Abstract No. 260 Does Y90 Dosing Affect Adverse Events in Y90 Segmentectomy?

Y90 radioembolization is widely used for hepatocellular carcinoma (HCC), but safe and tolerable dose varies among patients. This study investigates whether higher doses in segmental Y90 embolization lead to increased adverse events. A retrospective study evaluated 40 Y90 segmental radioembolizations from 05/19-04/22 at a single institution with institutional review board approval. All patients were selected for Y90 via liver tumor board. Average age was 71 (±10.6), 60% were male. SPECT imaging was utilized to generate voxel-based dosimetry analysis. Adverse events were defined with common terminology criteria for adverse events (CTCAE)v5. The cohort had an average ALBI score of -2.59 (±0.46), median MELD of 9.0 (8-11), and was composed of Child-Pugh A 85% (34) and Child-Pugh B 15% (6) patients. The delivery parameters analyzed were median Rx dose 220 Gy (180-450), median gross tumor volume (GTV) mean dose Gy 282.5 (167.3-465.5), median mean dose to healthy liver (DHL) 39.9 Gy (26.7-64.3) median GTV 55.0cc (21.9-92.2), GTV/Liver volume (LV) 0.031 (0.016-0.094), median conformality index (CI) 1.31 (0.51-2.89), median healthy tissue overtreatment factor (HOF) 0.858 (0.099-2.518), median activity density (AD) 28.86 (20.05-42.47), median homogeneity index (HI) 2.45 (1.52-3.92). Seven patients (17.9%) experienced G2 or higher adverse events within one month of radioembolization. Rx Dose, Mean GTV dose, Mean DHL, GTV, GTV/Liver volume, CI, HOF, AD, and HI did not show significant predictive value for ≥ G2 adverse events p=0.73, p=0.90, p=0.99, p=0.14, p=0.58, p=0.63, p=0.46, p=0.46, p=1.00 respectively. In a binary multivariable logistical regression analysis of Rx dose, HOF, GTV/LV, and Child-Pugh score, only Child-Pugh was a predictive factor for ≥G2 adverse events, with an Odds ratio of 142.8 p=0.018 (Table 260.1). In our study, higher Rx dose, mean GTV dose, DHL, GTV, GTV/Liver, CI, HOF, AI, and HI failed to reach statistical significance in predicting ≥G2 adverse events, whereas Child-Pugh was predictive. Limitations include retrospective design and small sample size.