Reproductive Endocrine Related Mood Disorders-Differential Sensitivity

National Institute Of Mental Health

Investigators

Linked publications & trials

Abstract

This report includes work arising from the following clinical protocols: NCT00005011, NCT00056901, NCT00059228, NCT00082043, NCT00100360, NCT00001177, NCT00001259, and NCT00001481. Although the mechanisms underlying the mood destabilizing effects of ovarian steroids in PMDD remain to be fully characterized, as does the source of susceptibility to this trigger, our findings provide a new target on which interventions could be focused, namely changes in steroid (or steroid metabolite) levels across the menstrual cycle. In summary, our work to date demonstrates that in women with PMDD, but not in women without PMDD, a change in ovarian steroid level permits the recurrence of a dysregulated affective state. Our studies and those of others have documented that PMDD symptoms are eliminated by ovarian suppression and recur after administration of ovarian steroids (yet appear in the context of levels of ovarian steroids indistinguishable from those in women without PMDD). We recently, replicated these findings of a differential behavioral sensitivity in PMDD in a larger independent sample of women with PMDD and control women. In this study, thirty-four women with PMDD (10 from the original cohort) and 76 healthy participants (15 from the original cohort) completed a daily rating form during three hormone conditions: ovarian suppression induced by the gonadotropin-releasing hormone agonist leuprolide, leuprolide+estradiol addback, and leuprolide+progesterone addback. Affective and somatic symptom scores during the last 8 of 12 weeks of leuprolide alone were compared with scores during the first 4 weeks of estradiol addback and the first 4 weeks of progesterone addback. For affective symptoms (anxiety, sadness, irritability, mood swings), there were significant increases in symptom severity scores during estradiol addback and progesterone addback compared with leuprolide treatment alone. Compared to healthy comparison participants, women with PMDD had significantly higher symptom scores during each addback. With regard to physical symptoms, women with PMDD had significantly more severe bloating and food cravings, regardless of hormone conditions. Breast pain was significantly more severe during estradiol addback compared with leuprolide alone or progesterone addback. Our study confirmed that ovarian suppression in women with PMDD eliminates symptom cyclicity, and that symptoms are precipitated during ovarian steroid addback in women with PMDD but not in healthy comparison women. In PMDD, irritability and mood swings are tied more closely to progesterone than estradiol. Despite the replication of this hormone-related behavioral phenotype in PMDD, the mechanisms underlying the presumed alteration in steroid signaling require further characterization. We also employed a machine learning platform in a large sample of over 800 women (women who met DSM criteria for PMDD, asymptomatic controls and women who presented with PMDD but who did not meet research criteria for PMDD) who completed daily symptom ratings of sadness, anxiety and irritability across at least two menstrual cycles. Included in this data set are multiple clinical characteristics of these women as well as laboratory tests when available. We will examine several questions in this data set including the specificity of the symptom of irritability in PMDD, the clinical characteristics associated with PMDD (versus those who do not meet criteria for PMDD) and whether there are distinct patterns of symptom expression that define specific phenotypes of PMDD that will be employed with other ongoing studies (e.g., development of polygenic scores) to better characterize this condition. Our collaborations with Elizabeth Bertone-Johnson and Donghao Lu will be critical to these questions given the identification of several epidemiologic risk factors and the development of polygenic risk scores in their epidemiologic studies of women with PMDD and premenstrual disorders. In women who have a past DSM-major or -minor depression during the postpartum (PPD) but who are currently euthymic, we demonstrated that PPD, like other RMDs, can be characterized by differential sensitivity to declining ovarian steroids. In our first study, we recapitulated several hormonal events characteristic of pregnancy and the postpartum. After suppressing ovarian steroid production with a GnRH agonist, we administered for two months, and then precipitously withdrew, daily supraphysiologic doses of estradiol (10 mg Estrace) and progesterone (1600 mg Prometrium). Asymptomatic women with a history of PPD, but not controls, experienced significant symptoms of depression when withdrawn from ovarian steroids. Nonetheless, several women with past PPD developed symptoms toward the end of the high dose phase (i.e., prior to hormone withdrawal), thus making it unclear whether steroid withdrawal triggered the depression (as is usually presumed) or whether symptoms emerged after prolonged exposure to ovarian steroids (and continued during withdrawal). Indeed, a recent study employing the identical design reported that women with past PPD, but not control women, experienced recurrences of negative mood symptoms during both (unblinded) ovarian steroid withdrawal and addback. These findings demonstrate and confirm the differences in behavioral response to ovarian steroids in women with PPD compared with control women, but suggest the non-specificity of ovarian steroid withdrawal per se in the ontogeny of PPD. We modified our original protocol to examine the role of ovarian steroid withdrawal in PPD under blinded conditions. We employed an almost identical hormone manipulation protocol as used previously in asymptomatic women with past PPD and asymptomatic control women (HCs) with no past history of affective disorder (including PPD); this protocol differed in that after the 8-week add back phase of this study, all women were randomized during the last month of the study (i.e., weeks 16 – 20) in a double-blind manner to receive either placebo (withdrawal as in the original study) or combined E2 and progesterone (continuation). This design permits assessment of the specific contribution of E2 and progesterone withdrawal (versus latency following exposure to chronic high dose combined ovarian steroids) to the triggering of depression in PPD. Sixteen women with past PPD and twelve HCs completed this protocol. Our preliminary results confirm the vulnerability to experience recurrent mood symptoms in women with past PPD, but not HCs, throughout the hormone manipulation protocol. In the women with past PPD, recurrences of depressive symptoms frequently occurred in the context of changes in circulating ovarian steroid levels (both planned and random occurrences); however, there were no differences in the pattern or severity of symptom recurrence between women with PPD randomized to ovarian steroid withdrawal (n=9) and those randomized to continuation (n=7). Thus, our preliminary data are consistent with those reported by Schiller et al and confirm differential sensitivity to ovarian steroids in women with PPD (compared with HCs) but now suggest that it is the exposure rather than the withdrawal that is the primary trigger of recurrent mood symptoms. In collaboration with Dr. David Goldman at NIAAA, we developed lymphoblastoid cell lines (LCLs) and human induced-pluripotent cell lines (h-IPSCs) from women with and without RMDs (i.e., women with PMDD, PPD, PPP and perimenopausal depression PMD). We are completing a major effort leveraging combined transcriptomic and genomic data to identify and characterize the shared and unshared genes of risk underlying reproductive mood disorders (RMDs). Our research question is, how do these genes and transcripts compare across these disorders, and to other mood disorders such as depression? Preliminary analysis of transcriptomic data on PMDD, perimenopausal depression (PMD), PPD LCLs suggests highly unique transcriptomic signatures for each condition.