2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2025 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
The SAVE regimen shows high response and MRD negativity rates in patients with AML, but myelosuppression and infectious complications remain key concerns.
The all-oral combination comprised of revumenib (Revuforj) plus decitabine/cedazuridine and venetoclax (Venclexta; SAVE regimen) elicited responses in patients with newly diagnosed acute myeloid leukemia (AML) although a concern for myelosuppression was reported, according to data from the phase 1/2 SAVE trial (NCT05360160).1
The findings which were presented at the 2025 ASH Annual Meeting & Exposition showed that the regimen in patients with newly diagnosed with AML (n = 21) induced an objective response rate (ORR) of 86%, which comprised a complete remission (CR)/CR with partial hematologic recovery (CR/CRh) rate of 81% and a CR with incomplete platelet count recovery (CRp) rate of 5%; the CR rate was 76% and the CRh rate was 5%. All patients who attained composite CR achieved minimal residual disease (MRD) negativity at a sensitivity of 10-4 by multiparameter flow cytometry. Two early deaths occurred, and 1 patient withdrew consent and was not response evaluable.
At a median follow-up of 9 months, the median duration of response (DOR) was not reached (NR) in those attaining CR or CRh, and the 12-month DOR rate was 70% (95% CI, 46%-100%). Moreover, the median overall survival (OS) was NR, with a 12-month OS rate of 57% (95% CI, 36%-90%); the median event-free survival (EFS) was also NR, with a 12-month EFS rate of 50% (95% CI, 29%-85%).
In terms of safety, the most common any-grade adverse effects (AEs) were vomiting (67%), elevated aspartate and alanine aminotransferase levels (62%), nausea (57%), and electrolyte disturbances. The most frequent grade 3 or higher AEs were febrile neutropenia (48%), thrombocytopenia (33%), neutropenia (24%), bacteremia (19%), and site-specific infections (lung, 14%; skin, 14%). Three grade 5 events were reported in the form of bacteremia (n = 2) and bronchopulmonary hemorrhage (n = 1). Any-grade differentiation syndrome (DS) occurred in 19% of patients, with 2 cases grade 3 or higher; all cases resolved with steroids.
“The all-oral combination of revumenib, decitabine/cedazuridine and venetoclax shows good activity in newly diagnosed AML with NPM1 mutations or KMT2A rearrangements. Response and MRD-negative rates were high, although the cohort size is small and the follow-up period is short,” Wei-Ying Jen, MA, BMBch, M Med, FRCPath said. “We recommend empiric, high-dose steroids at the earliest suspicion of differentiation syndrome, even within the context of combination therapy. There is concern for myelosuppression, as evidenced by the rate of infectious complications; this highlights the need to optimize menin inhibitor combinations to improve the overall risk-benefit profile—especially in NPM1-mutated AML.”
Jen is an assistant professor in the Department of Leukemia of the Division of Cancer Medicine at The University of Texas MD Anderson Cancer Center, in Houston, Texas.
“Menin is a scaffold protein which regulates gene expression. The interaction of menin and KMT2A is a critical dependency in KMT2A-rearranged or NPM1-mutated acute leukemias; this leads to aberrant gene expression and leukemogenesis,” Jen explained. “Revumenib is a potent and selective small molecule inhibitor of this interaction.”
In November 2024, the FDA cleared revumenib for the treatment of adult and pediatric patients aged 1 year and older with relapsed or refractory AML and a KMT2A translocation.2 The following year, in October 2025, the regulatory agency has approved the drug for use in adult and pediatric patients at least 1 year of age with relapsed or refractory AML and a susceptible NPM1 mutation who do not have satisfactory alternative treatment options.3
Jen reported that single-agent revumenib has elicited ORRs ranging from 43% to 60% in those with relapsed/refractory AML and KMT2A rearrangement or NPM1 mutation, with CR/CRh rates ranging from 21% to 30%, and DORs ranging from approximately 4 months to 6 months.1 “However, responses can be short lived; therefore, there is a need to improve response rates and decrease the risk of relapse with rational combination strategies,” Jen said.
Although hypomethylating agents in combination with venetoclax is standard treatment for patients who are older or unfit with AML and has improved outcomes, relapse is inevitable for most patients, Jen noted. Preclinical data have indicated that BCL-2 and menin inhibition improved survival outcomes.
