2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

SESSION

622. Lymphomas: Translational – Non-Genetic: Poster III

Program: Poster Abstract Session
Monday, December 14, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Imaging, Omics Technologies, Bioinformatics, Profiling, Treatment Considerations, Drug Development, Biological Therapies, Bispecific Antibody Therapy, (CAR)-T Cell Therapies, Biological Processes, Pathogenesis, Genomics, Hematopoiesis, Immune Mechanism, Metabolism, Microbiome, Molecular Biology, Multi-Systemic Interactions, Diseases, Immunology, Immunodeficiency, Infectious Diseases, Viral, Lymphoid Malignancies, Lymphomas, Hodgkin Lymphoma, Non-Hodgkin Lymphoma, B Cell Lymphoma, T Cell Lymphoma, Indolent Lymphoma, Aggressive Lymphoma, Research, Fundamental Science, Translational Research, Clinical Research, Study Population, Human, Adult, Animal Model
PAPER

5286 Spatial immune profiling reveals divergent myeloid and T-cell organization in primary versus secondary central nervous system lymphoma

Amelie Gasper, Melissa Klingeberg2, Matthias Barone, PhD3, Jana Ihlow, Dr. med.4, Helena Radbruch, MD5, Daniel Steiert6, Arik Horne7, Julia Onken, MD8, Peter Vajkoczy, MD9, Björn Chapuy, MD10, Simon Haas, PhD11, David Capper, MD12, Ulrich Keller13, Birgit Sawitzki, PhD14, Fabian Coscia, PhD15, Stefan Habringer, MD16
1 Charité – Universitätsmedizin Berlin, Berlin, Germany
2 Humboldt University of Berlin, Berlin, Germany, Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Berlin, Germany
3 Berlin Institute of Health (BIH) & Charite´ – Universitätsmedizin Berlin, Berlin, Germany
4 Charité - Universitätsmedizin Berlin, Berlin, Germany
5 Charité - Universitätsmedizin Berlin, Berlin, Germany
6 Charité – Universitätsmedizin Berlin, Berlin, Germany
7 Berlin Institute of Health (BIH) & Charite´ – Universitätsmedizin Berlin, Berlin, Germany
8 Charité - Universitätsmedizin Berlin, Berlin, Germany
9 Charité - Universitätsmedizin Berlin, Berlin, Germany
10 Charité – Universitätsmedizin Berlin, Berlin, Germany
11 Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center For Molecular Medicine in the Helmholtz Association, Berlin, Germany, Berlin Institute of Health (BIH) & Charite´ – Universitätsmedizin Berlin, Berlin, Germany, Charité – Universitätsmedizin Berlin, Berlin, Germany
12 Charité - Universitätsmedizin Berlin, Berlin, Germany
13 Charité – Universitätsmedizin Berlin, Berlin, Germany
14 Berlin Institute of Health (BIH) & Charite´ – Universitätsmedizin Berlin, Berlin, Germany
15 Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Berlin, Germany
16 Charité – Universitätsmedizin Berlin, Berlin, Germany, German Cancer Consortium (DKTK), Partner Site Berlin, and German Cancer Research Center (DKFZ), Berlin, Germany

Introduction

Primary and secondary central nervous system (CNS) lymphoma (PCNSL and SCNSL) are clinically distinct diseases with different presentations and therapeutic challenges. Yet, the biological basis for these differences remains incompletely understood. In particular, it is unknown whether their CNS tumor microenvironments share a common immune organization or retain disease-specific features. We therefore compared the immune composition, cellular phenotypes, and spatial architecture of PCNSL and SCNSL CNS lesions.

Methods

Multiplex immunofluorescence on the PhenoCycler platform was performed on 12 PCNSL specimens and 27 SCNSL CNS specimens. The SCNSL cohort included synchronous and metachronous presentations with 15 paired systemic samples. The findings presented here are comparing the CNS lesions. Cell populations and functional states were defined using a custom 55-plex panel covering lineage, differentiation, cytotoxicity, checkpoint, hypoxia-response, and phagolysosomal markers. Patient-level analyses compared cellular composition, marker expression, cell-state proportions, distance to blood vessels, same-lineage clustering, and local cellular neighborhoods. False-discovery-rate-adjusted q values were reported where needed.

Results

PCNSL contained a higher B-cell fraction than SCNSL (p<0.001). Macrophages in PCNSL lesions showed higher HIF1A (p=0.002, q=0.010) and iNOS (p=0.034, q=0.045), consistent with a hypoxia-associated inflammatory phenotype. This difference in HIF1A expression persisted when the analysis was restricted to macrophages classified as M1-like by a prespecified phenotyping scheme (p=0.008). These differences were not explained by vascular proximity, as macrophage-to-vessel distance was not significantly different between diseases, and HIF1A expression did not correlate significantly with vessel distance. Instead, PCNSL macrophages were positioned in B-cell-rich neighborhoods, supporting an association between dense tumor architecture and metabolic stress. The PCNSL myeloid phenotype was accompanied by preservation of less differentiated T-cell states. PCNSL contained more naive/stem-like and progenitor-exhausted CD8 T cells, with higher TCF-1 across several CD8 subsets. Effector and proliferating CD8 T cells in PCNSL also expressed higher granzyme B. Spatially, CD8 T cells were closer to blood vessels compared to CD8 T cells in SCNSL (p=0.010), while CD4 T cells were more clustered (p=0.041), defining a perivascular immune architecture with retained stem-like and cytotoxic features. In contrast, SCNSL macrophages expressed higher Mac2/Galectin-3 (p=0.034, q=0.045) and CD68 (p<0.001, q=0.003), consistent with a Galectin-3/CD68-high phagolysosomal state. Several CD8 subsets in SCNSL had greater local enrichment of these macrophages and showed a shift toward more differentiated, immune checkpoint-associated phenotypes. In SCNSL, LAG3 expression was higher in naïve/stem-like, progenitor-exhausted, and proliferating CD8 T cells. Activated CD4 T cells expressed higher PD-1 (p=0.031). Spatially, progenitor-exhausted and effector/cytotoxic CD8 T cells had more neighboring macrophages assigned to the Mac2/Galectin-3-associated phagocytic state. The myeloid findings in metachronous SCNSL were consistent with those in the overall cohort, while the smaller treatment-naïve synchronous subgroup showed similar directional trends.

Conclusions

PCNSL and SCNSL develop distinct macrophage-associated T-cell ecosystems. PCNSL combines an HIF1A/iNOS-high, B-cell-associated myeloid phenotype of TCF-1-high and cytotoxically competent T cells. SCNSL is characterized by a Galectin-3/CD68-high phagolysosomal macrophage phenotype associated spatially with more differentiated and checkpoint-pressured T-cells. These findings suggest that CNS localization alone does not produce a uniform immune microenvironment across primary and secondary CNS lymphoma. These disease-specific immune architectures may differentially influence endogenous T-cell–redirecting therapies, however, prospective clinical and functional validation is required.

Stefan Habringer, MD: AbbVie - Honoraria; OnkowissenTV - Honoraria; Miltenyi Biomedicine - Honoraria; Jazz - Honoraria; Incyte - Honoraria; Pentixapharm - Honoraria; OnkowissenTV - Consultancy (Includes expert testimony); Moonlight AI GmbH - Current Employment