H. Oh, Estrogen-induced Bcl-2 expression after intracellular Ca(2+) dynamics leading to ERK phosphorylation and estrogen spinal cord injury is mediated through phosphoinositide-3-kinase/Akt- neuroprotection in hippocampal neurons, Brain Res. 1172 (2007) 48–59. dependent CREB activation, J. Neurotrauma. 25 (2008) 1121–1131. [388] L. Zhao, T.W. Wu, R.D. Brinton, Estrogen receptor subtypes alpha and beta [382] P.P. Zandi, M.C. Carlson, B.L. Plassman, K.A. Welsh-Bohmer, L.S. Mayer, D.C. contribute to neuroprotection and increased Bcl-2 expression in primary Steffens, J.C.S. Breitner, Hormone replacement therapy and incidence of hippocampal neurons, Brain Res. 1010 (2004) 22–34. Alzheimer disease in older women: the Cache County Study, J. Am. Med. [389] L. Zhao, C. Jin, Z. Mao, M.B. Gopinathan, K. Rehder, R.D. Brinton, Design, Assoc. 288 (2002) 2123–2129. synthesis, and estrogenic activity of a novel estrogen receptor modulator – a [383] L. Zhang, D.R.C.A. Rubinow, G.-q. Xaing, B.-S. Li, Y.H. Chang, D. Maric, J.L. hybrid structure of 17beta-estradiol and vitamin E in hippocampal neurons, J. Barker, W. Ma, Estrogen protects against b-amyloid-induced neurotoxicity in Med. Chem. 50 (2007) 4471–4481. rat hippocampal neurons by activation of Akt, Neuroreport 12 (2001) 1919– [390] H. Zheng, H. Xu, S.N. Uljon, R. Gross, K. Hardy, J. Gaynor, J. Lafrancois, J. 1923. Simpkins, L.M. Refolo, S. Petanceska, R. Wang, K. Duff, Modulation of A(beta) [384] Y. Zhang, N. Champagne, L.K. Beitel, C.G. Goodyer, M. Triﬁro, A. LeBlanc, peptides by estrogen in mouse models, J. Neurochem. 80 (2002) 191– Estrogen and androgen protection of human neurons against intracellular 196. Atherosclerosis 130 (1997) 199–202

The effect of supraphysiologic doses of testosterone on fasting total homocysteine levels in normal men

Joseph M. Zmuda a, Linda L. Bausserman b, Denise Maceroni b, Paul D. Thompson a,*

a Pre6enti6e Cardiology, Suite 1212, Kaufmann Building, Uni6ersity of Pittsburgh Medical Center, 200 Lothrop St., Pittsburgh, PA 15213, USA b Miriam Hospital and Brown Uni6ersity, Pro6idence, RI, USA

Received 30 September 1996; received in revised form 15 November 1996; accepted 3 December 1996

Abstract

Elevated total homocysteine (tHcy) levels are associated with increased risk for atherosclerotic cardiovascular disease. tHcy levels are higher in men than in women, and estrogen replacement therapy may reduce tHcy levels in postmenopausal women. The effect of androgenic hormones on tHcy levels in men has not been examined. The present study determined the effect of supraphysiologic doses of testosterone, with or without its aromatization to estradiol, on fasting tHcy levels in 14 normal male weightlifters aged 19–42 years. Subjects received testosterone enanthate (200 mg/week intramuscularly), the aromatase inhibitor, testolactone (1 g/day orally), or both drugs together in a crossover design. Each treatment lasted 3 weeks and each treatment was separated by a 4-week washout. Both testosterone regimens increased serum testosterone levels, whereas estradiol increased only during testosterone alone. Mean tHcy levels were not significantly altered when testosterone was given alone or together with testolactone. Testolactone did not significantly influence tHcy levels. We conclude that short-term, high-dose testosterone administration does not affect fasting tHcy levels in normal men. © 1997 Elsevier Science Ireland Ltd.

Keywords: Estrogen; Homocysteine; Men; Testosterone

1. Introduction documented an inverse correlation between serum estradiol concentrations and postmethionine tHcy levels Homocysteine is a sulphur-containing amino-acid in premenopausal women, [4] but to our knowledge the formed by the demethylation of dietary methionine. effect of androgenic hormones on tHcy concentrations Homocysteine may promote atherosclerosis by injuring in men has not been examined. the vascular endothelium, [1] and elevated total homo- The androgenic hormone testosterone is normally cysteine (tHcy) levels are associated with increased risk aromatized to estradiol in liver, muscle, and adipose for atherosclerotic cardiovascular disease [2]. The ob- tissue, and peripheral aromatization of testosterone is servation that fasting tHcy concentrations are higher in the major source of circulating estrogen in men [6]. We men than in women has led to the suggestion that sex have previously examined the effect of testosterone steroid hormones may inﬂuence tHcy levels [3]. Evi- aromatization on serum lipids [7] in men by administer- dence to support this idea comes from studies showing ing testosterone alone or in combination with the aro- that estrogen plus progestin replacement therapy [4] matase inhibitor testolactone. Because of the and the estrogen agonist tamoxifen [5] reduce tHcy observations that tHcy concentrations are higher in concentrations in postmenopausal women. Others have men than in women [3] and that female sex hormones may affect tHcy levels [4,5], we used stored plasma samples from this prior study to determine the effects * Corresponding author. Tel.: +1 412 6486859; fax: +1 412 of testosterone and its aromatization to estradiol on 6486358. fasting tHcy levels in men.

2. Materials and methods 2.3. Biochemical assays 2.1. Study subjects Fasting tHcy levels were measured in plasma with the fluorimetric method of Vester and Rasmussen [10], Fourteen healthy non-smoking men between 19 and except that 20% methanol was used in buffer B in the 42 years of age (Mean S.D.; 27.1 7.4 years) pro- 9 9 high-performance liquid chromatography (HPLC) provided written informed consent and completed the cedure. lnterassay variation for tHcy determinations study. Their mean body weight was 86.6 20.6 kg 9 was avoided by analyzing all samples for an individual before the study, and body fat estimated from the sum subject in a single assay or autoanalyzer run. Intra-as- of three skinfold measurements [8] was 12.7 6.9% 9 say coefficient of variation for tHcy was less than 6.0%. (Table 1). All men had been weightlifting for approxi- Serum testosterone and estradiol were assayed in dupli- mately 6 years and exercised at least three times per cate using radioimmunoassay kits (Diagnostic Prod- week. None of the subjects had a history of renal, ucts, Los Angeles, CA). hepatic, or vascular disease, and no subject averaged more than one alcoholic beverage daily or took regular 2.4. Statistical analysis medications. All men denied current and prior androgen use. Baseline urinalysis confirmed that subjects had Data were analyzed with a treatment-by-time analy- not recently used either anabolic–androgenic steroids sis of variance with repeated measures on both factors. or testosterone [7]. The subjects were instructed to When interaction effects were significant, the effect of maintain their habitual level of physical activity and to time for each treatment and the differences between avoid altering their dietary habits and nutritional sup- treatments at each measurement point were tested plement use during the study. Subjects were reimbursed statistically. A modified Bonferroni procedure was em- for participation as approved by The Miriam Hospital ployed to adjust for multiple comparisons [11]. Pearson Clinical Research Review Board. simple and partial correlation coefficients were com- 2.2. Study design puted to examine the associations between baseline serum testosterone and estradiol levels and fasting tHcy Subjects were randomly assigned to a counterbal- concentrations with and without adjustment for base- anced cross-over design involving three treatments: line age and body weight. testosterone enanthate (E.R. Squibb and Sons, Prince- ton, NJ), 200 mg/week intramuscularly (i.m.); oral testolactone (E.R. Squibb and Sons), 250 mg four times 3. Results daily (QID); and both testosterone enanthate, 200 mg/ week i.m. and testolactone, 250 mg QID. This testos- Pretreatment serum testosterone concentrations were terone dose has been recently studied as a male in the normal physiologic range for young adult men. contraceptive [9]. Each treatment lasted 3 weeks, and Serum testosterone levels increased by 38% when testos- treatments were separated by a 4-week washout period. terone was given alone and by 102% when testosterone Blood samples were obtained from an antecubital vein and testolactone were combined (PB0.01 for both; before and after each 3 week treatment, between 06:00 Fig. 1). Testosterone also produced a 43% increase and 09:00, after a 12-h fast, and before testosterone (PB0.01) in serum estradiol levels, an effect that was injections. Plasma samples were collected in pre-chilled not observed when testosterone and testolactone were evacuated tubes that contained heparin as an anticoag- administered together (Fig. 1). Testolactone alone did ulant and were immediately placed on ice. These sam- not significantly change either testosterone or estradiol ples were separated by centrifugation usually within 1 h levels. These results indicate that testolactone inhibited (maximum 2 h), and frozen and stored at −70°C until aromatase activity and blocked the conversion of ex- analyzed. ogenous testosterone to estradiol. Mean pretreatment tHcy levels (5.091.1 vmol/l: Table 1 range, 3.3–7.0 vmol/l) were lower than previously re- Baseline characteristics of study subjects ported values for men of a similar age [12]. We suspect Mean S.D. that most of this difference is due to habitual use of B-vitamin-containing supplements, a common practice Age (years)27.1 7.4 among athletes. Fasting tHcy concentrations were not Weight (kg)86.6 20.7 significantly altered when testosterone was given alone Body fat (%) 12.7 6.9 or when testosterone and testolactone were combined Testosterone (nmol/l) 19.0 5.9 Estradiol (pmol/l) 145.7 52.7 (Table 2). Testolactone did not significantly affect Homocysteine (vmol/l) 5.0 1.1 plasma tHcy levels. Baseline serum testosterone (r=− 0.10; P=0.74) and estradiol (r=−0.13; P=0.65) con- sponses to acetylcholine. Although this reference referral normal subjects with endothelial dysfunc- group had normal coronary angiograms, referral due tion is necessary to assess the potential cardiovas- to chest pain sets these patients apart from a purely cular risk of this finding in a presumed low-risk voluntary group with normal coronary arteries. population. Much has been written about vasodilation of dis- eased and “control” human coronary arteries using an intravascular Doppler mounted catheter of the 1. Celermajer DS, Sorensen KE, Spiegelholter DJ, Georgakopoulos D, Robin- Millar Velocimeter (Houston, Texas) type.11-14Be- son J, Deanfield JE. Aging is associated with endothelial dysfunction in healthy cause of its catheter-based configuration, only prox- men years before the age-related decline in women. J Am Coil Cardiol 1994;24:471-476. imal straight segments of the coronary arteries may 2. Treasure CB, Klein L, Vita JA, Manoukian SV, Renwick GH, Selwyn AP, be safely and easily accessed with this technique. Ganz P, Alexander RW. Hypertension and left ventricular hypertrophy are associated with impaired endothelium mediated relaxation in human coronary This technique is also subject to a greater likelihood resistance vessels. Circulation 1993;87:86-93. of artifact and poorer signal quality secondary to the 3. Zeiher A, Drexler H, Wollschlager H, Just H. Modulation of coronary va- size and relative inflexibility of the catheter and dif- somotor tone in humans: progressive endothelial dysfunction with different early stages of coronary atherosclerosis. Cimdation 1991;83:391-401, ficulty in ensuring coaxial positioning. With use of 4. Drexler H, Zeiher AM. Endothelial function in human coronary arteries in this technique, endothelium-dependent and endothe- viva: focus on hypercholesterolemia. Hypertension 199 1;18( suppl):II-90-11. lium-independent vasodilation has been reported in 99. 5. Herrington DM, Braden GA, Williams JK, Morgan TM. Endothelial depen- control subjects defined variou~ly.~‘-‘~ In some dent coronary vasomotor responsiveness in postmenopausal women with and cases, mild coronary disease in the study vessel and/ without estrogen replacement therapy. Am J Cardiol 1994;73:951-952. 6. Quillen JE, Rossen JD, Oskarsson HJ, Minor RL Jr, Lopez AG, Winniford or severe disease in another vessel were allowed. MD. Acute effect of cigarette smoking on the coronary circulation: constriction Hypertension, hyperlipidemia, hyperglycemia, and of epicardial and resistance vessels. JAm Cob Cnrdiol 1993;22:642-647, current tobacco use are commonly seen in previously 7. Vita J, Treasure C, Nabel E. Coronary vasomotor response to acetylcholine relates to risk factors for coronary artery disease. Circulation 1990;8 1:49 l-497. reported control subjects. 8. Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990’s. Our study describes coronary relaxation proper- Nature 1993; 362:801-809. ties in a referral normal population, which we de- 9. Brown BG, Zhao X-Q, Sacco DE, Albers JJ. Lipid lowering and plaque regression: new insights into prevention of plaque disruption and clinical events fined simply as normotensive nondiabetic subjects in coronary disease. Circulation 1993;87:1781- 1791. without angiographic coronary artery disease. Our 10. Haskell WL, Alderman EL, Fair JM, Maron DJ, Mackey SF, Superko R, study suggests that an optimal intracoronary adeno- Williams PT, Johnstone IM, Champagne MA, Krauss RM, Farquhar JW. Effects of intensive multiple risk factor reduction on coronary atherosclerosis and clin- sine bolus infusion is 16 pg. Similarly, an optimal ical cardiac events in men and women with coronary artery disease: the Stanford intracoronary acetylcholine infusion rate appears to Coronary Risk Intervention Project (SCRIP). Circulation 1994;89:975-990. 11. Wilson RF, Laughlin DE, Ackell PH, Chilian WM, Holiday MD, Hartley be 30 pglmin. We found that 39% of referral normal CJ, Armstrong ML, Marcus ML, White CW. Transluminal, subselective mea- subjects had evidence of endothelial dysfunction, de- surement of coronary artery blood flow velocity and vasodilator reserve in man. fined as reduced endothelium-dependent relaxation Circulation 1985;72:82-92. 12. Houghton JL, Frank MJ, Carr AA, van Dohlen Tw, Prisant LM. Relations ( < 150% increase in coronary blood flow above among impaired coronary flow reserve, left ventricular hypertrophy and thallium baseline). perfusion defects in hypertensive patients without obstructive coronary artery disease. J Am CoU Cardiol 1990;15:43-51. In a referral normal cardiac population, endo- 13. Drexler H, Zeiher AM, Meinzer K, Just H. Correction of endothelial dysfunction in coronary microcirculation of hypercholesterolemic patients by L- thelium-independent coronary relaxation is nearly arginine. Lancer 1991;338:1546-1550. always normal, but endothelium-dependent relax- 14. Egashira K, Hirooka Y, Kai H, Sugimachi M, Suzuki S, Inou T, Tokeshita A. Reduction in serum cholesterol with pravastatin improves endothelium-de- ation may be depressed in a significant proportion pendent coronary vasomotion in patients with hypercholesterolemia. Circula- of patients. Further study of the natural history of tion 1994;89:2519-2524.

