Magrolimab Combination Therapy Demonstrates Promising Activity in High-Risk AML

Magrolimab in combination with azacitidine and venetoclax was well tolerated and yielded promising responses among patients with high-risk acute myeloid leukemia regardless of TP53 mutation status.

First-line magrolimab plus azacitidine (Vidaza) and venetoclax (Venclexta) produced promising response rates and was tolerated well among patients with high-risk de novo and secondary acute myeloid leukemia (AML) with or without TP53 mutations, according to data from a phase 1/2 trial (NCT04435691) presented during the 2022 American Society of Hematology (ASH) Annual Meeting.

In the frontline population (n = 43), the CR/CR with incomplete hematologic recovery (CRi) rate was 72% (n = 31), and the objective response rate (ORR) was 81% (n = 35). The minimal residual disease (MRD)–negative rate was 67% (n = 16/28). The median count recovery for platelets and absolute neutrophils was 32 days (range, 0-74) and 36 days (range, 16-88), respectively. The 4- and 8-week mortality rates were 0%.

In the relapsed/refractory cohort (n = 36), the CR/CRi rate was 11% (n = 2) in venetoclax-exposed patients and 44% (n = 8) in venetoclax-naïve patients. Only 1 patient without prior venetoclax exposure achieved a morphologic leukemia-free state, for an ORR of 49% (n = 9).

“The combination of azacitidine, venetoclax, and magrolimab was safe in the frontline setting in this very high-risk population,” said lead study author Naval Daver, MD, associate professor in the Department of Leukemia at The University of Texas MD Anderson Cancer Center in Houston, in a presentation of the data. “Activity in relapsed/refractory AML was modest.”

Azacitidine plus venetoclax represents a frontline standard in older and unfit patients with AML, but only 35% to 40% of patients remain alive for more than 2 years. Moreover, patients with TP53 mutations have worse outcomes, with a median survival of 5 to 7 months.

Prior single-arm studies have shown encouraging safety and activity with an hypomethylating agent (HMA) plus CD47 inhibition in frontline, TP53-mutant and TP53 wild-type disease. Moreover, the addition of the CD47 inhibitor magrolimab to azacitidine and venetoclax was previously found to have encouraging synergistic activity when examined in preclinical models.

The phase 1 dose-finding portion of the trial enrolled patients (n = 6) at least 18 years of age with relapsed/refractory AML, an ECOG performance status (PS) of 2 or less, adequate organ function, and a white blood cell count no greater than 15 x 109/L. In this group, no dose-limiting toxicities occurred. As such, the recommended phase 2 dose (RP2D) for magrolimab was established at 1 mg/kg on days 1 and 4, 15 mg/kg on day 8, and 30 mg/kg on day 11 of the first cycle and thereafter for subsequent doses.

The phase 2 portion of the research was split into 3 cohorts: frontline (de novo and secondary AML), relapsed/refractory and venetoclax-naïve, and relapsed/refractory with prior venetoclax exposure. In the frontline cohort, patients had to be at least 75 years of age, younger than 75 years and ineligible for intensive therapy, or be at least 18 years of age with TP53-mutant or adverse-risk cytogenetics regardless of fitness.

The primary objectives were to determine the maximum tolerated dose and RP2D and assess the rate of CR and CRi. Secondary objectives included the evaluation of ORR, duration of response (DOR), event-free survival, overall survival (OS), MRD-negative rate, 4- and 8-week mortality, and the number of patients transitioning to stem cell transplant.

Baseline characteristics in the frontline cohort reflected a median age of 70 years (range, 32-84); 37% (n = 16) of patients were female, and most patients (n = 40; 93%) had an ECOG PS of 1 or 2. Daver emphasized that 91% (n = 39) of patients had adverse risk and 65% (n = 28) had adverse cytogenetics per ELN 2017 criteria.

At a median follow-up of 14.5 months, the DOR in the frontline de novo population (n = 33) was not reached in patients with TP53-mutant and TP53 wild-type disease. The 18-month DOR rates were 52% and 74%, respectively. The 12-month OS rates in the TP53-mutant and TP53 wild-type populations were 53% and 83%, respectively.

At a median follow-up of 14.0 months, the DOR in the frontline secondary population (n = 10) was 5.9 months in those with TP53-mutant disease and 2.2 months in those with TP53 wild-type disease. The median OS was 7.6 months and 9.6 months, respectively.

Additionally, the variant allele fraction of TP53 was reduced from baseline to best response in responders (n = 15; P = .01) and non-responders (n = 6; P = .07).

To get a better sense of how the triplet compares with combinations of HMAs and venetoclax in the TP53-mutated population, investigators conducted a survival comparison of the 2 approaches using datasets evaluating azacitidine or decitabine plus venetoclax from MD Anderson Cancer Center (n = 150). The results favored the triplet, with a median OS of 10.4 months vs 5.7 months with HMA/venetoclax (adjusted HR for the triplet for death, 0.41; 95% CI, 0.18-0.88; P = .009). Similar findings were seen in a propensity-matched analysis, with a median OS of 10.4 months with the triplet vs 3.5 months with HMA/venetoclax (P = .02).

Additional findings demonstrated that 8 patients with de novo and 1 with secondary, untreated TP53-mutant disease underwent transplant. The median relapse-free survival (RFS) was 16.3 months in transplanted patients vs 4.2 in non-transplanted patients. Similarly, median OS was improved in transplanted patients, at 16.6 months vs 9.8 months in non-transplanted patients.

In the relapsed/refractory population, at a median follow-up of 17.9 months, the median RFS was 3.1 months in venetoclax-exposed patients vs 7.5 months in venetoclax-naïve patients. The median OS was 3.1 months and 5.6 months in venetoclax-exposed and venetoclax-naïve patients, respectively.

In terms of safety, all patients (n = 79) experienced at least 1 adverse effect (AE), and 90% (n = 71) had at least 1 grade 3 or greater AE.

Notably, there were no immunological AEs or treatment discontinuations because of treatment-related AEs. Eight patients (10%) experienced infusion reactions (grade 3, n = 3), which were well managed with dexamethasone prior to the next doses of therapy.

Eighteen patients (23%) had grade 3 or greater anemia on study, none of which led to life-threatening events or deaths. The median drop in haptoglobin after the first infusion of magrolimab was 1.2 g/dl (range, 0-3.9) in the frontline population.

Common non-hematologic treatment-emergent AEs included febrile neutropenia (any grade, 44%; grade ≥3, 44%), lung infection (any grade, 43%; grade ≥3, 35%), sepsis (any grade, 15%; grade ≥3, 15%), and hyperbilirubinemia (any grade, 52%; grade ≥3, 11%).

“There were no unexpected AEs,” Daver said. “Next steps include a randomized study that has been initiated to assess whether azacitidine, venetoclax, and magrolimab can improve on azacitidine and venetoclax in frontline patients in the phase 3 ENHANCE-2 [NCT04778397] and ENHANCE-3 [NCT05079230] trials,” Daver concluded.


Daver N, Senapati J, Maiti A, et al. Phase I/II study of azacitidine (AZA) with venetoclax (VEN) and magrolimab (Magro) in patients (pts) with newly diagnosed (ND) older/unfit or high-risk acute myeloid leukemia (AML) and relapsed/refractory (R/R) AML. Presented at: 2022 ASH Annual Meeting and Exposition; December 10-13, 2022; New Orleans, LA. Abstract 61