Ziftomenib Expands Role in Relapsed/Refractory NPM1-Mutated AML

Acute myeloid leukemia (AML) is an aggressive hematologic malignancy, and NPM1 mutations, present in approximately 30% of cases,¹ are among the most common genetic alterations driving the disease. Although patients with NPM1-mutated AML often respond to frontline chemotherapy, relapse remains common, and treatment options for relapsed/refractory disease, particularly among older or unfit patients who cannot tolerate intensive regimens, have remained limited.

The emergence of menin inhibitors, including ziftomenib (Komzifti), has introduced a targeted approach for this molecular subset, and investigators have since sought to build on single-agent activity by pairing ziftomenib with existing lower-intensity backbones to improve depth and durability of response. One such effort is the phase 1/2 KOMET-007 study (NCT06097832), data from which were recently published in Blood evaluating ziftomenib in combination with venetoclax (Venclexta) and azacitidine in a cohort of patients with relapsed/refractory NPM1-mutated AML.²

In an interview with Targeted Oncology, Eunice S. Wang, MD, chief of the Leukemia Service at Roswell Park Comprehensive Cancer Center and first author of the Blood publication, provided a comprehensive overview of KOMET-007 and described implications for sequencing and the evolving role of menin inhibition in AML treatment.

Targeted Oncology: Could you provide a general background of the KOMET-007 study, including the rationale and an overview of the study design?

Eunice S. Wang, MD: KOMET-007 is a study that’s designed to explore the efficacy and safety of the combination of the menin inhibitor ziftomenib plus different chemotherapy backbones. As we know, ziftomenib is a menin inhibitor which was approved last year for the treatment of patients [with AML] whose disease is characterized by a mutation in the NPM1 gene. In the relapsed/refractory setting, ziftomenib monotherapy leads to complete remission or complete remission without count recovery [CR/CRc] in about 23% of patients with a median duration of overall survival [OS] in the 3- to 4-month range. This doesn’t seem like much, but this is certainly something that was impressive given that this is a targeted therapy that is directed towards the underlying biology of disease in this particular subtype.

Obviously, we were seeking to improve upon these outcomes, and the publication in Blood in June 2026 reported the outcome of a cohort on the KOMET-007 study, which investigated the combination of ziftomenib plus a lower intensity backbone, venetoclax-azacitidine, for patients with relapsed/refractory [AML].

In this study, we examined 67 patients with the standard venetoclax-azacitidine backbone and treated them with the combination of chemotherapy and ziftomenib. There was a small dose escalation in 27 patients, where we gave them doses of ziftomenib ranging from 200 mg, 400 mg, to 600 mg and determined that the safest and potentially most effective dose was 600 mg. Subsequently, an additional 40 patients were treated with ziftomenib 600 mg and venetoclax-azacitidine, and that is the FDA-approved dose as a single-agent therapy. The key end point was looking at response rates, as well as duration of response, and the safety end points were looking at safety, tolerability, and adverse events.

What were the key efficacy and safety findings described in the Blood publication?

What we found is that there was an overall CRc rate of 48%, two-thirds of which were MRD-negative, and we saw particularly impressive results in…patients who had not received prior venetoclax-azacitidine who were then treated with the combination of ziftomenib and venetoclax-azacitidine in this clinical trial. In that particular group of patients, the overall response rate was over 80% with a CRc rate of 70%, and three-quarters of them were MRD [minimal residual disease] negative, and this was compared [with] very low responses with a CRc rate of only 24% in patients who had previously been exposed to a portion of the regimen, which is prior venetoclax therapy.

We found that the drug was very well tolerated. About one-third of patients had low blood counts. About 25% of patients had febrile neutropenia and neutropenia [each], and the special toxicities that we noted with ziftomenib monotherapy, ie, differentiation syndrome, were only found in 2 patients out of the 67. Those 2 patients had grade 3 events, but those were successfully mitigated with protocol-specified interventions.

So overall [we saw] improved efficacy—the median duration of response had increased from about 4 months with the ziftomenib monotherapy to 8.6 months, which was almost a doubling of the [OS], and we think that this new strategy with the combination didn’t add significantly to the toxicities of that backbone therapy alone.

What do the stronger outcomes in venetoclax-naive patients tell us about the optimal sequencing of therapies in NPM1-mutated AML?

I think for NPM1-mutant patients, if for example they receive treatment upfront with an intensive chemotherapy backbone that does not contain venetoclax, that this particular triplet regimen—venetoclax, azacitidine, and ziftomenib—would be a great option with extremely high response rates…. However, in patients who have previously received a lower intensive venetoclax-azacitidine regimen, just adding the ziftomenib [to] this particular regimen may or may not improve upon ziftomenib monotherapy alone, because the only new part of that triplet regimen is the ziftomenib, so that makes sense. If you’ve already gotten the key 2 of the 3 drugs and you failed, then you’re going to add the one drug, so it’s similar to just adding the one drug alone.