Testosterone Decreases Lipoprotein(a) in Men

Joseph M. Zmuda, MS, Paul D. Thompson, MD, Roberta Dickenson, BS, Linda L. Bausserman, PhD

ipoprotein (a) [ Lp (a)] has recently emerged as used lipid-lowering medications, only high-dose ni- L an important risk factor for atherosclerotic car- acin appears to reduce Lp(a) levels.2 Others have ,diovascular disease.’ Lp (a) levels are largely under demonstrated that estrogen 3 and the anabolic-andro- genetic control and have proved remarkably resistant genie steroid stanozolo14 reduce Lp (a) concentra- to therapeutic manipulations.’ Of the commonly tions in postmenopausal women, but to our knowledge, the effect of androgenic hormone administration on Lp(.a) in men has not been examined. Testoster- From the Divisions of Cardiology, The University of Pittsburgh, Pitts- burgh, Pennsylvania and The Miriam Hospital and Brown University, one is normally aromatized to estradiol, and periph- Providence, Rhode Island This study was sup orted by National In- eral aromatization of testosterone is the major source stitutes of Health Grant Ht.28467 and gifts Prom the Miriam Foun- of circulating estrogen in men.5 We recently exam- dation, the Haire family, Wrlliam Jakober, and the McNulty family. ined the effect of testosterone aromatization on se- Dr. Thompson’s address is: Preventive Cardiology, Suite 12 12, Kauf- mann Building, University of Pittsburgh Medical Center, 200 lothrop rum lipid and lipoprotein levels in men by admin- Street, Pittsbur h, Pennsylvania. Manuscript received November 29, istering testosterone alone or in combination with the 1995; revise f manuscript received and acceptedjanuary 9, 1996. aromatase inhibitor testolactone.6 Recent reports that

1244 0 1996 by Excerpta Medica, Inc 0002-9149/96/s 15.00 All rights reserved. PII SOOO2-9149(96)001769 estrogen decreases Lp( a)3 prompted us to examine report. Spear-mancorrelation coefficients were used Lp( a) levels from our study to determine the effects to examine the relation between initial Lp( a) values of testosterone and its aromatization to estradiol on and its subsequent change. Results are presented as Lp (a) levels. mean + SD...... Fourteen healthy nonsmoking male weightlifters Pretreatment serum testosterone levels were in the (aged 27 2 7 years; mean 2 SD) provided written normal physiologic range for young adult men. Se- informed consent and completed the study.6 Subjects rum testosterone levels increased by 39% when tes- weighed 85.4 + 21.0 kg before the study; body fat, tosterone was given alone and by 105% when tes- estimated from the sum of 3 skinfold measurements, tosterone and testolactone were combined (p l alcoholic bever- fect that was not observed when testosterone and tes- age daily or took regular medications, and all denied tolactone were administered together. Testolactone current and prior use of androgen. Baseline urinal- alone did not significantly change either testosterone ysis confirmed that the subjects had not recently used or estradiol levels. These results indicate that testo- either anabolic-androgenic steroids or testosterone.6 lactone inhibited aromatase activity and blocked the The subjects were instructed to maintain their habit- conversion of exogenous testosterone to estradiol. ual level of physical activity and avoid altering their Average Lp(a) values decreased by 37% during dietary habits during the study. Subjects were reim- treatment with testosterone, by 28% when testoster- bursed for their participation as approved by the Mir- one and testolactone were combined (p

for an individual subject in a single assay or autoan- Treatment Baseline Week 3 alyzer run. Intraassay coefficient of variation for Lp(a) was ~5%. Serum testosterone and estradiol Testosterone (nmol/L) were assayed in duplicate with radioimmunoassay Testosterone 18.7 % 5.5 26.0 t 4.9* kits (Diagnostic Products Corporation, Los Angeles, Testolactone 18.7 2 5.2 22.9 2 4.9 California). Testosterone + testolactone 18.4 ? 3.8 37.8 2 ll.l*‘f Data were analyzed with a treatment-by-time re- Estradiol (pmol/L)

peated measures analysis of variance. When inter- Testosterone 133 i 64 195 t 75* action effects were significant, the effect of time for Testolactone 133+32 118&50+ each treatment and the differences between treat- Testosterone + testolactone 130&44 113t22’

ments at each measurement point were tested statis- *Significant (p

BRIEF REPORTS 1245 Available online http://breast-cancer-research.com/content/11/5/212

Review Androgens and the breast Constantine Dimitrakakis1,2 and Carolyn Bondy1

1Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, CRC, Room 1-3330, 10 Center Drive, MSC-1103 Bethesda, Maryland 20892-1103, USA 21st Department of Ob/Gyn, Athens University Medical School, 80 Vas. Sophias Street, 11528, Athens, Greece

Corresponding author: Constantine Dimitrakakis, [email protected]

Abstract excess androgen exposure may increase the risk of breast Androgens have important physiological effects in women while at cancer in women [2]. the same time they may be implicated in breast cancer pathologies. However, data on the effects of androgens on mammary epithelial Experimental data suggest that conventional estrogen proliferation and/or breast cancer incidence are not in full treatment regimens, both as oral contraceptives (OCs) and agreement. We performed a literature review evaluating current hormone therapy (HT) [3], upset the normal estrogen/ clinical, genetic and epidemiological data regarding the role of androgen balance and promote ‘unopposed’ estrogenic androgens in mammary growth and neoplasia. Epidemiological studies appear to have significant methodological limitations and stimulation of mammary epithelial proliferation and, hence, thus provide inconclusive results. The study of molecular defects potentially breast cancer risk. This is due to suppression of involving androgenic pathways in breast cancer is still in its infancy. gonadotropins by exogenous estrogen treatment, resulting in Clinical and nonhuman primate studies suggest that androgens globally reduced ovarian steroidogenesis, so both inhibit mammary epithelial proliferation and breast growth while endogenous estrogen and androgen production are reduced, conventional estrogen treatment suppresses endogenous but only estrogens are provided by the treatment regimens. androgens. Abundant clinical evidence suggests that androgens normally inhibit mammary epithelial proliferation and breast growth. Additionally, estrogens, particularly in oral form, stimulate the Suppression of androgens using conventional estrogen treatment hepatic production of sex hormone binding globulin (SHBG), may thus enhance estrogenic breast stimulation and possibly which binds testosterone with high affinity, reducing breast cancer risk. Addition of testosterone to the usual hormone androgen bioavailability. As a result of this dual effect, total therapy regimen may diminish the estrogen/progestin increase in and bioavailable testosterone levels are significantly reduced breast cancer risk but the impact of this combined use on mammary gland homeostasis still needs evaluation. in women taking oral contraceptives or estrogen supplementation for ovarian insufficiency [4].

