JHEP Reports · Published 2026-06-12 · DOI 10.1016/j.jhepr.2026.101904
Marie Decraecker, Héloïse Bourien, Elhadji Malick Thiam, Julien Edeline, Alice Koudjou-Sonegueng, Philippe Merle, Jean-Frédéric Blanc, Giuliana Amaddeo, Thomas Uguen, Thomas Decaens, Nathalie Ganne-Carrie, Christine Silvain, Mohamed Bouattour, Ghassan Riachi, Manon Allaire, Stéphane Cattan, Xavier Adhoute, Marilyne Gratien, Rodolphe Anty, Jean-Marie Peron, Fréderic Oberti, Aurore Baron, Alina Pascale, Georges-Philippe Pageaux, Sylvain Manfredi, Jean-Baptiste Nousbaum, Marie Lequoy, Anaïs Jaillais, Isabelle Ollivier-Hourmand, Moana Gelu-Simeon, Jean-Pierre Bronowicki, Alexandra Heurgue, Anne-Laure Villing, Isabelle Rosa, Jean-François Cadranel, Gérard Ducournau, Olivier Ganry, Charlotte Costentin, Eric Nguyen-Khac
Background & Aims: Unresectable hepatocellular carcinoma (HCC) remains a major global health burden. Immunotherapy-based combinations have become the standard of care in first-line (1L) treatment, but evidence on subsequent therapies is limited. We aim at describing access to and outcomes of second-line (2L) treatments following atezolizumab–bevacizumab (AB) compared with sorafenib (Sor), using real-world data from the prospective, French CHIEF cohort. Methods: Patients were included in the CHIEF cohort, part of the prospective, real-world STRETCH study (Systemic TReatment sEquences in paTients with unresectable HCC). Adults with unresectable HCC who received 1L treatment with AB or Sor between September 2019 and September 2024 were analyzed. Median overall survival (mOS) and median progression-free survival were calculated from 2L treatment initiation. Results: Among 1,103 patients included (AB, n = 899; Sor, n = 204), baseline characteristics were broadly similar, with most patients having Child-Pugh A liver function (77.1% vs. 70.3%) and Barcelona Clinic Liver Cancer stage C disease (66.3% vs. 84.3%). The mOS was 22.3 (95% CI 18.5–27.4) months with AB and 9.4 (7.3–12.7) months with Sor (p <0.0001). After progression, 42.1% of AB-treated and 60.0% of Sor-treated patients received 2L treatment (p <0.001). After AB, 70.8% received tyrosine kinase inhibitors (TKIs) with mOS of 13.0 (9.8–15.5) months; patients receiving immunotherapy or combination regimens did not reach survival (p = 0.0009). The mOS with 2L TKIs was similar after AB or Sor (13.0 vs. 8.6 months; p = 0.082). Conclusions: In this prospective real-world cohort, access to 2L treatment was lower after AB than after Sor. Nonetheless, 2L TKIs achieved numerically similar survival outcomes irrespective of 1L therapy, whereas immunotherapy rechallenge yielded encouraging results in selected patients. Impact and implications: The increasing use of immunotherapy-based combinations in first-line treatment for unresectable hepatocellular carcinoma raises crucial questions about optimal sequencing strategies after progression. In this study, we provide prospective, real-world evidence on second-line treatment access and outcomes in the post-immunotherapy setting. These findings are important for clinicians and researchers seeking to refine treatment algorithms and ensure equitable access to effective therapies. In practice, the numerically similar survival achieved with tyrosine kinase inhibitors across treatment sequences and the promising results of immunotherapy rechallenge may inform individualized patient management and guide the design of future clinical trials evaluating sequential strategies.
Abstract from DOAJ. Public domain (CC0 1.0).
Read the article at the publisher →
Decraecker, M., Bourien, H., Thiam, E., et al. (2026). Second-line practices in the era of immunotherapy in HCC: The CHIEF cohort. JHEP Reports. https://doi.org/10.1016/j.jhepr.2026.101904