AUTHORS
Ariel Leyte-Vidal, BS, Muneeza Maqsood2, Dayna L Holroyd3, Vanie Seecharan4, Brian Farrell5, Ailin Alvarez6, Elias Lagonik III, PhD7, Nellie Ahmadieh8, Urmika Balaji9, Carlyn Leavitt10, Deborah White, PhD11, David Yeung, MD PhD12, Timothy P Hughes, MD, MBBS, FRACP, FRCPA13, Markus A Seeliger, PhD14, Susan Branford, PhD15, Neil P Shah, MD, PhD16
AFFILIATIONS
1 University of California San Francisco, San Francisco, California, United States, University of Miami, Miami, Florida, United States
2 University of Adelaide, Adelaide, Australia
3 University of Adelaide, Adelaide, South Australia, Australia
4 Stony Brook University, Stony Brook, New York, United States
5 Stony Brook University, Stony Brook, New York, United States
6 University of California San Francisco, San Francisco, California, United States
7 University of Adelaide, Adelaide, South Australia, Australia, University of Adelaide, Adelaide, South Australia, Australia
8 University of California San Francisco, San Francisco, California, United States
9 University of California San Francisco, San Francisco, California, United States
10 University of California San Francisco, San Francisco, California, United States
11 University of Adelaide, Adelaide, South Australia, Australia, University of Adelaide, Adelaide, South Australia, Australia
12 South Australian Health and Medical Research Institute, Adelaide, South Australia, Australia
13 South Australian Health and Medical Research Institute, Adelaide, South Australia, Australia
14 Stony Brook University, Stony Brook, New York, United States
15 University of Adelaide, Adelaide, Australia
16 University of California San Francisco, San Francisco, California, United States
ABSTRACT
Background
Kinase domain missense mutations that decrease affinity of orthosteric (ATP-competitive) tyrosine kinase inhibitors (TKI) relative to ATP represent the best documented mechanism of on-target resistance to TKIs used to treat a broad range of malignancies, including BCR::ABL1 TKIs for the treatment of chronic myeloid leukemia (CML) and ABL1/2-rearranged acute lymphoblastic leukemia (ALL). We and others recently demonstrated that BCR::ABL1 variants lacking ABL1 exon 2 display intrinsic resistance to the allosteric TKI asciminib. This exon encodes part of a regulatory domain required for asciminib to induce BCR::ABL1 kinase autoinhibition. This finding led us to hypothesize that deletion of select ABL1 exons might confer acquired resistance to asciminib and other allosteric TKIs.
Methods
RNA-based targeted next-generation sequencing, which detects abnormalities involving entire exons and intronic splice site variants, and/or Sanger sequencing was performed for 109 CML patients with resistance to TKI therapy, including 51 with asciminib resistance. Atypical splicing linked to ABL1 exon deletion was assessed using the SpliceChaser bioinformatic tool. Variant BCR::ABL1 isoforms were created in retroviral vectors that were used to transform Ba/F3 cells to growth factor independence and assessed for sensitivity to dasatinib, asciminib, and TGRX-678, a clinically active investigational allosteric TKI. Impacts of variants on asciminib target engagement were quantified by a NanoBRET assay. Splicing reporter minigenes containing splice site variants were created in pSPL3 and transiently transfected into HEK293T cells to evaluate for ABL1 exon skipping. AlphaFold3 and ICM-Pro were used to model the structural consequences of ABL1 variants within the kinase domain and assess effects on the myristoyl-binding pocket targeted by asciminib.
Results
Four CML patient samples with displacement of ABL1 exon 10 were identified (3 chronic phase; 1 lymphoid blast phase). Two patients had in-frame deletion of ABL1 exon 10 and two had ABL1 exon 9 duplication. These variants were exclusively detected at the time of acquired resistance to asciminib (4/51 patients, 7.8%). Exon 10 deletions were associated with intronic variants in the ABL1 exon 10 splice donor region. Strikingly, one case harbored five distinct splice donor variants. ABL1 exon 10 skipping of the BCR::ABL1 allele was confirmed using long-read nanopore sequencing. All detected ABL1 exon 10 splice donor variants caused efficient exon 10 skipping in splicing reporter minigene assays. In the context of p210 and p190 BCR::ABL1 isoforms, ABL1 exon 10 deletion and exon 9 duplication conferred complete resistance to asciminib and TGRX-678 (>1,000-fold increases in EC50) in vitro, while having no impact on sensitivity to the orthosteric TKI dasatinib. BCR::ABL1 p210 NanoBRET assays demonstrated that both variants substantially diminish asciminib target engagement. Asciminib is known to induce a bent conformation of the ABL1 αI’ helix in the kinase C-lobe that facilitates stabilization of the autoinhibited kinase conformation. Notably, this αI’ helix is encoded by the junction of ABL1 exons 9 and 10, which is disrupted by exon 10 deletion and exon 9 duplication. ABL1 exon 10 encodes two residues within this helix (V506 and E509) that confer asciminib resistance when mutated. Consistent with NanoBRET assays, AlphaFold3 modeling and ICM-Pro docking predicted that exon 10 deletion or exon 9 duplication alter the structure of the ABL1 myristoyl-binding pocket, suggesting profound impacts on asciminib binding affinity.
Conclusions
We present the first description of exon displacement as a novel mechanism of acquired on-target TKI resistance in any malignancy. ABL1 exon 10 deletion and exon 9 duplication variants confer high degrees of resistance to allosteric TKIs by impacting their targeted myristoyl-binding pocket, while retaining sensitivity to dasatinib. In contrast to orthosteric TKIs that target ATP-binding pockets, allosteric ABL1 TKIs bind regions that are dispensable for kinase activation and deletion of these regions does not substantially impair pathologic kinase activity. These findings have significant implications for methods employed worldwide to evaluate TKI resistance. Screening for ABL1 exon duplications and deletions in the kinase and regulatory domains in asciminib-resistant CML and ALL cases is warranted to reveal the full spectrum of resistant variants.
DISCLOSURES
Ariel Leyte-Vidal, BS: No relevant conflicts of interest to declare.