HT & Antidepressant Response

Released March 28, 2013

This e-newsletter presents reviews of important, recently published scientific articles selected by The North American Menopause Society (NAMS), the leading nonprofit scientific organization dedicated to improving women’s health and quality of life through an understanding of menopause and healthy aging. Each review has a commentary from a recognized expert that addresses the clinical relevance of the item. Oversight for this e- newsletter issue was by Chrisandra L. Shufelt, MD, MS, NCMP, Chair-elect of the 2012-2013 NAMS Profes- sional Education Committee. Opinions expressed in the commentaries are those of the authors and are not neces- sarily endorsed by NAMS or Dr. Shufelt.

HT & antidepressant response groups with respect to demographic and clinical characteristics (eg, postmenopausal women Kornstein SG, Toups M, Rush J, et al. Do menopausal were older, more likely to be divorced or wi- status and use of hormone therapy affect antidepressant dowed, less educated, and less likely to be em- treatment response? Findings from the Sequenced Treat- ment Alternatives to Relieve Depression (STAR*D) ployed; postmenopausal women also had overall study. J Womens Health. 2013;22(2):121-131. Level of longer duration of illness and greater likelihood evidence: II-2. of chronicity than premenopausal women). Notwithstanding, no significant differences in Summary. There is little doubt that women are treatment efficacy (response, remission rates) at higher lifetime risk than men to develop de- were found when menopausal staging and hor- pression.1 Overall, the risk for a major depres- mone use (either oral contraceptives or HT) sive disorder is 1.7 times higher among women were considered. than men. Because this difference appears to emerge after puberty and declines during the Comment. Like in other stages of life, depres- postmenopausal years, it has been postulated sion in midlife women is a complex, multifa- that gonadal hormones might play a role in ceted phenomenon influenced by numerous fac- women’s vulnerability to depression and their tors that include comorbid medical and psychia- therapeutic response to antidepressants. tric conditions, sleep changes, age, ethnicity, body mass index, cigarette smoking, and stress- Kornstein et al utilized data from the Sequenced ful life events, just to name a few.2 Kornstein et Treatment Alternatives to Relieve Depression al were interested in knowing whether treatment (STAR*D) study to examine whether meno- outcomes with an SSRI would differ based on pausal status and use of hormonal contracep- menopausal status or the use of hormonal con- tives or hormone therapy (HT) would have an traceptives or HT. This is a clinically relevant impact on treatment efficacy among women re- question, particularly in light of previous studies ceiving the selective serotonin reuptake inhibi- that had indicated a superior response to SSRIs tor (SSRI) citalopram. Premenopausal (N = 896) among younger versus older women and among and postmenopausal (N = 544) women were HT users versus nonusers.3-5 Moreover, accu- treated with citalopram for 12 to 14 weeks and mulated data suggested the existence of a “win- had their treatment response and remission as- dow of vulnerability” for the emergence of de- sessed with standardized instruments. There pression among peri- and early postmenopausal were significant differences between these two women6 and indicated a more robust antidepres- 2 sant response to estrogen-based therapies in pe- 6. Cohen LS, Soares CN, Vitonis AF, Otto MW, Harlow rimenopausal versus postmenopausal women.7,8 BL. Risk for new onset of depression during the menopausal transition: the Harvard study of moods and cycles. Such observations, along with the current know- Arch Gen Psychiatry. 2006;63(4):385-390. ledge regarding the effects of estrogen on mo- 7. Soares CN, Almeida OP, Joffe H, Cohen LS. Efficacy 9 noamines that regulate mood and behavior, of estradiol for the treatment of depressive disorders in support the authors’ hypothesis (and the aim of perimenopausal women: a double-blind, randomized, pla- previous investigations)10 that antidepressant cebo-controlled trial. Arch Gen Psychiatry. 2001;58(6): 529-534. response could in fact differ significantly based 8. Morrison MF, Kallan MJ, Ten Have T, Katz I, Tweedy on menopausal staging or hormone use. K, Battistini M. Lack of efficacy of estradiol for depression in postmenopausal women: a randomized, controlled Ideally, the management of midlife women with trial. Biol Psychiatry. 2004;55(4):406-412. depression should require a careful, tailored ap- 9. Lokuge S, Frey BN, Foster JA, Soares CN, Steiner M. Depression in women: windows of vulnerability and new proach; it should take into consideration the insights into the link between estrogen and serotonin. J presence and severity of menopause-associated Clin Psychiatry. 2011;72(11):e1563-1569. issues (sleep disturbances, vasomotor symp- 10. Soares CN, Thase ME, Clayton A, et al. Desvenlafax- toms, pain) as well as the tolerability and risks ine and escitalopram for the treatment of postmenopausal of various hormonal and nonhormonal strategies women with major depressive disorder. Menopause. 2010;17(4):700-711. for bone and cardiovascular health, cognition, and mood functioning. This present study CRP & fracture risk serves, however, as a reassurance that even in the absence of such systematic approach, satis- Ishii S, Cauley JA, Greendale GA, et al. C-reactive pro- factory results might be obtained with antide- tein, bone strength, and 9-year fracture risk: data from pressant agents that are commonly prescribed The Study of Women’s Health Across the Nation and generally well-tolerated. (SWAN). J Bone Miner Res. 2013 Mar 2. [Epub ahead of print] Level of evidence: II-2. Claudio N. Soares, MD, PhD, FRCPC Visiting Researcher Summary. Researchers analyzed data from a Former Professor of Psychiatry and multisite, multiethnic prospective cohort of Obstetrics & Gynecology 1,872 community-dwelling women, pre- or ear- McMaster University Ontario, Canada ly perimenopausal at baseline. Over a median follow-up of 9 years, incident nondigital and References: noncraniofacial fractures were recorded yearly. 1. Kessler RC, McGonagle KA, Swartz M, Blazer DG, They examined whether composite indices of Nelson CB. Sex and depression in the National Comor- femoral neck strength relative to load (which are bidity Survey. I: Lifetime prevalence, chronicity and re- currence. J Affect Disord. 1993;29(2-3):85-96. inversely associated with fracture risk) are also 2. Soares CN. Can depression be a menopause-associated inversely associated with C-reactive protein risk? BMC Med. 2010;8:79. (CRP) and whether that explains part of the as- 3. Thase ME, Entsuah R, Cantillon M, Kornstein SG. sociation between CRP and fracture risk. Relative antidepressant efficacy of venlafaxine and SSRIs: sex-age interactions. J Womens Health (Larchmt). 2005;14(7):609-616. In analyses adjusted for age, race/ethnicity, di- 4. Pinto-Meza A, Usall J, Serrano-Blanco A, Suarez D, abetes, menopause transition stage, body mass Haro JM. Gender differences in response to antidepres- index, smoking, alcohol use, physical activity, sant treatment prescribed in primary care. Does meno- medications, prior fracture, and study site, CRP pause make a difference? J Affect Disord. 2006;93(1- was associated inversely with each composite 3):53-60. 5. Schneider LS, Small GW, Clary CM. Estrogen re- strength index but not associated with femoral placement therapy and antidepressant response to sertra- neck or lumbar spine bone mineral density line in older depressed women. Am J Geriatr Psychiatry. (BMD). A total of 194 women had fractures 2001;9(4):393-399. during the follow-up. Fracture hazard increased linearly with log(CRP) for CRP levels ≥3 mg/L

