Azacitidine/Ivosidenib/Venetoclax Generates High ORRs in IDH1-Mutant Newly Diagnosed AML

The combination of azacitidine, venetoclax (Venclexta), and ivosidenib (Tibsovo) induced minimal residual disease (MRD) negativity and durable responses in patients with IDH1-mutated, newly diagnosed acute myeloid leukemia (AML), according to findings from a phase 1b/2 trial (NCT03471260) presented at the 2026 EHA Congress.¹

Among evaluable patients (n = 40), the overall response rate (ORR) with the combination was 95%, with a composite complete response (CRc) rate of 93%. Best responses included CR (60%), CR with partial hematologic recovery (CRh; 18%), CR with incomplete count recovery (CRi; 15%), morphologic leukemia-free state (MLFS; 2.5%), and no response (5%). The median number of cycles to best response was 2.

An MRD analysis by flow cytometry with a limit of detection of less than 0.1% showed that 91% of evaluable patients (n = 34) were MRD negative. The median time to MRD negativity was 2 cycles.

What are the current standard treatment regimens for patients with IDH1-mutated AML?

In October 2020, the FDA granted regular approval to venetoclax plus azacitidine, decitabine, or low-dose cytarabine for the treatment of patients with newly diagnosed AML who are at least 75 years of age or who have comorbidities that preclude the use of intensive induction chemotherapy, based on findings from the phase 3 VIALE-A (NCT02993523) and VIALE-C (NCT03069352) trials.² Then, in May 2022, the FDA approved ivosidenib plus azacitidine for this indication based on data from the phase 3 AGILE trial (NCT03173248).³

“Therefore, for our patients with IDH1-mutated AML, there are currently 2 standard-of-care regimens available to them, both using hypomethylating agents,” Jennifer Marvin-Peek, MD, said in a presentation of the data.¹ “Although the outcomes with these 2 doublet regimens are quite good, the responses were still not universal and were often not durable, with eventual relapse over time. Therefore, we designed a multicenter investigator-initiated phase 1/2 study looking at a triplet regimen to see if combining these 3 agents could improve response rates and clinical outcomes.”

Marvin-Peek is a medical oncology fellow at the University of Texas MD Anderson Cancer Center in Houston.

“The median age of patients at diagnosis with IDH1 mutations is approximately 70 years old, so when thinking about frontline therapies for these patients, that’s something we have to take into consideration, as well as whether these patients are fit enough for intensive chemotherapy,” Marvin-Peek contextualized in an interview with OncLive^®.

What was the design of the phase 1b/2 trial of an ivosidenib-based triplet in IDH1-mutated AML?

This phase 1b/2 trial enrolled patients at least 18 years of age with IDH1 R132–mutant, relapsed/refractory AML, newly diagnosed AML, or high-risk myelodysplastic syndromes and/or myeloproliferative neoplasms. Patients needed to have an ECOG performance status of 0 to 2 with adequate organ function. Patients could not have received prior venetoclax or ivosidenib.

In the phase 1b portion, patients received ivosidenib plus venetoclax at 1 of 2 dose levels, or ivosidenib plus venetoclax and azacitidine at 1 of 2 dose levels. Dose level 3, comprising oral ivosidenib at 500 mg continuously starting at cycle 1 day 15 plus oral venetoclax at 400 mg on days 1 through 14 plus intravenous or subcutaneous azacitidine at 75 mg/m² on days 1 through 7, was selected as the recommended phase 2 dose (RP2D). The phase 2 portion enrolled 40 patients with newly diagnosed AML to receive the RP2D.

Safety and tolerability, as well as ORR within 5 cycles, served as the primary end points. Secondary end points included overall survival (OS), duration of response (DOR), and MRD-negativity rate by flow cytometry.

Patients had a median age of 72 years (range, 51-80). In total, 53% of patients had de novo AML, and 38% of patients had secondary AML. The most common co-mutations reported were in DNMT3A, RUNX1, SRFF2, and NPM1. Additionally, 4 patients had RAS co-mutations, and 2 patients had TP53 co-mutations.

What additional efficacy findings were seen with ivosidenib plus venetoclax and azacitidine in the phase 1b/2 trial in AML?

At a median follow-up of 35 months (95% CI, 19.3-40.5), the median OS was not reached (NR; 95% CI, 52.8-NR), and the 3-year OS rate was 79% (95% CI, 64%-96%). The median DOR was NR (95% CI, 43.4-NR), and the 3-year DOR rate was 83% (95% CI, 64%-100%). Additionally, the 3-year cumulative incidence of relapse (CIR) rate was 9% (95% CI, 0%-20%).

As of the data cutoff, 30% of patients remained on the study. The most common reason for treatment discontinuation was for hematopoietic stem cell transplant (HSCT; 45%), although 5% of patients discontinued therapy due to adverse effects (AEs). In total, 3 patients experienced relapse, all of whom have IDH1-negative clones.

