Early C-reactive protein kinetics predict outcomes in unresectable hepatocellular carcinoma treated with immune checkpoint inhibitors: a multicenter, observational study comparing different C-reactive protein kinetic definitions

Résumé fourni par la source

Background and aims: Immune checkpoint inhibitors (ICIs) have become the standard therapy for unresectable hepatocellular carcinoma (HCC). However, reliable prognostic biomarkers for treatment response remain warranted. This study assessed the prognostic value of early C-reactive protein (CRP) kinetics as an on-treatment biomarker in patients with hepatocellular carcinoma receiving ICIs. Methods: In this multicenter retrospective study, 289 patients with HCC treated with ICIs between 2019 and 2023 were analyzed. Patients were stratified using two established CRP kinetic models. The Fukuda model classified patients as flare responders, responders, or non-responders based on early CRP kinetics after treatment initiation. The Ishihara model categorized patients as "normal", "normalized", or "non-normalized" according to baseline and on-treatment CRP levels. Both models were additionally evaluated at eight weeks for earlier risk stratification. Results: According to the Fukuda model, flare responders demonstrated significantly prolonged median overall survival (34.5 vs 12.7 months), progression-free survival (12.42 vs 5.32 months), and higher objective response rates (53.7% vs 25.2%) compared with non-responders. In the Ishihara model, the normalized and normal groups showed significantly improved outcomes with median overall survival of 28.35 and 25.1 months, progression-free survival of 12.48 and 9.89 months, and objective response rates of 54.2% and 43.4%, respectively, compared with the non-normalized group (7.46 months; 3.35 months; 18.8%). The adapted 8-week Ishihara model enabled earlier risk stratification prior to first radiologic staging. Conclusions: Early C-reactive protein kinetics provide robust prognostic information in patients with HCC treated with ICIs. Both CRP-based models identify subgroups with significantly improved outcomes, with the Ishihara model enabling earlier identification of patients at high risk of treatment failure and thereby supporting earlier risk stratification during immunotherapy.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.