81P - Single-cell spatial landscape of NSCLC reveals subtype specific immune features

Background

Single-cell technologies have revealed novel insights about the complexity of the tumor immune microenvironment. Most clinical strategies rely on histopathological stratification of tumor subtypes, yet the spatial context of cellular interactions within these stratified subgroups remains poorly understood. We previously applied imaging mass cytometry, a novel technology, to describe the tumor and immunological landscape of lung adenocarcinoma (LUAD) patient tumors across five histological subtypes of lung adenocarcinoma. We resolved over 1 million cells, enabling an understanding of the cellular spatial relationships associated with distinct clinical correlates, such as survival. Here, we compare the immune landscape of LUAD with lung squamous cell carcinoma (LUSC), a different subtype of lung cancer.

Methods

Imaging mass cytometry (IMC) is a novel technology which allows for the use of up to 50 antibodies labelled with metal isotopes to characterize the tumour and the tumour immune microenvironment. Our research group has optimized a panel of IMC markers delineating immune, tumour and structural cell types. Using IMC we have assessed the frequency of 16 different cell types, the interactions across these cells and their organization into cellular communities. We have compared these outputs across LUAD and LUSC patients.

Results

We show that LUAD and LUSC exhibit distinct immune profiles. Lung squamous cell carcinoma is characterized by a higher prevalence of neutrophils and a lower prevalence of macrophages compared to LUAD. In addition, our spatial analysis of immune lineages demonstrates contrasting cellular interactions across LUAD and LUSC as well as unique arrangements of immune cells into cellular neighborhoods. This highlights the spatial heterogeneity that exists across lung cancer subtypes beyond the prevalence of immune cell types alone.

Conclusions

Our results describe the importance of spatial interactions by demonstrating how distinct spatial profiles can exist across subtypes of lung cancer. Overall, we find unique immune profiles across LUAD and LUSC as it relates to immune frequencies, pairwise cellular interactions and communities of cells.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

The authors.

Funding

Rosalind and Morris Goodman Chair in Lung Cancer Research.

Disclosure

All authors have declared no conflicts of interest.