EBD-300 is a third-generation variant built on the backbone of GpA. As can be seen in the surface representations of the tetramers of the clinical bacterial asparaginases from E. coli (EcA) and Erwinia (ErA) (Figure 1A) as well as the percent identity graphs (Figure 1B), the bacterial enzymes are highly dissimilar (<25% identity) to hA, which is in line with their high immunogenicity. In contrast, GpA shares over 70% sequence similarity with hA, providing a strong foundation for reducing immunogenicity. To further mitigate this risk, we engineered EBD-300 by first truncating it to remove the C-terminal domain, which is absent in bacterial type II asparaginases and may carry additional enzymatic activities. The first generation, EBD-100, was generated through a structure-guided process that identified surface residues that could be mutated to the corresponding hA sequence without loss of activity, increasing the percent amino-acid identity to hA from 69.8% for GpA to 81.7% for EBD-100. Subsequently, by using a T-cell proliferation assay that used overlapping peptides containing non-human residues with human PBMCs to see which peptides were able to elicit a T-cell response, we identified several residues with the potential to induce T-cell stimulation, and these were mutated to the hA sequence, resulting in EBD-200 with 83% identity to hA. We recently reported our evaluation of EBD-200 in blood and select solid tumors that showed high anti-cancer efficacy (13). Finally, we employed a mass-spectrometry analysis of naturally processed EBD-200 to identify epitopes presented on MHC-II. While these epitopes are not guaranteed to trigger a T-cell response, elimination of these potential epitopes would abrogate the potential for a T-cell proliferation response. Therefore, many of those epitopes were likewise mutated to the hA sequence, resulting in EBD-300 that has 87.1% identity to hA. Throughout this humanization and de-immunization process we made sure to assess the functionality of the variants, with a major focus on the asparagine Km value (low Km = high affinity for substrate). To achieve complete and persistent depletion of blood asparagine, the Km value needs to be in the low micromolar range. With EBD-300 we achieved a human-like asparaginase that has a Km for asparagine of 41.2 µM, which is between that of EcA (15.0 µM) and ErA (47.5 µM) – Table 1. The rate of the reaction (k_cat) is also between these two clinically used asparaginases.

EBD-300 is highly unique in that it is an asparaginase that has a low Km for, the kinetic property required for sustained depletion of this amino acid in the blood that gives asparaginase its anti-cancer effect in the clinic, while also demonstrating no off-target glutaminase activity, even at high concentrations of enzyme and excess glutamine substrate (Supplementary Figure 1; Table 1). This is in stark contrast to the bacterial clinical asparaginases of which ErA and EcA demonstrate strong and moderate glutaminase activity, respectively (Supplementary Figure 1; Table 1). In addition to having the required kinetic properties to efficiently deplete blood of asparagine without impacting the all-important glutamine (thereby eliminating side effects due to perturbations in glutamine levels and high ammonia production), an additional property of EBD-300 that favors clinical use is its extended in vivo persistence (Table 1). The very short half-life of the naked bacterial asparaginases demands high doses given frequently (three times per week). The seven-fold longer half-life of EBD-300 suggests that, when administered at a similar dose, a single weekly treatment will achieve and maintain full asparagine depletion, which would improve patient experience and compliance, leading to improved outcomes. With this unique asparaginase at hand, we next aimed to explore its effectiveness in AML.

We quantified ASNS levels in two chromosome 7-deleted cell lines, OCI-AML6 and UCSD-AML1. We also measured ASNS in two cell lines without chromosome 7 deletions: MOLM-13 (VEN-sensitive) and THP-1 (VEN-resistant) (Figure 2A). Of note, we performed fluorescence in situ hybridization (FISH) on OCI-AML6 and MOLM-13 cell lines and confirmed chromosome loss in OCI-AML6, but not MOLM-13 (Figure 2B). The cell lines with 7/7q deletions showed lower ASNS expression compared to those without the deletions (Figure 2A), consistent with gene dose effect.

