Clinical utility of 2-[(18)F] fluoro-2-deoxy-D-glucose positron emission tomography in predicting World Health Organization grade in pancreatic neuroendocrine tumors.

Abstract

Pancreatic neuroendocrine tumors (PNETs) present various histologic stages, and their clinical behavior ranges from benign to highly aggressive. World Health Organization (WHO) grading categorizes PNETs into 3 groups (G1, G2, and G3) based on proliferative activity. The aim of this study was to assess the clinical utility of 2-[(18)F] fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) in preoperative differential diagnosis for various histologic grades of PNETs and in predicting postoperative prognosis in PNET patients. We investigated prospectively 36 patients who underwent preoperative FDG-PET, received surgery for PNETs, and had resected specimens examined via histology. The maximum standardized uptake value (SUVmax) was determined on FDG-PET, and differentiating power between G1/G2 and G3 PNETs was examined using various SUVmax cutoff levels. We also evaluated the prognostic impact of FDG-PET findings using postoperative survival data. SUVmax significantly correlated with WHO grade (Spearman rank correlation 0.584; P = .0018), and the SUVmax of G3 tumors (5.0 ± 2.5; n = 4) was significantly higher than that of G1/G2 tumors (2.7 ± 1.6; n = 32; P = .0159). Using 2.5 as a cutoff SUVmax, the sensitivity, specificity, and accuracy of differentiating G3 tumors from G1/G2 tumors were 100.0%, 62.5%, and 66.7%, respectively. Furthermore, the SUVmax of FDG-PET (<2.5 vs ≥2.5) was significantly related to postoperative disease-free survival (P = .0463). These results suggest that FDG-PET may be useful for differentiating G3 PNETs from G1/G2 PNETs and for predicting postoperative prognosis in PNET patients. This preliminary finding is expected to be confirmed by prospective validation with more patients.