A pathological study of tumour regression in oesophageal adenocarcinoma treated with preoperative chemoradiotherapy.

AIMS To measure residual tumour in oesophageal adenocarcinoma treated with preoperative chemoradiotherapy, to correlate specific pathological variables with survival, and to describe morphological changes in tumour and non-neoplastic tissue resulting from preoperative treatment. METHODS Resection specimens from 47 cases of oesophageal adenocarcinoma treated with preoperative 5-fluorouracil/cisplatin and radiotherapy were reviewed. Residual tumour was assessed in terms of tumour regression grade (TRG), pTNM stage, lymphovascular space invasion, and resection margin involvement. Survival analysis was performed using the Kaplan-Meier method and log rank test. Cox's proportional hazard model was used for multivariate analysis. RESULTS A complete pathological response (TRG1) was present in eight cases. The absence of residual tumour was confirmed by negative immunohistochemical staining for MNF116. Tumour corresponding to TRG2 was present in five cases, to TRG3 in nine, to TRG4 in 22, and to TRG5 in three. By multivariate analysis, pN0 status (n = 35) had a positive effect on survival (p = 0.04) and TRG had no significant effect on survival (p = 0.06). Patients with pN0 tumours had a median survival of 48 months versus eight months for those with pN1 tumours (log rank test, p < 0.0001). We found that giant fibroblasts were discernible from single large residual tumour cells on haematoxylin and eosin alone. CONCLUSION Response to preoperative chemoradiotherapy in oesophageal adenocarcinoma is variable. Although there are as yet no reliable predictors of response to treatment, patients who are identified at diagnosis as having negative loco-regional lymph nodes should benefit considerably from this treatment approach.