We next evaluated the efficacy of EBD-300 in inducing cell death in AML cell lines with and without VEN using Cell Titer Glo (CTG) assay. The anti-leukemic activity of VEN, EBD-300, and their combination (VEN+EBD-300) varied among the cell lines. In MOLM-13 cells, the combination of VEN and EBD-300 (IC50 = 6.84 nM, 95% CI: 6.29 - 7.64, n=3, Figure 2C) enhanced sensitivity compared to VEN alone (IC50 = 8.66 nM, 95% CI: 5.83 - 12.76, n=3, Figure 2C). A strong synergistic effect was observed in THP-1 cells, a VEN-resistant model (14). The VEN IC50 was markedly reduced from 2268.8 nM (95% CI: 1528.3 - 3120.8, n=3, Figure 2C) to 953.9 nM (95% CI: 642.6 - 1416.1, n=3, Figure 2C) when combined with EBD-300. Conversely, OCI-AML6 cells were highly sensitive to EBD-300 monotherapy (IC50 = 0.000101 IU/ml, 95% CI: 0.000084 - 0.000117, n=3, Figure 2C). Given the profound sensitivity to EBD-300 monotherapy, no appreciable reduction in the VEN IC50 was observed with the combination therapy; the IC50 remained at 312.5 nM (95% CI: 291.6 - 351.6, n=4, Figure 2C), comparable to VEN alone (291.6 nM, 95% CI: 291.6 - 321.3, n=4, Figure 2C), suggesting that EBD-300 use precluded any further sensitization to VEN. Lastly, UCSD-AML1 did not exhibit sensitivity to EBD-300 monotherapy with an IC50 of 2.43 IU/mL (95% CI: 0.29 - 20.71, n = 3, Figure 2C); however, its combination with VEN reduced the IC50 of VEN from 2,570.3 nM (95% CI: 1,912.4 - 3,454.2, n = 3, Figure 2C) to 1,055.7 nM (95% CI: 904.6 - 1,231.9, n = 4, Figure 2C), suggesting an additive effect between VEN and EBD-300.

To assess the ability of EBD-300 to reduce clonogenic potential of AML primary patient cells, we performed methylcellulose-based Colony Forming Unit (CFU) assays on CD34+ enriched primary AML samples from three patients with deletion 7: Patient 1: (JAK2, KRAS, RUNX1; -7); Patient 2: (NRAS, PTPN11; -7), and Patient 3: (BRINP3L, GATA2, TP53, U2AF1; -7) (Figure 3A). We hypothesized that deletion 7 (where ASNS gene is located) would confer increased susceptibility of AML cells to EBD-300 given their dependency on exogeneous asparagine. Primary cells were cultured in methylcellulose and with DMSO, VEN (100 nM), EBD-300 (0.5 IU/mL), and a combination of VEN+EBD-300 for 14 days. For Patient 1, there was a significant reduction in CFUs with a decrease of 24.3% in the VEN group (p = 0.0066), 58.9% in the EBD-300 group (p < 0.0001), and 63.9% in the VEN+EBD-300 group (p = 0.0002) compared to the DMSO control, (Figure 3B; Supplementary Figure 2). In Patient 2, no colonies were observed, only clusters. Yet, a significant reduction in number of clusters was noted: 44.1% in the VEN group (p = 0.0055), 79.6% in the EBD-300 group (p = 0.0012), and 99.4% in the VEN+EBD-300 group (p = 0.0001) (Figure 3C; Supplementary Figure 3). Patient 3 exhibited a 24.3% reduction in CFUs with EBD-300 treatment and a 70.8% reduction with the VEN+EBD-300 combination compared to the DMSO control, while VEN alone did not result in a reduction in CFUs relative to the DMSO control (Figure 3D; Supplementary Figure 4). These findings suggest that EBD-300, either as monotherapy or in combination with VEN, can effectively reduce colony forming potential of primary leukemic cells an average of 54.4% in EBD-300 and 78.0% in VEN+EBD-300 groups. Of note, all patients had cytogenetic confirmation of deletion 7 suggesting heightened sensitivity of these cells to asparagine depletion.

