Association of Preoperative Chemosensitivity With Postoperative Survival in Patients With Resected Gastric Adenocarcinoma

IMPORTANCE: Considering its low completion rate, the survival benefit associated with postoperative chemotherapy (PC) is unclear in patients with resectable gastric adenocarcinoma who received preoperative chemotherapy. OBJECTIVE: To determine whether preoperative chemosensitivity is associated with postoperative survival among patients with resectable gastric adenocarcinoma who receive PC. DESIGN, SETTING, AND PARTICIPANTS: This national, hospital-based cohort study used data from the National Cancer Database, which covers more than 70% newly diagnosed gastric adenocarcinomas in the US, between 2006 and 2017. Participants included patients with clinical stage II or III disease treated with preoperative chemotherapy and curative-intent resection, excluding radiotherapy. Preoperative chemosensitivity was defined as very sensitive (ypT0N0), sensitive (pathological TNM stage less than clinical, excluding ypT0N0), and refractory (pathological greater than or equal to clinical). Data were analyzed in April 2021. EXPOSURES: Receipt of PC or not. MAIN OUTCOMES AND MEASURES: Overall survival from surgical discharge. RESULTS: This study included 2382 patients (1599 men [67%]; median [IQR] age, 63 [54-70] years). Most patients (1524 patients [64%]) received no PC. Most patients (1483 patients [62%]) had refractory disease, followed by sensitive disease (727 patients [31%]) and very sensitive disease (172 patients [7%]). Patients with older age (odds ratio [OR], 0.99; 95% CI, 0.97-1.00), comorbidity (OR, 0.71; 95% CI, 0.57-0.90), longer time from chemotherapy initiation to surgery (OR, 0.99; 95% CI, 0.97-1.00), less sensitivity to preoperative chemotherapy (very sensitive vs refractory OR, 0.58; 95% CI, 0.37-0.89; sensitive vs refractory OR, 0.96; 95% CI, 0.76-1.20), and longer surgical hospitalization (OR, 0.95; 95% CI, 0.93-0.97) had a significantly lower likelihood of receiving PC. PC was not associated with improved survival in the whole group (hazard ratio [HR], 0.88; 95% CI, 0.75-1.02). Patients with refractory disease had the worst survival compared with patients with sensitive disease (HR, 0.39; 95% CI, 0.32-0.46) and those with very sensitive disease (HR, 0.12; 95% CI, 0.07-0.20). Preoperative chemosensitivity was significantly associated with the survival benefit from PC (P for interaction = .03). PC was significantly associated with longer survival in patients with sensitive disease (5-year survival rate, 73.8% in the PC group vs 65.0% in the no PC group; HR, 0.64; 95% CI, 0.46-0.91), but not in those with very sensitive disease (5-year survival rate, 80.0% in the PC group vs 90.8% in the no PC group; HR, 2.45; 95% CI, 0.81-7.43) and those with refractory disease (5-year survival rate, 41.8% in the PC group vs 40.7% in the no PC group; HR, 0.93; 95% CI, 0.79-1.10). CONCLUSIONS AND RELEVANCE: In this cohort study, preoperative chemosensitivity was associated with survival among patients with resectable gastric adenocarcinoma who received PC. These findings may help inform future studies to personalize postoperative therapy.