Co‐Targeting c‐Myc and Bcl‐2 by Oral Small Molecule Combination of WBC100 and Venetoclax Effectively Controls Acute Myeloid Leukemia in Preclinical Models

Uncontrolled proliferation and apoptosis evasion are two hallmarks of acute myeloid leukemia (AML), but the molecular mechanisms remain poorly understood. In this study, it is demonstrated that the double over‐expresser of oncoprotein c‐Myc and anti‐apoptotic protein Bcl‐2 is a critical “genetic overdrive” of proliferation and apoptosis evasion in AML and is associated with poor genetic alterations. Double‐knockdown of c‐Myc/ Bcl‐2 synergistically kills AML cells in vitro and in vivo. Moreover, a novel oral small molecule combination co‐targeting c‐Myc and Bcl‐2 with WBC100 and Venetoclax (VEN) at low doses are developed. Importantly, the study shows that this combination results in deep and durable remissions of AML, and its efficacy is superior to the frontline combination of Venetoclax and hypomethylation azacitidine (AZA) in AML mouse models and PDX models from relapsed or refractory AML patients. Mechanically, Bcl‐2 knockdown induces mitochondrial outer membrane permeabilization and c‐Myc‐knockdown impairs mitochondria biogenesis. Co‐targeting c‐Myc/Bcl‐2 reciprocally abrogates over‐proliferation and apoptosis resistance via forming a double hit to mitochondrial biogenesis and apoptosis machinery. The findings for the first time demonstrate that co‐targeting c‐Myc/Bcl‐2 by the novel oral small molecule combination of WBC100/Venetoclax is a promising and convenient therapy for AML and support its clinical trial.


Introduction
Despite the advances in the treatment of acute myeloid leukemia (AML), the fiveyear survival rate of AML patients remains poor. Chemotherapy can successfully induce remission, but most AML patients suffer from lethal disease relapse within 3 years. [1] Recently, the anti-apoptotic protein Bcl-2 inhibitor Venetoclax combined with the hypomethylating agent(HMA) azacitidine exhibits encouraging results for newly diagnosed AML, [2] but this combination remains less effective in patients with relapsed/refractory (R/R)AML, [3] highlighting the urgent need for novel therapies.
Over-proliferation and apoptosis evasion are two hallmarks of acute myeloid leukemia, [4] but the molecular mechanisms remain poorly understood. The transcription factor c-Myc is deregulated in most AMLs [5] and plays a pivotal role in the over-proliferation of AML by maintaining the undifferentiated state and self-renewal capacity of leukemia stem cells (LSCs). [6] Anti-apoptotic protein Bcl-2 is frequently overexpressed in LSCs and plays a central role in apoptosis evasion in AML. [7] Furthermore, overexpression of Bcl-2 can accelerate c-Myc-induced AML-like disease. [8] In diffuse large B-cell lymphomas (DLBCLs), concomitant expression of c-Myc/Bcl-2 proteins, known as "double expressers", has been associated with poor prognosis and aggressive disease. [9] Therefore, we suppose that dual over-expresser of c-Myc/ Bcl-2 proteins may directly lead to over-proliferation and apoptosis evasion of AML cells, and co-targeting c-Myc/Bcl-2 might be an effective therapeutic strategy.
To test the above hypothesis and develop a potent and convenient therapy, we first investigated whether double over-expresser of c-Myc/ Bcl-2 was associated with AML and then attempted to develop an oral small molecule combination to co-target c-Myc/Bcl-2. We found that double over-expresser of c-Myc/Bcl-2 was common in AML patients, and was associated with high genetic alterations frequency. These genetic alterations, often mutations of tumor suppressor and oncogenes mutations, including TP53 mutation, FLT3-ITD mutation, SF3B1 mutation, SRSF2 mutation, and so on, are associated with the increased risk of AML relapse and chemotherapy-resistant AML. [10] Our studies showed that co-targeting c-Myc/ Bcl-2 synergistically killed AML cells by impairing mitochondria biogenesis and apoptosis machinery. WBC100 is an effective c-Myc inhibitor directly degrading c-Myc protein through CHIP mediated 26S proteasome pathway and the mechanism is different from that reported previously. [11] Importantly, we demonstrated that the novel oral small molecule combination co-targeting c-Myc with WBC100 and Bcl-2 with Venetoclax (VEN) induced deep and durable remissions, and its efficacy was superior to the frontline combination treatment of VEN and azacytidine in various AML mouse models as well as AML PDX models with R/R AML.

