ADV score plus FDG-PET as preoperative predictors of post-resection prognosis in patients with hepatocellular carcinoma ≥10 cm: a retrospective cohort validation study.

[BACKGROUND] The ADV score, calculated by multiplying α-fetoprotein, des-γ-carboxy prothrombin, and tumor volume (TV), is a surrogate marker for post-resection prognosis in hepatocellular carcinoma (HCC). 2- 18 F-fluoro-2-deoxy- d -glucose (FDG) positron emission tomography (PET) reflects the tumor biology. This study aimed to validate prognostic prediction models (PPMs) that combined the ADV score and FDG-PET in patients undergoing resection for large HCC tumors (≥10 cm).

[METHODS] Of the 7905 patients who underwent resection for HCC between 2009 and 2018, 412 with solitary tumors ≥10 cm were analyzed.

[RESULTS] The median pathological and radiological tumor diameters and TVs were 12.0 and 11.0 cm and 575 and 696 mL, respectively. The median values of pathological, radiological and volumetric TVs converged at 2.8log mL; therefore, radiological TV was selected. At 1, 3, 5, and 10 years, the tumor recurrence (TR) rates were 60.7, 72.2, 76.6, and 85.1%, respectively, and the overall survival (OS) rates were 77.7, 56.5, 48.2, and 35.5%, respectively. Independent risk factors for TR included viral hepatitis, hypermetabolic PET, microvascular invasion (MVI), and satellite nodules. Hypermetabolic PET, MVI, satellite nodules, and cirrhosis were significant predictive factors of OS. The ADV scores, expressed in 2.0log-intervals, showed a clear prognostic separation of TR and OS. Combining the ADV score with FDG-PET activity further improved the predictive accuracy, leading to two PPMs: a two-tiered model (hypermetabolic PET with ADV ≥7.0log vs. others) and a four-tiered model (isometabolic PET and hypermetabolic PET stratified by ADV <7.0log, ≥7.0log and <11.0log, and ≥11.0log).

[CONCLUSIONS] These two PPMs reliably predicted post-resection outcomes in patients with large HCC, making them useful tools for guiding surgical decisions and postoperative surveillance in high-risk patients.