Dasatinib Overcomes AML Resistance to BCL2 Inhibition By Indirectly Inhibiting MCL1: From Ex Vivo drug Sensitivity/Resistance Profiling to a Phase II Clinical Trial: A Hemato-BIO Study

Introduction. BH3 mimetics such as the BCL2 specific inhibitor (-inh) venetoclax (VEN) and the BCL2/BLCxL-inh navitoclax (NAV), are small molecules inducing intrinsic apoptotic cell death by inhibiting antiapoptotic proteins. VEN has changed the paradigm of treatment of acute myeloid leukemia (AML) unfit for intensive chemotherapy based on the VIALE-A study. Nevertheless, approximately 1/3 of AML is primary refractory to VEN (VEN-R), with no current treatment option. Monocytic differentiation, signaling mutations and MCL1 antiapoptotic dependency are among the most frequent and interconnected resistance mechanisms.

Method. To find new therapeutics active in resistant AML, we collected clinical data and bone marrow or peripheral blasts at AML diagnosis in a prospective biobanking clinical trial named HEMATO-BIO-IPC 2013-015 (NCT02320656, PI: Prof. N Vey), between 2014 and 2019. We performed targeted DNA sequencing using a NGS panel of recurrently mutated AML genes and ex vivo drug sensitivity/resistance profiling (DSRP).

Results. Taking the first 108 AML analyzed samples, we identified that 17 (15.7%) were resistant to NAV (NAV-R), taking a Z score >0.75 as a threshold. The patient median age was 62.5 yo, median leukocyte count, platelet count, percentage of blasts were 29.7 G/L (ranges, 1.1-86.3), 67 G/L (ranges, 18-163) and 72% (ranges, 21-93), respectively. Twelve AML were classified as FAB4 (n=6) and FAB5 (n=6). Cytogenetics was normal in 9 cases. NPM1 mutation was found in 6 and FLT3-ITD in 4 cases. Signaling mutations, including NRAS, KRAS and PTPN11 mutations, were found in 3, 2 and 2 cases, respectively. No samples were found with KIT mutations and only 2 samples had a TP53 mutation. We noticed a strong anticorrelation between NAV-R and sensitivity to most of the tested kinase inhibitors, including the Pi3K-inh BKM120 (Buparlisib) and Idelalisib, the JAK2-inh Ruxolitinib, the MEK-inh Trametinib, the mTOR-inh Temsirolimus, the FLT3-inh Quizartinib, and the BCR-ABL-inh Imatinib and Dasatinib (DASA). On the other hand, we did not observe any anticorrelation with the EGFR-inh erlotinib nor gefitinib. Amongst the most active kinase inhibitors, DASA had the lowest median IC50 (Z-score=0,01511). As DASA has already been used in clinical settings in AML, we chose DASA for further tests. We first confirmed the anticorrelation between DASA and the BCL2/BCLxL-inh ABT797 in the BEAT-AML cohort. As NAV and ABT797 are both known BCL2/BCLxL-inh, we hypothesized that blasts sensitive do DASA (DASA-S) were in fact dependent on MCL1. Indeed, western blot (WB) analysis showed higher MCL1 and proapoptotic BIM protein levels. To link a possible dependence of DASA-S blasts to MCL1, we performed BH3 profiling in 25 AML samples. We confirmed that DASA-S samples had a strong MCL1 dependency as shown by a dose-dependent mitochondrial membrane depolarization using MS1 peptide and a high MS1/BAD ratio. We next hypothesized that DASA was targeting MCL1 through an indirect manner. We performed a WB of MCL1 using AML cell line K562 and HL60, and observed a dose dependent decrease in MCL1 protein levels upon DASA treatment while BCL2 expression was not altered. Collectively, these results suggest that DASA could be a potential MCL1 indirect inhibitor that could be used in the clinical settings to treat VEN-R AML. To prove this hypothesis, we designed a phase II clinical trial named VEN-R DASA-IPC 2022 067 (EUCT 2023-505846-24-00) that is currently enrolling patients failing a minimum of two VEN-AZA cycles.

Conclusion. DASA may be efficient in targeting VEN-R by inhibiting MCL1. Preclinical studies based on the AML collections HEMATIO-BIO allowed us to ask clinical questions that can be addressed in early-phase clinical trial.