Double Over-Expresser of c-Myc and Bcl-2 is Common in AML and Reveals a High-Risk Signature
The oncoprotein c-Myc and anti-apoptotic protein Bcl-2 are deregulated in hematological malignancies and play important roles in drug resistance and relapse, [12] but little is known about the co-overexpression of c-Myc and Bcl-2 (double overexpresser) in AML patients. By analyzing data from BloodSpot (http://www.bloodspot.eu/), we found that both c-Myc and Bcl-2 transcripts were relatively higher in the majority of AML patients ( Figure 1A). To further validate this result, we used western blotting to examine c-Myc and Bcl-2 protein levels in primary AML samples from patients and observed that the majority of AML samples showed elevated protein expressions of c-Myc and Bcl-2 ( Figure 1B), as compared to normal mononuclear cells isolated from umbilical cord blood (CB) ( Figure 1C). Detailed information on AML patients was shown in Table S1 (Supporting Information). The tested 36 AML patients were divided into two groups according to the median cut-off value of c-Myc/Bcl-2 expression: c-Myc high /Bcl-2 high group (high expression group), others with c-Myc low , Bcl-2 low or c-Myc low /Bcl-2 low (low expression group). To investigate whether the double over-expresser of c-Myc and Bcl-2 was implicated in relapse and refractory AML, we performed cytogenetic analyses and somatic gene mutational analyses related to relapsed/refractory AML [13] (Figure 1D). Interestingly, the c-Myc/Bcl-2 high expression group showed more adverse mutation events, such as TP53 mutation, FLT3-ITD mutation, SF3B1 mutation, and TET2 mutation than the low expression group (p = 0.0075). But there was no difference in chromosome aberrations between the two groups ( Figure 1D). We expanded this finding to a larger cohort from TCGA and got a similar result ( Figure 1E). Since these mutations are associated with limited response to drugs, we next investigated whether the dual expresser of c-Myc and Bcl-2 was correlated with poor outcomes based on studies from the Beat AML program (≈200 AML samples, Figure 1F). [14] Overall survival (OS) differed significantly between c-Myc/Bcl-2 low expression group and c-Myc or Bcl-2 high expression group (log-rank p = 0.031). In addition, we assessed overall survival time by analyzing small samples (23 patients) with c-Myc high /Bcl-2 high and low expression groups and found that patients with high expression of c-Myc/Bcl-2 had a poorer overall survival (log-rank p = 0.0155) than others ( Figure 1G). The distribution of demographic data, cytogenetic risk groups, and c-Myc/Bcl-2 coexpression pattern was presented in Table S2 (Supporting Information). The median survival time in the dual expresser of c-Myc and Bcl-2 group was 3 months, whereas the low-expresser of c-Myc and Bcl-2 group was 13 months ( Figure 1G).
Collectively, these observations indicate that double overexpresser of c-Myc/Bcl-2 is common in AML. Patients with c-Myc high /Bcl-2 high have an increased risk of gene mutations and dismal outcomes.

Double Over-Expresser of c-Myc and Bcl-2 Contributes to Over-Proliferation of AML Cells
To determine whether the double over-expresser of c-Myc and Bcl-2 was associated with the over-proliferation of AML cells, we compared the effects of single-knockdown and doubleknockdown of c-Myc and Bcl-2 on the proliferation of AML cells with doxycycline (DOX)-inducible specific shRNAs targeting c-Myc and Bcl-2 in MOLM-13 and Kasumi-1 cells, which highly express c-Myc and Bcl-2 (Figure 2A,B). Single knockdown of c-Myc inhibited the proliferation of MOLM-13 and Kasumi-1 cells by 50.8% and 76.8%, respectively ( Figure 2C), whereas single knockdown of Bcl-2 showed a minor effect with 10.7% and 25.6% growth inhibition rates ( Figure 2C). However, double-knockdown (double-KD) exhibited a more significantly synergistic effect with growth inhibition rates up to 94.7% and 87.4% in MOLM-13 and Kasumi-1 cells at day 7 ( Figure 2C).
