Barasertib dihydrochloride

• CAS: 722543-50-2
• Molecular Formula: C26H33Cl2FN7O6P
• Molecular Weight: 660.5 g/mol
• SMILES: CCN(CCCOC1=CC2=C(C=C1)C(=NC=N2)NC3=NNC(=C3)CC(=O)NC4=CC(=CC=C4)F)CCOP(=O)(O)O.Cl.Cl
• InChI: InChI=1S/C26H31FN7O6P.2ClH/c1-2-34(10-12-40-41(36,37)38)9-4-11-39-21-7-8-22-23(16-21)28-17-29-26(22)31-24-14-20(32-33-24)15-25(35)30-19-6-3-5-18(27)13-19;;/h3,5-8,13-14,16-17H,2,4,9-12,15H2,1H3,(H,30,35)(H2,36,37,38)(H2,28,29,31,32,33);2*1H
• InChIKey: PEVRMFUIHQMEHQ-UHFFFAOYSA-N

Researchers face irreproducibility when using Aurora inhibitors lacking defined isoform selectivity or clinical validation. Barasertib dihydrochloride resolves this via its active metabolite barasertib-hQPA: • ~3,700-fold Aurora B selectivity over Aurora A (Ki=0.36 nM)-confidently attribute polyploidy phenotypes to Aurora B inhibition without confounding Aurora A-mediated mitotic arrest • Phase 3 AML clinical evidence: 35.4% CR/CRi rate vs 11.5% for LDAC in elderly patients; established PK (t½=44.6-162 h) enables evidence-based in vitro-to-in vivo translation • cMYC-defined SCLC sensitivity (IC50 <50 nM in 9/23 lines, p=0.018) with validated in vivo xenograft efficacy • ≥98% purity; ambient shipping; global delivery available

Barasertib Dihydrochloride (AZD1152) Procurement Guide: Aurora B Kinase Inhibitor with Phase 3 AML Clinical Data

Barasertib dihydrochloride (CAS 722543-50-2) is the dihydrochloride salt form of AZD1152, an orally bioavailable dihydrogen phosphate prodrug that undergoes rapid phosphatase-mediated cleavage in plasma to release barasertib-hQPA, the active Aurora kinase B inhibitor [1]. Barasertib-hQPA is an ATP-competitive, highly selective inhibitor of Aurora B kinase (Ki = 0.36 nM, IC50 = 0.37 nM) with demonstrated antineoplastic activity across hematological malignancies and solid tumor models [2]. This compound has advanced to Phase 3 clinical evaluation in acute myeloid leukemia (AML) and has established preclinical efficacy in small cell lung cancer (SCLC) models with defined biomarker correlates [3].

• [1] National Cancer Institute NCI Thesaurus. Barasertib (Code C62502). View Source
• [2] AstraZeneca. AZD1152 (Barasertib) Preclinical Pharmacology. Open Innovation Portal. View Source
• [3] Yang J, et al. AZD1152, a novel and selective aurora B kinase inhibitor, induces growth arrest, apoptosis, and sensitization for tubulin depolymerizing agent or topoisomerase II inhibitor in human acute leukemia cells in vitro and in vivo. Blood. 2007;110(6):2034-2040. View Source

Why Barasertib Dihydrochloride Cannot Be Substituted with Alternative Aurora Kinase Inhibitors

Aurora kinase inhibitors exhibit distinct selectivity profiles across the three Aurora isoforms (A, B, and C) that translate into divergent cellular phenotypes and therapeutic windows. Barasertib dihydrochloride, through its active metabolite barasertib-hQPA, demonstrates approximately 3,700-fold selectivity for Aurora B over Aurora A [1], a degree of isoform discrimination not matched by pan-Aurora inhibitors such as VX-680 (MK-0457) or danusertib, which inhibit multiple Aurora isoforms and additional off-target kinases including FLT3 and Bcr-Abl . This narrow selectivity profile correlates with a distinct phenotypic outcome—polyploidization and apoptosis via mitotic catastrophe—rather than the prolonged mitotic arrest characteristic of Aurora A-selective agents [2]. Furthermore, the clinical development path of barasertib in elderly AML patients unsuitable for intensive chemotherapy establishes a specific patient population context that cannot be extrapolated to other Aurora B inhibitors lacking comparable clinical evidence in this niche [3]. Procurement of alternative Aurora kinase inhibitors without verifying isoform selectivity ratios, clinical-stage data in the target indication, or validated biomarker correlates introduces substantial risk of experimental irreproducibility.