3 in Cox proportional hazards analyses. Addition Where CRP levels fit in the spectrum of risk of any of the composite strength indices did factors at this point in time is unclear. Inflam- weaken the association and made it statistically mation may be an important part of the risk of nonsignificant. Researchers concluded that, secondary osteoporosis in some individuals. The above the threshold of 3 mg/L, fracture risk in- study also points out the differential suscepti- creases with increasing CRP. Composite bility of cortical versus trabecular bone to in- strength indices are inversely related to CRP flammation, with lumbar spine BMD explaining levels, and this partly explains the increased more of the CRP fracture association than did fracture risk with inflammation. femoral neck BMD. Lumbar spine, which con- tains trabecular bone and is highly vascular, Comment. Chronic inflammation is associated may be more susceptible to inflammation- with multiple conditions including cardio- induced bone fragility. vascular disease, diabetes, and dementia. Wom- en with osteoporosis are also at risk for cardi- Stuart L. Silverman, MD, FACP, FACR ovascular disease. It is therefore of interest to Medical Director Bone Center of Excellence study whether fracture risk is associated with Cedars-Sinai Medical Center inflammation. Elevated levels of an inflamma- Los Angeles, CA tory marker, CRP, have been associated with increased fracture risk. The mechanism of this Personal histories of association has been unclear because there have been conflicting results on the association of atypical femur fractures CRP and BMD. Schneider JP, Hinshaw WB, Su C, Solow P. Atypical femur fractures: 81 individual personal histories. J Clin Ishii et al studied 1,872 peri-menopausal com- Endocrinol Metab. 2012;97(12):4324-4328. Level of evi- munity-dwelling women (aged 42-53 y) from dence: III. the SWAN study with at least one menstrual period in the prior 3 months and tested the hy- Summary. A voluntary, anonymous online sur- pothesis that the association of CRP with in- vey documented the fracture history of 81 per- creased fracture risk is mediated by changes in sons nationwide. The mean duration of bisphos- bone strength. Using surrogate markers of fe- phonate treatment in patients was 9.5 years, and moral neck composite strength derived from prevention was the initial indication in 68% of bone density studies, they found a modest in- patients. Ninety-four percent started on alendro- verse association (0.035-0.041 SD in bone nate, 77% reported prodromal pain, 16% of strength per doubling of CRP). Fracture hazard these were diagnosed with incident stress frac- increased with CRP above a given threshold of tures, and 39.5% experienced a contralateral 3 mg/L, with no influence of BMD. The authors atypical femur fracture (AFF). In the 71 patients concluded that composite measures of bone with AFF, 38% reported delayed healing, 11% strength are related to CRP and suggested that had a complete contralateral AFF, and 22% un- inflammation plays a role in the risk of fragility derwent prophylactic rodding for a contralateral fracture. stress AFF. In patients with complete AFFs, 44% were continued on a bisphosphonate after Why are these findings important? They imply the fracture, and 35% underwent a metatarsal that inflammation plays a role in bone fragility fracture. Patients with AFF experienced delayed as reflected by composite bone strength indices. healing, prodromal pain, and persistant risk of a The study needs to be replicated in other popu- contralateral and/or other fracture. In patients on lations of different ages and using other tech- long-term bisphosphonates, evaluating femur niques of measuring bone strength, such as fi- pain may facilitate early diagnosis of stress nite element analysis. AFFs, reducing fracture risk.

