Magrolimab in Combination With Azacitidine in Patients With Higher-Risk Myelodysplastic Syndromes: Final Results of a Phase Ib Study

PURPOSE

Magrolimab is a monoclonal antibody that blocks cluster of differentiation 47, a don't-eat-me signal overexpressed on cancer cells. Cluster of differentiation 47 blockade by magrolimab promotes macrophage-mediated phagocytosis of tumor cells and is synergistic with azacitidine, which increases expression of eat-me signals. We report final phase Ib data in patients with untreated higher-risk myelodysplastic syndromes (MDS) treated with magrolimab and azacitidine (ClinicalTrials.gov identifier: NCT03248479).

PATIENTS AND METHODS

Patients with previously untreated Revised International Prognostic Scoring System intermediate-/high-/very high-risk MDS received magrolimab intravenously as a priming dose (1 mg/kg) followed by ramp-up to a 30 mg/kg once‐weekly or once‐every-2-week maintenance dose. Azacitidine 75 mg/m² was administered intravenously/subcutaneously once daily on days 1-7 of each 28-day cycle. Primary end points were safety/tolerability and complete remission (CR) rate.

RESULTS

Ninety-five patients were treated. Revised International Prognostic Scoring System risk was intermediate/high/very high in 27%, 52%, and 21%, respectively. Fifty-nine (62%) had poor-risk cytogenetics and 25 (26%) had TP53 mutation. The most common treatment-emergent adverse effects included constipation (68%), thrombocytopenia (55%), and anemia (52%). Median hemoglobin change from baseline to first postdose assessment was −0.7 g/dL (range, −3.1 to +2.4). CR rate and overall response rate were 33% and 75%, respectively. Median time to response, duration of CR, duration of overall response, and progression-free survival were 1.9, 11.1, 9.8, and 11.6 months, respectively. Median overall survival (OS) was not reached with 17.1-month follow-up. In TP53-mutant patients, 40% achieved CR with median OS of 16.3 months. Thirty-four patients (36%) had allogeneic stem-cell transplant with 77% 2-year OS.

CONCLUSION

Magrolimab + azacitidine was well tolerated with promising efficacy in patients with untreated higher-risk MDS, including those with TP53 mutations. A phase III trial of magrolimab/placebo + azacitidine is ongoing (ClinicalTrials.gov identifier: NCT04313881 [ENHANCE]).

INTRODUCTION

Patients with higher-risk myelodysplastic syndromes (HR-MDS) as classified by Revised International Prognostic Scoring System (IPSS-R; including intermediate-, high-, and very high-risk patients) have poor prognosis and high risk of progression to acute myeloid leukemia (AML).¹ Patients with HR-MDS receive hypomethylating agents (HMAs), including azacitidine and decitabine, as the most common frontline therapy with intensive chemotherapy or targeted therapies in selected patients (eg, young patients with excess blasts and available donors, those with AML-defining mutations). The only potentially curative therapy is allogeneic hematopoietic stem-cell transplant (allo-HSCT) in a minority of patients, which is associated with improved overall survival (OS) in patients up to age 75 years.^2-4 However, complete remission (CR) rates are low with single-agent HMAs in pivotal trials and a large meta-analysis (7%-17%),^5-12 with few exceptions.^13-15 Although median OS with azacitidine was 24.5 months in the AZA-001 trial,⁶ almost all subsequent published data sets have demonstrated median OS ranging from 10 to 19 months.^10-14,16 Notably, multiple completed trials combining investigational agents with HMAs have failed to meet primary end points.^17-22 Thus, there is a high unmet need to improve outcomes in newly diagnosed HR-MDS.

CONTEXT

• Key Objective

• Is the first-in-class anti–cluster of differentiation 47 monoclonal antibody magrolimab safe, well tolerated, and efficacious when combined with azacitidine in untreated patients with higher-risk myelodysplastic syndromes, a population with high unmet need?

• Knowledge Generated

• In the overall population and in the subset of patients with TP53 mutations, magrolimab + azacitidine produced encouraging overall and complete remission rate, duration of overall and complete remission, and overall survival; over one third of patients were able to receive allogeneic stem-cell transplant, the only potentially curative treatment for higher-risk myelodysplastic syndromes. Magrolimab plus azacitidine was generally well tolerated, with manageable expected on-target anemia with hemoglobin improvement and reduced transfusion requirement over time on treatment.

• Relevance (C.F. Craddock)

• Magrolimab uses a novel mechanism of action in myelodysplasia and in combination with azacitidine is well tolerated and demonstrates promising clinical activity. The combination is currently undergoing evaluation in a randomized phase III trial.*

• *Relevance section written by JCO Associate Editor Charles F. Craddock, MD.

Cluster of differentiation 47 (CD47), a cell surface molecule, functions as a phagocyte immune checkpoint when bound to its receptor, phagocyte cell surface signal-regulatory protein alpha, impeding phagocytosis and serving as a don't-eat-me signal (Data Supplement, online only).^23,24 CD47 is overexpressed on most cancer cells including blasts in MDS.²⁵

Magrolimab is a first-in-class humanized immunoglobulin G4 anti-CD47 antibody that blocks CD47 binding to signal-regulatory protein alpha and enhances the phagocytosis of tumor cells.²⁶ Magrolimab has promising activity in relapsed/refractory non-Hodgkin lymphoma in combination with rituximab.²⁷ Synergy with anti-CD47 therapies occurs with therapeutics that increase prophagocytic signals on tumor cells. Preclinical data demonstrated that azacitidine robustly upregulated a cell surface prophagocytic marker, calreticulin, on malignant myeloid cells.²⁸ Furthermore, magrolimab + azacitidine increased phagocytosis in vitro and in vivo in myeloid models, providing strong mechanistic rationale for investigating the combination in patients.^28-30 This report describes final results of the phase Ib combination study (5F9005) of magrolimab + azacitidine in patients with untreated HR-MDS (ClinicalTrials.gov identifier: NCT03248479).

PATIENTS AND METHODS

Eligibility Criteria

Adults with untreated MDS by WHO classification and an IPSS-R risk category of intermediate, high, or very high risk were eligible. Patients could not have received prior treatment with HMAs but could have received lower-risk MDS therapies, including growth factors, transfusions, lenalidomide, and/or hydroxyurea. Patients had an Eastern Cooperative Oncology Group performance status of 0-2, AST/ALT ≤ 5× the upper limit of normal, and a glomerular filtration rate of ≥ 40 mL/min/1.73 m². No baseline hemoglobin requirement was specified in the Protocol (online only). Full inclusion/exclusion criteria are available in the protocol (Data Supplement).

Study Oversight

All patients provided written informed consent before study participation. The study was conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonisation Good Clinical Practice guidelines. Institutional review board approval was required at each site before study conduct.

Study Design

Study 5F9005 was an open-label, single-arm, multicenter phase Ib trial (Data Supplement) to evaluate the safety/tolerability (primary objective) and efficacy (secondary objective) of magrolimab + azacitidine in patients with untreated HR-MDS. A safety run-in cohort with standard 3 + 3 design was conducted in patients with MDS to evaluate dose-limiting toxicities and confirm the dose and schedule for the expansion phase. In the expansion phase, magrolimab was administered intravenously at the selected expansion dose with a priming dose of 1 mg/kg on days 1 and 4, 15 mg/kg on day 8, and 30 mg/kg on days 11, 15, and 22, then 30 mg/kg once weekly in the initial expansion cohort (planned enrollment 36 patients), or with maintenance dose interval increased to once daily every 2 weeks beginning cycle 3 day 1 in the additional expansion cohort (planned enrollment 58 patients). Azacitidine was administered subcutaneously or intravenously at 75 mg/m² once daily on days 1-7 of each 28-day cycle in all cohorts. Treatment was continued until unacceptable toxicity, progression, or death. A study design schema is provided (Data Supplement).

