HIC# 0609001839
YALE UNIVERSITY SCHOOL OF MEDICINE YALE NEW HAVEN HOSPITAL HUMAN INVESTIGATION COMMITTEE
Application to Involve Human Subjects in Research
Title of Research Project: Symptom Onset Antidepressant Treatment for PMDD Principal Investigator: Kimberly A. Yonkers, MD Campus Address: 142 Temple Street, Suite 301, New Haven, CT 06510 Campus Phone: 764.6621 Fax: 764.6766 Pager: E-mail: [email protected] Protocol Correspondent Name & Address: Joanne Cunningham, PhD, 142 Temple Street, Suite 301, New Haven, CT 06510 Campus Phone: 764.5719 Fax: 764.6766 E-mail: [email protected]
SECTION I: PRINCIPAL INVESTIGATOR/FACULTY ADVISOR AGREEMENT
As the Principal Investigator or Faculty Advisor of this research project, I certify the following:
. The information provided in this application is complete and accurate. . That I assume full responsibility for the protection of human subjects and the proper conduct of the research. . That subject safety will be of paramount concern, and every effort will be made to protect subjects’ rights and welfare. . That the research will be performed according to ethical principles and in compliance will all federal, state and local laws, as well as institutional regulations and policies regarding the protection of human subjects. . That all members of the research team will be kept apprised of research goals. . That I will obtain approval for this research study and any subsequent revisions prior to initiation. . That I will report to the HIC any serious injuries or other unanticipated problems involving risk to participants.
9/3/09 ______Signature Date
SECTION II: FUNDING, TRAINING AND PROTOCOL-RELATED CONFLICT OF INTEREST
1. Funding Source: Please describe the funding source(s) for this study. Check all boxes that apply:
External* Department Name Investigator Initiated: ______Other: ______
*For externally funded studies, please supply the following information:
PI of Contract or Grant: Kimberly A. Yonkers Funding Source: National Institute of Health Contract or Grant Title: Symptom-Onset Antidepressant Treatment for PMDD Contract or Grant #: TBA - NA
If using more than one funding source for this study, list all funding sources on an attached sheet. For grants: attach those sections of your grant application/agreement that pertain to the technical and human subject’s portion of this protocol.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 1
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
2. Human Subject Protection Training: All investigators and key study personnel (persons involved in the design and/or conduct of research involving human subjects) are required to complete human subject protection training (HSPT). This training requirement can be met through the Yale web-based program at http://info.med.yale.edu/irbtraining or the NIH program at http://cme.cancer.gov/c01. Please note that investigators who have not completed this training requirement cannot participate in study activities until this training is completed.
3. Conflict of Interest Statement: All investigators and key study personnel (those persons involved in the design and/or conduct of the research involving human subjects) are required to read a copy of the Yale Human Investigation Committee Policy on Protocol-Related Conflict of Interest (“HIC COI Policy” – see http://info.med.yale.edu/hic/). Please note that the HIC COI Policy addresses protocol-related conflict of interest, and is distinct from the annual disclosure required by the Yale University Policy on Conflict of Interest and Conflict of Commitment.
All investigators and key study personnel are required to sign their name in the space provided below. Those who have answered “no” to all screening questions asked in the HIC COI Policy should indicate below that no Protocol-Related COI exists. Those who answered “yes” to any question in the HIC COI Policy should download a copy of the Protocol-Related Conflict of Interest Disclosure Form, which must be submitted to the HIC along with this Application.
Name Protocol-Related COI? HSPT Completed?
Principal Investigator Kimberly Yonkers, MD Yes No Yes No Co-Investigator Margaret Altemus, MD Yes No Yes No Co-Investigator Susan Kornstein, MD Yes No Yes No Study Personnel Ralitza Gueorguieva, PhD Yes No Yes No Study Personnel Kari VanSteenburgh Yes No Yes No Study Personnel Nathan Gotman Yes No Yes No Study Personnel Ariadna Forray, MD Yes No Yes No Study Personnel Geraldine Hawthorne-Jones Yes No Yes No Study Personnel Wendy Brunetto Yes No Yes No Study Personnel Sarah Saiano Yes No Yes No Study Personnel Amanda Pusey, MD Yes No Yes No Study Personnel Amanda Greco Yes No Yes No Study Personnel Cristine Hine Yes No Yes No
*** My signature here indicates that I have read and am in compliance and will continue to be with the HIC’s Protocol-Specific Conflict of Interest Policy.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 2
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
4. Department Chair’s Conflict of Interest Statement (to be completed by the Chair of each department with which the Principal Investigator and co-investigators are affiliated and/or which the research affects): Do you know of any real or apparent institutional conflict of interest (e.g., Yale University ownership of a sponsoring company) that might compromise this research? Yes No
______Signature of Chair*** Department
***My signature here indicates that I have read and am in compliance with the HIC’s Protocol-Related Conflict of Interest Policy. I further agree to submit a Protocol-Related Conflict of Interest Disclosure Form if I am aware of any real or apparent institutional conflict of interest.
APPROVED FOR SUBMISSION TO HIC:
______Signature of Primary Reviewer Please Print Name of Primary Reviewer Date
For HIC Use Only
______Date Approved Human Investigation Committee
Protocol is valid until: ______
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 3
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
SECTION III: GENERAL INFORMATION
1. Choose all that apply: (* See indicated section in HIC Guidelines for Investigators) Children/minors* (Section E.1) Pregnant women/fetuses/placenta Decisionally impaired* (Section E.2) Prisoners Females of childbearing potential Non-English Speaking Radioactive Materials Use of Employees IND # ______Students IDE # ______A or B (Section C.1)
2. Location of study: Please identify the hospital, in-patient or outpatient facility, school or other agency that will serve as the location of the research.
Yale University Yale-New Haven Hospital APT Foundation, Inc. West Haven VA Community Consultation Board, Inc. Haskins Laboratories John B. Pierce Laboratory, Inc. Connecticut Mental Health Center Hill Health Corporation Comprehensive Cancer Center General Clinical Research Center(s) MR-TAC Other locations, Specify: ______
Please indicate the location(s) within the hospital and/or Medical School where the research will take place:
1. Yale University, School of Medicine 2. Cornell University, Weill Medical College 3. Virginia Commonwealth University
Please note: when other institutions are engaged in the research, it may be necessary to secure the approval of their Institutional Review Boards (IRB) and/or to insure that the institution has obtained a Federal Wide Assurance (FWA). Institutions may not list the Yale HIC as their IRB of record unless the Federal government has approved their FWA and they have in place a fully executed IRB Authorization Agreement between their institution and Yale University.
3. Probable Duration of Project: Please state the expected duration of the project, including all follow-up and data analysis activities. [Please answer this and all other questions on the form with font size 12.]
The approximate duration of subject participation will be approximately 11 months. The total enrollment period will be approximately 36 months. The total study duration will be approximately 60 months or five year total.
4. Number of Subjects: Please state the number of subjects to be enrolled at Yale. For multi-center studies, indicate the total number of subjects to be enrolled across all sites. If different subject populations will participate, state the anticipated number in each group.
300 women will be randomized across three recruitment sites; Yale University will randomize 100 subjects.
SECTION IV: RESEARCH PLAN
1. Statement of Purpose: What are the scientific aims of the study, or the hypotheses to be tested?
The primary aims of this study is to assess whether the SRI, sertraline, is more effective than placebo at ameliorating symptoms of PMDD when capsules are taken from the onset of premenstrual symptoms through first few days of menses (i.e., “symptom-onset dosing”). In addition, this study
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 4
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
intends to evaluate whether stopping capsules at the onset of menses or shortly thereafter is associated with SRI discontinuation symptoms.
The secondary objectives of this study are to: (1) estimate whether improvement in functional impairment and quality of life is greater with symptom-onset dosing using sertraline than with similarly administered placebo; (2) explore whether greater overall symptom severity, suicidal symptoms and severe physical symptoms are moderators of response for symptom onset treatment; and (3) examine whether symptom onset treatment is effective in treating physical symptoms in addition to mood symptoms.
2. Background: Describe the background information that led to the plan for this project. Please provide references to support the expectation of obtaining useful scientific data. When available, previous work in animal and/or human studies should be included.
Premenstrual Dysphoric Disorder: PMDD is a severe form of PMS characterized by affective and behavioral symptoms (1). Women with PMDD experience depressed mood, mood swings, irritability, anxiety, and changes in sleep and appetite as well as disturbances in functioning at home and/or work. PMDD differs from other forms of PMS in that patients must suffer from at least 5 symptoms, one of which must be affective, and these experiences cannot be “merely an exacerbation” of another psychiatric or medical condition (1). Symptoms associated with PMDD are confirmed by longitudinal “prospective self-ratings” that are tracked for 2 complete menstrual cycles. Applying these criteria, between 3% and 8% of women endorse severe premenstrual symptoms (2-4). Although the prevalence of PMS differs by age (2, 5), there is no evidence that women from various age groups express symptoms differently. Rather, retrospective report of PMDD symptoms appears to be stable across age cohorts (6).
The length of the symptomatic period varies among women, lasting an average of 7 days nearly every menstrual cycle (7, 8). While it is common to trivialize premenstrual disturbances such as PMDD, the illness is serious for the women who are afflicted. The morbidity of PMDD is due to its severity, chronicity, impairment in interpersonal relationships, and diminished work productivity (1, 9, 10). When one considers that women who suffer from PMDD have severe symptoms each cycle beginning in their early twenties and enduring until menopause (mean of 51 years), they will have as many as 2800 cumulative days or 7-8 years of symptoms. This, in conjunction with the prevalence of ~5%, places PMDD in a similar burden of illness category as dysthymia, major depression disorder, and other common medical disorders (11). The economic burden of PMDD is shown by studies demonstrating higher rates of treatment-seeking (6, 8, 11-18), lower work productivity and higher work absenteeism (6, 12, 14, 15) among symptomatic vs. non-symptomatic women.
While women with PMDD experience difficulty completing home and office-based tasks during their symptomatic period, they are particularly distressed by the adverse impact the illness has on their family relationships (12). Women with PMDD report that premenstrual symptoms lead to arguments with family and friends, difficulty with sex, and feelings that life is not worth living (17). The severity of PMDD is reflected by the suicide attempts made by approximately 15% of women with this illness (6, 19, 20). The personal pain associated with severe premenstrual symptoms and the impact the illness has on women’s family and employment roles argue for the need to define adequate treatment modalities.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 5
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Biology of PMDD: The cause of PMDD remains unknown. Because symptoms occur repeatedly during the luteal phase of the menstrual cycle it was felt originally that alterations in the hormonal milieu led to the illness. However, data in support of this are limited and it is also possible that symptoms result from abnormal central nervous system response to usual hormonal cycling (21). This is consistent with research suggesting dysregulation in neurotransmitter systems, including, among others, serotonin, GABA, and vasopressin.
Studies implicating dysregulation of the serotonin system show that in women with PMDD: 1) binding to imipramine receptors is lower than in controls (22, 23); 2) symptoms can be provoked during the follicular phase or worsened during the premenstrual phase if subjects are depleted of the 5-HT precursor, tryptophan (24, 25), or administered the 5-HT receptor antagonist metergoline (26); 3) and abnormal prolactin and/or cortisol responses occur after administration of a number of serotonergic probes including L-tryptophan (27, 28), buspirone (29), m-CPP (30), and fenfluramine (31, 32).
Data showing that SRIs effectively treat PMDD (see below) also strongly suggest that the serotonin system is etiologically involved in PMDD. However, these compounds may work through other systems as well (33-41). For example, preclinical work shows that SRIs alter progesterone metabolism, thereby increasing the endogenous GABA/benzodiazepine agonist, allopregnanolone production (42). Androgenic hormones can promote aggression and irritability and some suggest that these hormones are unusually active in women with PMDD (43). A novel oral contraceptive with androgen antagonist properties has recently been found effective for treatment of PMDD (44) As support for involvement of this system, we present pilot data (see Section C) demonstrating reductions in 17-OH pregnenolone, an androgen hormone precursor, after SRI treatment. Irritability and aggression have also been associated with vasopressin, which is released to a greater extent in the luteal phase (45). There is evidence that central (46, 47) and circulating (48) pools of vasopressin are reduced by SRI treatment. Additionally, inflammatory mediators and progesterone can promote fatigue and impair concentration (49). Studies from the 1980’s indicate that administration of anti- inflammatory medications improve premenstrual dysphoria and irritability as well as physical symptoms (50-55). SRIs also suppress delayed type hypersensitivity and inflammatory cytokine release in laboratory models of inflammation and autoimmune disease (56-59). These anti- inflammatory effects of SRIs could be mediated through 5-HT receptors on immune cells or by suppression of HPA axis reactivity (60).
Dr. Altemus is currently conducting pilot clinical studies exploring the effects of SRI treatment on inflammatory mediators and hormones in the androgen synthesis pathway (R21 MH071543). Of relevance to the current application, these studies will yield sufficient pilot data to choose the more promising among these potential biological mediators of SRI action in PMDD. If this application is funded, we would be well positioned to immediately apply for supplemental funding to examine the most likely biological mediators of response from the above list. This could be explored in women who undergo brief (intermittent) treatment and chronic daily treatment (during the extension phase).
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 6
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
The linked image cannot be displayed. The file may have been moved, renamed, or deleted. Verify that the link points to the correct file and location.
Antidepressant Therapy for PMDD. Data showing the treatment efficacy of SRIs for PMDD are strong (61- 80). These findings are reflected in a meta-analysis of 15 SRI trials that found the odds of improving were 6.9 (CI=3.9-12.2) times higher with an SRI than with placebo (81) (see Figure 1). About 70% of women with moderate to severe PMS or PMDD can expect to feel well or nearly well after SRI treatment, compared with 30-35% given placebo (9). Improvement has been demonstrated in both mood (61, 66, 74, 82-85) and physical symptoms with SRIs (4,87) although response is somewhat more mixed for physical symptoms. Other predictors of response are understudied, although age is not associated with response (78). Antidepressant type is discriminating; SRIs are more effective for PMDD than antidepressants not active at the serotonin transporter (61, 62, 67).
Initial studies showing SRIs to be effective in PMDD stipulated daily medication administration throughout the menstrual cycle. Subsequent work with sertraline (63, 64, 76, 86), fluoxetine (77, 80, 87), citalopram (73), paroxetine (79, 88), and clomipramine (72) demonstrated the efficacy of luteal phase dosing, (i.e., commencing the SRI treatment at ovulation with discontinuation on the 1st day of menses or shortly thereafter). This suggests a rapid onset of action for the SRIs in PMDD. In one study, discontinuation symptoms were not found with the abrupt cessation of medication (89), possibly because the active treatment period was brief.
A few trials have compared luteal phase with continuous dosing and the efficacy between them appears generally comparable (73, 87, 90) (Figure 2 and Preliminary Studies). A potential advantage to intermittent treatment is that patients often prefer this method. The reduction in side effects may enhance compliance. In one large study, among those discontinuing SRI treatment for PMDD, 43% did so because of side effects and 19% discontinued out of fear of dependence (91). In fact, research shows that side effects are reduced when fluoxetine (87) and paroxetine (92) are taken intermittently; attrition is lower when citalopram is dosed intermittently (73). These studies support the efficacy and tolerability of intermittent SRI treatment for PMDD. Intermittent dosing is well suited to a disorder that is itself intermittent; the diminished drug exposure, lower cost, and possible reduction in side effects are clear benefits.
Figure 2: The Efficacy of Intermittent Is Similar to Continuous Illness Severity & Suicidality as Moderators of Paroxetine Response to Intermittent Treatment. The treatment platform in these “intermittent” studies Placebo Intermittent Paroxetine stipulated that subjects initiate therapy just prior 100 Continuous Paroxetine to ovulation and continue through the onset of 80 menses for ~16 days of treatment. Women 60 monitored ovulation with a predictor kit that 40 indicated when pill taking should commence. 20 APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 7 0 Study End Point
Intermittent vs placebo,P <.001; Continuous vs placebo,P <.001; Continuous vs Intermittent, NS. *Based on a CGI score of 1 or 2 during the luteal phase of each menstrual cycle. Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
However, in a recent pilot study, 27 women were randomized to 3 cycles of treatment with open-label escitalopram taken at either symptom onset (~6 days prior to and until menses) or beginning with ovulation and continuing until menses (~16 days) (93). While the proportion of subjects improving by >50% were equivalent in both groups, half cycle treatment was superior to symptom onset treatment among the more severely ill group.