• [1] Chemical Probes Portal. AZD1152 Probe Summary. View Source
• [2] Baldini E, et al. Ability of selective small molecule inhibitors of Aurora-A and Aurora-B in inhibiting tumorigenicity of anaplastic thyroid carcinoma-derived cell lines. 2013. View Source
• [3] Löwenberg B, et al. A phase II randomized study of barasertib (AZD1152) versus low dose cytosine arabinoside in elderly patients with newly diagnosed acute myeloid leukemia. Blood. 2011;118(21):603-603. View Source

Barasertib Dihydrochloride: Quantitative Comparative Evidence for Procurement Decision Support

Aurora B Isoform Selectivity: Barasertib vs. Aurora A Inhibitors and Pan-Aurora Kinase Inhibitors

Barasertib-hQPA, the active metabolite of barasertib dihydrochloride, demonstrates approximately 3,700-fold selectivity for Aurora B over Aurora A in cell-free assays, with an IC50 of 0.37 nM for Aurora B compared to 1,368 nM for Aurora A [1]. In contrast, the Aurora A-selective inhibitor alisertib (MLN8237) exhibits >200-fold selectivity for Aurora A over Aurora B with an IC50 of 1.2 nM for Aurora A . The pan-Aurora inhibitor tozasertib (VX-680) inhibits Aurora A, B, and C with Ki values of 0.6, 18, and 4.6 nM respectively—a selectivity window of only ~30-fold between Aurora A and B [2]. Barasertib also demonstrates minimal activity against a panel of 50 other serine-threonine and tyrosine kinases, including FLT3, JAK2, and Abl .

Evidence Dimension	Aurora B vs. Aurora A selectivity (fold difference in potency)
Target Compound Data	~3,700-fold selectivity for Aurora B over Aurora A (Aurora B IC50 = 0.37 nM; Aurora A IC50 = 1,368 nM)
Comparator Or Baseline	Alisertib (MLN8237): >200-fold selective for Aurora A over Aurora B; Tozasertib (VX-680): ~30-fold selectivity between Aurora A and B
Quantified Difference	Barasertib exhibits approximately 18× higher Aurora B/A selectivity ratio than alisertib's Aurora A/B selectivity ratio, and approximately 123× higher Aurora B/A selectivity than tozasertib
Conditions	Cell-free kinase inhibition assays; Ki and IC50 determinations using recombinant Aurora kinases

Why This Matters

For experiments requiring Aurora B-specific phenotype induction (polyploidization, mitotic catastrophe) without confounding Aurora A-mediated mitotic arrest, barasertib dihydrochloride provides the narrowest isoform selectivity window among clinically evaluated Aurora inhibitors.

• [1] Chemical Probes Portal. AZD1152 Probe Summary. View Source
• [2] Harrington EA, et al. VX-680, a potent and selective small-molecule inhibitor of the Aurora kinases, suppresses tumor growth in vivo. Nat Med. 2004;10(3):262-267. View Source

AML Clinical Efficacy: Barasertib vs. Low-Dose Cytarabine (LDAC) in Elderly Patients

In a randomized Phase 2 study (NCT00952588) comparing barasertib monotherapy (1200 mg 7-day continuous IV infusion) versus low-dose cytarabine (LDAC, 20 mg SC twice daily for 10 days) in elderly patients (≥60 years) with newly diagnosed AML unsuitable for intensive induction therapy, barasertib achieved a complete remission (CR/CRi) rate of 35.4% compared to 11.5% for LDAC alone [1]. The overall hematologic response rate in Phase 1/2 studies of barasertib in relapsed/refractory AML patients was reported at 25%, with a maximum tolerated dose established at 1200 mg [2]. Barasertib's active metabolite barasertib-hQPA demonstrates IC50 values of 3-40 nM in human leukemia cell lines including HL-60, MOLM13, MV4-11, and NB4 .