Marvin-Peek noted that patients without co-occurring signaling pathway mutations (n = 25; including mutations in KRAS, NRAS, PTPN11, NF1, BRAF, FLT3-ITD, FLT3-TKD, KIT, JAK2, and CSF3R) had superior responses vs those with these mutations (n = 15), with CRc rates of 100% vs 80%, respectively. In the patients without these signaling mutations, the rates of CR, CRh, CRi, MLFS, and no response were 72%, 8%, 20%, 0%, and 0%, respectively. In the patients with these mutations, these respective rates were 40%, 27%, 13%, 7%, and 13%. The median OS and DOR were NR in both these groups. The 3-year OS rates in these respective groups were 95% (95% CI, 86%-100%) vs 54% (95% CI, 30%-95%; P = .037). The 3-year DOR rates were 90% (95% CI, 78%-100%) vs 66% (95% CI, 40%-100%), respectively (P = .29).

Among patients with non-treated secondary AML (n = 33), the CRc rate was 97%, and the rates of CR, CRh, CRi, MLFS, and NR were 64%, 15%, 18%, 0%, and 3%, respectively. Among those with treated secondary AML (n = 7), the CRc rate was 71%, and the respective rates of CR, CRh, CRi, MLFS, and NR were 43%, 14%, 14%, 14%, and 14%. The median OS and DOR were both NR in both these groups. The 3-year OS rates in these respective groups were 83% (95% CI, 68%-100%) vs 60% (95% CI, 29%-100%; P = .50). The 3-year DOR rates were 82% (95% CI, 68%-100%) vs 66% (95% CI, 43%-100%), respectively (P = 1.00).

Among patients who did not receive subsequent allogeneic HSCT (allo-HSCT), the median age was 76 years (range, 51-80), and the median follow-up was 35 months. In these patients, the median OS and DOR were both NR, and the 3-year rates of OS, DOR, and CIR were 90% (95% CI, 73%-100%), 83% (95% CI, 64%-100%), and 17% (95% CI, 0%-26%), respectively. Among those who did receive subsequent allo-HSCT, the median age was 69 years (range, 62-76), and the median follow-up was 30 months. In these patients, the median OS and DOR were both NR, and the 3-year rates of OS, DOR, and CIR were 82% (95% CI, 66%-100%), 82% (95% CI, 65%-100%), and 0% (95% CI, 0%-100%), respectively.

Regarding treatment density, the average length of each cycle was 35 days.

“Ivosidenib was largely able to be given continuously, starting at cycle 1 day 15 for patients on concomitant azoles,” Marvin-Peek noted. “No dose modifications were made to azacitidine or ivosidenib, but close monitoring for AEs in azole levels [were] recommended. Venetoclax dose modifications were made from moderate and strong CYP3A inhibitors, as well as posaconazole.”

What was the safety profile of ivosidenib plus venetoclax and azacitidine in IDH1-mutated AML?

Grade 3/4 nonhematologic AEs were reported in 30% of patients and included infections (15%), tumor lysis syndrome (7.5%), QTc prolongation (5%), abdominal pain (2.5%), differentiation syndrome (2.5%), and myocardial infarction (2.5%).

“Overall, we feel like this regimen has been very well tolerated and is something we’re hoping to see patients on,” Marvin-Peek told OncLive.

What are the next steps for evaluating triplet regimens in IDH1-mutated AML?

Going forward, the phase 3 EVOLVE-1 trial (NCT07075016) is ongoing to investigate azacitidine plus ivosidenib and venetoclax in patients with newly diagnosed IDH1-mutated AML.

“These data will help shed light on what the optimal regimen is for [patients with] IDH1-mutated [disease] who are unfit or ineligible for intensive chemotherapy,” Marvin-Peek concluded.

Disclosures: Marvin-Peek reported consultancy/advisory board involvement with Servier, Rigel Pharmaceuticals, and Blueprint.

References

1. Marvin-Peek J, Garcia JS, Borthakur G, et al. A multicenter phase 1b/2 trial of azacitidine, venetoclax, and ivosidenib in IDH1-mutated acute myeloid leukemia (AML). Presented at: 2026 EHA Congress; June 11-14, 2026; Stockholm, Sweden. Abstract S127.
2. FDA grants regular approval to venetoclax in combination for untreated acute myeloid leukemia in adults unable to receive intensive chemotherapy. News release. ASCO. November 3, 2020. Accessed June 11, 2026. https://www.asco.org/practice-policy/policy-issues-statements/asco-in-action/fda-grants-regular-approval-venetoclax
3. Servier announces FDA approval of Tibsovo (ivosidenib tablets) in combination with azacitidine for patients with newly diagnosed IDH1-mutated acute myeloid leukemia. News release. Servier Pharmaceuticals. May 25, 2022. Accessed June 11, 2026. https://servier.us/blog/fda-approval-tibsovo-aml/