Early discordance between alpha-fetoprotein response and albumin-bilirubin deterioration refines response assessment in unresectable hepatocellular carcinoma receiving first-line immune-combination therapy: a real-world landmark cohort study
- 1Department of Hepatobiliary and Pancreatic Surgery, Affiliated Hospital of Nantong University, Nantong University Medical School
Abstract
Background: Early assessment during first-line immune-combination therapy for unresectable hepatocellular carcinoma (HCC) must consider both antitumor activity and hepatic tolerance. Existing integrated approaches have paired early AFP variation with baseline liver reserve, but they do not directly ask whether hepatic reserve is preserved or deteriorates over the same early treatment interval. We investigated whether early alpha-fetoprotein (AFP) response and change in modified albumin-bilirubin (ALBI) grade define clinically interpretable response-tolerance phenotypes.
Methods: This retrospective landmark cohort included AFP-evaluable patients with unresectable HCC who received first-line immune-combination therapy. Follow-up for the primary analysis began 70 days after treatment initiation. AFP response was defined as a landmark-to-baseline ratio of 0.80 or lower among patients with baseline AFP of at least 20 ng/mL; modified ALBI deterioration was defined as worsening by at least one category. Patients were classified as concordant favorable, response-liver vulnerability, nonresponse-liver preserved, or concordant failure. The primary endpoint was progression-free survival (PFS).
Results: Of 286 treated patients, 68 (23.8%) had baseline AFP below 20 ng/mL and were not AFP-evaluable; 166 patients entered the primary day-70 cohort. The four phenotypes included 58, 36, 43, and 29 patients, respectively. Relative to concordant favorable, concordant failure had the strongest adjusted association with shorter PFS (hazard ratio [HR] 2.85, 95% confidence interval [CI] 1.63–4.99; P <0.001), followed by nonresponse-liver preserved (HR 1.69, 95% CI 1.02–2.79; P = 0.040). The estimate for response-liver vulnerability was smaller and imprecise (HR 1.49, 95% CI 0.89–2.49; P = 0.130). The combined non-favorable group was associated with shorter PFS (HR 1.79, 95% CI 1.18–2.72; P = 0.006). The hepatic-decompensation estimate was imprecise after parsimonious adjustment (odds ratio 2.42, 95% CI 0.90–6.50; P = 0.081). Overall survival (OS) estimates were also imprecise; the largest point estimate was observed for concordant failure (HR 2.09, 95% CI 1.00–4.38; P = 0.050).
Conclusion: Early AFP-ALBI discordance distinguished different patterns of tumor-marker response and hepatic tolerance during immune-combination therapy. In AFP-evaluable patients with an ambiguous first assessment, a non-favorable biochemical pattern may support earlier multimodal reassessment of tumor status, liver function, and treatment feasibility, but it should not be used as a stand-alone rule for changing therapy.
Related
- Early on-treatment alpha-fetoprotein and des-gamma-carboxy prothrombin responses are associated with outcomes in unresectable hepatocellular carcinoma receiving transarterial chemoembolization, tyrosine kinase inhibitors and programmed cell death protein-1 inhibitors.
- Muscle strength and muscle mass phenotypes and survival in unresectable hepatocellular carcinoma receiving hepatic arterial infusion chemotherapy-based combination therapy
- Conversion Surgery After HAIC-Based Combination Therapy with Targeted Therapy and Immune Checkpoint Inhibitors for Initially Unresectable Hepatocellular Carcinoma: A Retrospective Cohort Study
- Revisiting alpha-fetoprotein thresholds for hepatocellular carcinoma surveillance in the metabolic dysfunction-associated steatotic liver disease era: time for an evidence-based recalibration.
- Association between alpha-fetoprotein elevation and recurrence in patients who achieved normalized alpha-fetoprotein after curative resection for hepatocellular carcinoma : A retrospective cohort study
BibTeX
@article{wu2026early,
title = {{Early discordance between alpha-fetoprotein response and albumin-bilirubin deterioration refines response assessment in unresectable hepatocellular carcinoma receiving first-line immune-combination therapy: a real-world landmark cohort study}},
author = {Haitao Wu and Pengju Tang and Jinbo Gong and Xintao Shi and Yiming Zhao and Zhong Chen},
journal = {Frontiers in Oncology},
year = {2026},
month = oct,
doi = {10.3389/fonc.2026.1954746},
url = {https://www.frontiersin.org/articles/10.3389/fonc.2026.1954746/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Early discordance between alpha-fetoprotein response and albumin-bilirubin deterioration refines response assessment in unresectable hepatocellular carcinoma receiving first-line immune-combination therapy: a real-world landmark cohort study
AU - Haitao Wu
AU - Pengju Tang
AU - Jinbo Gong
AU - Xintao Shi
AU - Yiming Zhao
AU - Zhong Chen
PY - 2026
DA - 2026/10/19
JO - Frontiers in Oncology
DO - 10.3389/fonc.2026.1954746
UR - https://www.frontiersin.org/articles/10.3389/fonc.2026.1954746/full
AB - Background: Early assessment during first-line immune-combination therapy for unresectable hepatocellular carcinoma (HCC) must consider both antitumor activity and hepatic tolerance. Existing integrated approaches have paired early AFP variation with baseline liver reserve, but they do not directly ask whether hepatic reserve is preserved or deteriorates over the same early treatment interval. We investigated whether early alpha-fetoprotein (AFP) response and change in modified albumin-bilirubin (ALBI) grade define clinically interpretable response-tolerance phenotypes. Methods: This retrospective landmark cohort included AFP-evaluable patients with unresectable HCC who received first-line immune-combination therapy. Follow-up for the primary analysis began 70 days after treatment initiation. AFP response was defined as a landmark-to-baseline ratio of 0.80 or lower among patients with baseline AFP of at least 20 ng/mL; modified ALBI deterioration was defined as worsening by at least one category. Patients were classified as concordant favorable, response-liver vulnerability, nonresponse-liver preserved, or concordant failure. The primary endpoint was progression-free survival (PFS). Results: Of 286 treated patients, 68 (23.8%) had baseline AFP below 20 ng/mL and were not AFP-evaluable; 166 patients entered the primary day-70 cohort. The four phenotypes included 58, 36, 43, and 29 patients, respectively. Relative to concordant favorable, concordant failure had the strongest adjusted association with shorter PFS (hazard ratio [HR] 2.85, 95% confidence interval [CI] 1.63–4.99; P <0.001), followed by nonresponse-liver preserved (HR 1.69, 95% CI 1.02–2.79; P = 0.040). The estimate for response-liver vulnerability was smaller and imprecise (HR 1.49, 95% CI 0.89–2.49; P = 0.130). The combined non-favorable group was associated with shorter PFS (HR 1.79, 95% CI 1.18–2.72; P = 0.006). The hepatic-decompensation estimate was imprecise after parsimonious adjustment (odds ratio 2.42, 95% CI 0.90–6.50; P = 0.081). Overall survival (OS) estimates were also imprecise; the largest point estimate was observed for concordant failure (HR 2.09, 95% CI 1.00–4.38; P = 0.050). Conclusion: Early AFP-ALBI discordance distinguished different patterns of tumor-marker response and hepatic tolerance during immune-combination therapy. In AFP-evaluable patients with an ambiguous first assessment, a non-favorable biochemical pattern may support earlier multimodal reassessment of tumor status, liver function, and treatment feasibility, but it should not be used as a stand-alone rule for changing therapy.
SN - 2234-943X
ER -