MCL1 inhibition exerts potent antileukemic activity by restoring apoptosis in T-lineage acute lymphoblastic leukemia∗ Authors: Giorgia Giordano, Olivia Arnold, Diren Arda Karaoglu, Jommel Macaraeg, Kathryn Shimamoto, Arjun Dhar, Ella Postich, Hamed R. Youshanlouei, Sinan Cetin, Girish Venkataraman, Anand A. Patel, Adam DuVall, Michael W. Drazer, David T. Teachey, James L. LaBelle, Wendy Stock, Caner Saygin Published: 2026 TO THE EDITOR: T-acute lymphoblastic leukemia (T-ALL) constitutes 20% of ALL cases and is diagnosed across the spectrum of all ages.1 Treatment with nelarabine-based regimens, modeled after pediatric multiagent chemotherapy protocols, has pushed survival to 85% for children and young adults.2 Unlike B-ALL, drug development for T-ALL has been considerably limited, with no molecularly targeted or cellular therapies approved to date. This translates into dismal survival outcomes for patients with relapsed T-ALL.1,3 To address the unmet need, several early phase clinical trials investigated the efficacy of BH3 mimetics venetoclax (B-cell lymphoma 2 [BCL-2] inhibitor) and navitoclax (BCL-2/B-cell lymphoma-extra large [BCL-xL] inhibitor) in relapsed/refractory ALL.4,5 In T-ALL, immature subsets (eg, early T-precursor) depend on BCL-2, whereas more mature subsets (eg, precursor T) depend on BCL-xL for survival.6 Combination of venetoclax, navitoclax, and chemotherapy led to 50% complete remission rate in relapsed T-ALL.4 However, the median duration of response and overall survival were 4 and 6 months, respectively. BH3 profiling identified changes in antiapoptotic protein dependence at relapse, whereas transcriptomic studies showed increased Myeloid Cell Leukemia 1 (MCL1) expression. These data suggest that restoring leukemic cell apoptosis by targeting BCL-2 family of proteins is a promising strategy for T-ALL, yet BCL-2/BCL-xL inhibition may not be sufficient for long-term disease control. Based on consistently high MCL1 expression in both treatment-naïve and relapsed/refractory disease, we hypothesize that MCL1 blockade can be an effective therapeutic strategy in T-ALL. To investigate antileukemic activity of MCL1 blockade in T-ALL, we studied the sensitivity of 10 different T-ALL primary patient samples to treatment with 3 different MCL1 inhibitors (Figure 1A; supplemental Methods). Across the samples obtained from patients with different age, sex, and molecular subtypes, ABBV-467 was more effective with lower 50% inhibitory concentration (IC50) values when compared with AZD-5991 and S63845 (Figure 1B; supplemental Table 1). Similarly, ABBV-467 was more potent than AZD-5991 in 6 T-ALL cell lines tested (supplemental Figure 1A-B). Primary cells that were more sensitive to MCL1 inhibition had higher MCL1 protein levels (Figure 1C). To study antileukemic efficacy and toxicity of ABBV-467 in vivo, we engrafted 10 Nod-scid-gamma (NSG) mice with T-ALL patient-derived xenograft (PDX) cells and treated with vehicle vs ABBV-467 (25 mg/kg weekly intraperitoneally) for 28 days. Mice treated with ABBV-467 had significantly prolonged survival (Figure 1D). Next, we treated similar number of PDX-engrafted NSG mice for 4 weeks and sacrificed to assess tumor burden and toxicity. After 4 weeks of treatment, MCL1 inhibitor-treated mice had significantly lower CD7+ leukemic cell burden in their bone marrows and spleens (Figure 1E-F; supplemental Figure 1C). Histopathological examination of liver, heart, kidney, and gut tissues did not demonstrate significant toxicity in ABBV467-treated mice (supplemental Figure 1D). Collectively, these data suggest that MCL1 inhibition