Ziftomenib Shows Clinically Meaningful Activity in Heavily Pretreated NPM1‑Mutated AML — A Targeted Option for a Molecularly Defined Relapse Population

Highlights

– Ziftomenib (600 mg once daily) met the primary endpoint in the KOMET‑001 phase II cohort for relapsed/refractory NPM1‑mutated AML: CR/CRh 22% (95% CI, 14–32; P = .0058).

– Overall response rate was 33% with a median response duration of 4.6 months; 61% of responders were MRD‑negative by the study’s assays.

– Safety was manageable: common grade ≥3 events included febrile neutropenia (26%), anemia (20%), and thrombocytopenia (20%); differentiation syndrome occurred in 25% (15% grade 3) and was treatable per protocol.

Background and Unmet Need

Acute myeloid leukemia (AML) is a heterogenous disease driven by diverse genetic abnormalities. NPM1 mutations define a distinct molecular subtype present in ~25–30% of adult AML and are associated with unique biology including aberrant HOX gene expression and a propensity to respond to conventional induction chemotherapy in first presentation. However, for patients with relapsed or refractory (R/R) NPM1‑mutated disease, outcomes are poor and effective salvage options are limited. There is a high unmet need for targeted agents that exploit the specific biology of NPM1‑mutated leukemias, especially in older or heavily pretreated patients who are ineligible for intensive approaches or transplantation.

Biological Rationale for Menin Inhibition

Menin is a scaffold protein that interacts with mixed‑lineage leukemia (MLL/KMT2A) complexes and sustains expression of HOX and MEIS1 genes, which drive leukemogenesis in MLL‑rearranged and NPM1‑mutated AML. In preclinical models, disruption of the menin–MLL interaction suppresses HOX/MEIS1 transcriptional programs, induces differentiation, and impairs leukemic propagation. This mechanism provides a compelling, pathway‑directed approach for patients whose leukemias are addicted to HOX/MEIS1 signaling, such as NPM1‑mutant AML. Ziftomenib is a potent, selective oral menin inhibitor developed to translate this biology into clinical benefit.

Study Design — KOMET‑001 Phase II (Registration‑Enabling) Overview

KOMET‑001 (ClinicalTrials.gov NCT04067336) is a first‑in‑human program of ziftomenib. After dose‑finding in phase I, the registration‑enabling phase II cohort enrolled patients with relapsed or refractory NPM1‑mutated AML to receive ziftomenib 600 mg once daily as monotherapy. Key eligibility criteria included documented NPM1 mutation and prior treatment exposure; patients were heavily pretreated, with a substantial proportion having received prior venetoclax. The primary end point for the phase II cohort was rate of complete remission with full hematologic recovery (CR) or CR with partial hematologic recovery (CRh). Secondary endpoints included overall response rate (ORR), duration of response (DoR), overall survival (OS), measurable residual disease (MRD) negativity among responders, and safety.

Key Results

Between January 26, 2023, and May 13, 2024, 92 patients (median age 69 years, range 33–84) with R/R NPM1‑mutated AML were treated in the phase II cohort. The primary and key secondary outcomes were:

• CR/CRh rate: 22% (95% CI, 14–32), P = .0058 versus the prespecified null; this met the primary endpoint.
• Overall response rate (CR+CRh+CRi+PR): 33% (95% CI, 23–43).
• Among responders, 61% were MRD‑negative by the trial assays, suggesting deep molecular remissions in a majority of patients who achieved CR/CRh.
• Median duration of response: 4.6 months (95% CI, 2.8–7.4).
• Median overall survival (intent‑to‑treat): 6.6 months (95% CI, 3.6–8.6).

Subgroup analyses prespecified in the protocol showed that CR/CRh rates were generally consistent irrespective of prior therapies — including prior venetoclax exposure — and independent of broad comutation patterns. This suggests that activity of ziftomenib was not restricted to a narrow prior‑therapy or co‑mutation subgroup within NPM1‑mutated AML.

Safety and Tolerability

Ziftomenib was generally well tolerated in this older, pretreated population. The most common grade ≥3 treatment‑emergent adverse events were febrile neutropenia (26%), anemia (20%), and thrombocytopenia (20%). Differentiation syndrome (DS), an expected on‑target effect reflecting rapid myeloid maturation, occurred in 25% of patients; 15% experienced grade 3 DS and there were no grade 4–5 DS events reported. DS events were manageable with the study’s predefined monitoring and mitigation strategies (e.g., corticosteroids, hydroxyurea, supportive care). Importantly, investigators observed no clinically significant QTc prolongation and a low rate of permanent discontinuation due to ziftomenib‑related adverse events (3%).

Interpretation of Efficacy Outcomes

The observed 22% CR/CRh rate is clinically meaningful in the context of R/R NPM1‑mutated AML, a population with limited salvage options and poor prognosis. MRD negativity in 61% of responders is notable because MRD clearance has been associated with improved relapse‑free survival in multiple AML settings and indicates depth of response beyond marrow blast reduction. However, the median DoR of 4.6 months and median OS of 6.6 months underscore that while ziftomenib induces objective and deep responses in a subset, durability remains limited for many, reflecting either disease biology, emergence of resistance, or challenges in delivering subsequent consolidative therapies (e.g., stem cell transplantation) in an older cohort.

