This study included 15 clinical trials involving 945 patients to evaluate the efficacy of PD-1/PD-L1 inhibitors in treating advanced OC. The pooled results showed that the ORR was 19%. Single PD-1/PD-L1 inhibitors showed limited efficacy, with an ORR of 9%, while combination with chemotherapy showed an increased ORR of 36%. In addition, PD-1/PD-L1 inhibitors had a higher ORR in platinum-sensitive OC than in platinum-resistant OC (31% vs 19%).
Immune checkpoint inhibitors, especially PD-1/PD-L1 inhibitors, are changing the treatment paradigm in certain cancers, such as melanoma and non-small cell lung cancer. The overall ORR with single PD-1/PD-L1 inhibitors across other cancers was approximately 20%, while it was 9% in OC. Previous studies have shown that PD-L1 expression (tumor cells and/or tumor-infiltrating lymphocytes), tumor mutational burden (TMB), microsatellite instability (MSI) and/or mismatch repair (MMR) deficiency are effective predictive biomarkers for anti-PD1/PD-L1 therapy. However, KEYNOTE-028 showed a poor ORR for PD-1 inhibitors, even in PD-L1-positive OC patients (14). Additionally, KEYNOTE-100 showed a low rate of MSI in OC. As for TMB, it was also very low in OC patients. Therefore, seeking an optimal treatment modality with PD-1/PD-L1 inhibitors seems necessary before identifying a better predictive biomarker.
Vascular endothelial growth factor (VEGF) creates an immunosuppressive microenvironment within cancers by suppressing dendritic cell maturation, increasing the Treg population and stimulating the growth of myeloid-derived suppressor cells in the tumor microenvironment (21, 22). Bevacizumab can reverse these VEGF-mediated immunosuppressive effects on the tumor microenvironment, potentially augmenting immune-mediated antitumor activity. Several studies have demonstrated the synergistic effect between antiangiogenic agents and PD-1/PD-L1 inhibitors in solid tumors, including renal cancer, non-small lung cancer, and endometrial cancer (23–27). OC is known to highly express VEGF, which serves as a major driver of tumor neovascularization and local immune suppression (28). Therefore, anti-VEGF agents could theoretically enhance the efficacy of immunotherapy in OC. This study also showed a high ORR of 30% in OC patients treated with antiangiogenic agents and PD-1/PD-L1 inhibitors.
In recent years, increasing evidence has shown that chemotherapy is not only a cytotoxic agent but also a stimulator of tumor-specific immune responses. Chemotherapy involves the stimulation of anticancer immunity either by initiating the release of immunostimulatory molecules from dying cancer cells or by mediating off-target effects on immune cell populations (29). On the one hand, chemotherapy could induce immunogenic cell death (ICD), enabling the release of neoantigens and signals to antigen-presenting cells; on the other hand, chemotherapy was found to reduce the number and activity of immune-suppressive cells, including myeloid-derived suppressor cells and Treg cells (30–32). Therefore, chemotherapy can theoretically initiate or restore anticancer immune responses by converting immunologically “cold” tumors into “hot” tumors. Several studies have shown clinical activities with a combination of immunotherapy and chemotherapy (33, 34). This study also showed a high ORR of 36% in OC patients treated with chemotherapy and PD-1/PD-L1 inhibitors.
Platinum-resistant OC is a dismal disease and has a low response to subsequent chemotherapy. In this study, we found that the pooled ORR was 25% in studies on platinum-resistant patients and 49% in studies on platinum-sensitive patients. This might be attributed to the immunosuppressive environment in platinum-resistant OC. Data on the tumor microenvironment of platinum-resistant OC showed low CD8 + T cell infiltration and highly activated CD4 + T cells (9, 35).
This study had some limitations. First, most of the included articles were noncomparable studies, and some of them had small sample sizes. Second, the PD-1/PD-L1 inhibitors were different among studies, which inevitably caused bias. Third, the complete data were hardly accessible in some studies to perform subgroup analysis.