Association of LR treatment response category with outcome of patients with hepatocellular carcinoma on explant pathology

Abstract

Objectives

Liver transplant (LT) is an effective treatment for hepatocellular carcinoma (HCC) in appropriately selected patients. Locoregional therapy (LRT) is often performed to extend a patient’s eligibility for LT. Imaging has a modest sensitivity of approximately 40–77% for detecting pathologically viable HCC in post-LRT patients. The impact on overall survival (OS) and disease-free survival (DFS) is unclear. We hypothesize that Liver Imaging Reporting & Data Systems Treatment Response (LI-RADS TR) category is equivalently correlated with long-term survival and overall disease-free progression when compared to explant pathology findings. We additionally hypothesize that neoadjuvant LRT can improve OS and DFS in LT patients initially within MC.

Methods

Patients found to have HCC on explant between January 2005 and December 2021 were included. A total of 167 patients were divided into treatment (any pre-LT LRT except for Y-90 therapy) and control (no pre-LT LRT) groups. Of the patients who received pre-LT LRT, imaging studies were reviewed by two abdominal radiologists using 2018 LI-RADS criteria. Statistical analysis was performed using Kaplan-Meier survival curves and Cox proportional hazard models to assess OS and DFS.

Results

No statistically significant difference in OS or DFS (p = 0.23 and p = 0.22 respectively) was initially found. Given significant difference in age between the groups (p < 0.0001), Cox proportional hazard models were used to adjust for age with statistical significance reached for better OS and DFS in the treatment group (p = 0.05 and p = 0.05 respectively). Contrary to our hypothesis, there was no difference between treatment response groups regarding overall survival or disease-free survival, presumably because of low number of HCC recurrences in our patient population (4%).

Conclusion

Despite not reaching statistical significance, LI-RADS TR categorization demonstrates a good interreader agreement (Kappa 0.6), helping radiologists feel comfortable that modest sensitivity of the LI-RADS TR treatment response category for detecting pathologically active malignancy does not confer a negative clinical outcome.