exerts potent antileukemic activity in T-ALL cell lines, primary cells, and PDX model.Figure 1.MCL1 inhibition is an effective treatment strategy in T-ALL models. (A) Curves showing viability of T-ALL primary patient samples treated with 3 different MCL1 inhibitors. (B) Whisker plots showing log(IC50) values for T-ALL primary patient cells treated with MCL1 inhibitors. (C) Immunoblots showing BCL-2 family of protein expression in T-ALL primary cells. (D) Kaplan-Meier overall survival curves of NSG mice engrafted with T-ALL PDX1 cells and treated with vehicle vs ABBV-467. (E) CD7+ leukemic cell percentages in bone marrow of NSG mice engrafted with T-ALL PDX1 and treated with vehicle vs ABBV-467. (F) Histopathological examination of bone marrow and spleen leukemic cell burdens in NSG mice engrafted with T-ALL PDX1 and treated with vehicle vs ABBV-467. (G) Immunoblots showing BAX, BAK and actin levels of isogenic control and BAX/BAK1 dKO MOLT4 cell line. (H) Curves showing viability of isogenic control and BAX/BAK1 dKO MOLT4 cell lines treated with MCL1 inhibitor. (I) Curves showing viability of T-ALL cell line MOLT4 and primary patient cells pretreated with pan-caspase inhibitor Q-VD-OPh for 4 hours and treated with ABBV-467. (J) Immunoblots showing procaspase and cleaved caspase levels of T-ALL cell lines and primary patient samples treated with BH3 mimetics in the presence or absence of Q-VD-OPh. ∗∗P < .01; ∗∗∗P < .001. dKO, double knockout; IC50, 50% inhibitory concentration. MCL1 inhibition is an effective treatment strategy in T-ALL models. (A) Curves showing viability of T-ALL primary patient samples treated with 3 different MCL1 inhibitors. (B) Whisker plots showing log(IC50) values for T-ALL primary patient cells treated with MCL1 inhibitors. (C) Immunoblots showing BCL-2 family of protein expression in T-ALL primary cells. (D) Kaplan-Meier overall survival curves of NSG mice engrafted with T-ALL PDX1 cells and treated with vehicle vs ABBV-467. (E) CD7+ leukemic cell percentages in bone marrow of NSG mice engrafted with T-ALL PDX1 and treated with vehicle vs ABBV-467. (F) Histopathological examination of bone marrow and spleen leukemic cell burdens in NSG mice engrafted with T-ALL PDX1 and treated with vehicle vs ABBV-467. (G) Immunoblots showing BAX, BAK and actin levels of isogenic control and BAX/BAK1 dKO MOLT4 cell line. (H) Curves showing viability of isogenic control and BAX/BAK1 dKO MOLT4 cell lines treated with MCL1 inhibitor. (I) Curves showing viability of T-ALL cell line MOLT4 and primary patient cells pretreated with pan-caspase inhibitor Q-VD-OPh for 4 hours and treated with ABBV-467. (J) Immunoblots showing procaspase and cleaved caspase levels of T-ALL cell lines and primary patient samples treated with BH3 mimetics in the presence or absence of Q-VD-OPh. ∗∗P < .01; ∗∗∗P < .001. dKO, double knockout; IC50, 50% inhibitory concentration. In addition to its canonical function as an inhibitor of apoptosis, MCL1 has noncanonical functions in cell cycle regulation, DNA repair and cell metabolism.7 To uncouple these functions, we generated isogenic control and BAX/BAK1 double knockout MOLT4 cell lines by using CRISPR/Cas9 technology (Figure 1G). Apoptosis-resistant double knockout cells lost their sensitivity to ABBV-467 (Figure 1H). Similarly, we treated T-ALL cell lines and primary patient samples with pan-caspase inhibitor Q-VD-OPh and assessed their sensitivity to potent MCL1 inhibitor, ABBV-467 (Figure 1I). Pretreatment of T-ALL cells with Q-VD-OPh for 4 hours before ABBV-467 abrogated their sensitivity to MCL1 blockade. Decreased sensitivity to