Interaction between severe postoperative complications and adjuvant chemotherapy on overall survival after upfront resection of pancreatic ductal adenocarcinoma.

Abstract

Objective: Severe postoperative complications (Clavien-Dindo grade ≥3) occur in 10%-30% of patients after curative resection of pancreatic ductal adenocarcinoma (PDAC). Whether adjuvant chemotherapy modifies their impact on long-term survival remains unclear. This study aimed to systematically evaluate the interaction between severe complications and adjuvant chemotherapy after PDAC resection.

Methods: We retrospectively analyzed 383 patients with pathologically confirmed PDAC who underwent upfront curative-intent resection at our center between January 2018 and January 2024. Our institutional database initially contained 398 surgical records; 15 records meeting the R2 margin-status or M1-metastasis exclusion criteria were excluded, leaving 383 patients for the primary analysis. Median overall survival (OS) was estimated using the Kaplan-Meier method, and group comparisons were performed with the log-rank test. Covariates adjusted for in the multivariable Cox regression model included age, sex, BMI, tumor differentiation, margin status, vascular invasion, perineural invasion, lymph node ratio, and tumor size. Stratified Cox regression and interaction-effect models were further constructed. Exploratory recurrence-free survival (RFS) analyses defined an event as recurrence or death from any cause. Sensitivity analyses excluded patients with OS <60 days and, additionally, evaluated a 90-day threshold.

Results: Severe complications occurred in 40 of 383 patients (10.4%). Adjuvant chemotherapy was received by 16 of 40 patients (40.0%) in the severe-complication group and 205 of 343 patients (59.8%) in the non-severe-complication group (P = 0.026). Among patients who did not receive chemotherapy, median OS was significantly shorter in the severe-complication group than in the non-severe-complication group (7.9 vs. 14.6 months, P < 0.001); among those who received chemotherapy, the difference was not statistically significant (18.8 vs. 20.5 months, P = 0.445). In the interaction model, severe complications were associated with increased mortality in the no-chemotherapy stratum (HR: 4.18, 95% CI: 2.49-7.01, P < 0.001) but not in the chemotherapy stratum (HR: 1.24, 95% CI: 0.66-2.34, P = 0.511). In separate stratified Cox models adjusted for the same covariates, adjuvant chemotherapy was associated with lower mortality in the severe-complication group (HR: 0.15, 95% CI: 0.04-0.57, P = 0.005), whereas the association was not statistically significant in the non-severe-complication group (HR: 0.76, 95% CI: 0.56-1.03, P = 0.075). Exploratory RFS analysis showed shorter RFS among patients with severe complications (median 6.4 vs. 11.1 months; adjusted HR: 1.58, 95% CI: 1.09-2.28, P = 0.016).

Conclusions: In this retrospective cohort, severe complications were associated with worse OS among patients who did not receive adjuvant chemotherapy, whereas the association was not statistically significant among those who received chemotherapy. Because the severe-complication subgroup was small and treatment receipt was subject to postoperative selection, the interaction should be considered hypothesis-generating. These findings do not establish that adjuvant chemotherapy offsets the adverse prognostic association of severe complications and require prospective validation before informing treatment decisions.