25 July 2024 in Journal of Cancer Research and Clinical Oncology
Background: Currently, there is a lack of effective indicators for predicting the efficacy of immunotherapy in patients with advanced hepatocellular carcinoma (HCC). This study aimed to investigate the expression and prognostic value of peripheral T lymphocyte subsets in advanced HCC. Methods: Patients with advanced HCC who were treated with immune checkpoint inhibitors (ICIs) from December 2021 to December 2023 were included in the study. Flow cytometry was used to detect lymphocyte subsets before treatment. The patients were divided into disease control (DC) and nondisease control (nDC) groups based on treatment efficacy. Relationships between the clinical characteristics/peripheral T lymphocytes and immunotherapy efficacy were analyzed. The effectiveness of peripheral T lymphocyte subsets in predicting immunotherapy efficacy for patients with advanced HCC was analyzed using receiver operating characteristic (ROC) curves. Results: A total of 40 eligible patients were included in this study. Non-DC was significantly associated with higher albumin-bilirubin (ALBI) scores. The percentages of γδ+Vδ2+PD1+ T cells and γδ+Vδ2+Tim3+ T cells were greater in the nDC group than in the DC group. Multivariable regression analysis revealed that the ALBI score and T lymphocytes expressing γδ+Vδ2+PD1+ and γδ+Vδ2+Tim3+ were founded to be independent influencing factors. The area under the ROC curve (AUC) values for these combinations was 0.944 (95% CI, 0.882 ~ 1.000). Conclusions: The calculation of the ALBI score and determination of the percentages CD3+γδ+Vδ2+PD1+ T lymphocytes and CD3+γδ+Vδ2+Tim3+ T lymphocytes in the peripheral blood of patients with advanced HCC are helpful for predicting the patients’ responses to ICIs, helping to screen patients who may clinically benefit from immunotherapy. Retrospectively registered: number: ChiCTR2400080409, date of registration: 2024-01-29.
10 September 2026 in SSRN Electronic Journal
Background: Postoperative adjuvant therapy has yielded inconsistent benefits for patients with hepatocellular carcinoma (HCC) considered at high risk of recurrence after resection. This may partly reflect the limited ability of conventional clinicopathological criteria to capture the multi-dimensional heterogeneity underlying postoperative recurrence and precisely identify patients at genuinely high risk. We aimed to develop and externally validate an artificial intelligence (AI)-enabled multimodal framework to characterize postoperative recurrence heterogeneity and refine risk stratification after HCC resection.Methods: This multicenter retrospective study included 556 patients across development, external validation, exploratory adjuvant therapy cohorts and selected TCGA cohort. Clinical variables, MRI-derived radiomic features, and AI-derived histopathological representations were integrated for recurrence risk stratification. Histopathological interpretation, bulk transcriptomics, and single-cell RNA sequencing were used to characterize high-risk recurrence. Results: The multimodal framework achieved a C-index of 0.844 (95%CI, 0.772-0.905) in the development cohort and 0.774 (0.664-0.867), 0.818 (0.736-0.886) across external cohorts. Multimodal stratification further resolved recurrence heterogeneity within conventional risk groups: conventionally low-risk patients reclassified as multimodal high risk had increased recurrence risk (HR, 8.76; 95% CI, 4.12-18.60), whereas conventionally high-risk patients reclassified as multimodal low risk had reduced risk (HR, 0.15; 95% CI, 0.10-0.22). High-risk recurrence phenotyping was associated with invasive histopathological features, dysfunctional CD8+ T-cell states, and tumor-associated macrophage programs.Interpretation: AI-enabled multimodal phenotyping identifies clinically and biologically distinct postoperative recurrence-risk states beyond conventional clinicopathological stratification. This approach may provide a biologically informed framework for postoperative risk assessment and support more refined risk enrichment and patient stratification in future adjuvant clinical trials.