Preclinical efficacy of the potent, selective menin-KMT2A inhibitor JNJ-75276617 (bleximenib) in KMT2A- and NPM1-altered leukemias

Abstract

The interaction between menin and histone-lysine N-methyltransferase 2A (KMT2A) is a critical dependency for KMT2A- or nucleophosmin 1 (NPM1)-altered leukemias and an emerging opportunity for therapeutic development. JNJ-75276617 (bleximenib) is a novel, orally bioavailable, potent, and selective protein-protein interaction inhibitor of the binding between menin and KMT2A. In KMT2A-rearranged (KMT2A-r) and NPM1-mutant (NPM1c) acute myeloid leukemia (AML) cells, JNJ-75276617 inhibited the association of the menin-KMT2A complex with chromatin at target gene promoters, resulting in reduced expression of several menin-KMT2A target genes, including MEIS1 and FLT3. JNJ-75276617 displayed potent antiproliferative activity across several AML and acute lymphoblastic leukemia (ALL) cell lines and patient samples harboring KMT2A or NPM1 alterations in vitro. In xenograft models of AML and ALL, JNJ-75276617 reduced leukemic burden and provided a significant dose-dependent survival benefit accompanied by expression changes of menin-KMT2A target genes. JNJ-75276617 demonstrated synergistic effects with gilteritinib in vitro in AML cells harboring KMT2A-r. JNJ-75276617 further exhibited synergistic effects with venetoclax and azacitidine in AML cells bearing KMT2A-r in vitro, and significantly increased survival in mice. Interestingly, JNJ-75276617 showed potent antiproliferative activity in cell lines engineered with recently discovered mutations (MEN1M327I or MEN1T349M) that developed in patients refractory to the menin-KMT2A inhibitor revumenib. A cocrystal structure of menin in complex with JNJ-75276617 indicates a unique binding mode distinct from other menin-KMT2A inhibitors, including revumenib. JNJ-75276617 is being clinically investigated for acute leukemias harboring KMT2A or NPM1 alterations, as a monotherapy for relapsed/refractory acute leukemia (NCT04811560), or in combination with AML-directed therapies (NCT05453903).

Graphical abstract

Figure 1.

Chemical and crystal structures of JNJ-75276617. (A) Chemical structure of JNJ-75276617. (B) Crystal structure of menin bound to JNJ-75276617 (Protein Data Bank Identifier [PDB ID]: in deposition). JNJ-75276617 is shown as ball and sticks, and selected side chains of menin are shown as sticks. Dotted lines represent key hydrogen-bonding interactions between the protein and the ligand, and ligand-binding pocket is depicted as gray surface. Menin residue numbering is according to transcript variant: NM_000244.

Figure 2.

JNJ-75276617 inhibits menin-KMT2A complex association with chromatin, downregulates menin-KMT2A target genes, and induces myeloid cell differentiation in KMT2A-altered and NPM1-mutant AML cells. (A) A chromatin immunoprecipitation–quantitative polymerase chain reaction (qPCR) assay was performed to assess the binding of menin to target gene promoters (MEIS1, HOXA9, and HOXA10) in response to treatment with JNJ-75276617 for 48 hours. All qPCRs were performed in triplicate. Signals for menin binding in the KMT2A-r MOLM-14 or NPM1c OCI-AML3 samples were normalized to input values, and binding events per 1000 cells were calculated. Compared with dimethyl sulfoxide (DMSO) control, menin binding was decreased in MOLM-14 and OCI-AML3 cells treated with JNJ-75276617 at concentrations from 0.1 to 1.0 μM (MEIS1, 3.2- to 9.8-fold in MOLM-14, 1.4- to 2.6-fold in OCI-AML3; homeobox gene A9 [HOXA9], 2.4- to 3.4-fold in MOLM-14, 14.0- to 24.7-fold in OCI-AML3; homeobox gene A10 [HOXA10], 1.6- to 2.2-fold in MOLM-14, 5.2- to 10.3-fold in OCI-AML3); in all cases, the fold decreases fall outside the 95% confidence interval (CI) of the DMSO control. (B) JNJ-75276617 inhibited expression of menin-KMT2A target genes and increased expression of differentiation genes in KMT2A-altered and NPM1-mutant AML cells. Cells were incubated with the indicated concentrations of JNJ-75276617 for either 48 (MOLM-14) or 72 (OCI-AML3) hours. Relative expression of menin-KMT2A target genes and differentiation markers was calculated by dividing the normalized values of the treated samples by the normalized value of the DMSO control. The experiment was performed 3 times, and the error bars represent the mean ± standard deviation. (C,D) Differential expression of menin-KMT2A target genes and myeloid cell differentiation signature in response to JNJ-75276617 in various leukemia cells. Samples were prepared in duplicate. Microarray analysis was performed on leukemia cell lines in response to JNJ-75276617 treatment for 48 hours. Representative menin-KMT2A PD markers are shown in volcano plots (C). Gene sets involved in myeloid cell differentiation were enriched on treatment of JNJ-75276617 in KMT2A-r (MOLM-14, MV4-11) and NPM1c (OCI-AML3) cell lines as denoted by enrichment score (red), whereas no gene set enrichment was found in KMT2A/NPM1-WT (HL-60, KO-52, and K-562) cells (D). The experiment was performed once.