To further test the correlation between loss of one ASNS allele on chromosome 7 and sensitivity to asparaginase, we first assessed the sensitivity of a PDX model (CTG-2456) with cytogenetics 46, XX, del(7)(q22q36). Sub-lethally irradiated female NOG mice were inoculated with 2x10⁶ CTG-2456 human AML cells. Mice were monitored post-AML inoculation to assess human AML engraftment using human-CD45 (hCD45), mouseCD45 (mCD45), human-CD33, human-CD3 antibodies and BD TruCountTM beads. When individual animals had ≥ 20% live human-CD45 cells in the bone marrow, they were randomized to vehicle control and EBD-300 (n = 10/group) and were treated intravenously for 28 days. EBD-300 was given at a dose of 750 IU/kg on Mondays and Wednesdays, and 1,500 IU/kg on Fridays to account for the extended interval of the weekend. Terminal blood and bone marrow samples were collected for flow cytometry on Day 28. An 8-color panel, which targeted human CD45+ leukocytes, AML monoblasts, and Leukemic Stem Blasts, was utilized to evaluate the anti-tumor activity of EBD-300. At Day 28, the human AML cells were mostly eradicated from both the blood and bone marrow (Figure 4A, p <0.0001 and Figure 4B, p<0.001). Mice treated with EBD-300 experienced moderate weight loss that allowed for uninterrupted treatments for the entire 28 days of the study (Figure 4C). This strong response to EBD-300 encouraged us to expand the studies to additional AML PDX models and to combination with VEN.

Following the demonstration of the invitro efficacy of VEN+EBD combination and in vivo EBD-300 single agent efficacy, we further tested the preclinical efficacy of VEN, EBD-300 and the combination in three PDX models with different mutations and cytogenetic profiles, described in Figure 5A. Briefly, following sublethal irradiation, immunodeficient NSG mice were injected with 1x10⁶ PDX cells. Engraftment was monitored weekly via measuring the hCD45 levels via flow cytometry in peripheral blood; and upon confirmation of engraftment, the mice were randomized into four groups for treatments: Vehicle control (VEH), Venetoclax (VEN), EBD and the combination of Venetoclax and EBD (VEN+EBD-300) as outlined in Figure 5B.

For PDX-1 (TP53-mutated; deletion 7; adverse risk), EBD-300 was administered at a dose of 750 IU/kg intraperitoneally twice a week for three weeks, while other treatments followed the dosages listed in Figure 5B. The average circulating hCD45 percentage was lower in the EBD-300 and VEN+EBD-300 groups, at 21.64% and 21.45% respectively, compared to 33.7% in the VEH group and 33.8% in the VEN group at week3 (Figure 5C). Notably, circulating hCD45% began to rise again by week 4 after the termination of treatment. Therefore, we resumed the EBD-300 treatment at week 5 to evaluate whether it could once again inhibit the progression of AML. After resuming treatment, circulating hCD45% further decreased to 47.0% in the EBD-300 group and 41.8% in the VEN+EBD-300 group by week 6, compared to 81.0% and 81.8% in the VEH and VEN groups (VEN+EBD-300 vs VEH, P<0.029, VEN+EBD-300 vs VEN, P<0.005, Figure 5C). Mice receiving EBD-300 (Groups EBD-300 and VEN+EBD) exhibited a 22.5% decrease in body weight by day 18 compared to the VEH group (Figure 5C, p <0.001). This is consistent with the asparaginase-based weight loss seen in murine studies (12). Moreover, the weight of mice in those groups recovered after halting treatment from Day 18 to Day 31, with an increase of 7.19% and 13.6% in the EBD-300 and VEN+EBD-300 groups, respectively. On Day 24, three mice from each group were sacrificed to assess the leukemia burden in the bone marrow, spleen, and liver. There were no significant differences in hCD45% across the VEN, EBD, and VEN+EBD-300 groups (Supplementary Figure 5). Nevertheless, the VEN+EBD-300 treatment group exhibited a prolonged median survival of 60 days, compared to 52 days, 54 days, and 49 days observed in the VEH, VEN, and EBD-300 groups, but this was not statistically significant (P = 0.364) (Supplementary Figure 5). This suggests that while the burden of the disease in the organs and the survival outcomes may not have been significantly different likely due to the aggressive nature of the disease but EBD-300 and VEN+EBD-300 still reduced the circulating hCD45%.

