Discovery of Mcl-1-specific inhibitor AZD5991 and preclinical activity in multiple myeloma and acute myeloid leukemia

Abstract

Mcl-1 is a member of the Bcl-2 family of proteins that promotes cell survival by preventing induction of apoptosis in many cancers. High expression of Mcl-1 causes tumorigenesis and resistance to anticancer therapies highlighting the potential of Mcl-1 inhibitors as anticancer drugs. Here, we describe AZD5991, a rationally designed macrocyclic molecule with high selectivity and affinity for Mcl-1 currently in clinical development. Our studies demonstrate that AZD5991 binds directly to Mcl-1 and induces rapid apoptosis in cancer cells, most notably myeloma and acute myeloid leukemia, by activating the Bak-dependent mitochondrial apoptotic pathway. AZD5991 shows potent antitumor activity in vivo with complete tumor regression in several models of multiple myeloma and acute myeloid leukemia after a single tolerated dose as monotherapy or in combination with bortezomib or venetoclax. Based on these promising data, a Phase I clinical trial has been launched for evaluation of AZD5991 in patients with hematological malignancies (NCT03218683).

Conflict of interest statement

A.E.T., M.A.B., A.A., B.M.A., E.C., J.C., K.J.E., E.G., F.D.G., D.H., J.A.H., J.W.J, S.L.K., J.G.K., M.L.L., M.J.P., B.P., P.B.R., D.W.R., A.G.S., N.S., W.Y., Q.Y., X.Z., J.P.S., E.A.C., D.M.W., S.E.F., and A.W.H. are or were employees of AstraZeneca at the time of conducting these studies. A.K.N. is a consultant/advisory board member for Amgen, GSK, BMS, Janssen, Celgene, Takeda, Adaptive and Spectrum Pharmaceuticals. L.H.B. received honoraria from AstraZeneca and is a consultant for Abbvie and AstraZeneca. R.W.J. receives research support from Abbvie, AstraZeneca, BMS, Roche and MecRX. G.P.G. receives honoraria from Roche and Gilead. S.M.M. declares no competing interests.

Figures

Fig. 1

Structure-based design of macrocyclic inhibitors of Mcl-1. a Crystal structure of 1 in complex with Mcl-1. Note 2:1 ligand:protein stoichiometry in binding site (pdb id: 6FS2). The protein surface around the highlighted residues has been hidden for clarity. b Literature compounds 1 and 2, dimeric target 3 and compound synthesized from byproduct, 4. c Crystal structure of 4 in complex with Mcl-1 (pdb id: 6FS1). The protein surface and alpha helix ribbon at the front of the pocket has been hidden for clarity

Fig. 2

Chemical and crystal structure of (Ra)-7. a Crystal structure of (Ra)-7 in complex with Mcl-1. Note coplanar alignment of Arg 263 and the carboxylic acid and the interaction between 17-Cl of (Ra)-7 and the backbone carbonyl of Ala227 (pdb id: 6FS0). The protein surface around the highlighted residues has been hidden for clarity. b Structure of (Ra)-7 and numbering of atoms. c1H NMR of (Ra)-7, highlighting the upfield chemical shift of H38 indicative of a conformation that is rigid and consistent with the active binding conformation depicted in a

Fig. 3

(Ra)-7 induces on-target intrinsic apoptosis. a Apoptosis induction in Eμ-Myc lymphoma cells stably expressing human Mcl-1, Bcl-2, Bcl-xL, Bfl-1/A1, or Bcl-w and treated with (Ra)-7 for 24 h. Data are shown as mean ± SD (n = 3). b Isothermal dose–response curve of Mcl-1 at 48 °C plotted against varying concentrations of (Ra)-7. I/I40 μM: ratio of the signal intensity for each particular sample to the signal intensity obtained with (Ra)-7 at 40 μM. Data are shown as mean ± SD (n = 3). c Caspase-3/7 activity in NCI-H23 cells treated with siRNA targeting Bak mRNA for 72 h before treatment with (Ra)-7 for 6 h. Representative data of two independent experiments are shown

