Pathologic complete response may not represent the optimal surrogate for survival after preoperative therapy for esophageal cancer

Background We designed a phase II trial to examine the benefit of preoperative hyperfractionated radiation therapy (XRT) and concurrent chemotherapy for patients with locally advanced esophageal cancer (LAEC). Aim of Study The pathologic complete response (pCR) was the primary endpoint to estimate efficacy. Methods Twenty-three patients with LAEC received twice-daily XRT during wk 1 and 5 once-daily XRT during wk 2–4 (59 Gy). Cisplatin (100 mg/m²) was given on d 1, while 5-fluorouracil (1000 mg/m²) was given by continuus infusion the first and fifth weeks of the XRT. Results The pCR for the 19 patients undergoing esophagectomy was 16%. The study was closed at the interim analysis having not met the required minimum pCR rate of 20%. Hematologic toxicities consisted of grades III and IV neutropenia observed in 33% and 14% of patients, respectively. Grade III nausea and vomiting was seen in 38% of patients. One grade V pulmonary toxicity occurred. The median survival was 44.6 mo with 65% of patients alive at 2 yr. Conclusions The pCR rate in this trial did not meet the predetermined statistical minimum. With the encouraging 2-yr survival, it is not clear that pCR is a reliable surrogate endpoint to discern treatment efficacy.