Multicenter randomized phase II study of chemoradiation (CRT) followed by surgery (S) and chemotherapy (CT) versus induction CT followed by CRT and S in high-risk rectal cancer: GCR-3 final efficacy and safety results

Abstract

4103 Background: In locally advanced RC the optimal therapeutic sequence remains an important clinical question. Induction CT prior to CRT and S may be associated with better efficacy and compliance. Methods: Eligible pts had medium or distal high risk RC defined by MRI and/or US: Tumors within 2mm of mesorectal fascia, distal T3 at/below levators, resectable T4 and T3N+. Pts, stratified by center, were randomized assigned to receive either Arm A : capecitabine (Cap) 825 mg/m2 BID 5 d/w, oxaliplatin (Ox) 50 mg/m2 IV weekly x 5 and concomitant RT: 50.4 Gy in 28 fractions. S was planned 5–6 w after CRT. Post-op four cycles of Cap 1,000 mg/m2 bid days 1 to 14; Ox 130 mg/m2 day 1 or Arm B: Induction CapOx followed by CRT and S. Two parallel, Simon 2-stage designs: α=0.05 β=0.1; 24 evaluable pts/arm 1st stage and 54 pts/arm for 2nd stage. Primary endpoint: pathological complete response (pCR). Secondary endpoints included toxicity and treatment compliance. Results: 108 Pts were randomly assigned (arm A/B, 52/56), and 103 were assessable (49/54) from 14 sites. Median age 62/60 years, Male 65/70%. During treatment period 6 pts died A/B: 2 vascular, 1 suicide/ 3 post-op. Pts with any grade ¾ toxicity during CRT were arm A/B: 29% (14/49) and 23% (12/53). Any grade ¾ toxicity during adjuvant/induction CT were 51% (19/37) and 17% (9/54); χ2,p= 0.0004. On an intent-to-treat basis the pCR for Arm A/B was achieved in seven (13.5%; 95% CI, 5.6%-25.8%) and eight (14.3%; 95% CI, 6.4%-26.2%). R0 resections were achieved in 92% (45/49) and 88% (48/54). 51% (25/49) and 93% (50/54) received all four cycles of adjuvant/induction CT (χ2;p<0.0001). Relative Median Dose intensity of adjuvant /induction CT was 0.74/0.96 (Wilcoxon; p<0.0001) for Cap and 0.75/1.0 (Wilcoxon; p<0.0001) for Ox. Conclusions: Induction CT prior to CRT has more favorable compliance and toxicity profiles. Furthermore, there is no compromise in pCR and R0 resection rates. Larger trials evaluating this strategy are justified. [Table: see text]