Immunotherapy has become front-line treatment for most non-small cell lung cancer (NSCLC) patients but has only marginally improved outcomes. The question of why some patients have durable responses to immune checkpoint blockade (20% of patients) but the majority have none or only a brief response, remains a fundamental knowledge gap. A likely possibility is that changes in the tumour microenvironment, immune cells or the tumour cells themselves drive differential responses to immunotherapy. The recent emerging technology of spatial -omics allows this to be addressed with unparalleled resolution, by integrating cell spatial organization with deep molecular profiling.
We have performed spatial proteomics and spatial transcriptomics analyses of a cohort of resected NSCLC samples. Spatial proteomics using the MIBIscope evaluated the expression of 38 protein markers in individual cells while keeping the tissue architecture intact. Spatial transcriptomic was performed using the Nanostring GeoMx platform to resolve spatially the tumour transcriptome. Integration of these two datasets will determine how a specific immune context can drive tumour evolution.
We have found that tumours that grow in an immune-inflamed environment (e.g. in ever-smoker patients) have a high prevalence of resident memory T cells. The presence of lung-resident memory T cells increased the recruitment of tumour infiltrating lymphocytes, yet accelerated the immune evasion in lung tumour cells. These results demonstrate that ever-smoker tumours are under immune pressure from their inception, forcing early onset of immune escape mechanisms in a subset of patients which ultimately leads to their insensitivity to immune checkpoint inhibitor treatment.