-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

200 The Impact of Lineage on CNS Infiltration Dynamics in Infant Leukaemia

Program: Oral and Poster Abstracts
Type: Oral
Session: 603. Lymphoid Oncogenesis: Basic: Molecular Insights into Acute Lymphoblastic Leukemias
Hematology Disease Topics & Pathways:
Research, Fundamental Science, Lymphoid Leukemias, ALL, Acute Myeloid Malignancies, Translational Research, Diseases, Immunology, Lymphoid Malignancies, Metabolism, Biological Processes, Pathogenesis
Saturday, December 7, 2024: 2:15 PM

Alasdair Duguid, MBBChir1*, Camille Malouf, PhD2*, Christina Halsey, MD, PhD3 and Katrin Ottersbach, PhD2*

1Institute for Regeneration and Repair, University of Edinburgh, Edinburgh, SCO, United Kingdom
2Institute for Regeneration and Repair, University of Edinburgh, Edinburgh, United Kingdom
3Wolfson Wohl Cancer Research Centre, School of Cancer Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, United Kingdom

Infant leukaemia is a rare, aggressive form of paediatric acute leukaemia commonly associated with KMT2A gene rearrangements. There are several notable features of KMT2A-rearranged infant leukaemia, including lineage plasticity and high rates of central nervous system (CNS) involvement. In a previous study, we identified two microRNAs, miR-128a and miR-130b, upregulated in KMT2A-AFF1+ infant leukaemia patient samples. From this, we developed two immunocompetent mouse models of infant leukaemia by overexpressing these miRs individually in KMT2A-AFF1+ mouse fetal liver haematopoietic stem and progenitor cells, resulting in a pro-B acute lymphoblastic leukaemia (B-ALL) phenotype and mixed lineage leukaemia phenotype (Malouf et al. Blood 2021). Our mixed lineage leukaemia model generated myeloid, B-lymphoid and mixed myeloid-B-lymphoid output. Through transcriptomic comparisons, we validated our mixed lineage infant leukaemia model against KMT2A-AFF1 mixed lineage leukaemia patient samples (Tirtakusuma et al. Blood 2022). This comparison demonstrated that the leukaemia propagating cells (LPC) from our model were indeed mixed lineage based on gene expression and did not demonstrate myeloid or B-lymphoid bias. Using these models, we focused on understanding the biology of CNS involvement of KMT2A-AFF1+ infant leukaemia, highlighting shared and lineage-specific functional and transcriptomic differences.

The functional properties of CNS-derived lineage-negative LPCs were explored in both miR overexpressing KMT2A-AFF1+ infant leukaemia models. Phenotypic analysis of the leukaemia blasts in both models showed that all populations of leukaemia present in the bone marrow (BM) were present in the CNS niche, suggesting a lack of niche-specific entry bias. Transplantation of CNS-derived LPCs demonstrated lineage-specific differences. In contrast to the pro-B ALL model, where all CNS-derived LPCs engrafted in the BM niche and resulted in terminal leukaemia, in the mixed lineage leukaemia model, engraftment of CNS-derived LPCs was significantly impaired with few transplant recipients developing terminal leukaemia. We speculate that CNS-derived KMT2A-AFF1+ mixed lineage infant leukaemia LPCs undergo irreversible adaptation to the CNS niche that may limit their ability to result in systemic relapse, unlike the more plastic adaptations that occur to the CNS niche in pro-B ALL LPCs.

We then performed transcriptomic comparisons between BM and CNS-derived LPCs from our infant leukaemia models. These comparisons have particular novelty and insight into CNS-leukaemia biology in mixed lineage leukaemia as no comparable human or murine datasets exist to our knowledge. Although there was evidence of niche-dependent regulation of some shared biological processes between the two models, there were very few common differentially expressed genes. The small number of shared differentially expressed genes included potentially targetable pathways such as the upregulation of growth factor receptors Met and Pdgfrb and the transcriptional repressor Bcl6 by CNS-derived LPCs.

In both models, CNS-derived LPCs demonstrated altered lipid and cholesterol metabolism, although through the regulation of different pathways. These metabolic adaptations in the pro-B ALL model shared many features with published data from B-ALL xenograft models (Savino et al. Nat Cancer 2020; Cousins et al. Leukemia 2022). Unique to this immunocompetent model system was the differential expression of genes implicated in leukaemia-immune cell interactions and suppression of T-cell activity. This was also lineage-specific. CNS-derived LPCs from the mixed lineage leukaemia model upregulated the immune checkpoints Btla and Cd83, whereas CNS-derived LPCs from the pro-B ALL model upregulated the immune checkpoints Ctla4 and Cd96.

Differentially regulated processes by niche unique to the mixed lineage leukaemia model included hypoxia response, IL2 STAT signalling and inflammatory signalling through TNF via NFkB. With the inherent plasticity of KMT2A-AFF1+ infant leukaemia, these lineage-specific adaptations to the CNS niche may provide a route for leukaemia cell survival and should be considered in the development of novel CNS leukaemia-specific therapies in this context.

Disclosures: Halsey: Autolus: Membership on an entity's Board of Directors or advisory committees; Jazz Pharmaceuticals: Consultancy.

*signifies non-member of ASH