the induction of apoptosis was also evidenced by the absence of caspase 3 cleavage in T-ALL cells that are cotreated with BH3 mimetics (navitoclax or ABBV-467) and Q-VD-OPh (Figure 1J). Altogether, these data demonstrate that the therapeutic efficacy of MCL1 inhibition in T-ALL depends on its canonical antiapoptotic function. Despite their promising antileukemic activity, acquired resistance to BH3 mimetics accounts for relapse and shortened survival in ALL.4,5 To gain insights into the mechanisms of acquired resistance to MCL1 inhibition in T-ALL, we derived isogenic control and ABBV-467–resistant (A467-R) MOLT4 and MOLT16 T-ALL cell lines by passaging parental cells with progressively higher concentrations of ABBV-467 over the course of 2 months (Figure 2A). A467-R cells had >100-times higher 50% inhibitory concentration than their control counterparts. We observed increased MCL1 and BCL-xL expression in A467-R cells (Figure 2B). Next, we investigated changes in selective antiapoptotic protein dependencies of control vs A467-R T-ALL cells through BH3 profiling. In this functional assay, isogenic cells were treated with BH3 peptides that target different members of the BCL-2 family and mitochondrial depolarization was measured by cytochrome c release (supplemental Methods). A467-R MOLT4 T-ALL cells were characterized by decreased apoptotic priming (ie, lower response to BIM), response to MCL1 inhibition (ie, MS1 peptide stimulation), and increased sensitivity to BCL-xL inhibition (ie, HRK peptide stimulation) (Figure 2C). To understand the mechanism by which MCL1 inhibitor resistance increases BCL-xL dependence in T-ALL, we performed RNA-sequencing to compare the transcriptomic profiles of MOLT4 control vs A467-R cells (Figure 2D). A467-R cells had upregulated T-cell receptor (TCR) signaling when compared with controls (Figure 2E). We have previously shown that pre-TCR signaling mediates increased expression of BCL-xL via NF-κB pathway, and ACK1-AKT signaling axis enhances BH3 mimetic resistance in T-ALL.6 We observed a similar mechanism of increased ACK1 and AKT signaling in MOLT4 and MOLT16 A467-R cells (Figure 2F). As a result, MCL1 inhibitor-resistant MOLT4 and MOLT16 cells were susceptible to BCL-2/BCL-xL inhibition with navitoclax or pre-TCR/ACK1 signaling blockade with dasatinib (supplemental Figure 2A). We validated these findings in primary T-ALL patient samples (T-ALL2, T-ALL8) by demonstrating synergy between ABBV-467 and navitoclax, dasatinib, or capivasertib (AKT inhibitor) (Figure 2G; supplemental Figure 2B-E). Next, we investigated the combination of dual MCL1 and BCL-2/BCL-xL blockade in vivo, which resulted in acute toxicity and fatality in mice treated with ABBV-467 plus navitoclax. The toxicity was seen even in mice without engrafted leukemic cells (supplemental Figure 3). Therefore, we tested the combination of ABBV-467 with capivasertib in T-ALL PDX2 model. A total of 32 mice were engrafted with T-ALL PDX2 cells and randomized to 4 groups to receive vehicle, ABBV-467 (25 mg/kg weekly intraperitoneally), capivasertib (75 mg/kg orally daily), and the combination of ABBV-467 plus capivasertib at same doses (Figure 2H). Mice in single-agent groups had improved survival when compared with vehicle control, but the combination arm had significantly better survival. We did not observe any significant toxicity with the combination of ABBV-467 and capivasertib. Altogether, these data suggest that upregulated pre-TCR/ACK1/AKT signaling drives MCL1 inhibitor resistance in T-ALL, and the combination of capivasertib and ABBV-467 is synergistic in vitro and in