In our earlier studies we have developed experimental models for the triggering of symptoms in both premenstrual dysphoric disorder (PMD) and postpartum depression that we continue to employ in our efforts to identify both the underlying biology of these conditions as well as the development of new safe and effective therapies for women with these conditions. As a continuation of these studies, we have demonstrated that women with PMD who respond to gonadotropin releasing hormone (GnRH) agonist-induced ovarian suppression experience a recurrence of PMD after the initial re-exposure to combined estradiol and progesterone (but not placebo). However, recurrent symptoms do not occur once hormone levels are stabilized over the subsequent two months of continuous replacement therapy. These observations are of both clinical and scientific importance, as they identify promising phenotypes and suggest the physiologic basis for the susceptibility to experience PMD and will also provide alternative hormone-based therapies for women with this condition. Approximately 60-70% of women with PMD show symptomatic improvement in response to the GnRH agonist-induced ovarian suppression. It has been very difficult to predict why some women respond, while others do not. We identified a suite of statistical descriptors of pre-treatment mood rating data to have high sensitivity and specificity for predicting the clinical response to ovarian suppression in PMD by applying chaos-based Approximate Entropy (ApEn) modeling to mood rating data. Relatively regular and non-spiky pre-trial dynamics of mood ratings predict a positive response to ovarian suppression with high probability. In contrast, mean symptom levels were indistinct between the subgroups. We are presently in the latter stages of a follow-up study employing these same statistical measures involving the response of women with PMD to fluoxetine. In contrast to our findings with leuprolide, none of the statistical measures distinguished responders from nonresponders to fluoxetine. Thus, these data suggest that the pattern of symptoms cyclicity predicts a differential response to hormonal therapy versus a psychotropic agent. These statistical measures may have broad applicability to behavioral studies for many psychiatric disorders, facilitating both accurate diagnosis and the prediction of response to treatment. As an indirect measure of the relevance of declining ovarian steroid secretion in PPD, we examine the efficacy of estradiol in the treatment of PPD. Women with PPD (both those nursing and those who are bottle-feeding their babies) are randomized in a double-blind, parallel design to receive either 17 beta estradiol (100 mcgs daily by skin patch) or placebo for six weeks. Preliminary results suggest that mood rating scores are improved compared with both baseline and scores in women receiving placebo. Differences between estradiol and placebo treatments were apparent by four weeks, reminiscent of the relatively rapid antidepressant effects of estradiol therapy in depressed perimenopausal women. If these findings are confirmed in a larger sample, estradiol treatment may not only provide a safe and effective alternative to traditional antidepressants in women with postpartum depression (both nursing and non-nursing mothers), but it may also suggest the relevant hormonal trigger for the development of this condition. In addition to our studies on the behavioral effects of changes in sex steroids across the menstrual cycle and during the postpartum, we employ a number of methodologies to investigate the underlying biological mechanisms of these conditions including studies employing multimodal neuroimaging such as positron emission tomography (PET), structural magnetic resonance imaging (MRI), and functional magnetic resonance imaging (fMRI). Our neuroimaging protocols demonstrated for the first time in humans a differential reward-related pattern of brain activation in the orbital frontal cortex and the amygdala during the luteal phase compared with the follicular phase of the menstrual cycle using fMRI technology. We also perform O15 PET studies in women who are participating in the GnRH agonist-induced hypogonadism study. In this study, cognitive activation is achieved using the N-back test which allows us to vary the cognitive load and the effort required to perform the task. Differences in brain activation between the 2-back and 0-back (control) tasks were compared across hormone conditions in women with and those without PMD. Preliminary results suggest that during the N-back testing, women with PMD show abnormal recruitment of the dorsolateral prefrontal cortex (dlPFC) with greater activation in the right dlPFC, and reduced activation in the left DLPFC an asymmetry not observed in the asymptomatic controls. These findings are robust and observable in both 015PET and fMRI techniques. Interestingly, scores on the global assessment of function scale, a measure of the severity of the functional impairment accompanying each womans PMD, correlate with activity within the medial frontal and precentral gyri. Future findings in these studies may identify disturbances of a process that is critical to the clinical phenomenology of PMD, a functional disturbance of the normal modulatory effect of gonadal steroids, and a locus of the disturbance. Our therapeutic clinical trials have demonstrated that the 5-alpha reductase inhibitor, dutasteride, a medication that inhibits neurosteroid synthesis has no effect on either the symptoms of PMD or the luteinizing hormone (LH) surge prior to ovulation. We are evaluating the effects of a higher dose of dutasteride (i.e., 2.5 mg daily) in women with PMD and an asymptomatic comparison group. However, our preliminary findings suggest that neither alterations in the plasma levels nor a deficiency of neurosteroids, such as allopregnanolone, play a critical role in the pathophysiology of PMD. We are collaborating with Dr. Rima Kaddurah-Daouk at Duke University to more comprehensively examine neurosteroid metabolism with metabolomic tools. Our strategy is to perform metabolomic studies in women with PMD who respond to Lupron with elimination of symptoms and who experience return of symptoms during progesterone or estradiol replacement. We now have the capacity to examine steroid and steroid metabolites employing a new platform designed for lipid-soluble molecules. Thus, we now want to specifically examine the effects of both estradiol and progesterone on the pattern of steroid metabolites in women with PMD and an asymptomatic comparison group who participate in the Lupron paradigm. Finally, we have completed our study of HPA axis function in women with and without premenstrual dysphoria under conditions of GnRH agonist-induced hypogonadism and estradiol and progesterone replacement. We have adequately demonstrated and confirmed previous findings that even under conditions of dexamethasone suppression progesterone results in an increase in several measures of cortisol and ACTH secretion. However, the effects of progesterone (or either hypogonadism or estradiol) on HPA axis function did not differ between women with PMD and controls. Thus, although mood symptoms in women with PMD were largely confined to the progesterone add back phase and progesterone increased HPA axis response, the regulation of HPA axis function did not differ in PMD compared with controls. These data suggest that HPA axis function in women with PMD is normal.
Showing the most recent 10 out of 24 publications