Primary end points were adverse events (AEs) and serious AEs per Common Terminology Criteria for AEs Version 4.03, and efficacy per investigator-assessed rate of CR by International Working Group 2006 MDS response criteria.³¹ Secondary and exploratory end points are listed in the Data Supplement. Treatment-emergent AEs (TEAEs) were assessed from first dose until 30 days after last dose of study drugs. Bone marrow evaluations for response assessment were conducted at screening and every two cycles beginning on cycle 3 day 1 and then every three cycles beginning on cycle 7 day 1 (Data Supplement). Overall response rate was a secondary end point defined per protocol as CR + marrow CR (mCR) + partial remission + stable disease with any hematologic improvement (HI). Minimal residual disease (MRD) negativity was centrally assessed using multiparameter flow cytometry (Hematologics, Inc, Seattle, WA) with a 0.02% lower limit of detection. Cytogenetics were evaluated locally using IPSS-R criteria.^1,32 TP53 and other prespecified mutations were identified using next-generation sequencing (NGS) conducted locally per each institution's standard practice.

Statistical Analysis

Statistical analyses were based on all patients who received at least one dose of magrolimab. Descriptive statistics were calculated for continuous variables; frequency and percentage were determined for categorical variables with 95% CIs using the Clopper and Pearson method for primary and secondary efficacy end points. Formal hypothesis testing of the CR rate used the chi-square test at a two-sided significance level of 0.05 (see the Data Supplement for details of sample size and power assumptions). Time-to-event variables were assessed by Kaplan-Meier estimates and corresponding two-sided 95% CIs for the median. Durations of response include response maintained in the post-transplant settings of patients who received allo-HSCT. Data cutoff was December 1, 2021.

RESULTS

Patients

This study enrolled 95 patients between July 10, 2018, and August 4, 2020, including one patient from the safety run-in cohort who received the same regimen as the expansion cohort. Median age was 69 (range, 28-91) years, and 69.5% had Eastern Cooperative Oncology Group performance status 1/2 (Table 1). A total of 62.1% had poorer-risk cytogenetics (ie, poor/very poor by IPSS-R criteria^1,32) and 27.4% had complex cytogenetics (≥ 3 cytogenetic abnormalities); 22.1% had therapy-related MDS (Table 1). IPSS-R risk was intermediate, high, or very high in 27.4%, 51.6%, and 21.1% of patients, respectively; 26.3% (n = 25) had a TP53 mutation. Median exposure was six cycles (range, 1-27 cycles), and as of data cutoff, four patients were ongoing on treatment and 40 remained in follow-up. Primary reasons for treatment discontinuation in 91 who discontinued included progressive disease (30.5%), AE (6.3% for magrolimab, 7.4% for azacitidine), patient decision, physician decision, lack of efficacy (5.3% each), and consent withdrawn (4.2%); 34 patients (35.8%) discontinued and received allo-HSCT; of these, five received alternative MDS-directed therapy before allo-HSCT and were censored for CR (or objective response [OR]) duration at the last response assessment before new MDS-directed therapy.

TABLE 1.

Baseline Demographics and Characteristics (N = 95)

Safety

The most common all-grade TEAEs regardless of attribution were constipation (68.4%), thrombocytopenia (54.7%), anemia (51.6%), neutropenia (47.4%), nausea (46.3%), and diarrhea (43.2%; Table 2). The most common grade 3/4 TEAEs regardless of attribution were anemia (47.4%), neutropenia (46.3%), and thrombocytopenia (46.3%; Data Supplement). Serious TEAEs documented in ≥ 5% of patients were febrile neutropenia (24.2%), pneumonia (9.5%), anemia (8.4%), bacteremia (6.3%), pyrexia (5.3%), and infusion-related reaction (IRR; 5.3%; Data Supplement).

TABLE 2.

