Abstract:2576 Background: Tumor microenvironment (TME) heterogeneity leads to the discrepancy of survival prognosis and clinical treatment response for HCC patients. Although the documented molecular subtypes moderately describe the TME heterogeneity and characteristics, their clinical application is constrained by several issues. Methods: We integrated three single-cell datasets from 39 HCC patients to describe the TME landscape and identified prognosis-related cell subclusters. Unsupervised clustering of subcluster-specific markers (SSMs) was performed to generate transcriptomic subtypes. The predictive value of these molecular subtypes for prognosis and treatment response was explored in multiple external HCC cohorts and Xiangya real-world HCC cohort with patients who received immune checkpoint blockade (ICB) therapy. Cancer stemness was estimated using bioinformatic methods and in vivo and in vitro experiments. The features of TME were further validated using single-cell RNA-seq and immune repertoire sequencing, mass cytometry (CyTOF) and multiplex immunofluorescence (mIF). Ultimately, we constructed prognosis-related score (PRS)based on machine learning algorithm, and identified the potential therapeutic targets and agents for high-PRS patients. Results: The comprehensive single-cell analysis provided a high-resolution depiction of TME heterogeneity in HCC and confirmed 6 cell subclusters of prognostic relevance. Five transcriptomic subtypes were constructed using SSMs, which possessed different clinical prognosis, stemness characteristics, immune landscape and therapeutic responses. Class 1 indicated an inflamed phenotype with better clinical outcomes, while Class 2 and class 4 demonstrated immune-deserted phenotypes lacking T cells infiltration. Class 5 and class 3 indicated inhibitory tumor immune microenvironment enriched regulatory T cells and suppressive immune checkpoints. Multiple therapeutic cohorts suggested that Class 5 and class 3 were sensitive to ICB and targeted therapy, while Class 1 and Class 2 were more responsive to transcatheter arterial chemoembolization treatment (TACE). Class 4 displayed resistant to all conventional HCC therapies. The PRS performed well in prognostic prediction in multiple HCC cohorts. Four potential therapeutic agents and four targets were further identified for high-PRS HCC patients. Conclusions: Our study generated a clinically valid molecular classification, thereby providing guidance for precision medicine in HCC patients.

Peeling Off the Hidden Genetic Heterogeneities of Cancers Based on Disease-Relevant Functional Modules

Unraveling the hidden heterogeneities of breast cancer based on functional miRNA cluster.

Unravelling the hidden heterogeneities of diffuse large B-cell lymphoma based on coupled two-way clustering

Identification of Molecular Subtypes for Integrated Multi-Omics Analysis for Use in Guiding Precision Medicine in Hepatocellular Carcinoma.

Identifying Disease Feature Genes Based on Cellular Localized Gene Functional Modules and Regulation Networks

A New Strategy for Exploring the Hierarchical Structure of Cancers by Adaptively Partitioning Functional Modules from Gene Expression Network

Molecular Subtyping of Cancer Based on Distinguishing Co-Expression Modules and Machine Learning

Predicting Stage-Specific Cancer Related Genes and Their Dynamic Modules by Integrating Multiple Datasets

Functional modules with disease discrimination abilities for various cancers

Identification of breast cancer risk modules via an integrated strategy

Modularity-based Credible Prediction of Disease Genes and Detection of Disease Subtypes on the Phenotype-Gene Heterogeneous Network

Identifying Cancer Prognostic Modules by Module Network Analysis

[Identification of Gene Functional Modules Shared by Cancers Based on Biclustering].

An integrated approach to identify causal network modules of complex diseases with application to colorectal cancer.

A Cancer Gene Module Mining Method Based on Bio-Network of Multi-Omics Gene Groups

Identifying Cancer Patient Subgroups by Finding Co-Modules from the Driver Mutation Profiles and Downstream Gene Expression Profiles

Integration of multi-omics data to mine cancer-related gene modules

Identifying Disease Genes and Module Biomarkers by Differential Interactions

Towards precise classification of cancers based on robust gene functional expression profiles

Exploring Phenotype-Associated Modules in an Oral Cavity Tumor Using an Integrated Framework

Identifying Lymph Node Metastasis-Related Factors in Breast Cancer Using Differential Modular and Mutational Structural Analysis