Abstract 1994: Defining tumor infiltrating myeloid microenvironment landscapes in HPV+ and HPV- head and neck cancers

Abstract Head and neck cancer (HNC) has a <20% response rate to immunotherapy, making it imperative to learn what causes high treatment resistance and how to overcome these challenges. One way to generate a holistic view of both human papillomavirus positive (HPV+) and negative (HPV-) HNC is to study the tumor immune microenvironment (TIME). Since HPV+ HNC often results in more favorable outcomes than HPV- HNC, assessing differences between these two disease states may provide critical mechanistic insight into what causes discrepancies in treatment response. This study uses publicly available and in-house collected single-cell transcriptome (scRNA-seq) data from preclinical murine models and human HNC to assess the differences between HPV+ and HPV- HNC TIME. Recent scRNA-Seq data showed extensive characterization of infiltrating T and B cells between HPV+ and HPV- HNC, but little is known about the myeloid compartment. As myeloid cells with both pro- and anti-tumoral functions make up a bulk of the HNC TIME, we sought to identify the differences in myeloid cells between HPV+ vs. HPV- HNC and their roles in response to immune checkpoint inhibitor (ICI) therapy. We identified higher myeloid heterogeneity than previously reported from both published and our generated scRNA-Seq. In human HNC patients, we identified four macrophage, seven monocyte, and five dendritic cell subsets. One macrophage subset expressed high levels of CXCL9 in both humans and our murine HNC model. Interestingly, recent studies identified CXCL9 as a biomarker for immunotherapy response. Observing conserved Cxcl9-expressing macrophages between murine and human HNC could help assess treatment response in cancer patients. We will assess the translational value of murine HPV+ and HPV- HNC models compared to their human counterparts. To this end, we are generating scRNA-Seq data from different HNC murine models: MOC1, MOC2, mEERL. The MOC2 and MOC1 cell lines recapitulate a responsive and non-responsive HPV- HNC tumor phenotype, respectively. The mEERL cell line recapitulates an HPV+ HNC phenotype. Our analysis shows high concordance between mouse models with immunological cold TIME (MOC2) and a proportion of human HNCs. We also identified changes in myeloid landscapes between immunological cold-to-hot TIME mice receiving ICI (anti-PD1 + anti-CTL4) therapies. We will present an in-depth comparative analysis to provide mechanistic insights into observed differences in the TIME between mice and humans. We will also present a novel bioinformatics analysis to interrogate how cell-cell signaling interactions between myeloid and T cell subsets influence their downstream gene regulatory network and tumor immune phenotypes. This analysis provides a reference for future studies in HNC immunology. Citation Format: Athena E. Golfinos, Wei Wang, Aisha Mergaert, Paul F. Lambert, Huy Q. Dinh. Defining tumor infiltrating myeloid microenvironment landscapes in HPV+ and HPV- head and neck cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1994.