The phase 1/2 study enrolled patients with relapsed or refractory AML or mixed phenotype acute leukemia who were at least 12 years old. Patients were required to have an ECOG performance status ranging from 0 to 2 and have acceptable organ function. Additionally, the frontline phase 2 cohort included patients who were not eligible for intensive chemotherapy whose tumors harbored KMT2A rearrangement, NPM1 mutation, or NUP98 rearrangement.
The phase 1 portion leveraged a 3+3 dose-escalation design. Revumenib was given at 110 mg plus CYP3A4 inhibition or 160 mg without CYP3A4 inhibition (DL 0) or 160 mg plus CYP3A4 inhibition or 270 mg without CYP3A4 inhibition (DL 1). “The recommended phase 2 dose [RP2D] of revumenib was identified as 160 mg twice daily with strong CYP3A4 inhibitors, or 270 mg twice daily with strong CYP3A4 inhibitors,” Jen noted.
Revumenib was administered on days 1 to 28. Decitabine/cedazuridine was given on days 1 to 5, and venetoclax was given at a target dose of 400 mg with ramp up on days 1 to 14.
The primary objective for phase 1 in patients with relapsed/refractory disease was to evaluate safety and identify the maximum tolerated dose and RP2D; for phase 2, it was to evaluate the efficacy of the regimen in those with frontline and relapsed/refractory disease. Secondary objectives for the phase 2 portion looked at OS, relapse-free survival, complete response duration, and MRD. Exploratory objectives focused on MRD at 10-5 and resistance.
During the first cycle, day 14 bone marrow was performed. “If bone marrow blasts were less than 5%, revumenib was held after day 21,” Jen noted. “Revumenib monotherapy was resumed as maintenance post-transplant for a planned duration of one year.”
At the meeting, Jen shared the outcomes of the newly diagnosed cohort of the study. In all patients, the median age was 70 years (range, 60-83) with 52% aged 70 years or older. Most patients were female (71%) and almost one-fourth had secondary AML (24%). In terms of comutations, 19% of patients had NRAS/KRAS, 19% had FLT3 (ITD, 5%; TKD, 14%), IDH1/2 (19%), or MDS-associated mutations in genes like ASCL1, BCOR, EZH2, SF3B1, STAG2, U2AF1, ZRSR2, and RUNX1 (43%).
In those with NPM1 mutations (n = 14), the ORR with the SAVE regimen was 86%, which comprised a CR/CRh rate of 79%; the CR rate was 71% and the CRh rate was 7%. The CRp rate was 7% in this group. Most patients (86%) were MRD negative 10-4 by multicolor flow cytometry. Two patients (10%) experienced early death. The median DOR was NR, and the 12-month DOR rate was 71% (95% CI, 43%-100%). The median OS was also NR, with a 12-month OS rate of 53% (95% CI, 29%-97%).
In patients with KMT2A rearrangements (n = 7) the SAVE regimen induced an ORR of 86%, which comprised a CR/CRh rate of 86%; the CR rate was 86%. Fourteen percent of patients were not evaluable. Again, 86% of patients achieved MRD negativity 10-4. The median DOR was NR, with a 12-month DOR rate of 80% (95% CI, 52%-100%). The median OS was also NR, with a 12-month OS rate of 69% (95% CI, 40%-100%).
Jen reported that most patients achieved MRD negativity at the end of cycle 2, with clearance rates improving over time. She added that two patients did not have NPM1 MRD clearance and went on to relapse. In those who achieved a CR/CRh, the median DOR was NR. “An estimated 70% of patients are still in remission at 1 year,” Jen said. “At a median follow-up of 9 months, the median OS and EFS have not been reached.”
Jen noted that a ddPCR assay was designed to identify eight MEN1 point mutations, which confer resistance to revumenib. “Synthetic gBlocks, representing each mutation and wild-type controls, were generated for assay development and optimization. Of note, there was false-positive detection of some variants at a low mutation burden, likely corresponding to binding of the wild-type MEN1 gene.”
She added that at the time of relapse, MEN1 mutation testing was done for two of the three cases of relapse. One MEN1 M327B mutation was detected with a variant allele frequency of 17%. “This mutation has been described in vivo and in vitro with exposure to revumenib as well as other menin inhibitors,” she noted, adding that the emergent MEN1 mutation suggests that the combination regimen may not fully address this mechanism of relapse and that there are likely other resistance mechanisms inherent to the combination that need to be further examined.
Disclosures: Jen did not disclose any financial relationships.
Related Content:
