The authors applied single-cell spatial transcriptomics to compare HER2-low and HER2-high breast tumors with the goal of defining microenvironmental features associated with treatment sensitivity and resistance. This approach allowed simultaneous resolution of malignant, stromal, and immune compartments while preserving spatial relationships among cell types within the tumor microenvironment.
Analysis identified a heterogeneous landscape across tumor, stromal, and immune populations. Among immune populations, dendritic cells emerged as a notably remodeled compartment in HER2-low tumors. The spatial data revealed not only the presence of multiple immune cell types but also their organization into distinct spatial niches, some of which were enriched for myeloid cells while others maintained intermixing with effector immune populations.
Focused analysis resolved multiple dendritic cell states within the HER2-low tumors. The reported states included:
These distinct states represent a spectrum from homeostatic or classical cDC identities to interferon-activated and plasmacytoid phenotypes, and the presence and relative abundance of these states differed between resistant and sensitive HER2-low tumors.
Spatial proximity analysis demonstrated marked architectural differences correlating with therapeutic response:
Resistant HER2-low tumors exhibited increased segregation of tumor epithelial cells from effector immune populations. This segregation manifested as an enrichment of myeloid-rich immune niches and an overall immune-restricted spatial architecture.
In resistant tumors, there was enrichment of homeostatic and classical cDCs together with depletion of IFN-activated cDCs and pDCs. The study also found altered communication patterns between tumor cells, myeloid populations, and T cells in these resistant contexts.
By contrast, sensitive HER2-low tumors retained immune-intermixed niches that were enriched for antigen presentation and interactions characteristic of effector immune responses.
These spatial patterns suggest that not only the identity but the spatial organization of dendritic cell states and myeloid niches are associated with therapeutic resistance versus sensitivity.
The investigators performed independent validation using TCGA BRCA data to test the clinical relevance of the dendritic-cell states observed in the spatial dataset. Key associations reported included:
Elevated homeostatic cDC2 signatures predicted poorer survival.
Inflammatory dendritic cell signatures, including signatures related to IFN-activated and plasmacytoid states, were associated with more favorable outcomes.
These external data support the notion that dendritic cell state composition and associated transcriptional programs have prognostic implications in breast cancer beyond the discovery cohort.
The coordinated spatial remodeling of dendritic cell states and the emergence of myeloid-rich, immune-restricted niches in resistant HER2-low tumors indicate potential translational opportunities:
Dendritic-cell myeloid niche organization may serve as a candidate biomarker to stratify HER2-low patients by likely therapeutic response.
Therapeutic strategies that shift dendritic cell states toward IFN-activated or inflammatory phenotypes, or that disrupt myeloid-rich resistant niches, could represent potential vulnerabilities to overcome resistance in HER2-low disease.
The authors propose that understanding and targeting the spatial immune microenvironment, particularly dendritic cell programs and myeloid-T cell communications, may improve outcomes in HER2-low breast cancer.
Details such as cohort size, specific therapeutics evaluated, experimental parameters, and statistical metrics were not reported in the abstract and would need to be consulted in the full preprint for comprehensive assessment. The preprint lists no competing interests. Funding was declared from the NIH Common Fund (1R01GM163238). The work was posted as a bioRxiv preprint and the authors granted bioRxiv a license to display the manuscript in perpetuity.