Phase I Study of Stereotactic Body Radiotherapy followed by Ipilimumab with Nivolumab vs. Nivolumab alone in Unresectable Hepatocellular Carcinoma

SBRT is an established treatment option for patients with unresectable hepatocellular carcinoma (HCC) with high rates of local control. Stimulating immune tumor surveillance with SBRT followed by checkpoint blockade may lead to increased treatment response rates or duration. We conducted a phase 1 randomized study to examine the safety and tolerability of SBRT followed by Ipilimumab with Nivolumab (Ipi/Nivo) or Nivolumab (Nivo) alone. Child-Pugh (CP) Class A patients with advanced HCC not eligible for resection or transplant were randomized to receive 40 Gy in 5 fractions (minimum 48-hour interfraction interval) SBRT to liver tumor followed by either Nivo or Ipi/Nivo. Radiation was prescribed to up to 100 cc of tumor, or the entire tumor if able to meet normal tissue liver constraints. Nivo (240 mg) was given every 2 weeks while Ipi (1 mg/kg) was given every 6 weeks. Immunotherapy began within 14 days after SBRT and continued until progression, dose-limiting toxicity (DLT), or for up to 2 years. The primary endpoint was 6-month DLT as pre-defined by protocol. Stopping guidelines were established for number of allowable DLTs in each treatment arm as enrollment increased. Secondary endpoints included response as assessed using iRECIST criteria. Pre-SBRT and post-SBRT biopsies of the radiated lesion were obtained for exploratory analyses. 14 patients received SBRT plus at least one cycle of Ipi/Nivo (n = 8) or Nivo (n = 6); the study was terminated early for slow accrual. 13 patients had CP Score A5 and one had A6. 6 patients had extra-hepatic disease at enrollment. The median mean liver dose was 11.21 Gy and the median V30 Liver was 13%. Only 2 patients had dose-limiting toxicity in first 6 months of treatment (grade 3 hepatotoxicity). 9 DLTs would have been deemed unsafe for number enrolled. Overall response was 36% PR, 36% SD, and 28% PD. Rates of PR/SD/PD were 50%/37.5%/12.5% for Ipi/Nivo and were 12.5%/37.5%/50% for Nivo alone. All 5 patients with response are alive at last follow-up; 4 of the responders (all on Ipi/Nivo arm) have durable response of least 18 months (18.2 mo. – 27.2 mo.) after SBRT. With a median follow-up of 8.3 months, median OS in the nivo arm was 6.7 months whereas the median OS in ipi/nivo arm was not reached. 2 of 6 in Nivo arm and 1 of 8 in Ipi/Nivo arm died of disease progression in the first six months after SBRT. Overall, only one patient progressed in the radiated lesion, which was noted at first post-SBRT imaging, and the patient died of progressive disease. The addition of SBRT to immunotherapy was well-tolerated. SBRT followed by combined checkpoint blockade with Ipi/Nivo was associated with a high rate of durable response in a population with advanced disease. Local control in radiated field was excellent. Pre/post-SBRT tissue analysis may help provide biomarkers for response. SBRT with combined Ipi/Nivo warrants further prospective study in HCC.