Predictive value of PD-L1 expression in response to immune checkpoint inhibitors for gastric cancer treatment: a systematic review and meta-analysis

Background Which patients may benefit more from immune checkpoint inhibitors (ICIs) is still an important question. The present study aimed to investigate the role of the PD-L1 molecule in predicting the effectiveness of PD-1/PD-L1 inhibitors in gastric cancer patients.Method We searched PubMed, Scopus, Web of Science, and EMBASE databases as of 25 March 202225 March 2022.Results Ten articles were included. When we used the TPS method for PD-L1 expression, none of the patients in tumor proportion score (TPS)≥1% and TPS < 1% groups took advantage of ICI therapy in terms of OS and PFS. However, gastric cancer patients with combined-positive score (CPS)≥1, CPS ≥ 5, and CPS ≥ 10 tumors represented superior OS for ICIs over the control agents, while their counterparts (i.e. patients with CPS < 1, CPS < 5, and CPS < 10 tumors) did not. In the subgroup analysis when patients with CPS ≥ 1 were selected, Nivolumab improved the OS and PFS remarkably by 26% and 25% when compared with control agents, respectively. However, Pembrolizumab significantly increased the rate of disease progression by 47% relative to the control medications.Conclusion Among patients suffering from gastric cancer, considering PD-L1 CPS thresholds seems to be a more reliable predictive factor than TPS threshold for lower rate of mortality when PD-1/PD-L1 inhibitors are administered.