Abstract 1177: Tumor-associated macrophages and clinical outcomes to immunotherapy

Abstract Tumor-associated macrophages (TAMs) have often been classified as either inflammatory M1 or immunosuppressive M2 macrophages. This simplified model had predominantly been informed by in vitro polarization assays and in situ analyses relying on a limited set of cellular markers. The advent of single-cell transcriptomics enabled a more holistic and unbiased investigation of TAMs. We investigated cell subpopulations that are associated with beneficial outcomes in NSCLC patients (n = 219) receiving a weekly dose of 10 mg/kg anti-PD-L1 within the NCT02576574 trial. To leverage the power of single cell technologies, we used a bioinformatical method called Scissor, which identifies distinct cell subsets from single cell data integrated with bulk expression and clinical outcome data. Indication-matched published single cell expression data from three independent studies were used to this end. In all single cell datasets tested, we identified a distinct subset of TAMs that was associated with improved progression-free survival in patients treated with anti-PD-L1. Interestingly, this cell population was a subset of a larger cell cluster that had been previously classified as M2 macrophages due to their expression of CD163. We confirmed our findings using an independent published dataset consisting of 135 NSCLC patients treated with immune checkpoint inhibitors and bulk RNA-seq data with similar results. When further characterizing the TAM subset within the single cell expression data, we found these cells expressed TREM2, C1Qs, FOLR2 and APOE consistent with lipid-associated macrophages. This subset was distinct from TAM subpopulations expressing marker genes such as SPP1 and VCAN (angiogenic TAMs). Notably, these TAMs produced chemokines CXCL9, a gene that had been strongly associated with clinical response to immunotherapy, as well as CCL18, which was shown to promote cancer invasiveness. To investigate the spatial distribution of CD163+ TAMs in general we turned to the IMMUcan dataset. IMMUcan is a public-private EU funded consortium profiling the human tumor microenvironment (TME) in 2, 500+ patients and currently hosting one of the largest cancer-related imaging mass cytometry (IMC) datasets to date. Within the advanced NSCLC cohort we found that CD163+ TAMs were frequently in direct contact with tumor cells. Finally, patients with high CD163+ TAMs in the TME were distinct from patients with high CD8+ T cell infiltration. These data suggest that macrophage subpopulations that were previously considered to have an immunosuppressive role can be associated with clinical benefit to immunotherapy. Targeting these TAMs as part of immune checkpoint inhibitor blockade may be worthwhile given their ambiguous role. Citation Format: Henoch S. Hong, Marzena Mura, Jing Yang, Aslihan Gerhold-Ay, Thomas Mrowiec, Daniel Schulz, Bernd Bodenmiller, Robin Liechti, Marie Morfouace, Parantu Shah, Alex Rolfe. Tumor-associated macrophages and clinical outcomes to immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1177.