Growth inhibition caused by double-knockdown of both genes was further validated in vivo using an orthotopic AML model in NSG mice with MOLM-13-luciferase (MOLM-13-Luc) cells stably expressing DOX-inducible c-Myc and Bcl-2 shRNAs. Five days after transplantation, AML mice were treated with doxycycline by daily oral gavage to induce knockdown of c-Myc or /and Bcl-2. The tumor burden was evaluated by monitoring the tumor signal using a bioluminescence imaging (BLI) system ( Figure 2D). Consistent with in vitro results, single-knockdown of c-Myc or Bcl-2 greatly suppressed AML growth, and the tumor signal quantitated by photon intensities was 3.33 × 10 7 and 1.32 × 10 8 respectively, while the control group showed a rapid tumor progression with photon intensities of 1.34× 10 10 at day 19 ( Figure 2E). However, double-knockdown of c-Myc/Bcl-2 genes resulted in more . When analyzing chromosome abnormalities, normal karyotype was plotted as "0", complex cytogenetic was plotted as "2" and others were plotted as "1". E) Analysis of gene mutation events in c-Myc high /Bcl-2 high (n = 70) and low-expression (n = 41) samples from TCGA. F) Survival analysis of patients with AML obtained from the Beat AML program showing differences between the c-Myc high /Bcl-2 high group and the low-expression group (statistical significance was evaluated using the log-rank test). G) Kaplan-Meier curve analysis of overall survival in AML patients by the expression of c-Myc and Bcl-2. Patients were divided into high-exp (n = 9) or low-exp (n = 14) by the median value of c-Myc and Bcl-2 (statistical significance was evaluated using the log-rank test). The p values were calculated using a 2-tailed, unpaired Student t-test or one-way ANOVA for multiple comparisons: * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001; ns, not significant. synergistic inhibition with photon intensities of 2.91× 10 6 at day 19, as compared to that of single-knock-down of c-Myc or Bcl-2 ( Figure 2E). Consistently, the overall survival time of the double-KD group was longer than that of the single c-Myc-KD group or single Bcl-2-KD group (log-rank p = 0.009 for double-KD vs Bcl-2-KD, log-rank p = 0.017 for double-KD vs c-Myc-KD) ( Figure 2F). These results clearly show that co-targeting c-Myc and Bcl-2 synergistically inhibits the growth of AML cells and is an ideal potential therapeutic strategy.

Oral Small Molecule c-Myc Inhibitor WBC100 Synergizes with Bcl-2 Inhibitor Venetoclax to Kill AML Cells In Vitro
Based on the findings above, we next sought to determine whether oral small molecule inhibitors of c-Myc and Bcl-2 could display a synergistic role in eliminating AML cells. Our recent studies have shown that WBC100 is a novel orally active smallmolecule c-Myc inhibitor that potently regresses c-Myc-high tumors via the degradation of c-Myc protein. [15] Venetoclax (VEN) is an oral Bcl-2 inhibitor, but single-agent VEN results in only ≈20% overall response rate. [16] Therefore, we used a panel of various AML cell lines to evaluate synergistic activity between two oral small molecules WBC100 and VEN. Karyotype and mutational characteristics, including complex karyotypes, TP53, and FLT3-ITD mutations were shown in Table S3 (Supporting Information). IC50s of WBC100 and VEN in all tested cell lines were shown in Figure S1A (Supporting Information). We observed that the combination of WBC100 and VEN had a significant synergy in cell proliferation inhibition in tested cell lines ( Figure 3A; Figure S1B, Supporting Information), which was assessed by the combination index (CI<1) according to the median effect method of Chou and Talala ( Figure 3B; Figure S1C, Supporting Information). Flow cytometry analysis showed that combined treatment with WBC100 and Venetoclax for 48 h resulted in a higher apoptosis rate (71.2%) than that of single-agent treatment (45.0% for WBC100 and 36% for Venetoclax) ( Figure 3C; Figure S1D, Supporting Information).
