Purpose: The diagnostic accuracy of [(18)F]fluorodeoxyglucose (FDG) PET is insufficient to characterise hepatocellular carcinoma (HCC) in liver masses and to diagnose all cases of recurrent HCC. HCC has been reported to take up [(11)C]acetate, but routine use of this tracer is difficult. Choline is another tracer of lipid metabolism, present in large amounts in HCC. In a proof-of-concept study, we evaluated [(18)F]fluorocholine (FCH) uptake by HCC and compared FCH PET/CT with FDG PET/CT.
Methods: Twelve patients with newly diagnosed (n=8) or recurrent HCC (n=4) were prospectively enrolled. HCC was assessed by histology in eight cases and by American Association for the Study of Liver Diseases (AASLD) criteria in four cases. All patients underwent whole-body PET/CT 10 min after injection of 4 MBq/kg FCH. Within 1 week, 9 of the 12 patients also underwent whole-body FDG PET/CT 1 h after injection of 5 MBq/kg FDG.
Results: The per-patient analysis showed a detection rate of 12/12 using FCH PET/CT for both newly diagnosed and recurrent HCC. The median signal to noise ratio was 1.5+/-0.38. There was a trend towards a higher FCH SUV(max) in well-differentiated HCC (15.6+/-7.9 vs 11.9+/-0.9, NS). Of the nine patients who underwent FCH and FDG PET/CT, all nine were positive with FCH whereas only five were positive with FDG.
Conclusion: FCH provides a high detection rate for HCC, making it potentially useful in the initial evaluation of HCC or in the detection of recurrent disease. The favourable result of this proof-of-concept study opens the way to a phase III prospective study.