Efficacy and toxicity of neoadjuvant immune checkpoint inhibitors in resectable gastric cancer: A meta-analysis and systematic review.

291 Background: Immune checkpoint inhibitors (ICI) have made significant breakthroughs in late-stage gastric cancer. It is an attractive issue whether ICIs also function in the neoadjuvant setting. Methods: A systematic review was performed using combined terms related to “stomach cancer”, “gastric cancer”, “gastroesophageal cancer”, “immune checkpoint inhibitor”, “PD-1”, “PD-L1” and “neoadjuvant” in PubMed and annual meeting of ASCO, ASCO GI, ESMO, and ESMO GI from 2019 to 2021. Complete pathological response (CPR) rates, major pathological response (MPR) rates, R0 resection rates, and side effects were pooled and analyzed. MPR was defined as ≤10% viable tumor cells and included CPR. Study outcomes were pooled using the function METAPHOR in the META package in R.3.6.1. The I 2 and P statistics were used to evaluate heterogeneity among studies. Funnel plots were used for publication bias assessment. Results: A total of 13 phase I/II clinical trials, including 332 resectable gastric cancer (T2-4 or N+) patients with neoadjuvant ICI-containing treatments, were collected. The pooled rates of CPR, MPR, and R0 resection were 0.16 (95% credible intervals (CI), 0.12-0.22), 0.36 (95% CI, 0.24-0.51), and 0.97 (95% CI, 0.94-0.99), respectively. As a comparison, outcomes from 25 studies on neoadjuvant chemotherapy were also pooled, with rates of CPR 0.08 (95% CI, 0.06-0.11), MPR 0.22 (95% CI, 0.19-0.26), and R0 resection 0.84 (95% CI, 0.80-0.87). Besides, the overall grade 3 or higher toxicity rates were 0.24 (95% CI 0.03-0.54) vs. 0.28 (95% CI 0.13-0.47) in ICI-based treatment and chemotherapy groups. Stratified by treatment strategies, ICI alone showed the lowest efficacy (CPR 0.07, 95% CI 0.02-0.19 and MPR 0.16, 95% CI 0.03-0.29). Addition of chemotherapy to ICI promoted the CPR (0.15, 95% CI 0.10-0.22) and MPR (0.36, 95% CI 0.22-0.50). Further addition of radiotherapy displayed the highest CPR (0.35, 95% CI 0.21-0.52) and MPR (0.74, 95% CI 0.54-0.93). Only four trials reported the outcomes of dMMR/MSI-H patients, with superior CPR (0.39, 95% CI 0.20-0.62) and MPR (0.82, 95% CI 0.82-1.00), while those of pMMR/MSS patients were 0.05 (95% CI 0.00-0.13) and 0.20 (95% CI 0.20-0.31) respectively, among which 2 trials used ICI only. Conclusions: Neoadjuvant ICI plus chemotherapy/radiotherapy, instead of ICI alone, has better pathological responses and R0 resection rates than chemotherapy. dMMR/MSI-H is a superior biomarker for neoadjuvant ICI therapy for resectable gastric cancer. The value or biomarker of immunotherapy in pMMR/MSS patients remains exploring.[Table: see text]