Suicidal symptoms may be an additional measure of illness severity, although it was not considered in the aforementioned study. We are unaware of other published PMDD treatment studies that have explored baseline suicidal thoughts as a moderator of response, possibly because most clinical trials exclude these subjects. In an ongoing effectiveness study (PI: KA Yonkers) (R21MH6237901) suicidal subjects were not excluded and baseline suicidal thoughts conferred a lower likelihood of response (see Preliminary Data, Study #5). While this was non-significant in our small study (n=69), suicidal thoughts deserve further exploration since they may be important moderators of clinical response. In fact, a recent study in adolescents found that suicidal thoughts at baseline were associated with a lower likelihood of response to major depressive disorder (MDD) (94).
Efficacy of Less than Half-Cycle Treatment Compared to Placebo. Aside from our pilot data (see Preliminary Studies), we are aware of only a single study that has compared placebo and an intermittent platform shorter than the full 2-week luteal phase. Enteric-coated fluoxetine (90 mg) was administered weekly either once at 7 days or twice at 14 and 7 days before expected menses (80). Women who received active medication for 2 weeks experienced greater overall improvement as well as significantly greater improvement in mood and social functioning compared with placebo. Outcomes for women who took medication for 1 week did not differ significantly from placebo. While it is possible that the lower efficacy for 1 relative to 2 weeks of active treatment means that shorter treatment is inadequate, other factors could be responsible. First, enteric coated, weekly fluoxetine yields similar plasma levels to daily fluoxetine after 1-2 weeks of administration. Given that the ½ life of fluoxetine is 24-72 hours, and the ½ life of its major bioactive metabolite is 4-16 days, women in the 2-week treatment arm were exposed to a higher overall dose of medication during the final luteal phase week. Second, the quoted study did not initiate pills at “symptom onset”. Hence, the failure of a single dose treatment administered 7 days prior to menses may have resulted from starting pills after the onset of symptoms for those subjects who develop symptoms earlier in the luteal phase. Finally, differences between groups were small and the study may have been underpowered.
In any case, findings from the fluoxetine study contrast sharply with our pilot work and preliminary work of others (93, 95). These conflicting findings suggest that a definitive study with adequate power and duration is needed to determine whether a symptom onset treatment strategy can be recommended to women with PMDD. Given the novelty of this treatment modality, such a study should ideally assess whether efficacy is enduring and whether it is effective for individuals with more severe, including suicidal symptoms. There are a number of potential advantages to symptom-onset dosing, not the least of which is that it is likely to result in improved compliance relative to luteal phase dosing, especially for women who either have irregular cycles or difficulty estimating when ovulation occurs.
Efficacy of Intermittent Treatments for Physical Symptoms. The preponderance of studies of full cycle SRI dosing for PMDD have shown significant improvement in physical as well as mood symptoms, (96, 97). One full cycle study suggested that response for physical symptoms may require higher SRI doses than needed for mood symptoms (98). However, some studies of luteal phase SRI dosing suggest that with intermittent dosing physical symptoms are less responsive than mood symptoms (73, 77, 80, 99) although this is not always the case(100). Since symptom-onset treatment is even briefer APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 8
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
than half-cycle dosing, the experience of severe physical symptoms premenstrually, may moderate overall response.
Risk of Discontinuation Symptoms with Intermittent Treatment. Intermittent dosing requires that an individual take pills for several days to weeks and stop medication abruptly. A concern with this mode of administration is that abruptly stopping a short-acting SRI may herald the onset of a discontinuation or withdrawal syndrome (101-103). These symptoms typically emerge within 1-5 days of stopping treatment and generally resolve within 5 to 7 days but may endure 1-2 weeks. Symptoms most often include dizziness, insomnia, vivid dreams, nausea, fatigue, headaches, parasthesias, anxiety, and irritability. Several double-blind placebo substitution studies have shown that among the SRIs, the risk of discontinuation symptoms appears to be higher with paroxetine, venlafaxine, and sertraline than with fluoxetine (101-103). We did not find evidence of discontinuation symptoms associated with half-cycle sertraline use (89). Nonetheless, only some of the symptoms associated with discontinuation were evaluated in our report, indicating a need to replicate and extend our findings.
Evidence Supporting Rapid Onset of Action for Antidepressant Agents. A number of scholarly articles suggest that antidepressants must be taken for approximately 4 weeks before they show therapeutic action. On the other hand, as noted in a recent review and meta-analysis (104), many investigators maintain that antidepressants ameliorate depressive symptoms within days (105-108). There are substantial data from PMDD studies that support a rapid onset of action such that treatment initiated 2 weeks before menses is effective. Challenge studies with m-CPP, a 5HT2C agonist, also provide indirect evidence supporting a rapid response of PMDD symptoms to serotonin reuptake inhibition. Within one hour of administration, mCPP produces improvement in PMDD symptoms (30), suggesting that the acute increase in availability of synaptic serotonin during SRI treatment may act through 5HT2C receptors. In an unpublished study conducted by Graca Cardoso, MD and Peter Schmidt, MD from the Behavioral Endocrinology Branch at the NIMH intramural program (personal communication), women with prospectively diagnosed PMDD were given 20 mg/d fluoxetine in an open-label trial aimed at evaluating the time course for therapeutic action of fluoxetine on PMDD symptoms. Fluoxetine was started in the AM during the luteal phase of the menstrual cycle, after the development of PMDD symptoms, and continued for several cycles as maintenance therapy. The symptom response to initiation of fluoxetine treatment was monitored hourly during waking hours until several days after the onset of menses. To date, 5 women have been evaluated, and all had a marked decrease in symptoms within the first 48 hours of starting fluoxetine.
PRELIMINARY STUDIES
Recruitment Experience (Yale). Prior to moving to Yale, Dr. Yonkers ran a program at the UTSW Medical Center conducting treatment protocols for patients with PMDD. A PMDD treatment program has been established at Yale and has been ongoing for the past 6 years, with over 1000 women screened via questionnaire. Among women who were evaluated at Yale and entering a PMDD treatment protocol, 81% were Caucasian, 9% were African-American, 9% were Hispanic, and 1% were other ethnicities. Dr. Yonkers has an active presence in the Obstetrics & Gynecology clinics at Yale New Haven Hospital, the Hospital of St. Raphael, and Bridgeport Hospital. In a recent pilot study of mental health care utilization among pregnant women, she recruited a cohort of 529 racially diverse women, including 12% non-Hispanic white, 40% African American, 44% Hispanic, and 4% Asian and other groups (109). In the just recently completed pilot study on symptom onset treatment presented below, 65% of women were Caucasian, 20% were Hispanic and 15% were African American (112). Attrition for long-term PMDD treatment trials is illustrated by pooling retention data from two ongoing APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 9
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
studies that included 5 and 6 menstrual cycles of treatment. One compared fluoxetine, placebo, and calcium, while the second was an open-label effectiveness study of sertraline (25 mg, 50 mg, or 100 mg). Using data from both studies, 81% of subjects assigned placebo (n=13), and 72% of subjects assigned to sertraline (n=79) completed their protocol.
Recruitment Experience (Cornell). Dr. Altemus heads the Neuroendocrinology Research Laboratory in the Department of Psychiatry at Weill Medical College and is Attending in the Payne Whitney Women’s Program. Over the past 4 years, Dr. Altemus’ program recruited subjects with PMDD for industry treatment trials and for investigator-initiated treatment, neuroimaging and cognitive studies. She is now PI of a NIMH-funded trial of a hormonal treatment for PMDD and an investigation of the effects of SRIs on gonadal steroid hormones. Her collaborator on these projects is Dr. James Kocsis, who directs the Depression Studies at Weill Medical College. Dr. Kocsis has been conducting treatment studies in mood disorders since 1979. He has extensive experience recruiting and retaining subjects for long-term treatment studies, including trials that have randomized depressed subjects to active treatment or placebo for up to 24 months. His experience in conducting long-term treatment trials will be an asset to the project at Cornell as well as other sites. PMDD referral sources include the internal medicine, ob-gyn, and psychiatry clinics of New York Presbyterian Hospital-Weill Cornell Medical Center, and the NY Women’s Mental Health Consortium, a network of over 150 local clinicians who specialize in treating mental health disorders in women. Over the past 6 months, 6 patients per month have initiated screening for PMDD, and 2 per month have qualified. The ethnic/racial distribution over the past year has been 52% Caucasian, 10% Hispanic, 27% African American and 10% other.
Recruitment Experience (VCU). Dr. Kornstein is Executive Director of the Virginia Commonwealth University (VCU) Mood Disorders Institute, where she has been conducting treatment protocols in mood disorders, including PMDD, since 1992. She is also Executive Director of the VCU Institute for Women’s Health, which was designated by the US Department for Health and Human Services as a National Center of Excellence in Women’s Health. She has had extensive clinical experience treating women with PMS and PMDD in her role for over 10 years as Co-Director of a multidisciplinary PMS Clinic in the Women’s Health Center at VCU. She is a PI and Regional Site Director for the NIMH Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study and the new NIMH Depression Treatment Network. Her site had the highest minority enrollment among the 14 regional centers participating in the STAR*D trial. Of 238 participants enrolled at VCU, 32% were Caucasian, 59% were African American, 7% Native American, 3% Hispanic, and 2% Asian and other groups. She has conducted many clinical trials with study durations over six months, and has had excellent recruitment and retention rates. In a recent 9-month depression study, her site enrolled 56 participants, of which 45 (80%) completed. Subjects for this study will be recruited through advertisements and referrals from the Women’s Health Center and the OB/GYN and primary care clinics at VCU Medical Center.
Study 1: Symptom-onset treatment of women with PMDD using either paroxetine-(CR) or placebo. Aims: This study was designed to 1) obtain preliminary data regarding the efficacy of symptom-onset treatment with an SRI compared to like-administered placebo; 2) determine whether women who take paroxetine intermittently experience withdrawal symptoms; 3) evaluate whether women accurately estimate when to take capsules such that they avoid having clinically meaningful symptoms (at least a 3 on a 6-pt scale). (110).
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 10
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Methods: Two cycles of daily ratings diagnosed Figure 3: Crossover Trial of Paroxetine vs Placebo Administered at Symptom Onset (n=20) PMDD; requirements were the same as for the study proposed in this application (see Research 30 Design and Methods). Subjects and evaluators Group 1: Placebo Followed by Paroxetine 28 were masked to treatment condition; 20 women 26 Group 2: Paroxetine folowed by Placebo 25.1 23.7 were randomized to either 1 cycle of paroxetine 24 22 (25 mg) followed by 1 cycle of placebo or 1 cycle 20 18.2 of placebo followed by 1 cycle of paroxetine (25
SCORE 18 17.7 mg). Women took capsules when they started to 16 feel symptoms and continued them until the 1-3 14 14.1 12 12.7 day interval after the onset of menses, depending 10 on when their symptoms typically remit. Capsules Baseline Cycle 1 Cycle 2 were dispensed in bottles with a microchip embedded in the cap that determined the day and time bottles were opened. Efficacy estimates of paroxetine compared with placebo were tested using a paired t-test. To evaluate whether women commenced medication when needed, i.e. proximate to the onset of premenstrual symptoms, we determined the number of women who experienced at least 5 mild (3 on 6-pt scale) symptoms during the 3rd, 2nd, and 1st day prior to commencing capsules. To assess whether women in this pilot study suffered from a possible discontinuation syndrome, we looked at our data in two ways. We reasoned that discontinuation symptoms occurring after stopping medication would prompt a higher score on the Daily Record of Severity of Problems (DRSP) during the 3-5 days post medication discontinuation. Thus, the average total daily DRSP scores during the 5- day interval post pill discontinuation were compared for women who had stopped either paroxetine or placebos after the first cycle of randomized treatment. Since not all DRSP symptoms are found with a discontinuation syndrome, we repeated this analysis using symptoms found by others to be associated with SRI withdrawal. We used 9 DRSP items (irritability, anxiety, depressed mood, feeling out of control, insomnia, poor concentration, fatigue, headache, and muscle aches) (101) to compute an SRI- withdrawal symptom factor. Results: Average PMTS scores for baseline and post-treatment are presented in Figure 3. There is a trend for a difference between conditions even among these 20 women. During the 1st cycle, the difference was 5.5 (sd=6.5; P=.08). Women who took paroxetine after placebo improved only slightly, illustrating a diminished drug placebo difference during the 2nd cycle. However, women who were switched from paroxetine to placebo became nearly twice as symptomatic. These data provide preliminary support for the efficacy of paroxetine compared with placebo when capsules are taken at symptom onset. Women took capsules for an average of 94 days including capsules taken during the first few days of menses. Data on pill taking are shown in Figure 4; only 3 women (18%) had 5 symptoms that were at least mild on the 3rd day
The linked image cannot be displayed. The file may have been moved, renamed, or deleted. Verify that the link points to the correct file and location. prior to starting her capsules. Two (12%) women had at least 5 mild symptoms two days before beginning capsules, and 4 (24%) subjects had at least 5 mild symptoms one day before commencing capsules. When we looked at greater severity of symptoms (4 or moderate) in relation to pill taking, no women had at least 5 symptoms of moderate severity either three or two days before commencing pill taking, while 3 women (18%) had at least 5 moderate symptoms on the day prior APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 11
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
to commencing pill taking. These data illustrate that in this small pilot study, women were able to estimate when to begin capsules in relation to the onset of their premenstrual symptoms. In the analysis evaluating discontinuation symptoms the average daily DRSP score for the 5-day interval post ceasing capsules was 39.9 12.8 for women who stopped paroxetine and 40.721.9 for women who stopped placebo. When the cluster of 9 symptoms associated with possible withdrawal were compared, the mean was 14.34.6 for women stopping paroxetine, and 16.59.2 for women who stopped placebo. Thus, symptoms were slightly better among women who were treated with paroxetine and abruptly discontinued medication. Conclusions: Although the number of women in this pilot is limited, subjects taking paroxetine improved substantially more than did subjects who took placebo. Women initiated pill taking within 1 or 2 days of symptoms becoming mild or worse. No problems with discontinuation symptoms as determined prospectively using the DRSP were identified in this small cohort. A larger and longer study is needed to confirm efficacy findings for this treatment modality. Moreover, a careful assessment of discontinuation symptoms is needed before clinicians can recommend that this modality as safe.
