Outcomes following sequential trials of stereotactic body radiotherapy (SBRT) for hepatocellular carcinoma (HCC).

Abstract

4111 Background: SBRT can precisely deliver ablative doses of radiation while sparing normal tissues. Here are our institution’s outcomes following HCC SBRT. Methods: Two trials of SBRT for patients with active HCC unsuitable for standard local-regional therapies (resection, RFA or TACE) were conducted from 2004-07 (TRIAL 1, phase I/II) and 2007-10 (TRIAL 2, phase II). All had class Child-Pugh A, liver enzymes < 6xULN, platelets ≥ 50x10e9/L, and ≥ 700 ml of uninvolved liver. SBRT dose was 24-54 Gy in 6 fractions. Dose was dependent on a radiation-induced liver toxicity model and proximity of HCC to luminal GI tissues. Primary endpoint was local control at 1 year (LC1y), defined as no progressive disease (PD) of irradiated HCC by RECIST. Fisher’s exact test, Kaplan-Meier, log-rank and Cox model were used for analyses. Results: Total accrual was 102 patients (TRIAL 1: n=51, TRIAL 2: n=51). Underlying liver disease was: hepatitis B, 39%; hepatitis C, 40%; alcohol, 25%; other, 14%; none, 7%. Prior therapies were delivered in 50% of patients (9% surgery, 34% RFA, 22% TACE, 16% other, no sorafenib). Baseline CLIP score was ≥2 in 52% of patients. Median number of lesions was 2 (1 to >10). Median sum of treated liver lesions (STLL) was 10 cm (1.8-43.3 cm). Tumor thrombus (TT) was present in 55%, and extra-hepatic disease in 12%. Median follow-up was 15 months (m). Best RECIST response was: CR11%, PR43%, SD44%. LC1y was 79% (CI 95% 66-87%). STLL was significantly associated with LC (HR=1.09, p=0.02). TRIAL 2 patients had a trend to improved LC (HR=0.45, p=0.09). Toxicity ≥grade 3 was seen in 33%. In 7 patients, death was likely related to treatment (1.1 – 7.8 m post SBRT): 5 liver failures (1 had PD), 1 cholangitis (HCC in bile duct), 1 GI bleed. Median overall survival was 16.8 m (CI 95% 9.6-20.4 m). On multivariate analysis of survival, only TT (21.5 m without TT vs. 10.6 m with TT; HR=2.21, p=0.02) and trial were significant. Median survival of patients treated on TRIAL 2 was 25.1 m vs. 10.6 m in TRIAL 1, HR=0.45, p=0.006. There was a trend for smaller STLL (p=0.09) and use of sorafenib at time of progression (p=0.07) in TRIAL 2 vs. 1. Conclusions: There is strong motivation for studying SBRT for HCC in a randomized trial.