Evaluation of PET response to neoadjuvant therapy for patients with gastric and GEJ adenocarcinoma.

Abstract

114 Background: The implications of response to neoadjuvant treatment as measured by PET remain poorly defined for patients with gastric and gastro-esophageal junction (GEJ) adenocarcinoma. Our aims are to determine if changes in PET avidity correlate with histologic response, and to determine the best predictor(s) of disease-free survival (DFS) and overall survival (OS). Methods: We reviewed a prospective database to identify patients with gastric and GEJ adenocarcinoma who were evaluated with PET imaging prior to andfollowing neoadjuvant treatment. Spearman correlation and Cox proportional hazards were utilized. Results: Since 2002, 216 patients of median age 63 years met our criteria. At last follow-up (median 22 months, range: 0 - 119), 118 patients recurred or died. The median DFS and OS for expired patients were 7.5 months (range: 0-62) and 14 months (range: 0 - 69), respectively. Between baseline and follow-up PET imaging (median 63 days, range: 15 - 454), 170 patients were treated with chemotherapy and 46 patients with chemoradiotherapy. The median change in SUV was 43% (range: -300 - 100.0%) and the median histologic tumor response was 50% (range: 0 - 100%). No association was identified with the use of chemoradiation (as compared to chemotherapy alone) and change in SUV (p=0.8). We identified a significant relationship between change in SUV and histologic response (r=0.32, p<0.01). Furthermore, the change in SUV was related to both DFS and OS on univariate analysis, as was tumor response and pathologic stage (Table). On multivariate analysis only pathologic stage, and specifically the presence of lymph node metastases, was related to DFS (p<0.01) or OS (p<0.01). Conclusions: Following neoadjuvant therapy for gastric and GEJ adenocarcinoma, PET response is prognostic, although pathologic nodal status is the best predictor of outcome. [Table: see text]