HCC Heterogeneity Hepatocellular carcinoma (HCC) is not a single disease but a collection of molecularly distinct tumors that happen to arise in the liver. This heterogeneity explains why treatments that work for some patients fail in others - patients with different molecular subtypes may need fundamentally different treatment strategies.
Prior Classification Efforts Previous studies had proposed two or three HCC molecular classes based primarily on mRNA expression patterns. However, these classifications were limited by sample size, data type (expression only), or geographic scope. A comprehensive multi-dimensional molecular classification using large patient cohorts was lacking.
Study Objective This study combined data from 183 Japanese patients (TMDU cohort) and 373 TCGA patients to perform an integrated molecular analysis incorporating genomics, transcriptomics, proteomics, and immunological profiling, culminating in a robust three-subtype HCC classification.
Clinical Importance A validated molecular classification that integrates immune features with genomic and transcriptomic data is essential for matching HCC patients to emerging immunotherapies and targeted drugs, and for stratifying patients in clinical trials.
Cohort Description The 183-patient TMDU cohort provided Japanese patients primarily with HBV and HCV-related HCC, while the 373 TCGA patients represented Western populations with more diverse etiologies. The combination increased statistical power and geographic generalizability.
Data Integration The analysis integrated whole exome sequencing for mutations and copy number variations, RNA-seq for transcriptomic profiling, and computational deconvolution of tumor immune cell content from gene expression data. Unsupervised clustering identified reproducible molecular patterns across both cohorts.
Immune Profiling A key methodological contribution was systematic characterization of the tumor immune microenvironment (TIME) in each molecular subtype, using immune cell deconvolution algorithms to estimate the abundance and activation state of T cells, NK cells, macrophages, and other immune populations.
Validation Strategy Molecular subtypes identified in the TMDU cohort were validated in the TCGA cohort and vice versa, ensuring that the classification is not cohort-specific and can be applied to HCC patients from different populations.
MS1 - Mitogenic and Stem-Like Molecular Subtype 1 (MS1) is characterized by frequent TP53 mutations, chromosomal instability, and activation of cell cycle and DNA replication pathways. MS1 tumors show stem-like gene expression features and are the most aggressive, with the worst prognosis.
MS2 - CTNNB1-Mutated with Immunosuppression MS2 is defined by activating mutations in CTNNB1 (beta-catenin gene) and Wnt pathway activation. These tumors have a distinctive immunosuppressed microenvironment with low T cell infiltration, which may explain why HCC patients with CTNNB1 mutations have historically shown poor responses to immune checkpoint inhibitors.
MS3 - Metabolic Disease-Associated with Immune Infiltration MS3 tumors are enriched in patients with non-alcoholic steatohepatitis (NASH) and metabolic liver disease. In contrast to MS2, MS3 shows substantial immune infiltration - particularly T cells and macrophages - suggesting that these tumors may be particularly responsive to immunotherapy.
Survival Differences The three subtypes show significantly different overall survival outcomes, with MS1 having the poorest prognosis, MS3 intermediate, and MS2 relatively better (despite CTNNB1 mutations being targetable). This survival difference reflects the combined influence of tumor aggressiveness and immune context.
MS2 Immune Exclusion The CTNNB1-mutated MS2 subtype shows a characteristic pattern of immune exclusion - despite the presence of immune cells in the surrounding liver, the tumor tissue itself lacks significant T cell infiltration. Beta-catenin is known to drive expression of immune exclusion factors, explaining this immune 'cold' phenotype.
MS3 Immune Infiltration In contrast, MS3 tumors show 'hot' immune features with abundant CD8+ T cells and macrophages. However, the presence of immune cells does not guarantee anti-tumor activity - many infiltrating immune cells may be exhausted or suppressed by checkpoint molecules.
Checkpoint Molecule Expression Across all three subtypes, the expression levels of PD-L1, PD-1, and CTLA-4 vary substantially, suggesting that different subtypes may have different sensitivities to checkpoint inhibitors (anti-PD-1, anti-CTLA-4 antibodies) that are now approved for HCC.
Macrophage Polarization M2 macrophages (pro-tumorigenic, immunosuppressive) are enriched in MS1 and MS2, while MS3 shows a mix of M1 and M2 macrophages. Therapeutic strategies to repolarize macrophages toward the anti-tumor M1 phenotype could be particularly valuable in the MS1 and MS2 subtypes.
MS2 and Wnt Targeted Therapy The high frequency of CTNNB1 mutations in MS2 makes this subtype a candidate for Wnt/beta-catenin pathway inhibitors. Several such inhibitors are in clinical development, and the molecular definition of MS2 could guide patient enrichment in trials testing these agents.
MS3 and Immunotherapy The immune-inflamed phenotype of MS3 makes these patients likely responders to checkpoint inhibitors. Biomarker-driven enrollment of MS3 patients in immunotherapy trials could improve response rates and help define which HCC patients truly benefit from currently approved atezolizumab plus bevacizumab.
MS1 and Combination Strategies MS1's stem-like features and chromosomal instability suggest potential vulnerability to inhibitors of stem cell pathways or DNA damage response. However, the poor prognosis also indicates that MS1 patients may need aggressive combination approaches.
Etiology-Treatment Intersection The enrichment of NASH in MS3 creates an interesting connection - treating the underlying metabolic disease might reduce the oncogenic inflammatory signals that drive MS3 HCC, potentially combining systemic and hepatic treatment strategies.
Companion Diagnostic Development For the molecular subtype classification to guide treatment decisions, it must be translated into a clinically practical diagnostic test - ideally using RNA expression from biopsy material that can be obtained at diagnosis. Existing platforms like NanoString or clinical RNA-seq services could be adapted.
Prospective Biomarker Trials The hypothesis that MS3 patients respond better to immunotherapy and MS2 patients respond poorly requires prospective testing. Planned clinical trials should stratify HCC patients by molecular subtype and analyze outcomes accordingly.
Mechanisms of Immune Exclusion in MS2 The beta-catenin-driven immune exclusion in MS2 represents both a challenge and an opportunity. Understanding the specific molecular mechanisms of this exclusion could identify ways to convert immune-cold MS2 tumors into immune-responsive ones, potentially enabling immunotherapy in this currently refractory subtype.
Longitudinal Subtype Stability Whether HCC molecular subtypes change over time with disease progression or treatment is unknown. Studying biopsy pairs from the same patients at different disease stages would determine whether molecular classification needs to be repeated during the disease course.