Background Although immune checkpoint blockade revolutionized cancer therapy, response rates have been mixed in gynecological malignancies. While uterine endometrial cancer with high microsatellite instability (MSIHI) and high tumor mutational burden (TMB) respond robustly to checkpoint blockade, high-grade serous ovarian cancer (HGSOC) with low TMB respond modestly. Currently, there has been no known immune signature or T cell phenotype that predicts clinical response in gynecological tumors.

Methods To dissect the immune landscape and T cell phenotypes in gynecological cancer patients receiving PD-1 blockade, we used high-dimensional cytometry (flow cytometry and mass cytometry (CyTOF)). We performed longitudinal deep immune profiling of PBMC from patients with recurrent uterine endometrial cancer receiving single-arm nivolumab, and HSGOC patients receiving neoadjuvant nivolumab plus platinum-based chemotherapy prior to debulking surgery.

Results Chemotherapy-resistant MSI-H uterine cancer patients treated with nivolumab had a proliferative T cell response 2–4 weeks post PD-1 blockade, consistent with responses seen in high TMB melanoma and lung cancer. The responding Ki67+ CD8 T cell population was largely CD45RAloCD27hi or CD45RAloCD27lo and highly expressed PD1, CTLA-4, and CD39, consistent with the phenotype of exhausted T cells (TEX). These exhausted-like cells are enriched in responders, whereas early expansion Tregs are enriched in non-responders. Unlike patients with uterine endometrial cancer, patients with TMBlo ovarian cancer did not have a clear proliferative CD8 T cell response after neoadjuvant nivolumab plus chemotherapy treatment, suggesting systemic immune suppression. At baseline, ovarian cancer without recurrence have more terminally differentiated effector-like CD8 T cells, and patients with recurrence have more naive-like cells. Thus, both high and low TMB gynecological tumors have distinct immune landscapes associated with clinical response. Additionally, in MSI-H uterine endometrial cancer patients, the length of time between the prior chemotherapy and the initiation of immunotherapy was negatively correlated with T cell reinvigoration post immunotherapy and clinical response. This suggests the importance of optimize therapeutic timing to maximize the therapeutic efficacy when combining immunotherapy and chemotherapy.

Conclusions Collectively, our immune profiling revealed the distinct immune signatures associated with clinical response to PD-1 blockade in gynecological cancers. Our results also suggest that TMBhi inflamed versus TMBlo cold tumor microenvironment, and timing of chemo/immunotherapy could impact differentiation and functions of T cells.

Ethics Approval The study was approved by MSKCC Ethics Board, approval number 17–180 and 17–182.