Amir T. Fathi, MD:
For several decades, intensive induction chemotherapy with cytarabine and anthracycline (7+3) has been the standard of care for the management of fit patients with newly diagnosed AML.¹ However, substantial toxicity has been associated with this regimen, and treatment outcomes have remained suboptimal. For patients who are not eligible for intensive chemotherapy, the combination of venetoclax, a potent BCL2 inhibitor, with a hypomethylating agent, azacitidine, has become the standard of care based on the overall survival (OS) advantage over azacitidine alone demonstrated in the phase III VIALE-A study.² In fit patients with AML, multiple retrospective studies have suggested that the combination of a hypomethylating agent, such as azacitidine or decitabine, with venetoclax, may result in superior efficacy and better tolerability when compared with intensive chemotherapy.^3,4 Hence, this hypothesis was tested prospectively in the randomized phase II PARADIGM study.⁵

PARADIGM is a multicenter, randomized phase II trial that investigated azacitidine plus venetoclax vs conventional intensive induction chemotherapy for fit patients 18 years of age or older with previously untreated AML.⁵ Of importance, several subsets of patients with AML were excluded from participation in this trial, including patients with t(15;17), promyelocytic leukemia (PML)–retinoic acid receptor alpha fusion (otherwise known as the alteration that causes acute PML), patients with core binding factor (CBF) AML, and those with FLT3 ITD/TKD mutations (≥5% variant allele frequency). Patients younger than 60 years of age with disease harboring NPM1 mutations, those with mixed phenotype, and those harboring BCR/ABL1 fusions were also excluded from participation in this study.

Eligible patients were randomly assigned to receive azacitidine plus venetoclax (n = 86) or conventional induction chemotherapy, either 7+3 or CPX-351. Of note, the choice to receive either 7+3 or CPX-351 was established prior to randomization. All patients were stratified by age (≥65 vs <65) prior to randomization.

Following midtreatment bone marrow assessment on Days 13-16, patients on the conventional induction chemotherapy arm with residual disease had the option to receive a second induction treatment with 5+2 or CPX-351 followed by consolidation therapy with cytarabine or CPX-351 if response is achieved. On both treatment arms, patients were allowed to undergo HSCT if response was achieved. This is an important endpoint in terms of assessments in this study.

The primary endpoint was EFS, and the secondary endpoints included response, OS, measurable residual disease (MRD), hospitalization metrics, toxicity and quality of life.

Amir T. Fathi, MD:
The baseline characteristics were fairly well balanced between arms. The median age was 65 years (range: 23-79) on the azacitidine/venetoclax arm and 64 years (range: 28-79) on the induction chemotherapy arm. On both arms, the majority of the patients were White (approximately 80%). Approximately 6% were Black, approximately 8% were Hispanic and approximately 9% were Asian. At least 50% of the patients were ≥65 years of age on either arm. More than 90% of the patients had an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0 or 1, demonstrating that most of the patients on the study were fit. Approximately 70% of patients had adverse risk by European LeukemiaNet (ELN) 2022 classification, with only one third having favorable or intermediate risk. This can be explained by the exclusion criteria, which precluded the participation of patients with CBF AML, FLT3 ITD/TKD mutations, and those younger than 60 years of age with NPM1 mutations.

Amir T. Fathi, MD:
After a median follow-up of 21.9 months, the primary endpoint of EFS was met with a superior EFS of 14.6 months demonstrated on the azacitidine/venetoclax arm vs 6.15 months on the induction chemotherapy arm (HR: 0.57; P = .0022). The 1-year EFS rate was 53.4% for azacitidine/venetoclax vs 36% for intensive chemotherapy. In a multivariate analysis, the adjusted HR for EFS with azacitidine/venetoclax vs induction chemotherapy was significant despite adjustments for factors such as age, ELN 2022 risk, and certain mutations (HR: 0.66; P = .0231).

Amir T. Fathi, MD:
The ORRs were improved in the azacitidine/venetoclax arm vs the induction chemotherapy arm (P <.0001). Although the complete remission (CR) rate was numerically higher with azacitidine/venetoclax (55.8%) compared with induction chemotherapy (48.8%), there was no statistically significant difference between arms (P = .2227). There was a statistically significant improvement on the azacitidine/venetoclax arm (77.9%) vs the induction chemotherapy arm (53.5%), when the rates of CR, CR with partial hematologic recovery (CRh), and CR with incomplete blood count recovery (CRi) were combined (P = .0006).