Introduction This literature review compares the findings that androgens in Treatment of women with physiological testosterone women promote the risk for breast cancer versus the supplementation to remedy hypoactive sexual desire disorder evidence that androgens protect the mammary gland from is an area of great interest at present [1]. While it seems hormone-induced stimulation, increased proliferation and evident that testosterone treatment increases sexual activity, neoplasia. the risk-benefit ratio for such treatment remains unclear. Androgen receptors are found in virtually every tissue in Normal breast development: estrogens and women as well as men, including breast, bone and brain, androgens indicating that androgens and their metabolites may play an Estrogens stimulate, while androgens inhibit breast develop- important role in normal tissue homeostasis and possibly in ment independently of genetic sex. Breast tissue is similar in pathologies like breast cancer, osteoporosis, libido and prepubertal boys and girls. Pubertal rises in estrogen levels cognitive decline. Thus, testosterone treatment to improve cause breast growth in girls and frequently in boys sexual function may have unintended effects in diverse (transiently). Estradiol levels are significantly higher in girls tissues. A continuing area of concern is the notion that with premature thelarche than in normal prepubertal girls. An

AR = androgen receptor; DHEA = dehydroepiandrosterone; DHEA-S = DHEA sulfate; DHT = dihydrotestosterone; ER = estrogen receptor; HSD = hydroxysteroid dehydrogenase; HT = hormone therapy; OC = oral contraceptive; PR = progesterone receptor; SHBG = sex hormone binding globulin; WHI = Women’s Health Initiative.

Page 1 of 9 (page number not for citation purposes) Breast Cancer Research Vol 11 No 5 Dimitrakakis and Bondy

association between expression of a high activity isoform of The mammary gland is capable of synthesizing both estradiol the testosterone metabolizing CYP3A4 and the early onset and testosterone. All the steroidogenic enzymes necessary of thelarche has been documented, suggesting that for the formation of androgens and estrogens from steroid decreasing testosterone levels may also trigger early breast precursors - steroid sulfatase, 17β-hydroxysteroid dehydroge- growth [5]. Conversely, androgen excess due to adrenal nases (17β-HSDs), 3β-HSDs, 5α-reductases and aromatase - tumor or hyperplasia suppresses normal breast develop- have been reported in normal mammary tissues, breast ment in girls, despite apparently adequate estrogen levels cancer specimens or cell lines [10]. Androgens stimulate or [6]. In castrated male-to-female transsexuals, feminizing inhibit the growth of breast cancer cells in vitro depending on estrogen therapy stimulates breast growth with full acinar the cell line and clone under study according to former data and lobular formation and estrogen-treated genetically male [11]. Breast cancer cell lines and tissue specimens express breast tissue exhibits normal female histology. Estrogens the enzymes involved in DHT as well as estradiol synthesis. In taken to treat prostate cancer also lead to breast a histochemical study, expression of 5α-reductase was development in men with suppressed gonadal function and significantly correlated with androgen receptor expression reduced testosterone levels. Conversely, androgen use by and 17β-HSD and 3β-HSD immunoreactivities and the female athletes and female-to-male transsexuals leads to abundance of this androgenic molecular assembly was breast atrophy. inversely correlated with tumor size, histological grade and proliferative index [12], suggesting an inhibitory role for DHT Supporting the normal inhibitory role of endogenous in tumor growth. androgens on breast growth, androgen receptor (AR) blockade with flutamide causes gynecomastia and rarely breast adeno- Androgen receptor carcinoma [7]. Males may also develop gynecomastia when the Androgen agonists such as testosterone and DHT function estrogen/androgen ratio is increased due to decreased through binding to the intracellular AR. This is a member of androgen production or increased aromatization. the nuclear hormone receptor super-family comprising classic DNA-binding, hormone binding and activation domains The balance between stimulatory effects of the estrogens and (Figure 1). AR expression is abundant in normal mammary inhibitory effects of the androgens is the critical factor that epithelium and in the majority of breast cancer specimens regulates mammary cell proliferation both in normal and in and cell lines. The AR is co-localized with estrogen and cancer tissues [8]. It has not been possible to identify progesterone receptors in epithelial cells but not detected in specific estrogen/androgen ratios predictive of breast mammary stroma or myoepithelium [13]. The co-expression of stimulation or inhibiting effects for several reasons. Estradiol estrogen receptor (ER) and AR in mammary epithelial cells and testosterone assays have not been very sensitive nor suggests that the effects of estrogen and androgen on accurate in the lower ranges, while both hormones bind to mammary epithelial proliferation are integrated within the SHBG, so total values may not be as informative as ‘free’ or mammary epithelial cell. The AR gene is located on the X bioavailable hormone [4]. Moreover, single hormone chromosome with no corresponding allele on the Y, so it measurements may not be very informative about tissue functions solely as a single copy gene, as shown by the exposure over time. Both estradiol and testosterone levels complete loss of androgen effect in XY individuals with an vary hourly in response to diurnal rhythm, diet, stress and inactivating mutation of the AR [14]. exercise, so a single value may be inadequate to assess true tissue exposure. In addition, estradiol and testosterone may Binding of testosterone or DHT triggers a cascade of signa- be synthesized locally in peripheral tissues from circulating ling events, including phosphorylation and conformational precursors such as the sulfate of dehydroepiandrosterone changes in the receptor, which dissociates from cytoplasmic (DHEA-S) and androstendione [9]. The conjugated products proteins and migrates to the cell nucleus. Ligand-activated of steroid metabolism find their way into the circulation after AR regulates gene expression through binding to androgen peripheral action and provide evidence as to the proportion of response elements located in a gene’s enhancer or promoter the precursor pools of steroids utilized as androgen or region. As with other similar receptors, the AR functions in estrogen. Analysis of these metabolites by Labrie and transcriptional regulation in concert with a host of nuclear colleagues [9] and Sasano and colleagues [10] suggested proteins, which may serve as co-activators or co-repressors. that the major proportion of androgen effectors in women Interestingly, the BRCA1 gene product has been identified as derive from such an endocrine mode of action, which will not an AR co-activator [15]. The BRCA1 protein binds to the AR be detected by assays of circulating testosterone or and potentiates AR-mediated effects, suggesting that dihydrotestosterone (DHT). Interestingly, while circulating BRCA1 mutations may blunt androgen effects. However, levels of testosterone and DHT are five- to ten-fold higher in other studies have not confirmed these findings [16]. men than women, the abundance of androgen metabolites is less than two-fold higher in men, suggesting that local tissue AR has a highly polymorphic CAG repeat in exon 1 that production and action of androgens in women may be more encodes a polyglutamine stretch (Figure 1). There is evidence significant than historically suspected. that longer CAG repeats are associated with earlier age of

Page 2 of 9 (page number not for citation purposes)

Available online http://breast-cancer-research.com/content/11/5/212

Figure 1

Schematic design of the androgen receptor gene (top) and protein (below). The polymorphic trinucleotide repeat site (CAG) is indicated in green at the left. Trans-activating function (TAF), DNA-binding (DBD) and ligand-binding domains (LBD) are labeled. breast cancer onset [16]. However, other studies have not remains unclear. It is possible that the steroid receptor confirmed this finding [17]. In a study nested within the contributes differently in healthy compared to cancerous Nurses’ Health Study cohort [18], no relation was found breast tissue; thus, a number of unanswered questions between AR genotype and breast cancer risk among post- remain and further studies are needed before safe conclu- menopausal Caucasian women overall, but an increased risk sions are drawn. was observed when analysis was limited to those individuals with a first-degree family history of breast cancer. Germ-line The hypothesis that androgens are directly involved in breast mutations in the AR gene conferring variable degrees of carcinogenesis is based on the presence of ARs in the androgen insensitivity have been associated with the majority of breast cancers. It is proposed that androgens, occurrence of breast cancer in men [19]. Another study [20] through binding to their receptors, act independently to provides evidence that the association with the long AR-CAG produce tumors with specific clinical behaviors [24]. Clinical was observed only in postmenopausal and not premeno- data support that a significant number of poorly differentiated pausal women, which may explain the insignificant results in breast carcinomas are ER-negative and progesterone studies restricted to young women. In other studies, reduced receptor (PR)-negative but AR-positive, or patients with AR- risk was observed with another trinucleotide repeat, GGC, in positive tumors experience a better disease-free survival. young women. AR repeat length might be partly responsible These associations constitute important clinical and for the increased risk of early onset breast cancer in women pathologic prognostic information. Recently, AR expression in using oral contraceptives or HT [21]. a tumor is considered as an indicator of lower malignancy potential; this provides a new range of therapeutic targets for Emphasis should be given to the fact that none of these poorly differentiated cancers [25]. studies had sufficient statistical power to implicate or exclude specific AR defects in breast cancer risk. A recent epidemio- Androgens and breast cancer: logical meta-analysis concludes that there is no association epidemiological data between AR genetic variations and breast cancer risk among Long-term estrogen treatment increases the risk of breast Caucasian women [17]. cancer in both males and females through estrogenic stimulation of mammary epithelial proliferation. Additional AR-CAG repeat length was inversely associated with testos- carcinogenic effects by estrogen metabolites have been terone levels in both pre- and postmenopausal normal women proposed [26]. The most widely accepted risk factor for [22]. Lillie and colleagues [23] evaluated the association breast cancer is the cumulative dose of estrogens that breast between AR-CAG repeat length and mammographic density, epithelium is exposed to over time. However, it has been a strong breast cancer risk factor. They found that in difficult to correlate breast cancer risk with isolated serum postmenopausal estrogen/progesterone users, carriers of the estrogen levels in epidemiological studies, probably secon- less active AR had a higher mean percentage of density than dary to problems using single random hormone levels for the carriers of the more active AR. This suggests that AR evaluation of tissue-specific exposure as discussed above. genotype modifies hormone-induced proliferation as reflected in mammographic density and explains the mechanism by Attempts to correlate adrenal precursor steroids with breast which estrogen/progesterone use increases breast cancer cancer incidence have been relatively successful, or at least risk. However, the exact mechanisms and metabolic paths in consistent, perhaps reflecting the importance of local tissue which AR participates in normal tissues are still obscure. The conversion. Many years ago, reduced 17-ketosteroid excre- role of AR in oncogenesis or breast tumor proliferation tion was noted in the urine of pre-menopausal women with