Figure 2.

JNJ-75276617 inhibits menin-KMT2A complex association with chromatin, downregulates menin-KMT2A target genes, and induces myeloid cell differentiation in KMT2A-altered and NPM1-mutant AML cells. (A) A chromatin immunoprecipitation–quantitative polymerase chain reaction (qPCR) assay was performed to assess the binding of menin to target gene promoters (MEIS1, HOXA9, and HOXA10) in response to treatment with JNJ-75276617 for 48 hours. All qPCRs were performed in triplicate. Signals for menin binding in the KMT2A-r MOLM-14 or NPM1c OCI-AML3 samples were normalized to input values, and binding events per 1000 cells were calculated. Compared with dimethyl sulfoxide (DMSO) control, menin binding was decreased in MOLM-14 and OCI-AML3 cells treated with JNJ-75276617 at concentrations from 0.1 to 1.0 μM (MEIS1, 3.2- to 9.8-fold in MOLM-14, 1.4- to 2.6-fold in OCI-AML3; homeobox gene A9 [HOXA9], 2.4- to 3.4-fold in MOLM-14, 14.0- to 24.7-fold in OCI-AML3; homeobox gene A10 [HOXA10], 1.6- to 2.2-fold in MOLM-14, 5.2- to 10.3-fold in OCI-AML3); in all cases, the fold decreases fall outside the 95% confidence interval (CI) of the DMSO control. (B) JNJ-75276617 inhibited expression of menin-KMT2A target genes and increased expression of differentiation genes in KMT2A-altered and NPM1-mutant AML cells. Cells were incubated with the indicated concentrations of JNJ-75276617 for either 48 (MOLM-14) or 72 (OCI-AML3) hours. Relative expression of menin-KMT2A target genes and differentiation markers was calculated by dividing the normalized values of the treated samples by the normalized value of the DMSO control. The experiment was performed 3 times, and the error bars represent the mean ± standard deviation. (C,D) Differential expression of menin-KMT2A target genes and myeloid cell differentiation signature in response to JNJ-75276617 in various leukemia cells. Samples were prepared in duplicate. Microarray analysis was performed on leukemia cell lines in response to JNJ-75276617 treatment for 48 hours. Representative menin-KMT2A PD markers are shown in volcano plots (C). Gene sets involved in myeloid cell differentiation were enriched on treatment of JNJ-75276617 in KMT2A-r (MOLM-14, MV4-11) and NPM1c (OCI-AML3) cell lines as denoted by enrichment score (red), whereas no gene set enrichment was found in KMT2A/NPM1-WT (HL-60, KO-52, and K-562) cells (D). The experiment was performed once.

Figure 3.

