Importance: The effect of adjuvant chemotherapy following resection of pancreatic adenocarcinoma after preoperative (m)FOLFIRINOX (combination leucovorin calcium [folinic acid], fluorouracil, irinotecan hydrochloride, and oxaliplatin in full or modified dosing) chemotherapy on overall survival (OS) is unclear because current studies do not account for the number of cycles of preoperative chemotherapy and adjuvant chemotherapy regimen.
Objective: To investigate the association of adjuvant chemotherapy following resection of pancreatic adenocarcinoma after preoperative (m)FOLFIRINOX with OS, taking into account the number of cycles of preoperative chemotherapy and adjuvant chemotherapy regimen.
Design, setting, and participants: This retrospective cohort study included patients with localized pancreatic adenocarcinoma treated with 2 to 11 cycles of preoperative (m)FOLFIRINOX followed by resection across 48 centers in 20 countries from 2010 to 2018. Patients who died within 3 months after surgery were excluded (landmark). Data were analyzed from February 1 to December 31, 2023.
Exposures: Preoperative (m)FOLFIRINOX chemotherapy followed by resection and eventually followed by adjuvant chemotherapy.
Main outcomes and measures: The primary outcome was OS, calculated from the 3-month landmark. Cox regression analysis, including interaction analyses, was performed to investigate the association of adjuvant chemotherapy with OS.
Results: Overall, 767 patients were included after resection of pancreatic adenocarcinoma (median [IQR] age, 62 [55-67] years; 404 [52.7%] male). Adjuvant chemotherapy was independently associated with prolonged OS (hazard ratio [HR], 0.66; 95% CI, 0.49-0.87), confirmed by adjusted OS curves. The interaction analysis to assess estimated treatment effect across subgroups was not statistically significant. The forest plot and interaction test suggest that the association of adjuvant chemotherapy was lower among patients receiving 8 or more cycles of preoperative (m)FOLFIRINOX, those who had radiological response, and those with ypN0 disease. Compared to no adjuvant chemotherapy, both adjuvant (m)FOLFIRINOX (HR, 0.57; 95% CI, 0.40-0.80) and other multiagent adjuvant regimens (HR, 0.61; 95% CI, 0.41-0.92) were associated with prolonged OS, whereas single-agent adjuvant chemotherapy was not (HR, 0.75; 95% CI, 0.55-1.03).
Conclusions and relevance: In this cohort study, adjuvant (m)FOLFIRINOX and other multiagent chemotherapy regimens were associated with improved OS following resection of localized pancreatic adenocarcinoma after preoperative (m)FOLFIRINOX, whereas single-agent adjuvant chemotherapy was not. The impact of adjuvant chemotherapy on OS may be lower in subgroups such as patients with 8 or more preoperative cycles of (m)FOLFIRINOX, those having radiological response, and those with ypN0.