OBJECTIVE  To compare and evaluate the diagnostic efficacy of ¹⁸F-FAPI PET/CT and ¹⁸F-FDG PET/CT in hepatic malignancies, analyze the maximum standardized uptake value(SUV_max) for the two tracers, and explore their diagnostic value in hepatic malignancies.

METHODS  Clinical and imaging data of 20 patients(16 males, 4 females, median age 65 years) who underwent ¹⁸F-FAPI-04 PET/CT and ¹⁸F-FDG PET/CT at the Affiliated Hospital of Jiangnan University from February 2023 to July 2023 were retrospectively analyzed. All patients underwent both ¹⁸F-FAPI-04 PET/CT and ¹⁸F-FDG PET/CT with maging interval time≤7 d. Histopathology served as the gold standard for final diagnosis; for cases unsuitable for tissue diagnosis, imaging follow-up was used to determine the presence of hepatic malignancies. The PET/CT imaging findings of the two tracers were compared. The Wilcoxon signed-rank test was applied to compare the differences in SUV_max, metabolic tumor volume(MTV), total lesion glycolysis(TLG) and total lesion FAP uptake(TLF) between ¹⁸F-FAPI PET/CT and ¹⁸F-FDG PET/CT. The Kappa test evaluated the consistency of lesion identification between two observers. The Wilcoxon paired rank sum test for SUV_maxM(P25, 75) was used to assess the image contrast and correlation between 1 h and 2 h imaging.

RESULTS For 18 positive lesions(9 primary, 9 metastatic), the SUV_max of ¹⁸F-FAPI PET/CT and ¹⁸F-FDG PET/CT were 11.74(7.65, 15.51) and 8.05(4.91, 10.33), respectively, with a statistically significant difference(z=−2.461, P=0.014). In 9 primary liver lesions, the SUV_max were 13.06(7.30, 18.22) and 5.94(3.01, 8.82), respectively, with a statistically significant difference(z=−2.429, P=0.015); in 9 metastatic liver lesions, the SUV_max were 10.42(7.70, 14.17) and 10.16(6.53, 13.86), respectively, with no significant difference(z=−0.533, P=0.594). For 18 positive lesions(9 primary, 9metastatic), the MTV_4.0 were 29.29(3.41, 34.72) and 18.02(0.85, 31.05)(z=−2.243, P=0.025), and TLF_4.0 were 208.73(19.49, 262.61) and 111.48(3.73, 213.38)(z=−2.373, P=0.018), showing significant differences. In 9 primary liver lesions, TLF_2.5 were 337.06(78.41, 462.48) and 191.25(10.41, 352.41)(z=−2.073, P=0.048), and TLF_3.0 were 307.95(55.89, 393.50) and 160.46(0.05, 279.24)(z=−2.429, P=0.015), with significant differences. No significant differences were found in MTV and TLF of 9 metastatic liver lesions under different fixed thresholds and percentage thresholds. The visual analysis consistency of liver lesions between two observers was 100%(Kappa=1.000, z=4.88, P<0.001). The consistency in lesion number was 88.89%(Kappa=0.8252, z=5.36, P<0.001), and the FDG imaging analysis consistency was 88.89%(Kappa=0.8269, z=6.24, P<0.001). There was no significant difference in image contrast between 1 h and 2 h imaging(P=0.1289), while the Pearson correlation coefficient between 1 h SUV_max and 2 h SUV_max was 0.7928(P<0.05), indicating a significant correlation.

CONCLUSION ¹⁸F-FAPI shows higher target-to-background contrast in primary liver lesions, which can compensate for the limitations of ¹⁸F-FDG PET/CT in detecting primary liver cancer. Therefore, the highly selective uptake of FAPI in liver tumors may open up new application fields for non-invasive tumor diagnosis, staging, and therapeutic efficacy monitoring.