Next, we determined two important anti-apoptotic proteins MCL-1 and Bcl-xL, which confer resistance to Venetoclax in AML treatment. As expected, the combination resulted in a significant decrease of MCL-1/Bcl-xL proteins as compared to the control or single agent. Consistently, the decrease of MCL-1 and Bcl-xL proteins was concomitantly accompanied by an increase of cell apoptotic molecules cleaved caspase-3 and PARP ( Figure 3F). We also observed that WBC100 dramatically decreased c-Myc protein levels in MOLM-13 and Kasumi-1 cells ( Figure 3F).
Taken together, the combination of two oral small molecule inhibitors WBC100 and VEN exhibits a significantly synergistic anti-leukemia activity despite various cytogenetic abnormalities, indicating a broader use in AML treatment. Furthermore, the combination can overcome resistance to Venetoclax by downregulating MCL-1 and Bcl-xL.

Double-Knockdown of c-Myc and Bcl-2 Impairs the Synthesis and Function of Mitochondria
Given that c-Myc and Bcl-2 play critical roles in mitochondriamediated chemotherapy resistance of cancers, [17] we next performed an RNA-seq experiment to reveal global gene expression changes caused by c-Myc and Bcl-2 double-KD. As detected by flow cytometry, 24 h induction of shRNAs by DOX showed no significant cell death ( Figure S2A, Supporting Information). When we combined the two-time points (24 h and 72 h) of our screen, we found that c-Myc/Bcl-2 double KD markedly impaired mitochondrial gene expression, mitochondrial translation, and protein localization to the mitochondrion by GO analy-sis ( Figure 4A). Gene set enrichment analysis (GSEA) revealed that c-Myc and Bcl-2 double-KD caused marked disruption of the mitochondrial membrane, including inhibiting mitochondrial gene expression and facilitating membrane permeability ( Figure 4B). We next labeled the mitochondria of MOLM-13 cells with Mito-Tracker, a dye that accumulates in active mitochondria, after c-Myc/Bcl-2 double-KD by DOX induction for 72 h to observe mitochondrial changes in morphology or function. Compared to the SCR group, c-Myc/Bcl-2 double-KD exhibited a significant decrease in Mito-Tracker fluorescence intensity, suggesting that c-Myc and Bcl-2 double-KD can reduce the mitochondrial amount ( Figure S2B, Supporting Information). The transmission electron microscope (TEM) study further showed that c-Myc/Bcl-2 double-KD induced a 60% decrease in the mitochondrial number compared to the SCR group (p = 0.0003) ( Figure 4C,D).
c-Myc is essential in mitochondrial biogenesis [18] and Bcl-2 is an antiapoptotic protein anchor to the outer mitochondrial membrane. [19] KEGG pathway analysis of the Bcl-2-KD group showed significant enrichment in inflammation-related signaling pathways, including cytokine/cytokine receptor interaction, and chemokine signaling pathway while c-Myc-KD downregulated many genes involved in mitochondria biogenesis, such as metabolic pathways, and DNA replication [5,20] (Figure S2C, Supporting Information). To further demonstrate if either or both c-Myc or Bcl-2-KD, specifically influenced the mitochondrial number in AML cells, we performed a TEM assay and directly observed mitochondrial morphology changes upon c-Myc or Bcl-2-KD, with increased hyperfused and swollen mitochondrial phenotype, which was more pronounced in the double-KD group with aberrant mitochondrial morphology ( Figure 4E). Previous researches report elongated mitochondrial networks impede proper nucleoid distribution and removal of damaged mtDNA, [21] so we next investigated whether c-Myc or Bcl-2 KD contributes to mtDNA reduction by determining mtDNA copy numbers with real-time quantitative PCR. As compared to controls, total mtDNA decreased in the c-Myc or Bcl-2-KD group and reduced more significantly in the double-KD group by 1/2-2/3 ( Figure 4F). These results indicate simultaneous silence of both MYC and BCL2 genes shows additive effects on mitochondria disruption and impairs mitochondrial homeostasis.