JNJ-75276617 inhibits proliferation and induces differentiation and apoptosis of AML and B-ALL cells with KMT2A alteration and NPM1 mutations and morphologic differentiation of KMT2A-AF9–transduced mouse bone marrow cells. (A) The antiproliferative activity of JNJ-75276617 was determined in a panel that included 8 AML cell lines (KMT2A-r [MOLM-14, MOLM-13, MV4-11, and THP-1], KMT2A–partial tandem duplication [PTD; EOL-1], NPM1c [OCI-AML3], KMT2A/NPM1-WT [KO-52 and HL-60]), 1 leukemia B-ALL cell line with KMT2A-r (RS4:11), and a KMT2A/NPM1-WT chronic myeloid leukemia (CML) cell line (K562). MOLM-14, MOLM-13, MV4-11, and THP-1 were originally derived from pediatric patients. Cells were treated with JNJ-75276617 for 8 days, and spheroid-like growth was measured in real time by live-cell imaging. Results are shown from a representative experiment in which all cell lines were evaluated in parallel (A). Absolute IC₅₀ and the mean ± standard deviation (SD) values were calculated as percentage change in confluence to DMSO-treated cells. The experiment was performed at least 3 times for each cell line (B). (C,D) Flow cytometry analysis of differentiation and apoptosis in KMT2A-r and NPM1c AML cells following treatment with various concentrations of JNJ-75276617 for 7 days (C). Expression of differentiation markers CD14 and CD11b was evaluated in the viable cell population only, and apoptosis was evaluated in the total cell population (D). Duplicate samples were tested for each condition, and bars represent the mean ± SD. The experiment was performed 3 times. (E) Effect of JNJ-75276617 on morphologic differentiation of KMT2A-AF9–transformed mouse BM cells was examined by May-Grünwald Giemsa staining. KMT2A-AF9–transformed mouse BM cells were treated with DMSO or 200 nM JNJ-75276617 for 10 days. Condensed nuclei, a readout of neutrophil-like morphologic differentiation, were counted after 10-day JNJ-75276617 treatment from 3 independently captured images. Representative images are shown. The experiment was performed once.

Figure 4.

JNJ-75276617 inhibits proliferation and induces differentiation and apoptosis of KMT2A-altered and NPM1-mutant primary AML or B-ALL patient samples. (A) The effect of JNJ-75276617 on viability of primary AML patient samples was tested in cocultures of mononuclear cells from primary AMLs, enriched by CD34⁺ or CD117⁺ selection, and MS5 murine stromal cells. Cell number was counted after 5 different human AML patient samples (with the noted mutations) were treated with JNJ-75276617 at the indicated concentrations for 7 days. CD11b staining was performed to assess differentiation in response to treatment. Test conditions were run in triplicate. Average values of DMSO control are indicated by horizontal dashed lines. Significant difference compared with DMSO control is indicated (∗P < .001, ∗∗P < .01, ∗∗∗P < .05). (B) The effect of JNJ-75276617 on viability of primary KMT2A-r B-ALL patient sample was tested in liquid culture. Cell number was counted after treatment of JNJ-75276617 at the indicated concentrations for 8 days. Test conditions were run in triplicate. Average values of DMSO control are indicated by horizontal dashed lines. Significant difference compared with DMSO control is indicated (∗P < .001). (C) Flow cytometry analysis of apoptosis in primary KMT2A-r B-ALL patient sample following treatment with various concentrations of JNJ-75276617 for 8 days. On day 8, cells were stained with CD45-PECy7 and annexin V APC, in a 96-well plate, and were incubated for 30 minutes at 4°C. Apoptosis was evaluated in the total cell population. Triplicate samples were tested for each condition, and bars represent the mean ± standard deviation. The experiment was performed once.

Figure 5.

JNJ-75276617 alters menin-KMT2A target gene expression and prolongs survival of mice bearing NPM1c and KMT2A-r xenografts. (A) Plasma concentrations following a single oral administration of 5 mg/kg (gray squares) or 50 mg/kg (black squares) in mice. The dashed black line represents the IC₅₀ from the in vitro MEIS1 expression assay (45 ± 6 nM, supplemental Figure 3B). (B) Mice bearing subcutaneous (SC) KMT2A-r MOLM-14 xenografts were treated with JNJ-75276617 at the dose levels indicated for 5 weeks. Line underneath the x axis indicates the dosing period. Data are displayed while at least two-thirds of animals remained in the group (n = 9-10/group). ∗ denotes significant tumor regression (P ≤ .05) from initial tumor volume after JNJ-75276617 treatment. (C) Mice bearing SC MOLM-14 xenografts were treated with JNJ-75276617 at the dose levels indicated for 3 or 11 days (n = 2-5/group). Tumors were harvested 16 hours after the last dose, and MEIS1 mRNA levels were assessed. Expression values are calculated relative to tumors treated with vehicle. Bar graphs represent the mean ± standard deviation (SD). ∗P < .0001. (D-G) Mice engrafted with patient-derived AML xenografts harboring KMT2A-r mutation: CBAM-68552 (D,E) or NPM1c: LEXFAM-2734 (F) or AM7577 (G) were treated with JNJ-75276617 at the dose levels indicated. (E) Bone marrows from CBAM-68552–engrafted mice following 4 weeks of drug treatment were analyzed for differentiation markers (human CD11b, CD13, and CD14) by flow cytometry (n = 3). (H,I) Mice engrafted with B-ALL xenografts with KMT2A-r (CBAB-62871; n = 7/group) were treated with JNJ-75276617 for 6 weeks. (I) Bone marrows from CBAB-62871-engrafted mice following 3 weeks of drug treatment were analyzed for differentiation markers (human CD11b, CD13, and CD14) by flow cytometry (n = 3). Individual values are shown as a dot plot, whereas bar graphs represent the mean ± SD. (D,F,G,H) Line underneath the x axis indicates the dosing period. ∗ denotes significant difference (P ≤ .05) in survival between treatment with JNJ-75276617 and vehicle control (n = 7-10/group). (E,I) ∗ denotes a significant difference vs vehicle control (P < .05).

