Tumor metabolism derived from (18)F-FDG PET/CT in predicting the macrotrabecular-massive subtype of hepatocellular carcinoma

BACKGROUND: The recently described pathological subtype of hepatocellular carcinoma (HCC), named macrotrabecular massive (MTM), is associated with an unfavorable prognosis. This study aimed to evaluate the potential for tumor metabolism obtained by β-2-[18F] fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography ((18)F-FDG PET/CT) to be used as a preoperative imaging indicator for predicting MTM-HCCs. METHODS: This study was designed to be cross-sectional. Patients who underwent preoperative (18)F-FDG PET/CT and who had surgically-diagnosed HCC between June 2015 and June 2021 were retrospectively included. Tumor metabolism was determined by the tumor-to-normal liver standardized uptake value ratio (TLR) of the primary tumor as shown on (18)F-FDG PET/CT. Clinical, pathological, and PET/CT characteristics were compared between non-MTM-HCCs and MTM-HCCs. Univariate analyses were used to screen the predictive factors of MTM-HCCs, then multivariate binary logistic regression analyses were performed. A regression-based diagnostic model was then established. Substantial necrosis was assessed to compare the predictive performance between traditional imaging and TLR measured on (18)F-FDG PET/CT. The receiver operating characteristic (ROC) curve analyses and the DeLong test were used to assess the predictive performance. RESULTS: A total of 93 patients (mean age, 52.6±11.3 years; 81 male) with 36 MTM-HCCs were included. Multivariate binary logistic regression analyses identified higher platelet count [PLT; ≥118.5×10(3)/µL; odds ratio (OR), 3.63; 95% confidence interval (CI), 1.13–12.87; P=0.035], higher aspartate transaminase (AST; ≥52 IU/L; OR, 4.15; 95% CI: 1.34–14.33; P=0.017), and larger TLR (≥2.2; OR, 5.55; 95% CI: 1.90–17.56; P=0.002) as independent predictors of MTM-HCCs. A TLR ≥2.2 helped to identify 72.2% of the MTM-HCCs with a specificity of 75.4%. The AUC of the regression-based diagnostic model for predicting MTM-HCCs was 0.835 (95% CI: 0.746–0.923), with a sensitivity of 80.6% and a specificity of 78.9%. Substantial necrosis enabled the identification of MTM-HCCs with 52.8% sensitivity and 87.7% specificity, with an AUC of 0.702 (95% CI: 0.588–0.817). There was no statistical difference between TLR and substantial necrosis in predicting MTM-HCCs using the DeLong test (P>0.05). CONCLUSIONS: Tumor metabolism determined by TLR on (18)F-FDG PET/CT is a valuable imaging indicator for MTM-HCCs. Noninvasive prediction of this subtype can achieve good sensitivity and excellent predictive performance based on the regression model of AST, PLT, and TLR.