Given that the mitochondrial DNA (mtDNA) stress response can induce an inflammatory response and the activation of a variety of inflammatory mediators, [22] these results explained enhanced inflammatory responses in Bcl-2-KD cells, which might be a result of mtDNA release to the cytosol as a consequence of incomplete apoptosis and mitochondrial outer membrane permeabilization (MOMP). To test this hypothesis, we extracted mtDNA from the cytosol and quantitative analysis showed that cytosolic mtDNA content increased by 2-3 fold upon Bcl-2 deletion ( Figure  S2D, Supporting Information). These results suggest that Bcl-2-KD induces mitochondrial outer membrane permeabilization (MOMP), which is required for mtDNA release into the cytosol.
To examine whether the mechanism observed in the genetic inhibition applies to the pharmacological synergism, we evaluated mitochondrial dysfunction. Mito-Tracker Red staining of the WBC100/VEN combination group revealed a remarkable reduction of mitochondria, as shown by a significant decrease in mean fluorescence intensity compared to the control(p = 0.006) or single agent group ( Figure S2E, Supporting Information; Figure  4G). Since active mitochondria are the site of oxidative phosphorylation (OXPHOS), to further demonstrate if OXPHOS is specifically inhibited by the combination treatment in AML cells, we detected reactive oxygen species (ROS), the main byproduct of OXPHOS, using DCFH-DA and flow cytometry. As expected, the WBC100/VEN combination group showed a 70% ROS decline in Kasumi-1 and a 50% ROS decline in MV4-11 cells compared to controls ( Figure S2F, Supporting Information; Figure 4H).
Overall, our findings suggest that co-targeting c-Myc and Bcl-2 concurrently acts as a double hit to induce mitochondrial outer membrane permeabilization and disrupt mitochondrial homeostasis, contributing to impaired mitochondrial respiratory function, and leading to cell death.

Combination of Oral Small Molecule Inhibitors Co-targeting c-Myc and Bcl-2 Exhibits Deep and Durable Remissions in Refractory AML Models
Given the marked in vitro synergy of two oral small molecules WBC100 and VEN against AML cells, we next evaluated in vivo anti-leukemia efficacy of WBC100 or VEN alone and in combination in the NSG mouse model. Mice were engrafted with a refractory MOLM13-Luc cell line via the tail vein. We selected the MOLM-13 cell line since it is resistant to AML standardof-care chemotherapeutic agent idarubicin (IDA) as we previously reported. [15] After confirmation of AML engraftment, the mice were randomly assigned to four groups and orally administered with the vehicle, low doses of WBC100 (0.4 mg kg −1 ) or VEN (50 mg kg −1 ), or both. After 2 weeks of treatment, quantitation of MOLM-13-Luc luminescence via in vivo imaging showed that the leukemic burden in the WBC100/VEN combination group nearly vanished (p<0.0001) while the single agent VEN group (p = 0.0523, non-significant) only showed slightly lower photon intensity (leukemia burden) than the vehicle group and soon succumbed to death ( Figure 5A). Single-agent WBC100 showed potent efficacy in vivo (p = 0.0004) compared to the vehicle. However, a significant difference was observed in photon intensity between single agent WBC100 and WBC100/VEN combination arms at day 14, with photon intensities of 1.52 × 10 7 in the WBC100 group, compared to 8.04 × 10 5 in the WBC100/VEN combination group (p = 0.005) ( Figure 5A,B). Furthermore, the WBC100/VEN combination group didn't show significant weight loss compared to the control (p = 0.9813) or single agent when dosed for 2 weeks (Figure 5C), suggesting that the WBC100/VEN combination was well tolerated. As expected, the mice in the WBC100/VEN combination group survived the longest while mice in other groups all died of AML invasion in a short time ( Figure 5A). These results were further supported by an additional, more aggressive in vivo model with Kasumi-1-Luc cells. The leukemic burden in the vehicle and VEN groups progressed rapidly, whereas WBC100 and WBC100/VEN combination groups demonstrated great suppression of AML growth ( Figure 5D). Since the mice receiving WBC100 or WBC100/VEN combination had no detectable tumor signals ( Figure 5E) as measured by bioluminescent imaging, dosing was stopped at day 20 to observe disease relapse. We found that the efficacy superiority of the WBC100/VEN combination over WBC100 monotherapy was evident at day 40 (20 days after drug withdrawal) ( Figure 5F). These results show that the WBC100/VEN combination exhibits more deeper and durable remissions in refractory AML models as compared with WBC100 or VEN monotherapy.

