Abstract
638
Objectives The outcome of advanced gastric cancer remains poor, and early detection of recurrence is important. We hypothesized that heterogeneity of gastric cancer biology influences the accuracy of FDG PET/CT for recurrence detection. We thus investigated the performance of FDG PET/CT for detecting gastric cancer recurrence in patients categorized according to FDG avidity of the primary tumor.
Methods Subjects were 372 patients (mean age 57.3 ± 11.5 y, male 254) with advanced gastric cancer who underwent curative surgery. All subjects had FDG PET/CT for initial staging and for recurrence surveillance after surgery. Primary tumors were classified as FDG-avid if they displayed focal uptake with SUVmax ≥ 4, or were otherwise classified as non-FDG-avid. Follow up FDG PET/CT were evaluated for recurrence. The presence of recurrence was determined by medical records with > 11 mo of follow-up.
Results Of the 372 subjects, 240 had FDG-avid primary tumors (64.5%; 59.7 y, male 175), whereas 132 had non-FDG-avid primary tumors (35.5%; 52.9 y, male 79). During follow-up, 71 patients (19.3%) were confirmed to have recurrence. Among 62 cases with eligible follow-up PET/CT, 41 (3 at anastomosis site) had FDG-avid primary tumors and 21 (2 at anastomosis site) had non-FDG-avid tumors. For all recurrences, PET/CT sensitivity was 34/41 (82.9%) for the FDG-avid group and 11/21 (52.3%) for the non-FDG-avid group (P = 0.01). For recurrences outside the anastomosis site, PET/CT sensitivity was 41/57 (71.9%) for all subjects, 32/38 (84.2%) for the FDG-avid group, and 9/19 (47.4%) for the non-FDG-avid group (P = 0.004). PET/CT specificity for recurrence was 499/515 (96.9%), 301/311 (96.8%), and 198/204 (97.1%) for respective groups (P = n.s.).
Conclusions Primary tumor FDG avidity substantially influences the diagnostic performance of PET/CT for recurrent gastric cancer. Hence, follow-up FDG PET/CT appears to have greater value for recurrence surveillance in patients with high tumor FDG uptake on initial PET/CT.