A phase II trial of preoperative concurrent chemotherapy/radiation therapy plus bevacizumab/erlotinib in the treatment of localized esophageal cancer.

PURPOSE To evaluate the efficacy of bevacizumab (Avastin, Genentech) and erlotinib (Tarceva, Genentech/Roche) when added to preoperative chemoradiation therapy with paclitaxel, carboplatin, and infusional 5-fluorouracil (5-FU) in the treatment of localized cancers of the esophagus or gastroesophageal (GE) junction. The primary endpoint was the pathologic complete response (pCR) rate. METHODS Eligible patients had previously untreated localized squamous cell, adenocarcinoma, or adenosquamous carcinoma of the esophagus or GE junction, and were considered surgical candidates at enrollment. Daily erlotinib (100 mg orally) was administered on days 1-42 of preoperative treatment. Patients received paclitaxel (200 mg/m2 intravenously [IV]), carboplatin (area under the curve [AUC] 5.0 IV), and bevacizumab (15 mg/kg IV) on days 1 and 22, and 5-FU by continuous infusion (225 mg/m2/day IV) on days 1-35, with radiation therapy in 1.8-Gy single fractions, Monday-Friday (to a total of 45 Gy). Those who were deemed surgical candidates proceeded to resection during weeks 12-14. RESULTS Between February 2007 and September 2009, 62 patients (median age, 64 years; 92% male; 94% adenocarcinoma) were enrolled; 44 patients (71%) completed neoadjuvant treatment and proceeded to surgery. Eighteen patients (29%) achieved pCR, with partial pathologic remission in an additional 22 patients (35%). Common grade 3/4 toxicities included leukopenia (64%), neutropenia (44%), mucositis/stomatitis (42%), diarrhea (27%), and esophagitis (27%). There were 40 instances of treatment-related hospitalization, and 2 postoperative deaths. CONCLUSIONS The addition of bevacizumab and erlotinib to neoadjuvant chemoradiation did not demonstrate survival benefit or improved pCR rate over similar regimens. While the overall rates of toxicity were not increased, targeted agent-specific toxicity was evident. Further study of this specific regimen is not warranted.