PD-L1 Expression on Circulating Tumour-Derived Microvesicles as a Complementary Tool for Stratification of High-Grade Serous Ovarian Cancer Patients

Simple Summary PD-1/PD-L1 axis blockade immunotherapy leverages on assessment of tumour PD-L1 status by immunohistochemistry for identifying cancer patients' who would benefit most from this therapy. Because some PD-L1 negative cancer patients also benefit from the treatment, it has been postulated that spatial heterogeneity of PD-L1 expression in tumour tissue may reduce the predictive value of the test. Here we demonstrate for the first time the presence of PD-L1+ circulating tumor-derived microvesicles in the blood stream of high grade serous ovarian cancer patients and propose the assessment of tumor-derived circulating PD-L1+ microvesicles to be inserted in predictive algorithms for complementing tumour PD-L1 status. Background: Ovarian cancer (OC) has recently attracted attention for the use of PD-1/PD-L1 axis blocking agents, with durable activity reported only in a subset of patients. The most used biomarker for sensitivity to the PD-1/PD-L1 axis blockade is tumour PD-L1 status by immunohistochemistry. However, patient stratification using this method suffers from intrinsic heterogeneity of OC, likely contributing to the unsatisfactory results obtained so far. Cells communicate with each other by releasing microvesicles (MVs) that carry parental cell surface features. Thus, we hypothesised that PD-L1(+) tumour cells (TC) and infiltrating PD-L1(+) leukocytes should shed MVs carrying surface PD-L1 that may serve as a proxy for the whole tumour PD-L1 status. Results: We showed for the first time the presence of measurable amounts of TC- and leukocyte-derived PD-L1(+) MVs (range: 1.4-178.8 MVs/mu L and 6.2-504.8 MVs/mu L, respectively) in the plasma of high-grade serous OC (HGSOC) patients (n = 63), using a sensitive flow cytometry platform. The concentration of PD-L1(+) MVs of either origin did not associate with the PD-L1 status of TCs and leukocytes in the tumour biopsies, suggesting that the circulating PD-L1(+) MVs also included ones from locations not selected for immunohistochemistry analysis and represented the PD-L1 status of the whole tumour mass. In this study, we also describe the serendipitous discovery of circulating PD-L1(+) MVs of platelet origin (10.3-2409.6 MVs/mu L). Conclusions: The enumeration of circulating PD-L1(+) MVs in HGSOC patients may provide a novel direction for assessing the tumour PD-L1 status and contribute to HGSOC patient stratification for immunotherapy interventions. The presence of circulating PD-L1(+) MVs of platelet origin, a finding not yet reported in HGSOC patients, warrants further studies.