A three-step strategy of induction chemotherapy then chemoradiation followed by surgery in patients with potentially resectable carcinoma of the esophagus or gastroesophageal junction

BACKGROUND, Patients with locoregional carcinoma of the esophagus or gastro esophageal junction have a poor survival rate after surgery. Preoperative c hemotherapy or chemoradiotherapy has not improved the outcome for these pat ients. Our study was designed to assess the feasibility of preoperative ind uction combination chemotherapy in addition to chemoradiotherapy to improve the curative resection rate, local control, and survival. PATIENTS AND METHODS. Patients having histologic proof of localized carcino ma (either squamous cell carcinoma or adenocarcinoma) of the esophagus or g astroesophageal junction underwent full classification including endoscopic ultrasonography (EUS). Patients first received up to two courses of induct ion chemotherapy consisting of 5-fluorouracil at 750 mg/m(2)/day as continu ous infusion on Days 1-5, cisplatin at 15 mg/m(2)/day as an intravenous bol us on Daps 1-5, and paclitaxel at 200 mg/m(2) as a 24-hour intravenous infu sion on Day 1. The second course was repeated on Day 29. This was followed by radiotherapy (45 grays in 25 fractions) and concurrent admission of 5-fl uorouracil (300 mg/m(2)/day as a continuous infusion 5 days/week) and cispl atin (20 mg/m(2) on Days 1-5 of radiotherapy). After chemoradiotherapy, pat ients underwent surgery. The feasibility of this approach, curative resecti on rates, patient survival, and patterns of failure were assessed. RESULTS. Thirty-seven of 38 patients enrolled were evaluable for toxicity a nd survival. Adenocarcinoma and distal esophageal location of carcinoma wer e observed frequently. Thirty-five (95%) of the 37 patients underwent surge ry, all of whom had an RO (curative] resection. A pathologic complete respo nse was noted in 11 (30%) of the 37 total patients. In addition, 5 patients (14%) had only microscopic carcinoma. According to EUS classification, 31 [89%) of the 35 patients who underwent surgery had a T3 carcinoma whereas a ccording to pathologic classification only 3 (9%) had a T3 carcinoma (P les s than or equal to 0.01). Similarly, according to EUS classification, 23 pa tients (66%) had an N1 carcinoma, whereas according to pathologic classific ation only 7 patients (20%) had an N1 carcinoma (P less than or equal to 0. 01). At a median follow-up of 20 months (minimum follow-up, 13+ months; max imum follow-up, 36+ months), the median survival duration for the 37 patien ts had not pet been reached. In addition, there were two deaths related to surgery. CONCLUSIONS. These data show that the three-step strategy of preoperative p aclitaxel-based induction chemotherapy then chemoradiotherapy followed by s urgery is feasible and appears quite active in patients having locoregional carcinoma of the esophagus or gastroesophageal junction. Future investigat ions should focus on substituting cisplatin with less toxic agents and incl uding more systemic therapy with newer classes of agents. (C) 2001 American Cancer Society.