Page 3 of 9 (page number not for citation purposes) Breast Cancer Research Vol 11 No 5 Dimitrakakis and Bondy

breast cancer and subsequent studies have documented estrogen levels in premenopausal women, androstendiol reduced DHEA and its sulfate (DHEA-S) in the serum of pre- could exhibit anti-estrogenic effects, while in the hypo- menopausal breast cancer patients. In the first prospective estrogenic postmenopausal milieu, the agonist effect may study in this field, levels of androgen metabolites in urine predominate. This view remains speculative, and other were found to be abnormally reduced in premenopausal possibilities still exist. It is possible that the high estrogen women who subsequently developed breast cancer [27], environment in premenopausal women promotes androgenic indicating a protective role of androgens on the breast. In enzyme and AR expression in mammary tissue, allowing contrast, in a recent prospective study of pre-menopausal androgenic effects by DHEA metabolites, while in postmeno- women [28], no association was found between plasma pausal women, an estrogen-deficient tissue microenviron- adrenal androgen levels and risk of breast cancer. In the ment may favor estrogenic effects. Nurses’ Health Study II, no correlation between DHEA and DHEA-S levels and breast cancer risk overall was found but, Genetic variation in CYP19 and SHBG genes was found to interestingly, among premenopausal women there was a contribute to the variance in circulating hormone levels in positive association, especially for tumors that express both postmenopausal women, but none was statistically signifi- ERs and PRs [29]. Also, among premenopausal women, cantly associated with breast cancer risk [32]. higher levels of testosterone and androstendione were associated with increased risk of invasive ER+/PR+ tumors, In some prospective epidemiological studies, age-adjusted although with a non-statistically significant increase in overall mean values of total and free testosterone and estradiol were risk of breast cancer [29]. significantly higher pre-diagnostically in postmenopausal breast cancer cases compared with controls, and estradiol Several epidemiological studies have examined the correla- and total testosterone were elevated in other case-control tion of circulating androgens, such as testosterone, and risk studies of postmenopausal breast cancer. It was observed for breast cancer. A major limitation of these studies was that that elevated serum levels of both estrogens and androgens the androgen assays used were developed primarily to contribute to a greater risk of breast cancer [33] and a meta- measure the higher levels found in men and lack reliability in analysis of nine prospective studies revealed that breast the low ranges found in normal women [4]. Testosterone and cancer risk increases with increasing concentrations of androstenedione levels demonstrate substantial daily almost all sex hormones [34]. variability, while most of the epidemiological data are based on a single blood sample collected at non-standard times. None of these studies manage, however, to disconnect the Another problem using serum testosterone levels to gauge risk associated with increased estradiol levels from the androgenic effects at the tissue level is that most of the androgen component, and since androgens are the obligate circulating testosterone is tightly bound to SHBG, while only precursors for estradiol synthesis, this is a major confounding the free hormone is bioactive. SHBG, and thus total testos- factor in assessing the role of androgen independently of the terone levels, vary widely based on genetic, metabolic and known cancer-promoting estrogen effect. In line with these endocrine influences, and it is now accepted that measure- observations, a recent study [35] concluded that increased ment of free or bioavailable testosterone predicts androgenic breast cancer risk with increasing body mass index among effects more accurately than total testosterone levels [4]. postmenopausal women is largely the result of the associated Finally, most androgenic activity in women originates from the increase in estrogens. The association of androgens with peripheral conversion of precursors such as DHEA into breast cancer risk did not persist after adjustment for estrone, androgens within the cells of target tissues, and this activity the estrogen most strongly associated with the risk. Other will not be detected by measurement of circulating androgens. authors conclude that conversion of DHEA to estrogens, particularly estradiol, is required to exert a mitogenic In a recent study [30], levels of testosterone and DHEA-S in response [36]. These results suggest that the contribution of saliva were statistically significantly lower in breast cancer androgens to breast cancer risk is largely through their role as patients compared to controls and these differences were substrates for estrogen production. more profound in postmenopausal women. Breast cancer patients, when compared to controls, presented with an Other studies have found no association between androgens androgen insufficiency and a relative imbalance of sex steroid and breast cancer [37,38]. Recent epidemiological studies hormones in favor of estrogens. on the association between androgen levels and breast cancer risk are summarized in Table 1. Several studies have revealed, however, that adrenal androgens are increased in postmenopausal women with breast The above mentioned observations indicate the difficulty cancer [31]. A possible explanation regarding the divergence separating potential direct effects of circulating testosterone between pre- and postmenopausal findings is that one from its potential to be aromatized into estradiol. An interesting adrenal ‘androgen’, androstenediol, also known as ‘herma- research topic would be to investigate levels of testosterone phrodol’, is a weak ER agonist. In the presence of high and DHT metabolites in these studies in order to assess more

Page 4 of 9 (page number not for citation purposes) Available online http://breast-cancer-research.com/content/11/5/212

Table 1

Epidemiological studies on the association between androgens and breast cancer risk

Study Conclusion

Elevated levels of androgens (and estrogens) are associated with increased risk of breast cancer Tamimi et al. 2006 [47] Prospective cohort study in Nurses’ Health Study with over a million person-years studied: women receiving postmenopausal hormones with testosterone had a 17.2% increased risk of breast cancer per year of use Micheli et al. 2007 [54] Breast cancer patients (n = 194) with high testosterone had significantly lower event-free survival than those with low testosterone (P = 0.004) and a significantly higher risk of breast cancer events with an adjusted hazard ratio of 1.77 (95% CI, 1.06 to 2.96) The Endogenous Hormones Meta-analysis: breast cancer risk increases statistically significantly with increasing concentrations of almost all and Breast Cancer sex hormones Collaborative Group 2002 [34] Tworoger et al. 2006 [31] Prospective nested case-control study within the Nurses’ Health Study II: adrenal androgens are positively associated with breast cancer among predominately premenopausal women (for example, for DHEA: RR, 1.6; 95% CI, 0.9 to 2.8; P = 0.09) Eliassen et al. 2006 [29] Nurses’ Health Study II, nested: higher levels of testosterone and androstenedione in 18,521 premenopausal women are associated with insignificant overall increase in breast cancer risk, but increased risk of invasive and ER+/PR+ cancers (for example, RR = 2.9; CI = 1.4 to 6.0)

Androgen levels acting with protective patterns Hofling et al. 2007 [51] Randomized, double-blind, placebo-controlled study: testosterone use inhibited exogenous estrogen-induced breast tissue proliferation in 99 postmenopausal women (P < 0.001) Dimitrakakis et al. 2004 [48] Retrospective, observational study that followed 508 postmenopausal women receiving testosterone in addition to usual hormone therapy: incidence of breast cancer in testosterone users was substantially less than in women receiving estrogen/progestin in the WHI study and in the Million-woman study Suzuki et al. 2001 [55] Intratumoral dihydrotestosterone inhibits cancer cell proliferation in hormone-dependent human breast carcinoma Haiman et al. 2002 [18] A case-control study nested within the Nurses’ Health Study cohort (cases, n = 727; controls, n = 969): longer CAG repeat alleles of AR increases breast cancer risk (odds ratio, 1.70; 95% CI, 1.20 to 2.40; P = 0.04) MacLean et al. 2004 [19] Forty-one male breast cancers were studied: incidence of longer CAG repeats in AR was significantly higher in the breast cancer group than in the normal population (P < 0.05) Ogawa et al. 2008 [25] In 227 primary breast cancers, AR expression was significantly higher in breast tumors with favorable characteristics Dimitrakakis et al. 2009 [30] Testosterone and DHEA-S salivary levels were statistically significantly lower in breast cancer patients compared to controls (n = 541)

No association between serum concentrations of androgens and breast cancer risk Ness et al. 2009 [46] A group of postmenopausal participants in the WHI study used testosterone combined with estrogens: testosterone addition had no statistically significant effect on breast cancer occurrence Cox et al. 2006 [17] Among postmenopausal women, common variants of the AR gene are not associated with risk of breast cancer Page et al. 2004 [28] Prospective observational study: no relationship between serum DHEA or DHEA-S and subsequent breast cancer in middle-aged women Olson et al. 2007 [32] No association with breast cancer risk was detected for individual variants of CYP19 mutation in 750 cases Adly et al. 2006 [37] Serum levels of steroids in 331 women: androgen levels were not independently associated with increased risk of breast cancer Beattie et al. 2006 [38] Case-cohort design including 135 postmenopausal women with and 275 without breast cancer enrolled in the NSABBP P-1 trial: risk of breast cancer was not associated with sex hormone levels

AR, androgen receptor; CI, confidence interval; DHEA, dehydroepiandrosterone; DHEA-S, DHEA sulfate; ER, estrogen receptor; NSABBP, National Surgical Adjuvant Breast and Bowel Project; PR, progesterone receptor; RR, relative risk; WHI, Women’s Health Initiative. directly tissue exposure to androgens. As noted above, a single levels directly contribute to breast cancer, then women with serum hormone measurement seems unlikely to be informative clinically evident long-term hyperandrogenism - for example, about a woman’s true long-term exposure to that hormone or polycystic ovary syndrome and congenital adrenal hyperplasia - her specific risk of developing breast cancer. Nor does it seem should experience increased rates of breast cancer, but this is to be a biologically plausible mechanism that androgens acting not the case [39]. Moreover, androgen levels are chronically as androgens could promote breast cancer, since virtually all elevated in men, who have a breast cancer risk less than 1% of clinical data suggest just the opposite. If elevated androgen that of women. This is despite the fact that lifetime estradiol

Page 5 of 9 (page number not for citation purposes) Breast Cancer Research Vol 11 No 5 Dimitrakakis and Bondy

levels are not much lower in men than in women. In fact, Figure 2 decreased androgen levels found, for example, in Kleinefelter’s syndrome and other hypogonadal syndromes increase the risk of breast cancer in males [40]. Epidemiological studies in men indicate that low urinary androsterone and serum free testosterone levels are related to early onset of breast cancer, a much higher relapse rate and a worse response to endocrine therapy [41].