Study 2: Intermittent and Continuous treatment with sertraline or placebo for PMS. Aims: This study evaluated sertraline in the treatment of moderate-to-severe PMS (not PMDD) using 3 different dosing strategies: luteal phase (14 days), continuous, and symptom-onset dosing. Methods: 314 women from 22 US sites were randomly assigned to 4 cycles of fixed-dose treatment with sertraline (25 or 50 mg/d) or placebo. ~ 80% of women met criteria for PMDD, while the rest were subsyndromal (PMS). Drs. Kornstein and Yonkers participated in this study. Once assigned, patients underwent the following sequential treatment regimen: 1) two cycles of luteal phase dosing, beginning on Day 14 of the cycle and continued until the onset of menses; 2) one cycle of continuous daily dosing throughout the menstrual cycle; and 3) one cycle of dosing at the onset of premenstrual symptoms (as determined by the patient) and continuing until the onset of menses (symptom-onset dosing). Assessments included the Daily Symptom Report (DSR) and Clinical Global Impressions (CGI) ratings. Table 1. Treatment response with different dosing regimens in women with PMS
# Luteal Phase Dosing (n=252) Continuous Dosing (n=225) Symptom-Onset Dosing (n=168)
DSR-total score: Baseline Sertraline-25 mg 26.5 + 10.7 -12.6 + 1.2* -16.0 + 1.1** -14.2 + 1.5* Sertraline-50 mg 28.2 + 10.0 -12.1 + 1.1* -13.3 + 1.1 -13.0 + 1.5 Placebo 26.2 + 10.5 -10.8 + 1.1 -10.7 + 1.1 -10.0 + 1.5 CGI-Improvement Sertraline-25 mg 2.4 + 0.1* 2.3 + 0.1** 2.3 + 0.1** Sertraline-50 mg 2.5 + 0.1 2.5 + 0.1† 2.4 + 0.1** Placebo 2.7 + 0.1 2.8 + 0.1 2.9 + 0.1
#Least Square-mean (+SE) change scores (except for CGI-I) based on repeated measures ANCOVA of either 1 cycle (continuous and symptom-onset dosing) or 2 cycles (luteal phase dosing); *p<.05, **p<.01; †p<.08 Results: Subjects treated with 25 mg of sertraline achieved significant improvement in daily symptom total scores and CGI-I across all dosing regimens (Table 1). Conclusions: These results provide preliminary evidence that low doses of sertraline may be effective for moderate-to-severe PMS using a symptom-onset dosing strategy. Limitations are that the study
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 12
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
design was sequential, attrition occurred before the symptom-onset cycle was initiated, the treatment Day of Cycle -15 -10 -5 -2 1 phase endured for only one cycle and
90% 80.8% not all women had PMDD. 80% 72.8% 69.5% 70% 58.2% 60% Study 3: Duration of premenstrual 50% 45.9% Symptoms in women with PMDD. 40% 32.9% 34.4% 30% 24.8% Aims: The goals of this study were to 20% 15.3%
Patients meeting PMDD meeting Patients 11.3% 10% 1) identify the duration of moderate to Symptom Severity Criteria, % Criteria, Symptom Severity 0% severe symptoms in women with 75% increase over DRSP follicular score 5 or more DSM-IV symptoms of at least moderate severity PMDD; 2) determine when in the DRSP Symptom Severity Criteria for PMDD cycle symptoms begin or become problematic for PMDD women; and 3) investigate the temporal sequence for the Figure 6. Mean total DRSP symptom severity score across onset of various symptoms (see Manuscript the menstrual cycle (n=276) #2 in Appendix A). 90 80 Methods: This was a database reanalysis of 70 a mulit-center study using sertraline. 60 Entrance criteria were similar to the criteria 50 for the proposed study. All women 40 Mean DRSP Total Score 30 prospectively confirmed premenstrual
20 symptoms for at least 2 cycles, underwent a 10 single-blind placebo cycle, and continued 0 daily ratings during the course of the 3-cycle -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 Day of Menstrual Cycle study. Women took pills for approximately Mean mid -follicular DRSP total score = 33.5 Onset of menses 16 days, starting with a positive ovulation indicator test through the onset of menses (86). For this analysis, the 2 qualifying cycles were used to analyze the onset and offset of symptoms. The outcome measure to determine symptom expression was the DRSP. Drs. Yonkers and Kornstein participated in data collection.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 13
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Results: Figure 5 illustrates that the vast majority of women with PMDD experience an onset of
The linked image cannot be displayed. The file may have been moved, renamed, or deleted. Verify that the link points to the correct file and location. moderate to severe symptoms less than 10 days before menses; only 11-15% of women experience moderate to severe symptoms for >10 days. The minimum possible score on the DRSP is 21 (1= none for all 21 items), and the average follicular phase score in this study was 33.5. Figure 6 illustrates that symptoms worsen by about 50% approximately 6 days before the onset of menses in the majority of women. Figure 7 shows that there is similarity in the timing among groupings of symptoms, but irritability has a slightly earlier onset. About half of women experience moderate to severe symptoms (DRSP scores of 4-5) for anger and irritability during the 5 days before the onset of menses. Conclusions: The majority of women experience <10 days of moderate to severe symptoms. Symptoms worsen by 50% approximately 6 days before menses and peak 2 days before menses. Anger and irritability show the greatest symptom increase and may worsen earlier than other symptoms. Symptoms typically continue into menses for 2-3 days. These data suggest that if the therapeutic benefit of SRIs is immediate, symptom-onset dosing would result in significantly less medication exposure for women with PMDD. Study 4: The Effects of SRI Treatment on Pregnenolone and Progesterone Aims: The goal of this study (R21 MH071543; M Altemus) was to determine the impact of SRI treatment on hormones in the androgen synthetic pathway.
Methods: In a cross-sectional study, 5 women with OCD chronically treated with SRIs (fluoxetine, paroxetine, sertraline), 3 women with OCD not taking SRIs, and 11 healthy female control subjects had a blood sample drawn between 8 and 10 AM, in the early follicular phase of the menstrual cycle (3-7 days after onset of menses). Mean age was 35+/- 5 years and did not differ between the SRI- treated and medication-free groups.
Figure 8: SRIs Effects on Pregnenolone & Progesterone Results: Compared to female control subjects, medicated women with OCD 17-OH-PREGNENOLONE 17-OH –PROGESTERONE taking SRIs had significant reductions in 180 9 160 8 basal 17-OH -pregnenolone, and 17-OH- 140 7 120 6 progesterone. Differences were not 100 OCD 5 * OCD 80 CONTROL 4 CONTROL statistically significant for DHEA and * 60 3 androstenedione, possibly due to increased 40 2 20 1 variability among SRI-treated subjects in 0 0 SRI no SRI SRI no SRI these measures. There was no effect of diagnosis on any of the hormone levels. Within subject comparisons rather than a cross sectional study may reveal changes in these and perhaps other hormones during SRI treatment. A supplemental application would allow us to test this both acutely and chronically.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 14
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Conclusions: SRIs may have a unique non-serotonergic effect on the pathophysiology of PMDD by influencing hormones in the androgen synthetic pathway.
Study 5: The Effects of Illness Severity and Suicidality on Treatment Response
Aims: The goal of this ongoing study (R21 MH6237901; K Yonkers), has been to determine the real- world effectiveness of half-cycle SRI treatment for PMDD and subthreshold PMDD.
Methods: Women (n=69) are recruited from typical practice settings and treated with half-cycle sertraline. Treatment instructions are given in a manner typical of clinical practice and medication is dispensed in bottles with medication event monitoring capsules. For this analysis, we assessed response (Clinical Global Impression Improvement Scale score of 1) in women who did or did not endorse suicidal symptoms at baseline. Also, we explored whether improvement in physical symptoms was significantly correlated with improvement in mood (as determined by the PMTS).
Results: 60% of subjects took 50mg and felt that dose adequate while 33% increased to 100mg during the symptomatic phase. Only 7% felt that 25 mg was an adequate dose while 2% could not tolerate more than 25mg and felt the dose inadequate. Three out of 14 (21%) subjects with suicidal symptoms and 20 out of 54 (37%) without suicidal symptoms were classified as responders (p=0.35). There was significant correlation between improvement in mood and improvement in physical symptoms (r=0.45, p<.0001) and a significant effect of treatment on improvement in mood (t (19.3)= 3.32, p=0.003,) but not in physical symptoms.
Conclusion: These results are in the same direction of our hypotheses moderator (suicidal symptoms may moderate response). Additionally, physical symptoms are correlated with mood symptoms but are less affected by treatment. The limited sample size renders inadequate power for some tests.
3. Research Plan: Please provide an orderly scientific description of the study design and research procedures as they directly affect the subjects.
This is a 3-center, randomized, placebo-controlled, double-blind, clinical trial, which is comprised of 11 study visits within 3 stages: 2 cycles of screening; 6 cycles of double-blind placebo or sertraline (50 – 100 mg/day) self-administered when subjects feel symptoms have commenced; and 3 cycles of daily sertraline treatment as an extension phase for subjects who have completed per protocol treatment, have found intermittent treatment problematic (i.e. discontinuation symptoms) or who have failed to respond adequately to intermittent treatment (Figure 9).
The linked image cannot be displayed. The file may have been moved, renamed, or deleted. Verify that the link points to the correct file and location.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 15
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Study Visit Procedures
Table 2: Symptom Onset Study Measures and Procedures
Measure Screening Baseline Visit Visit Visit Visit Visit Visit 7 Visit 8 Visit 9 Visit (Visit 1) 2 3 4 5 6 (End Double (Daily Open) Daily Open Blind) End Month Beginning 3 4 5 6 7 8 9 10 12 of month 1 Pre Randomization X Double Blind X X X X X X Intermittent Daily Open Label X X Randomization Event X In Office X X X X X X X X By Phone X X DRSP X X X X X X X X X X Weight X X X Pregnancy Test X X X X X X X MINI X PMTS X X X X X X X X X IDS-C X X X X X X X X X CGI X X X X X X X X X HASS II X X X X X X X X Q-LES-Q X X X X Ease of Pill Taking X X X X X X Concomitant X X X X X X X X X Medications/Adverse Events
Screening Procedures:
Subjects will be recruited by a variety of venues including posters, mailings, advertisements (print, radio, television), referral from gynecological practices and clinics and other referrals. Subjects who call in responding to a referral or advertisement will be informed of study procedures and we will request verbal consent to conduct a brief screening for eligibility (eg, currently pregnant or planning pregnancy, age and other exclusionary criteria, ongoing psychiatric or general medical treatment and self reported symptoms). Women who preliminarily appear to be eligible will be invited for an in person assessment. Women who are provisionally eligible will be sent daily rating forms (the Daily Rating of Severity of Problems; DRSP), directions to the research office and a note confirming the appointment.
Screening Visit: During the full screening visit, the Research Assistant (RA) will review study procedures, answer questions, and obtain written and verbal consent. The interview will include a discussion of the need to use contraception during the trial. The RA will review in detail the process of recording daily ratings and answer questions that may have arisen while previously keeping ratings. Subjects will continue daily using the DRSP form and will be asked to return forms weekly via mail, fax or e-mail to obtain 2 qualifying cycles of daily ratings (Cycles A and B) (113). Paper forms will have headers with patient IDs rather than subject name to protect confidentiality.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 16
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
The RA will follow up with all potential subjects one week later and on additional occasions, as necessary, by phone. She will note the estimated onset of menses and will call women again after menses to ensure that they remain interested and to remind them to return forms. She will review daily rating forms with the study subject so that the subject is comfortable with the anchors set on the scale. Regular contact with study personnel enhances participant and proper screening.
As the daily rating forms are returned, the RA will enter results of the qualifying cycles into TrialDB, an internet-accessible generic open-source clinical study data management system, which will automatically calculate follicular and luteal phase averages as well as follicular and luteal phase differences for all DRSP symptoms. Screening forms will be reviewed by The PI, who will identify medical or safety issues that would preclude further participation. The TrialDB system will confirm that subjects are eligible by daily ratings and other criteria. For a definitive PMDD diagnosis, DSM-IV stipulates 2 cycles of daily ratings to confirm luteal symptoms. Based on daily symptom scores, subjects must have a minimal average luteal phase score of mild (> or =3 on a 6-point scale) for at least 5 PMDD symptoms during the 5 most symptomatic of the final 7 luteal phase days and the first 2 days of menses onset, and the average follicular phase score must not be >2 on these same items. Women who continue to be eligible will be scheduled for a baseline visit (Visit 1).
Women will be allowed to chart for an additional cycle if one of the cycles does not meet criteria, and if the subject reports that the cycle was unusual for her (e.g., not as severe as usual in the luteal phase or more severe because of other factors during the follicular phase). The study psychiatrist will review ratings and if necessary, interview the subject by phone to determine whether she should be allowed to rate the additional cycle. Because medication dosing will not begin for at least 10-14 days after the onset of menses, daily ratings and other eligibility criteria from the 2-month screening phase can be reviewed early in the following cycle with sufficient time to start treatment by symptom onset in the post-screening cycle.
Should a subject show symptoms consistent with MDD or another psychiatric condition, she may be given immediate referrals over the phone or in person. Subjects who do not show sufficient severity after several months of daily ratings will end participation at this point but will be provided referrals. In some instances, forms will not be returned prior to Baseline Visit 1 and these procedures will occur at that visit.
Visit 1: Baseline Visit (Beginning of Treatment Cycle 1): The RA will collect any additional DRSP forms. In addition to patient interview, the DRSP will be reviewed to inform these ratings. The RA will administer the MINI Neuropsychiatric Interview and review this with the PI or a psychiatric co- Investigator (111). Subjects with exclusionary conditions (substance dependence, bipolar disorder) will be provided appropriate referrals.
The blinded rater, who like the RA is blind to treatment but also study design, will obtain efficacy measures. This rater will assess luteal phase symptom severity, with the luteal phase defined as the 7 days prior to the onset of menses for this and subsequent visits, using the Premenstrual Tension Scale (PMTS) (112), the IDS-C (113) and the Clinical Global Impressions (CGI)-Severity scale. Health- related quality of life and functional impairment during the 7 days prior to the onset of menses will be evaluated at this visit and every 2 to 4 months using the Quality of Life, Enjoyment and Satisfaction Questionnaire (Q-LES-Q) (114). These results along with the DRSP, IDS-C and MINI, will be entered into TrialDB. The RA will review these results and perform TrialDB quality control procedures to APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 17
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
ensure complete capture of data. In addition, the RA will check the IDS-C for a positive response to the question regarding suicidal thoughts.
Prior to randomization, the subject will be interviewed by a study psychiatrist, who will confirm results collected by the RA. Subjects who continue to meet entrance criteria will have a urine pregnancy test, the results of which must be negative. Women who are not eligible but desire treatment will be given appropriate referrals.
Some subjects will have suicidal ideation prior to entering the study or develop it during the study, necessitating procedures for assessing and managing their care. All subjects will be closely monitored for significant suicidal risks by completing the IDS at each visit (115). Those who endorse suicidal thoughts on the IDS as indicated by a score of 2 or higher (IDS question 18) will be administered the Harkavy-Asnis Suicide Survey II (HASS-II). If the HASS II confirms suicidal ideation, the subject will be assessed by the PI or Co-Investigator immediately. If the clinician considers the subject a “significant suicide risk” after assessing current risk factors (e.g. new life stressor, poor social support, poor impulse control, severe hopelessness, etc.), the subject will be referred for more intensive outpatient or inpatient treatment but will be followed by the study team until transfer to another clinician is complete. If they are not deemed a significant suicide risk, they may be eligible but will be monitored carefully.
The randomization list for each site (in blocks of 6), stratified by OC use and associated bottle numbers will be prepared by Dr. Gueorguieva, the Biostatistician, and will be entered electronically into TrialDB. A program running within TrialDB will utilize the study's eligibility rules to confirm or reject eligibility of these subjects for randomization. For eligible subjects, the software will automatically contact the RAs at each site via E-mail, transmitting information regarding the appropriate bottle number to be dispensed. Subjects will be given a bottle with the appropriate number that will contain either placebo or tablets with 25 mgs of sertraline. This bottle will have sufficient medication for 1 menstrual cycle if dosed at 50 mg for 14 days. We start at a dose of 50 mgs since this is the minimal dose that has been used in half cycle dosing for women with PMDD.
Some subjects may have difficulty taking the starting dose of 50 mgs (2 tablets). If a subject is intolerant of this dose, it may be lowered to either one tablet (25 mgs or similar appearing placebo) or one-half tablet (12.5 mg). She will be titrated over the course of the next several cycles up to 1 pill (25 mgs) or 2 pills (50 mgs) depending upon her starting dose. Over subsequent cycles she will be titrated up to 2 pills (50 mgs). However, if she is unable to tolerate the 50 mg dose, she may continue with the 12.5 mg or 25 mgs dose intermittently. Subjects and the investigative team will indicate when the dose change occurred and the number of pills taken (see below).
To guide subjects regarding the timing for starting tablets, the RA will review ratings with the subject and discuss the symptom pattern during the previous two cycles. The subject and RA will determine a “target” day that she can use to decide whether to commence taking tablets. This target day will only be a guide, and the subject will be told that she may start tablets earlier if symptoms commence earlier than anticipated or wait longer if symptoms have not yet been present and/or problematic. Subjects will also be told to call the research office immediately if their medications change between visits.
With RA guidance, subjects will determine when they are bothered by their premenstrual symptoms and commence taking tablets. This close monitoring by the investigators and subjects will allow them to modify recommendations regarding when women should start taking tablets to improve treatment APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 18
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
efficacy. The 6-cycle treatment phase will allow sufficient time for the subject to determine when it is best to start taking tablets, a procedure that reflects clinical practice. This approach also allows us to statistically explore whether there is a minimum number of days before the onset of menses by which most women will have had to start tablets to show a response to treatment. Findings from such an analysis may show, for example, that tablets must be taken for 7 or 10 days, or must be taken before a woman scores a 4 (moderate) on a certain number of symptoms to achieve maximum benefit. Findings such as this can then be incorporated into clinical practice recommendations.
