Posttransplant Dordaviprone Maintenance Is Safe With Low Cytopenia Rates in High-Risk AML and MDS

Maintenance therapy with the first-in-class imipridone dordaviprone (Modeyso) administered after allogeneic hematopoietic stem cell transplant (HSCT) was safe, generating no dose-limiting toxicities (DLTs), no graft failures, and low rates of grade 3/4 cytopenias in patients with high-risk acute myeloid leukemia (AML) or myelodysplastic neoplasm (MDS), according to findings from a phase 1 trial (NCT03932643) published in the American Journal of Hematology.¹

Among the 20 patients treated, no DLTs were observed at doses up to the maximum weekly dose of 625 mg, and no patients discontinued treatment due to adverse effects (AEs).

At 2 years, the relapse-free survival (RFS) rate was 60% (95% CI, 42%-86%), and the overall survival (OS) rate was 70% (95% CI, 53%-93%). The cumulative incidence of non-relapse mortality was 6.2%, and the cumulative incidence of relapse was 41.9% at 2 years.

Grade 3 or higher AEs occurred in 45% of patients, including 15% who experienced treatment-related AEs. Grade 3/4 cytopenias comprised anemia (15%), thrombocytopenia (15%), and neutropenia (10%). Grade 3/4 acute graft-vs-host disease (GVHD) was reported in 5% of patients.

“The use of dordaviprone [following HSCT] was well tolerated with no observed DLTs and a manageable safety profile,” lead study author Vijaya Raj Bhatt, MD, of the University of Nebraska Medical Center in Omaha, and coauthors wrote in the publication. “The 2-year OS of 70% in this high-risk population is encouraging, particularly in the context of using a weekly oral drug.”

What was the rationale for investigating dordaviprone post-HSCT in AML and MDS?

Relapse after allogeneic HSCT occurs in up to 40% to 60% of patients with high-risk AML and MDS and remains the principal cause of posttransplant mortality, the study authors noted. Randomized evidence supporting posttransplant maintenance has been limited largely to FLT3 inhibition or a combination of low-dose decitabine and granulocyte colony-stimulating factor, and maintenance approaches carry a substantial risk of myelosuppression and other nonhematologic toxicity.

Dordaviprone, formerly known as ONC201, is a first-in-class small-molecule imipridone that acts as a selective antagonist of the dopamine receptor DRD2 and an allosteric agonist of the mitochondrial protease caseinolytic protease P. In preclinical models, it has demonstrated activity against leukemia cells harboring a TP53 mutation or complex karyotype and against leukemia stem cells, while sparing normal bone marrow cells.

Dordaviprone is currently approved by the FDA for the treatment of adult and pediatric patients 1 year of age and older with diffuse midline glioma harboring an H3 K27M mutation with progressive disease following prior therapy.²

How was the phase 1 study conducted?

The single-center, phase 1 trial was conducted at the Fred & Pamela Buffett Cancer Center at the University of Nebraska Medical Center and enrolled 20 adult patients with high-risk AML or MDS who had undergone allogeneic HSCT within the preceding 6 to 20 weeks.¹ Key eligibility criteria included a Karnofsky performance status of at least 70%, less than 5% bone marrow blasts, an absolute neutrophil count above 1000/μL, and a platelet count of at least 50,000/μL.

High-risk disease was defined by adverse-risk AML per 2017 European LeukemiaNet criteria, poor- or very poor-risk MDS per the revised International Prognostic Scoring System, or response-based features such as relapsed or refractory disease or measurable residual disease (MRD). Patients with a history of grade 3/4 acute GVHD or those receiving prednisone at 0.25 mg/kg per day or higher were excluded.