Hormone therapy and androgens Exposure to endogenous and exogenous estrogens is thought to contribute to increased breast cancer risk. Since the introduction of combined OCs 40 years ago, many changes in doses and their biochemical structures have taken place and the possibility that OCs may increase the risk of breast cancer has been the subject of intense research. Although many epidemiological studies in the past have not linked OC use to breast cancer risk, several more recent studies have found an association, either overall or especially in subgroups of Average estradiol (E2) and testosterone (T) levels across the female women. A large meta-analysis on previously published studies lifespan. Y-axis, level in picograms; X-axis, age in years. Dashed lines [42] calculated a small but significant increase in relative risk predict changes in T and E2 hormone levels resulting from estrogen of breast cancer (RR = 1.24) in current OC users while other replacement therapy (ERT) beginning at menopause. publications [43] have not associated current or former OC use with an increased risk of breast cancer. However, because pill users are young, this represents a very small therapy or to women who never used postmenopausal increase in absolute risk. It is not yet known if lower dose and hormone formulations [47]. variable OC formulations are associated with a similar increase in risk, making comparisons very difficult. If androgens are protective against breast cancer as many studies suggest, then conventional HT may promote breast The bulk of the currently available evidence supports a causal cancer not only by increasing estrogen exposure but also by relationship between the use of HT and breast cancer. Recent decreasing endogenous androgen activity. Oral estrogen and long-term users of HT are associated with higher risk. The therapy reduces free androgens by stimulating hepatic effect of concurrent progestin use appears to further increase production of SHBG and through suppression of luteinizing risk above that with estrogens alone. The most important hormone, thus inhibiting ovarian androgen production [4]. randomized clinical trial providing information about this issue Thus, institution of pharmacological estrogen therapy at is the Women’s Health Initiative (WHI) study [44] reporting menopause may result in a drastic reduction in the increased risk of breast cancer in women who took estrogen testosterone/estradiol ratio, which is normally maintained at plus progestin but not in women who took estrogen alone relatively high levels throughout a woman’s lifespan [45]. The results from observational studies are generally (Figure 2). consistent with those of the WHI trial, reporting no significant variation in the risk of breast cancer with use of different Pertinent to these results, our published article [48] provides estrogens, progestins, doses, or routes of administration. important information about the addition of testosterone to the HT regimen. In our evaluation of 508 postmenopausal A group of postmenopausal participants in the WHI study women in Australia receiving testosterone in addition to usual used testosterone combined with estrogens. In this group, HT, the incidence of breast cancer in testosterone users was testosterone addition for a period of a year had no statistically substantially less than in women receiving estrogen/progestin significant effect on breast cancer occurrence, suggesting at in the WHI study and in the Million-woman study [49]. Breast least that androgen induction did not increase the number of cancer rates in the testosterone users was closer to that breast cancer cases in this trial [46]. In the same study, rates reported for HT never users, and their age-standardization of breast cancer were lower in longer-term compared to rate was the same as for the general population in South shorter-term users of estrogen plus testosterone. On the Australia. These observations suggest that the addition of other hand, in a prospective study of over a million person- physiological doses of androgen to OCs and HT could years with 24 years of follow-up within the Nurses’ Health protect the breast from ‘unopposed’ estrogenic effects. Study, current users of estrogen plus testosterone have shown a 2.5-fold increased risk of developing breast cancer Suppression of normal endogenous androgen may be an compared to menopausal women who used estrogen-only adverse consequence of pharmacological estrogen therapy, if

Page 6 of 9 (page number not for citation purposes) Available online http://breast-cancer-research.com/content/11/5/212

Figure 3

Mammary epithelial proliferation shown by Ki67 immunoreactivity (brown dots) in ovariectomized monkeys treated with (a) placebo (Con), (b) estradiol (E2), (c) E2 and progesterone (P4), (d) tamoxifen (Tam) and (e) E2 and testosterone (T). (f) Quantification of the Ki67 proliferation index. Proliferation is increased with E2 or E2 and P4 (E/P), while this increase is attenuated by the addition of T to E2 (E/T). All differences are statistically significant when compared to the placebo group. Data from Zhou and colleagues [53]. androgens are indeed protective against estrogen-induced influence mammographic breast density like conventional HT mammary proliferation. We have shown that addition of low [51,52]. The antiproliferative effects of androgens on breast physiological doses of testosterone (producing serum levels tissue may occur either indirectly, via downregulation of other in the mid-normal range for women as well as rhesus receptors like PRs, or directly, through breast AR stimulation. monkeys) to estrogen therapy in ovariectomized rhesus monkeys significantly inhibits HT-induced mammary epithelial Women, and particularly postmenopausal women, have been proliferation [3] (Figure 3). Additionally, testosterone treat- treated with testosterone for female sexual dysfunction for ment significantly reduced mammary epithelial estrogen almost six decades. The exact biological role of androgens in receptor expression, thus suggesting a potential mechanism the restoration of libido in hypoactive sexual desire disorder in for the growth inhibitory effect. Moreover, we have found that females is still unclear. The main safety concern for women treatment of intact cycling monkeys with the AR antagonist who have undergone years of this therapy has been breast and flutamide resulted in a significant increase in mammary endometrial cancer risk related to androgens. In a recent trial of epithelial proliferation [3], adding to the burden of evidence 814 sexually hypoactive women, the results for breast cancer that endogenous androgens normally limit mammary prolifera- risk were inconclusive [1]. Nevertheless, current experience tion and, hence, cancer risk. Other studies on primates also does not confirm a positive correlation between testosterone suggest that inclusion of testosterone with estrogen/ use and breast cancer occurrence; thus, androgens do have a progesterone use may counteract breast cell proliferation place in female sexual dysfunction treatment. [50]. In a recent randomized, double-blind, placebo-controlled study, testosterone use inhibited exogenous estrogen- Conclusion induced breast tissue proliferation in postmenopausal women This review focuses on the role of androgens with regard to [51]. There is also evidence that testosterone does not breast growth and neoplasia. Measurement of circulating sex

Page 7 of 9 (page number not for citation purposes) Breast Cancer Research Vol 11 No 5 Dimitrakakis and Bondy

