Clinical predictors of pathologically response after neoadjuvant chemoradiotherapy for esophageal squamous cell carcinoma: long term outcomes of a phase II study.

Abstract

A pathologically complete response (pCR) or near pCR to neoadjuvant chemoradiotherapy (NCRT) might imply a better prognosis in patients with esophageal cancer. The aim of the study is to identify clinical factors associated with a pCR or near pCR. We retrospectively analyzed 40 patients with radical esophagectomy after NCRT for esophageal squamous cell carcinoma (ESCC) from January 2001 to December 2006 in Sun Yat-sen University Cancer Center. Clinical factors included age, gender, weight loss, dysphagia, drinking status, smoking status, tumor location, tumor length, tumor grade, cT status, cN status, the regimen of chemotherapy and the interval between NCRT and surgery as potential predictors for a pCR or near pCR. Logistic regression was used to estimate the independent factors for a pCR or near pCR. After surgical resection, 22.5% of the patients obtained the pCR. Patients with pCR had a better prognosis than those with non-pCR. However, there was no statistically significantly difference between the two groups (P=0.124). We separated the patients into pCR or near pCR (good responders, GRs) and poor responders (PR) based on the histology. GR showed better overall survival (OS) than PR (P=0.014). Univariate analysis indicated that short tumor length, good tumor grade and never drinking were associated with GR to NCRT. Using logistic regression analysis, good tumor grade was the only independent factor for the GR to NCRT (P=0.021). Cox regression revealed that weight loss, drinking status and GR were independent factors in ESCC patients with radical esophagectomy after NCRT. Our study indicated that good tumor grade were an independent significant factor for the GR to NCRT. Weight loss, drinking status and GR were independent factors in patients with radical esophagectomy after NCRT. GR may improve OS of ESCC patients receiving NCRT.