Characterisation of tumour microenvironment remodelling following oncogene inhibition in preclinical studies with imaging mass cytometry

Febe Van Maldegem,Dina Levi,Harshil Patel,Karishma Valand,Charles Swanton,Mihaela Angelova,Victoria Siu Kwan Tsang,Nourdine Bah,Gavin Kelly,Miriam Molina-Arcas,Julian Downward,Edurne Mugarza,Emma Colliver,Sareena Rana,Philip Hobson,Megan Cole,Katey Enfield,Christopher Moore

doi:10.1038/s41467-021-26214-x

Abstract

Mouse models are critical in pre-clinical studies of cancer therapy, allowing dissection of mechanisms through chemical and genetic manipulations that are not feasible in the clinical setting. In studies of the tumour microenvironment (TME), multiplexed imaging methods can provide a rich source of information. However, the application of such technologies in mouse tissues is still in its infancy. Here we present a workflow for studying the TME using imaging mass cytometry with a panel of 27 antibodies on frozen mouse tissues. We optimise and validate image segmentation strategies and automate the process in a Nextflow-based pipeline (imcyto) that is scalable and portable, allowing for parallelised segmentation of large multi-image datasets. With these methods we interrogate the remodelling of the TME induced by a KRAS G12C inhibitor in an immune competent mouse orthotopic lung cancer model, highlighting the infiltration and activation of antigen presenting cells and effector cells.

Highlights

Mouse models are critical in pre-clinical studies of cancer therapy, allowing dissection of mechanisms through chemical and genetic manipulations that are not feasible in the clinical setting
Highly multiplexed imaging technologies, such as imaging mass cytometry (IMC) that is based on unique atomic mass are very attractive, permitting in depth characterisation of the tumour microenvironment (TME) with a metal-conjugated antibody panel of up to 40 markers while retaining the spatial context[5,6]
Two recent studies reported enhanced survival when combining the KRAS G12C inhibitors with anti-PD-1 immune checkpoint blockade in an immunotherapy sensitive syngeneic KRAS G12C mutant CT26 colon carcinoma subcutaneous tumour model[15,16]. This prompted us to investigate the effects of tumour-specific KRAS inhibition on the TME in the context of a preclinical model of lung cancer, the 3LL ΔNRAS cell line, a KRAS G12C mutant and NRASknockout Lewis lung carcinoma derivative that we have previously shown to be sensitive to KRAS G12C inhibition[17]

Summary

Introduction

Mouse models are critical in pre-clinical studies of cancer therapy, allowing dissection of mechanisms through chemical and genetic manipulations that are not feasible in the clinical setting. We optimise and validate image segmentation strategies and automate the process in a Nextflow-based pipeline (imcyto) that is scalable and portable, allowing for parallelised segmentation of large multi-image datasets With these methods we interrogate the remodelling of the TME induced by a KRAS G12C inhibitor in an immune competent mouse orthotopic lung cancer model, highlighting the infiltration and activation of antigen presenting cells and effector cells. We present a complete IMC workflow, including a validated 27-marker antibody panel, automated and optimised image segmentation using our imcyto pipeline and showcase various spatial analyses We applied these methods to study the effects of MRTX1257, a mutant-specific inhibitor of the KRAS G12C oncoprotein, on the TME of an immunotherapy refractory KRAS G12C mutant lung cancer. Using this methodology we gained detailed quantitative insight into the phenotypes and spatial relationships of immune cells and stromal compartments of the mouse lung TME, and how KRAS G12C inhibition promotes remodelling into a more immune activated state

Methods

Results

Conclusion