Abstract IA002: Hodgkin lymphoma tumor microenvironment

Abstract

This presentation focuses on the comprehensive, multidimensional characterization of the tumor microenvironment (TME) in classic Hodgkin lymphoma (cHL), integrating malignant cell genotyping, single-cell transcriptomics, and spatial imaging to refine disease taxonomy and identify biologically and clinically relevant subtypes. With recent broadening of immunotherapy options for cHL patients, most recent work addresses the need for a more precise molecular classification of the disease by dissecting the complex interplay between malignant Hodgkin and Reed–Sternberg (HRS) cells and their surrounding immune ecosystem. Previous single-cell studies by multiple groups revealed distinct immune cell populations within the TME, including specialized subsets of T cells, macrophages, stromal elements and B cells. Notably, cHL is enriched for unique regulatory T-cell populations expressing immune checkpoint molecules such as LAG3 and CTLA4, confirming the importance of immune evasion mechanisms. Spatial characterization using imaging mass cytometry (IMC) demonstrated that the organization of these immune subsets is closely linked to tumor cell features, including MHC class II expression, highlighting spatially dependent immune interactions. An integrative analysis of 114 cHL tissues identified four molecular subclasses (CST, CN913, STB and CN2P) with distinct clinical, genetic and microenvironmental characteristics, revealed through spatial transcriptomics and IMC. A molecular classifier (HLGen) was shown to robustly call these classes from tissue samples enriched for HRS cells and ctDNA-derived sequencing data. Moreover, correlative analysis in primary tissues and functional in vitro studies demonstrated that CSF2RB mutations, that are characteristic of the CST subclass, hyperactivated JAK2-STAT5 signaling, was associated with TARC (encoded by CCL17) and IL13 expression, and enhanced recruitment of CCR4+ regulatory T cells. These findings support a model in which tumor cell genetics are linked to, and actively shape, the immunosuppressive TME. Towards clinical assay development, suitable for decision making at the time point of cHL relapse, a spatial model (RHL4S), incorporating both malignant cell and microenvironmental features, demonstrated strong prognostic value for post-transplantation outcomes. Additionally, our work on single cell transcriptomic profiling of relapse biopsies highlights an important role for non-malignant B cells and a pathogenesis model in which early relapse cases showed enrichment of naïve B cells in close proximity to HRS cells and exhausted T-cell populations. Overall, the presented body of work underscores the importance of multimodal study designs and establishes a framework in which tumor genetics, immune cell composition, and spatial organization are linked, with direct implications for prognostication and the development of subtype-specific therapeutic strategies.