Tablets will be dispensed in bottles with caps that have a microchip embedded into the cap (Medication Event Monitoring System (MEMS) caps). Women will be asked to open the bottle only when they plan to take medication and to remove one dose per day. In pilot work, we have found that women are able to comply with this request. We will ask women to stop taking medication within 3 days after the onset of menses (Days 1-3 of the menstrual cycle). We allow up to 3 days post menses because prior work shows that women with PMDD continue to experience moderate to severe symptoms during the first several days of menses (see Preliminary Studies, Study # 3). This flexibility in the stop date will provide for adequate treatment if women have a few days of symptoms during the follicular phase. We considered allowing a wider window for the medication stop date, but our pilot data show that few women continue having symptoms beyond 3 days into the menstrual phase of the cycle. Further, we found that women experience substantial relief from follicular phase symptoms even if they stop taking tablets at Day 1 or 2. Women who feel they need to take tablets longer will be re- evaluated for another clinical condition (eg, MDD, dysthymic disorder, GAD). Such women will be asked to continue efficacy measures since this is an “intent-to-treat” cohort.
Subjects will continue daily ratings, including an addendum assessing withdrawal symptoms (see below), throughout the trial. They will also indicate on the form when they took tablets and how many were taken. This will both provide a back-up to the MEMS cap information in case the cap fails and provide additional information regarding the number of tablets taken.
After completing the visit, subjects will be given $65 ($50 for 2 full menstrual cycles of DRSP charting, $10 for the Baseline Visit and $5 for mileage costs) and another appointment will be scheduled.
Visit 2 (Beginning of Treatment Cycle 2): Subjects will be seen for Visit 2 approximately 5 days after menses associated with the first treatment cycle (between Days 5 and 7 of Cycle 2) to capture all luteal phase events and to assess discontinuation symptoms. Rating scales will be administered (PMTS, CGI- I, IDS-C, HASS-II) by a blinded rater, subjects’ perceptions of the ease of pill taking elicited (EPT), and the daily self-ratings (DRSP) will be reviewed by the RA and clinician. Subjects will be reimbursed $10 for completing this first double-blind treatment cycle and $5 for mileage costs for a total of $15. Ratings will be entered into TrialDB immediately. The RA will conduct a tablet count, record the number of tablets taken each day and will download information from the MEMS cap. TrialDB data quality control procedures will be followed and used to confirm accuracy of the source documents. Study medication sufficient for the next menstrual cycle will be dispensed.
Discontinuation symptoms can occur with short-acting SRIs and serotonin-norepinephrine reuptake inhibitors. In a systematic, long-term, 28-week study involving panic disorder patients, abrupt discontinuation of sertraline was primarily associated only with insomnia (15.7% of patients randomized to placebo vs. 4.3% continuing on sertraline) and dizziness (4.3% of patients continuing to
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 19
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
take sertraline and 16.4% switched to placebo) (116). This is in line with findings by Michelson and colleagues, who found that dizziness was the complaint most often identified by women who had stopped sertraline (101). Discontinuation symptoms may endure for weeks, although this is unusual. Typically, symptoms cease within 3-8 days post cessation of tablets (101-103). Discontinuation symptoms associated with short-acting SRIs peak on the 4th day post-discontinuation (101). Because prior studies find discontinuation symptoms within 5 days of stopping tablets, we will evaluate this time interval. However, because subjects collect symptoms daily throughout the trial, we can also explore other intervals as well as possible diminution in these effects over consecutive menstrual cycles.
Subjects in our pilot did not cite discontinuation symptoms. However, our pilot assessed a limited number of symptoms, and some difficulties associated with discontinuation (e.g. dizziness) were not evaluated. Since the goal of this study is to inform clinicians and patients about the benefits and risk of symptom-onset treatment, we plan to thoroughly investigate whether women are at risk for discontinuation symptoms. Accordingly, we will add a short addendum to the DRSP that asks additional symptoms associated with discontinuation. This will allow us to complete the 17-item Michelson SSRI withdrawal checklist daily. We will have information on symptoms during medication or placebo treatment and the days after tablets have been stopped. Since these ratings are collected daily, we will be able to assess whether discontinuation effects occurred during the 5 days after tablets were stopped, but we will also be able to plot our data and look for symptom spikes during other intervals (101). Should a woman have discontinuation symptoms that are moderate to severe, the tablets will be restarted, and she will be tapered off over the next week. If the subject is unable to tolerate the symptom-onset method of medication administration due to the severity of withdrawal symptoms or other side effects, she will be removed from the symptom-onset treatment phase and offered open-extension phase daily treatment. Similarly, if the subject fails to respond and would like to discontinue per protocol medication, she may be released from the symptom-onset treatment phase and offered open-extension phase treatment.
If a subject shows an insufficient response at Visit 2 (CGI-I > 1), her dose may be increased from 2 tablets (50 mg of sertraline or placebo) each luteal phase day after symptom-onset to 4 tablets (100 mg of sertraline or placebo). In order to enhance tolerability, subjects will be asked to take 2 tablets (50 mg of sertraline or placebo) each day for 2 days and then increase to 4 tablets (100 mg of sertraline or placebo) through day 1 of menses. For day 2 and 3 of menses, the subject will be asked to reduce the dose to 2 pills (50 mg of sertraline or placebo) per day. For subsequent cycles, unless directed otherwise by the clinician, the subject will take 2 pills for the first 2 luteal phase days after symptom- onset, then increase to 4 pills (100 mg of sertraline or placebo) per day beginning on day 3. She will continue on 4 pills per day through the first day of menses. For days 2 and 3 after menses, she will decrease pills to 2 pills per day. The RA will also review daily ratings and note when the subject commenced taking tablets. The subject will keep a chart indicating when she took 2 tablets and when she took 4. These procedures have been used successfully in two pilot studies (see Preliminary Studies; Study #1 and Study #5). If the subject waited to start tablets until after being symptomatic for several days, the RA will discuss the need to commence treatment earlier in the menstrual cycle. The RA will ensure that she has sufficient medication.
Subjects who have experienced worsened suicidal ideation will be assessed immediately at this visit. If they have intent to harm themselves, they will have an acute intervention such as hospitalization. For subjects who go into the open label extension treatment, this will constitute the last cycle and final study assessments will be completed.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 20
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Visit 3 (Beginning of Treatment Cycle 3): Subjects will be seen on Day 5-7 of menses during Cycle 3. The RA will count tablets and will download information from the MEMS cap. Rating and safety scales will be administered by the blinded rater (PMTS, CGI-I, IDS-C, HASS-II, EPT) to determine efficacy as well as treatment impact on health-related quality of life (Q-LES-Q). Data will be entered and checked using TrialDB data quality control procedures. Women who have not achieved full response (CGI-I of 1) may have their dose increased to a maximum of 100 mg, as described above for Visit 2. Other strategies to ensure that treatment has been maximized include a review of pill taking and symptoms to ensure that treatment has occurred for the entire symptomatic phase. Despite taking 100 mg and initiating treatment to encompass all symptomatic days, some subjects may feel they have had insufficient benefit and wish to withdraw from the study. They will be referred to the open-label extension phase unless they have developed a contraindication to sertraline (eg. mania, rash). Subjects who leave the study (per protocol treatment and do not enter extension) will continue being followed until the 6-month point since this is an intent to treat trial. The RA will ensure that she has sufficient medication for the next cycle. Subjects will be reimbursed $10 for completing this second double- blind treatment cycle and $5 for mileage costs, for a total of $15.
Visit 4 (Beginning of Treatment Cycle 4): Rating and safety scales will be administered (PMTS, CGI-I, IDS-C, HASS-II, EPT) by the blinded rater to determine efficacy. The RA will conduct a tablet count and will download information from the MEMS cap. The RA will ensure that she has sufficient medication for the next 3 cycles. Since clinically, some women require a dosage increase after several months of treatment, we allow a titration to a maximum dose of 100 mg at this visit as well. In the event that a woman requires additional medication, she will have tablets refilled from a stock bottle that has her number. She will continue taking tablets from the MEMS cap bottle and record when she took tablets onto her daily rating form. Subjects will be reimbursed $10 for completing this treatment cycle and $5 for mileage costs for a total of $15
Visits 5,6 (Beginning of Treatment Cycles 5, 6): It is anticipated that a subject’s dose will be stabilized by this point. In previous work, we found that phone visits, after a period of stabilization, minimizes subject burden and increases interest in participation. Accordingly, subjects will be contacted by telephone, and ratings (PMTS, CGI-I, IDS-C, HASS-II, EPT) will be conducted by the blinded rater for the subsequent 2 menstrual cycles. Data will be entered and its accuracy monitored by TrialDB quality control procedures. Error checks will be run and missing data or errors will be corrected immediately. We will ask subjects to send their daily ratings by mail and will remind them about this during phone contact. Any subjects having side effects or problematic symptoms that require more active management (e.g., suicidal thoughts or severe symptoms that do not resolve in the follicular phase) will be scheduled for a face-to-face visit within 3 days of the phone visit. The same procedures as outlined above will be followed for suicidal subjects or those wishing to discontinue. Subjects will be reimbursed $10 for completing each double-blind treatment cycle.
Visit 7: Final ratings (see Visit 3) will be collected at this visit. Subjects completing the randomized portion of the trial will be referred to the open label extension phase for 3 months of single-blind daily treatment from the PI or co-Investigator. Subjects will be reimbursed $10 for completing this sixth double-blind treatment cycle and $5 for mileage costs, plus receive a randomized control trial completion bonus of $150. Subjects who attend all visits, completing the baseline and all six randomized cycles will thus receive a $70 reimbursement for visits, $25 in mileage costs for 5 in- person office visits, plus a bonus of $150 for completing the study.) APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 21
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Open-Label Extension Phase (Visits 8 & 9): All study completers as well as subjects who felt their response was insufficient or were unable to tolerate the on-off effects of symptom onset treatment will be eligible for the extension phase. Excluded will be acutely suicidal subjects who require additional intervention and subjects for whom either additional (eg psychotic symptoms) or alternative (eg for mania or substance dependence) treatment is required. Subjects entering the extension phase must all have had their final study visit and assessments. Visits in the extension phase will occur 1 and 3 cycles after extension phase entry but additionally as needed, for example to assess safety. Subjects will be reimbursed $10 for completing each open-label extension treatment cycle ($20 total) and $5 in mileage costs for 2 in-person visits ($10 total). The maximum possible reimbursement for this entire study for subjects who complete all randomized visits and extension phase visits is $325 ($10 for the baseline visit, $50 for completion of 2 full cycles of DRSP ratings, $60 for six double blind visits, $150 randomized control trial study completion bonus at visit 7, $20 for 2 extension phase visits, and $35 for mileage costs). Subjects will receive active treatment with sertraline dosed flexibly between 12.5 mg and 100 mg per day in the extension phase. At Visit 8, the DRSP, IDS, PMTS, CGI and HASS II will be collected. For Visit 9, the same measures that were used at Visit 3 will be collected, with the exception of the ease of pill taking questions (EPT) which will no longer be pertinent as subjects will be receiving continuous open-label dosing. Subjects will be referred to follow-up treatment with an outside clinician after completion of the extension phase. While data collected from the extension protocol will not be used for the Primary Aims of the current protocol, they will be used to estimate effect sizes for daily treatment should a future study comparing symptom onset and daily dosing be indicated.
Quality Control
Prior to beginning of the study and randomly throughout the study, we will invite women to be videotaped while we administer the MINI, PMTS and IDS-C. This will allow us to provide training and practice to the evaluators and will allow us to determine initial and ongoing reliability for our ratings. The videotapes will be kept entirely confidential and will be viewed only by the doctors and staff who work on this project. In addition, they will be destroyed upon expiration of consent. We will obtain separate consent for videotaping.
The principal investigators at the three sites will view and obtain consensus on ratings for the first 12 videotapes. The first 6 tapes will be used for training at their respective sites. An additional 6 tapes will be rated by each of the blinded evaluators. By the 6th tape, inter-rater reliability must be at least 0.8 for each of the evaluators and for each instrument. Those who fail to achieve this level of reliability will under-go retraining and will rate additional tapes until they achieve acceptable reliability.
Please note: Investigators should take care to distinguish clearly between any procedures that are experimental and those that are part of subjects’ standard clinical care.
4. Statistical Considerations: Describe the statistical analyses that support the study design. This section should include:
. The number of subjects expected to enter the study. . A statement about the statistical power of the study to test the major hypothesis. . A summary of the plans for data analysis. APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 22
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
300 women will be randomized equally across the three recruitment sites; therefore, Yale University will randomize 100 subjects.
Statistical Analyses: Descriptive statistics and graphs will be used to summarize the data on all randomized subjects. Baseline demographic and clinical characteristics for the 2 randomized groups will be compared using chi-square tests for categorical variables, and using t-tests or Mann-Whitney tests for continuous variables. All continuous variables will be examined for adherence to the normal distribution using normal probability plots and Kolmogorov-Smirnov tests. If normality is not satisfied and transformations do not help with achieving normality, alternative analytic strategies will be considered such as generalized estimating equations or nonparametric methods for repeated measures analysis (117). Transformations will be applied if needed. Dropouts and completers will also be compared on baseline characteristics using the same approach. All statistical tests will be performed at a two-tailed alpha level of 0.05. All statistical analyses will be performed in SAS Version 9.1 or later.
General Data Analysis Strategy: Mixed-effects regression models (118) will be used for the comparison of the two treatment groups (sertraline vs placebo) to address the specific aims below. These models allow for different numbers of observations per subject, use all available data on each subject, and are unaffected by randomly missing data. They also provide flexibility in modeling the correlation structure of the data. In each model, we will include fixed effects of time, treatment and time by treatment, and random subject effects. Since preliminary data on the type of trends over time (linear, quadratic) are not available, we will first treat time as a categorical predictor and will then test for polynomial trends over time. We will consider several different error structures (e.g., AR1, independence, unstructured) and select the best fitting one based on information criteria. SAS PROC MIXED will be used for the mixed-model analyses. Data on women who discontinue the study protocol but have subsequent measurements will be used in the primary analyses as randomized. All data on women who are lost to follow-up during the study will also be used in the model. In two studies from the Yale site that endured for 5-6 months and enrolled 130 subjects, 73% on active SRI treatment and 81% of placebo completed. If there are concerns of informative dropout and/or informative intermittent missing data, we will use pattern mixture models (119) to perform sensitivity analyses to our main analyses. We will also perform exploratory analyses with compliance as an additional covariate.
Hypothesis 1a: We hypothesize that subjects who take active medication during their symptomatic days will derive significantly greater symptomatic benefit than subjects who take placebo capsules in the same manner. The PMTS will be the primary outcome measure for this specific aim. We will use mixed effects models to test this hypothesis. We anticipate that the treatment by time interaction will be statistically significant. If the interaction is statistically significant, we will perform follow-up comparisons between the treatment groups. If the relationship is linear over time, we will estimate separate slope coefficients for each group and provide estimates and associated 95% confidence intervals of the change over time. If the relationship is not linear, we will perform multiple comparisons between the groups at each time point.
Hypothesis 1b: We hypothesize that subjects on active medication will improve over the course of the study (i.e., consecutive cycles) as they become better able to estimate when to start medication. This hypothesis will be tested based on the same models as in Hypothesis 1a. A linear trend with a statistically significant negative slope in the active group will be considered supportive of this
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 23
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
hypothesis. The placebo group may experience an initial improvement in symptoms (a negative slope) but may level off or worsen; therefore, a curvilinear trend over time is expected.
Hypothesis 2: We hypothesize that the rate of discontinuation symptoms will be no higher with active medication than with placebo. The primary outcome measure for this specific aim will be a subscale of symptoms from the DRSP but we will conduct a secondary analysis to confirm findings using items from the DRSP and addendum that comprise the Michelson SRI withdrawal checklist (101). Because of the additional symptoms in the Michelson SRI withdrawal list, it is possible that this measure will be better able to detect differences in discontinuations symptoms, but we do not have pilot data to support this. To test the 2nd hypothesis, we will first test the equivalence of the treatment and control group at each time point. Mixed effects models will then be fitted to estimate the trends in discontinuation symptoms over time. We anticipate that both the treatment and treatment by time interaction will be non-significant in this analysis. However, to assess the clinical significance of the observed differences in withdrawal symptom severity, we will estimate the difference between active treatment and placebo at each cycle with 95% confidence intervals.