vivo.Figure 2.Pre-TCR/ACK1/AKT signaling drives MCL1 inhibitor resistance in T-ALL. (A) Curves showing viability of isogenic control and A467-R T-ALL cells treated with ABBV-467. (B) Immunoblots showing MCL1, BCL-xL and BCL-2 levels of isogenic T-ALL control and A467-R cell lines. (C) Graphs showing mitochondrial depolarization upon stimulation with BIM, MS1, BAD, and HRK peptides in BH3 profiling of MOLT4 control and A467-R cells. (D) Dot plot shows genes that are upregulated in A467-R MOLT4 cells when compared with control cells in RNA-seq analysis. (E) Enrichment plot for T-cell receptor complex signaling in A467-R T-ALL cells. (F) Immunoblots showing ACK1 and AKT signaling pathway activity in control vs A467-R T-ALL cell lines. (G) ZIP synergy plots of T-ALL primary cells treated with capivasertib and ABBV-467. (H) Kaplan-Meier overall survival curves NSG mice engrafted with T-ALL PDX2 cells and treated with vehicle vs ABBV-467 vs capivasertib vs the combination of ABBV-467 plus capivasertib (n = 8 mice per group). ∗∗P < .01; ∗∗∗P < .001. Pre-TCR/ACK1/AKT signaling drives MCL1 inhibitor resistance in T-ALL. (A) Curves showing viability of isogenic control and A467-R T-ALL cells treated with ABBV-467. (B) Immunoblots showing MCL1, BCL-xL and BCL-2 levels of isogenic T-ALL control and A467-R cell lines. (C) Graphs showing mitochondrial depolarization upon stimulation with BIM, MS1, BAD, and HRK peptides in BH3 profiling of MOLT4 control and A467-R cells. (D) Dot plot shows genes that are upregulated in A467-R MOLT4 cells when compared with control cells in RNA-seq analysis. (E) Enrichment plot for T-cell receptor complex signaling in A467-R T-ALL cells. (F) Immunoblots showing ACK1 and AKT signaling pathway activity in control vs A467-R T-ALL cell lines. (G) ZIP synergy plots of T-ALL primary cells treated with capivasertib and ABBV-467. (H) Kaplan-Meier overall survival curves NSG mice engrafted with T-ALL PDX2 cells and treated with vehicle vs ABBV-467 vs capivasertib vs the combination of ABBV-467 plus capivasertib (n = 8 mice per group). ∗∗P < .01; ∗∗∗P < .001. The clinical development of MCL1 inhibitors has been hampered by their dose-limiting cardiotoxicity.8,9 This is attributed to the essential role of MCL1 in cardiomyocyte survival through its canonical and noncanonical functions. Our results reinforce that MCL1 represents a bona fide therapeutic target in T-ALL, and its canonical function as a gatekeeper of mitochondrial apoptosis is indispensable for leukemic cell survival. The observed synergy between MCL1 inhibitors and capivasertib, coupled with its favorable toxicity profile, offers a compelling combinatorial approach that circumvents the toxicity we observed with cotargeting MCL1 and BCL-xL. Our findings have immediate translational implications, supporting the advancement of MCL1-targeted therapies in T-ALL through rationally designed combination regimens. Future efforts must prioritize the development of next-generation MCL1 inhibitors that focus on blocking canonical MCL1 function and improving therapeutic indices. Conflict-of-interest disclosure: D.T.T. is the Garrett Brodeur Endowed Chair in Oncology at the Children’s Hospital of Philadelphia; serves on advisory boards (unpaid) for AbbVie, Amgen, AstraZeneca, Autolus, Beam Therapeutics, J&J Innovation, Jazz, Merck, Novartis, Pfizer, Servier, Sobi, Syndax, and NeoImmuneTech; receives research funding (to his laboratory) from Beam Therapeutics, and NeoImmuneTech; and has patents or patents pending for CART. J.L.L. reports grant support from AbbVie; and consultancy for Newave Pharmaceutical Inc.