Ali Afrasiabi, Bahareh Gholami, Andrew M Moon, Alex Villalobos, Johannes L Du Pisanie, Bryan Harris, Hyeon Yu, David Mauro, Paolo Varela, Michael Mohnasky, Sandra Gad, Nima Kokabi, FRCPC
In this retrospective cohort, hepatic decompensation following segmental Y-90 TARE appeared more strongly associated with impaired baseline hepatic reserve than with the post-treatment dosimetry parameters evaluated. Baseline hypoalbuminemia is a strong independent predictor of early hepatic decompensation and should be carefully considered during patient selection for radioembolization.
BACKGROUND: Transarterial radioembolization (TARE) with yttrium-90 (Y-90) microspheres is an established treatment for unresectable hepatocellular carcinoma (HCC). Although segmental TARE offers a favorable safety profile, hepatic decompensation remains a clinically important complication. This study aimed to identify clinical, laboratory, and post-treatment dosimetric predictors of hepatic decompensation following segmental Y-90 TARE.
METHODS: In this retrospective cohort study, 102 patients with HCC who underwent segmental Y-90 TARE between 2015 and 2025 were analyzed. Baseline demographic, clinical, laboratory, and imaging data were collected. Hepatic decompensation was defined as new-onset ascites, new or worsening hepatic encephalopathy, or total bilirubin > 3× the upper limit of normal within 2-4 months after TARE, in the absence of radiographic tumor progression. Post-treatment dosimetry was performed using SPECT/CT images. Univariate and multivariable logistic regression models were used to identify predictors of hepatic decompensation.
RESULTS: Hepatic decompensation occurred in 14 patients (13.7%). On univariate analysis, decompensation was associated with baseline hypoalbuminemia (< 3.5 g/dL), higher MELD and ALBI scores, Child-Pugh class B/C, INR ≥ 1.2, lower white blood cell count, and higher alkaline phosphatase levels. Because target doses and normal-parenchyma tolerances differ between microsphere types, post-treatment dosimetry and volumetry were analyzed separately by platform. In the glass microsphere cohort, patients who developed decompensation received a lower mean absorbed dose to the treated segment (236 Gy vs. 411 Gy, p = 0.03), whereas no significant dosimetric or volumetric differences were observed in the resin microsphere cohort. In multivariable analysis, baseline hypoalbuminemia (< 3.5 g/dL) was the only independent predictor of hepatic decompensation (OR = 7.98, 95% CI 1.16-80.39, p = 0.04).
CONCLUSION: In this retrospective cohort, hepatic decompensation following segmental Y-90 TARE appeared more strongly associated with impaired baseline hepatic reserve than with the post-treatment dosimetry parameters evaluated. Baseline hypoalbuminemia is a strong independent predictor of early hepatic decompensation and should be carefully considered during patient selection for radioembolization.