Using a 3+3 design, patients received oral dordaviprone once weekly at a starting dose of 250 mg, with escalation by 125-mg increments to a maximum of 625 mg weekly in the absence of DLTs. Weekly dosing continued for up to 13 four-weekly cycles. Ten patients received doses of 250 mg to 500 mg (one of whom was unevaluable for DLT after stopping the drug before the DLT window), and the remaining 10 patients received 625 mg. Patients received a median of 8.5 cycles (range, 1-13). GVHD prophylaxis consisted of tacrolimus and methotrexate in all patients, with low-dose antithymocyte globulin used in 80% of patients.

The primary end points were the rate of DLTs during the first cycle and the rate of grade 3 or higher AEs. Secondary end points included overall AEs, relapse, RFS, non-relapse mortality, and OS, with disease relapse and mortality captured over 2 years. Exploratory end points assessed changes in health-related quality of life and functional status.

Patients had a median age of 68 years (range, 39-75), 40% were older than 70 years, and 60% were men. Sixty percent had AML and 40% had MDS; among patients with AML, 50% had adverse-risk disease, and 41% had MRD before HSCT, and among patients with MDS, 75% had very poor–risk disease. A TP53 mutation was present in 36% of patients and a monosomal karyotype in 35%; other common high-risk mutations included RUNX1 (26%), BCOR (21%), U2AF1 (16%), STAG2 (15%), and ASXL1 (10%). Sixty percent of patients had an HSCT comorbidity index of at least 3, and 40% had impaired instrumental activities of daily living at enrollment. Seventy percent received reduced-intensity conditioning and 30% received myeloablative conditioning; all patients received peripheral blood grafts, mostly from matched unrelated (55%) or matched sibling (35%) donors.

What additional data were reported?

The median follow-up was 26.5 months (range, 8-31). Seven patients relapsed—2 with AML and 5 with MDS—and all patients with relapsed MDS had very high-risk features such as a TP53 mutation, a monosomal karyotype, or both. Among 7 patients with TP53-mutated AML or MDS, 5 relapsed, and 3 were alive at 2 years. Ten of 12 patients with AML did not relapse, including many with MRD before transplant or one or more high-risk mutations.

Over 2 years, 6 patients died, most (n = 5) from the underlying cancer; 1 patient died of late-onset acute liver GVHD following a tacrolimus taper that the investigators deemed unlikely to be related to the study drug. Any-grade AEs occurred in 95% of patients (70% related), and serious AEs were reported in 40% (15% related).

The most common grade 3 or higher AEs occurring in more than 5% of patients, beyond cytopenias, included generalized muscle weakness (10%) as well as supraventricular tachycardia, fall, hypertriglyceridemia, hypotension, thromboembolic event, and increased alanine aminotransferase, alkaline phosphatase, and aspartate aminotransferase (5% each). Most patients maintained or improved measures of physical function from baseline to 3 months, with stabilization or improvement in Karnofsky performance status, instrumental activities of daily living, and short physical performance battery in 95%, 80%, and 79% of patients, respectively.

The study authors noted that the findings support further clinical testing of dordaviprone to prevent posttransplant relapse, potentially preferentially in high-risk AML based on the early signal in that population, with a separate trial in MDS considered subsequently. They added that a future randomized trial will be needed to establish efficacy and should consider TP53 mutation and monosomal karyotype status as stratification factors, and that combination strategies—such as with azacitidine, given preclinical synergy—warrant evaluation, particularly for disease harboring a TP53 mutation or monosomal karyotype.

“Given its favorable safety profile, weekly oral dosing, and preclinical synergy with other drugs such as azacitidine, this trial demonstrates that dordaviprone is safe for further clinical testing to prevent posttransplant relapses of AML and MDS,” the study authors concluded.

References

1. Bhatt VR, Wichman CS, Bouska A, et al. Dordaviprone maintenance after allogeneic HCT for high-risk acute myeloid leukemia and myelodysplastic neoplasm. Am J Hematol. 2026;101(9):2288-2296. doi:10.1002/ajh.70428
2. Modeyso. Prescribing information. Updated January 2026. Accessed August 17, 2026. https://pp.jazzpharma.com/pi/modeyso.en.USPI.pdf