Comment. Important clinical lessons can be References: derived from this review of AFF cases asso- 1. Whitaker M, Guo J, Kehoe T, Benson G. Bisphospho- nates for osteoporosis—where do we go from here? N ciated with bisphosphonate therapy. Almost all Engl J Med. 2012;366(22):2048-2051. of these cases were seen before the recent FDA 2. The North American Menopause Society. 2010 posi- guidance regarding the duration of bisphospho- tion statement of The North American Menopause Socie- nate therapy, as follows:1 ty. Management of osteoporosis in postmenopausal women. Menopause. 2010;17(1):25-54. 3. National Osteoporosis Foundation. Clinician's Guide to 1. The treatment with bisphosphonates should Prevention and Treatment of Osteoporosis 2013. Availa- be reserved for patients at high risk of fracture ble at: http://www.nof.org/hcp/practice/practice-and- and in whom the benefit-risk ratio is very favor- clinical-guidelines/clinicians-guide. Accessed March 22, able, in accord with The North American Me- 2013. nopause Society2 and National Osteoporosis 4. McClung MR, Harris ST, Iller PD, et al. Bisphospho- 3 nate therapy for osteoporosis: benefits, risks, and drug Foundation guidelines. Limit therapy to 3 years holiday. Am J Med. 2013;126(1):13-20. (or one dose of zoledronic acid) if used to prevent bone loss in women in early menopause or Menopause in BRCA1/2 carriers in those discontinuing estrogen therapy.

