The aim of this study was to predict the survival of patients with hepatocellular carcinoma (HCC) by examining metabolic PET parameters, apparent diffusion coefficients (ADCs), and clinical parameters.Methods
We retrospectively reviewed 52 patients with pathologically confirmed HCC (age, 57.9 ± 10.7 years; 43 men) who underwent MRI and 18F-FDG PET/CT. The tumor-to-normal liver SUV ratio (TLR), the mean ADC of each tumor, and other clinical data were obtained. Survival analysis was performed.Results
Thirty-two patients died during the follow-up period. There was an inverse correlation between the mean SUV and the mean ADC of a tumor (r = −0.402, P = 0.020). Among HCC patients, disease-specific survival was significantly associated with each of the following factors: high TLR (TLR ≥ 2; hazard ratio [HR], 3.78; P = 0.001), high mean ADC (mean ADC ≥ 1.250 × 10−3 mm2/s; HR, 0.45; P = 0.028), AFP ≥ 400 ng/mL (HR, 3.48; P = 0.001), PIVKA-II ≥ 100mAU/mL (HR, 6.39; P = 0.011), tumor size (HR, 1.13; P < 0.001), number of tumors (HR, 2.16; P = 0.031), tumor stage (HR, 3.08; P < 0.001), and surgery for initial treatment (HR, 0.06; P < 0.001). The results of multivariate analysis show that DSS was significantly associated with each of the following factors: TLR ≥ 2 (HR, 2.46; P = 0.044), PIVKA-II ≥ 100mAU/mL (HR, 5.11; P = 0.037), tumor stage (HR, 3.01; P < 0.001), and surgery for initial treatment (HR, 0.04; P < 0.001).Conclusions
High TLRs and low mean ADCs were associated with poor outcomes. The TLR was an independent prognostic factor in patients with HCC, but the mean ADC was not. A negative correlation was found between the mean ADC and the mean SUV of a tumor.