Ziftomenib plus 7+3 yielded high CR and MRD-negativity rates with manageable safety in newly diagnosed NPM1/KMT2A-mutated AML.
Ziftomenib (Komzifti) combined with 7+3 showcased favorable tolerability and robust activity in the form of deep and durable responses in patients with newly diagnosed NPM1-mutated and KMT2A-rearranged
In terms of safety, any grade 3 or higher treatment-emergent adverse effects (TEAEs) occurred in 96% of all patients (n = 99), and 53% of TEAEs were determined to be related to ziftomenib. Presenting author Amer M. Zeidan, MBBS, MD, noted that the toxicities observed are in line with what is seen with 7+3 and primarily include blood count suppression, febrile neutropenia, some gastrointestinal effects, and pruritus. The addition of ziftomenib to 7+3 did not appear to result in increased risk of complications vs 7+3 historical controls, he added, and no new or unexpected AEs were observed.
The complete response (CR) rate, defined as a CR with full count recovery, achieved with the combination was 94% in patients with NPM1-mutated disease (n = 49) and 82% in those with KMT2A-rearranged disease (n = 50); in all patients, this rate was 88%. The composite CR (CRc) rates across these groups were 96%, 90%, and 93%, respectively. The objective response rates (ORRs) were 98%, 92%, and 95%, respectively. The median duration of CR was not reached (NR) in the NPM1-mutated cohort and 12.0 months (95% CI, 6.0-not evaluable [NE]) in the KMT2A-rearranged cohort.
At a median follow-up of 17.6 months (range, 1.0-23.5) in the NPM1-mutated cohort, the median overall survival (OS) was NR (95% CI, NR-NR), and the 12-month OS rate was 94%. Moreover, the 60-day mortality rate was 2% (n = 1). At a median follow-up of 11.0 months (range, 0.9-21.9) in the KMT2A-rearranged cohort, the median OS was also NR (95% CI, 12.8 months-NE); the 12-month OS rate was 71%. The 60-day mortality rate in this group was 4% (n = 2).
“This phase 1 study, the KOMET-007 study, supports the combination of ziftomenib with 7+3 to be safe, well tolerated, and associated with excellent preclinical activity,” Zeidan said in the presentation. “The AEs of special interest, such as differentiation syndrome and QTC [prolongation], are very well mitigated. It does not seem that ziftomenib adds additional myelosuppression, as evidenced by the low incidence of severe infectious complications, as well as the excellent time to count recovery. On the efficacy front, the CR rates are exceptionally high for [those with] NPM1[-mutated disease].”
Zeidan is chief of the Division of Hematologic Malignancies at the Yale School of Medicine; as well as director of Early Therapeutics Research, director of the Leukemia and Myeloid Malignancies Program, assistant medical director of the Clinical Trials Office, and co-leader of the Leukemia and Myeloid Malignancies Clinical Research Team at Yale Cancer Center in New Haven, Connecticut.
It is known that NPM1 mutations and KMT2A rearrangements drive leukemogenesis in approximately 35% to 40% of all AML cases. Ziftomenib is an oral menin inhibitor, which “are among the most exciting drugs we have,” Zeidan said.
In November 2025, the
At the time of the decision, Harry Erba, MD, PhD, told OncLive®:3 “In our study, we had patients achieving remission after single-agent ziftomenib who were able to then transition and proceed with allogeneic transplant, which would be the only potentially curative option for this group of patients.”
LEARN MORE: You can listen to the
“[Ziftomenib] is now being explored based on preclinical data suggesting synergy in combination with standard therapies for AML with NPM1 [mutations] and KMT2A [rearrangements],” Zeidan noted in his presentation.1
The multicohort, open-label, dose-escalation and -expansion study was launched to evaluate ziftomenib in combination with standard therapies in adult patients with newly diagnosed or relapsed/refractory NPM1-mutated or KMT2A-rearranged AML.
Previous data from the study revealed that when
At EHA 2026, Zeidan shared longer follow-up data from the portion of the trial evaluating ziftomenib combined with 7+3 in adult patients with newly diagnosed AML and NPM1 mutations or KMT2A rearrangements.1
Ziftomenib was administered on day 8 of cycle 1 and continuously thereafter throughout induction and consolidation. Then, patients were permitted to receive maintenance therapy with the drug either after completion of consolidation chemotherapy or after transplant, Zeidan explained. Patients received standard doses of 7+3; cytarabine was given on days 1 through 7 of cycle 1, and daunorubicin was given on days 1 through 3 of cycle 1. For the dose-escalation (phase 1a) portion of the study, patients started with 200 mg of ziftomenib once daily for dose level 1; dose level 2 was 400 mg once daily, and dose level 3 was 600 mg once daily.
“What I’m presenting today is [findings with] the 600-mg dose, which is what has been approved [for the] monotherapy [regimen], but also what has been chosen as the optimal dose for expansion,” Zeidan noted.
The primary end points of the analysis were AE, dose-limiting toxicities in the phase 1a portion, and CR. Secondary end points comprised CRc, ORR, and duration of response (DOR).