2. Review need for continued bisphosphonate Lin WT, Beattie M, Chen LM, et al. Comparison of age therapy after 3 to 5 years. Continue treatment in at natural menopause in BRCA1/2 mutation carriers patients at high risk—those with previous verte- with a non–clinic-based sample of women in northern bral fracture or those whose bone density is still California. Cancer. 2012 Jan 29. [Epub ahead of print] very low. Discontinue therapy in patients not at Level of evidence: II-3. high risk.1,4 3. Counsel patients on therapy for more than Summary. This study examined whether 3 years to report new thigh pain. If stress frac- BRCA1/2 mutation carriers have an earlier on- tures or reactions are identified, discontinue bis- set of natural menopause than unaffected wom- phosphonates and institute medical and surgical en. The subject included 382 white carriers of therapy to prevent the occurrence of the com- the BRCA1/2 gene identified within the Breast plete fracture. Cancer Risk Program Registry at the University 4. If patients on long-term bisphosphonates ex- of California at San Francisco and 765 other perience a stress fracture of other bones, per- white women in northern California. Median form imaging studies to evaluate the presence of age at natural menopause in BRCA1/2 carriers femoral stress reactions. was significantly younger than in the unaffected sample (50 vs 53 y). In the future, we will likely be able to identify by genetic or metabolic testing a subset of pa- The unadjusted hazard ratio for natural meno- tients who are at particular risk for developing pause was 4.06 (95% CI, 3.03-5.45); after ad- atypical femoral fracture while receiving potent justing for smoking, parity, and oral contracep- antiresorptive drugs. Until then, our improved tive use, it was 3.98 (95% CI, 2.87-5.53). For understanding of the epidemiology and clinical BRCA1/2 carriers who were current heavy features of these atypical fractures allows us to smokers (≥20 cigarettes/d), median age at natu- identify most patients before they experience a ral menopause was 46 years; for nonsmokers, it complete fracture and, in that manner, limit the was 49 years. BRCA1/2 mutation was asso- risk of long-term bisphosphonate therapy. ciated with a significantly earlier age of natural menopause, and heavy smoking increased the Michael R. McClung, MD, FACP risk, suggesting a risk of earlier infertility in Director, Oregon Osteoporosis Center BRCA1/2 carriers. Portland, OR Comment. The clinical definition of menopause is cessation of cyclic bleeding for 1 year. Al-

5 though there is marked variability in the onset of New oral medication for menopause, the median age continues to be 51.4 vulvar & vaginal atrophy years. Lin et al examine a cellular abnormality that causes the loss of viable follicles by DNA US Food and Drug Administration. FDA approves Os- damage in germ cells. Normally, the further ac- phena for postmenopausal women experiencing pain dur- tion of germ cells may be quiescence, recruit- ing sex. [press release]. Feb 26, 2013. ment for development and ovulation, or apopto- www.fda.gov/NewsEvents/Newsroom/PressAnnounceme sis. Without regeneration in the ovary, the fol- nts/ucm341128.htm. Accessed March 28, 2013. licles are depleted, which leads to a perimeno- Summary. The FDA has approved ospemifene pausal state. This decline may result in infertili- (Osphena 60-mg tablets), an oral tissue-selective ty and irregularity. estrogen agonist/antagonist, for managing dys-

pareunia caused by vulvar and vaginal atrophy In this carefully documented study, DNA muta- in menopausal women. Ospemifene acts as an tions, repair genes, and BRCA1/2 are used as estrogen agonist in vaginal tissues. In clinical markers. Evidence from many studies indicates trials, ospemifene alleviated pain with sexual DNA repair genes increase the risk of primary intercourse and increased vaginal mucosal ma- ovarian insufficiency. Also, it is assumed that if turation more than placebo. oocytes are more prone to DNA damage and experience accelerated follicular depletion, this Contraindications include estrogen-dependent would lead to a higher incidence of infertility neoplasia and active or prior venous thrombo- and early menopause. The white BRCA1/2 Cali- embolism, stroke, or myocardial infarction. Al- fornia women had definite spontaneous (natural) though ospemifene also acts on the endome- menopause onset 3 to 4 years earlier than the trium as an estrogen agonist, no cases of endo- other women who were not carriers. Women metrial cancer have been noted in clinical trials. who smoked had a much earlier onset as well. The most commonly reported adverse reactions The authors suggested that the earlier meno- were hot flashes (8%), vaginal discharge (4%), pause in women with the BRCA1/2 mutation and muscle spasms (3%). The approval includes may be because of an even narrower reproduc- a boxed warning about endometrial cancer and tive window, as shown in some statistics for in- cardiovascular disorders. fertility and miscarriage.

