The relationship of lifetime history of depression on the ovarian tumor immune microenvironment

Depression is associated with a higher ovarian cancer risk. Prior work suggests that depression can lead to systemic immune suppression, which could potentially alter the anti-tumor immune response. We evaluated the association of pre-diagnosis depression with features of the anti-tumor immune response, including T and B cells and immunoglobulins, among women with ovarian tumor tissue collected in three studies, the Nurses’ Health Study (NHS; n = 237), NHSII (n = 137) and New England Case-Control Study (NECC; n = 215). Women reporting depressive symptoms above a clinically relevant cut-point, antidepressant use, or physician diagnosis of depression at any time prior to diagnosis of ovarian cancer were considered to have pre-diagnosis depression. Multiplex immunofluorescence was performed on tumor tissue microarrays to measure immune cell infiltration. In pooled analyses, we estimated odds ratios (OR) and 95% confidence intervals (CI) for the positivity of tumor immune cells using a beta-binomial model comparing those with and without depression. We used Bonferroni corrections to adjust for multiple comparisons. We observed no statistically significant association between depression status and any immune markers at the Bonferroni corrected p-value of 0.0045; however, several immune markers were significant at a nominal p-value of 0.05. Specifically, there were increased odds of having recently activated cytotoxic (CD3+CD8+CD69+) and exhausted-like T cells (CD3+Lag3+) in tumors of women with vs. without depression (OR = 1.36, 95 %CI = 1.09–1.69 and OR = 1.24, 95 %CI = 1.01–1.53, respectively). Associations were comparable when considering high grade serous tumors only (comparable ORs = 1.33, 95 %CI = 1.05–1.69 and OR = 1.25, 95 %CI = 0.99–1.58, respectively). There were decreased odds of having tumor infiltrating plasma cells (CD138+) in women with vs. without depression (OR = 0.54, 95 %CI = 0.33–0.90), which was similar among high grade serous carcinomas, although not statistically significant. Depression was also related to decreased odds of having naïve and memory B cells (CD20+: OR = 0.54, 95 %CI = 0.30–0.98) and increased odds of IgG (OR = 1.22, 95 %CI = 0.97–1.53) in high grade serous carcinomas. Our results provide suggestive evidence that depression may influence ovarian cancer outcomes through changes in the tumor immune microenvironment, including increasing T cell activation and exhaustion and reducing antibody-producing B cells. Further studies with clinical measures of depression and larger samples are needed to confirm these results.