By Pooja Toshniwal PahariaReviewed by Lauren HardakerFeb 11 2026By charting immune ecotypes across squamous cell carcinoma, basal cell carcinoma, and melanoma at single-cell resolution, the study shows how metabolic reprogramming and macrophage polarization reshape the tumor microenvironment, helping explain immune evasion and pointing toward more precise patient stratification strategies.
Image credit: Pixel-Shot/Shutterstock.com
In a recent study published in Clinical and Translational Medicine, researchers present an integrative immune atlas of skin cancers, combining single-cell transcriptomic analyses with complementary functional analyses across multiple independent cohorts.
Single-cell atlas reveals immune ecotypes
The study identifies shared and tumour-specific immune remodelling programmes, including immune-evasive tumour cell subsets and distinct macrophage polarization states. Notably, a melanoma subpopulation with major histocompatibility complex-1 (MHC-I) downregulation was associated with poor survival, while secreted phosphoprotein 1-positive (SPP1+) macrophages correlated with advanced disease, highlighting immune heterogeneity as a key driver of progression and therapeutic response.
The authors further report that acral melanoma (AM) exhibits persistently low MHC-I activity even at early stages, whereas cutaneous melanoma (CM) shows stage-associated MHC-I downregulation during progression. Together, these findings position immune “ecotypes” as a unifying framework linking tumour metabolism, immune architecture, and clinical outcome across skin cancer subtypes.
Distinct immune architectures across skin cancers
Skin cancers comprise a biologically diverse group of malignancies, including squamous cell carcinoma (SCC), basal cell carcinoma (BCC), acral melanoma (AM), and cutaneous melanoma (CM), which differ markedly in clinical presentation, metastatic behavior, and patient outcomes. Variation in immunotherapy efficacy across these subtypes highlights fundamental differences in immune architecture and resistance pathways.
During disease progression, the tumour microenvironment (TME) undergoes remodelling involving lymphoid and myeloid reprogramming, spatial reorganization, and alterations in T lymphocyte activity. However, the mechanisms by which these immune ecosystems evolve across skin cancer types and stages to promote immune escape remain unclear.
Multi-cohort analysis across 70 patients
In the present study, researchers conducted a comprehensive analysis of immune remodelling across major skin cancer subtypes using single-cell transcriptomic and integrative multi-cohort approaches, integrating newly generated and publicly available scRNA-seq datasets with bulk RNA-seq prognosis cohorts and publicly available spatial transcriptomic resources. They obtained single-cell ribonucleic acid (RNA) sequencing (scRNA-seq) data from 102 samples derived from 70 patients, including SCC, BCC, AM, and CM, across varying stages to capture immune ecosystem changes during disease progression. They also sampled adjacent normal tissues for comparison.
Clinical staging followed the American Joint Committee on Cancer (AJCC) TNM classification, eighth edition. The team classified carcinomas ≤2.0 cm without high-risk pathological features as early stage, while larger or high-risk lesions were considered advanced. For melanoma, they defined localized tumours without lymph node involvement or metastasis as early-stage, whereas those with lymphatic or metastatic spread were late-stage.
The researchers identified immune ecotypes, analysed cell–cell communication networks, and estimated immune cell subtype enrichment in bulk datasets using single-cell-derived signatures. Single-cell data were integrated with bulk RNA-seq prognosis cohorts to link immune cell states and tumour subpopulations with clinical outcomes. Immunofluorescence staining validated SPP1+ macrophage accumulation and stage-associated spatial patterns, while in vitro experiments using human (THP-1 and SK-MEL-28) and murine cell lines (RAW264.7 and B16) examined immune–tumour interactions.
The team performed quantitative real-time polymerase chain reaction (qRT-PCR) and transcriptomic profiling to identify differentially expressed genes. Gene set enrichment analysis (GSEA) characterized functional programmes. Lastly, the researchers queried clustered regularly interspaced short palindromic repeats (CRISPR)-associated protein 9 (Cas9) dependency datasets to assess the functional relevance of candidate genes.