Amir T. Fathi, MD:
A significantly higher proportion of patients proceeded to HSCT after receiving azacitidine plus venetoclax compared with intensive chemotherapy (n = 52 vs n = 34; P = .009). There was no statistically significant difference in OS between arms (21.5 months vs 18.6 months; P = .1873). Of note, proceeding to HSCT was an important factor that contributed to improvements in outcomes in both arms. In a multivariate analysis after adjustments for HSCT, the EFS improvement with azacitidine/venetoclax over induction chemotherapy remained significant (HR: 0.67; P = .0302).

Amir T. Fathi, MD:
At Week 2, it is not surprising that patients who received azacitidine/venetoclax reported significantly better quality of life  (P = .001), depression symptoms (P = .0007), and overall symptom burden (P = .0186) compared with those who received induction chemotherapy. In the first 30 days of treatment, patients who received azacitidine/venetoclax were less likely to require care in the intensive care unit (P = .003), and the index hospitalization duration was greater among patients who received induction chemotherapy (14.8 days vs 35.9 days; P <.001). During the first 6 months of treatment, the length of hospital stay was significantly higher among patients who received induction chemotherapy compared with those who received azacitidine/venetoclax (58.5 days vs 41 days; P <.001).

Amir T. Fathi, MD:
On either of the arms, treatment led to myelosuppression, including grade ≥3 febrile neutropenia, decreased platelet count, anemia, decreased neutrophil count, and decreased white blood cell and lymphocyte counts. The incidence of cytopenias on both arms was fairly similar. However, azacitidine/venetoclax was associated with lower rates of grade ≥3 infection (15.1% vs 20.8%; OR: 0.53; P = .04) and bleeding complications (1.3% vs 6.3%; OR: 0.16; P = .01) compared with induction chemotherapy. On the azacitidine/venetoclax arm, a lower 60-day mortality rate was also noted (0 vs 4.7%; P = .0802).

Courtney D. Dinardo, MD, MSCE:
The interaction between menin and KMT2A is a key dependency in acute leukemias resulting from KMT2A rearrangements, NPM1 mutations, or NUP98 rearrangements. The inhibition of this interaction is known to promote leukemic cell differentiation and apoptosis.^6-8 Of note, NUP98 fusions are associated with upregulated HOXA9 and susceptible to menin inhibition.⁸ Therefore, the SAVE trial investigated a nonintensive combination including a hypomethylating agent, venetoclax, and a menin inhibitor for patients with AML harboring NPM1 mutations, KMT2A rearrangements, or NUP98 rearrangements who are not eligible to receive intensive chemotherapy.⁹

The phase I/II trial investigated the efficacy and safety of the all oral combination of revumenib with decitabine/cedazuridine and venetoclax for patients 12 years of age or older with newly diagnosed or relapsed/refractory AML or myeloid mixed phenotype acute leukemia (MPAL) harboring NPM1 mutations, KMT2A rearrangement, or NUP98 rearrangements.⁹ To be eligible for participation, patients with newly diagnosed disease had to be ineligible to receive high-intensity chemotherapy. At ASH 2025, results from the cohort of patients with newly diagnosed AML were presented. It is important to note the uniqueness of the patient cohort in that it included children and young adolescents. Revumenib was dosed at 270 mg when administered without a strong CYP3A inhibitor, and at 160 mg when administered with a strong CYP3A inhibitor. All patients had to undergo bone marrow assessment on Day 14 of cycle 1, in addition to the end of cycle assessment. If bone marrow blasts were <5%, revumenib treatment was held after Day 21. This design allowed for count recovery and tolerability during that first cycle of treatment. Another interesting design of the study is that patients who proceeded to receive stem cell transplant (SCT) received maintenance revumenib for 1 year post SCT.

The primary endpoints in the phase I portion of the study were safety, maximum tolerated dose, and determination of the recommended phase II dose. For the phase II portion of the study, the primary endpoint was efficacy. Secondary endpoints included OS, recurrence-free-survival, and MRD.