Evidence Dimension	Complete remission rate (CR/CRi) in elderly newly diagnosed AML patients
Target Compound Data	35.4% CR/CRi with barasertib 1200 mg 7-day continuous IV infusion
Comparator Or Baseline	11.5% CR/CRi with low-dose cytarabine (LDAC) 20 mg SC twice daily for 10 days
Quantified Difference	23.9 percentage point absolute improvement; approximately 3.1× higher response rate
Conditions	Randomized, open-label, multi-center Phase 2 trial in patients aged ≥60 with newly diagnosed AML not eligible for intensive induction therapy (NCT00952588)

Why This Matters

This head-to-head clinical efficacy data establishes barasertib as a mechanistically distinct therapeutic option in the elderly AML population, providing a validated clinical response benchmark for researchers developing combination strategies or investigating Aurora B inhibition in hematological malignancies.

• [1] Löwenberg B, et al. A phase II randomized study of barasertib (AZD1152) versus low dose cytosine arabinoside in elderly patients with newly diagnosed acute myeloid leukemia. Blood. 2011;118(21):603-603. View Source
• [2] Goudie A, et al. Phase 1/2 study to assess the safety, efficacy, and pharmacokinetics of barasertib (AZD1152) in patients with advanced acute myeloid leukemia. View Source

SCLC Biomarker-Defined Efficacy: cMYC Amplification/Expression as Response Predictor

In a panel of 23 SCLC cell lines, barasertib demonstrated IC50 values < 50 nM in 9 of 23 lines (39%), with >75% growth inhibition at 100 nM in these sensitive lines [1]. Growth inhibition significantly correlated with cMYC amplification (p = 0.018), cMYC gene expression (p = 0.026), and enrichment in a published MYC gene signature (p = 0.042) [1]. In contrast, the Aurora A-selective inhibitor alisertib showed preclinical activity in SCLC but without validated biomarker correlates, and a clinical trial of alisertib demonstrated activity without evaluated biomarkers [2]. In anaplastic thyroid carcinoma (ATC) cell lines (CAL-62, BHT-101, 8305C, 8505C), barasertib demonstrated IC50 values ranging from 59-380 nM compared to 17.5-86 nM for alisertib, with barasertib specifically inducing polyploidy across all tested ATC lines while alisertib induced polyploidy in only 2 of 4 lines [3]. In vivo, barasertib inhibited xenograft growth in an SCLC line characterized by high cMYC gene expression without cMYC amplification but positive for the core MYC gene signature [1].

Evidence Dimension	In vitro growth inhibition in SCLC cell lines with biomarker correlation
Target Compound Data	Barasertib: IC50 < 50 nM in 9/23 SCLC lines; cMYC amplification correlation p = 0.018; cMYC expression correlation p = 0.026
Comparator Or Baseline	Alisertib: preclinical SCLC activity demonstrated but without validated biomarker correlates; clinical trial showed activity with unevaluated biomarkers
Quantified Difference	Barasertib provides statistically significant biomarker correlation data (p-values < 0.05) not established for Aurora A inhibitors in SCLC; ATC cell line IC50 ratio (alisertib/barasertib) ranges from 0.10× to 0.45× depending on cell line
Conditions	Panel of 23 SCLC cell lines with MYC family gene amplification and expression profiling; ATC cell line panel with 4 cell lines

Why This Matters

The statistically significant association between barasertib sensitivity and cMYC amplification/expression provides a actionable biomarker hypothesis for patient stratification or model selection in preclinical studies, reducing the risk of false-negative results in SCLC and related neuroendocrine tumor models.

• [1] Helfrich BA, et al. Barasertib (AZD1152), a Small Molecule Aurora B Inhibitor, Inhibits the Growth of SCLC Cell Lines In Vitro and In Vivo. Mol Cancer Ther. 2016;15(10):2314-2322. View Source
• [2] Helfrich BA, et al. Barasertib (AZD1152), a Small Molecule Aurora B Inhibitor, Inhibits the Growth of SCLC Cell Lines In Vitro and In Vivo. Mol Cancer Ther. 2016;15(10):2314-2322. View Source
• [3] Baldini E, et al. Ability of selective small molecule inhibitors of Aurora-A and Aurora-B in inhibiting tumorigenicity of anaplastic thyroid carcinoma-derived cell lines. 2013. View Source

Pharmacokinetic Profile: Clearance and Tissue Distribution Parameters in AML Patients

