This content is rewritten in English to be completely unique in wording and sentence structure while preserving the original meaning and key information. It remains thorough, expands where helpful, and keeps a friendly, professional tone with clear explanations for beginners. It also opens with a provocative hook to engage readers and invites discussion about contested points at the end.
Bold opening: A targeted, all-oral combination can drive impressive responses in newly diagnosed AML, but safety concerns—especially myelosuppression and infection risk—remain a central debate about its real-world value.
A recent phase 1/2 study analyzed the all-oral regimen SAVE, which combines revumenib (Revuforj) with decitabine/cedazuridine and venetoclax for patients with newly diagnosed AML. The trial (NCT05360160) reported high response rates and a high rate of minimal residual disease (MRD) negativity, but it also highlighted notable safety challenges.
Key findings from the 2025 ASH presentation include: in the cohort of 21 patients with newly diagnosed AML, the overall response rate (ORR) reached 86%. This comprised a complete remission (CR) or CR with partial hematologic recovery (CRh) rate of 81%, and a CR with incomplete platelet recovery (CRp) rate of 5%. The CR/CRh subset accounted for 76% CR and 5% CRh. Importantly, all patients who achieved a composite CR also achieved MRD negativity at a sensitivity of 10^-4 by multiparameter flow cytometry. There were two early deaths, and one patient withdrew from response assessment.
At a median follow-up of about 9 months, the duration of response (DOR) for those achieving CR or CRh had not yet reached a median value. The 12-month DOR estimate was 70%. Median overall survival (OS) was not reached, with a 12-month OS of 57%. The median event-free survival (EFS) was also not reached, and the 12-month EFS was 50%.
Safety-wise, the most common any-grade adverse events included vomiting (67%), elevated liver enzymes (AST/ALT, 62%), nausea (57%), and electrolyte disturbances. Grade 3 or higher events were dominated by febrile neutropenia (48%), thrombocytopenia (33%), neutropenia (24%), bacteremia (19%), and infections at specific sites (lung 14%; skin 14%). Three treatment-related deaths (grade 5) occurred due to bacteremia (n=2) and bronchopulmonary hemorrhage (n=1). Differentiation syndrome of any grade occurred in 19% of patients, with two cases reaching grade 3 or higher; all such cases resolved with steroids.
Commenting on these results, Wei-Ying Jen noted that the combination shows strong activity in AML patients with NPM1 mutations or KMT2A rearrangements, but the small cohort and short follow-up temper the conclusions. She emphasized the importance of early empiric high-dose steroids when differentiation syndrome is suspected—even in combination therapy—and highlighted myelosuppression and infectious complications as key safety concerns that call for optimization of menin-inhibitor combinations to improve the overall risk-benefit balance, particularly in NPM1-mutated AML.
Jen is an assistant professor at The University of Texas MD Anderson Cancer Center, in Houston, Texas.
How are menin inhibitors like revumenib being evaluated in leukemias?
SAVE Regimen: Revumenib Plus Decitabine/Cedazuridine and Venetoclax in AML
- The all-oral SAVE regimen demonstrated high response and MRD negativity rates in newly diagnosed, menin-susceptible AML.
- Early follow-up suggested durable remissions, with median OS, EFS, and DOR not yet reached.
- Safety concerns centered on myelosuppression and infections, underscoring the need to refine menin-inhibitor combinations for better safety.
Jen explained that menin acts as a scaffold protein regulating gene expression. The interaction between menin and KMT2A is a critical dependence in leukemias driven by KMT2A rearrangements or NPM1 mutations, contributing to abnormal gene expression and leukemogenesis. Revumenib is a potent and selective inhibitor of this interaction.
Regulatory updates followed: in November 2024, the FDA approved revumenib for relapsed or refractory AML with a KMT2A translocation in patients aged 1 year and older. A subsequent approval in October 2025 extended use to adults and pediatric patients with relapsed/refractory AML and a susceptible NPM1 mutation when no satisfactory alternatives exist.
