Efficacy endpoints of radiation therapy group protocol 0247: a randomized, phase 2 study of neoadjuvant radiation therapy plus concurrent capecitabine and irinotecan or capecitabine and oxaliplatin for patients with locally advanced rectal cancer.

PURPOSE: To report secondary efficacy endpoints of Radiation Therapy Oncology Group protocol 0247, primary endpoint analysis of which demonstrated that preoperative radiation therapy (RT) with capecitabine plus oxaliplatin achieved a pathologic complete remission prespecified threshold (21%) to merit further study, whereas RT with capecitabine plus irinotecan did not (10%).

METHODS AND MATERIALS: A randomized, phase 2 trial evaluated preoperative RT (50.4 Gy in 1.8-Gy fractions) with 2 concurrent chemotherapy regimens: (1) capecitabine (1200 mg/m(2)/d Monday-Friday) plus irinotecan (50 mg/m(2)/wk × 4); and (2) capecitabine (1650 mg/m(2)/d Monday-Friday) plus oxaliplatin (50 mg/m(2)/wk × 5) for clinical T3 or T4 rectal cancer. Surgery was performed 4 to 8 weeks after chemoradiation, then 4 to 6 weeks later, adjuvant chemotherapy (oxaliplatin 85 mg/m(2); leucovorin 400 mg/m(2); 5-fluorouracil 400 mg/m(2); 5-fluorouracil 2400 mg/m(2)) every 2 weeks × 9. Disease-free survival (DFS) and overall survival (OS) were estimated univariately by the Kaplan-Meier method. Local-regional failure (LRF), distant failure (DF), and second primary failure (SP) were estimated by the cumulative incidence method. No statistical comparisons were made between arms because each was evaluated individually.

RESULTS: A total of 104 patients (median age, 57 years) were treated; characteristics were similar for both arms. Median follow-up for RT with capecitabine/irinotecan arm was 3.77 years and for RT with capecitabine/oxaliplatin arm was 3.97 years. Four-year DFS, OS, LRF, DF, and SP estimates for capecitabine/irinotecan arm were 68%, 85%, 16%, 24%, and 2%, respectively. The 4-year DFS, OS, LRF, DF, and SP failure estimates for capecitabine/oxaliplatin arm were 62%, 75%, 18%, 30%, and 6%, respectively.

CONCLUSIONS: Efficacy results for both arms are similar to other reported studies but suggest that pathologic complete remission is an unsuitable surrogate for traditional survival metrics of clinical outcome. Although it remains uncertain whether the addition of a second cytotoxic agent enhances the effectiveness of fluorouracil plus RT, these results suggest that further study of irinotecan may be warranted.