Imaging Features of Hepatocellular Carcinoma: Association with Transarterial Chemoembolization Outcomes

Hepatocellular carcinoma (HCC) accounts for up to 90% of all primary liver cancer worldwide, is one of the five most frequently diagnosed cancers in the world, and is the second leading cause of cancer death worldwide. Diagnosing HCC involves use of the Liver Reporting & Data System (LI-RADS), which allows Radiologists to calculate a score corresponding to a relative chance of HCC for any given hepatic observation based on features observed on cross- sectional imaging. Curative therapies for HCC are surgical resection, liver transplantation, and thermal ablation, however for those HCCs not amenable to these therapies, transarterial chemoembolization (TACE) has developed as a standard of care treatment. Transarterial chemoembolization takes advantage of the unique dual blood supply of the liver, which allows hepatic arterial embolization to selectively injure the target tumors, while the portal vein supplies the unaffected surrounding parenchyma. However, while TACE is widely used and has demonstrated improved 2-year survival for unresectable HCC patients, the overall survival benefits are modest. For this reason, the need to accurately predict post procedure outcomes and stratify patient risk becomes more essential. To date, there has been little investigation into the individual imaging features of the LI-RADS score and how each of these imaging features correlates with clinical outcomes after TACE. In this study, we tested the hypothesis that certain HCC features seen on cross-sectional imaging will be predictive of post-TACE outcomes. This study retrospectively analyzed 91 patients with confirmed HCC who underwent conventional TACE from 2015-2020. Computed tomography (CT) and magnetic resonance (MR) imaging was performed before and after TACE and overall post-TACE time-to-progression (TTP) and transplant free survival (TFS) were calculated with respect to individual factors of the LI-RADS score (non-rim arterial phase hyperenhancement, APHE); portal venous. phase or later phase washout, observation size, an increase of 10 mm or more in diameter within 1 year and enhancing capsule) as well as other imaging and clinical features. In accordance with the study hypothesis, there were several imaging features which were predictive of prolonged post-TACE TTP and TFS upon univariate analysis. These features were absence of tumor in vein (TIV), non-infiltrative growth pattern, tumor size less than 3 cm, presence of non-rim APHE, and left lobe tumor. Furthermore, several clinical and laboratory features were predictive of prolonged TTP and TFS. These included Child-Pugh (CP) score A, post-TACE alpha-fetoprotein (AFP) less than 400 ng/mL, and absence of hepatic encephalopathy (HE). With regards to number of TACE procedures and its relationship to TTP, the median TTP was 649 days and 310 days for patients who received one or multiple TACE procedures respectively (p = 0.0419). Upon multivariate analysis, multiple TACE procedures were associated with decreased TTP (HR = 1.007, p = 0.0094). With regards to CP class and its relationship to TTP, the median TTP was 352 days and 251 days for CP class A and B respectively (p = 0.0257). Upon multivariate analysis, it was found that CP class B was associated with decreased TTP (HR = 1.9, p = 0.0244) as was as CP class C (HR = 1.93, p = 0.4407), though the latter was not statistically significant. It was also found that CP class was associated with decreased TFS, with CP class B having a median survival time of 1145 days (p = 0.0005), and upon multivariate analysis CP class B increased ones risk of death (HR = 11.283, p = 0.0022) compared to CP class A. This study investigated the role that individual imaging features have on post TACE survival outcomes. The results of the study demonstrate that there are tumor imaging features that exist which can portend survival outcomes in patients receiving TACE. A more robust study in a larger cohort should be conducted to better elucidate these relationships.