LAUSR

BACKGROUND In the treatment of esophageal squamous cell carcinoma (ESCC), the optimal use of 18fluorodeoxyglucose positron emission tomography (PET) in measuring metabolic tumor response is undetermined. We launched a phase II trial to evaluate early metabolic response to one-cycle induction chemotherapy in patients with locally advanced ESCC. METHODS ESCC patients in stage classification T3N0, N1M0, or M1a (American Joint Committee on Cancer, 6th edition) received one-cycle chemotherapy comprising paclitaxel, cisplatin, and 24-h infusional 5-fluorouracil and leucovorin on days 1 and 8, followed by neoadjuvant chemoradiotherapy, 40 Gy, with paclitaxel/cisplatin and then esophagectomy. PET was performed at baseline and day 14 of chemotherapy. The primary endpoint was pathologic complete response (pCR). We hypothesized early metabolic responders with >35% reduction in maximum standardized uptake value (SUVmax), would have better pCR Results. RESULTS Sixty-six patients were enrolled. The median progression-free survival (PFS) and overall survival (OS) were 16 months (95% confidence interval [CI], 9-27) and 22 months (16-40), respectively. The early metabolic response rate was 55%; and the pCR rate was 34% in the esophagectomy population. The early metabolic response was not associated with pCR or survival. In an exploratory analysis, the postchemotherapy SUVmax was an independent prognostic factor for pCR, PFS, and OS. CONCLUSION Our study failed to validate the predefined early metabolic response for pCR to neoadjuvant chemoradiotherapy in locally advanced ESCC patients. However, postchemotherapy SUVmax may be prognostic and predictive, and warrants further study.