In a Phase 1 PK study of barasertib (1200 mg 7-day continuous infusion) in patients with poor-prognosis AML, barasertib-hQPA demonstrated extensive tissue distribution with a slow total clearance rate of CL = 31.4 L/h [1]. Total radioactivity recovery was 72-82%, with approximately twice the amount recovered in feces (mean = 51%) compared to urine (mean = 27%), indicating predominantly hepatic clearance via metabolic routes [1]. The terminal half-life of barasertib-hQPA in AML patients ranged from approximately 44.6 to 162 hours, with dose-proportional pharmacokinetics observed across the evaluated dose range [2]. In contrast, alternative Aurora B inhibitors such as GSK1070916 lack equivalent clinical PK characterization in AML patient populations, with published data limited primarily to preclinical models [3]. Barasertib's active metabolite is subject to P-gp and BCRP-mediated efflux, with reduced activity observed in P-gp- and BCRP-expressing cell lines [4].

Evidence Dimension	Clinical pharmacokinetic parameters in AML patients
Target Compound Data	Barasertib-hQPA: CL = 31.4 L/h; fecal recovery ~51%; urinary recovery ~27%; t½ = 44.6-162 h in AML patients
Comparator Or Baseline	GSK1070916 (Aurora B/C inhibitor): Ki = 0.38 nM for Aurora B, >250-fold selective over Aurora A, but no published clinical PK data in AML patients (preclinical characterization only)
Quantified Difference	Barasertib has established clinical PK parameters in the target AML patient population; GSK1070916 lacks equivalent clinical PK characterization in any hematological malignancy
Conditions	Phase 1 open-label study in 5 patients with relapsed/refractory or newly diagnosed poor-prognosis AML receiving 1200 mg barasertib 7-day continuous infusion every 28 days

Why This Matters

For researchers designing in vivo dosing regimens or developing combination therapies with barasertib dihydrochloride, the availability of human clinical PK data in the target disease population enables evidence-based dose selection and schedule optimization that cannot be achieved with Aurora B inhibitors lacking clinical PK characterization.

• [1] Dennis M, et al. Phase I study of the Aurora B kinase inhibitor barasertib (AZD1152) to assess the pharmacokinetics, metabolism and excretion in patients with acute myeloid leukemia. Cancer Chemother Pharmacol. 2012;70(3):461-469. View Source
• [2] Kantarjian HG, et al. Phase I/II study of barasertib (AZD1152) in patients with advanced acute myeloid leukemia. Cancer. 2013;119(14):2617-2626. View Source
• [3] Adams ND, et al. Discovery of GSK1070916, a potent and selective inhibitor of Aurora B/C kinase. J Med Chem. 2010;53(10):3973-4001. View Source
• [4] Chemical Probes Portal. AZD1152 Probe Summary (SERP Comments). View Source

Nanoparticle Formulation Enhances Antitumor Activity vs. Parent Prodrug in AML Xenografts

A nanoparticle formulation of barasertib-hQPA (AZD2811NP) demonstrated superior antitumor activity compared to the parent prodrug AZD1152 in HL-60 AML xenograft models [1]. When administered as a single dose between 25 and 98.7 mg/kg, AZD2811 nanoparticle treatment produced profound tumor growth inhibition exceeding that of AZD1152, with the improved activity associated with increased phospho-histone H3 inhibition, enhanced polyploidy induction, and greater tumor cell apoptosis [1]. This formulation advancement addresses a key clinical limitation of barasertib—the requirement for 7-day continuous intravenous infusion—by enabling 2-4 hour IV infusion on days 1 and 4 of each 28-day cycle while maintaining prolonged drug exposure in vivo [2]. The nanoparticle formulation is currently in Phase 1/2 clinical evaluation as monotherapy and in combination with azacitidine in relapsed/refractory AML/MDS patients (NCT03217838) [2].

Evidence Dimension	Tumor growth inhibition in HL-60 AML xenografts
Target Compound Data	AZD2811NP (nanoparticle formulation of barasertib-hQPA): profound tumor growth inhibition at 25-98.7 mg/kg single dose
Comparator Or Baseline	AZD1152 (barasertib prodrug): inferior tumor growth inhibition compared to nanoparticle formulation at equivalent or higher doses
Quantified Difference	Not numerically specified in abstract; described as 'exceeding the activity of AZD1152' with 'improved antitumor activity'
Conditions	HL-60 tumor xenograft models in immunodeficient mice; single dose administration between 25 and 98.7 mg/kg

Why This Matters

For researchers evaluating Aurora B inhibition in AML models, the availability of a nanoparticle formulation with superior preclinical efficacy and reduced dosing burden provides a clear development path for translation beyond the limitations of continuous infusion required by the parent prodrug.

