Abstract 867: Multi-omic characterization of tumor microenvironment evolution in LBCL across treatment lines: Insights from Atlas classification and axicabtagene ciloleucel trials

Abstract Axicabtagene ciloleucel (axi-cel), an autologous anti-CD19 chimeric antigen receptor (CAR) T-cell therapy, was approved for patients with relapsed or refractory large B-cell lymphoma (LBCL) after two or more lines of systemic therapy (ZUMA-1, NCT02348216), as well as for those refractory to or relapsing within 12 months of first-line chemoimmunotherapy (ZUMA-7, NCT03391466). Recent molecular classifications of LBCL based on genetic factors shown to associate with response to R-CHOP and other chemo-immunotherapy based standard of care regimens have not been shown to associate with response to axi-cel in LBCL. Methods: To advance our understanding of the impact of tumor biology on response to axi-cel, we performed a detailed characterization of the tumor microenvironment (TME) using a multi-omic approach, integrating data from proteomics, transcriptomics and genomics offering a comprehensive view of biological systems, which we named Atlas. To construct the LBCL Atlas, we applied multiple technologies to 267 vendor-procured tumor samples from LBCL patients at diagnosis: (i) IHC panels, capturing spatial information at the protein and cellular level by counting and positioning cells of interest, (ii) Transcriptomics via RNA sequencing (RNA-seq) and Nanostring hybridization technology, (iii) Genomics, focusing on somatic mutation analysis through whole-exome sequencing (WES). Advanced mathematical and bioinformatics processing was employed to extract biological insights by combining both linear (multimodal factor analysis) and non-linear (self-organizing maps) methodologies. This effort resulted in a graphical representation of both complementary and overlapping signatures. Using samples with different prognostic outcomes and data from ZUMA-1 and ZUMA-7 axi-cel clinical trials, we further explored the evolution of the TME throughout disease progression and different lines of treatment. We focused on both adaptive immunity (lymphocytes, NK cells, etc.) and innate immunity (macrophages, MDSCs), examining the TME clusters defined by Atlas to characterize distinct tumor microenvironments. This Atlas provided an alternative to the cell-of-origin classification and other genetically based classifications of LBCL, based on TME. We will present evolutionary changes though lines of therapy to less favorable tumor microenvironments and, through the projection of clinical trial samples, demonstrate that the Atlas is a powerful tool for exploring mechanisms behind treatment response and disease progression. This includes the association of immune cell populations, such as T-cell and macrophage phenotypes and overall immune cell contexture within the TME, and other biomarkers with disease progression and treatment response. Citation Format: Michael D. Mattie, Regis Perbost, Aurelie Auguste, Alexia Papadopoulos, Jenny J. Kim, Ioana Kloos, Gayatri Tiwari, Justin Budka, Darawan Rinchai, Simone Filosto, Rhine Shen, Davide Bedognetti, Jérôme Galon. Multi-omic characterization of tumor microenvironment evolution in LBCL across treatment lines: Insights from Atlas classification and axicabtagene ciloleucel trials [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 867.