Results: Mapping 259,000 cells across subtypes
The analysis yielded 259,761 high-quality single cells, including melanocytes, epithelial cells, T/NK cells, monocytes/macrophages, dendritic cells (DCs), fibroblasts, endothelial cells, and B/plasma cells. The TME showed marked cancer type- and stage-specific immune remodelling. SCC exhibited the highest abundance of monocytes/macrophages and DCs, while advanced BCC and melanoma samples showed lower immune infiltration than early-stage tumours.
Mitochondrial reprogramming defines aggressive melanoma
Comparative analyses revealed distinct transcriptional programmes across tumour types. BCC and SCC demonstrated strong p53 pathway activation, while SCC and early-stage CM showed enhanced interferon (IFN) signalling and MHC-related antigen presentation. Contrastingly, advanced melanomas demonstrated increased angiogenesis and epithelial–mesenchymal transition (EMT) scores, alongside enrichment of mitochondrial oxidative phosphorylation programmes, indicating increased invasiveness and immune evasion.
The researchers identified a malignant melanoma cell population expressing NADH:ubiquinone oxidoreductase subunit C2 (NDUFC2) and mitochondrial asparaginyl-tRNA synthetase 2 (NARS2), with decreased MHC-I levels, that was abundant in advanced disease. This population was associated with treatment resistance and poor survival. Functional dependency analyses confirmed NDUFC2 and NARS2 as essential for melanoma cell proliferation, supporting a link between mitochondrial metabolic reprogramming and immune evasion.
Two macrophage states with opposing outcomes
Tumour-associated macrophages (TAMs) arose from recruited monocytes and tissue-residing macrophages and adopted two prognostically distinct states. Pro-inflammatory macrophages expressing C-X-C motif chemokine ligand (CXCL)-9 and 10 predominated in SCC, whereas tissue-remodelling SPP1⁺ macrophages were enriched in advanced, metastatic melanomas with poor prognosis. Melanoma-conditioned media promoted macrophage expression of M2-associated markers, accompanied by increased SPP1.
Experimental knockdown or overexpression of SPP1 in macrophages modulated M2 marker expression, supporting a regulatory role for SPP1 in macrophage polarisation. Cell–cell communication analyses further implicated tumour-derived signalling pathways in shaping SPP1⁺ macrophage states within melanoma microenvironments.
Five reproducible immune ecotypes identified
The team identified five immune ecotypes, T-cell-dominant, stromal-enriched, balanced, desert, and myeloid-enriched states, that were reproducible across cohorts and skin cancer subtypes. A shift toward immune-desert ecotypes marked disease progression in BCC, AM, and CM, whereas SCC uniquely transitioned to a myeloid-enriched state. The ecotypes showed distinct signalling profiles, with MHC-I activity defining T-cell-dominant tumours and elevated SPP1-associated macrophage signalling characterizing immune-desert microenvironments, underscoring their prognostic and therapeutic relevance.
Translating atlas insights into precision immunotherapy
The integrative analysis establishes a comprehensive immune remodelling atlas across skin cancers, revealing shared and tumour-specific mechanisms of immune evasion with clear clinical relevance. The identification of an immune-evasive NARS2⁺NDUFC2⁺ melanoma subpopulation links mitochondrial metabolic programmes to reduced MHC-I expression and poor prognosis, highlighting opportunities to restore tumour immunogenicity.
The study also identifies SPP1⁺ TAMs as key drivers of immune-desert microenvironments and adverse outcomes. By defining ecotype-based immune landscapes, the findings support improved patient stratification and inform personalized immunotherapy strategies. However, further validation using larger cohorts, in vivo lineage tracing, and functional validation of myeloid- and metabolism-targeted interventions is required for clinical translation.
Journal Reference
Huang, L. et al. (2026). Integrative single-cell analysis uncovers distinct tumour microenvironment ecotypes and immune evasion across skin cancers. Clinical and Translational Medicine, 16(2), e70611. DOI: 10.1002/ctm2.70611. https://onlinelibrary.wiley.com/doi/10.1002/ctm2.70611
Scientists Uncover How Cell Growth and Pressure Shape Pancreatic Ducts