We hypothesized that a higher dose of EBD-300 may better control the disease. We thus tested PDX-2 (IDH1,NPM1;46,XX [20]), where we increased EBD-300 dosage to 1,500 IU/kg three times a week for three weeks in both the EBD-300 and VEN+EBD-300 groups. By Day 18, mice in the EBD-300 and VEN+EBD-300 groups exhibited a weight reduction of 21.2% and 13.4%, respectively, from baseline (Day 1) (P<0.0001) (Supplementary Figure 6A). Additionally, a significant reduction in circulating hCD45% was observed in the VEN and VEN+EBD-300 groups compared to the VEH group (p < 0.0001) (Figure 6A). On Day 24, three mice from each group were sacrificed, and hCD45% was assessed in the spleen, bone marrow, and liver (Figure 6A). There was a significant reduction in leukemia burden in the bone marrow where hCD45% levels were significantly lower in the VEN+EBD-300 group compared to the VEH group, specifically, 25.0% and 74.4%, respectively (P<0.0256) (Figure 6A). The VEN group had 58.9% and the EBD-300 group had 36.2% in bone marrow. Similarly, a significant reduction in leukemia burden was observed in the spleen in the VEN+EBD-300 group compared to the VEH group, (7.5% and 22.6%, respectively (P = 0.0302) (Figure 6A) The VEN and EBD-300 groups had a spleen leukemia burden of 19.9% and 15.4%, respectively. Furthermore, a significant decrease in leukemia burden was observed in the liver in both the EBD and VEN+EBD-300 groups compared to the VEH group, with hCD45% levels of 22.4% (P = 0.0051) and 11.0% (P = 0.0002), respectively, compared to 53.7% in the VEH group. The VEN group had an hCD45% of 54.7% (Figure 6A). Survival analysis for the remaining 5 mice in each group indicated no significant differences in the probability of survival between the VEN, EBD-300, VEN+EBD-300, and VEH groups (p = 0.266) (Supplementary Figure 6A).

Subsequently, we evaluated the drug efficacy in another PDX model, PDX-3, a deletion 7q model with IDH1 mutation. EBD-300 dosage of 1,500 IU/kg was planned to be administered three times a week for three weeks. However, due to the aggressive nature of this leukemia, some mice started dying during the treatment, and thus the treatment was terminated at the end of the second week, and mice were subsequently assessed for survival by Kaplan-Meier analysis. Day 11 post-treatment, the EBD-300 and VEN+EBD-300 groups experienced weight losses of 15.8% and 15.6%, respectively, compared to the baseline (p=0.0203 Supplementary Figure 6B). Moreover, reductions in circulating hCD45% were observed in EBD-300 and VEN+EBD-300 groups with an average circulating hCD45% of 48.0% and 42.8%, respectively, in comparison to 71.8% in VEH and 82.1% in VEN groups by week 2. Survival analysis for PDX-3 demonstrated a statistically significant but modest increase in median survival to 47 days in VEN+EBD in comparison to 45.5 days, 42.5 days, and 44 days in EBD-300 (HR: 0.9362,95% CI: 0.3395 to 2.582, p=0.0233), VEH (HR: 0.9681,95% CI: 0.3510 to 2.670, p=0.0173) and VEN (HR: 0.9043,95% CI: 0.3279 to 2.494, p=0.0054), respectively (Figure 6B), highlighting the efficacy of VEN+EBD-300 in prolonging survival even in aggressive, high risk PDX AML models.

