Chronic myeloid leukemia (CML) is a success story in cancer management. The development of tyrosine kinase inhibitors (TKIs) as a targeted therapy inhibiting BCR-ABL1 onco-kinase, the hallmark of CML, was as a “game changer” in CML management which has allowed most patients to achieve a durable remission and almost normal life expectancy. However, a minority of CML patients progress from the “controllable” chronic phase (CP) to the aggressive blast phase (BP) with a dismal outcome. BP-CML remains as the main clinical challenge in CML management, despite the development of more potent 2nd and 3rd generation TKIs. BP-CML has been linked to the acquisition of somatic mutations in selected leukemia-associated genes, of which RUNX1 mutations being among the most common. Nevertheless, the mechanisms underlying BP-CML pathogenesis remain largely elusive.
Characterization of more defined molecular subtypes would enable tailoring better treatment modalities for BP-CML patients. Given the low incidence rates of BP-CML in the TKI era, forming a BP-CML patient cohort with predictive power is troublesome. To overcome this challenge, we have collaborated with Susan Branford’ lab from University of South Australia, to expand our cohort and include data from additional BP-CML patients. In total, we performed comprehensive genetic and drug sensitivity profiling of 20 RUNX1mut and RUNX1wt BP-CML patients. In contrast to RUNX1wt BP-CML patients, the mutational landscape of RUNX1mut BP-CML patients revealed enrichment of PHF6 and BCORL1 mutations, as well as IKZF1 deletions comparable to RUNX1mut AML patients. Furthermore, we demonstrated a specific role of AID/RAG axis in mediating structural rearrangements in RUNX1mut BP-CML patients. On the transcriptional level, RUNX1mut BP-CML patients showed upregulation of stem cell and B-lymphoid specific transcription factors, associating with distinct phenotype of aberrant lymphoid marker expression (ex: CD19 and CD7), in comparison to RUNX1wt BP-CML patients. Similar to RUNX1mut AML, RUNX1mut BP-CML showed deregulation of interferon, TNF and coagulation pathways confirming RUNX1mut specific transcriptional signature. These findings were confirmed in CML cell lines using CRISPR/Cas9 gene editing technology.
The personalized approach that integrates genetic data and drug testing, represents a powerful strategy for identification of promising therapeutic options for leukemia. We performed high throughput drug sensitivity and resistance testing (DSRT) for a cohort of eight BP-CML patients using a large oncology compounds library involving 255 of both approved and emerging drugs. Blasts from RUNX1mut patients were more responsive for mTOR-, BCL2-, and VEGFR-inhibitors and glucocorticoids compared to blast from RUNX1wt patients. Interestingly, combination of the selected drugs with imatinib showed potential synergistic effect inhibiting RUNX1mut leukemia cells. DSRT findings from RUNX1mut patients were also validated in homozygous RUNX1-/- and heterozygous RUNX1-/mut cells lines. Finally, we investigated the potential targeting of aberrant CD19 expression in RUNX1mut BP-CML by CAR T-cells immunotherapy. Notably, in ex-vivo assays RUNX1mut blasts exhibited sensitivity to CD19-CAR T-cells, both alone and in combination with imatinib.
Our study highlights the power of personalized approach as a potentially promising strategy for management of BP-CML patients. The study represents comprehensive characterization of RUNX1mut subset of BP-CML patients and provides insights into the role of RUNX1 mutations in CML progression by induced transcriptional reprogramming and aberrant mutagenic AID/RAG activity. Integration of genomic and drug sensitivity data highlighted novel therapeutic options involving targeted therapies and CD19-CAR T-cell immunotherapy approach, which open new avenues to improve management of BP-CML patients. Given the high prevalence of RUNX1 mutations in hematological malignancies, our findings may have importance also in other RUNX1mut leukemias.