Timing of Adjuvant Chemotherapy After Gastrectomy in Patients With Stage II to III Gastric Cancer: A Target Trial Emulation Study.

PURPOSE: Uncertainty persists regarding the optimal interval between D2 gastrectomy and initiation of adjuvant chemotherapy (AC) for stage II to III gastric cancer (GC). We investigated whether delaying treatment beyond 6 weeks compromises overall survival.

MATERIALS AND

METHODS

Using multicenter observational data (2010-2022), we emulated a pragmatic noninferiority target trial to compare AC initiation at ≤6 versus 7-12 weeks postgastrectomy. Eligible patients had received curative-intent surgery for stage II to III GC. The AC regimen comprised platinum-based agents plus fluoropyrimidines. The primary end point was 5-year all-cause mortality from surgery to death or administrative censoring (30 June 2025). The Clone-censor-weighting approach was used, and inverse probability weighting achieved balance across baseline and postoperative time-varying covariates (standardized mean difference <0.1). Weighted pooled logistic regression estimated cumulative mortality; risk differences (RDs) and risk ratios (RRs) were computed from 500 bootstrap samples. The prespecified noninferiority margin was an absolute 10% increase in 5-year mortality.

RESULT: Among 1,637 eligible patients, 35.9% initiated AC within 6 weeks (median 37 days), whereas 17.5% began between 7 and 12 weeks postoperatively. Delayed initiation (7-12 weeks) met the noninferiority criterion: the absolute RD in 5-year mortality was -0.52% (95% CI, -7.95 to 7.41), with the upper confidence limit below the 10% margin. The noninferiority finding was robust to weight trimming and to alternative imputation strategies. No significant difference in mortality was observed across clinically relevant subgroups.

CONCLUSION

Initiating AC 7-12 weeks after D2 gastrectomy appeared to be noninferior to initiation within 6 weeks with respect to 5-year mortality in stage II to III GC; however, residual confounding cannot be excluded and the findings require validation in further studies.