Chronicle: a Phase III Trial in Locally Advanced Rectal Cancer After Neoadjuvant Chemoradiation Randomising Postoperative Adjuvant Capecitabine/Oxaliplatin Versus Control

Abstract

Background: In stage III colon cancer oxaliplatin/5FU-based adjuvant chemotherapy (FOLFOX) improves disease-free survival (DFS) and overall survival (OS). Despite a lack of evidence in rectal cancer, guidelines recommend postoperative chemotherapy following neoadjuvant 5FU-based chemoradiation. We examined the benefit of 6 cycles of chemotherapy with capecitabine and oxaliplatin (XELOX) following neoadjuvant chemoradiation in curatively resected patients compared to observation alone. Methods: Eligible patients were aged >18 years of age with histologically confirmed rectal carcinoma, defined as a tumour located <15 cm from the anal verge or below the peritoneal reflection. Complete resection of the primary tumour was performed before patients were randomly assigned to either observation or XELOX for 6 cycles over 4 months, to start within 12 weeks of surgery and with no evidence of metastatic disease. Patients were permitted to receive up to 6 weeks of neoadjuvant fluoropyrimidine-based chemoradiation with a planned total dose of at least 45Gy. The primary endpoint was DFS; secondary endpoints were toxicity and OS. 390 patients were required in each arm to detect a difference in 3-year DFS from 40% to 50.5%, with 85% power and two-sided 5% significance level. Results: The study closed prematurely in 2008 because of poor accrual; of the intended 800 patients, 113 were randomly assigned to either observation (Group 1, n = 59) or XELOX for 6 cycles over 4 months (Group 2, n = 54). Compliance was poor, 93% allocated chemotherapy started but only 48% completed 6 cycles of chemotherapy. Dose reductions according to protocol in Group 2 were 39%, and levels of G3/G4 toxicity 40%. After a median follow-up of 44.8 months, 16 patients (27%) in Group 1 had relapsed or died compared with 12 patients (22%) in Group 2. The 3-year DFS rate was 78% with XELOX and 71% with observation (hazard ratio [HR] for DFS = 0.80; 95% CI, 0.37-1.68; P = 0.56). The 3-year OS for XELOX and observation were 89% and 88% respectively (HR for OS = 1.18; 95% CI, 0.42-3.37; P = 0.75). Conclusion: Post-operative randomisation between combination chemotherapy and observation led to poor accrual, either because patients did not wish to embark on a new treatment plan, or clinicians had decided whether or not to administer post-operative chemotherapy. The total number of patients was too small to detect a difference in DFS or OS and this was compounded by poor compliance with chemotherapy. Trials that test the role of neoadjuvant chemotherapy should be explored, and future trials investigating complex treatment pathways should look at alternative strategies to improve recruitment.