Back

ASXL1 truncating mutations drive leukemic resistance to T cell attack

McCurry, D.; Bachireddy, P.

2025-06-01 cancer biology
10.1101/2025.05.29.656798 bioRxiv
Show abstract

We previously found that specific exhausted T cell subsets defined response, but not resistance, to donor lymphocyte infusions (DLI), a curative immunotherapy for leukemic relapse following allogeneic stem cell transplant (SCT). To identify leukemia molecular pathways that drive resistance, we analyzed whole exome and targeted mutation panel sequencing in two independent cohorts of DLI-treated patients, nominating oncogenic, truncating mutations in ASXL1 (ASXL1MUT) as the genetic basis for DLI resistance. Deep interrogation of 138,152 bone marrow single myeloid cell transcriptomes (scRNA-seq) from this cohort linked DLI resistance to a transcriptional state notable for leukemic stem cell identity and HLA-I suppression. In silico analysis of publicly available scRNA- and ATAC-seq data in acute myeloid and chronic myelomonocytic leukemias, respectively, confirmed an association between ASXL1MUT and HLA-I suppression across myeloid malignancies. CRISPR correction of the endogenous ASXL1MUT in the K562 leukemic cell line increased HLA-I, but not HLA-II, surface protein expression through increased deposition of the activating H3K4Me3 mark with only modest effect on the repressive H3K27Me3 mark, suggesting a Polycomb-independent mechanism of action. Indeed, inhibitors of EZH2, a critical component of the PRC2 complex, significantly upregulated HLA-I surface protein expression independently of ASXL1MUT, suggesting that EZH2 inhibition could bypass ASXL1MUT-mediated HLA-1 suppression. Importantly, ASXL1CORRECTION significantly increased CD8+ T cell recognition, activation and killing, and ASXL1MUT-mediated T cell suppression could be overcome by EZH2 inhibition. Thus, by integrating molecular analyses with immuno-functional studies, we define a novel oncogene-driven pathway of immune evasion and propose a therapeutic strategy to re-engage T cell killing in ASXL1MUT tumors.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.