Oral Small Molecule Combination of WBC100/VEN is Superior to the Frontline Combination of VEN/Azacitidine in AML Mouse Models
Treatment with VEN in combination with azacytidine results in an equally poor ≈20% overall objective response rate in AML, with only half of the responses being complete remissions in the relapsed setting. [23] To investigate whether the novel oral small molecule combination WBC100/VEN could rescue the resistance to the Venetoclax-HMA combination, we treated primary leukemia cells from relapsed AML patients with resistance to Venetoclax and azacytidine in vitro (Figure 6A,B). Relapsed AML samples showed reduced sensitivity to Venetoclax ( Figure 6A, IC50>3.2 μM and 2.8 μM, respectively), and the addition of WBC100 at 40 nM increased apoptosis sensitivity to Venetoclax with a synergistic effect (CI<1) (Figure 6B,C). These findings suggest that WBC100/VEN combination is still efficacious in the setting of Venetoclax-HMAs resistance. To further confirm these results in vivo, we next generated an orthotopic AML model with resistant MOLM-13-Luc cells and compared the efficacy between WBC100/VEN combination and AZA/VEN combination. After establishing AML models, mice were dosed daily with vehicle only, WBC100/VEN combination, and AZA-VEN combination, respectively. Weekly imaging quantitation of AML burden demonstrated an advantage of the WBC100/VEN combination over the AZA/VEN combination in reducing leukemia burden(p<0.0001 for WBC100-VEN vs AZA-VEN) ( Figure 6D,E). Drugs were withdrawn until vehicle-treated mice became moribund (≈14 days since the first treatment). Mice were followed and survival was evaluated by Kaplan-Meier analysis. As expected, WBC100/VEN combination provided a significant survival advantage (42 days) over the AZA-VEN combination (30 days) or the vehicle control group (21 days),    3). D) Serial bioluminescence images of mice bearing Kasumi-1-Luc cells after treatment with vehicle, WBC100, VEN, or the combination (n = 3 mice per group). E) Quantification of tumor loads of experimental mice during treatment (day 0 to day 20, n = 3). F) Quantification of tumor loads of experimental mice after drug withdrawal (day 20 to day 40, n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 as determined by one-way ANOVA; ns, not significant. consistent with tumor burden measurements ( Figure 6F). Of note, the AZA-VEN combination led to a 10% body weight loss within the first week, but WBC100/VEN combination did not exhibit obvious body weight loss ( Figure 6G).
These findings indicate that the novel oral small molecule combination of WBC100/VEN is highly active in R/R AML with ideal tolerability, and its efficacy is deeper and more durable as compared with the frontline azacitidine/VEN combination.

