Editors' ChoiceCancer

Cancer-immune topology influences lung cancer evolution

See allHide authors and affiliations

Science Translational Medicine  10 Jun 2020:
Vol. 12, Issue 547, eabc8944
DOI: 10.1126/scitranslmed.abc8944


The extent of lymphocyte infiltration controls cancer subclone divergence, and a single immune cold region can increase the risk of relapse.

Ecological constraints within the tumor microenvironment shape the emergence of genetically divergent cancer cell clones and subclones. In addition to intratumor heterogeneity at the cancer cell level, immune and stromal cells exhibit irregular patterns of organization in the tumor microenvironment, resulting in a complex tumor-stroma-immune interface. Integrating spatial profiles of immune cell infiltration with cancer cell genetic divergence analyses offers a new approach to elucidate the forces driving cancer evolution and identify predictors of disease progression.

AbdulJabbar et al. developed a computational pathology pipeline to classify single cells from histology images and quantify the fractions of cancer, stromal, and immune cells in lung tumors from the TRAcking Cancer Evolution through Therapy (Rx) (TRACERx) study. This scalable and unbiased approach allowed for classification of distinct regions within each tumor into immune hot or immune cold and was validated using RNA-sequencing signatures of immune cell infiltration. By analyzing the genomic distance between subclones within a single tumor, the authors found that dominant tumor cell clones diversified faster in immune hot regions compared with immune cold regions. Interestingly, this association was only present in lung adenocarcinoma subtypes and not in lung squamous cell carcinoma. Analysis of the number of immune cold regions within a tumor revealed an association with disease-free survival that was validated in an independent cohort. Compared with immune hot regions, immune cold regions exhibited an irregular and complex tumor-stroma interface (high fractal dimension) but no difference in the percentage of stroma cells. Lymphocyte numbers at the tumor boundary were higher for samples with a high clonal neoantigen score, and an enrichment of the effector/regulatory T cell ratio (CD8+/CD4+FOXP3+) was found adjacent to the tumor nests compared with locations distal from the tumor boundary.

The link between spatial variability of the tumor-immune-stromal interface and intratumor genetic heterogeneity provides important insights into how tumor microenvironment selection pressures shape tumor evolution. Future studies are needed to investigate the functional role of different immune and stromal cell subpopulations in immune evasion mechanisms. In addition, a better understanding of antigen presentation mechanisms in tumor cells from different lung cancer subtypes should aid the development of new therapeutic targets.

Highlighted Article

View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article