Magrolimab plus azacytidine in patients with myelodysplastic syndromes and acute myeloid leukemia

During the American Society of Clinical Oncology (ASCO) Virtual Meeting 2020, David Sallman from the Moffitt Cancer Center, US, presented an abstract on 5F9005, a phase Ib study (NCT03248479), investigating the safety and efficacy of magrolimab plus azacytidine (AZA) in patients with myelodysplastic syndromes (MDS), and acute myeloid leukemia (AML).¹ He presented data from the expansion cohort.

Magrolimab (Hu5F9-G4), which was last year granted fast track designation by the U.S Food and Drug Administration, is a first-in-class antibody against CD47, a macrophage immune checkpoint protein. This antibody has been shown to block phagocytic evasion, inducing tumor phagocytosis, which facilitates the elimination of malignant cells. In preclinical models, a synergy of magrolimab with AZA has been observed. The 5F9005 study was designed to examine whether such synergy can be seen in patients, and to assess the tolerability and clinical activity of the combination.

Study design and patient characteristics

Patients with untreated AML who were ineligible for induction therapy or with untreated intermediate/very high-risk MDS were eligible for the study. The study was designed as a dose escalation (N = 6) followed by cohort expansion. Magrolimab was administered in an escalating dose from (1–30 mg/kg) for the first 2 weeks. Then at a maintenance dose of 30 mg/kg weekly (Q1W) throughout, or once every 2 weeks (Q2W) from Cycle 3+. AZA (75 mg/m²) was given on Day 1–7 of each cycle.

The primary objectives of the study were safety and efficacy assessed by the International Working Group 2006 (MDS), and the European Leukemia Net 2017 (AML) criteria. Secondary objectives included pharmacokinetics, pharmacodynamics, and immunogenicity, while exploratory objectives aimed to assess CD47 receptor occupancy, markers of immune cell activity, and molecular profiling in AML/MDS.

Based on pharmacokinetics, pharmacodynamics, and a similar receptor occupancy with Q1W and Q2W, Q2W dosing was selected.

Baseline patient and disease characteristics are presented in Table 1.

Table 1. Selected baseline patient and disease characteristics in untreated MDS and AML

AML, acute myeloid leukemia; MDS, myelodysplastic syndromes; MRC, myelodysplasia-related changes; NA, not applicable; RS, ringed sideroblasts; WHO, World Health Organization.
Characteristic	MDS (N = 39)	AML (N = 29)
Median age (range), years	70 (47–80)	74 (60–89)
Cytogenetics risk, % Intermediate Poor Unknown	28 64 8	7 72 21
WHO AML classification, % MRC Recurrent genetic abnormalities Therapy related Not otherwise specified	NA	66 7 10 17
WHO MDS classification, % RS and single/multi-lineage dysplasia Multi-lineage dysplasia RS with multi-lineage dysplasia Excess blasts Unclassifiable/unknown	3 18 8 56 15	NA
Therapy related MDS, %	31	NA
Harboring TP53 mutation	13	45

Results for the expansion cohort

Safety

• The maximum tolerated dose was not reached
• The safety profile of the magrolimab plus AZA combination was consistent with AZA monotherapy
• The most frequent adverse event (AE) was anemia, which significantly improved over time in the majority of patients
• There were no significant cytopenias, infections, or immune-related AEs
• Most patients had improved neutrophil and platelet counts on combined therapy
• Discontinuation due to drug-related AEs was reported in 1/68 (1.5%) of patients from the combination treatment who experienced a severe infusion-related reaction

Efficacy

• The median follow-up was 5.8 (2.0–15.0) and 9.4 (1.9–16.9) months for patients with MDS and AML, respectively
• Overall, efficacy with magrolimab plus AZA was higher than with AZA monotherapy
• The combination induced high overall response rates (ORR) in patients with MDS and AML (Table 2)
  • ORR of 91% and 64%, including 42% and 56% of patients with complete response, respectively
  • 35% and 50% of patients achieved a complete cytogenetic response
• Responses deepened over time, with a median duration of response not reached
• Median overall survival was not reached
• Median time to response was 1.9 months
• In total, 9/58 (16%) of patients proceeded to receive allogeneic stem cell transplant
• The combination was also effective in patients with TP53 mutations (Table 3)

Table 2. Responses with magrolimab plus AZA

Best overall response	MDS (N = 33)	AML (N = 25)
ORR, %	91	64
CR, %	42	40
CRi, %	NA	16
PR, %	3	4
SD, %	9	32
PD, %	0	4
MRD negative status, %	20	50
Median DoR, months	NR	NR
Median follow-up (range), months	5.8 (2.0–15.0)	9.4 (1.9–16.9)
AML, acute myeloid leukemia; CR, complete response; CRi, CR with incomplete hematologic recovery; DoR, duration of response; MDS, myelodysplastic syndromes; MRD, measurable residual disease; NA, not applicable; NR, not reached; ORR, overall response rate; PD, progressive disease; PR, partial response; SD, stable disease.

 

Table 3. Responses in patients with TP53 mutations

AML, acute myeloid leukemia; CR, complete response; CRi, CR with incomplete hematologic recovery; DoR, duration of response; MRD, measurable residual disease; MDS, myelodysplastic syndromes; NR, not reached.
Best overall response	MDS TP53 mutant (n = 4)	AML TP53 mutant (n = 12)
ORR, %	75	75
CR, %	50	42
CRi/marrow CR, %	25	33
Complete cytogenetic response, n (%)	3/3 (100)	4/8 (50)
MRD negative status of responders, n, (%)	0	4/9 (44)
Median DoR, months	NR	NR
6-months survival probability, %	100	91
Median follow-up (range), months	7 (4.2–12.2)	8.8 (1.9–16.9)

Conclusion

The data indicate that magrolimab, a first-in-class anti-CD47 antibody, in combination with AZA has a manageable toxicity profile and encouraging efficacy in patients with MDS and AML. Moreover, the combination demonstrated activity in patients with TP53 mutations, a high unmet need population with poor outcomes. A further study to explore the efficacy of magrolimab in the TP53 mutated AML is planned.

The cohort expansion study (NCT03248479) in patients with MDS and AML is ongoing, as are the registrational studies for MDS.