Figure 6.

JNJ-75276617 exhibits synergistic effects with gilteritinib or venetoclax and azacitidine in AML. (A) KMT2A-r MOLM-13 cells were incubated with the indicated drug combination concentrations for 6 days (JNJ-75276617 and gilteritinib) in technical triplicates. The combination effect was calculated by extended Biochemically Intuitive Generalized Loewe (BIGL) package, with results shown under highest single agent (HSA) null model. Data from 3 independent experiments were pooled and analyzed. Combination effect is represented by the contour plots, with indicating antagonism (red area) and synergy (blue area) as represented by the intensity scale. Dot is proportional to the increase in effect compared with the null, and color intensity is statistical effect. (B) KMT2A-r MOLM-13 cells were treated with JNJ-75276617 for 8 days and venetoclax for 4 days in quadruplicate. The combination effect was calculated by extended BIGL package, with results shown under HSA null model. Data from 2 independent experiments were pooled and analyzed. Combination effect is represented by the contour plots, with indicating antagonism (red area) and synergy (blue area) as represented by the intensity scale. Dot is proportional to the increase in effect compared with the null and color intensity is statistical effect. (C) KMT2A-r MOLM-13 cells were incubated with the indicated drug combination concentrations for 8 days (JNJ-75276617 and azacitidine) and for 4 days (venetoclax) in quadruplicate. The combination effect was calculated by extended BIGL package, with results shown under HSA null model. Data from 2 independent experiments were pooled and analyzed. Combination effect is represented by the contour plots, with indicating antagonism (red area) and synergy (blue area) as represented by the intensity scale. Dot is proportional to the increase in effect compared with the null and color intensity is statistical effect. (D) Mice bearing disseminated MOLM-13 xenografts were treated with JNJ-75276617, venetoclax, azacitidine, or combinations at the dose levels indicated. Lines underneath the x axis indicate the dosing period for each drug. ∗ denotes significant difference (P ≤ .05) in survival compared with the vehicle control group.

Figure 7.

JNJ-75276617 exhibits potent antiproliferation activity in MEN1-mutant cell lines resistant to other menin-KMT2A inhibitors. (A) Fluorescence polarization assay demonstrating the dose-dependent displacement of a KMT2A peptide from WT, M327I-mutant, and T349M-mutant menin on treatment with JNJ-75276617. Data are represented as mean ± standard deviation (SD), n = 3 independent replicates. (B) Evaluation of JNJ-75276617 for antiproliferation activity in MEN1^M327I and MEN1^T349M mutant KMT2A-r MV4-11 cell lines. Drug response was calculated on the basis of the luminescence for each menin inhibitor concentration relative to the DMSO sample after 10 days of JNJ-75276617 treatment. Data are represented as mean ± SD, n = 3 independent experiments. (C) Superimposition of a docking model of JNJ-75276617 (green sticks; docked on M327I-mutant menin chain A; Protein Data Bank Identifier [PDB] ID: 8E90; second rotamer of W346 [chain B] is displayed in gray lines). JNJ-75276617 bound to wild-type menin (this study; PDB ID: in deposition) is shown for reference in orange and its binding pocket as gray surface. Side chain of M327/I327 is shown as sticks. The distance between selected atoms of T349 and JNJ-75276617 is shown as orange dashed line. Residue numbering is according to transcript variant: NM_000244.