Back

Integrated in silico and in vitro approaches identify SNX.2112 as a drug vulnerability in t(7;12) AML stem-like cells

Ciciro, Y.; Ragusa, D.; Johnson, E. M.; Johns, A.; Kanannejad, S.; Gurung, R.; Oporto Espuelas, M.; Suen, C.-W.; Torregrosa-Cortes, G.; Giustacchini, A.; Tosi, S.; Pina, C.

2025-09-11 cancer biology
10.1101/2025.09.06.674620 bioRxiv
Show abstract

The t(7;12) translocation is a chromosomal rearrangement characteristic of infant Acute Myeloid Leukemia (AML) that results in ectopic overexpression of the homeobox gene MNX1. The resulting AML originates in utero and MNX1 transforms fetal, but not adult, hematopoietic progenitors. Using a 3-dimensional (3D) gastruloid model of blood development, we recently showed that t(7;12)-AML originates at the endothelial-to-hematopoietic transition, explaining its characteristic gene expression signature. Herein, we employ that signature to interrogate the transcriptional profiles of hundreds of human cell lines against the GDSC database of drug sensitivities and identify 12 candidate drugs with putative activity against t(7;12)-AML. We employ a cell line in which we engineered the t(7;12) translocation and systematically test the candidate drugs cell surface phenotype and clonogenic assays. Importantly, we identify SNX.2112 as a potential therapeutic agent against t(7;12)-AML. SNX.2112 selectively eliminates leukemia progenitors and colony-initiation in vitro, and selectively decreases MNX1 expression. The same effects are recapitulated in the 3D gastruloid model. Critically, the signature targeted by SNX.2112 uniquely maps to t(7;12) progenitors in single-cell RNA-seq analysis of leukaemia patients. Combinatorial treatment of t(7;12)-engineered cells with cytarabine or mitoxantrone reveals synergy with conventional chemotherapeutic agents, suggesting the potential of SNX.2112 as a targeted and cytotoxicity-sparing therapeutic approach. Overall, we successfully use an integrated computational and multi-model experimental approach to identify a drug vulnerability of t(7;12)-AML which can be explored for clinical potential.

Matching journals

The top 6 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.