Impact of center-level protocol adherence on neoadjuvant treatment outcome in pancreatic ductal adenocarcinoma.

INTRODUCTION: Chemotherapy dose reductions and delays reduce relative dose intensity (RDI) but their prognostic impact in pancreatic ductal adenocarcinoma (PDAC) remains unclear and may be influenced by heterogeneous adherence to protocol guidelines. Hence, we investigated the impact of RDI, protocol violations, and center-level protocol adherence on Event Free Survival (EFS). MATERIALS AND METHODS: This is a post hoc exploratory analysis of the preoperative part of the phase III CASSANDRA-PACT21 trial, including 260 patients with resectable and borderline resectable (R/BR) PDAC treated with neoadjuvant mFOLFIRINOX (fluorouracil, leucovorin, irinotecan, oxaliplatin) or PAXG (cisplatin, nab-paclitaxel, capecitabine, and gemcitabine) at 13 Italian centers. Chemotherapy RDI was calculated for individual agents, for platinum salts and for fluoropyrimidines grouped together. Protocol violations were defined as unjustified dose reductions or delays according to trial guidelines. RESULTS: Protocol violations were significantly more frequent in the PAXG arm compared to mFOLFIRINOX (27% vs 16%; p = 0.047). A significant association with EFS was observed only for platinum salts RDI (10.3 vs 15.8 months for RDI<or≥80%, respectively; p = 0.039; HR 0.73, 95%CI, 0.54-0.98). Protocol violations per se did not affect EFS, whereas violations resulting in reduced platinum salts RDI led to shorter EFS (9.3 versus 15.1 months; p = 0.014; HR 0.64, 95%CI 0.45-0.91). Patients treated at higher violation rate centers experienced significantly shorter EFS than those at lower violation centers (10.3 vs 14.5 months, respectively; p = 0.040; HR 0.70, 95%CI 0.50-0.99). CONCLUSIONS: The association of platinum RDI with EFS was observed only in univariable analysis while center-level protocol adherence was independently related to EFS in multivariable analysis.