# Efficacy and safety of perioperative immune checkpoint inhibitors combined with chemotherapy versus chemotherapy alone for locally advanced gastric or gastroesophageal junction adenocarcinoma: a systematic review and meta-analysis of randomized phase III trials

## TL;DR

Adding immune checkpoint inhibitors to chemotherapy improves cancer response in gastric patients but increases side effects.

## Contribution

This study provides a meta-analysis showing that combining ICIs with chemotherapy improves pathological response in locally advanced gastric cancer.

## Key findings

- Combining ICIs with chemotherapy significantly increased pathological complete response rates.
- Grade ≥3 toxicities were higher with ICI-chemotherapy combinations.
- Excluding a specific trial reduced the estimated toxicity risk.

## Abstract

Phase III trials evaluating perioperative immune checkpoint inhibitors (ICIs) combined with chemotherapy for locally advanced gastric/gastroesophageal junction (G/GEJ) adenocarcinoma have reported mixed efficacy results. We performed a systematic review and meta-analysis to synthesize the current evidence.

We systematically searched for randomized phase III trials comparing perioperative ICI plus chemotherapy with chemotherapy alone. Only phase III RCTs were included to minimize bias and avoid overestimation from phase II studies (see Methods for rationale).Primary outcomes were pathological complete response (pCR). Secondary outcomes were grade ≥3 treatment-related adverse events (TRAEs). Pooled odds ratios (ORs) and hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated using random-effects models for pCR due to clinical heterogeneity,fixed-effect models were used for safety as sensitivity. Heterogeneity was assessed using the I² statistic.

Three trials (DRAGON IV, MATTERHORN, KEYNOTE-585) involving 2,271 patients were included. The addition of ICIs significantly improved the pCR rate (random-effects OR 4.04, 95% CI 2.60 to 6.27; p < 0.00001; I² = 49%). Grade ≥3 TRAEs were numerically higher in the ICI-chemotherapy group (OR 1.61, 95% CI 0.96 to 2.71; p = 0.07; I² = 77%).Sensitivity analysis excluding the TKI-containing DRAGON IV trial attenuated the toxicity estimate (OR 1.23,95% CI 0.71-2.14;I2 = 58%).

The incorporation of ICIs into perioperative chemotherapy for locally advanced G/GEJ adenocarcinoma significantly enhances pathological response, albeit with an increased risk of grade ≥3 toxicities. These findings support perioperative ICI-chemotherapy as a promising and emerging strategy, pending confirmation from mature overall survival data.