The Novel Combination of WBC100/VEN Potently Regresses Tumors In Patient-derived Xenografts Established from R/R AML Patients
To further confirm the efficacy of the WBC100/VEN combination and reveal its action mechanisms, we constructed patientderived xenograft transplantation models (PDX) by injecting primary cells from two patients with high c-Myc and Bcl-2 expression subcutaneously into NSG mice (AML patient #32 and AML patient #36). AML patient #32 showed a multi-gene mutation phenotype including FLT3-ITD, TET2, NPM1, and SRSF2 mutations. AML patient #36 harbored SF3B1, TP53 mutations with complex karyotype. Both tested samples were categorized into the adverse risk group according to the 2022 European LeukemiaNet (ELN) risk classification. After xenograft tumors reached ≈150 mm 3 , the mice were randomly assigned to four groups to receive the vehicle, WBC100, VEN, and WBC100/VEN combination for 15 days and then euthanized. Consistently, the WBC100/VEN combination was also highly effective in eliminating AML cells from patient #32, but the single-agent Venetoclax group and vehicle controls displayed rapid leukemia progression, as assessed by tumor volume and weight (Figure 7A,B). The tumor growth inhibition (TGI) rates in WBC100, VEN, and the WBC100/VEN combination groups were 62.42%, 37.09%, and 76.27%, respectively ( Figure 7A). Notably, in AML patient #36 bearing TP53 mutation with complex karyotype, the combination group showed almost complete loss of the tumor burden (tumor volume range = 0-4 mm 3 ) while the control group proliferated rapidly (tumor volume range = 91.2-557.5 mm 3 ) ( Figure 7C-E). Consistently, western blotting analysis revealed positive correlations between tumor reduction and c-Myc protein decreases ( Figure 7F), indicating that the combination of WBC100/VEN killed leukemia cells in tumor tissues via targeting c-Myc in vivo. Besides, tumor tissues were processed for H&E staining, immunostaining of Ki-67 and c-Myc, and CD34 staining for leukemic stem cells (LSCs). The vast majority of tumor cells in the combination-treated animals exhibited downregulated c-Myc protein levels, reduced proliferative activity, and attenuated CD34+ LSCs proportion ( Figure 7G). These results further demonstrate an impressive anti-leukemia activity of the WBC100/VEN combination against aggressive AML in PDX models.

Discussion
To date, AML still is a deadly hematological malignancy, and novel effective therapy is a highly unmet need. Our studies show that double over-expresser of c-Myc and Bcl-2 is a critical "genetic overdrive" of over-proliferation and apoptosis evasion in AML. We found that double over-expresser of c-Myc and Bcl-2 is frequently present in AML patients with poor genetic alterations and overall survival. Double-KD of c-Myc/Bcl-2 by shRNAs exhibited a significantly synergistic anti-leukemia effect both in vitro and in vivo.
Mechanically, we observed that Bcl-2 KD promoted mt DNA release from damaged mitochondrial into the cytosol and caused a lethal inflammatory phenotype. These results indicate that Bcl-2 KD disrupts mitochondria apoptosis machinery by inducing mitochondrial outer membrane permeabilization. c-Myc-KD impaired mitochondria biogenesis by abrogating mt DNA synthe-sis and disrupting mitochondrial translation. Double-KD of c-Myc/Bcl-2 showed a deep synergistic suppression of mitochondrial biogenesis and apoptosis machinery impairment in AML cells. These findings indicate that co-targeting c-Myc and Bcl-2 is an ideal therapeutic strategy for AML.
VEN is a potent orally active Bcl-2 inhibitor, but single-agent VEN has a limited anti-leukemia effect. The combination of VEN/AZA has yielded encouraging antileukemia activity, but its duration remains short and even less effective for R/R AML. Recently, we developed a novel oral small molecule c-Myc inhibitor with potent anti-tumor activity. Therefore, in this study, we attempted to develop a novel oral small molecule combination with WBC100 and oral Bcl-2 inhibitor VEN for co-targeting c-Myc/Bcl-2 in AML. We found that the novel combination of WBC100 and VEN exerted significantly synergistic lethality in AML cell lines and primary AML cells bearing genetic alterations, which were identified as essential target genes contributing to Venetoclax resistance and predictors of refractory response to VEN-AZA. [24] Importantly, the novel combination of WBC100/VEN induced deep and durable remissions in resistant AML models and PDX models without obvious side effects. Moreover, WBC100/VEN combination has more significant reductions in leukemia burden with long-term survival benefits as compared to the frontline VEN-AZA combination. In addition, this combination is an orally active small molecule combination and can be conveniently administered and provides timing and location of flexibility for patients.

Conclusion
Overall, our studies for the first time show that the double overexpresser of c-Myc/Bcl-2 is a critical overdrive of AML overproliferation and apoptosis evasion. Co-targeting c-Myc and Bcl-2 by the novel oral small molecule combination of WBC100/VEN can induce deep and durable remission for AML, and support its clinical trial, especially for R/R AML.