Exploratory Hypothesis 3a: We hypothesize that subjects who take active medication during their symptomatic days will show more improvement in functional impairment than subjects who take placebo capsules in the same manner. The outcome measure for this hypothesis will be the sum of the three impairment items for the Daily Record of Severity of Problems. We will use mixed effects models to test this hypothesis. We anticipate that the treatment by time interaction will be statistically significant. If the interaction is statistically significant, we will perform follow-up comparisons between the treatment groups as outlined for specific aim 1.
Exploratory Hypothesis 3b: We hypothesize that subjects who take active medication during their symptomatic days will show more improvement in quality of life than subjects who take placebo capsules in the same manner. This hypothesis will also be tested using mixed effects models as specified above. The Q-LES-Q scale will be the outcome measure.
Exploratory Hypothesis 4: We hypothesize that women with lower symptom scores on the DRSP, women with no suicidal symptoms and women with less Table 3: Sample Size Requirements* severe physical symptoms will be more likely to respond Effect Size Dropout N Per Group Rate to active treatment than will be women with higher scores 0.4 10% 112 on the DRSP. To test these moderator hypotheses, we will 0.4 20% 125 follow the approach of Kraemer et al. (120). We will fit 0.4 30% 143 three separate regression models for each potential 0.5 10% 72 moderator. Treatment, potential moderator, and their 0.5 20% 80 interaction will be included in the models as fixed effects 0.5 30% 92 and CGI improvement scale symptom score will be the 0.6 10% 50 response measure. We anticipate that there will be a 0.6 20% 57 significant interaction between treatment and potential 0.6 30% 65 0.7 10% 38 moderator. We expect to see greater effect of treatment at 0.7 20% 43 lower DRSP baseline scores, in the absence of suicide 0.7 30% 49 symptoms (as determined by the HASS-II) and for less 0.8 10% 29 severe physical symptoms. 0.8 20% 33 0.8 30% 38 Exploratory hypothesis 5: We hypothesize that treatment *Power=.80; Alpha=.05; 2-Sided T-Test will improve physical symptoms and that improvement in physical symptoms will be less than improvement in mood APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 24
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
symptoms. For the first part of this hypothesis we will use the same model as in Hypothesis 1 with the physical symptom items on the PMTS as the response variable. For the second part we will calculate change scores for improvement in physical symptoms and improvement in mood items (as measured by the DRSP) and we will test whether the mean improvement in physical symptoms is less that the mean improvement in mood symptoms.
Power Analyses and Sample Size Determination: The sample size was based on the power calculations for the hypotheses of Primary Aims 1 and 2. The estimated effect size for the difference between active and placebo in our pilot data from the cross-over study in the first cycle was d=0.8 (d is the absolute means difference over the standard deviation) (121). In studies of half-cycle and full-cycle SSRI administration, the estimated effect sizes between treatment and placebo were 0.4 (Study 3 in Preliminary Studies). Since it is difficult to predict the exact shape of the change over time, we conservatively base the power calculation on the first and last time points in the study and assume 10-30 % dropout. This estimate is in line with the dropout rate of 23% on active drug and 19% on placebo seen in our other studies that endured for 6 cycles of treatment. Table 3 provides sample size estimates for the between-group comparison of mean changes from baseline to Cycle 6 of the study (Hypothesis 1a), assuming effect sizes between 0.4 and 0.8 and different dropout rates. Using a standard deviation estimate of 6 and within-subject correlation of 0.3 based on our pilot data, effect sizes of 0.4 and 0.8 correspond to mean difference of 3.5 and 7 points, respectively, on the PMTS.
We also observed a very large effect size in our pilot data based on CGI-I scores (70% of the subjects randomized to sertraline were classified as responders; 10% of the subjects randomized to placebo were considered responders). To detect such an effect as statistically significant with 80% power at 0.05 level using a chi-square test, 9 subjects per group are needed. With the highest estimated effect size based on the PMTS measure (n=143 per group) and assuming 30% dropout, we can detect, for example, the following differences in proportions with at least 80% power: 10% vs 25%, 20% vs 38%, 30% vs 50%. These are small but clinically meaningful differences.
For Hypothesis 2, we consider a 20-30% increase in discontinuation symptoms in the treatment group as compared with the placebo group to be clinically meaningful. This is a very modest increase in symptoms due to discontinuation. Since lower mean discontinuation symptom scores in the active group than in the placebo group are supportive of our hypothesis, we base the power calculation on a one-sided non-inferiority test. In our pilot data, the mean discontinuation DRSP subscale (9 items) were 14.3 (SD=4.6) for paroxetine and 16.54 (9.3) for placebo. Table 4 provides sample size estimates for this hypothesis based on different combinations of placebo group means, standard deviations and upper equivalency bounds. We used the most conservative estimate for our sample size estimate. As Table 4: The Sample Size Requirements* can be seen from Mean SD Upper Bound** Sample Size Sample Size Tables 3 and 4, the test (Placebo) without dropout with 30% dropout of the second 16 5 19.2 (20%) 33 47 hypothesis requires the 16 5 20.8 (30%) 15 21 16 9 19.2 (20%) 105 150 largest sample size 16 9 20.8 (30%) 47 76 estimate. We will 17 5 20.4 (20%) 29 41 recruit 100 women per 17 5 22.1 (30%) 13 19 17 9 20.4 (20%) 93 133 site for a total of 300 17 9 22.1 (30%) 42 60 women. With this *Power=.80; Alpha=.05; 1-Sided T-Test sample size estimate, **% Increase From Placebo Mean we will have greater than 80% power if our APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 25
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
drop out rate is higher than we usually experience and the effect size is lower than our pilot data with paroxetine (0.9 in our study but 0.4 in other datasets). If we have an effect size consistent with our pilot data (0.9) and a drop-out rate of 20%, then power will be over 90%.
Power for Secondary Aims: The analysis plan for Specific Aim 3 and Specific Aim 5 will be the same as for Specific Aim 1. Hence, with the chosen sample size, we will have more than 80% power to detect a slightly smaller than medium effect size (d=0.4) for the functional impairment items on the DRSP and on the QOL measure, and for the physical symptoms item on the DRSP. Such effect sizes are considered clinically meaningful. For the moderator analyses in Specific Aim 4 with the chosen sample size of 300 women, we have more than 80% power to detect an effect size f=0.2 (f=0.25 is considered medium) for the F-test of the interaction between treatment and the potential moderator in ANOVA. Future Directions: Symptom-onset dosing for PMDD is becoming increasingly common in clinical practice despite the lack of placebo-controlled study to determine its efficacy. Completion of the work described in this application, whether the hypotheses are confirmed or refuted, will have significant impact on clinical care of women with PMDD. The findings of this study, designed to reflect real-life implementation (i.e., symptom onset), will assist in developing treatment guidelines for women with PMDD. Such guidelines, informed by our work, including our moderator analyses, may be tested in actual practice settings to determine the overall effectiveness of this modality compared to other treatments and other methods of medication administration. Finally, supplemental studies (e.g., mediator-moderator studies using biological markers) will be possible upon the completion of Dr. Altemus’ work and with the help of a supplemental application.
SECTION V: RESEARCH INVOLVING DRUGS, DEVICES OR BIOLOGICS
Please note: protocols using chemicals, hormones, other natural substances, or devices not regulated by the U.S. Food and Drug Administration (FDA) must still complete this section of the application form. Based upon the information provided in items 1-4, the HIC will determine whether an Investigational New Drug (IND) or Investigational Device Exemption (IDE) application must be submitted to the FDA.
1. Identification of Drug, Device or Biologic: What is the name of the drug, device or biologic being used? Please identify whether FDA approval has been granted, and for what indication(s).
Sertraline (Zoloft®) is an FDA-approved medication for the treatment of PMDD, as well as depressive and anxiety disorders.
2. Background Information: Please provide a description of previous human use, known risks, and data addressing dosages, intervals, routes of administration, and any other factors that might influence risks.
Typical dosing for the diagnosis of PMDD range from 50mg daily to 100mg of Sertraline daily, and this is the dosing range for treatment in this study.
Discontinuation symptoms can occur with short-acting SRIs and serotonin-norepinephrine reuptake inhibitors. Because prior studies find discontinuation symptoms within 5 days of stopping capsules, we will evaluate this time interval. However, because subjects collect symptoms daily throughout the trial, we can also explore other intervals as well as possible diminution in these effects over consecutive menstrual cycles.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 26
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Suicidal ideation, in severe cases, can be a symptom of Premenstrual Dysphoric Disorder. With caution, such study subjects will be monitored in this study. Subjects who experience any worsening of suicidal ideation will be assessed within 24 hours. If they actively intend to harm themselves, they will have an acute intervention such as hospitalization. For subjects who are in the open label extension treatment, this will constitute the last cycle and final study assessments will be completed.
3. Source: Please identify the source of the drug, device or biologic to be used.
Study medication will be provided by Pfizer Pharmaceuticals, the maker of this drug. They will also provide matching placebo, encapsulated to look identical to the Zoloft.
4. Preparation and Use: Please describe the method of preparation, storage, stability information, and for parenteral products, method of sterilization and method of testing sterility and pyrogenicity.
Study medication (active and placebo) will be stored in a locked exam room cabinet at a steady office temperature (between 59-86 degrees). This medication does not need to be refrigerated.
5. Use of an Investigational Drug, Device or Biologic: Protocols which utilize a drug, device or biologic not approved by the FDA must provide the following information. N/A
. What is the Investigational New Drug (IND) or Investigational Device Exemption (IDE) number assigned by FDA?
. For IDE’s: Did the FDA approve this IDE as a Category A (experimental/investigational) or as a Category B (non-experimental/investigational)?
. Who holds the IND or IDE?
. Is the drug or device provided free of charge by the Sponsor? Yes _X__ No ___
6. The FDA requirements to qualify for an exemption from filing an IND are as follows [21 CFR 312.2(b)]: Exemptions – The clinical investigation of a drug product that is lawfully marketed in the United States is exempt from the requirements filing for an IND if ALL the following apply: i. The investigation is not intended to be reported to FDA as a well controlled study in support of a new indication for use nor intended to be used to support any other significant change in the labeling for the drug; ii. If the drug that is undergoing investigation is lawfully marketed as a prescription drug product, the investigation is not intended to support a significant change in the advertising for the product; iii. The investigation does not involve a route of administration or dosage level or use in populations or other factor that significantly increases the risks (or decreases the acceptability of the risks) associated with the use of the drug product; iv. The investigation is conducted in compliance with the requirements for institutional (HIC) review and with the requirements for informed consent of the FDA regulations (21 CFR Part 50 and 21 CFR Part 56); v. The investigation is conducted in compliance with the requirements regarding promotion and charging for investigational drugs and vi. The investigation is not emergency research (21 CFR Part 50.24).
7. Use of Placebo: Does any part of the study involve placebo? Yes _X__ No ___ a. If yes, please address each one of the following: i. The safety and efficacy of other available therapies (if any). Sertraline and fluoxetine are FDA approved therapies for the medical condition of PMDD, in the continuous daily-dosing format. Over-the-counter remedies have been found helpful for less severe cases of premenstrual syndrome. APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 27
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
ii. The maximum total length of time a participant may receive placebo on study. A participant may be randomized to take placebo for the premenstrual phase (as many as 14 days) of 6 menstrual cycles. Thus, a participant may take a maximum of 84 days of placebo study pills. iii. The greatest potential harm that may come to a participant as a result of not receiving effective therapy (immediate or delayed onset). Although some subjects perceive short-term benefit from taking a placebo, there is no reason to expect that a subject’s PMDD would improve while taking a placebo. Potential harm includes the likelihood that symptoms will not improve, and thus recovery will be delayed while she participates in this study. There is no reason to believe that PMDD symptoms would worsen while taking a placebo. iv. Protocols in place to safeguard participants receiving placebo. At every study visit the subject and the study coordinator will evaluate the subject’s willingness to continue participating as a study subject. Participants have the right to withdraw from this study at any time. The Principal Investigator also has the right to remove a subject who does not evidence sufficient clinical improvement while taking study medication. Further, whether taking placebo or the active study medication, if a patient should show clinical deterioration at any point in the study, or if, in the psychiatrist's opinion, she requires a different treatment for her condition, she will be removed from the study and closely monitored by the psychiatrist per standard medical care outside of study participation.
SECTION VI: HUMAN SUBJECTS
1. Recruitment Procedures: How will potential subjects be identified, contacted and recruited? Attach copies of any recruitment materials – such as flyers, telephone scripts, or introductory letters – that will be used.
Please note that a researcher may not use an individual’s Protected Health Information (PHI) for recruitment into research without first obtaining an authorization from the individual, or a Waiver of Authorization from the HIC. A treating provider does, however, have the option to: . Discuss with his/her own patients the option of enrolling in a study. . Obtain written authorization from the patient for referral into a research study. . Provide background information about the study to the patient so that the patient can initiate contact with the researcher. . Provide the individual’s PHI to a researcher without authorization when the researcher has obtained an approved Waiver of Authorization for recruitment purposes from the HIC.
If PHI will be accessed without subject authorization, please state whether any member of the research team has an existing clinical relationship with the potential subject. Researcher-clinicians are permitted to access the PHI of their own patients, or patients of co-investigators listed on the protocol, for recruitment purposes. Alternately, please submit a completed Yale University Request for HIPAA Waiver of Authorization for Research Form (available at http://info.med.yale.edu/hic/hipaa/documentation.html). For further information, see the Yale HIPAA website at http://info.med.yale.edu/hipaa/.
Subjects may be connected to our study by a number of methods. Our PMS Research Program’s flyers are distributed throughout the community in areas where menstruating women are likely to frequent. Collaborative gynecologists who know of our work often are willing to post a poster in their waiting room or another high-traffic area of their clinic. Women hear of our services through word of mouth, from other study subjects. We will also purchase mailing lists from InfoUSA in order to do mass mailings at least 4 times per year until we reach are recruitment goal. For each mass mailing, a letter of introduction will be mailed to between 8,000 and 10,000 households in various towns throughout New Haven County along with a secure self- APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 28
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
addressed postage free envelope. Women who are interested in hearing more about our study will provide their contact information on the bottom portion of the letter and return to us using the envelope provided. Lastly, subjects who complete other PMDD treatment trials of ours may choose to continue on into this study. A copy of our poster and flyer (already approved for other studies) are attached for HIC review.
2. Inclusion/Exclusion Criteria: What are the criteria for subject inclusion or exclusion? How will eligibility be determined, and by whom?
Inclusion Criteria: Female outpatients who: 1. Are between the ages of 18 and 48, 2. Are menstruating and have cycles between 21 and 35 days (inclusive), 3. Meet DSM-IV criteria for PMDD, a. For a definitive PMDD diagnosis, DSM-IV stipulates 2 cycles of daily ratings to confirm luteal symptoms. Based on daily symptom scores, subjects must have a minimal average luteal phase score of mild (≥3 on a 6-point scale) for at least 5 PMDD symptoms during the 5 most symptomatic of the final 7 luteal phase days and the first 2 days of menses onset, and the average follicular phase core must not be >2 on these same items. 4. Have exhibited symptoms of PMDD in at least 9 of 12 menstrual cycles during the year prior to screening, 5. In the investigators opinion, are using an adequate method of birth control, and 6. Have provided verbal consent and then must have read and signed an Informed Consent form explaining this study.
Exclusion Criteria: Any potential candidate who:
1. Fulfills current (within the last 6 months) MINI DSM-IV criteria for major depressive episode, a psychotic disorder (schizophrenia, schizoaffective, etc), bulimia nervosa, anorexia nervosa, or a substance abuse disorder, and in the case of alcohol abuse, the candidate does not believe she can moderate her use or refrain from alcohol consumption during the study period. 2. Fulfills MINI DSM-IV criteria for a substance dependence disorder within the last 6 months. 3. Fulfills MINI DSM-IV criteria for bipolar disorder, lifetime. 4. Shows follicular phase symptoms consistent with a diagnosis of major depression, or bipolar disorder on DRSP forms. 5. Has a severe, clinically significant co-existing condition that in the Investigators’ opinion, renders the patient unsuitable for the study, 6. Evidences a significant risk of suicide, 7. Is taking ongoing antidepressant or other psychotropic medication, 8. Has a history of hypersensitivity or adverse reaction to sertraline, 9. Is lactating, pregnant, or who is planning to become pregnant during the course of the study. 10. Is undergoing treatment with a depot hormonal preparation or any other medication that would lead to lack of menses or markedly irregular menses. 11. Is using a hormonal contraceptive pill or hormonal device for less than 6 months. 12. Is taking a hormonal contraceptive pill that includes 20 micrograms of ethinyl estradiol and 3 mgs of drosperinone (Yaz® or Ocella® ). 13. Has been in individual psychotherapy or individual counseling for 3 months or less at the time of the initial screen.