Fig. 4

Kinetics of apoptosis and cell death induction by (Ra)-7. a Whole cell extract (bottom) and IP (top) of Mcl-1 from lysates of MOLP-8 cells treated with (Ra)-7 at indicated concentrations followed by immunoblot analysis. b Summary of the kinetic of (Ra)-7 effect on Mcl-1:Bak complex disruption, MOMP, cellular loss of ATP, caspase-3/7 activity, phosphatidyl-serine externalization and cell membrane permeability in MOLP-8 cells upon treatment with (Ra)-7 at 500 nM. Representative data of two independent experiments are shown

Fig. 5

Hematological cell lines are preferentially sensitive to AZD5991. a Viability (n = 142) and b caspase-3/7 induction (n = 154) evaluated in cancer-derived cell lines treated with AZD5991 for 24 or 6 h, respectively. Data are shown as median with 95% confidence interval

Fig. 6

AZD5991 exhibits potent anti-tumor efficacy in MM models. a Subcutaneous tumor growth in the MOLP-8 tumor model treated with a single i.v. dose of AZD5991 10–100 mg kg−1. Tumor volumes are presented as mean ± SEM, seven mice were evaluated per group. b MOLP-8 tumor lysates prepared from mice dosed with a single i.v. dose of AZD5991 at 10, 30, or 100 mg kg−1 were evaluated for expression of cleaved caspase-3 by MesoScale discovery assay. Mean ± SD are shown. c In vivo plasma concentration of AZD5991 following a single i.v. dose of AZD5991 10–100 mg kg−1 to mice with subcutaneous MOLP-8 tumors. AZD5991 plasma concentrations were assessed through the first 24 h following compound administration. Concentrations of AZD5991 (ng  mL-1) are plotted on a log10 scale as mean ± SD (n = 3). d Apoptosis in mononuclear cells isolated from bone marrow aspirate of MM patients (n = 48) treated with increasing concentrations of AZD5991 for 24 h and evaluated by Annexin V by flow. Each dot represents a unique patient sample. center line indicates the median, bounds of box denote 25% (lower) and 75% (upper) percentile, and whiskers encompass 5–95 percentile. e NCI-H929 cells were treated with bortezomib at the indicated concentrations, whole-cell lysates prepared after 2 or 4 h of treatment and protein expression evaluated by immunoblotting. f Subcutaneous tumor growth in the NCI-H929 tumor model treated with AZD5991 in combination with bortezomib. Both drugs were dosed intravenously. Arrows indicate day of dosing for AZD5991 (blue) and bortezomib (green). Tumor volumes are presented as mean ± SEM, six mice were evaluated per group

Fig. 7

AZD5991 causes tumor regression in AML models. a Subcutaneous tumor growth in the MV4-11 tumor model treated with a single i.v. dose of AZD5991 at 10, 30, or 100 mg kg−1. Values are presented at mean ± SEM, six mice were tested per group. b MV4-11 tumor lysates prepared from mice dosed with a single i.v. dose of AZD5991 at 30 mg kg−1 (n = 4), 100 mg kg−1 (n = 4), or vehicle control (n = 2) were evaluated for expression of caspase 3 and cleaved PARP by western blotting. Leukemic cells (HLA-ABC+ hCD45+) were assessed by flow cytometry in peripheral blood (c) or bone marrow (d) obtained from mice engrafted with MOLM-13 leukemia cells and treated with vehicle (n = 6), venetoclax at 100 mg kg−1 per oral daily (n = 3) or AZD5991 at 100 mg kg−1 i.v. once weekly (n = 3). Analysis were performed on day 10 after treatment initiation. A non-parametric, unpaired, two-tailed t-test was used to calculate significance. e Heatmap representing EC50 values for caspase activation at 6 h in 11 AML cell lines after treatment with AZD5991 or venetoclax monotherapy or combination. EC50 values represented for combination were determined with venetoclax at 160 nM and variable concentrations of AZD5991. f OCI-AML3 whole-cell lysates prepared before or after 3, 6, or 24 h of treatment with venetoclax, AZD5991, or combination were evaluated for expression of indicated proteins by immunoblotting. g Subcutaneous tumor growth in the OCI-AML3 tumor model treated with AZD5991 (i.v.) in combination with venetoclax (oral) or corresponding single agents. Tumor volumes are presented as mean ± SEM, 10 mice were evaluated per group