TEAEs Occurring in ≥ 10% of Patients Regardless of Causality (N = 95)

Immune-related reactions possibly/probably related to magrolimab were infrequent (2.1%; one grade 2 pneumonitis and one grade 3 pneumonitis). Treatment-related TEAEs occurring in ≥ 10% of patients by grade are shown in the Data Supplement. Magrolimab-related TEAEs were anemia in 37.9% of patients (3.2% grade 2, 34.7% grade 3, no grade 4/5). Of the patients with magrolimab-related anemia, 33.3% had grade 1, 50.0% had grade 2, and 13.9% had grade 3 anemia at baseline. In cycle 1, median (range) hemoglobin change from baseline to the first post–magrolimab infusion sample was −0.7 g/dL (−3.1 to +2.4 g/dL), and median maximum drop was −1.1 g/dL (−5.6 to +2.0 g/dL) between the first and second magrolimab doses and −0.5 g/dL (−4.9 to +5.4 g/dL) between the second and third magrolimab doses. A total of 27.2% of patients had a ≥ 2 g/dL drop and 10.9% had a ≥ 3 g/dL drop from baseline between magrolimab doses 1 and 3. Hemoglobin change from baseline over time on treatment is shown in Figure 1 and the number of RBC/whole blood units transfused over time in the Data Supplement.

FIG 1.

Hemoglobin changes from baseline over time on treatment with magrolimab + azacitidine in patients with HR-MDS (N = 95). Data shown are median (Q1, Q3) change in blood samples drawn before each magrolimab dose. Analysis includes all patients who received at least one dose of magrolimab. HR-MDS, higher-risk myelodysplastic syndrome; Q1, first quartile; Q3, third quartile.

Antidrug antibodies were detected in 2.2% of patients treated with magrolimab; these were transient, not neutralizing, and without clinical sequelae. TEAEs led to magrolimab and azacitidine dose delays in 52.6% and 49.5% (the most common reason for magrolimab and azacitidine dose delays was neutropenia: 14.7% and 12.6%, respectively), and dose reduction in 0% and 17.9%, respectively (most commonly because of neutropenia: 11.6%; Data Supplement). TEAEs led to treatment discontinuation in 10 patients (10.5%; Data Supplement). Mortality was 2.1% at 60 days from treatment initiation. During the TEAE assessment period, most patients died because of progressive disease (20.0%) or AE (8.4%; Data Supplement).

Efficacy

The primary efficacy end point of CR with magrolimab + azacitidine was achieved by 32.6% (95% CI, 23.4 to 43.0) of the overall population by intention-to-treat analysis, with an overall response rate of 74.7% (95% CI, 64.8 to 83.1; Fig 2A; Table 3). Median (range) time to first response was 1.9 months (0.7-10.9) and median time to CR was 3.7 months (1.7-7.2). Median CR duration was 11.1 months (95% CI, 7.6 to 13.4) and median OR duration was 9.8 months (95% CI, 8.8 to 12.9). Among patients who achieved a CR, four patients had responses that deepened to CR between 6 and 8 months of therapy (Fig 3). Of those who achieved CR, 41.9% were flow cytometric MRD-negative. An mCR with or without HI was achieved by 16.8% and 14.8%, respectively. Of 65 patients with abnormal cytogenetics at baseline, a complete cytogenic response was achieved in 19 of 47 (40.4%) evaluable patients. Among 20 patients with abnormal cytogenetics who achieved CR, the complete cytogenic response rate was 50.0%. Of 26 patients with complex cytogenetics, 12 (46.2%) achieved CR. Any HI was achieved in 58.9% (HI in neutrophils, 22 [23.2%]; platelets, 38 [40.0%]; and erythroid, 36 [37.9%]). Overall, 37 patients (38.9%) were RBC transfusion-dependent at baseline, defined as transfusion within 4 weeks before first study treatment; 13 of these patients (35.1%) converted to RBC transfusion independence, defined as no transfusions for at least 8 consecutive weeks. Median duration of transfusion independence was 5.2 months (range, 1.9-18.0) as of data cut‐off.

FIG 2.