Comment. Symptomatic vulvar and vaginal Based on these data, healthcare providers con- atrophy is common but undertreated; thus, new sulting these women on fertility may recom- treatment options are welcome. Ospemifene mend earlier childbearing. The statistics are may be particularly appealing to women who sound, and the results are documented and ex- prefer not to use vaginal cream, tablets, or the plained. Perhaps an important aspect is this vaginal ring. However, unlike vaginal estrogen study’s design using a marker system to trans- therapy, ospemifine is associated with hot flash- late the presence of a common problem in wom- es and may (like its cousins tamoxifen and ra- en. With increased longevity, the trend for late loxifene) raise risk for venous thrombo- pregnancies must be questioned in light of these embolism. As with vaginal estrogen, ospemi- results. Gestation should be reckoned with in fene’s package labeling does not explicitly rec- any case where early perimenopause is a poten- ommend concurrent progestin therapy to pre- tial medical hazard. vent endometrial neoplasia in women with an

Bernard A. Eskin, MS, MD intact uterus. However, endometrial monitoring Professor of Obstetrics and Gynecology should be considered in long-term users, and Drexel University College of Medicine any vaginal bleeding should be evaluated. Be- Philadelphia, PA cause ospemifene has not been adequately studied in women with breast cancer, the FDA re-

6 commends that, as with vaginal estrogen, ospe- Robert A. Wild, MD, MPH, PhD; Chunyuan Wu, MS; mifene “should not be used in women with J. D. Curb, MD, MPH; Lisa W. Martin, MD; Lawrence Phillips, MD; Marcia Stefanick, PhD; known or suspected breast cancer or with a his- Maurizio Trevisan, MD; and JoAnn E. Manson, MD, tory of breast cancer.” DrPH. ♦ Andrew M. Kaunitz, MD Relapse of vasomotor symptoms after discontinuation Professor and Associate Chairman of the selective serotonin reuptake inhibitor escitalo- Department of Obstetrics and Gynecology pram: results from the Menopause Strategies: Finding University of Florida College of Medicine-Jacksonville Lasting Answers for Symptoms and Health Research Jacksonville, FL Network. Hadine Joffe, MD, MSc; Katherine A. Guthrie, PhD; Published in Journal Watch Women's Health at Joseph Larson, MS; Lee S. Cohen, MD; http://womens-health.jwatch.org/ March 21, 2013. Janet S. Carpenter, PhD, RN, FAAN; Andrea Z. LaCroix, PhD; and Ellen W. Freeman, PhD. ♦ Menopause Editor’s picks Racial differences in perception of healthy body from March 2013 weight in midlife women: results from the Do Stage Transitions Result in Detectable Effects study. Semara Thomas, MD; Roberta B. Ness, MD, MPH; NAMS spotlights selections from the most Rebecca C. Thurston, PhD; Karen Matthews, PhD; recent issue of the Society’s official journal, Chung-Chou Chang, PhD; and Rachel Hess, MD, MS. Menopause, chosen by its Editor-in-Chief, ♦ Isaac Schiff, MD. Effects of bazedoxifene alone and with conjugated equine estrogens on coronary and peripheral artery atherosclerosis in postmenopausal monkeys. Coronary heart disease events in the Women’s Health Thomas B. Clarkson, DVM; Kelly F. Ethun, DVM, PhD; Initiative hormone trials: effect modification by meta- Haiying Chen, MD, PhD; Debbie Golden, BS; bolic syndrome: a nested case-control study within the Edison Floyd, BS; and Susan E. Appt, DVM. Women’s Health Initiative randomized clinical trials.

First to Know® is a registered trademark of The North American Menopause Society Copyright © 2013 The North American Menopause Society All rights reserved 5900 Landerbrook Drive, Suite 390 • Mayfield Heights, OH 44124 • USA Tel 440/442-7550 • Fax 440/442-2660 • [email protected] • www.menopause.org

The level of evidence indicated for each study is based on a grading system that evaluates the scientific rigor of the study design, as developed by the US Preven- tive Services Task Force. A synopsis of the levels is presented below.

Level I Properly randomized, controlled trial. Level II-1 Well-designed controlled trial but without randomization. Level II-2 Well-designed cohort or case-control analytic study. Level II-3 Multiple time series with or without the intervention (eg, cross-sectional and uncontrolled investigational studies). Level III Meta-analyses; reports from expert committees; descriptive studies and case reports.

Accurate, Concise Summaries About Menopause MenoNotes

Six free handouts summarize some of the most confusing menopause-related topics. Hot flashes, vaginal dryness, bioidentical hormone therapy, and a menstrual calendar (in English, French, and Spanish) are clearly explained with the most up-to-date information. More topics will be added regularly so check back often.

Free to download on the NAMS website at: www.menopause.org/publications/publications-for- women/menonotes