steroids and their metabolites demonstrates that androgen parameters in breast cancer patients. J Steroid Biochem Mol activity is normally abundant in healthy women throughout Biol 2009, 113:195-201. 13. Zhou J, Anderson K, Bievre M, Ng S, Bondy CA: Primate their entire lifetime. Epidemiological studies investigating mammary gland insulin-like growth factor system: cellular testosterone levels and breast cancer risk have major localization and regulation by sex steroids. J Investig Med 2001, 49:47-55. theoretical and methodological limitations and do not provide 14. Avila DM, Zoppi S, McPhaul MJ: The androgen receptor (AR) in consensus. The molecular epidemiology of defects in syndromes of androgen insensitivity and in prostate cancer. J pathways involved in androgen synthesis and activity in Steroid Biochem Mol Biol 2001, 76:135-142. 15. Park JJ, Irvine RA, Buchanan G, Koh SS, Park JM, Tilley WD, Stall- breast cancer hold great promise, but investigation of these is cup MR, Press MF, Coetzee GA: Breast cancer susceptibility still in the early stages. Clinical observations and experimental gene 1 (BRCAI) is a coactivator of the androgen receptor. data indicate that androgens inhibit mammary growth, and Cancer Res 2000, 60:5946-5949. 16. Spurdle AB, Antoniou AC, Duffy DL, Pandeya N, Kelemen L, Chen they have been used in the past with success to treat breast X, Peock S, Cook MR, Smith PL, Purdie DM, Newman B, Dite GS, cancer. It is of concern that current forms of estrogen treat- Apicella C, Southey MC, Giles GG, Hopper JL, Chenevix-Trench G, Easton DF; EMBRACE Study Collaborators: The androgen ment in OCs and for ovarian failure result in suppression of receptor CAG repeat polymorphism and modification of endogenous androgen activity considering that the addition breast cancer risk in BRCA1 and BRCA2 mutation carriers. of testosterone to the HT regimen ameliorates the stimulating Breast Cancer Res 2005, 7:R176 17. Cox DG, Blanché H, Pearce CL, Calle EE, Colditz GA, Pike MC, effects of estrogen/progestin on the breast. Research Albanes D, Allen NE, Amiano P, Berglund G, Boeing H, Buring J, addressing the role of androgens in breast cancer prevention Burtt N, Canzian F, Chanock S, Clavel-Chapelon F, Feigelson HS, Freedman M, Haiman CA, Hankinson SE, Henderson BE, Hoover and the efficacy of hormonal supplementation with R, Hunter DJ, Kaaks R, Kolonel L, Kraft P, LeMarchand L, Lund E, physiological androgen to maintain estrogen/androgen ratios Palli D, Peeters PH, Riboli E, et al.: A comprehensive analysis of typical of normal women is warranted. the androgen receptor gene and risk of breast cancer: results from the National Cancer Institute Breast and Prostate Cancer Cohort Consortium (BPC3). Breast Cancer Res 2006, 8:R54. Competing interests 18. Haiman CA, Brown M, Hankinson SE, Spiegelman D, Colditz GA, The authors declare that they have no competing interests. Willett WC, Kantoff PW, Hunter DJ: The androgen receptor CAG repeat polymorphism and risk of breast cancer in the Nurses’ Health Study. Cancer Res 2002, 62:1045-1049. References 19. MacLean HE, Brown RW, Beilin J, Warne GL, Zajac JD: 1. Davis SR, Moreau M, Kroll R, Bouchard C, Panay N, Gass M, Increased frequency of long androgen receptor CAG repeats Braunstein GD, Hirschberg AL, Rodenberg C, Pack S, Koch H, in male breast cancers. Breast Cancer Res Treat 2004, 88:239- Moufarege A, Studd J; APHRODITE Study Team: Testosterone 246. for low libido in postmenopausal women not taking estrogen. 20. Giguere Y, Dewailly E, Brisson J, Ayotte P, Laflamme N, Demers N Engl J Med 2008, 359:2005-2017. A, Forest VI, Dodin S, Robert J, Rousseau F: Short polygluta- 2. Schover LR: Androgen therapy for loss of desire in women: is mine tracts in the androgen receptor are protective against the benefit worth the breast cancer risk? Fertil Steril 2008, 90: breast cancer in the general population. Cancer Res 2001, 61: 129-140. 5869-5874. 3. Dimitrakakis C, Zhou J, Wang J, Belanger A, LaBrie F, Cheng C, 21. Abbas S, Brauch H, Chang-Claude J, Dunnebier T, Flesch-Janys Powell D, Bondy C: A physiologic role for testosterone in limit- D, Hamann U, Hein R, Justenhoven C, Salazar R: Polymorphisms ing estrogenic stimulation of the breast. Menopause 2003, 10: in genes of the steroid receptor superfamily modify post- 292-298. menopausal breast cancer risk associated with menopausal 4. Lobo RA: Androgens in postmenopausal women: production, hormone therapy. Int J Cancer 2009, in press. possible role, and replacement options. Obstet Gynecol Surv 22. Westberg L, Baghaei F, Rosmond R, Hellstrand M, Landen M, 2001, 56:361-376. Jansson M, Holm G, Bjorntorp P, Eriksson E: Polymorphisms of 5. Kadlubar F, Berkowitz G, Delongchamp R, Green B, Wang C, the androgen receptor gene and the estrogen receptor beta Wolff MS: The putative high activity variant, CYP3A4*1B, pre- gene are associated with androgen levels in women. J Clin dicts the onset of puberty in young girls. Proc Am Assoc Endocrinol Metab 2001, 86:2562-2568. Cancer Res 2001, 42:408. 23. Lillie EO, Bernstein L, Ingles SA, Gauderman WJ, Rivas GE, 6. Forsbach G, Guitron-Cantu A, Vazquez-Lara J, Mota-Morales M, Gagalang V, Krontiris T, Ursin G: Polymorphism in the andro- Diaz-Mendoza ML: Virilizing adrenal adenoma and primary gen receptor and mammographic density in women taking amenorrhea in a girl with adrenal hyperplasia. Arch Gynecol and not taking estrogen and progestin therapy. Cancer Res Obstet 2000, 263:134-136. 2004, 64:1237-1241. 7. Karamanakos P, Mitsiades CS, Lembessis P, Kontos M, Trafalis D, 24. Nicolas Diaz-Chico B, German Rodriguez F, Gonzalez A, Ramirez Koutsilieris M: Male breast adenocarcinoma in a prostate R, Bilbao C, Cabrera de Leon A, Aguirre Jaime A, Chirino R, cancer patient following prolonged anti-androgen monother- Navarro D, Diaz-Chico JC: Androgens and androgen receptors apy. Anticancer Res 2004, 24:1077-1081. in breast cancer. J Steroid Biochem Mol Biol 2007, 105:1-15. 8. Labrie F: Dehydroepiandrosterone, androgens and the 25. Ogawa Y, Hai E, Matsumoto K, Ikeda K, Tokunaga S, Nagahara H, mammary gland. Gynecol Endocrinol 2006, 22:118-130. Sakurai K, Inoue T, Nishiguchi Y: Androgen receptor expression 9. Labrie F, Luu-The V, Lin SX, Simard J, Labrie C, El-Alfy M, Pelletier in breast cancer: relationship with clinicopathological factors G, Belanger A: Intracrinology: role of the family of 17 beta- and biomarkers. Int J Clin Oncol 2008, 13:431-435. hydroxysteroid dehydrogenases in human physiology and 26. Yager JD, Davidson NE: Estrogen carcinogenesis in breast disease. J Mol Endocrinol 2000, 25:1-16. cancer. N Engl J Med 2006, 354:270-282. 10. Sasano H, Suzuki T, Nakata T, Moriya T: New development in 27. Bulbrook RD, Thomas BS, Utsunomiya J, Hamaguchi E: The intracrinology of breast carcinoma. Breast Cancer 2006, 13: urinary excretion of 11-deoxy-17-oxosteroids and 17-hydroxy- 129-136. corticosteroids by normal Japanese and British women. J 11. Birrell SN, Bentel JM, Hickey TE, Ricciardelli C, Weger MA, Hors- Endocrinol 1967, 38:401-406. fall DJ, Tilley WD: Androgens induce divergent proliferative 28. Page JH, Colditz GA, Rifai N, Barbieri RL, Willett WC, Hankinson responses in human breast cancer cell lines. J Steroid SE: Plasma adrenal androgens and risk of breast cancer in Biochem Mol Biol 1995, 52:459-467. premenopausal women. Cancer Epidemiol Biomarkers Prev 12. Suzuki M, Ishida H, Shiotsu Y, Nakata T, Akinaga S, Takashima S, 2004, 13:1032-1036. Utsumi T, Saeki T, Harada N: Expression level of enzymes 29. Eliassen AH, Missmer SA, Tworoger SS, Spiegelman D, Barbieri related to in situ estrogen synthesis and clinicopathological RL, Dowsett M, Hankinson SE: Endogenous steroid hormone

Page 8 of 9 (page number not for citation purposes) Available online http://breast-cancer-research.com/content/11/5/212

concentrations and risk of breast cancer among pre- Curb D, Gass M, Hays J, Heiss G, Hendrix S, Howard BV, Hsia J, menopausal women. J Natl Cancer Inst 2006, 98:1406-1415. Hubbell A, Jackson R, Johnson KC, Judd H, Kotchen JM, Kuller L, 30. Dimitrakakis C, Glaser RL, Zava DT, Tsigginou A, Marinopoulos S, LaCroix AZ, Lane D, Langer RD, Lasser N, Lewis CE, Manson J, Antsaklis A: The protective role of androgens: salivary Margolis K, Ockene J, et al.: Effects of conjugated equine estro- hormone levels in newly diagnosed breast cancer patients In gen in postmenopausal women with hysterectomy: the 26th Annual Miami Breast Cancer Conference: March 4-7, Women’s Health Initiative randomized controlled trial. JAMA 2009; Miami, FL. 2009. [http://www.cancerlearning.com/index. 2004, 291:1701-1712. cfm/fuseaction/conference.showOverview/id/5/conference_id/93] 46. Ness RB, Albano JD, McTiernan A, Cauley JA: Influence of estro- 31. Tworoger SS, Missmer SA, Eliassen AH, Spiegelman D, Folkerd gen plus testosterone supplementation on breast cancer. E, Dowsett M, Barbieri RL, Hankinson SE: The association of Arch Intern Med 2009, 169:41-46. plasma DHEA and DHEA sulfate with breast cancer risk in 47. Tamimi RM, Hankinson SE, Chen WY, Rosner B, Colditz GA: predominantly premenopausal women. Cancer Epidemiol Bio- Combined estrogen and testosterone use and risk of breast markers Prev 2006, 15:967-971. cancer in postmenopausal women. Arch Intern Med 2006, 166: 32. Olson JE, Ingle JN, Ma CX, Pelleymounter LL, Schaid DJ, Pankratz 1483-1489. VS, Vierkant RA, Fredericksen ZS, Wu Y, Couch FJ, Vachon CM, 48. Dimitrakakis C, Jones RA, Liu A, Bondy CA: Breast cancer inci- Sellers TA, Weinshilboum RM: A comprehensive examination dence in postmenopausal women using testosterone in addi- of CYP19 variation and risk of breast cancer using two haplo- tion to usual hormone therapy. Menopause 2004, 11:531-535. type-tagging approaches. Breast Cancer Res Treat 2007, 102: 49. Beral V: Breast cancer and hormone-replacement therapy in 237-247. the Million Women Study. Lancet 2003, 362:419-427.50. 33. Eliassen AH, Missmer SA, Tworoger SS, Hankinson SE: Endoge- Somboonporn W, Davis SR: Testosterone effects on the nous steroid hormone concentrations and risk of breast breast: implications for testosterone therapy for women. cancer: does the association vary by a woman’s predicted Endocr Rev 2004, 25:374-388. breast cancer risk? J Clin Oncol 2006, 24:1823-1830. 51. Hofling M, Hirschberg AL, Skoog L, Tani E, Hagerstrom T, von 34. Key T, Appleby P, Barnes I, Reeves G; Endogenous Hormones Schoultz B: Testosterone inhibits estrogen/progestogen- and Breast Cancer Collaborative Group: Endogenous sex hor- induced breast cell proliferation in postmenopausal women. mones and breast cancer in postmenopausal women: Menopause 2007, 14:183-190. reanalysis of nine prospective studies. J Natl Cancer Inst 52. Conner P: Breast response to menopausal hormone therapy - 2002, 94:606-616. aspects on proliferation, apoptosis and mammographic 35. Key TJ, Appleby PN, Reeves GK, Roddam A, Dorgan JF, Long- density. Ann Med 2007, 39:28-41. cope C, Stanczyk FZ, Stephenson HE Jr, Falk RT, Miller R, 53. Zhou J, Ng S, Adesanya-Famuiya O, Anderson K, Bondy CA: Schatzkin A, Allen DS, Fentiman IS, Key TJ, Wang DY, Dowsett Testosterone inhibits estrogen-induced mammary epithelial M, Thomas HV, Hankinson SE, Toniolo P, Akhmedkhanov A, proliferation and suppresses estrogen receptor expression. Koenig K, Shore RE, Zeleniuch-Jacquotte A, Berrino F, Muti P, Faseb J 2000, 14:1725-1730. Micheli A, Krogh V, Sieri S, Pala V, Venturelli E, et al.: Body mass 54. Micheli A, Meneghini E, Secreto G, Berrino F, Venturelli E, Caval- index, serum sex hormones, and breast cancer risk in post- leri A, Camerini T, Di Mauro MG, Cavadini E, De Palo G, Veronesi menopausal women. J Natl Cancer Inst 2003, 95:1218-1226. U, Formelli F: Plasma testosterone and prognosis of post- 36. Schmitt M, Klinga K, Schnarr B, Morfin R, Mayer D: Dehy- menopausal breast cancer patients. J Clin Oncol 2007, 25: droepiandrosterone stimulates proliferation and gene expres- 2685-2690. sion in MCF-7 cells after conversion to estradiol. Mol Cell 55. Suzuki T, Darnel AD, Akahira JI, Ariga N, Ogawa S, Kaneko C, Endocrinol 2001, 173:1-13. Takeyama J, Moriya T, Sasano H: 5alpha-reductases in human 37. Adly L, Hill D, Sherman ME, Sturgeon SR, Fears T, Mies C, breast carcinoma: possible modulator of in situ androgenic Ziegler RG, Hoover RN, Schairer C: Serum concentrations of actions. J Clin Endocrinol Metab 2001, 86:2250-2257. estrogens, sex hormone-binding globulin, and androgens and risk of breast cancer in postmenopausal women. Int J Cancer 2006, 119:2402-2407. 38. Beattie MS, Costantino JP, Cummings SR, Wickerham DL, Vogel VG, Dowsett M, Folkerd EJ, Willett WC, Wolmark N, Hankinson SE: Endogenous sex hormones, breast cancer risk, and tamoxifen response: an ancillary study in the NSABP Breast Cancer Prevention Trial (P-1). J Natl Cancer Inst 2006, 98:110- 115. 39. Gadducci A, Gargini A, Palla E, Fanucchi A, Genazzani AR: Poly- cystic ovary syndrome and gynecological cancers: is there a link? Gynecol Endocrinol 2005, 20:200-208. 40. Giordano SH, Buzdar AU, Hortobagyi GN: Breast cancer in men. Ann Intern Med 2002, 137:678-687. 41. Bulbrook RD, Thomas BS: Hormones are ambiguous risk factors for breast cancer. Acta Oncol 1989, 28:841-847. 42. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Collaborative Group on Hormonal Factors in Breast Cancer. Lancet 1996, 347:1713-1727. 43. Marchbanks PA, McDonald JA, Wilson HG, Folger SG, Mandel MG, Daling JR, Bernstein L, Malone KE, Ursin G, Strom BL, Norman SA, Wingo PA, Burkman RT, Berlin JA, Simon MS, Spirtas R, Weiss LK: Oral contraceptives and the risk of breast cancer. N Engl J Med 2002, 346:2025-2032. 44. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J; Writing Group for the Women’s Health Initiative Investigators: Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women’s Health Initiative randomized controlled trial. JAMA 2002, 288:321-333. 45. Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, Bonds D, Brunner R, Brzyski R, Caan B, Chlebowski R,

