Tumor Tissue Spatial Structure Determines Immunotherapy Efficacy

News

Tumor Tissue Spatial Structure Determines Immunotherapy Efficacy

While immunotherapy has already found application in solid tumors, its application in breast cancer, especially triple-negative breast cancer, has remained somewhat unclear. Triple-negative breast cancer lacks various receptor expressions, is highly invasive, and often shows resistance to multiple treatment methods, making the need for new treatment approaches urgent.

In the case of triple-negative breast cancer, targeting PD-1 and PD-L1 has proven effective for some patients. However, reliable biomarkers for identifying treatment-sensitive patients have been lacking, and the interaction between immune checkpoint receptors and ligand-expressing cells often results in T-cell dysfunction.

Immunotherapy involving checkpoint blockade can activate dysfunctional T-cells. However, the interaction between the spatial organization of tumor tissue and the effectiveness of checkpoint blockade therapy has been a relatively unexplored area of research.

Recently, a team led by H. Raza Ali from the University of Cambridge published an article in Nature titled “Spatial Predictors of Immunotherapy Response in Triple-Negative Breast Cancer.” This study generated dynamic cell maps of triple-negative breast cancer in the context of checkpoint blockade therapy and found a correlation between the spatial structure of cell interactions and treatment response.

Tumor Tissue Spatial Structure Determines Immunotherapy Efficacy

To elucidate the relationship between tissue structure in triple-negative breast cancer and the response to immunotherapy, the authors collected patient samples at three different time points: before treatment (baseline, n = 243), on the first day of the second treatment cycle (during treatment, n = 207), and after surgical resection (post-treatment, n = 210). The authors used mass cytometry to analyze the expression of 43 genes in patient samples.

Additionally, they created in-situ platinum maps and found that platinum levels were significantly higher in samples during and after treatment compared to baseline, with most of the drug accumulating in macrophages.

Ultimately, the authors obtained 1855 high-quality tissue images. To precisely analyze the cell phenotypes in situ, they segmented individual cells in the maps and performed cluster analysis to identify various cell types, including epithelial cells, immune cells, and endothelial cells, as well as fibroblasts.

They evaluated spatial heterogeneity by analyzing correlations between cells in different images of the same tumor.

Tumor Tissue Spatial Structure Determines Immunotherapy Efficacy

Subsequently, the authors began investigating the relationship between untreated tumor tissue structure and immunotherapy responsiveness.

They first analyzed the relationship between cell density and reactivity in different epithelial cells and the tumor microenvironment, discovering that only PD-L1+IDO+ antigen-presenting cells could predict patient responsiveness to checkpoint blockade therapy.

The authors then examined whether different cell interactions were associated with treatment.

They considered cell-to-cell contact as an interaction.

Their analysis revealed a significant correlation between interactions of epithelial cells with the tumor microenvironment and treatment responsiveness, with the most significant correlations found in interactions between epithelial cells and CD20+B cells (Pinteraction = 0.003, FDR = 0.06) and epithelial cells and CD8+GZMB+T cells (Pinteraction = 0.006, FDR = 0.06).

The authors found that high expression of key activation markers in T cells that came into contact with tumor cells was highly associated with treatment responsiveness, and these contact T cells exhibited higher proliferation rates.

These findings suggest that the interaction between tumor and the immune system is a crucial indicator of immunotherapy responsiveness and indicates that checkpoint blockade therapy can enhance the potential for anti-tumor immune responses within the tumor.

Further analysis revealed that the proliferation level of tumor tissue could predict immunotherapy efficacy.

The authors calculated the correlation between the proliferation score of each cell type and immunotherapy, with the highest predictive factor being the proliferation level of CD8+TCF1+T cells.

The analysis of samples from patients during treatment revealed unique changes in tumor tissue induced by immunotherapy.

Immunotherapy induced the clonal expansion of CD8+ T cells, and the presence of CD15+ cells (including neutrophils and monocytes) during treatment could predict treatment resistance. CD15+ cells and CD15- cells were closely mixed together in a mosaic pattern.

However, the mechanism by which CD15+ cells induce treatment resistance remains unclear.

In summary, this study used mass cytometry to create dynamic cell maps of triple-negative breast cancer in the context of immunotherapy and revealed that tumor phenotype, activation status, and spatial organization jointly determine treatment responsiveness.