What this suggests to us is that the best efficacy, given the high efficacy that we saw with the venetoclax-naive patients, is to bring this for the older, unfit patients who are not eligible for intensive chemotherapy…into the upfront setting. In those venetoclax-naive patients, you have an overall response of 80%; that’s pretty good. That’s better than venetoclax-azacitidine alone, potentially, in that upfront setting. So, I think that is the next step, and this is currently occurring in a phase 3 trial that is accruing [KOMET-017; NCT07007312], and there’s a subsequent cohort of the same trial, KOMET-007, that was done in newly diagnosed patients. I think we’re looking forward to both of the results, both this single cohort in the newly diagnosed setting, as well as the phase 3 study, which is examining in a randomized setting ziftomenib plus venetoclax-azacitidine vs venetoclax-azacitidine alone to see whether this could successfully make its way into upfront therapy for older unfit individuals.

How do you see MRD evolving as a treatment goal in NPM1-mutated AML?

It’s interesting because in this particular subset of AML…we have a lot of data suggesting that when you treat younger, fit patients [who] have this particular subset of disease, MRD negativity using a highly sensitive qPCR for that particular mutant NPM1 gene can predict outcomes to intensive chemotherapy.³ These data suggest that we might be able to translate that particular biomarker into less intensive chemotherapy regimens and potentially use it as a surrogate even for our patients that are treated with less intensive therapies.

That MRD negativity is something that we’re not only looking at in terms of overall prognosis to chemotherapy—high or low [intensity of] chemotherapy—but also something that we’re actively exploring, and that our colleagues use in the transplant world to predict outcomes and risk of relapse following allogeneic stem cell transplantation. With current developments and advances in stem cell technology, more and more of our older patients are able to undergo an allogeneic stem cell transplantation. Those transplanters like to see that patients are MRD negative, because going into the transplant, you want patients to have as little disease as possible so that you can reduce the risk of subsequent recurrence.

MRD negativity is increasingly, particularly for this subset, being focused on in our phase 3 study looking at intensive chemotherapy backbones. The plan is to use the MRD negativity as a known biomarker of response in those patients and explore whether we could use that same biomarker similarly in this type of regimen, less intensive therapy with or without transplant.

How do you see the role of menin inhibitiors evolving over the next several years?

There’s definitely evidence that menin inhibitors are effective not only for patients [with] NPM1-mutant [AML], as we’ve seen with ziftomenib, but also in patients who have a much [rarer] type of acute leukemia, those characterized by rearrangements in the KMT2A or MLL gene, and in that subset there also [are] very positive studies using revumenib [Revuforj], which is a different menin inhibitor for patients. Revumenib also has activity in patients [with] NPM1-mutant [AML], as we’ve seen here, so we do think that menin inhibitors as a class of agents [have] already become a standard, and increasingly going to become a more important addition to our armamentarium for treatment of newly diagnosed patients [with AML].

With the addition of menin inhibitors, as well as the existence of other targeted therapies—IDH1, IDH2, [and] FLT3 inhibitors—we now have targeted therapies for well over half of patients with newly diagnosed AML, and that is a great thing; it’s making precision medicine for this particular cancer type a reality, and emphasizes the importance of doing a thorough and extensive next-generation sequencing or molecular characterization at the time of the disease diagnosis and relapse to better target these lesions and prolong survival.

We’re starting to see improvements, we hope, in the overall survival of patients with the addition of all these agents, and I think menin [inhibition] is an important step in the right direction for precision medicine. [Ziftomenib] also is an oral agent, it’s well tolerated overall, it doesn’t seem to add a lot of toxicity, as in this study, to our standard low-intensity backbone, and that makes it an option for the majority of our patients who present with this disease; the median age is 70 and rising. We’re seeing a lot of secondary AMLs from other cancers, so I think these new drugs are an important step forward.

REFERENCES

1. Diamantidis MD, Vlachou MS, Katsikavela A, Kalomoiri S, Bartzi V, Ikonomou G. NPM1-mutated AML: deciphering the molecular and clinical puzzle in the era of novel treatment strategies. Cancers (Basel). 2025;17(13):2095. Published 2025 Jun 23. doi:10.3390/cancers17132095

2. Wang ES, Erba HP, Zeidan AM, et al. Ziftomenib with venetoclax and azacitidine in relapsed/refractory NPM1-mutated acute myeloid leukemia. Blood. Published online June 2, 2026. doi:10.1182/blood.2026034043

3. Othman J, Tiong IS, O’Nions J, et al. Molecular MRD is strongly prognostic in patients with NPM1-mutated AML receiving venetoclax-based nonintensive therapy. Blood. 2024;143(4):336-341. doi:10.1182/blood.2023021579