Page 9 of 9 (page number not for citation purposes) Testosterone Testosterone Testolactone 40r Testokctone

35 FIGURE 1. Mean and individual lipoprotein(a) (LpIal) values dur- 30 ing the study. Statistical analyses were performed an log-trans- Lp(4 25 formed values. ‘Si nificant (p mg/dl

with the nonaromatizable androgen stanozolol.4 TABLE II Lipid, Lipoprotein, and Lipoprotein(a) Concentrations During Three Drug Conditions Consequently, it appears that androgens can reduce Lp (a) in both men and women. Treatment Boseline Week 3 The original aim of this study was to examine the Cholesterol (mg/dl) effects of testosterone, with or without aromatization to estradiol, on HDL cholesterol levels.6 The obser- Testosterone 176233 173 % 27 Testolactone 178 _t 29 171 k28 vation that estrogen decreases Lp (a) 3 prompted us Testosterone + testolactone 187225 177+28 to examine the effects of testosterone, testolactone, or the combination on Lp (a) levels. Because supra- LDL Cholesterol (mg/dl) physiologic doses of testosterone increase estrogen

Testosterone 108227 110221 levels, the use of testolactone provided an opportu- Testolactone 108 2 24 105 2 22 nity to examine the androgenic effects of testoster- Testosterone + testolactone 119222 118?24 one with and without its aromatization to estradiol. Testosterone alone increased serum estradiol levels HDL Cholesterol (mg/dl) by 47%, whereas estradiol levels did not change sig- Testosterone SO? 15 42+13* nificantly when testosterone and testolactone were Testolactone 482 15 46115 combined. Lp (a) decreased 37% during testosterone Testosterone + testolactone so-+ 14 40+12*+ and 28% during testosterone and testolactone. These decreaseswere not significantly different, suggesting lipoprotein(a) (mg/dl) that most of the decrease in Lp(a) is mediated by Testosterone 9.4 2 1 1.6 5.9 k 8.7* testosterone and not by its conversion to estradiol. Testoloctone 7.4 -c 8.8 6.9 + 8.9 The decrease in Lp( a) during testosterone and Testosterone + testolactone 7.2 t 9.3 5.2 + 7.9*+ testosterone plus testolactone treatment was greatest

*Significant (p < 0.01) difference from baseline within treatment. in men with the highest initial Lp(a) levels and ‘Significant (p < 0.01) difference from testolactone treatment. largely confined to the 7 men with baseline Lp( a) Cholesterol, LDL, and HDL results have been published previously6 and ore levels >5 mg/dl. These results suggest that the effect included to allow evaluation of effect of testosterone on overall cardiac risk. of testosterone may be most pronounced in subjects HDL = high-density lipoprotein; LDL = low-density lipoprotein. Values are expressed (IS mecm 5 SD. with the highest pretreatment Lp (a) levels. The mechanisms by which testosterone treatment reduced Lp( a) concentrations are not clear. LDL both; Table II). Total cholesterol, LDL cholesterol, cholesterol levels did not change in the present study, and triglyceride levels did not change significantly suggesting that the effect of testosterone on Lp(a) during any of the drug conditions. was independent of changes in LDL metabolism. . . . Lp (a) synthesis, rather than catabolism, is thought This report adds testosterone to the brief list of to be the primary metabolic determinant of Lp( a) interventions known to affect Lp( a) levels. Neither concentrations in humans, l3 and testosterone has re- hydroxymethylglutaryl coenzyme A reductase inhib- cently been documented to reduce ape(a) gene ex- itors, lo bile acid sequestrants, l1 fibrates, l1 diet, l1 or pression in a transgenic mouse model expressing the vigorous exercise l2 appear to be effective in reduc- human ape(a) gene.14Thus, it is possible that tes- ing Lp (a) levels. Recent reports suggest that estro- tosterone lowers Lp (a) levels by decreasing apo (a) gen3 and high-dose niacin’ lower Lp( a) concentra- synthesis. tions. One prior study reported a 65% decrease in Androgens have been assumed to increase ath- Lp (a) in postmenopausal women treated for 6 weeks erosclerotic disease risk by reducing HDL-C.6,15 The

1246 THE AMERICAN JOURNAL OF CARDIOLOGY@ VOL. 77 JUNE I, 1996 present results suggest that the effect of testosterone 6. Zmuda JM, Fahrenbach MC, You&in BT, Bausserman LL, Terry RB, Catlin DH, Thompson PD. The effect of testosterone aromatization on high-density on vascular disease risk factors is more complex, and lipoprotein cholesterol level and postheparin lipolytic activity. Metabolism. that its deleterious effect on HDL cholesterol may 1993;42:446-450. be offset by potentially beneficial effects on Lp(a). 7. Handelsman DJ, Farley TMM, Waives GMH. Contraceptive efficacy of testosterone-induced azoospermia in normal men. Zancet. 1990;336:955-959. Testosterone, like estrogen, l6 may also have favor- 8. Santen RJ. The testis. In: Felig P, Baxter JD, Broadus AE, Frohman LA, eds. able effects on coronary vasomotor function, be- Endocrinology and Metabolism. 2nd ed. New York: McGraw-Hill Book Co; cause exogenous testosterone decreases exercise-in- 1987:821-905. 9. Glantz SA. Primer of Biortatistics. 1 r d ed. New York: McGraw-Hill. Inc; duced ST segment depression in men with angina 1992. pectoris I7 and vasodilates rabbit l8 coronary arteries. 10. Kostner GM, Gavish D, Leopold B, Bolzano K, Weintraub MS, Breslow JL. HMG CoA reductase inhibitors lower LDL cholesterol without reducing Such results may have important implications for the Lp( a) levels. Circulation. 1989;80:1313- 13 19. prolonged use of testosterone in hypogonadism” and 11. Brewer HH. Effectiveness of diet and drugs in the treatment of patients as a male contraceptive,7 and to prevent fraility as with elevated Lp(a) levels. In: Scam AM, ed. Lipoprotein(a). San Diego, Calif: Academic Press, Inc; 1990:21 I-221. men age.20 12. Dufaux B, Order U, Mullet R, Hollmann W. Delayed effects of prolonged exercise on serum lipoproteins. Metabolism 1986;35:105-109. In summary, the results of the present study in- 13. Rader DJ, Cain W, Zech LA, Usher D, Brewer HB. Variation in lipopro- dicate that testosterone reduces Lp (a) concentra- tein(a) concentrations among individuals with the same apolipoprotein (a) iso- tions in normal men primarily by an androgenic form is determined by the rate of Lipoprotein(a) production. J Clin invest 1993;91:443-447. effect and not by its conversion to estradiol. 14. Frazer KA, Narla G, Zhang JL, Rubin EM. The apolipoprotein(a) gene is regulated by sex hormones and acute-phase inducers in YAC transgemc mice. 1. Scanu AM. Lipoprotein(a): a genetic risk factor for premature coronary heart Nature Genetics 1995;9:424-431. disease. JAMA 1992;267:3326-3329. 15. Thompson PD, Cullinane EM, Sady SP, Chenevert C, Saritelli AL, Sady 2. Carlson LA, Hamsten A, Asplund A. Pronounced lowering of serum levels MA, Herbert PN. Contrasting effects of testosterone and stanozolol on serum of lipoprotein Lp(a) in hyperlipidaemic subjects treated with nicotinic acid. J lipoprotein levels. JAMA 1989;261: 1165- 1168. InternMed 1989;226:271-276. 16. Gilligan DM, Quyyumi AA, Cannon RO III. Effects of physiological levels 3. Sacks FM, McPherson R, Walsh BW. Effect of postmenopausal estrogen of estrogen on coronary vasomotor function in postmenopausal women. Cir- replacement on plasma Lp(a) lipoprotein concentrations. Arch Intern Med culation 1994;89:2545-255 1. 1994; 154:1106-l 110. 17. Jaffe MD. Effect of testosterone cypionate on postexercise ST segment 4. Albers JJ, Taggart HM, Applebaum-Bowden D, Hafner S, Chestnut CH, depression. Br Heart J 1977;39: 1217- 1222. Hazard WR. Reduction of lecithin-cholesterol acyltransferase, apolipoprotein 18. Yue P, Chatterjee K, Beale C, Poole-Wilson PA, Collins P. Testosterone D and the Lp(a) lipoprotein with the anabolic steroid stanorolol. Biochim Bio- relaxes rabbit coronary arteries and aorta. Circulation 1995; 9 I : 1 I54- I 160. phys Acta 1984;795:293-296. 19. Bhasin S. Androgen treatment of hypogonadal men. J Clin Endocrinul 5. MacDonald PC, Madden JD, Brenner PF, Wilson JD, Siiteri PK. Origin of M&ah 1992;74:1221-1225. estrogen in normal men and in women with testicular feminization. J Clin En- 20. Bardin CW, Swerdfoff RS, Santen RJ. Androgens: risks and benefits. J Clin docrinol Metah 1979;49:905-916. Endocrinol Metab 1991;73:4-7.