Potential candidates may also be excluded from the study if they are taking any of the following medications: Daily NSAID/aspirin – bleeding APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 29
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
oxycodone/tramadol – decrease in asthesia opiate use, except for acute accident or medical procedure astimizole – cardiotoxity, Torsade de Pointe fentanyl, levomethadyl – cardiotoxicity flecanmide (increased flecanamide levels) (dex)fenfluramine – serotonin syndrome droperidol – serotonin syndrome phenytoin – phenytoin toxicity furazolidone/linezolid (MAO like) oral steroids (mood altering) DHEA-mania benzodiazepine >2/week antidepressant (participants must be off antidepressant medications for two months of symptom charting, with cessation occurring not less than 2 days before first follicular phase of the charting cycle.) antipsychotic mood stabilizer 5-HTP calcium >800mg/day Yasmin, Yaz-24, Ocella Spironolactone, other diuretics OTC menstrual agents (pamprin, midol) Chaste tree
Caution Triptans –serotonin syndrome Benzodiazepines – increased sedation Cimetidine – increased sertraline level Darunavir/ritonavir - reduced sertraline level Erythromycin – serotonin syndrome Methadone –increased methadone level Metcloprimide (reglan) – EPS risk
Potential Candidates may also be excluded from the study if they have any of the following exclusionary medical conditions: Exclusion Cancer (skin cancer will be included, except melanoma) Pancreatitis Hx of elevated prolactin Impaired liver or kidney function Autoimmune disease (MS, lupus, rheumatoid arthritis) Chronic pain requiring meds at least 5 days per week Untreated or unstable migraine, adrenal, thyroid, polycystic ovarian syndrome, diabetes, AIDS or seizure disorder.
3. Subject Population: Provide a detailed description of the proposed involvement of human subjects. Describe the characteristics of the subject population, including their anticipated number, age range and health status.
The selection of subjects should be equitable. Generally speaking, the subject selection should reflect a reasonable cross- section of the population that is being studied. In research that requires a more restricted population, the rationale for this need should be fully justified. Investigators must also provide scientific justification for the exclusion of underrepresented populations such as women, children, or minorities.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 30
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Subject population includes 18-48 year old females, as noted above in inclusion/exclusion criteria. 100 subjects will be enrolled at each of three sites. Our research program has a proven track record enrolling a diverse range of women, as measured by age, race and ethnicity. Relationships with community gynecologists in a diverse range of area towns and city neighborhoods, including particularly strong collaborative relationships with area Community Health Centers, suggest a high likelihood of receiving a diverse range of patient referrals.
4. Vulnerable Subjects: Certain populations are considered to be vulnerable and require special protections when asked to participate in a research study. (Vulnerable populations include, but are not limited to pregnant women, fetuses, human embryos, prisoners, children, and cognitively impaired individuals or persons with questionable capacity to consent. Others which may require special consideration include elderly persons, economically disadvantaged persons and educationally disadvantaged persons.) Children include all persons who have not attained the legal age for consent to treatments or procedures involved in the research under the applicable law of the jurisdiction in which the research will be conducted (the age of majority is 18 years in the Connecticut). Parental permission and the child’s assent is required for participation in the study. (Assent is defined as “a child’s affirmative agreement to participate in research” and should be sought in addition to parental permission when the minor subject is sufficiently mature to understand the nature of his or her participation in a research study.) Please refer to the HIC Guidelines for Investigators at http://www.med.yale.edu/hic/forms/forms/guidelines.pdf
Will vulnerable subjects be enrolled in the study? If so, identify the vulnerable population and provide a justification for their involvement. Also address any additional safeguards necessary to protect the rights and welfare of vulnerable subjects.
Vulnerable subject populations are not targeted for this study. Women must be age 18 or over to sign informed consent and participate in this study. Pregnant women are excluded from this study, and urine testing is conducted at study visits to confirm that they are not pregnant. The study and alternatives to participation will be fully explained by qualified staff to all potential participants so that informed consent may be obtained. On any occasion when the cognitive capacity of a subject is in question, she will not be enrolled in this study. The study is likely to enroll women or all ranges of education and financial security; all women will receive the same thorough consenting process.
SECTION VII: CONSENT/ASSENT PROCEDURES
1. Consent Personnel: Please list all personnel who will be obtaining consent.
Consent will be obtained by Joanne Cunningham, PhD (Study Coordinator)or Kimberly Yonkers, MD (Principal Investigator).
2. Assessment of Capacity to Consent: For research involving subjects with limited decision-making capacity, how will the capacity to consent be assessed?
The Principal Investigator will meet personally with the potential subject to assess her capacity to provide consent. If there is any question about a potential subject’s ability to understand the purpose of procedures of the study, she will not be enrolled.
3. Process of Consent: Describe the setting and conditions under which consent will be obtained, including any steps taken to enhance subjects’ independent decision-making.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 31
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Informed Consent is conducted in our research office in a private office at the initial screening visit. Subjects must chart their symptoms for two menstrual cycles (roughly two months) before commencing study medication, during which time there is telephone and in-person contact with the Research Associate coordinating the study and opportunity to review questions the subject may have about her participation in the study. Details of the trial are again reviewed at Visit 1 before the subject begins taking study medication.
4. Non-English-Speaking Subjects: For research involving non-English-speaking subjects, fully explain provisions in place to ensure comprehension. In addition, please submit translated copies of all consent materials.
Patients who are not fluent in English are not enrolled in this study.
5. Parental Permission and Assent: For research involving minors, please explain how parental permission and child assent will be obtained.
N/A
6. Documentation of Consent: Specify the forms that will be used among the following: adult consent form, parental permission form, LAR (Legally Authorized Representative) permission form, adult assent form, adolescent assent form (ages 13-17 inclusive), child assent form (ages 7-12 inclusive), and information sheet. Copies of all forms should be appended to the protocol, in the same format that they will be given to subjects.
Adult Consent Compound Authorization only.
7. Waiver of Consent: Will you request either a waiver of consent, or a waiver of signed consent, for this study? If so, please address the following: No.
Waiver of consent: a) Does the research pose greater than minimal risk to subjects? b) Will the waiver adversely affect subjects’ rights and welfare? c) Why would the research be impracticable without the waiver? d) How will pertinent information be returned to subjects, if appropriate at a later date?
Waiver of signed consent: a) Does the research pose greater than minimal risk? If so, does a breach of confidentiality constitute the principal risk to subjects? b) Would the signed consent form be the only record linking the subject and the research? c) Does the research include any activities that would require signed consent in a non-research context?
8. HIPAA Authorization: If the research involves the creation, use or disclosure of PHI, separate authorization is required under the HIPAA Privacy Rule. Please provide the HIPAA Research Authorization Form and/or a request for waiver of HIPAA authorization. (For further information, see the Yale HIPAA website at http://info.med.yale.edu/hipaa/).
SECTION VIII: PROTECTION OF RESEARCH SUBJECTS
1. Risks: What are the reasonably foreseeable risks, discomforts, or inconveniences associated with participation in the research?
Please note: Potential research risks include more than physical harm; risks may also include, for example, emotional or psychological harm, risk of social stigmatization, economic or legal risk. APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 32
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
There may be mild discomfort from the interviews or questionnaires that ask personal questions but this is not expected to be problematic. Discontinuation risks will be reviewed and discontinuation symptoms will be measured carefully as part of this study. Patients will be monitored for worsening of depressive symptoms and suicidality closely by staff throughout the study. Because patients may have side effects when they are placed on active medication, they will be closely monitored for side effects.
2. Minimizing Risks: How will the above-mentioned risks be minimized?
Women can request not to answer any question which makes her feel uncomfortable; study staff will determine on a case-by-case basis if this lack of response compromises her ability to participate in the study. At all times, women will be reminded that their participation in this study is voluntary and they have the right to withdraw at their discretion.
The risk of medication discontinuation will be minimized by dosing the lowest effective dosage (50mg) of study medication and by titrating study pills up and down over several days, for those taking the higher dose (100mg).
Prior to study enrollment, a Mini International Neuropsychiatric Interview, a structured interview is conducted on each subject to screen for psychiatric disorders. All subjects with a diagnosis of major depressive disorder will be excluded from the study. All subjects considered too ill or at risk will not be included in the study; therefore, women only with a diagnosis of PMDD will be eligible to enroll. During study participation, subjects are seen once a month and are thoroughly reevaluated for depression and suicidal ideation at every visit. If a patient should show clinical deterioration at any point in the study, or if, in the psychiatrist's opinion, the patient requires a different treatment for her condition, she will be removed from the study and closely monitored by the psychiatrist per standard medical care outside of study participation.
We will also ask for permission to contact a family member or close friend if we feel that the patient is at risk during the time she is in the study. Consent to contact this person will be limited to emergency interventions, which includes the onset and/or worsening of depressive or suicidal symptoms. Subjects will be given an emergency card with phone numbers for the study team so that they or another person can contact the study team 24 hours a day if there is a general medical or psychiatric emergency and information about study medication is necessary.
3. Data and Safety Monitoring Plan: Please include a Data and Safety Monitoring Plan (DSMP) that includes an explicit statement of overall risks, addresses attribution and grading of adverse events and describes procedures for monitoring the ongoing progress of the research and reporting adverse events. For more information, please see the HIC website: http://info.med.yale.edu/hic/templates/DSMP.doc.
The principal investigator will be responsible for monitoring the data, assuring protocol compliance, conducting the safety reviews, and the specified frequency of the reviews at a minimum of every 6 months (including when reapproval of the protocol is sought). During the review process, the principal investigator (monitor) will evaluate whether the study should continue unchanged, require modification/amendment, continue or close to enrollment. Either the principal investigator, the HIC or NIMH have the authority to stop or suspend the study or require modifications.
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 33
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Study staff will review subjects' safety at each subject visit and telephone rating session and will present subjects' clinical status and adverse experiences at the weekly Study Personnel meeting at each site. Entrance criteria of all subjects are reviewed at this meeting. The referral to appropriate care for subjects who have had a clinical deterioration will be monitored at this meeting.
The risks associated with the current study are deemed moderate for the following reasons because: We do not view the risks associated with the receiving a placebo as minimal, and given the now established safety and validity of the FDA-approved medication sertraline, we do not view the proposed studies as high risk. Patients randomized to the active study medication, Sertraline, are receiving an FDA-approved medication that has established efficacy in the treatment of PMDD. The medication has been available by prescription for over 10 years for treatment of other disorders and substantial data support its safety. Although we are prescribing sertraline at a lower exposure than has been established as safe, sertraline is associated with side effects that require monitoring. Further, the integrity of the data collection process requires monitoring.
Although we have assessed the proposed study as one of moderate risk, the potential exists for anticipated and/or unanticipated adverse events, serious or otherwise, to occur since it is not possible to predict with certainty the absolute risk in any given individual or in advance of first-hand experience with the proposed study methods. Therefore, we provide a plan for monitoring the data and safety of the proposed study as follows:
Adverse events will be monitored for each subject participating in the study and attributed to the study procedures / design by the principal investigator, Dr. Yonkers, according to the following categories: a.) Definite: Adverse event is clearly related to investigational procedures(s)/agent(s). b.) Probable: Adverse event is likely related to investigational procedures(s)/agent(s). c.) Possible: Adverse event may be related to investigational procedures(s)/agent(s). d.) Unlikely: Adverse event is likely not to be related to the investigational procedures(s)/agent(s). e.) Unrelated: Adverse event is clearly not related to investigational procedures(s)/agent(s).
The following scale will be used in grading the severity of adverse events noted during the study:
1 Mild adverse event 2 Moderate adverse event 3 Severe unanticipated adverse event resulting inpatient hospitalization or prolongation of existing hospitalization, a persistent or significant disability/incapacity, or a congenital anomaly/birth defect. 4 Life-threatening adverse event 5 Fatal adverse event
The investigator will report the following types of adverse events to the HIC; a) serious AND unanticipated AND possibly, probably or definitely related events; and b) anticipated adverse events occurring with a greater frequency than expected. These adverse events will be reported to the HIC within 48 hours of it becoming known to the investigator, using HIC Form 6A. Adverse events will be deemed serious in nature if graded as 3 or higher according to the scale in item #4 above. All non- serious, unexpected adverse and all non-serious, expected adverse effects of study medication will be reported to the DSMB on a semi-annual basis (see below). APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 34
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
Dr. Yonkers will conduct a review of all adverse events upon completion of every study subject. The principal investigator will evaluate the frequency and severity of the adverse events and determine if modifications to the protocol or consent form are required.
At each phone and face-to-face visit, the research associate will collect concomitant medication and adverse events (AEs) information. The standard AE form includes information on the onset and termination of the event, severity (mild, moderate or severe), whether corrective therapy was required, whether the AE was possibly related to the study drug and the action taken (i.e. patient discontinued
the study or continued in the study). PIs will see patients at their sites for all AEs that warrant specific management. AE information will be kept in each patient’s study medical chart and will be computerized. Special procedures will be followed in the instance of a serious AE. First, a serious AE form will be constructed that lists the individuals and agencies to be contacted should a serious AE occur. The first person on the list is the PI and as above, that person will follow the management of the patient. Serious AEs will be reported immediately (within 48 hours) to the patient’s primary care ob-gyn, the Human Investigation Committee, the FDA and Pfizer. Progress of these notifications will be documented on the AE form. The HIC will be notified with the institutionally approved forms.
Finally, we will request from Pfizer Pharmaceuticals, who supplies medication for the study, that we be informed of serious AEs reported to them so that the information can be shared with our respective human subjects boards and patients (if necessary). Similarly, we will inform the manufacturer and NIMH of any serious AEs.
The Weill Medical College Data Safety Monitoring Board (DSMB) will evaluate the data and safety to women enrolled in the study. The DSMB will review the study semiannually. The DSMB will evaluate recruitment, the progress of the trial, subject retention, data quality, and confidentiality. In addition, they will review subjects' clinical status, rates of adverse events, and whether or not there have been any changes in risk to participating subjects. This semi-annual review will ensure that subject risk does not outweigh study benefits. The DSMB will meet with PIs for one day in Year 3 for an interim-study meeting. A report generated from each of these meetings will be retained at the study site and will be forwarded to the IRB and to NIMH.
The DSMB will be available to convene outside of the semi-annual meeting, if necessary, due to concerns regarding a particular subject or due to any troublesome developments in subjects' experiences during the study. The DSMB will make appropriate recommendations for changes in the study protocol, if needed.
4. Confidentiality: a) Will private identifiable information about individuals be collected and used? All study-related forms will have headers with patient IDs rather than subject name to protect confidentiality. To prevent possible breaches of confidentiality, information will be encoded and exclusively accessible to study staff. Information that is computerized will be entered by patient number. No names and/or uniquely identifying information will be used in publications. b) How will research data be collected and recorded? Research data will be collected and entered into the TrialDB system. c) What methods and procedures will be used to safeguard the confidentiality of subjects and their data? Subjects’ paper files are kept in locked cabinets at our research program’s office, as are study medications. Access to those files are limited to study-specific personnel only. Should a patient APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 35
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
request information and/or records be released to another health care professional, she must first sign a written information release request. d) What mechanisms are in place to ensure proper use and continued protection of these data? Subject files will be kept onsite for 7 years after the completion of the study, after which it will be properly shredded. e) Do any limits to confidentiality exist? If a participant indicates significant suicide risk or risk of harm to others, researchers will violate her confidentiality, as mandated by law, to protect the safety of individuals. f) What will be done with the data when the research is completed? Data containing personal information and identifiers will be stored in locked file cabinets for 7 years, stripped of personal identifying information. Aggregate data will be analyzed via the TrialDB system. g) Will any external individuals or agencies (such as the study sponsor, FDA, etc.) have access to study data? Annual summary reports will be submitted to NIH, the sponsor of this study, for purposes of accountability to the funder. However, these reports are in aggregate fashion and do not divulge any individual subject’s information. NIH does not have access to individual subject files. The HIC (the Committee that reviews, approves and monitors human subjects' research) also may have access to study records. h) If appropriate, has a Certificate of Confidentiality been obtained? For more information, please see the HIC website: http://www.med.yale.edu/hic/policy/CofC.pdf We do not intend to apply for a Certificate of Confidentiality.
