In contrast to progress for autism, bipolar disorder, and schizophrenia, discerning the biological basis of major depressive disorder (MDD) has been difficult. Genetic studies of MDD using genome-wide linkage, candidate gene, and GWAS (with Ns >20,000 subjects) have not been successful in identifying risk loci that meet contemporary standards for replication. Major lessons are that genetic approaches for higher prevalence/lower heritability diseases like MDD may be suboptimal given the likely role of etiological heterogeneity. We propose here an alternative strategy to identify networks and pathways involved in the etiology of MDD. To minimize heterogeneity, we will study postpartum depression (PPD), a more homogenous type of MDD. Although most women have mild mood symptoms in the first few days to weeks postpartum (the """"""""baby blues""""""""), these symptoms usually resolve spontaneously. In contrast, PPD is a severe and persistent form of MDD that is one of the most frequent complications of childbirth (prevalence 15-20%) and is the leading cause of maternal death. Practically, pregnant women are straight-forward to identify, as they have frequent contact with the healthcare system and are willing to participate in research on perinatal problems. The study of PPD should decrease heterogeneity in multiple ways (females only, age-banded, all subjects exposed to the same biopsychosocial event). Moreover, we have moved to the rigorous study of the biomarker space to identify the distinguishing features of PPD of potential clinical relevance. DNA methylation markers are appealing because methylation is directly related to gene expression. RNA expression signatures add complementary data on the state of a tissue and even of an organism. Finally, the comprehensive study of neuroendocrine hormones is of clear salience for PPD. We propose: (1) to ascertain and sample 1,000 PPD cases and 1,000 euthymic controls, all self-reported African-American;(2) conduct discovery analyses using an unbiased/screening biomarker assessment for 500 PPD cases and 500 controls (methylomics and transcriptomics via next-generation technologies on peripheral blood), and state-of-the-art statistical analyses will identify multivariate biomarker signatures for PPD;and (3) conduct validation assays of biomarkers meeting liberal significance criteria in independent samples (500 PPD cases and 500 controls). Our focus on African American women is optimal as this US minority group is manifestly under-studied, the prevalence of PPD is higher in African-American women, and in contrast to genetic studies where ancestry heterogeneity is a major source of false positive signals, it can be a strength in biomarker studies. MDD is a first-rank public health problem due to the associated morbidity, mortality, and personal/societal costs. Studying the MDD sub-form PPD is particularly appealing due to the inherent control for multiple sources of heterogeneity - we propose to study women of child-bearing years who are all exposed to the major risk factor of pregnancy. Moreover, PPD is itself an important and under-studied human health concern, particularly in African-American women. Successful completion of the proposed research will yield strong and replicable biomarker signatures for PPD. This new knowledge would provide insight into state-related alterations characteristic of PPD and potentially to trait-related vulnerabilities to MDD (particularly in combination with GWAS findings). Future research could evaluate whether a biomarker signature is of predictive utility and, if so, rational primary prevention may become feasible.
Postpartum depression causes enormous human suffering and cost to society. Our goal is rapidly to learn more about the biological basis of postpartum depression by studying the blood of African-American women with and without postpartum depression.