Rationale for the SAVE combination
Single-agent revumenib has produced ORRs ranging from 43% to 60% in relapsed/refractory AML with KMT2A rearrangement or NPM1 mutation, and CR/CRh rates from 21% to 30%. However, responses tended to be short-lived, prompting exploration of rational combinations to improve depth and duration of response.
Although hypomethylating agent and venetoclax combinations are standard for older or less fit AML patients, relapse remains common. Preclinical data suggested that adding BCL-2 inhibition (venetoclax) to menin inhibition could improve outcomes.
Trial design and patient population
The phase 1/2 trial enrolled patients with relapsed/refractory AML or mixed-phenotype acute leukemia aged 12 or older, with ECOG 0–2 and adequate organ function. A frontline phase 2 cohort included patients unsuitable for intensive chemotherapy whose tumors carried KMT2A rearrangement, NPM1 mutation, or NUP98 rearrangement.
Phase 1 used a 3+3 dose-escalation strategy to identify the recommended phase 2 dose (RP2D) of revumenib. The regimen involved revumenib plus decitabine/cedazuridine and venetoclax with a planned schedule: revumenib on days 1–28, decitabine/cedazuridine on days 1–5, and venetoclax with ramp-up through day 14.
In newly diagnosed patients (n=21, median age 70), the cohort skewed older (52% were 70+), with most being female. Comutations included NRAS/KRAS, FLT3, IDH1/2, and various MDS-associated mutations, revealing a heterogeneous group.
Efficacy by genotype
- NPM1-mutated cohort (n=14): ORR 86% with CR/CRh at 79% (CR 71%, CRh 7%), CRp 7%, and MRD negativity in 86% at 10^-4 by flow cytometry. Early deaths occurred in 2 patients, with DOR not yet reached and 12-month DOR at 71%. 12-month OS was 53%.
- KMT2A-rearranged cohort (n=7): ORR 86% with CR/CRh 86% (CR 86%), all evaluable for MRD negativity at 10^-4 (86%). DOR median not reached, 12-month DOR 80%, and 12-month OS 69%.
MRD trends in NPM1 patients
Most patients achieved MRD negativity by the end of cycle 2, with clearance improving over time. Two patients did not clear NPM1 MRD and relapsed. Among those achieving CR/CRh, DOR remained unreached, with approximately 70% estimated to be in remission at 1 year. Median OS and EFS were not reached at a 9-month follow-up.
Resistance mechanisms observed
A ddPCR assay targeted eight MEN1 mutations associated with resistance to revumenib. One relapse case revealed a MEN1 M327B mutation at a 17% variant allele frequency, consistent with known resistance patterns to menin inhibitors. This suggests that resistance may involve more than a single mechanism and that additional resistance pathways require investigation.
Disclosure: No financial relationships were disclosed by Jen.
References and context
1. Jen WY, DiNardo CD, Short NJ, et al. Phase II study of the all-oral combination of revumenib with decitabine/cedazuridine and venetoclax (SAVE) in newly diagnosed AML. 2025 ASH Annual Meeting. Abstract #47.
2. FDA approval updates for revumenib in KMT2A-translocated AML (Nov 2024).
3. FDA approval updates for revumenib in relapsed/refractory NPM1-mutant AML (Oct 2025).
Newsletter note
Stay informed on the latest and most practice-changing oncology data by subscribing to our newsletter for updates and expert analyses.
What this means for practitioners and patients
- The SAVE regimen shows promising activity in specific genetic subgroups of AML, particularly NPM1 mutations and KMT2A rearrangements, with high MRD negativity rates suggesting deep remissions.
- Safety concerns around myelosuppression and infectious complications call for careful patient selection, proactive infection prevention, and potential regimen optimization to improve the safety profile.
- The evolving role of menin inhibitors in AML raises important questions: can combination approaches sustain responses longer and with acceptable toxicity, and how should resistance mechanisms be monitored and addressed in ongoing trials?
Controversial take and closing thought
Does the high early efficacy justify broader use of an all-oral, multi-agent approach given the clear safety signals, or should these regimens remain experimental until larger, longer-term data confirm durable benefit with manageable risk? Share your view below: should frontline AML treatment tilt toward aggressive, targeted combinations or prioritize safer, proven regimens until more evidence accumulates?
