Integrating single-cell and bulk RNA sequencing data to characterize the heterogeneity of glycan-lipid metabolism polarization in hepatocellular carcinoma.

backgroundHepatocellular carcinoma (HCC) is high heterogeneity and remains an unmet medical challenge, but their metabolic heterogeneity has not been fully uncovered and required clinical applicable translational strategies.

methodsBy analyzing the RNA sequencing data in the in-house cohort and public HCC cohorts, we identified a metabolic subtype of HCC associated with multi-omics features and prognosis. Multi-omics alterations and clinicopathological information between different subtypes were analyzed. Gene signature, radiomics, contrast-enhanced ultrasound (CEUS), serum biomarkers were tested as potential surrogate methods for high throughput technology-based subtyping. Single-cell RNA sequencing analyses were employed to evaluate the immune characteristics changes between subtypes.

resultsBy utilizing metabolic-related pathways, we identified two heterogeneous metabolic HCC subtypes, glycan-HCC and lipid-HCC, with distinct multi-omics features and prognosis. Kaplan-Meier and restricted mean survival time analyses revealed worse overall survival in glycan-HCCs. And glycan-HCCs were characterized with high genomic instability, proliferation-related pathways activation and exhausted immune microenvironment. Furthermore, we developed gene signatures, radiomics, CEUS and serum biomarkers for subtypes determination, which showed substantial agreement with high-throughput-based classification. Single-cell RNA-seq showed glycan-HCCs were associated with multifaceted immune distortion, including exhaustion of T cells and enriched SPP1 + macrophages.

conclusionCollectively, our analysis demonstrated the metabolic heterogeneity of HCCs and enabled the development of clinical translation strategies, thus promoting understanding and clinical applications about HCC metabolism heterogeneity.