Efficacy of magrolimab + azacitidine in patients with HR-MDS (N = 95). Assessments of efficacy were based on bone marrow aspirate and trephine bone marrow biopsy collected every two cycles starting with cycle 3 day 1, then every three cycles starting with cycle 7 day 1. Analyses include all patients who received at least one dose of magrolimab. (A) Best change from baseline in percent of bone marrow blasts for each patient. Bars are labeled by TP53 mutation status; those labeled as missing represent patients with TP53 mutation status unknown. (B) KM curves of PFS for the overall population and by TP53 mutation status. (C) KM curve of OS for the overall population and by TP53 mutation status. (D) KM curve of OS for patients who did and did not undergo allo-HSCT after magrolimab + azacitidine treatment. allo-HSCT, allogeneic hematopoietic stem-cell transplantation; HR-MDS, higher-risk myelodysplastic syndrome; KM, Kaplan-Meier; NR, not reached; OS, overall survival; PFS, progression-free survival.

TABLE 3.

Efficacy Outcomes

FIG 3.

Swimmer plot of response over time, treatment duration, and next treatment initiation for individual patients treated with magrolimab + azacitidine who achieved a CR. Y-axis indicates TP53-mut status for each patient. Time of first response assessment and results of subsequent response assessments are indicated by vertical lines. Stem-cell transplant is shown by black dots. Blue stars indicate time of next treatment initiation, and blue arrows show patients who were ongoing on magrolimab treatment at the time of data cutoff. CR, complete remission; mCR, marrow CR; mut, mutation; PD, progressive disease; PR, partial remission; Pt, patient; SCT, stem-cell transplantation; SD, stable disease; wt, wild-type.

With median (Q1, Q3) duration of follow-up for survival of 17.1 months (10.8, 21.2), median progression-free survival was 11.6 months (95% CI, 9.0 to 14.0) and median OS was not reached (NR; 95% CI, 16.3 to NR; Figs 2B and 2C). Survival rates at 1 and 2 years were 74.6% and 51.6%, respectively. Median OS was NR in MRD-negative or MRD-positive patients with separation of survival curves favoring achievement of MRD negativity (Data Supplement). Fourteen patients (14.7%) progressed to AML in a median of 26.0 months (95% CI, 26.0 to NR).

Efficacy of magrolimab + azacitidine was assessed in patients with and without a detectable TP53 mutation at trial entry (Table 3). In the 25 TP53-mutant patients, 10 (40.0%) achieved CR, with a median CR duration of 7.6 months (95% CI, 3.1 to 13.4), a median OR duration of 9.2 months (95% CI, 5.0 to 12.2), and a median OS of 16.3 months (95% CI, 10.8 to NR); 1-year survival rate was 61.5% with a median follow-up of 12.5 months (Fig 2C). In TP53-wild-type patients, 19 (31.1%) achieved CR, with a median (95% CI) CR duration of 12.9 months (95% CI, 8.0 to NR), a median OR duration of 9.8 months (95% CI, 8.5 to 18.5), and a median OS NR (95% CI, 21.3 to NR); 1-year survival rate was 79.4% with a median follow-up of 19.0 months (Fig 2C).

Overall, 34/95 patients received allo-HSCT. Most patients were in CR or mCR at the last assessment before transplant (Data Supplement). Median OS was NR (95% CI, NR to NR), and 1-year survival was 91.2% (95% CI, 75.1 to 97.1) in allografted patients, compared with 15.0 months (95% CI, 12.5 to 21.3) and 1-year survival of 64.7% (95% CI, 50.7 to 75.6) in those without allo-HSCT (Fig 2D; Data Supplement). Four patients with TP53 mutation received allo-HSCT (Data Supplement). Median OS in these patients was 19.7 months (95% CI, 9.8 to 19.7) versus 13.4 months (95% CI, 9.8 to NR) in patients without allo-HSCT. Seven of 34 patients were MRD-negative at the assessment before allo-HSCT; median (95% CI) OS was NR (19.7 months to NR) in these patients and NR (NR to NR) in MRD-positive patients (Data Supplement).

The proportion of patients who achieved CR or mCR was consistent across NGS-identified driver mutations at screening (Data Supplement).

