Pairwise comparison of 18F-FDG and 18F-FCH PET/CT in prostate cancer patients with rising PSA and known or suspected second malignancy

OBJECTIVE: This study aimed to evaluate the usefulness of combining fluorine-18 choline (F-FCH) and fluorine-18 fluorodeoxyglucose (F-FDG) PET/computed tomography (CT) in patients with rising prostate-specific antigen and known or suspected second malignancy. MATERIALS AND METHODS: F-FCH and F-FDG PET/CT were performed 15±9 days apart on the same PET/CT system and acquisition and reconstruction parameters. A mean standardized uptake value (SUVmean) was computed for every lesion that could be discriminated with both tracers. PET results were confirmed by histology (eight patients) and clinical and imaging follow-up (mean±SD: 15±9 months). RESULTS: Of 77 consecutive patients who underwent F-FCH PET/CT scans for suspected prostate cancer recurrence, 10 (13%) were suspected to have a second malignancy because of F-FCH PET pattern inconsistency with that of prostate cancer (n=6), because of a history of a second malignancy with similar metastatic patterns (n=2) or inconsistency between disease burden and prostate-specific antigen value (n=2). Seventy lesions were studied, with a final diagnosis of prostate cancer, other cancers and benign disease in 55, nine and six lesions, respectively. F-FCH SUVmean and F-FCH/F-FDG SUVmean ratios were significantly different between prostate cancer, nonprostate cancer and benign disease (P<0.0001 and P=0.04, respectively). Receiving operating characteristic analysis showed that the F-FCH/F-FDG ratios were not better than F-FCH SUVmean in discriminating prostate cancer from nonprostate cancer and benign diseases (sensitivity, specificity and area under the curve were 69%, 80%, 0.71 and 84%, 80% and 0.89, respectively). CONCLUSION: We found that F-FCH/F-FDG SUVmean ratios cannot differentiate prostate cancer recurrences from other cancer types when both diagnoses are suspected. Doubtful lesions should be biopsied.