Experimental Section
Primary AML Cells: Leukemic primary patient samples used in vitro and in vivo were obtained after obtaining written informed consent under The Second Affiliated Hospital of Zhejiang University Ethics Committeeapproved guidelines (approval number 2022-0678). Mononuclear cells were isolated from peripheral blood or bone marrow samples by Ficoll density centrifugation and then cultured in Iscove-modified Dulbecco medium containing 15% fetal bovine serum.
Cell Lines: All leukemia cell lines were cultured in RPMI-1640 media supplemented with 10% fetal bovine serum (FBS) and 1% Pen/Strep at 37°C in a 5% CO2 humidified incubator. All cell lines were tested for authenticity by short tandem repeat (STR) profiling and routinely tested for mycoplasma.
Cell Proliferation and Drug Cytotoxicity Assays by MTT: MTT studies were performed as previously described [25] and were detailed in the supplemental methods. Briefly, cells were cultured in the presence of DOX or various concentrations of drugs. At indicated time points, cells were incubated with MTT for 4 h at 37°C. Lysis buffer was added and absorbance was measured at 562 nm after 16 h.
Confocal Laser Microscope Imaging: Cells were stained with 25 nM Mito-tracker and 10 μg mL −1 Hoechst for 30 min at 37°C. Cells were then washed twice with PBS. Images were recorded on Zeiss 710 laser scanning confocal microscope with a 63 × 1.40 oil immersion objective using Zeiss Transmission Electron Microscopy: MOLM-13 cells stably transduced with shRNAs were prefixed with 2.5% glutaraldehyde (4°C, overnight) and postfixed with 1% osmium tetroxide (room temperature,1.5 hours). The samples were then dehydrated in increasing concentrations of ethanol and 100% acetone. After that, samples were impregnated with graded concentrations of acetone/Spurr resin mixture[1:1 (vol/vol) acetone/Spurr resin for 1 h, 1:3 acetone/Spurr resin for 3 h, and 100% Spurr resin for 24 h] and then embedded in fresh resin to polymerize at 60°C. Tecnai G2 Spirit 120 kV transmission electron microscope (FEI Company) was used to image sample sections.
Detection of Mitochondrial Copy Number: Methods for extraction of total cellular DNA and cytosolic mtDNA could be found in supplemental materials. Mitochondrial copy number was normalized to nuclear DNA by quantitative real-time RT-PCR assay. More details and primer sequences for RT-PCR could be seen in supplemental methods.
Mouse Tumor Histopathology and IHC Staining: Tumors were fixed in 4% formalin; dehydrated and embedded in paraffin, and sectioned and processed for hematoxylin and eosin (H&E) staining. The sections were deparaffinized; heated for antigen retrieval, and incubated with the primary antibodies Ki-67, c-Myc, and CD34. The slides were scanned on a NanoZoomer (C13210-01).
Mouse Studies: Leukemia cell lines labeled with firefly luciferase were inoculated into the tail vein of NSG mice while patient-derived xenograft (PDX) cells were injected subcutaneously in the right flank of mice. For the growth inhibition experiment of MOLM-13 cells in vivo, mice received doxycycline (100 mg kg −1 ) by oral to induce shRNAs expressions. For combinational therapy, mice were dosed with vehicle (sterilized deionized water) or VEN (50 mg kg −1 , p.o. daily) and/or WBC100 (0.4 mg kg −1 , p.o. daily) and/or azacitidine (3 mg kg −1 , i.p. daily). Tumor BLI, body weight changes, tumor volume, and clinical symptoms were measured regularly until experiment termination. All animal experiments were approved by The Second Affiliated Hospital of Zhejiang University Ethics Committee (approval number AIRB-2021-1487).
Statistical Analysis: All statistical analyses were performed using GraphPad Prism, version 8.0. Data are represented as mean ± SD. Oneway ANOVA and Student's t-test were used to analyze differences. A pvalue < 0.05 was considered statistically significant.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.