Clinical Outcomes of Patients With Hepatocellular Carcinoma Treated With MR-guided Stereotactic Body Radiation Therapy With Online Adaptive Planning

Abstract

<h3>Purpose/Objective(s)</h3> Stereotactic body radiation therapy (SBRT) for primary liver tumors may be limited by proximity to organs-at-risk (OAR). MR-guided SBRT (MRgSBRT) with online adaptive planning has shown promise in delivering ablative doses to the liver safely. We reported the clinical outcomes of patients with unresectable hepatocellular carcinoma (HCC) treated with MRgSBRT with online adaptive planning. <h3>Materials/Methods</h3> Patients with unresectable HCC treated with MRgSBRT between 2015 and 2019 with minimum 3 months of follow-up were retrospectively reviewed at a single institution. Local progression was defined by mRECIST. Survival analysis was performed with the Kaplan-Meier method. Toxicity was assessed using Common Terminology Criteria for Adverse Events Version 5. <h3>Results</h3> Thirty-five patients with a total of 41 HCC lesions, all treated to 50 Gy in 5 fractions, were analyzed. The median tumor size was 2.9 cm (IQR: 2.0-4.5). Barcelona-Clinic Liver Cancer (BCLC) Staging was 0, A, B, C, and D in 2 (6%), 8 (23%), 18 (51%), 6 (17%), and 1 (3%), respectively. Twenty-seven (77%) patients had previous liver-directed therapy (chemoembolization, radioembolization, cryoablation) to the target lesion with evidence of residual disease or progression on follow-up imaging. The median follow-up from completion of MRgSBRT was 21 months (IQR, 10-34). Online adaptive planning was performed for 19% of all fractions. It was done due to exceeding dose constraints to OARs in 26/33 (79%) fractions, and to improve tumor coverage in 7/33 (21%) fractions. Clinical response assessed at a median of 2.4 months (IQR: 2.0-3.0) after MRgSBRT revealed a complete response in 26% (9/35), partial response in 23% (8/35), stable disease in 40% (14/35), and progressive disease in 9% (3/35). One (3%) patient did not have repeat imaging at the time of initial clinical follow-up. The 1- and 2-year freedom from local progression (FFLP) rates were both 91% (95% CI: 83-100%). The 1- and 2-year freedom from progression elsewhere in the liver (FFELP) rates were 58% (95% CI: 44-78%) and 50% (95% CI: 35-72%), respectively. The 1- and 2-year overall survival were 83% (95% CI: 71-96%) and 58% (95% CI: 43-77%), respectively. The most commonly noted acute grade ≤2 clinical toxicities were fatigue, abdominal pain, and nausea. Four patients had grade 3 toxicity: 1 (3%) ascites, 1 (3%) portal hypertension, 2 (6%) hepatic failure. One patient (3%) had grade 4 gangrenous cholecystitis. Eleven (31%) patients later underwent an orthotopic liver transplant. <h3>Conclusion</h3> MRgSBRT with online adaptive planning allows delivery of ablative doses to HCC with excellent local control and acceptable toxicity. Additional studies evaluating the role of MRgSBRT in the treatment of liver tumors are warranted.