Carving Out New Combination Therapies in Stem Cell Disorders


Despite the influx of novel agents that have received regulatory approval in recent years in stem cell disorders, there is still room for improvement.

William B. Donnellan, MD

Despite the influx of novel agents that have received regulatory approval in recent years in stem cell disorders, there is still room for improvement, explained William B. Donnellan, MD, director of Leukemia/Myelodysplastic Syndrome Research at Sarah Cannon Research Institute and Tennessee Oncology.

In a virtual presentation during the 2020 ASCO Direct HighlightsTM webcast, a program developed by Physicians’ Education Resource®(PER®), LLC, Donnellan discussed the trials that are pushing the needle forward in stem cell disorders.

A Novel Combination in MDS/AML

CD47 is expressed on tumor cells and releases a “do not eat me” signal, and increased expression confers worse prognosis in patients with acute myeloid leukemia (AML). However, magrolimab, a first-in-class CD47 monoclonal antibody, can suppress that signal, enabling the macrophage to engulf the tumor cell, said Donnellan. The activity of magrolimab was found to be enhanced in combination with azacitidine in preclinical models.

The combination was subsequently evaluated in the phase 1b/2 5F9005 trial in patients with untreated AML ineligible for induction chemotherapy or untreated intermediate- or high-risk myelodysplastic syndrome (MDS). Patients received 1 mg/kg and 30 mg/kg of magrolimab in week 1 and 2, respectively, followed by 30 mg/kg every 2 weeks thereafter, plus 75 mg/m2 of azacitidine on days 1 to 7.1

Regarding baseline characteristics (n = 68), 64% to 72% of patients with AML and MDS had poor cytogenetic risk, 66% of patients with AML had underlying myelodysplasia, 31% of patients with MDS had therapy-related disease, and 45% of patients with AML harbored TP53 mutations.

Regarding safety, the combination was well tolerated. No maximum-tolerated dose was reached, and the safety profile of the combination was consistent with azacitidine alone. No significant cytopenias, infections, or immune-related adverse effects (AEs) were reported, said Donnellan.

Patients did experience a mild decrease in hemoglobin at the onset of treatment, which is an expected on-target toxicity of CD47 inhibitors, said Donnellan. However, the toxicity was mitigated with stepwise dosing over the course of several weeks, after which anemia and red blood cell transfusion dependence improved over time.

Treatment discontinuation due to drug-related AEs occurred in 1 patient, but no deaths occurred during the first 60 days on study.

The overall response rate (ORR) was 91% in the MDS cohort and 64% (n = 33) in the AML cohort (n = 25), the majority of which were complete responses (42% vs 40%), respectively. These rates compare favorably to azacitidine monotherapy, with which the CR rate ranges from 6% to 17%, said Donnellan. The median time to response was 1.9 months. Additionally, responses deepened over time; the 6-month CR rate was 56% in the MDS cohort.

Patients with TP53 mutations did relatively well compared with the overall population, added Donnellan, with an ORR of 75% in both cohorts. The CR rates in the AML and MDS cohorts were 42% and 50%, respectively. The estimated 6-month survival was 91% and 100%, respectively. Moreover, the median duration of response (DOR) and survival had not been reached, which compares favorably with current therapies.

“While the median follow-up is relatively short, at 6 months in the MDS cohort and 9 months in the AML cohort, many of these patients are still responding and are in cytogenetic and minimal residual disease negative remissions,” said Donnellan.

The results of the trial served as the basis for the phase 2 ENHANCE trial, set to open in the second half of 2020, in which patients with intermediate- or high-risk MDS will be randomized to the combination versus azacitidine alone.

Addressing Unmet Needs in Higher-Risk MDS/CMML and Low-Blast AML

Novel and safe combination therapies that improve outcomes represent an unmet need for patients with higher-risk MDS/chronic myelomonocytic leukemia (CMML) or low-blast AML (LB AML) who are ineligible for high-intensity therapy or transplant. Attempts to enhance the effects of azacitidine monotherapy with other therapies have been limited by toxicity. Pevonedistat, a first-in-class small-molecule inhibitor of the NEDD8-activating enzyme (NAE), binds to and inhibits NAE, disrupting cell proliferation and survival.

Investigators launched a phase 2 study (NCT02610777), in which patients with higher-risk MDS/CMML or LB AML were randomized to 20 mg/m2 of intravenous (IV) pevonedistat on days 1, 3, and 5 plus 75 mg/m2 of IV or subcutaneous azacitidine (n = 58) on days 1 to 5, 8, and 9, or azacitidine alone (n = 62) on the same schedule in 28-day cycles.

