Treatment of hepatocellular carcinoma with tyrosine kinase inhibitors: Real-world outcomes from the West of Scotland.

Abstract

390 Background: Sorafenib, lenvatinib and regorafenib have proven efficacy in patients with HCC in randomised clinical trials. First line sorafenib significantly improves overall survival (OS) and disease control rates vs. placebo (SHARP) whilst lenvatinib (REFLECT) is non-inferior to sorafenib in terms of OS. Second line regorafenib improves OS versus placebo (RESORCE). Real world outcomes are often inferior to randomised controlled trial data. Consequently we investigated the outcomes of real world patients with HCC treated with multi-kinase inhibitors in our regional tertiary-referral centre. Methods: All patients with advanced HCC treated with sorafenib, lenvatinib or regorafenib as monotherapy from 4 December 2015 to 16 September 2020 were included. The Information Governance (Caldicott) guardian approved access to patients records. Median OS (mOS) and treatment duration (mDoT) were calculated using Kaplan-Meier estimate; baseline patient characteristics and information on dosing regimen, dose interruptions and modifications, were also collected. Results: The SHARP trial achieved mOS and mDoT of 10.7 and 5.3 months. 26% required dose reductions and 44% dose interruptions. Our patients achieved mOS of 7.98 and mDoT 2.62 months; 35% required a dose reduction and 48% a dose interruption. Mean daily dose (MDD) was not given in SHARP but was 663.8mg in REFLECT; 728mg sorafenib was achieved in West of Scotland (WOS). In REFLECT, lenvatinib patients had mOS and mDOT of 13.6 and 5.7 months; our patients had a mOS of 9.55 and mDoT of 3.68 comparatively. 37% and 40% had dose reductions and interruptions in REFLECT vs. 50% and 44% in WOS. Our patients had a MDD of 10.9 mg (12mg/day) and 6.88mg (8mg/day) lenvatinib vs. 10.5mg and 7.0mg in REFLECT. In RESORCE, mOS and mDOT of 10.6 and 3.6 months was acheived in regorafenib patients vs. 9.52 and 3.42 months in WOS. 68% needed drug interruption or reduction vs. 53% in WOS and MDD of regorafenib was 144·1mg vs. 148.3mg in our population. Conclusions: Consistent with other real world experiences our data demonstrate reduced mOS and mDoT times compared to trials. This is despite maintaining good dose intensity. On average, mOS was 2.6 months less in our population vs. trial data. This may be explained by etiological differences of underlying liver disease or treatment of less highly selected patients. Our patients tended to be older, have poorer PS and Child Pugh scores compared with trials. Our data lack control arms, preventing interpretation of whether improved survival outcomes over placebo are maintained. Interestingly our data mirror the results of REFLECT where lenvatinib treatment was maintained for longer periods compared to sorafenib, despite higher numbers of dose interruptions. Our data also confirm that observed within RESORCE where, remarkably, patients treated with second line regorafenib achieved mOS similar to those treated in the first line setting.