Prognostic Value of Preoperative Positron Emission Tomography-Computed Tomography in Surgically Resected Gastric Cancer

Abstract

The diagnostic value of PET-CT, in gastric cancer is well known, but the prognostic value of pretreatment PET-CT has not been adequately evaluated. This study aimed to investigate the preoperative prognostic value of PET-CT in gastric cancer patients. A total of 107 patients underwent surgical treatment for gastric cancer from April 2007 to December 2010 at Dong-A University Medical Center after confirming the presence of F-18 fluorodeoxyglucose (FDG) uptake on preoperative PET-CT. Among these patients, the following subjects were excluded: follow-up loss (13), palliative resection (5), neoadjuvant chemotherapy (1), and unrelated death (1). The remaining 87 patients were included in this study and data were collected by retrospectively reviewing the medical records. The median follow-up duration, defined as the period from operation to last imaging study date, was 34.2±14.8 months. FDG uptake values were represented by maximal standardized uptake value (SUVmax). In order to assess the correlation between SUVmax and recurrence, Kaplan-Meier's survival analysis with log-rank test and cox proportional hazard model were performed. Receiver operating characteristic (ROC) curve was employed to determine the optimal cutoff value of SUVmax. The result of Kaplan-Meier's survival analysis with log-rank test were significantly different between high SUVmax group and low SUVmax group (p=0.035), the cutoff value of which was 5.6. However, in multivariate analysis with cox proportional hazard model, T-staging, N-staging and SUVmax did not show statistical significance (p=0.190, p=0.307, and p=0.436, respectively). High SUVmax on PET-CT in gastric cancer can be a useful prognostic factor.