Abstract
BACKGROUND: The prognostic impact of preoperative (18)F-FDG PET/CT in advanced gastric cancer (AGC) remains a matter of debate. This study aims to evaluate the prognostic impact of SUV(max) in preoperative (18)F-FDG PET/CT of AGC according to histologic subtype, with a focus on the differences between tubular adenocarcinoma and signet ring cell (SRC) carcinoma. METHODS: As a discovery set, a total of 727 AGC patients from prospective database were analyzed according to histologic subtype with Cox proportional hazard model and p-spline curves. In addition, another 173 patients from an independent institution was assessed as an external validation set. RESULTS: In multivariate analysis, high SUV(max) in preoperative (18)F-FDG PET/CT of AGC was negatively correlated with disease-free survival (DFS) and overall survival (OS) in patients with diffuse type (DFS: HR 2.17, P < 0.001; OS: HR 2.47, P < 0.001) or SRC histology (DFS: HR 2.26, P = 0.005; OS: HR 2.61, P = 0.003). This negative prognostic impact was not observed in patients with intestinal type or well or moderately differentiated histology. These findings have been consistently confirmed in a validation set. The p-spline curves also showed a gradual increase in log HR as SUV(max) rises only for SRC histology and for diffuse-type AGC. Finally, a novel predictive model for recurrence of AGC with diffuse type or SRC histology was generated and validated based on the preoperative SUV(max). CONCLUSIONS: Preoperative high SUV(max) of AGC is a poor prognostic factor in those with diffuse type or SRC histology. This study is the first to demonstrate the differential prognostic impact of preoperative PET/CT SUV(max) in AGC according to histologic subtype and provide a clue to explain previous discrepancies in the prognostic impact of preoperative PET/CT in AGC. Prospective studies are required to validate the role of preoperative SUV(max) in AGC.