5. Potential Benefits: Please identify any benefits that may be reasonably expected to result from the research, either to subjects or to society at large.
The information that is obtained during this study may be useful scientifically and thus be helpful to others requiring treatment for premenstrual symptoms. Many subjects will experience a reduction in premenstrual symptoms, but this cannot be guaranteed for all.
SECTION IX: RESEARCH ALTERNATIVES AND ECONOMIC CONSIDERATIONS
1. Alternatives: For studies offering treatment, what treatment alternatives are available outside of the research? Please note: Some categories of non-treatment research may also require a section outlining alternatives to participation. For example, a study that provides screening for a particular illness or condition should state whether testing is available outside of the research.
Women may seek care from any medical provider for this diagnosis; alternative treatments include prescription medication from their provider, lifestyle changes, or herbal supplements. Women do not need to participate in this study to receive this medication.
2. Payments for participation (Economic Considerations): Describe any payments that will be made to subjects (including direct monetary payment, payment in the form of a gift, or reimbursement for costs such as travel, parking, childcare, etc.), and the conditions for receiving this compensation.
All tests, examinations, study pills and medical care required, as part of this study will be provided at no cost to study subjects. Subjects are reimbursed $10 for completing the video-taped interview, regardless of whether or not they continue on in the research intervention. Subjects are reimbursed $50 for the timely completion of Daily Rating of Problem Severity (DRSP) forms for 2 complete menstrual cycles. Subjects are reimbursed $10 for the baseline visit, $10 per treatment cycle visit (visits 2 through 7= 6 visits and extension phase visits 8 through 9 =2 visits) up to $90. Subjects will be
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 36
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
reimbursed $5 for costs associated with 5 in-person randomized office visits (Visits 1-4, 7) and 2 extension phase visits (Visit 8 & 9) for at total of $35 in mileage reimbursement costs. Study completers will receive a bonus $150 at visit 7. Therefore, study completers can receive a maximum total of $325 compensation ($50 for timely DRSP completion, $90 for study participation, $35 for mileage costs, and a $150 bonus for completing the randomized control trial.)
Costs for participation (Economic Considerations): Clearly describe the subject’s costs associated with participation in the research. If it is possible that the subject’s insurance, health plan benefits, or other third party payers will not cover research procedures or tests, this should be indicated. Clearly describe the parts of the research visits (drugs, tests, procedures, etc.) that will be provided at no cost to the subjects. Please note: If payment will be prorated for subjects who do not complete the study, this should be clearly explained. If payment is conditional on completing the study, this should be clearly explained.
Subjects should not incur any expenses related to their participation in this study. All medical procedures outlined in the research methods are provided free of cost.
3. In Case of Injury: Will medical treatment be available if injury occurs? Where and from whom may treatment be obtained? Are there any limits to the treatment being provided? Who will pay for this treatment? How will it be accessed by subjects? (Please refer to the Compensation and Medical Therapy sections of the HIC Guidance for Investigators Manual available on the HIC web site: http://www.med.yale.edu/hic/forms/forms/guidelines.pdf )
Subjects experience medication side effects, or not experiencing anticipated benefit from the study, will be removed and treated openly by the Principal Investigator for two cycles until their care can be transition to their usual medical provider, at which time care would be at the usual expense of their insurance carrier.
SECTION X: FORMATTING
Please ensure that your protocol contains the header and footer formatting as demonstrated on this document and version date. Consent, assent, permission, and information sheets must also contain these headers, footers and version dates.
References
1. Association, A. P.: Diagnostic and Statistical Manual of Mental Disorders-DSM-IV. Washington, D.C., American Psychiatric Association, 1994 2. Woods, N. F., Most, A., Dery, G. K.: Prevalence of perimenstrual symptoms. American Journal of Public Health 1982; 72:1257-1264 3. Johnson, S. R., Mcchesney, C., Bean, J. A.: Epidemiology of premenstrual symptoms in a nonclinical Sample. I. Prevalence, natural history and help-seeking behavior. Journal of Reproductive Medicine 1988; 33(4):340-346 4. Rivera-Tovar, A. D., Frank, E.: Late luteal phase dysphoric disorder in young women. The American Journal of Psychiatry 1990; 147:1634-1636 5. Ramacharan, S., Love, E. J., Fick, G. H., Goldfien, A.: The epidemiology of premenstrual symptoms in a population-based sample of 2650 urban women: Attributable risk and risk factors. Journal of Clinical Epidemiology 1992; 45:377-392 6. Wittchen, H., Becker, E., Lieb, R., Krause, P.: Prevalence, incidence and stability of premenstrual dysphoric disorder in the community. Psychological Medicine 2002; 32:119-132
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 37
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
7. Pearlstein, T., Yonkers, K., Fayyad, R., Gillespie, J.: Pretreatment pattern of symptom expression in premenstrual dsyphoric disorder. Journal of Affective Disorders 2005; 85:275- 282 8. Sternfeld, B., Swindle, R., Chawla, A., Long, S., Kennedy, S.: Severity of premenstrual symptoms in a health maintenance organization population. Obstetrics and Gynecology 2002; 99:1014-1024 9. Yonkers, K. A., Halbriech, U., Freeman, E., Brown, C., Endicott, J., Frank, E., Parry, B., Pearlstein, T., Severino, S., Stout, A., Stone, A., Harrison, W.: Symptomatic improvement of premenstrual dysphoric disorder with sertraline treatment. Journal of American Medical Association 1997; 278:983-988 10. Pearlstein, T.: Non-Antidepressant treatment of Premenstrual Syndrome. The Journal of Clinical Psychiatry 2000; 60:1-4 11. Halbreich, U., Borenstein, J., Pearlstein, T., Kahn, L.: The prevalence, impairment, impact, and burden of premenstrual dysphoric disorder (PMS/PMDD). Psychoneuroendocrinology 2003; 28(S3):1-23 12. Hylan, T., Sundell, K., Judge, R.: The impact of premenstrual symptomatology on functioning and treatment-seeking behavior: Experience from the United States, United Kingdom, and France. Journal of Women's Health and Gender-Based Medicine 1999; 8:1043-1052 13. Robinson, R., Swindle, R.: Premenstrual symptom severity: Impact on social functioning and treatment-seeking behaviors. Journal of Women's Health and Gender-Based Medicine 2000; 9:757-768 14. Borenstein, J., Dean, B., Endicott, J., Wong, J., Brown, C., Dickerson, V., Yonkers, K.: Heath and economic impact of the premenstrual syndrome. Journal of Reproductive Medicine 2003; 48:515-524 15. Chawla, A., Swindel, R., Long, S., Kennedy, S., Sternfeld, B.: Premenstrual dysphoric disorder: Is there an economic burden of illness? Medical Care 2002; 40:1101-1112 16. Kraemer, G., Kraemer, R.: Premenstrual syndrome: Diagnosis and treatment experiences. Journal of Women's Health 1998; 77:893-907 17. Campbell, E. M., Peterkin, D., O'grady, K., Sanson-Fisher, R.: Premenstrual symptoms in general practice patients. Prevalence and treatment. Journal of Reproductive Medicine 1997; 42:637-646 18. Singh, B., Berman, B., Simpson, R., Annechild, A.: Incidence of premenstrual syndrome and remedy usage: A national probability sample study. Alternative Therapies in Health and Medicine 1998; 4:75-79 19. Chaturvedi, S. K., Chandra, P. S., Gururaj, G., Pandian, D., Beena, M. B.: Suicidal ideas during premenstrual phase. Journal of Affective Disorders 1995; 34:193-199 20. Stout, A. L., Steege, J. F., Blazer, D. G., George, L. K.: Comparison of lifetime psychiatric diagnoses in premenstrual syndrome clinic and community samples. Journal of Nervous and Mental Disease 1986; 174:517-522 21. Steiner, M., Pearlstein, T.: Premenstrual dysphoria and the serotonin system: pathophysiology and treatment. The Journal of Clinical Psychiatry 2000; 61(Supplement 12):17-21 22. Rojansky, N., Halbreich, U., Zander, K., Barkai, A., Goldstein, S.: Imipramine receptor binding and serotonin uptake in platelets of women with premenstrual changes. Gynecologic and Obstetric Investigation 1991; 31:146-152
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 38
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
23. Steege, J. F., Stout, A. L., Knight, B. S., Nemeroff, C. B.: Reduced platelet tritium-labeled imipramine binding sites in women with premenstrual syndrome. American Journal of Obstetrics and Gynecology 1992; 167:168-172 24. Menkes, D. B., Coates, D. C., Fawcett, J. P.: Acute tryptophan depletion aggravates premenstrual syndrome. Journal of Affective Disorders 1994; 32:37-44 25. Heath, A. C., Yonkers, K. A., Orsulak, P., Bennett, M. J., Koonz, R., Rush, A. J.: Tryptophan depletion in premenstrual dysphoric disorder, in Society of Biological Psychiatry 1998 Annual Meeting & Scientific Program. Westin Harbour Castle, Toronto, Canada, 1998 26. Roca, C., Schmidt, P., Smith, M., Danaceau, M., Murphy, D., Rubinow, D.: Effects of metergoline on symptoms in women with premenstrual dysphoric disorder. The American Journal of Psychiatry 2002; 159:1876-1881 27. Bancroft, J., Cook, A., Davidson, D., Bennie, J., Goodwin, G.: Blunting of neuroendocrine responses to infusion of L-tryptophan in women with perimenstrual mood change. Psychological Medicine 1991; 21:305-312 28. Rasgon, N., Mcguire, M., Tanavoli, S., Fairbanks, L., Rapkin, A.: Neuroendocrine response to an intravenous L-tryptophan challenge in women with premenstrual syndrome. Fertility and Sterility 2000; 73:144-149 29. Yatham, L., Barry, S., Dinan, T. G.: Serotonin receptors, buspirone, and premenstrual syndrome. The Lancet 1989; 1:1447-1448 30. Su, T.-P., Schmidt, P. J., Danaceau, M., Murphy, D. L., Rubinow, D. R.: Effect of menstrual cycle phase on neuroendocrine and behavioral responses to the serotonin agonist m- chlorophenylpiperazine in women with premenstrual syndrome and controls. Journal of Clinical Endocrinology and Metabolism 1997; 82:1220-1228 31. Fitzgerald, M., Malone, K. M., Li, S., Al, E.: Blunted serotonergic response to fenfluramine challenge in premenstrual dysphoric disorder. Am J Psychiatry 1997; 154:556- 32. Steiner, M., Ln, Y., Coote, M., Wilkins, A., Lepage, P.: Serotonergic dysfunction in women with pure premenstrual dysphoric disorder: is the fenfluramine challenge test still relevant. Psychiatry Research 1999; 87:107-115 33. Halbreich, U., Petty, F., Yonkers, K., Kramer, G. L., Rush, A. J., Bibi, K. W.: Low plasma y- aminobutyric acid levels during the late luteal phase of women with premenstrual dysphoric disorder. The American Journal of Psychiatry 1996; 153:718-720 34. Epperson, N., Haga, K., Mason, G., Sellers, E., Gueorguieva, R., Zhang, E., Weiss, E., Rothman, D., Krystal, J.: Cortical GABA levels across the menstrual cycle in healthy women and those with premenstrual dysphoric disorder: a proton magnetic resonance spectroscopy study. Archives of General Psychiatry 2002; 59:851-858 35. Sundstrom, I., Nyberg, S., Backstrom, T.: Patients with premenstrual syndrome have reduced sensitivity to midazolam compared to control subjects. Neuropsychopharmacology 1997; 17:370-381 36. Sundstrom, I., Ashbrook, D., Backstrom, T.: Reduced benzodiazepine sensitivity in patients with premenstrual syndrome: A pilot study. Psychoneuroendocrinology 1997; 22:25-38 37. Sundstrom, I., Backstrom, T.: Citalopram increases pregnanolone sensitivity in patients with premenstrual syndrome: An open trial. Psychoneuroendocrinology 1998; 23:73-88 38. Bicikova, M., Dibbelt, L., Hill, M., Hampl, R., Starka, L.: Allopregnanolone in women with premenstrual syndrome. Hormone and Metabolic Research 1998; 30:227-230 39. Girdler, S., Straneva, P., Light, K., Pederson, C., Morrow, A.: Allopregnanolone levels and reactivity to mental stress in premenstrual dysphoric disorder. Biological Psychiatry 2001; 49:788-797 APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 39
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
40. Montelone, P., Luisi, S., Tonetti, A., Bernardi, F., Genazzani, A., Luisi, M., Petraglia, F., Genazzani, A.: Allopregnanolone concentrations and premenstrual syndrome. European Journal of Endocrinology 2000; 142:269-273 41. Rapkin, A. J., Morgan, M., Goldman, L., Brann, D. W., Simone, D., Mahesh, V. B.: Progesterone metabolite allopregnanolone in women with premenstrual syndrome. Obstetrics and Gynecology 1997; 90(5):709-714 42. Uzunova, V., Sheline, Y., Davis, J. M., Rasmusson, A., Uzunov, D., Costa, E., Guidotti, A.: Increase in the cerebrospinal fluid content of neurosteroids in patients with unipolar major depression receiving fluoxetine or fluvoxamine. Proceedings of the National Academy of Sciences of the United States of America 1998; 95:3239-3244 43. Eriksson, E., Sundblad, C., Lisjo, P., Modigh, K., Andersch, B.: Serum levels of androgens are higher in women with premenstrual irritability and dysphoria than in controls. Psychoneuroendocrinology 1992; 17(2/ 3):195-204 44. Yonkers, K., Brown, C., Pearlstein, T., Foegh, M., Sampson-Landers, C., Rapkin, A.: Efficacy of a new low-dose oral contraceptive with drospirenone in premenstrual dysphoric disorder. Obstet Gynecol 2005; 106:492-501 45. Altemus, M., Roca, C., Galliven, E., Leong, Y. M., Gold, P. W., Murphy, D. L., Deuster, P.: Increased vasopressin and ACTH responses to stress in the mid-luteal phase of the menstrual cycle. Annual Meeting, American College of Neuropsychopharmacology 1997 46. Altemus, M., Swedo, S., Leonard, H., Richter, D., Rubinow, D., Potter, W., Rapoport, J.: Changes in cerebrospinal fluid neurochemistry during treatment of obsessive-compulsive disorder with clomipramine. Arch Gen Psychiatry 1994; 51:794-803 47. Debellis, M. D., Geracioti, T. D., Listwak, S. J., Gold, P. W., Kling, M. A.: Fluoxetine reduces CSF CRH and AVP in major depression. Am. J. Psychiatry 1993; 150:656-657 48. Altemus, M., Cizza, G., Gold, P. W.: Chronic fluoxetine treatment reduces hypothalamic vasopressin secretion in vitro. Brain Res. 1992; 593(2):311-313 49. Maes, M.: The depressogenic effects of cytokines: Implications for the psychological and organic aetiology and treatment of depression. The International Journal of Neuropsychopharmacology 2002; 5:329-331 50. Wood, C., Jakubowicz, D.: The treatment of premenstrual syndrome with mefenamic acid. Br J Obstet Gynaecol 1980; 87:627-630 51. Budoff, P.: The use of prostoglandin inhibitors for the premenstrual syndrome. J Reprod Med 1983; 28:469-478 52. Gunston, K.: Premenstrual syndrome in Cape Town: part II: A double-blind placebo-controlled study of the efficacy of mefanamic acid. S Afr Med J 1986; 70:159-160 53. Jakubowicz, D. L., Godard, E., Dewhurst, S. J.: The treatment of premenstrual tension with mefenamic acid: analysis of prostaglandin concentrations. British Journal of Obstetrics and Gynaecology 1984; 91:78-84 54. Mira, M., Mcneil, D., Fraser, I. S., Vizzard, J., Abraham, S.: Mefenamic acid in the treatment of premenstrual syndrome. Journal of The American College of Obstetricians and Gynecologist 1986; 68(3):395-398 55. Facchinetti, F., Fioroni, I., Sances, G., Romano, G., Nappi, G., Genazzani, A. R.: Naproxen sodium in the treatment of premenstrual symptoms. A placebo-controlled study. Gyncol Obstet Invest 1989; 28:205-208 56. Descotes, J., Tedone, R., Evreux, J. C.: Different effects of psychotropic drugs on delayed hypersensitivity responses in mice. J Neuroimmunology 1985; 9:81-85