As of the data cutoff of April 10, 2026, 23 patients in the NPM1-mutated cohort were still receiving ziftomenib treatment; 10 had received on-study transplant, and 31 were receiving maintenance ziftomenib (8 of whom were being treated post-transplant). A total of 26 patients had discontinued treatment, mostly due to relapsed/refractory disease (n = 8) or physician decision (n = 8). In the KMT2A-rearranged cohort, 20 patients were still receiving ziftomenib; 32 had received on-study transplant, and 22 were receiving ziftomenib maintenance (21 after transplant). Thirty total patients had discontinued treatment, and the primary reason was due to relapsed/refractory disease (n = 10).
Ninety percent of patients in the NPM1-mutated cohort and 62% of those in the KMT2A-rearranged cohort were still on the study.
In all patients, the median age was 53 years (range, 18-71).
“The NPM1-mutated cohort was a little bit older than what you expect,” Zeidan said. “The median age was 60 years [range, 30-71], and that’s because the dose-escalation part of the study, which had the 600-mg dose as the final dose, actually was required to [enroll] either older patients, or [those with] adverse risk or therapy-related disease, meaning that those patients are higher risk than the typical NPM1-mutated patients, which is very important to note.”
A little more than half of all patients were female (57%), and most were White (69%) and had an ECOG performance status of 0 or 1 (93%). Moreover, 14% of patients had FLT3 co-mutations, and 20% of patients had IDH1/2 co-mutations. Twelve percent of patients had therapy-related AML.
The most common grade 3 or higher TEAEs reported in all patients included febrile neutropenia (63%), thrombocytopenia (60%), anemia (37%), neutropenia (31%), leukopenia (28%), hypokalemia (15%), sepsis (13%), lymphopenia (14%), increased alanine aminotransferase level (11%), and pruritus (10%).
The grade 3 or higher AEs of interest included differentiation syndrome and QTc prolongation. There were 4 total cases of grade 3 differentiation syndrome; 1 was in the NPM1-mutated cohort, and 3 were in the KMT2A-rearranged cohort. All events were resolved with protocol-specified mitigation, and 3 of the 4 patients continued to receive ziftomenib.
“This is very important, because the combination with 7+3 not only increases the efficacy, but it also reduces the risk of differentiation syndrome, which is something that we have been paying particular attention to for menin inhibition,” Zeidan said.
A total of 3 grade 3 cases of QTc prolongation were observed per investigator assessment; 1 was in the NPM1-mutated cohort, and 2 were in the KMT2A-rearranged cohort. None of these cases were determined to be related to ziftomenib; all resolved, and all patients continued treatment with the menin inhibitor.
Moreover, time to neutrophil and platelet recovery was comparable with that observed with intensive chemotherapy regimens, Zeidan added.
“Confirming the relative safety of this combination is that the time to count recovery was actually in line with what you see with 7+3,” he explained. ‘[The median times] to absolute neutrophil count [of at least 1.0 x 109/L] and platelet recovery [to at least 100 x 109/L were both] 28 days, and again, that suggests that the combination with ziftomenib is not adding additional myelosuppression.”
In the NPM1-mutated cohort, the CR minimal residual disease (MRD) negativity rate by local testing was 85%, and the CRc MRD negativity rate was also 85%; the median times to CR MRD and CRc MRD negativity were both 1.5 months (range, 0.5-12.2). In the KMT2A-rearranged cohort, the CR MRD negativity rate was 86%, and the CRc MRD negativity rate was 82%; the median times to CR MRD negativity and CRc MRD negativity were both 0.9 months (range, 0.5-2.8). In all patients, the CR MRD and CRc MRD negativity rates were 85% and 84%, respectively, and the median times to both were 1.2 months (range, 0.5-12.2) and 1.1 months (range, 0.5-12.2), respectively.
For the first time, Zeidan also reported on central molecular MRD negativity in the NPM1-mutated cohort. At a threshold of less than .01%, the CRC MRD negativity rate was 56%, and the median time to negativity was 2.3 months (range, 0.8-3.1). At the lower, less sensitive threshold of less than .1%, the CRc MRD negativity rate was 79%, and the median time to negativity was 2.2 months (range, 0.7-2.6). All this was achieved within 2 cycles of treatment, Zeidan noted.
“[These] data nicely support the ongoing phase 3 trial of this combination, which is KOMET-017 [NCT07007312],” Zeidan concluded. “[This] has 2 trials in 1 protocol. Both of them are phase 3, independently powered, registration-enabling trials that will enroll [approximately] 1300 patients between them.”
The non-intensive therapy portion includes patients with NPM1-mutated AML who are at least 75 years of age or 18 to under 75 years of age with comorbidities. They are randomly assigned 1:1 to receive ziftomenib plus venetoclax and azacitidine or placebo and venetoclax plus azacitidine. The primary end points are CR and OS.
The intensive portion includes patients with NPM1-mutated and KMT2A-rearranged AML who are FLT3 wild-type or have an ITD ratio below .05. They must be at least 18 years of age and have an ECOG performance status that ranges from 0 to 2. These patients are randomly assigned 1:1:1 to receive ziftomenib continuously for induction and consolidation with 7+3 induction followed by maintenance with ziftomenib, ziftomenib continuously for induction and consolidation with 7+3 induction followed by placebo maintenance, or placebo continuously for induction and consolidation with 7+3 induction followed by placebo maintenance. The primary end points are CR MRD negativity in the NPM1-mutated cohort and event-free survival.
