Loading…

Abstract 3863: CODEX high-multiplex imaging reveals distinct tumor microenvironment in mouse melanoma models associated with response to immunotherapy

In order to identify the determinants of melanoma resistance to immunotherapies and predictive biomarkers, we developed a series of immunocompetent syngeneic mouse models that represent diverse subtypes of human melanoma exhibiting a range of sensitivity to immune checkpoint blockade (ICB) therapies...

Full description

Saved in:
Bibliographic Details
Published in:Cancer research (Chicago, Ill.) Ill.), 2020-08, Vol.80 (16_Supplement), p.3863-3863
Main Authors: Kedei, Noemi, Pérez-Guijarro, Eva E., Chen, Jin-Qiu, Day, Chi-Ping, Malik, Mariam Q., Goldstein, David J., Merlino, Glenn T.
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In order to identify the determinants of melanoma resistance to immunotherapies and predictive biomarkers, we developed a series of immunocompetent syngeneic mouse models that represent diverse subtypes of human melanoma exhibiting a range of sensitivity to immune checkpoint blockade (ICB) therapies (Pérez-Guijarro et al, BioRxiv 2019). Comparative genomic and immunological analysis identified that melanoma plasticity as well as T cell dysfunction and exclusion programs strongly correlated with ICB resistance. However, how interactions between tumor and immune cells influence therapeutic efficacy is still unknown. To address this question, we used CODEX high-multiplex imaging technology that enables quantitative detection of 20+ markers at single cell level resolution preserving spatial context. CODEX does not require cell isolation that may result in variable cell loss. Furthermore, the whole tissue imaging provides insights into tumor/stroma and cellular heterogeneity. Fresh frozen tissues (n=3 for each model) were stained with a panel of 16 antibodies to identify major immune cell phenotypes including cytotoxic and helper T cells, B cells, and subsets of myeloid cells. Quantitative single cell level analysis of images was performed with HALO (Indica Labs) and multiplex analysis viewer (MAV) (Akoya Biosciences) software. Preliminary analysis identified unique tumor architecture, immune cell densities and distribution in each model. Although we found high diversity in the vasculature (CD31+), proliferation (Ki67+) and number of infiltrating leukocytes (CD45+) between the models, these were not associated with resistance. In contrast, high infiltration of cytotoxic T cells and dendritic cells was associated with response to ICBs. These results validated tumor analyses by FACS and gene signatures of immune cells. In addition, CODEX imaging revealed heterogeneity and complex spatial organization in the tumor microenvironment. Importantly, ICB-resistant tumors exhibited compact tumor structure with less stroma infiltration and lack of melanin expression, while ICB-sensitive tumors had complex, nodular and pigmented structures, indicating different differentiation status. Our data also suggested that myeloid phenotypes and functional compartment interactions may contribute to the response to ICBs. The study is expected to provide a higher resolution profiling of tumor-immune cell interactions and facilitate mechanistic understanding of resistance to immune che
ISSN:0008-5472
1538-7445
DOI:10.1158/1538-7445.AM2020-3863