Safety Context and Management Considerations

The safety profile aligns with expectations for an agent that induces differentiation of leukemic blasts. Differentiation syndrome is a class effect of therapies that induce rapid leukemic maturation (well described with APL therapy and now seen with menin inhibitors). The incidence (25%) and severity (15% grade 3; no grade 4–5) reported here indicate that DS is common but manageable with early recognition and protocolized treatment such as corticosteroids, temporary treatment delay, and supportive measures. Hematologic toxicities and febrile neutropenia were expected in an R/R AML population and require proactive infection prophylaxis and monitoring. The low discontinuation rate (3%) due to drug‑related AEs supports tolerability in a real‑worldish older cohort.

Strengths of the Study

• Molecularly targeted approach conducted in a genetically defined population (NPM1‑mutated), increasing biological plausibility of observed effects.
• Registration‑enabling phase II design with a prespecified primary endpoint and statistical plan that the study met.
• High rates of MRD negativity among responders, indicating depth of response beyond hematologic remission.
• Manageable and predictable toxicity profile with low treatment discontinuation.

Limitations and Unanswered Questions

• Single‑arm design without randomized comparator limits assessment of relative benefit versus standard salvage regimens or best supportive care; historical comparisons are imperfect in a heterogenous R/R AML population.
• Median DoR and OS remain modest; mechanisms of acquired resistance to menin inhibition in NPM1‑mutated AML need elucidation.
• Follow‑up remains limited for long‑term outcomes, including post‑response pathways such as bridging to transplant and long‑term survival among MRD‑negative responders.
• Real‑world applicability to younger, fitter patients or to frontline settings requires separate evaluation.

Clinical and Research Implications

Ziftomenib provides proof‑of‑concept that targeting the menin–MLL axis can yield clinically meaningful responses in R/R NPM1‑mutated AML. Practical implications include:

• Consideration of molecular testing for NPM1 and related markers at relapse to identify candidates for targeted menin inhibition in clinical trials or future approvals.
• Need for standardized DS monitoring and management protocols wherever menin inhibitors are used, including education of treating teams about early corticosteroid use and rapid supportive care initiation.
• Rationale for combination strategies: preclinical data support synergism of menin inhibition with hypomethylating agents, BCL‑2 inhibitors (venetoclax), or other targeted therapies; combining ziftomenib with such agents may deepen and prolong responses and merits controlled trials.
• Potential role as a bridge to allogeneic hematopoietic stem cell transplantation in responders, particularly those achieving MRD negativity, but prospective evidence is required.

Future Directions

Key next steps include randomized trials comparing ziftomenib (alone or in combination) with standard salvage regimens, exploration of combination regimens aimed at improving durability, investigation of mechanisms of resistance (genomic and transcriptomic profiling of relapses), and trials in earlier disease settings including frontline therapy for patients unfit for intensive induction. Biomarker research to predict which NPM1‑mutant patients will derive the most durable benefit (beyond presence of NPM1 mutation alone) is also critical.

Conclusion

The registration‑enabling cohort of KOMET‑001 demonstrates that ziftomenib, an oral menin inhibitor, delivers clinically meaningful responses in a difficult‑to‑treat population of relapsed/refractory NPM1‑mutated AML, with a manageable safety profile dominated by anticipated differentiation syndrome and hematologic events. While median response durability and survival remain limited for many patients, the observed depth of response (including frequent MRD negativity) and tolerability justify further development in randomized and combination studies and position menin inhibition as a promising targeted strategy in molecularly defined AML.

Funding and Trial Registration

The KOMET‑001 trial is registered at ClinicalTrials.gov (NCT04067336). Funding sources and disclosures are reported in the primary publication (Wang et al., J Clin Oncol 2025).

References

1. Wang ES, Montesinos P, Foran J, et al; KOMET‑001. Ziftomenib in Relapsed or Refractory NPM1‑Mutated AML. J Clin Oncol. 2025 Nov;43(31):3381‑3390. doi: 10.1200/JCO‑25‑01694. Epub 2025 Sep 25. PMID: 40997296; PMCID: PMC12573682.

2. Papaemmanuil E, Gerstung M, Bullinger L, et al. Genomic Classification and Prognosis in Acute Myeloid Leukemia. N Engl J Med. 2016 Jun 9;374(23):2209‑21. doi:10.1056/NEJMoa1516192. PMID: 27160978.

Author note

This article synthesizes the principal findings of the KOMET‑001 phase II cohort and places them in clinical and translational context. Clinicians should refer to the full peer‑reviewed manuscript and protocol for detailed methods, subgroup definitions, MRD assay characteristics, and safety management algorithms before applying these observations to practice.