Meta-Analysis of the Use of Low-Dose Beta-Adrenergic Blocking Therapy in Idiopathic or Ischemic Dilated Cardiomyopathy

Dawn G. Zarembski, PharmaD, Paul E. Nolan, Jr., PharmaD, Marion K. Slack, PhD, and Charles Y. Lui, MD

everal compensatory neurohormonal systems are pensatory mechanisms become inadequate to main- S stimulated in congestive heart failure (CHF) in tain ventricular function.’ In addition, the increase in an attempt to maintain cardiac function. The sym- plasma norepinephrine levels observed in CHF can pathetic nervous system is activated with resultant be directly correlated to severity and mortality of increases in circulating plasma norepinephrine levels CHF.4 Contemporary treatment of CHF is focused that produce tachycardia, vasoconstriction, and in- on altering these neurohormonal systems. Therapy creased force of contraction.1,2 The renin-angioten- aimed at attenuating excess sympathetic nervous sys- sin-aldosterone system is also activated, leading to tem activity through P-adrenergic blocking agents sodium and water retention and further increases in has been proposed in a number of small clinical tri- vasoconstriction secondary to angiotensin II synthe- als.5-23 Given the small sample size in each of the sis.3 Eventually, as the disease progresses these com- trials, the ability of these trials to impact medical practice has been limited. To optimize the available data, we conducted a meta-analysis designed to as- From the Department of Pharmacy Practice, Chicago College of Phar- sess the ability of p blockers to improve quality of macy, the Department of Pharmacy Practice 2nd Scrence, College of Pharmacy, Section of Cardiology, Department of Internal Medicine, Col- life and hemodynamic indexes in patients with idio- lege of Medicine, and University Heart Center, The University of Arizona, pathic or ischemic cardiomyopathy. Tucson. Dr. Zarembskr was sponsored by a grant from the American . . . Socrety of Hospital Pharmacists Research and Education Foundation. Dr. Prospective, randomized, placebo-controlled trials Nolan’s address is: Department of Pharmacy Practice 2nd Science, The University of Arizona, College of Pharmacy, 1703 E. Mabel, Tucson, were gathered from a review of the reports. Current Arizona 8572 1. Manuscript received July 10, 199.5; revised manuscript Contents: Clinical Practice was reviewed and MED- received and accepted December 20, 1995. LINE files from 1960 to November 1995 were searched.

aromatization to estradiol does not affect tHcy concentrations in eugonadal men. The observation that fasting tHcy concentrations are higher in men than in women [3] has led to the suggestion that sex steroid hormones may explain the gender difference in tHcy levels [3]. Support for this idea comes from studies showing that estrogen plus progestin replacement therapy [4] and the estrogen agonist tamoxifen [5] reduce tHcy concentrations in postmenopausal women. Others have documented an inverse correlation between serum estradiol concentrations and post-methionine tHcy levels in premenopausal women [4]. The present study is the first, to our knowledge, to examine the effects of androgenic hormones on tHcy levels in men. Serum testosterone and estradiol levels were not strongly associated with pretreatment tHcy levels. Fur- thermore, testosterone administration, with or without its aromatization to estradiol, did not significantly alter tHcy levels. These results suggest that neither testosterone nor estradiol have an important influence on tHcy levels in normal young men, and raise the possibility that factors other than testosterone contribute to Fig. 1. Mean (9S.E.) serum testosterone (upper panel) and estradiol higher fasting tHcy levels in men than in women. (lower panel) levels before and 3 weeks after treatment with testos- A more likely explanation for the gender difference terone, testolactone, and testosterone plus testolactone. * Significant in fasting tHcy levels may relate to the fact that cre- difference (PB0.01) from pretreatment levels. † Significant difference ‡ atine–creatinine production is directly coupled to s- (PB0.01) from testosterone treatment. Significant difference (PB 0.01) from testolactone treatment. adenosylhomocysteine generation from s-adenosylmethionine [13], and that lean body mass centrations were not strongly associated with tHcy lev- and creatine–creatinine production tend to be higher in els. Similar correlation coefficients were observed after men than in women. Indeed, plasma tHcy levels corre- controlling for baseline age and body weight. These late directly with serum creatinine concentrations in results indicate that testosterone, with or without its men and women [14]. Moreover, the gender difference aromatization to estradiol, does not influence fasting in tHcy levels disappeared in one recent study when tHcy levels in men. men and women were matched for serum creatinine concentrations [14]. These results suggest that the higher mean fasting tHcy levels in men compared with women are most likely a result of the direct relationship 4. Discussion between homocysteine production and creatine–creatinine synthesis [14]. The original aim of this study was to examine the The present study has several limitations. First, a effects of testosterone, with or without aromatization to longer treatment duration and larger sample size may estradiol, on high-density lipoprotein cholesterol levels be required to reveal a testosterone effect on mean in men [7]. Recent reports suggesting that estrogens fasting tHcy levels. Two previous reports demonstrat- may decrease the atherogenic amino-acid tHcy in post- ing that female sex hormones decrease plasma tHcy menopausal women [4,5] prompted us to examine the levels in women included at least 27 study subjects [4,5], effects of testosterone, testolactone or the combination and a significant decline in tHcy in one study was not on fasting tHcy levels. Because supraphysiologic doses detectable until after 3–4 months of treatment [5]. of testosterone increase estrogen levels in men, the use Therefore, the present study may have underestimated of testolactone provided an opportunity to examine the the true magnitude of testosterone’s effect on tHcy androgenic effects of testosterone with and without its levels. Since fasting tHcy levels are only weakly related aromatization to estradiol. Testosterone alone increased to the tHcy response to methionine loading [15], we serum estradiol levels by 43%, whereas estradiol levels also cannot exclude the possibility that testosterone did not change significantly when testosterone and administration influences post-methionine thcy levels. testolactone were combined. Fasting tHcy levels were Finally, all of our study subjects had normal total not significantly altered during either testosterone con- testosterone levels, so it is possible that men with dition, however, suggesting that testosterone and its hypogonadism would have experienced a change in 202 J.M. Zmuda et al. / Atherosclerosis 130 (1997) 199–202

Table 2 Plasma homocysteine concentrations before and after 3 weeks of testosterone, testolactone, and testosterone plus testolactone

Testosterone Testolactone Testosterone+Testolactone

Homocysteine (vmol/l) Before 4.791.1 4.690.8 4.891.1 After 4.690.9 4.991.1 4.890.9

Values are mean9S.D. There were no significant differences within or between treatments. thcy with testosterone treatment. Additional studies are menopausal women with breast cancer treated with tamoxifen. Int needed to examine these possibilities. J Cancer 1995;60:365. [6] MacDonald PC, Madden JD, Brenner PF, Wilson JD, Siiteri PK. Origin of estrogen in normal men and in women with testicular feminization. J Clin Endocrinol Metab 1979;49:905. Acknowledgements [7] Zmuda JM, Fahrenbach MC, Younkin BT, Bausserman LL, Terry RB, Catlin DH, Thompson PD. The effect of testosterone This study was supported by the National Institutes aromatization on high-density lipoprotein cholesterol level and postheparin lipolytic activity. Metabolism 1993;42:446. of Health grant HL-28467 and gifts from the Miriam [8] Jackson AS, Pollock ML. Generalized equations for predicting Foundation, the Haire family, William Jakober, and body density of men. Brit J Nutr 1978;40:497. the McNulty family. [9] Handelsman DJ, Farley TMM, Waites GMH. Contraceptive efficacy of testosterone-induced azoospermia in normal men. Lancet 1990;336:955. [10] Vester B, Rasmussen K. High performance liquid chromatogra- References phy method for rapid and accurate determination of homocysteine in plasma and serum. Eur J Clin Chem Clin Biochem 1991;29:549. [1] McCully KS. Homocysteine and vascular disease. Nature Med [11] Glanz SA. Primer of Biostatistics, 3rd edn. New York: McGraw- 1996;2:386. Hill, 1992. [2] Boushey CJ, Beresford SAA. A quantitative assessment of plasma [12] Ueland PM, Refsum H, Stabler SP, Malinow MR, Andersson A, homocysteine as a risk factor for vascular disease. J Am Med Allen RH. Total homocysteine in plasma or serum: methods and Assoc 1995;274:1049. clinical applications. Clin Chem 1993;39:1764. [3] Nyga˚rd O, Vollset SE, Refsum H, Stensvold I, Aage T, Nor- [13] Mudd SH, Poole JR. Labile methyl balances for normal humans drehaug JE, Ueland PM, Kva˚le G. Total plasma homocysteine on various dietary regimens. Metabolism 1975;24:721. and cardiovascular risk profile. The Hordaland Homocysteine [14] Brattstrom L, Lindgren A, lsraelsson B, Andersson A, Hultberg Study. J Am Med Assoc 1995;274:1526. B. Homocysteine and cysteine: determinants of plasma levels in [4] Van Der Mooren MJ, Wouters MGAJ, Blom HJ, Schellekens LA, middle-aged and elderly subjects. J Int Med 1994;236:633. Eskes TKAB, Rolland R. Hormone replacement therapy may [15] Bostom AG, Jacques PF, Nadeau MR, Williams RR, Ellison RC, reduce high serum homocysteine in postmenopausal women. Eur Selhub J. Postmethionine load hyperhomocysteinemia in J Clin Invest 1994;24:733. persons with normal fasting total plasma homocysteine: initial [5] Anker G, Lonning PE, Ueland PM, Refsum H, Lien EA. Plasma results from The NHLBI Family Heart Study. Atherosclerosis levels of the atherogenic amino acid homocysteine in post- 1995;116:147.