P-157 Assessing fitness for systemic therapy in hepatocellular carcinoma: Evaluating time on treatment according to Child-Pugh classification, model for end-stage liver disease score, and AFP level

Abstract

Clinical trials of systemic therapies for hepatocellular carcinoma (HCC) have classically used Child-Pugh classification (CPC) as enrollment criteria to assess severity of hepatic dysfunction related to cirrhosis, and patients outside of CPC A are generally excluded. Information to guide the care of patients with HCC and CPC B and C who might otherwise be candidates for treatment is thus minimal. This is of increased importance as the number of treatment options and effectiveness of these therapies has increased. We sought to further evaluate the impact of pre-treatment characteristics including CPC on the duration of time on that treatment, to take into account both treatment efficacy and tolerability in the setting of underlying cirrhosis. We conducted a retrospective review of medical records from patients with HCC who received systemic therapy at the University of Colorado Cancer Center. Specific therapy and duration of treatment were collected, as well as baseline CPC (A, B, or C), model for end-stage liver disease (MELD) score (<10 / 10-19 / 20-29 and <16/>16), and AFP (<400 / >400) at time of treatment initiation. The mean, median, standard deviation and interquartile range of the duration of treatment were reported for each of these factors, and non-parametric Kruskal-Wallis and Wilcox tests were conducted to compare the duration of treatments according to each. Subgroup analysis was performed by treatment type, defined as immunotherapy and non-immunotherapy, using the same method. The significance level was set to 0.05. Charts from 101 patients with HCC were reviewed. Of these, 90 received systemic therapy, and 86 patients had data sufficient to determine duration of treatment. A total of 5 patients received more than 1 line of systemic therapy, for 91 unique treatment events assessed. Of these, 62% occurred in the first-line setting, 31% in second-line, and 6% in third-line. Treatments were classified as immunotherapy in 55% and non-immunotherapy in 45%. The duration of treatment was significantly different according to the baseline CPC (p = 0.039) by the Kruskal-Wallis test, with CPC C associated with a significantly shorter duration of treatment as compared to CPC A (p = 0.0226). There was not a significant difference in pairwise comparison of CPC A and B. There was also a significant difference in duration of treatment when assessed by MELD score of <16 and >16 (p=0.019), though not according to other MELD cutoffs, nor baseline AFP level or type of treatment. In subset analysis according to treatment type (immunotherapy and non-immunotherapy), there was not a significant difference in duration of treatment according to any of the baseline factors analyzed. In this retrospective assessment, duration of systemic treatment was not significantly different in HCC patients with underlying CPC A and B liver disease, though was shorter in patients with CPC C disease. Baseline MELD score of <16 and >16 was also associated with a longer time on treatment. Further confirmation of this data may help guide treatment consideration for those patients with HCC and liver disease beyond CPC A.