Courtney D. Dinardo, MD:
At ASH 2025, data from 21 patients with newly diagnosed AML were presented. Out of the 21 patients, 14 had NPM1-mutant AML, and 7 had KMT2A-rearranged AML harboring 5 different KMT2A rearrangements. The median age was 70 years (range: 60-83), demonstrating an older, chemotherapy-ineligible population of patients. In the cohort of 14 patients with NPM1-mutant disease, comutations with NRAS/KRAS, FLT3, or IDH1/2 were reported. Of interest, 4 of the patients with NPM1-mutant AML also had FLT3 mutations, including 3 patients harboring FLT3 TKD mutations and 1 harboring FLT3 ITD mutations.

In the cohort of 7 patients with KMT2A-rearranged AML, 2 patients harbored comutations with NRAS/KRAS, and 1 patient harbored comutations with IDH1/2.

Courtney D. Dinardo, MD:
It is already recognized that the use of azacitidine plus venetoclax results in a 30% to 40% rate of cytopenias and cytopenia-related toxicities. As expected, the SAVE trial showed that 48% of the patients experienced febrile neutropenia (grade 4: 48%) and 38% experienced thrombocytopenia (grade ≥3: 33%). These safety data are at par with what is expected with the use of a hypomethylating agent in combination with venetoclax. Also, 43% of patients experienced QTc prolongation, all of which were grade 1/2. Differentiation syndrome, a known adverse event (AE) of menin inhibitors, was also reported in 19% of patients (grade ≥3: 10%).

Courtney D. Dinardo, MD:
Even though a sample size of 21 patients is small, an ORR of 86% demonstrated with the all-oral regimen studied in the SAVE trial was impressive. The CR/CRh rate was 81% in the overall population. Among patients with KMT2A-rearranged AML, a particularly adverse risk group of patients, the ORR was 86% with a CR/CRh rate of 86%. Also, the MRD negativity rate by flow cytometry (10^-4) was 86%.

These response rates are encouraging and deeper than what is normally seen in this patient population with the use of a hypomethylating agent and venetoclax alone without the addition of a third agent. Although the SAVE trial is not a randomized controlled trial and the sample size is small, these results are still impressive and encouraging.

There were 2 deaths within 30 days of treatment, and both occurred in patients with NPM1-mutant AML, indicating that even a nonintensive regimen of a hypomethylating agent and venetoclax plus revumenib can have early toxicity in older patients who mostly have preexisting comorbidities. This underscores the need for careful monitoring of patients when they are receiving this regimen.

Courtney D. Dinardo, MD:
In addition to MRD determination by flow cytometry, MRD was also determined by next-generation sequencing (NGS), a more sensitive tool. Among patients with NPM1-mutant AML, 6 out of 9 achieved MRD <5 x 10^-5 at the end of cycle 2, and 4 out of 5 achieved MRD <5 x 10^-5 at the end of cycle 3. Two patients who did not achieve MRD negativity by the end of cycle 2 subsequently experienced disease progression. Overall, it appears that the determination of MRD status by NGS is an appropriate tool to determine the durability of response achieved with the SAVE regimen.

Courtney D. Dinardo, MD:
After a relatively short median follow-up of 9 months, the median duration of response (DoR), OS, and EFS had not been reached. However, in the evaluable population of patients, the rates of DoR, OS, and EFS at 1 year were 70%, 57%, and 50%, respectively.

Amir T. Fathi, MD:
Similar to revumenib, ziftomenib is a menin inhibitor. Ziftomenib acts by disrupting the formation of the menin-KMT2A complex and blocking the oncogenic activity of mutant NPM1, resulting in terminal differentiation of leukemia blasts. By so doing, it elicits response and remission in leukemia cells.¹⁰ Recently, ziftomenib received FDA approval as monotherapy for patients with relapsed/refractory AML with a susceptible NPM1 mutation who have no satisfactory alternative treatment options.¹¹ Ziftomenib has also demonstrated activity in combination with other regimens in NPM1-mutant AML.^12-14 Therefore, KOMET-007 is investigating ziftomenib-containing triplet combination in patients with AML.¹⁵

KOMET-007 is a multicenter, open-label, phase I dose-escalation and dose-expansion study for patients 18 years of age or older with newly diagnosed or relapsed/refractory NPM1-mutant AML (NCT05735184). KOMET-007 is investigating venetoclax/azacitidine or venetoclax in combination with ziftomenib or standard induction cytarabine/daunorubicin (7+3) chemotherapy in combination with ziftomenib.