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 40
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
57. Sluzewska, A., Nowakowska, E., Gryska, K., Mackiewicz, A.: Interleukin-6 serum levels in depressed patients before and after treatment with fluoxetine. Ann NY Acad Sci 1994; 762:474-476 58. Bianchi, M., Rossini, G., Sacerdote, P., Panerai, A., Berti, F.: Effects of clomipramine and fluoxetine in subcutaneous carrageenin-induced inflammation in the rat. Inflammation Res 1995; 44:466-469 59. Maes, M., Song, C., Lin, A., Bonaccorso, S., Kenis, G., Dejonghj, R., Bosmans, E., Scharpe, S.: Negative immunoregulatory effects of antidepressants: inhibition of interferon-gamma and stimulation of inerleukin-10 secretion. Neuropsychopharm 1999; 20:370-379 60. Dhabhar, F. S., Mcewen, B. S.: Enhancing versus suppressive effects of stress hormones on skin immune function. PNAS 1999; 96:1059-64 61. Eriksson, E., Hedberg, M. A., Andersch, B., Sundblad, C.: The serotonin reuptake inhibitor paroxetine is superior to the noradrenaline reuptake inhibitor maprotiline in the treatment of premenstrual syndrome. Neuropsychopharmacology 1995; 12:167-176 62. Freeman, E., Rickels, K., Sondheimer, S., Polansky, M.: Differential reponse to antidepressants in women with premenstrual syndrome/premenstrual dysphoric disorder: a randomized controlled trial. Archives of General Psychiatry 1999; 56(10):932-939 63. Halbreich, U., Smoller, J. W.: Intermittent luteal phase sertraline treatment of dysphoric premenstrual syndrome. The Journal of Clinical Psychiatry 1997; 58:399-402 64. Jermain, D. M., Preece, C. K., Sykes, R. L., Kuehl, T. J., Sulak, P. J.: Luteal phase sertraline treatment for premenstrual dysphoric disorder: Results of a double-blind, placebo-controlled, crossover study. Archives of Family Medicine 1999; 8:328-332 65. Menkes, D. B., Taghavi, E., Mason, P. A., Spears, G. F. S., Howard, R. C.: Fluoxetine treatment of severe premenstrual syndrome. British Medical Journal 1992; 305:346-347 66. Ozeren, S., Corackci, A., Yucesoy, I., Mercan, R., Ethan, G.: Fluoxetine in the treatment of premenstrual syndrome. Eur J Obstet Gynecol Reprod Biol 1997; 73:167-170 67. Pearlstein, T. B., Stone, A. B., Lund, S. A., Scheft, H., Zlotnick, C., Brown, W. A.: Comparison of fluoxetine, buproprion, and placebo in the treatment of premenstrual dysphoric disorder. J Clin Psychopharmacology 1997; 17:261-265 68. Steiner, M., Steinberg, S., Stewart, D., Carter, D., Berger, C., Reid, R., Grover, D., Streiner, D.: Fluoxetine in the treatment of premenstrual dysphoria. New England Journal of Medicine 1995; 332:1529-1534 69. Stone, A. B., Pearlstein, T. B., Brown, W. A.: Fluoxetine in the treatment of late luteal phase dysphoric disorder. The Journal of Clinical Psychiatry 1991; 52:290-293 70. Su, T.-P., Schmidt, P. J., Danaceau, M. A., Tobin, M. B., Rosenstein, D. L., Murphy, D. L., Rubinow, D. R.: Fluoxetine in the treatment of premenstrual dysphoria. Neuropsychopharmacology 1997; 16:346-356 71. Sundblad, C., Modigh, K., Andersch, B., Eriksson, E.: Clomipraine effectively reduces premenstrual irritability and dysphoria: a placebo-controlled trial. Acta Psychiatr Scand 1992; 85:39-47 72. Sundblad, C., Hedberg, M. A., Eriksson, E.: Clomipramine administered during the luteal phase reduces the symptoms of premenstrual syndrome: A placebo-controlled trial. Neuropsychopharmacology 1993; 9:133-145 73. Wikander, I., Sundblad, C., Andersch, B., Dagnell, I., Sylberstein, D., Bengtsson, F., Eriksson, E.: Citalopram in premenstrual dysphoria: is intermittent treatment during luteal phases more effective than continuous medication throughout the menstrual cycle? Journal of Clinical Psychopharmaocology 1998; 18:390-398 APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 41
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
74. Wood, S. H., Mortola, J. F., Chan, Y.-F., Moossazadeh, F., Yen, S. S. S.: Treatment of premenstrual syndrome with fluoxetine: a double-blind, placebo-controlled, crossover study. Obstetrics and Gynecology 1992; 80:339-344 75. Yonkers, K. A., Halbreich, U., Freeman, E. W., Brown, C. S., Pearlstein, T. B.: Sertraline in the treatment of premenstrual dysphoric disorder. Psychopharmacology Bulletin 1996; 32:41-46 76. Young, S. A., Hurt, P. H., Benedek, D. M., Howard, R. S.: treatment of premenstrual dysphoric disorder with sertraline during the luteal phase: A randomized, double-blind, placebo- controlled crossover trial. The Journal of Clinical Psychiatry 1998; 59:76-80 77. Cohen, L., Miner, C., Brown, E., Freeman, E., Halbreich, U., Sundell, K., Mccray, S.: Premenstrual daily fluoxetine for premenstrual dysphoric disorder: a placebo-controlled, clinical trial using computerized diaries. Obstetrics and Gynecology 2002; 100:435-444 78. Freeman, E. W., Sondheimer, S. J., Polansky, M., Garcia-Espagna, B.: Predictors of response to sertraline treatment of severe premenstrual syndromes. The Journal of Clinical Psychiatry 2000; 61:579-584 79. Landen, M., Sorvik, K., Ysander, C., Allgulander, C., Nissbrandt, H., Gezelius, B., Hunter, B., Eriksson, E.: A placebo-controlled trial exploring the efficacy of paroxetine for the treatment of premenstrual dysphoria, in Presented at the APA. Philadelphia, PA, 2002 80. Miner, C., Brown, E., Mccray, S., Gonzales, J., Wohlreich, M.: Weekly luteal phase dosing with enteric-coated fluoxetine 90 mg in premenstrual dysphoric disorder: a randomized, double-blind, placebo-controlled clinical trial. Clinical Therapeutics 2002; 24:417-433 81. Dimmock, P., Wyatt, K., Jones, P., O'brien, S.: Efficacy of selective serotonin-reuptake inhibitors in premenstrual syndrome: a systematic review. The Lancet 2000; 356:1131-1136 82. Yonkers, K. A., Halbreich, U., Freeman, E., Brown, C., Endicott, J., Frank, E., Parry, B., Pearlstein, T., Severino, S., Stout, A., Stone, A., Harrison, W.: Symptomatic improvement of premenstrual dysphoric disorder with sertraline treatment. JAMA 1997; 278:983-988 83. Menkes, D., Taghavi, E., Mason, P., Howard, R.: Fluoxetine's spectrum of action in premenstrual syntrome. Int Clin Psychopharmacol 1993; 8:95-102 84. Su, T. P., Schmidt, P. J., Marry, M. O., Danaceau, M. A., Tobin, M. B., Rosenstein, D. L., Al, E.: Fluoxetine in the treatment of premenstrual dysphoria. Neuropsychopharmacology 1997; 16:346-356 85. Freeman, E., Rickels, K., Yonkers, K., Etal: Venlafaxine in the treatment of premenstrual dysphoric disorder. Obstet Gynecol 2001; 98:737-744 86. Halbreich, U., Bergeron, R., Yonkers, K., Freeman, E., Stout, A., Cohen, L.: Efficacy of intermittent, luteal phase sertraline treatment of premenstrual dysphoric disorder. Obstretrics and Gynecology 2002; 100:1219-1229 87. Steiner, M., Korzekwa, M., Lamont, J., Wilkins, A.: Intermittent fluoxetine dosing in the treatment of women with premenstrual dysphoria. Psychopharmacology Bulletin 1997; 33:771- 774 88. Gee, M., Bellew, K., Holland, F., Van Erp, E., Perera, P., Mccafferty, J.: Luteal phase dosing of paroxetine controlled-release (CR) is effective in treating premenstrual dysphoric disorder (PMDD), in American Psychiatric Association. San Francisco, CA, 2003 89. Yonkers, K. A., Pearlstein, T., Fayyad, R., Gillespie, J. A.: Luteal phase treatment of premenstrual dysphoric disorder (PMDD) improves symptoms that continue into postmenstrual phase. Journal of Affective Disorders 2005; 85(3):317-321 90. Freeman, E. W., Rickels, K., Arredond, F., Kao, L. C., Pollack, S., Sondheimer, S.: Full- or half cycle treatment of severe premenstrual syndrome with a serotonergic antidepressant. Journal of Clinical Psychopharmacology 1999; 19(1):3-8 APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 42
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
91. Sundstrom-Poromaa, I., Bixo, M., Bjorn, I. And Nordh, O.: Compliance to antidepressant drug therapy for treatment of premenstrual syndrome. Journal of Psychosomatic Obstetrics and Gynecology 2000; 21(4):205-211 92. Sundblad, C., Wikander, I., Andersch, B., Eriksson, E.: A naturalistic study of paroxetine in premenstrual syndrome: efficacy and side effects during ten cycles of treatment. European Neuropsychopharmacology 1997; 7:201-206 93. Freeman, E., Sondheimer, S., Sammel, M., Ferdousi, T., Lin, H.: A preliminary study of luteal phase versus symptom-onset dosing with escitalopram for premenstrual dysphoric disorder. The Journal of Clinical Psychiatry 2005; 66(6):769-773 94. Barbe, R. P., Bridge, J., Birmaher, B., Kolko, D., Brent, D. A.: Suicidality and its relationship to treatment outcome in depressed adolescents. Suicide and Life-Threatening Behavior 2004; 34:44-55 95. Freeman, E., Rickels, K., Sondheimer, S., Polansky, M., Xiao, S.: Continuous or intermittent dosing with sertraline for patients with severe premenstrual syndrome or premenstrual dysphoric disorder. The American Journal of Psychiatry 2004; 161(2):343-351 96. Steiner, M., Romano, S., Babcock, S., Dillon, J., Shuler, C., Berger, C., Carter, D., Reid, R., Stewart, D., Steinberg, S., Judge, R.: The efficacy of fluoxetine in improving physical symptoms associated with premenstrual dysphoric disorder. British Journal of Obstetrics and Gynaecology 2001; 108:462-468 97. Steiner, M., Kumar, R.: Serotonin reuptake inhibitors in the treatment of premenstrual dysphoria: Current state of knowledge. Journal of Psychiatry in Clinical Practice 1997; in press 98. Cohen, L., Soares, C., Yonkers, K., Bellew, K., Bridges, I., Steiner, M.: Paroxetine controlled release for premenstrual dysphoric disorder: a double-blind, placebo-controlled trial. Psychosom Med 2004; 66:707-713 99. Halbriech, U., Bergeron, R., Freeman, E., Al, E.: Intermittent luteal phase doing of sertraline effective in PMDD. Int J Neuropsychopharmacol 2000; 3:S249 100. Cohen, L., Soares, C., Lyster, A., Cassano, P., Brandes, M., Leblanc, G.: Efficacy and tolerability of premenstrual use of venlafaxine (flexible dose) in the treatment of premenstrual dysphoric disorder. J Clin Psychopharmacol 2004; 24:540-543 101. Michelson, D., Fava, M., Amsterdam, J., Apter, J., Londborg, P., Tamura, R., Tepner, R. G.: Interruption of selective serotonin reuptake inhibitor treatment: Double-blind, placebo- controlled trial. The British Journal of Psychiatry 2000; 176:363-368 102. Rosenbaum, J., Fava, M., Hoog, S., Ascroft, R., Krebs, W.: Selective serotonin reuptake inhibitor discontinuation syndrome: A randomized clinical trial. Biological Psychiatry 1998; 44:77-87 103. Judge, R., Parry, M. G., Quail, D., Jacobson, J. G.: Discontinuation symptoms: Comparison of brief interruption in fluoxetine and paroxetine treatment. International Clinical Psychopharmacology 2002; 17(5):217-225 104. Posternak, M., Zimmerman, M.: Is there a delay in the antidepressant effect? A meta-analysis. The Journal of Clinical Psychiatry 2005; 66(2):148-158 105. Kuhn, R.: The treatment of depressive states with G 22355 (imipramine hydrochloride). the American Journal of Psychiatry 1958; 115:459-464 106. Pollack, B.: Clinical findings in the use of Tofranil in depressive and other psychiatric states. The American Journal of Psychiatry 1959; 116:312-317 107. Keup, W., Apolito, A., Olinger, L., Al., E.: Tofranil (imipramine) in the treatment of depressive states. Journal of Nervous and Mental Disease 1960; 130:146-150
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 43
Downloaded From: https://jamanetwork.com/ on 09/23/2021 HIC# 0609001839
108. Angst, J.: Mechanisms of action of imipramine and its therapeutic use. Bern, Switzerland, Stampfli and Cie, 1970 109. Smith, M. V., Cavaleri, M. A., Howell, H. B., Poschman, K., Rosenheck, R. A., Yonkers, K. A.: Screening for and detection of depression, panic disorder, and PTSD in public-sector obstetrical clinics. Psychiatric Services 2004; 55(4):407-414 110. Yonkers, K. A., Holthausen, G. A., Poschman, K., Howell, H. B.: Symptom- onset treatment for women with premenstrual dysphoric disorder. J of Clinical Psychopharmacology 2006; 26:198-202 111. Sheehan, D., Lecrubier, Y., Janavs, J., Knapp, E., Weiller, E.: Mini International Neuropsychiatric Interview (MINI). University of South Florida, Tampa, 1996 112. Steiner, M., Haskett, R. F., Carroll, B. J.: Premenstrual tension syndrome: The development of research diagnostic criteria and new rating scales. Acta Psychiatrica Scandinavica 1980; 62:177-190 113. Rush, A. J., Gullion, C. M., Basco, M. R., Jarrett, R. B., Trivedi, M. H.: The Inventory of Depressive Symptomatology (IDS) Psychometric properties. Psychological Medicine 1996; 26:477-486 114. Endicott, J., Nee, J., Harrison, W., Blumenthal, R.: Quality of Life Enjoyment and Satisfaction Questionnaire: A new measure. Psychopharmacology Bulletin 1993; 29:321-326 115. Rush, A. J., Hiser, W., Giles, D. E.: A comparison of self-reported versus clinician-rated symptoms in depression. The Journal of Clinical Psychiatry 1987; 48(6):246-248 116. Rapaport, M., Wolkow, R., Rubin, A., Hackett, E., Pollack, M., Ota, K.: Sertraline treatment of panic disorder: Results of a long-term study. Acta Psychiatrica Scandinavica 2001; 104:289- 298 117. Brunner, E., Domhof, S., Langer, F.: Nonparametric Analysis of Longitudinal Data in Factorial Experiments. New York, NY, Wiley, 2002 118. Diggle, P., Heagerty, P., Liang, K., Zeger, S.: Analysis of Longitudinal Data, in Oxford University Press, 2002 119. Hedeker, D., Gibbons, R.: Application of random-effects pattern-mixture models for missing data in longitudinal studies. psychological Methods 1997; 2(1):64-78 120. Kraemer, H., Wilson, G., Fairburn, C., Agras, W.: Mediators and moderators of treatment effects in randomized clinical trials. Archives of General Psychiatry 2002; 59(10):877-883 121. Cohen, J.: Statistical power analysis for the behavioral sciences. New Jersey, Lawrence Erlbaum Associates, 1988
APPROVED BY THE YALE UNIVERSITY HIC 05-OCT-2011 44
Downloaded From: https://jamanetwork.com/ on 09/23/2021