Incidence and risk factors for sustained hepatic function toxicity 6 months after radioembolization: Analysis of the radiation-emitting sir-spheres in non-resectable liver tumor (RESIN) registry

Background: To quantify rates and risk factors for toxicity after hepatic radioembolization using resin yttrium-90 microspheres. Methods: Radiation-Emitting SIR-Spheres in Non-resectable liver tumor (RESIN) registry enrollees were reviewed with 614 patients included. Mean patient age was 63.1±12.5 years. The majority of patients were male (n=375, 61%) and white (n=490, 80%). Common tumor types were hepatocellular (n=197, 32%), colorectal (n=187, 30%) and neuroendocrine (n=56, 9%). Hepatotoxicity was measured using the Common Terminology Criteria for Adverse Events (CTCAE v 5). Potential risk factors for hepatotoxicity were tested using the Kruskal-Wallis or Pearson Chi-squared tests, and multivariate linear regressions. Results: At 6 months, 115 patients (18.7%) died (n=91, 14.8%), entered hospice (n=20, 3.3%) or sought treatment elsewhere (n=4, 4%). Seven (1.1%) deaths were from liver decompensation. Grade 3 toxicity rates were: bilirubin (n=85, 13.8%), albumin (n=28, 4.6%), ALT (n=26, 4.2%) and AST (n=37, 6.0%). For each of these liver function test components, baseline abnormal labs predicted Grade 3 toxicity at followup by Kruskal-Wallis test (P<0.001) and linear regression (all P<0.03). Other significant factors predicting toxicity at regression included elevated Body-Mass Index (albumin P=0.0056), whole liver treatment (bilirubin P=0.046), and lower tumor volume (ALT and INR, P<0.035 for both). Conclusions: Baseline liver function abnormalities prior to radioembolization is the strongest predictor of post-treatment Grade 3 toxicity with rates as high as 13.8%. Toxicity rates for specific lab values are affected by large volume treatments especially with low tumor volumes.