Metabolic–clinical subtypes predict survival in BCLC-C hepatocellular carcinoma receiving synchronous TACE plus targeted and immunotherapy
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background The synchronous combination of TACE, TKIs, and ICIs has improved outcomes for patients with BCLC-C HCC, yet treatment response varies widely. This study aimed to identify metabolically distinct phenotypes that predict response to synchronous triple therapy. Methods We conducted a retrospective cohort study of 1,792 patients with BCLC-C HCC. Propensity score matching and L2-penalized multivariate Cox regression were performed, followed by XGBoost with SHAP transformation and K-means clustering to define metabolic subtypes. Subgroup analyses with interaction testing were conducted across clinical strata to validate the robustness of the clustering system. Results After propensity score matching (716 pairs), synchronous triple therapy significantly improved overall survival compared with pre-immunotherapy TACE (median, 28.0 vs. 8.0 months; P < 0.001). Eight metabolic and clinical factors (AFP, TP, GGT, LDH, P, metastasis, HCO3, RBC) were identified as independent predictors. K-means clustering resolved four distinct metabolic phenotypes: metabolically fit (Cluster 1, 58.5%), metabolically compensated (Cluster 2, 24.6%), tumor-metabolism dominant (Cluster 3, 14.5%), and metabolically exhausted (Cluster 4, 2.4%). The metabolically fit phenotype demonstrated exceptional survival (median OS, 38.0 months; 1-year OS, 82.7%), while the metabolically exhausted phenotype showed a dismal prognosis (median OS, 4.0 months; 1-year OS, 4.2%). Subgroup analyses confirmed that the favorable prognosis of the metabolically fit phenotype remained consistent across all clinical strata. Conclusions The metabolic fitness-based clustering framework identifies patients with BCLC-C HCC most likely to benefit from synchronous triple therapy, suggesting that routine laboratory tests may help guide precision patient selection.
Source: BMC Cancer
Authors: Liqing Li, Zhitao Lin, Junbao Wei
Language: English
Subfield: Hepatology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,390
OpenAlex ID: W7213279646