As noted in myriad studies, asparaginase often causes severe weight loss in mice. A recent article correlated that weight loss to an in increase in growth differentiation factor 15 (GDF15), a hormone that impacts hunger (15). Hence, the weight loss observed in mice seems to be due to lack of appetite because of increased GDF-15 levels. Of note, even mice that have lost ~20% bodyweight under EBD-300 treatment were observed to be highly active and well groomed. Given that liver injury is a frequent toxicity of clinical asparaginases, we conducted a repeat dose toxicity study in CD-1 mice in which EBD-300 was administered once/week for 4 weeks at a dose of 2,000 IU/kg. At this dose, the nadir average asparaginase activity after 1 week was determined to be ~0.5 IU/ml, which is 5-fold higher than the accepted asparaginase activity of 0.1 IU/ml required for asparagine depletion. Clinical chemistry analysis revealed normal liver values, suggesting that in mice at this dose EBD-300 does not cause liver toxicity (Figure 7).

Primary human leukemia cells were acquired via an IRB protocol named “Multimodal analysis to dissect the tumor intrinsic and microenvironmental characteristics of myeloid neoplasms” with protocol number 2022–0576 at MD Anderson Cancer Center. Mononuclear cells were purified by Ficoll-Hypaque density centrifugation and washed with complete RPMI media containing 20% FBS. CD34⁺ AML hematopoietic stem/progenitor cells (HSPCs) were purified by immunomagnetic beads conjugated with anti-CD34 antibody (StemCell Technologies) prior to utilization in colony-forming unit (CFU) assays.

The human AML cell lines MOLM-13 and THP-1 were generously provided by Dr. Marina Konopleva from The University of Texas MD Anderson Cancer Center. OCI-AML6 and UCSD-AML1 were purchased from DSMZ (Germany). Viable cell counts were determined using trypan blue staining. Cell lines were incubated at 37 °C with 5% CO2 with saturating humidity. according to DSMZ recommendations. MOLM-13 and THP-1 were maintained in 90% RPMI 1640 supplemented with 10% heat-inactivated FBS and 1% Streptomycin/Penicillin. OCI-AML6 was cultured in 80% alpha-MEM (containing ribonucleosides and deoxyribonucleosides), 20% heat-inactivated FBS, and 20 ng/mL GM-CSF and 1% Streptomycin/Penicillin. UCSD-AML1 was cultured in 80% RPMI 1640, 20% heat-inactivated FBS, and 20 ng/mL GM-CSF and 1% Streptomycin/Penicillin. All cells were tested for mycoplasma contamination using an Enzyme-based detection assay. Moreover, human cell line authentication was confirmed via short tandem repeat (STR) analysis at the Cytogenetics and Cell Authentication Core at the University of Texas MD Anderson Cancer Center within 6 months of conducting the experiments.

For in vitro studies, VEN was purchased from Chemgood (Henrico, VA) in powder form, dissolved in DMSO to create 1–10 mM stock solutions, and stored at -80 °C. EBD-300 was supplied by Enzyme By Design (Chicago, Illinois) at a concentration of 400 IU/mL and stored at -80 °C.

For in vivo studies, EBD-300 was supplied by Enzyme By Design at 400 IU/mL and stored at -80 °C. It was diluted in the following sterile buffer solution: 25 mM Tris Buffer, pH 7.5; 200mM β-mercaptoethanol (BME) at the appropriate solution concentration. VEN powder was stored at 4 °C and then formulated fresh on each week of dosing at 10% DMSO, 30% PEG400, 60% Phosal PG50.

A CytoCell Del(7q) Deletion probe specifically designed to detect chromosomal deletions at regions 7q22.1-q22.2 and 7q31.2 on chromosome 7, was used for this study. It utilizes a red-labeled probe (Texas Red) that covers a 396kb region at 7q22.1-q22.2 (including the telomeric end of the RELN gene), and a green-labeled probe (FITC) that covers a 203kb region at 7q31.2 (including the TES gene).

The signal count was performed by manually scoring the distinct fluorescent spots in each cell nucleus, up to 200 nuclei per sample to ensure statistical accuracy (as per MD Anderson’s CLIA approved standards).

Cell lines were seeded into 96-well plates and treated with VEN, EBD, and their combination. Plates were incubated for 96 hours at 37 °C with 5% CO2 with saturating humidity. Subsequently, the plates were read using a Tecan Infinite plate reader (Tecan, Männedorf, Switzerland). Data were analyzed using GraphPad Prism Software (Graphpad, La Jolla, CA).