DISCUSSION

Although immunotherapy has shifted the paradigm in solid tumors and Hodgkin lymphoma, targeting traditional immune checkpoints in patients with MDS or AML (eg, with anti-programmed cell death protein 1, anti-programmed death-ligand 1, or cytotoxic T-cell lymphocyte-4 therapy) with or without azacitidine has shown significant immune-related toxicities and no clear evidence of synergy, possibly because of a functionally abrogated T-cell response.^22,33-35 Magrolimab offers a dual mechanistic approach not only to augment the innate immune response by macrophage-mediated phagocytosis but also to eradicate the underlying leukemia stem cell and augment adaptive immunity.^36,37 Importantly, in preclinical models, the combination with azacitidine is synergistic, mediated via upregulation of prophagocytic signals such as calreticulin.²⁸

Magrolimab + azacitidine was well tolerated in this study, and although dose delays of both treatments occurred in about 50% of patients, no patients had magrolimab dose reductions and few (6.3%) discontinued magrolimab treatment because of TEAEs. Magrolimab and azacitidine dosing are decoupled in more recent ongoing trials to avoid potential loss of efficacy and receptor saturation that may be associated with magrolimab dose delays beyond a 3-week interdose interval. Importantly, only one significant immune-related reaction was noted with magrolimab treatment (a grade 3 pneumonitis), which is consistent with no or very few immune-related AEs noted in prior magrolimab studies.^27,38 IRRs to the first doses were observed, which were well managed with acetaminophen and diphenhydramine premedication (with corticosteroid for grade 3 IRR) in the subsequent three to four doses with no need for continued premedication in subsequent doses.

Anemia is a known on-target side effect of anti-CD47 treatment.^26,39 Red cells express CD47 and are naturally removed from the circulation by macrophages as they lose CD47 expression and express higher levels of prophagocytic signals with increasing red cell age.^24,40 The priming dose removes older RBCs that express higher levels of eat-me signals with compensatory reticulocytosis,^26,38 which allows consistent delivery of higher maintenance doses without subsequent anemia. In this study, anemia related to magrolimab was reported in 37.9% (3.2% grade 2, 34.7% grade 3) of patients, 97.2% of whom had grade 1-3 anemia at baseline; however, grade 4 events were rare (1 anemia and 2 hemolytic anemia) and these were not considered magrolimab related. A drop in hemoglobin ≥ 2 g/dL between the first and third doses of magrolimab was seen in 27.2% of patients, and 10.9% had a ≥ 3 g/dL drop. This highlights the importance of close hemoglobin monitoring during the initiation of magrolimab treatment (Data Supplement). Anemia events because of magrolimab in later treatment cycles are uncommon. Importantly, 35.1% of patients who were transfusion-dependent at baseline converted to transfusion independence on treatment, and the overall need for RBC/whole blood unit transfusions decreased during the study.

Azacitidine is currently standard of care for HR-MDS,⁴¹ yet CR rates were < 20% in pivotal studies,^5,6 with most responders relapsing within 2 years and an OS of approximately 17-19 months.^11-14,16 In this context, the CR rate of 32.6% is encouraging. More importantly, the median OS was NR with a 17.1-month median follow-up, which is encouraging since 62.1% of patients had poor/very poor cytogenetic risk and 26% had TP53 mutations, populations that tend to have a worse prognosis.^1,12

Mutations in TP53 are observed in 10%-28% of patients with MDS,⁴² and are associated with short response durations to currently available therapies.⁴³ In this phase Ib study, magrolimab + azacitidine demonstrated promising efficacy overall and in TP53-mutated MDS. CR rate was 32.6% overall, with CR rates of 40.0% in TP53-mutated and 31.1% in TP53-wild-type patients. HI was achieved in 56.0% and 60.7% in TP53-mutated and TP53-wild-type subgroups, respectively, compared with 23%-36% of azacitidine-treated patients with MDS as published in Cancer and Leukemia Group B studies.⁸ Clinical outcomes are dismal in patients with TP53-mutated MDS, with a median OS of 5-10 months in TP53-mutated patients on any available therapies.^43-46 In this study, we see an encouraging median OS of 16.3 months. Notably, the high prevalence of TP53 mutation in this study (26.3%) makes the median OS of the entire cohort more promising (NR with a median follow-up of 17.1 months), with > 60% of TP53-wild-type patients alive at data cutoff. Of course, the efficacy of this combination needs confirmation in the ongoing registration study, given the single-arm phase Ib design of this study.