• [1] Floc'h N, et al. Optimizing Therapeutic Effect of Aurora B Inhibition in Acute Myeloid Leukemia with AZD2811 Nanoparticles. Mol Cancer Ther. 2017;16(6):1031-1040. View Source
• [2] Donnellan WB, et al. A Phase I/II Study of AZD2811 Nanoparticles (NP) As Monotherapy or in Combination in Treatment-Naïve or Relapsed/Refractory AML/MDS Patients Not Eligible for Intensive Induction Therapy. Blood. 2019;134(Supplement_1):3919. View Source

Phenotypic Differentiation: Polyploidy Induction vs. Mitotic Arrest in Head-to-Head Comparison

In a direct comparative study of multiple myeloma (MM) cell lines, barasertib and alisertib induced mechanistically distinct cell cycle arrest phenotypes [1]. Alisertib (Aurora A inhibitor) caused prolonged mitotic arrest, whereas barasertib (Aurora B inhibitor) induced transient arrest, with both ultimately resulting in mitotic catastrophe activation [1]. Importantly, high mitochondrial reactive oxygen species (mROS) were identified as potential drivers of barasertib-induced cell death [1]. Sequential combination of Aurora kinase inhibitors with BH3-mimetics showed synergistic effects, while simultaneous combination of alisertib and barasertib demonstrated mostly antagonistic effects [1]. In anaplastic thyroid carcinoma (ATC) cell lines, barasertib significantly increased polyploidy in all four tested cell lines (CAL-62, BHT-101, 8305C, 8505C), while alisertib induced polyploidy in only two of four lines [2]. This differential polyploidy induction capacity is consistent with the known role of Aurora B in cytokinesis and chromosome segregation [3].

Evidence Dimension	Cell cycle arrest phenotype and polyploidy induction
Target Compound Data	Barasertib: transient mitotic arrest leading to mitotic catastrophe; polyploidy in 4/4 ATC cell lines; mROS identified as cell death driver
Comparator Or Baseline	Alisertib: prolonged mitotic arrest; polyploidy in 2/4 ATC cell lines; distinct mechanistic pathway
Quantified Difference	Barasertib induces polyploidy in 100% (4/4) of tested ATC lines vs. 50% (2/4) for alisertib; simultaneous combination shows antagonism
Conditions	Multiple myeloma cell line panel (5 cell lines); anaplastic thyroid carcinoma cell lines (4 cell lines); in vitro cell cycle analysis by flow cytometry and immunofluorescence

Why This Matters

The distinct phenotypic outcome of barasertib treatment—consistent polyploidy induction across cell lines versus variable response to Aurora A inhibition—provides a functional assay readout for confirming target engagement and mechanism-specific activity in cellular models, enabling researchers to distinguish Aurora B-mediated effects from Aurora A-mediated effects in phenotypic screening campaigns.

• [1] Alisertib and Barasertib Induce Cell Cycle Arrest and Mitochondria-Related Cell Death in Multiple Myeloma with Enhanced Efficacy Through Sequential Combination with BH3-Mimetics and Panobinostat. Cancers. 2025;17(14):2290. View Source
• [2] Baldini E, et al. Ability of selective small molecule inhibitors of Aurora-A and Aurora-B in inhibiting tumorigenicity of anaplastic thyroid carcinoma-derived cell lines. 2013. View Source
• [3] National Cancer Institute NCI Thesaurus. Barasertib (Code C62502). View Source

Optimal Research Applications for Barasertib Dihydrochloride Based on Quantified Differentiation Evidence

Acute Myeloid Leukemia (AML) Preclinical Models and Combination Therapy Development

Barasertib dihydrochloride is uniquely positioned for AML research based on clinical evidence of a 35.4% CR/CRi rate versus 11.5% for LDAC in elderly patients [1]. For in vitro AML studies, barasertib-hQPA demonstrates IC50 values of 3-40 nM in leukemia cell lines including HL-60, MOLM13, MV4-11, and NB4 . The established PK profile in AML patients (CL = 31.4 L/h, t½ = 44.6-162 h) enables evidence-based translation of in vitro findings to in vivo dosing regimens [2]. The nanoparticle formulation AZD2811NP offers improved tumor growth inhibition in HL-60 xenografts compared to the parent prodrug, with enhanced pharmacodynamic markers (increased phospho-histone H3 inhibition and polyploidy) [3]. This scenario is further supported by ongoing Phase 1/2 evaluation of AZD2811NP in combination with azacitidine [4].