The combination led to a trend toward improved event-free survival (EFS) and a numerical improvement in overall survival (OS) versus azacitidine alone, said Donnellan. The median EFS was 21.0 months with pevonedistat and azacitidine versus 16.6 months with azacitidine alone in the intent-to-treat (ITT) population, translating to approximately a 34% reduction in the risk of death or transformation to AML (HR, 0.665; 95% CI, 0.423-1.047; P = .076). The median OS was 21.8 months and 19.0 months, respectively (HR, 0.802; 95% CI, 0.512-1.256; P = .334).2

In patients with higher-risk MDS, the median EFS was also prolonged with the combination versus azacitidine alone. Here, the median EFS was 20.2 months in the combination arm versus 14.8 months in the azacitidine-alone arm (HR, 0.539; 95% CI, 0.292-0.995; P = .045). The median OS was also numerically longer, at 23.9 months and 19.1 months, respectively (HR, 0.701; 95% CI, 0.386-1.273; P = .240).

Among 108 response-evaluable patients, the ORR was 70.9% in the combination arm, consisting of a 40.0% CR rate. The ORR in the azacitidine-alone arm was 60.4%, which consisted of a 30.2% CR rate. The median DOR was 20.6 months (95% CI, 10.71-34.60) in the combination arm versus 13.1 months (95% CI, 12.62—NE) in the azacitidine-alone arm.

According to a subgroup analysis, patients with higher-risk MDS (n = 59) and higher-risk CMML (n = 17) who received the combination experienced a higher ORR versus azacitidine alone, respectively, at 79.3% versus 56.7% and 77.8% versus 75.0%, respectively. Patients with LB AML (n = 32) experienced an ORR of 52.9% versus 60.0%, respectively.

In patients with higher-risk MDS, the median DOR was 34.6 months (95% CI, 11.53-34.60) versus 13.1 months (95% CI, 12.02—NE) in the combination and azacitidine-alone arms, respectively. The rate of transfusion independence was 69.2% with the combination versus 50.0% with azacitidine alone.

The most common any-grade AEs in the combination and azacitidine-alone arms, respectively, included pyrexia (38% vs 40%), cough (38% vs 34%), constipation (36% vs 47%), nausea (34% vs 45%), neutropenia (34%. vs 49%), diarrhea (33% vs 27%), anemia (31% vs 45%), febrile neutropenia (26% vs 29%), and fatigue (21% vs 40%).

The most common grade 3 or higher AEs in the combination and azacitidine-alone arms, respectively, included neutropenia (33% vs 27%), febrile neutropenia (26% vs 29%), decreased neutrophil count (21% vs 10%), anemia (19% vs 27%), thrombocytopenia (19% vs 23%), and pneumonia (12% vs 10%).

“The drug was very well tolerated. There really was not much in the way of toxicity beyond [what we see with] azacitidine alone,” said Donnellan.

The combination is being further evaluated versus azacitidine alone in the randomized phase 3 PANTHER trial (NCT03268954) in patients with higher-risk MDS/CMML and LB AML.

Dose Optimization in Chronic-Phase CML

A post hoc analysis modeling the data from the pivotal PACE trial suggested that a lower dose of ponatinib (Iclusig), 1 of 5 approved TKIs approved for the treatment of patients with chronic phase CML (CP-CML), could reduce the risk of arterial occlusive events (AOEs) while maintaining efficacy. To that end, investigators launched the phase 2 OPTIC trial to evaluate 1 of 3 starting doses of ponatinib––45 mg (n = 75), 30 mg (n = 73), or 15 mg (n = 68)––in patients with CP-CML resistant or intolerant to 2 or more prior TKIs.

The majority of patients had received at least 2 prior TKIs, for which resistance was the primary cause, said Donnellan.

Patients in the 45-mg arm achieved the highest 12-month BCR-ABLIS rate of 1% or lower, at 38.7% (range, 27.6%-50.6%), followed by patients in the 30-mg arm (27.4%; range, 17.6%-39.1%), and 15-mg arm (26.5%; range, 16.5%-38.6%).3

When the dose was reduced to 15 mg in the maintenance setting, 75% of patients in the 45 mg arm maintained BCR-ABLIS rate of 1% or lower. All 9 lost responses occurred within the first 90 days of the dose reduction, of which approximately 30% to 50% of patients were able to regain their response when they were re-escalated to 45 mg, said Donnellan.

Notably, patients with BCR-ABL T315I mutations achieved a higher BCR-ABLIS rate of 1% or lower in the 45 mg arm, at 42% versus 36% in those without the mutation. Conversely, patients with BCR-ABL T315I mutations had a lower BCR-ABLIS rate of 1% or lower in the 30 mg and 15 mg arms versus those without the mutation.

Regarding AEs of special interest, the reduction in AOEs, serious AOEs and grade 3 to 5 AOEs was most evident in the 45-mg arm. In the 45-mg arm, the rate of AOEs was 5.3%, which compares favorably with historical rates of 10% to 20%, which was first seen when the drug received regulatory approval, said Donnellan.

“Starting at 45 mg daily and then dropping down to 15 mg once a BCR-ABLIS rate of 1% or lower is achieved is the dosing scheme that provides the best benefit-risk profile,” said Donnellan.