Cells were lysed in protein lysis buffer (0.25 mol/L Tris-HCl, 2% sodium dodecyl sulfate (SDS), 4% β-mercaptoethanol, 10% glycerol, 0.02% bromophenol blue; 0.2 × 106 million cells in 10 μL). Equal amounts of protein samples were resolved by 10% or 12% SDS–polyacrylamide gel electrophoresis (PAGE),and then transferred to nitrocellulose membrane. Immunoblotting was performed with primary antibodies: ASNS (from Cell Signaling Technology, #20843) with 1:1000 dilution and β-actin 1:15,000 dilution. The samples were then incubated in secondary antibodies for 1 hour and washed 3 times in 1× phosphate-buffered saline with Tween 20 (PBST). Blots were scanned with Film Imaging System.

Fresh primary AML peripheral blood samples were processed using the Ficoll density gradient centrifugation method as described previously. CD34+ cell enrichment was performed using the EasySep Human CD34 Positive Selection Kit from StemCell Technologies (#17856) following the manufacturer’s protocol. The enriched cells were resuspended in 2% IMDM medium at a concentration of 0.2 × 10⁶ cells/mL.

For the colony-forming unit (CFU) assays, 200,000 cells were plated in MethoCult H4435 medium from StemCell Technologies (#04435) in 12-well plates under four treatment conditions (VEH, VEN, EBD, VEN+EBD), with each condition tested in triplicate. Each well contained a total volume of 2000 μL methylcellulose medium, 100 μL cell suspension, and 100 μL drug suspension. The plates were incubated at 37 °C with 5% CO₂ and 95% humidity for 14 days. Colonies were subsequently counted using an inverted microscope. For CFU scoring, aggregates of ≥100 cells were classified as colonies, while aggregates containing <100 cells were recorded as clusters. CFU counts were performed in a blinded manner with respect to treatment group.

NOD/SCID mice were purchased from Jackson Laboratory (Sacramento, CA) and housed under standard conditions as approved by the 00002373-RN00 IACUC protocol at UT MD Anderson Cancer Center. Mice were randomized into treatment groups based on initial engraftment percentage, ensuring that each group had a balanced baseline engraftment range across cages. Randomization was performed by coded allocation to maintain concealment until treatment administration. Group sizes (n=8 mice per arm) were selected based on prior experience with AML PDX models, in which 3 mice were sacrificed on Day 24 to measure leukemia burden and other relevant markers, while the remaining 5 mice were followed for survival. This design provided sufficient power to detect treatment effects while minimizing animal use.

Humane endpoints were followed according to IACUC guidelines and included weight loss exceeding 20% of baseline, persistent recumbency, severe lethargy, inability to reach food, significant dehydration, dyspnea or abnormal breathing, and severe diarrhea. Mice meeting endpoint criteria were euthanized by CO₂ inhalation followed by cervical dislocation. Supportive care, such as hydration gel and softened diet, was provided by veterinary technicians as needed based on daily observation. No analgesia was used in this study, as no surgical procedures were performed. All mice meeting endpoint criteria were excluded from the study and reported. All animal procedures were performed in accordance with MD Anderson IACUC protocols.

Four-six-week-old female NOG mice were sub-lethally irradiated with 150 cGy whole-body irradiation (RS 2000 – X-ray Biological Irradiator, RadSources Technologies, Inc.) 4 hours prior to inoculation with AML cells from model CTG-2456. Irradiated mice were injected with 2 × 10⁶ viable human AML cells (0.2 mL in PBS) into the lateral tail vein. Starting two weeks after the cell inoculation, blood and bone marrow AML burden was analyzed by flow cytometry using a custom AML Flow panel at 1–3 time points prior to the estimated engraftment window based upon previously known engraftment kinetics. Once sufficient engraftment levels were observed in surrogate animals (%hCD45+ of live cells averages at ≥20%), the remaining pre-study animals were randomized to a vehicle control and EBD-300 groups with animals of similar body weight (n=10/group). EBD-300 was given iv at a dose of 750 IU/kg on Mondays and Wednesdays, and 1,500 IU/kg on Friday, for 28 days. Mice were monitored daily via clinical observations. Body weights were taken 3 times during the week post-implant and once weekly thereafter. At Day 28, as much whole blood as possible was collected via cardiac puncture from all groups. Blood was transferred to K2EDTA tubes and gently mixed by inversion (by hand) 8–10 times. After inversion, sample tubes were stored on wet ice until processed for flow analysis using the Standard AML panel. Bone marrow was collected for flow analysis by flushing the marrow through the tibia and femur from both legs with MACS media and pooled into collection tubes for each mouse. Samples were stored on wet ice until processed for flow cytometry.