Historically, the proportion of patients with MDS who proceed to allo-HSCT is < 10%.⁴⁷ Limited data support improved outcomes with HMA treatment before allo-HSCT in patients with HR-MDS,^3,48,49 although ideally, improving depth of response as defined by MRD negativity will improve OS after allo-HSCT.⁵⁰ Among the 34 patients (35.8%) who proceeded to allo-HSCT in our study, the median OS and 95% CI were NR. Four patients (16.0%) with TP53 mutations had allo-HSCT, and the median OS of 19.7 months (95% CI, 9.8 to 19.7) is highly encouraging albeit with very small numbers precluding any definitive conclusions or recommendations. With the generally poor outcomes in HR-MDS, heightened further in patients with TP53 mutations,^42,46,51 the potential that magrolimab + azacitidine treatment may enable more patients to proceed with allo-HSCT could be highly desirable and requires further study in the ongoing registration study. Adding venetoclax to magrolimab + azacitidine may further increase both CR and CR/CR with incomplete blood count recovery rates in AML on the basis of early data from single-arm studies,⁵² potentially allowing more patients to be bridged to allo-HSCT; this strategy is under investigation in the ENHANCE-3 AML trial (ClinicalTrials.gov identifier: NCT05079230) and planned for HR-MDS, especially patients with excess blasts.

Limitations of this study include being a single-arm, nonrandomized study done at larger academic centers consistent with the phase Ib design, lack of centralized NGS assessment to annotate molecular responses, and lack of sufficient numbers to definitively delineate efficacy and outcomes in individual molecular subgroups.

To our knowledge, this phase Ib study represents the largest, most comprehensive data set for magrolimab with azacitidine in myeloid malignancies. On the basis of the interim results from this phase Ib trial, the combinations of magrolimab + azacitidine and placebo + azacitidine are being evaluated in patients with HR-MDS in the multinational, randomized phase 3 ENHANCE trial, which is recruiting as of this report (ClinicalTrials.gov identifier: NCT04313881). If successful, this combination could be an important addition for patients with HR-MDS, a population of significant unmet need.

DATA SHARING STATEMENT

Gilead Sciences shares anonymized individual patient data upon request or as required by law or regulation with qualified external researchers on the basis of submitted curriculum vitae and reflecting nonconflict of interest. The request proposal must also include a statistician. Approval of such requests is at Gilead Science's discretion and is dependent on the nature of the request, the merit of the research proposed, the availability of the data, and the intended use of the data. Data requests should be sent to datarequest@gilead.com.

AUTHOR CONTRIBUTIONS

Conception and design: David A. Sallman, Mark Chao, Paresh Vyas, Naval G. Daver

Administrative support: Anderson Tan

Provision of study materials or patients: David A. Sallman, Monzr M. Al Malki, Adam S. Asch, Eunice S. Wang, Joseph G. Jurcic, Terrence J. Bradley, Ian W. Finn, Daniel A. Pollyea, Suman Kambhampati, Tiffany N. Tanaka, Joshua F. Zeidner, Guillermo Garcia-Manero, Deepa Jeyakumar, Rami Komorokji, Jeffrey Lancet, Hagop M. Katarjian, Paresh Vyas, Naval G. Daver

Collection and assembly of data: Anderson Tan, Mark Chao, Carol O'Hear, Giridharan Ramsingh, Indu Lal

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Magrolimab in Combination With Azacitidine in Patients With Higher-Risk Myelodysplastic Syndromes: Final Results of a Phase Ib Study

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).