At ASH 2025, data from a cohort of 40 patients with newly diagnosed NPM1-mutant AML treated with ziftomenib in combination with venetoclax and azacitidine were presented.¹⁵ Ziftomenib at a dose of 600 mg per day was initiated on Day 8 of cycle 1 following the completion of the 7 days of azacitidine and ongoing venetoclax dosing in 28-day cycles. The coprimary endpoints were CR rate and safety, and the secondary endpoints were composite CR (CRc) rate, DoR, ORR, and MRD.

Amir T. Fathi, MD:
The median age was 75 years (range: 53-93). Out of 40 patients, 53% were females, 78% were White, 3% had an ECOG PS of 0, 40% had an ECOG PS of 1, and 58% had an ECOG PS of 2. Comutations with NPM1 mutations included FLT3 mutations (35%) and IDH1/2 mutations (23%).

Amir T. Fathi, MD:
Ziftomenib-related treatment-emergent adverse events (TEAEs) of any grade occurring in ≥25% of the patients were reported in 63%. The majority of the TEAEs were cytopenias, including thrombocytopenia (20%), neutropenia (23%), and leukopenia (13%). Grade ≥3 ziftomenib-related TEAEs occurring in ≥10% of patients included neutropenia (20%), thrombocytopenia (18%), anemia (13%), leukopenia (10%), and lymphocytopenia (3%).

Differentiation syndrome related to ziftomenib (grade 2) was reported in 1 patient; however, symptoms resolved with protocol-specified mitigation, and the patient was able to resume treatment with ziftomenib. Grade 3 QTc prolongation was reported in 1 patient, but this AE involved concomitant significant electrolyte abnormalities. The patient was able to resume ziftomenib therapy following symptom resolution. Of note, unlike revumenib, ziftomenib is not as associated with QTc prolongation.

Amir T. Fathi, MD:
The CRc rate was promising at 86%, including 73% of the patients who achieved a CR. Of note, the ORR was 89%. Among evaluable patients with comutations, the CR/CRh rates were 77% in those with FLT3 mutations and 89% in those with IDH1/2 mutations.

Assessment of MRD in 25 evaluable patients demonstrated MRD negativity by central NGS at a threshold of ≤0.1% in 68%. The MRD negativity rate was 44% when determined at a threshold of ≤0.01%. By the end of cycle 2, the rate of MRD negativity was approximately 64% to 71%. All patients achieved MRD negativity by cycle 4 of treatment

Amir T. Fathi, MD:
The median time to absolute neutrophil count (ANC) recovery of ≥0.5-1.0 x 10⁹/L was approximately 36-37 days (range: 1-69). This is akin to what is expected with azacitidine and venetoclax. The median time to platelet recovery ≥50 x 10⁹/L was 24 days (range: 0-84), and the median time to platelet recovery to ≥100 x 10⁹/L was 30 days (range: 20-77). After treatment, 5 patients proceeded to undergo HSCT, after which 3 received ziftomenib maintenance. At the time of data analysis, 27 out of 40 patients (68%) who received ziftomenib plus venetoclax and azacitidine were still alive and on study.

Courtney D. Dinardo, MD:
Survival outcomes with azacitidine and venetoclax in patients with FLT3-mutated AML are dismal.¹⁶ Therefore, a single-center, proof-of-principle study evaluated triplet therapy with azacitidine and venetoclax in combination with the FLT3 inhibitor, gilteritinib. This trial demonstrated the efficacy of the triplet combination with a CR rate of 90% and a 1-year OS rate of 83% in patients with newly diagnosed FLT3-mutant AML who are ineligible for intensive induction chemotherapy.¹⁷ It is important to further determine if outcomes can be significantly improved with the addition of gilteritinib to azacitidine plus venetoclax in a larger cohort of patients with FLT3-mutant AML. The efficacy and safety of azacitidine plus venetoclax and gilteritinib in the VICEROY trial for patients with FLT3-mutant AML were presented at ASH 2025.¹⁸

VICEROY is a randomized, dose-ranging/dose-expansion phase I/II study investigating the combination of gilteritinib with azacitidine and venetoclax dosed at either 200 mg or 400 mg daily.¹⁸ It is very important to point out that this is a multicenter trial involving more than 9 different institutions.

