Tailored Treatment Strategy for Locally Advanced Rectal Carcinoma Based on the Tumor Response to Induction Chemotherapy: Preliminary Results of the French Phase II Multicenter GRECCAR4 Trial

BACKGROUND: Preoperative radiochemotherapy and total mesorectal excision are the standard-of-care for locally advanced rectal carcinoma, but some patients could be over-or undertreated. OBJECTIVE: This study aimed to assess the feasibility of radiochemotherapy tailored based on the tumor response to induction chemotherapy (FOLFIRINOX) to obtain a minimum R0 resection rate of 90% in the 4 arms of the study. DESIGN: This study is a multicenter randomized trial (NCT01333709). SETTING: This study was conducted at 16 French cancer specialty centers. PATIENTS: Two hundred six patients with locally advanced rectal carcinoma were enrolled between 2011 and 2014. INTERVENTIONS: Good responders (>= 75% tumor volume reduction) were randomly assigned to immediate surgery (arm A) or standard radiochemotherapy (Cap 50:50 Gy irradiation and 1600 mg/m(2) oral capecitabine daily) plus surgery (arm B). Poor responders were randomly assigned to Cap 50 (arm C) or intensive radiochemotherapy (Cap 60, 60 Gy irradiation, arm D) before surgery. OUTCOME MEASURES: The primary end point was a R0 resection rate (circumferential resection margin >1 mm). STATISTICAL CONSIDERATIONS: The experimental strategies were to be considered effective if at least 28 successes (R0 resection) among 31 patients in each arm of stratum I and 34 successes among 40 patients in each arm of stratum II were reported (Simon 2-stage design). RESULTS: After induction treatment (good compliance), 194 patients were classified as good (n = 30, 15%) or poor (n = 164, 85%) responders who were included in arms A and B (16 and 14 patients) and arms C and D (113 and 51 patients). The trial was prematurely stopped because of low accrual in arms A and B and recruitment completion in arms C and D. Data from 133 randomly assigned patients were analyzed: 11, 19, 52, and 51 patients in arms A, B, C, and D. Good responders had smaller tumors than poor responders (23 cm(3) vs 45 cm(3); p < 0.001). The surgical procedure was similar among groups. The R0 resection rates [90% CI] were 100% [70-100], 100% [85-100], 83% [72-91], and 88% [77-95]. Among the first 40 patients, 34 successes were reported in arms C and D (85% R0 resection rate). The circumferential resection margin <= 1 rates were 0%, 0%, 12%, and 5% in arms A, B, C, and D. The rate of transformation from positive to negative circumferential resection margin was 93%. LIMITATIONS: There was low accrual in arms A and B. CONCLUSION: Tailoring preoperative radiochemotherapy based on the induction treatment response appears safe for poor responders and promising for good responders. Longterm clinical results are needed to confirm its efficacy. See Video Abstract at http://links.lww.com/DCR/A359.