Outcomes of MR-guided Stereotactic Body Radiotherapy (SBRT) or yttrium-90 Transarterial Radioembolization for Hepatocellular Carcinoma Treated at an Urban Liver Transplant Center

Background: There are overlapping indications for both stereotactic body radiotherapy (SBRT) and yttrium-90 (Y90) trans-arterial radioembolization as locoregional treatments for hepatocellular cancer, though most centers preferentially use one modality over the other. MR-guided radiation allows both effective on-table localization and integrated motion management as compared with many traditional linear accelerators, allowing SBRT to be done more easily. Y90 radioembolization has been a well-established modality to deliver highly conformal dose due to the localization of the microspheres to the vascular supply of a tumor. We looked at patient characteristics and treatment outcomes for patients receiving MR-guided SBRT or Y90 at an urban transplant center. Objectives: To compare patient characteristics and treatment outcomes of MR-guided SBRT with Y90 transarterial radioembolization in a liver transplant center. Methods: This retrospective single-institution study analyzed patients with HCC treated with SBRT or Y90 from August 2017 to September 2020. To select a patient population eligible for either treatment modality, any Y90 procedures for lesions \u3e 10 cm or for treatment volumes \u3e 1000 cc were omitted from the cohort. A total of 239 patients were included in the analysis, receiving a total of 98 courses of SBRT and 187 courses of Y90 treatment. Local control (LC), freedom from liver progression (FFLP), and overall survival (OS) rates were measured from treatment completion date to death date or last follow-up. All outcomes were censored at time of loss to follow-up; LC and FFLP were censored at time of liver transplant if applicable. Cox regression models were used for survival, with significant factors on the univariate analysis further analyzed with a multivariate model. Results: Median time to follow-up was 11 months (0-44 mo). The mean size of lesions treated with SBRT were smaller than those treated with Y90 (2.7 cm vs 4.3 cm, P\u3c0.01). The groups of patients differed in liver disease characteristics, with SBRT patients having fewer Child-Pugh A disease (62% vs 80%, P\u3c0.01), more having received locoregional treatments to the liver in the past (81% v 35%, P\u3c0.01), and more disease in previously treated liver (57% vs 25%, P\u3c0.01). Dose of radiation for SBRT was 45-50 Gy administered in 5 fractions; dose of Y90 radiation to tumor was prescribed to a median of 235.2 Gy (range 55.8-512.3 Gy). There was a higher rate of one year LC in the SBRT cohort (77% vs 57%, P\u3c0.01), while median FFLP (9 mo vs 8 mo, P=NS) and median OS were not significantly different (24 mo vs 21 mo, P=NS). Multivariate analysis revealed size of largest lesion (P\u3c0.01) was correlated with decreased local control; a 1 cm increase in tumor size was associated with a 25% increased risk of local failure. Subsequent transplant (P\u3c0.01) was the remaining significant factor. Treatment modality did not remain an independent predictor of LC. Predictors of OS in multivariate analysis included age (P=0.01), prior liver treatments (HR 2.86, P\u3c0.01), size of largest lesion (P\u3c0.01), Child-Pugh stage (P\u3c0.01), portal vein thrombosis (HR 1.6, P=0.04), and subsequent liver transplant (HR 0.08, P\u3c0.01). Conclusions: These findings support the effectiveness of both MR-guided SBRT and Y90 transarterial radioembolization in locoregional management of HCC at a single institution despite clear differences in the patient cohorts. Though survival outcomes were comparable, local control differences favored the cohort treated by SBRT, in large part due to differences in tumor size. This data supports further investigation in a randomized study between SBRT and Y90