Although phase II trials have shown that adding docetaxel to S-1/oxaliplatin (DOS) improves pathological response in HER2-negative locally advanced gastric cancer (LAGC), robust real-world evidence comparing DOS with the standard S-1/oxaliplatin (SOX) regimen is lacking. We retrospectively reviewed 1283 consecutive patients with cT2N+ or cT3-4NanyM0, HER2-negative gastric adenocarcinoma treated with ≥2 cycles of DOS (n = 461) or SOX (n = 822) at four high-volume Chinese centers between 2010 and 2024. To address confounding, we applied propensity-score matching (PSM, 1:1) and inverse probability-of-treatment weighting (IPTW). Major pathological response (MPR) was the primary endpoint; secondary endpoints included disease-free survival (DFS), overall survival (OS), and perioperative safety. DOS yielded a higher MPR rate than SOX (25.7% vs. 17.8%; OR 1.59, 95% CI 1.14-2.23; E-value =1.83 in PSM cohort) without increasing severe postoperative complications (2.2% each). In multivariable Cox models, DOS reduced progression risk (HR 0.74, 95% CI 0.59-0.93 in IPTW cohort), while OS was similar after a median 34.8-month follow-up. The DFS advantage was consistent across prespecified subgroups, particularly in patients aged ≥65 years, those with cT2-3 or cN0-1 disease, and proximal tumors. These findings indicate that neoadjuvant DOS induces deeper pathological regression and longer DFS than SOX without compromising surgical safety, supporting taxane-based triplets as a preferred perioperative option for HER2-negative LAGC; longer follow-up will clarify any impact on OS.
Keywords: DOS; SOX; gastric cancer; neoadjuvant chemotherapy; real‐world study.
© 2025 UICC.