Efficacy and safety of commercial Chinese polyherbal preparation combined with oxaliplatin-based chemotherapy for gastric cancer: a systematic review and network meta-analysis

Objective To evaluate the efficacy, safety, and quality of evidence for commercial Chinese polyherbal preparations (CCPPs) in combination with oxaliplatin-based chemotherapy (OX) for gastric cancer (GC). Methods We searched six databases for randomized controlled trials (RCTs) comparing CCPPs plus OX to other active treatments for GC. The risk of bias was assessed using a modified version of Cochrane risk of bias tool. A frequentist network meta-analysis using a random-effects model was performed to estimate the relative effectiveness between treatments. We used GRADE to assess the certainty of evidence, to categories the interventions, and to present the findings. Results One-hundred-and-eighty-nine RCTs involving 15,320 participants that reported 27 CCPPs were identified. Moderate to high certainty evidence showed that Compound Mylabris preparations (Disease Control Rate (DCR): 1.51, 1.13–2.03) is among the most effective CCPPs for improving DCR, while Kangai Injection (objective response rate (ORR): 1.40, 1.28 to 1.77; quality of life (QoL): 1.46, 1.11–1.94), Huachansu preparations (ORR: 1.28, 1.15–1.43), Ya Dan Zi Oil Emulsion Injection (QoL: RR 1.26, 1.06–1.51), Aidi Injection (QoL: 1.30, 1.13–1.50), Lentinan (QoL: 1.27, 1.05–1.54), and Yangzheng Xiaoji Capsules (ORR: 1.36, 1.05–1.77) showed intermediate efficacy for ORR and QoL. Regarding immune function improvement, with moderate to high certainty evidence, Shenmai Injection (CD3 + : 10.03, 1.69 to 18.37; CD4 + : 8.33, 0.64–16.02), Ginseng Polysaccharide Injection (CD3 + : 10.55, 1.89–19.21), Huachansu preparations (CD3 + : 7.42, 2.51–12.34), Kangai Injection (CD3 + : 11.65, 6.81 to 16.50; CD4 + /CD8 + : 0.25, 0.02–0.47), Lentinan (CD4 + : 9.43, 3.78–15.08), Shenqi Fuzheng Injection (CD4 + : 5.72, 3.68–7.76), and Yangzheng Xiaoji Capsules (CD3 + : 9.32, 0.84–17.80) showed improvements in specific immune parameters. Conclusion No CCPP was optimal for all endpoints, Compound Mylabris showed superior tumor response, and Kangai Injection offered the most favorable risk-benefit profile, providing broad efficacy and statistically significantly reducing AEs. The choice of adjuvant CCPP should be individualized based on specific therapeutic priorities, balancing efficacy and safety. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/view/CRD42025646173 .