Patients aged 18 years of age or older with newly diagnosed FLT3-mutant AML who are not candidates for intensive chemotherapy were eligible to participate in this trial. Also, patients 75 years of age or older and younger patients with prohibitive comorbidities were eligible to participate. In total, 44 patients were randomly assigned to receive the standard dose of azacitidine plus gilteritinib dosed at 80 mg daily in combination with 200 mg or 400 mg of venetoclax. Of note, the use of gilteritinib at 80 mg is lower than the FDA-approved daily dose of gilteritinib (120 mg) when used as a single agent.¹⁹ Being a dose-finding study, 2 different doses of venetoclax were investigated in VICEROY. Bone marrow tests were performed on Day 14 of cycle 1. If <5% blasts or hypocellular bone marrow is detected, the administration of venetoclax and gilteritinib is interrupted after Day 14.

The primary endpoints were dose-limiting toxicities (DLTs), safety, CR rate, and pharmacokinetics. The secondary endpoints were CRc rate, OS, and MRD.

Courtney D. Dinardo, MD:
The baseline characteristics were fairly well balanced between the 2 treatment arms. The median age was 73 years (range: 60-84) in the 200 mg arm, and 77 years (range: 59-88) in the 400 mg arm. This confirms that the patient population was older and chemotherapy ineligible.

In both arms, approximately 60% of patients had an ECOG PS of 0-1, and approximately 40% had an ECOG PS of 2. The majority of patients had AML harboring FLT3 ITD mutations, detected in 65% of patients on the 200 mg arm and in 83% of those on the 400 mg arm. Co-occurrence of FLT3 mutations and NPM1 mutations was identified in 50% of patients on the 200 mg arm vs 33% on the 400 mg arm. On either arm, mutations in DNMT3A, IDH2 and WT1 were also detected. Four patients had AML harboring CEBPA mutations, all of whom were on the 400 mg.

Courtney D. Dinardo, MD:
The reported safety events were febrile neutropenia and increased aspartate transaminase/alanine transaminase. There were no DLTs reported on the 200 mg arm. However, 2 cases of DLTs occurred with 400 mg of venetoclax, including prolonged grade 4 neutropenia (5%), prolonged grade 4 thrombocytopenia (5%), and tumor lysis syndrome (5%). All DLTs and safety events resolved, and none led to treatment discontinuation.

Courtney D. Dinardo, MD:
Common all-cause TEAEs on both arms included febrile neutropenia, pneumonia, sepsis, hypokalemia, and pyrexia. Drug-related TEAEs leading to treatment discontinuation on the 200 mg arm were pericardial effusion (5%) and colitis (5%). On the 400 mg arm, neutropenic sepsis led to treatment discontinuation in 1 patient. There were no deaths within 30 days of treatment on either arm. However, within 60 days of treatment, the mortality rate on the 400 mg arm was 4%.

Courtney D. Dinardo, MD:
It is important to note that the median duration of azacitidine dosing was 7 days in cycles 1-4, especially on the venetoclax 200 mg arm. On both treatment arms, the median number of gilteritinib dosing days was 17 in cycle 1, and this returned to a continuous 28 in cycles 2-4. This highlights the fact that most patients were clearing bone marrow blasts to <5% on Day 14 of cycle 1. On average, the median number of venetoclax dosing days in all patients in cycle 2 was 14, and this reduced in cycles 3 and 4. The median time to ANC recovery >0.5 x 10⁹/L and platelet recovery >50 x 10⁹/L ranged from 34-50 days in cycles 1 to 4 on both arms.

Courtney D. Dinardo, MD:
The CR rate was 70% in both arms, whereas the CRc rate was 90% in the 200 mg of venetoclax arm vs 91% in the 400 mg of venetoclax arm. The median time to CR was 1 month on both arms, and the median time to CRc was 0.6 months in the 200 mg arm vs 1 month in the 400 mg arm. These results are impressive considering that the patient population being studied had FLT3-mutant AML, with the majority harboring FLT3 ITD mutations.

After a median follow-up of 19 months, the median OS was 23 months on the 200 mg arm. After a median follow-up of 16 months, the median OS was 22 months on the 400 mg arm. These results are promising. Of note, gilteritinib dose was not affected by the dose of venetoclax studied in each arm. There were 9 deaths in the 200 mg arm vs 7 deaths in the 400 mg arm.