Clinicopathologic Predictors of Long Term and Disease-Free Survival in Esophageal Adenocarcinomas with Complete Pathologic Response to Neoadjuvant Chemoradiation
Agoston T Agoston, Yifan Zheng, Raphael Bueno, Gregory Y Lauwers, Robert D Odze, Amitabh Srivastava. Brigham and Women's Hospital, Boston, MA; Massachusetts General Hospital, Boston, MA
Background: Neoadjuvant chemoradiation treatment has improved outcomes for esophageal adenocarcinoma (EAC). Patients with pathologic complete response (pCR) to therapy have favorable long term and disease-free survival but there remains significant heterogeneity in outcome within this group. The aim of this study was to determine clinicopathologic predictors of long term and disease-free survival among patients with EAC who had pCR to neoadjuvant chemoradiation.
Design: 93 patients with EAC who had pCR to neoadjuvant chemoradiation were identified. A pre-determined set of clinicopathologic variables was examined for each case including: age, gender, tumor type and grade (pre-treatment biopsy), tumor location, tumor size (pre-treatment endoscopy), pre-treatment endoscopic ultrasound T and N stage, extent of pathologic sampling (# of sections from tumor bed, complete histologic examination of entire tumor bed), presence and depth of acellular mucin pools in the tumor bed, hyalinization, fibrosis, and acellular mucin pools within lymph nodes, and number of lymph nodes evaluated. Follow up data including overall survival (OS) and time to recurrence (TTR) were obtained by medical chart review, and were analyzed by a Cox Proportional Hazards Model.
Results: Of the 93 patients (mean age 62.7 years, M:F ratio 7.5:1), 57 patients (61%) died and 32 patients (34%) had documented local (n=2) or distant (n=30) recurrence (mean follow up time: 6.3 yrs). Histologic examination of the entire tumor bed was the most significant predictor of favorable outcome in this group (Hazard Ratio [HR]=0.39 for OS and HR=0.40 for TTR; p-values 0.0005 and 0.0115, respectively). The presence of any poorly differentiated component in the pre-treatment biopsy was associated with an unfavorable outcome (HR=1.88 for OS, HR=1.67 for TTR; p-values 0.02 and 0.15, respectively). None of the other clinicopathologic parameters examined were significantly different in those with or without an adverse outcome.
Conclusions: Complete histologic examination of the entire tumor bed and poorly differentiated phenotype in pre-treatment biopsies are important predictors of disease-free survival in patients with EAC and pCR. Our findings also confirm heterogeneous outcomes in pCR patients and suggest that histologic examination of the entire tumor bed is necessary to identify the subset of complete responders with a uniformly favorable outcome within this group.
Monday, March 4, 2013 1:00 PM
Poster Session II # 116, Monday Afternoon