Background: With the development of chemoradiotherapy for esophageal cancer, the complete response (CR) rate would become an important surrogate end-point. However, the Response Evaluation Criteria in Solid Tumors (RECIST) do not provide criteria for a response at the primary site of esophageal cancer. The objective of this study was to assess the validity of the endoscopic CR criteria for the primary site of esophageal cancer treated with chemoradiotherapy.
Methods: We reviewed 139 patients with T1-T4, N0-1, M0-1a esophageal cancer treated with definitive chemoradiotherapy from August 1992 to April 1999. CR was tentatively defined upon endoscopic observation of the entire esophagus as: (i) disappearance of the tumor lesion; (ii) disappearance of ulceration (slough); and (iii) absence of cancer cells in biopsy specimens.
Results: CR at the primary site (primary-CR) was achieved in 80 patients (58%). Of these, 71 (89%) were evaluated as having primary-CR within 6 months from the start of therapy. With a median follow-up of 53 months, a remarkable difference in the 5-year survival rate was observed between patients evaluated as having primary-CR and having non-CR (46 and 6%, P < 0.0001). Local failure was observed in 15 patients and the local control rate in patients with primary-CR was 78% at 5 years.
Conclusions: These criteria appear to represent an appropriate surrogate end-point because they are convenient to apply, require only a short time before a primary-CR can be declared and their fulfillment can predict long-term survival. It is recommended that RECIST include precise endoscopic findings for primary lesions in esophageal cancer in the CR criteria.