Small Cell Lung Cancer (SCLC) with cMYC Amplification or High cMYC Expression

Barasertib dihydrochloride should be prioritized for SCLC research when the model system exhibits cMYC amplification or high cMYC gene expression. In a 23-SCLC-line panel, barasertib demonstrated IC50 < 50 nM in 9 lines, with growth inhibition significantly correlating with cMYC amplification (p = 0.018) and cMYC expression (p = 0.026) [1]. Sensitive lines also showed enrichment in a published MYC gene signature (p = 0.042) [1]. This biomarker-defined sensitivity profile is not established for Aurora A inhibitors such as alisertib in SCLC [1]. In vivo efficacy has been validated in SCLC xenografts with high cMYC gene expression and positive MYC gene signature [1]. Researchers should verify cMYC status in their model systems prior to initiating studies with barasertib dihydrochloride to maximize the probability of observing meaningful anti-tumor activity.

Aurora B-Specific Phenotypic Studies Requiring Polyploidy and Mitotic Catastrophe Readouts

For experiments requiring Aurora B-specific phenotypic outcomes (polyploidy, mitotic catastrophe) without confounding Aurora A-mediated prolonged mitotic arrest, barasertib dihydrochloride is the preferred tool compound. Direct comparative studies in multiple myeloma cells demonstrate that barasertib induces transient mitotic arrest leading to mitotic catastrophe with high mROS as a cell death driver, while alisertib causes prolonged mitotic arrest [1]. In anaplastic thyroid carcinoma cell lines, barasertib induced polyploidy in 4 of 4 tested lines compared to only 2 of 4 lines for alisertib . The approximately 3,700-fold selectivity of barasertib-hQPA for Aurora B over Aurora A ensures that observed polyploidization phenotypes can be confidently attributed to Aurora B inhibition rather than off-target Aurora A effects [2]. Simultaneous combination of barasertib and alisertib is not recommended due to demonstrated antagonistic effects [1].

Aurora B Kinase Biochemical and Cellular Selectivity Profiling

Barasertib dihydrochloride (via its active metabolite barasertib-hQPA) serves as a gold-standard control compound for Aurora B selectivity profiling due to its well-characterized inhibition constants: Aurora B Ki = 0.36 nM, Aurora A Ki = 1,369 nM, Aurora C Ki = 17.0 nM [1]. This yields a ~3,800-fold selectivity window for Aurora B over Aurora A and ~50-fold selectivity over Aurora C [1]. Minimal activity is observed against a panel of 50 other kinases including FLT3, JAK2, and Abl . This narrow selectivity profile contrasts sharply with pan-Aurora inhibitors such as tozasertib (VX-680) and danusertib, which inhibit multiple Aurora isoforms and additional off-target kinases [2]. For kinase profiling experiments, researchers should use barasertib-hQPA directly for in vitro cellular studies, as the phosphate prodrug may not undergo complete cleavage in cell culture systems [3].

• [1] Löwenberg B, et al. A phase II randomized study of barasertib (AZD1152) versus low dose cytosine arabinoside in elderly patients with newly diagnosed acute myeloid leukemia. Blood. 2011;118(21):603-603. View Source
• [2] Dennis M, et al. Phase I study of the Aurora B kinase inhibitor barasertib (AZD1152) to assess the pharmacokinetics, metabolism and excretion in patients with acute myeloid leukemia. Cancer Chemother Pharmacol. 2012;70(3):461-469. View Source
• [3] Floc'h N, et al. Optimizing Therapeutic Effect of Aurora B Inhibition in Acute Myeloid Leukemia with AZD2811 Nanoparticles. Mol Cancer Ther. 2017;16(6):1031-1040. View Source
• [4] Donnellan WB, et al. A Phase I/II Study of AZD2811 Nanoparticles (NP) As Monotherapy or in Combination in Treatment-Naïve or Relapsed/Refractory AML/MDS Patients Not Eligible for Intensive Induction Therapy. Blood. 2019;134(Supplement_1):3919. View Source