Viable frozen PDX cells were thawed and washed with AlphaMEM supplemented with 20% fetal bovine serum (FBS). The following thawing medium was then prepared: 10 mL of warm AlphaMEM with 20% FBS, 200 μL of heparin, 200 μL of DNase, and 500 μL of MgSO₄. The thawed PDX cells were incubated in this medium for 15 minutes at 37 °C. Following incubation, cells were centrifuged at 1500 rpm for 5 minutes, and the supernatant was discarded. The cells were then washed with 10 mL of PBS and filtered through a cap filter flow tube. After filtering, the cells were counted and prepared for injection. A total of 32 female NSG mice, aged 6–8 weeks, were used for the study. The mice were subjected to sublethal irradiation at 250 cGy 24 hours before cell injection. Each mouse was injected with 1 × 10⁶ PDX cells. After confirming robust engraftment, the mice were randomized into four groups of eight. Treatment began on the day of randomization with the following regimens: VEH control, Venetoclax (VEN) administered orally at 100 mg/kg, 5 days per week for two weeks, EBD-300 administered intraperitoneally at 750–1500 IU/kg, two to three times a week for three weeks, or a combination of VEN and EBD. Engraftment was monitored weekly by collecting peripheral blood and assessing human CD45 (hCD45) and murine CD45 (mCD45) levels. On Day 24, three mice from each group were sacrificed to measure leukemia burden in the spleen, bone marrow, and liver using hCD45 and mCD45 markers. The rest of the mice (5 mice per group) were monitored for survival.

Female CD-1 mice (n=5) were injected i.p. with EBD-300 at a dose of 2,000 IU/kg once a week for 4 doses. After the last dose was given, the mice were fasted for 18 hours and then sacrificed, and blood was collected for clinical chemistry analysis.

For the viability and Annexin V/PI assays conducted in AML cell lines, data are presented as the standard error of the mean (SEM), with a significance level set at p<0.05. Dose-response curves were generated, and IC50 values were calculated by nonlinear regression analysis using a four-parameter logistic (4PL) model. Each value represents the result of 3 to 4 independent biological replicates. To compare the viability percentages among different treatment conditions, a one-way ANOVA followed by Tukey’s post hoc test was performed. Similarly, for the colony-forming unit (CFU) assays, a one-way ANOVA with Tukey’s post hoc test was used.

For the weight chart, which tracks the weights of each mouse in the different groups (VEH VEN, EBD-300, VEN+EBD-300) over the study timeline, a two-way ANOVA followed by Tukey’s post hoc test was utilized.

Lastly, mouse survival was analyzed using the Kaplan-Meier method, and comparisons between the four treatment groups were made. All statistical parameters, including sample sizes and p-values, are reported in the figures and their corresponding legends.

Healthy wild-type CD-1 female mice (8–10 weeks) were injected intravenously with 4,000 IU/kg EBD-300 or intraperitoneally with either 2,500 IU/kg Spectrila (EcA) or 7,500 IU/kg ErA. For the activity determinations, 50 µL of peripheral blood was collected in EDTA-coated tubes at regular intervals: 10 time points were collected over 120 hours for EBD-300; 6 time points were collected over 24 hours for EcA; and 8 time points were collected over 10 hours for ErA. Samples were processed by centrifugation at 4 °C at 4,600 rpm (2000xg) for 10 min. The plasma was either tested immediately for asparaginase activity or flash-frozen in liquid nitrogen and stored at -80 °C.