Does neoadjuvant FOLFOX chemotherapy improve the prognosis of high-risk Stage II and III colon cancers? Three years' follow-up results of the PRODIGE 22 phase II randomized multicentre trial.

Abstract

Neoadjuvant chemotherapy has proven valuable in locally advanced resectable colon cancer (CC) but its effect on oncological outcomes is uncertain. The aim of the present paper was to report 3-year oncological outcomes, representing the secondary endpoints of the PRODIGE 22 trial. PRODIGE 22 was a randomized multicentre phase II trial in high-risk T3, T4 and/or N2 CC patients on CT scan. Patients were randomized between 6months of adjuvant FOLFOX (upfront surgery) or perioperative FOLFOX (four cycles before surgery and eight cycles after; FOLFOX perioperative). In wild-type RAS patients, a third arm testing perioperative FOLFOX-cetuximab was added. The primary endpoint was the tumour regression grade. Secondary endpoints were 3-year overall survival (OS), disease-free survival (DFS), recurrence-free survival (RFS) and time to recurrence (TTR). Overall, 120 patients were enrolled. At interim analysis, the FOLFOX-cetuximab arm was stopped for futility. The remaining 104 patients represented our intention-to-treat population. In the perioperative group, 96% received the scheduled four neoadjuvant cycles and all but one had adjuvant FOLFOX for eight cycles. In the control arm, 38 (73%) patients received adjuvant FOLFOX. The median follow-up was 54.3months. Three-year OS was 90.4% in both arms [hazard ratio (HR)=0.85], 3-year DFS, RFS and TTR were, respectively, 76.8% and 69.2% (HR=0.94), 73% and 69.2% (HR=0.86) and 82% and 72% (HR=0.67) in the perioperative and control arms, respectively. Forest plots did not show any subgroup with significant difference for survival outcomes. No benefit from adding cetuximab was observed. Perioperative FOLFOX has no detrimental effect on long-term oncological outcomes and may be an option for some patients with locally advanced CC.