Effect of preoperative concurrent chemoradiotherapy on survival of patients with resectable esophageal or esophagogastric junction cancer: Results from a multicenter randomized phase III study.

4004 Background: Based on encouraging results of a phase II study we initiated a phase III study comparing preoperative chemoradiotherapy (CRT) followed by surgery versus surgery alone in patients (pts) with esophageal or esophagogastric junction cancer. Methods: Pts with resectable (T2-3N0-1M0) tumors received preoperative CRT consisting of weekly administrations of paclitaxel 50 mg/m2 and carboplatin AUC = 2 for 5 weeks and concurrent radiotherapy (41.4 Gy in 23 fractions, 5 days per week) followed by surgery versus surgery alone. Stratification parameters included performance status, histology, lymph node status. Results: Between end 2004 and 2008 363 pts were enrolled: male/female 284/79, median age 60 years (36-79), median WHO performance status 0 (0-1), adeno/squamous/other carcinoma 273/86/4. Major toxicities (grade ≥ 3) in the CRT arm: hematologic - leukopenia 7%, nonhematological toxicities were all below 5%. In 162 of the 188 pts (86%) randomized for surgery alone a resection could be performed versus 158 of the 175 pts (90%) who received CRT. The reported R0 resection rate was 92.3% in the CRT arm versus 64.9% in the surgery alone arm. In 132 revised specimens the pCR rate was 32.6%. In-hospital mortality was 3.7% in the surgery alone arm versus 3.8% in the CRT arm. With a median follow-up of 32 months 70 and 97 pts had died in the CRT group versus surgery alone group, respectively. The overall survival was significantly better (p = 0.011) in the group of pts treated with CRT (HR 0.67 [95% CI 0.50-0.92]). Median survival was 49 months in the CRT arm versus 26 months in the surgery alone arm. One, 2 and 3 years survival rates are 82%, 67% and 59% in the CRT arm and 70%, 52% and 48% in the surgery alone arm. Conclusions: Weekly administrations of carboplatin and paclitaxel with concurrent radiotherapy improves overall survival compared to surgery alone. Since toxicity was also acceptable, this regimen can be considered the standard of care for patients with resectable esophageal or esophagogastric junction cancer. No significant financial relationships to disclose.