Targeted Therapy in IDH2-Mutant AML

IDH2 mutations occur in approximately 10% to 15% of patients with AML, the prevalence of which increases with age, said Donnellan. IDH2 mutations are often thought of as founder mutations but can also be acquired at the time of progression.

In order to ascertain the value of adding the IDH2 inhibitor enasidenib (Idhifa) to azacitidine in patients with newly diagnosed AML who are ineligible for intensive chemotherapy, investigators conducted a phase 1b/2 trial, in which patients were ultimately randomized 2:1 to 100 mg of enasidenib daily plus azacitidine (n = 68) versus azacitidine alone (n = 33).

The majority of patients had intermediate or adverse cytogenetic risk, an IDH2-R140 mutation, and co-mutations, the most common of which included DNMT3A, ASXL1, and RUNX1.

The ORR was significantly higher in the combination arm versus the azacitidine-alone arm, at 71% versus 42%, respectively (P = .0064). Moreover, investigators reported approximately a 5-fold increase in the CR rate in the combination arm, at 53% versus 12% in the azacitidine-alone arm (P = .0001).4

Patients received a median of 10 cycles in the combination arm versus 6 cycles in the azacitidine-alone arm. The most common treatment-related AEs (TRAEs) in the combination and azacitidine-alone arms, respectively, included thrombocytopenia (62% vs 44%), nausea (69% vs 38%), anemia (53% vs 44%), and vomiting (49% vs 47%).

Common grade 3/4 TRAEs in the combination and azacitidine-alone arms, respectively, included thrombocytopenia (37% vs 19%), neutropenia (35% vs 22%), anemia (19% vs 22%), febrile neutropenia (15% vs 16%), and IDH differentiation syndrome (IDH-DS; 10% vs 0%). A total of 12 patients (18%) in the combination arm experienced IDH-DS at a median of 28.5 days. The median time to IDH-DS resolution was 11.5 days.

At a median follow-up of 14 months, the median OS was 22 months in the combination arm versus 22.3 months in the azacitidine-alone arm (HR, 0.99; 95% CI, 0.52-1.87; P = .9686). The study, already not powered to detect an OS difference, could have also been impacted by the fact that 21% of patients in the azacitidine-alone arm went on to receive subsequent treatment with enasidenib monotherapy, said Donnellan.

Among patients in the combination arm who achieved a CR, the median OS was not reached, and the 1-year survival rate exceeded 90%. The median EFS was 17.2 months and 10.8 months, respectively (HR, 0.59; 95% CI, 0.30-1.17; P = .1278).

“Currently, my preference would be to start with venetoclax (Venclexta)-based therapy, and at the time of progression, switch to an IDH2 inhibitor,” said Donnellan. “Though, this opens up the possibility of a new treatment for patients with IDH2 mutations.”


  1. Sallman DA, Malki MA, Asch AS, et al. Tolerability and efficacy of the first-in-class anti-CD47 antibody magrolimab combined with azacitidine in MDS and AML patients: phase 1b results. J Clin Oncol. 2020;38(15 suppl):7507. doi:10.1200/JCO.2020.38.15_suppl.7507
  2. Ades L, Watts JM, Radinoff A, et al. Phase II study of pevonedistat (P) + azacitidine (A) versus A in patients (pts) with higher-risk myelodysplastic syndromes (MDS)/chronic myelomonocytic leukemia (CMML), or low-blast acute myelogenous leukemia (LB AML) (NCT02610777). J Clin Oncol. 2020;38(15 suppl):7506. doi:10.1200/JCO/2020.38.15_suppl.7506
  3. Cortes JE, Lomaia E, Turkina A, et al. Interim analysis (IA) of OPTIC: A dose-ranging study of three ponatinib (PON) starting doses. J Clin Oncol. 2020;38(15 suppl):7502. doi:10.1200/JCO.2020.38.15_suppl.7502
  4. DiNardo CD, Schuh AC, Stein EM, et al. Effect of enasidenib (ENA) plus azacitidine (AZA) on complete remission and overall response versus AZA monotherapy in mutant-IDH2 (mIDH2) newly diagnosed acute myeloid leukemia. J Clin Oncol. 2020;38(15 suppl):7501. doi:10.1200/JCO.2020.38.15_suppl.7501
Related Videos
Tycel Phillips, MD
Ajai Chari, MD
A panel of 4 experts on hematologic malignancies
A panel of 4 experts on chronic graft-versus-host disease
A panel of 4 experts on chronic graft-versus-host disease
Alexey Danilov, MD, PhD
Hayder Saeed, MD
Corey Cutler, MD, MPH, and Hana Safah, MD, experts on GvHD
Corey Cutler, MD, MPH, and Hana Safah, MD, experts on GvHD
James K. McCloskey, MD, an expert on leukemia