The remarkable success of checkpoint control inhibitors in treating a variety of different cancers has necessitated a deeper assessment of the tumor and its microenvironment at the genetic and phenotypic level. Data from recent clinical trials have unequivocally established that the tumor microenvironment significantly impacts the efficacy of immune-oncology drugs.
We have taken a gene expression signature-based approach to qualitatively and quantitatively assess the epithelial, stromal and immune content of tumors from RNA-seq data. The immune cell content of the tumors was further stratified to determine the infiltration pattern of nine different immune cell types including CD8+/CD4+ T-cells, Treg cells, NK cells, dendritic cells, B-cells, myeloid-derived suppressor cells (MDSC) and M1/M2 macrophages in the tumors using gene signatures specific to each immune cell type.
We applied these signatures singly, or in combination on the TCGA RNA-seq data from 33 cancers. Our analysis supports that there is an underlying molecular symmetry in tumors with higher CD8+ T-cell infiltration across most cancers characterized by the upregulation of a common set of genes that mapped to interferon signaling, antigen presentation and a pro-inflammatory cytokine network pathways. The upregulated genes were strongly correlated with patient survival in skin cutaneous melanoma (SKCM) and other cancers providing a framework for identifying biomarkers of patient response to checkpoint control inhibitors.
We present a case study to support the idea that gene expression signatures can address a critical unmet need in the immuno-oncology space, which is to create a framework for treating tumors that carry less mutation burden combined with poor T-cell infiltration. As an example, we analyzed 476 skin cutaneous melanoma (SKCM) and 80 uveal melanoma (UVM) samples from TCGA. The UVM melanoma has ∼10-fold lower median mutational burden compared to SKCM, which correlates with a lower (<10-fold) T-cell neo-epitope content in these tumors. As expected, immune cell infiltration of UVM was significantly lower compared to SKCM and so were the infiltration of different immune cell types, indicating that UVMs are immunologically barren compared to SKCMs. However, we observed interesting differences in the immune cell content of the two tumors. Whereas CD8+ T-cell infiltrated tumors were far fewer in UVM <5% compared to SKCM (∼30%), the immune microenvironment was qualitatively different in these tumors. In UVM CD8 T-cell infiltration was correlated positively with pro-inflammatory antigen presenting M1 macrophages and dendritic cells as well as immune-suppressive MDSCs, Treg cells and M2 macrophages. By contrast, CD8+ T-cell infiltrated SKCM tumors had significantly lower levels of MDSCs and M2 macrophages and were enriched in dendritic cells, M1 macrophages and Treg cells. Significantly, in UVM, the macrophage content was dominated by M2 macrophages (M1:M2, 1:2), whereas in SKCM they were similar. Taken together, our analysis suggests that in UVM, the immunesuppressive tumor microenvironment is contributed by the combined presence of MDSCs and M2 macrophages, whereas in SKCM it is contributed by the infiltration of Treg cells.
Based on our findings, we propose that therapies targeting MDSC cells, or those that can shift the balance towards increased M1 macrophage content over M2 are likely to show efficacy in UVM melanoma, which are largely unresponsive to checkpoint inhibitor molecules.
Citation Format: Ravi Gupta, Nitin Mandloi, Ashwini Patil, Malini Manoharan, Rekha Sathian, Kiran V. Paul, Amitabha Chaudhuri. Analysis of tumor microenvironment identifies pathways predicting response to checkpoint control inhibitors: A case study comparing the immune microenvironment of uveal melanoma vs skin cutaneous melanoma [abstract]. In: Proceedings of the Second CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; 2016 Sept 25-28; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(11 Suppl):Abstract nr A121.
- ©2016 American Association for Cancer Research.