Pre-Treatment Trends in Child-Pugh Score as an Indicator of Post-Treatment Survival in Patients Receiving Liver SBRT for HCC

Abstract

We sought to understand the value of Child-Pugh (CP) score trends prior to undergoing stereotactic body radiation therapy (SBRT) for hepatocellular carcinoma in patients with advanced cirrhosis, as an indicator of post-treatment morbidity and mortality. We hypothesize that an uptrend in CP score prior to RT increases risk of post-treatment cirrhosis progression, or death. We retrospectively reviewed all patients who underwent definitive SBRT for HCC between 2014-2022 in our health system. Most patients were referred for treatment in the 2nd or 3rd line setting, allowing us to determine CP score at multiple points before SBRT. Response to treatment was evaluated with RECIST or mRECIST criteria. Acute treatment toxicities were assessed by CTCAE v5. OS was assessed using the Kaplan-Meier method, and differences between groups were evaluated using log-rank test. A total of 61 patients were identified, 26 had an immediate pre-SBRT CP score of A5, 11 with A6, 10 with B7, 4 with B8, 4 with B9, 5 with C10, and 1 with C11. Median prescribed dose was 40Gy in 5fx. Median OS of all patients was 24 months. When stratified by CP category, median OS for CP A was not reached (NR) at time of analysis, CP B had a median OS of 14.8mo, and CP C with 1.9mo (p < 0.01). Local control rate at 6mo was 87.5%. 5 patients (8.3%) had CTCAE grade 2 acute toxicity. 10 patients had CP score progression of 2pts or more within 6mo after treatment, and 7 patients had a cause of death attributed to end stage liver disease. Median time interval between HCC diagnosis and start of RT was 338 days. In this time interval, patients with a CP score increase of 2 pts or more prior to starting RT had a median OS of 362 days, compared to NR for patients without increase (p = 0.02). When excluding the CP A group, patients with 2 pt increase had median OS of 362 days vs 445 days (p = 0.34). Again excluding the CP A group, patients with a 1 pt increase had median OS of 362 days vs NR (p = 0.23). Of 9 patients who had a pre-RT CP score increase of 2 pts or more, 7 experienced continued CP progression after treatment. In this multi-institutional retrospective analysis, we found 24 patients with advanced CP B/C cirrhosis who underwent SBRT, none of whom received orthotopic liver transplant. While CP C patients did relatively poorly, we find that some patients with CP B cirrhosis may tolerate SBRT well with good oncologic effect. When taking pre-treatment CP score increase into consideration, we saw a trend but no statistical significance indicating that a pre-SBRT increase in CP score may be associated with worse OS after treatment among CP B/C patients. We conclude that there may be a subset of patients with advanced cirrhosis who, if well selected, are appropriate candidates for SBRT. We suggest a novel use of the CP score and pre-RT trends as an additional clinical tool to aid in decision making when selecting patients. We also suggest that patients with marked up-trending scores pre-treatment have a high likelihood of continued cirrhosis progression after treatment.