Enasidenib significantly improves complete remission and overall response when combined with azacitidine compared with azacitidine alone in patients with newly diagnosed acute myeloid leukemia with IDH2 mutations.
Courtney D. DiNardo, MD
Enasidenib (Idhifa), an oral small molecule inhibitor of mutant IDH2 proteins, significantly improves complete remission and overall response when combined with azacitidine (Vidaza) compared with azacitidine alone in patients with newly diagnosed acute myeloid leukemia (AML) with IDH2 mutations, according to data from an interim phase II trial presented at the 2019 ASH Annual Meeting.1
Results from the ongoing AG221-AML-005 trial revealed that at a median follow-up of 14 months, median event-free survival (EFS) was 17.2 months in patients randomized to receive the combination of enasidenib and azacitidine compared with 10.8 months in those who were given azacitidine alone (HR, 0.59; 95% CI, 0.30-1.17; P = .1278), said Courtney D. DiNardo, MD, at the meeting.
No significant difference in overall survival (OS) was observed between the 2 groups, but 21% of patients in the azacitidine alone arm received subsequent treatment with enasidenib monotherapy after discontinuation, noted DiNardo, an assistant professor in Department of Leukemia, Division of Cancer Medicine at the University of Texas MD Anderson Cancer Center.
IDH2 mutations occur in around 8 to 19% of patients with AML;2 they are often considered to be founder mutations with stability over the course of therapy. “It is now well recognized that IDH2 mutations can be acquired at the time of progression or of transformation myelodysplastic syndrome, or even more commonly myeloproliferative neoplasms,” she said. 3 IDH2 mutations lead to the accumulation of 2- hydroxyglutarate (HG), an oncometabolite that competitively inhibits alpha-ketoglutarate, which is a key required substrate for many dioxygenase reactions.
Enasidenib, already approved by the FDA for use in adults with IDH2-mutant relapsed/refractory AML, indirectly reduces DNA methylation by suppressing the 2-HG, thereby restoring function to alpha-ketoglutarate—dependent TET family enzymes, among other substrates.
In a large phase 1/2 study, enasidenib monotherapy led to an overall response rate (ORR) of 30.8% in patients with newly diagnosed AML and a CR rate of 18%.4 In vitro, the combination of enasidenib and azacitidine enhanced single-agent effects on releasing differentiation block and reduced leukemic stem/progenitor cell populations.5 Tolerability and efficacy data from the phase Ib dose-finding portion of the current study supported further testing of the combination, according to DiNardo.
The phase 2 portion of the trial enrolled 101 patients with newly diagnosed IDH2-mutant newly diagnosed AML who were ineligible to receive intensive chemotherapy. Participants were randomized in a 2:1 ratio to receive either enasidenib plus azacitidine or azacitidine alone in repeated 28-day cycles. All patients received subcutaneous azacitidine at 75 mg/m2/day for the first 7 days of each treatment cycle; patients who were randomized to the combination also received continuous enasidenib at 100 mg once daily.
The median age of the study population was 75 years (range, 62-85). About one-fourth of patients enrolled in the trial had secondary AML. At baseline, about one-fourth of patients had an IDH2-R140 mutation and about one-fourth had IDH2-R172 mutation. Forty-five patients in the combination arm and 18 in the azacitidine monotherapy arm had co-mutations. The most common co-occurring mutations were DNMT3A, ASXL1, and RUNX1. “[There was] a notable underrepresentation of patients with either FLT3-ITD or NPM1 compared with expectations,” said DiNardo. Eighty percent of patients had intermediate-risk cytogenetics in the combination arm.
At the time of data cutoff, August 19, 2019, 94% of patients in the azacitidine-alone arm and 69% in the combination arm discontinued therapy. The most common reason for discontinuation in both arms was disease progression: 52% in the azacitidine-alone arm and 31% in the combination arm. Discontinuation due to adverse events was rare in both arms, at 6%.
The ORR rate was 71% in the combination arm (95% CI, 58-81) and 42% in the azacitidine-alone arm (95% CI, 26-61; P = .0064); the CR rates were 53% (95% CI, 41-65) versus 12% (95% CI, 3-28; P = .0001), respectively. Time to first response was about 2 months in each arm and the time to CR was 5.5 months (range, 0.7-19.5) with enasidenib/azacitidine and 3.7 months (range, 3.0-4.1) with azacitidine alone. In patients who responded, the median duration of response was 24.1 months (range, 11.1-not reached) with combination enasidenib/azacitidine and 12.1 months (range, 2.8-14.6) with azacitidine alone.
“Responses were observed in patients with RAS pathway co-mutations, which have been associated with resistance to enasidenib monotherapy,” DiNardo said.
The median number of treatment cycles received was 10 in the combination arm and 6 in the azacitidine-only arm. The most common treatment-emergent adverse events (TEAEs) with enasidenib/azacitidine and azacitidine alone were thrombocytopenia (62% and 44%), nausea (69% and 38%), anemia (53% and 44%), and vomiting (49% and 47%). Grade ≥3 AEs were similar between arms, with the exception of IDH differentiation syndrome (IDH-DS). In the combination arm, IDH-DS occurred in 18% of patients at a median of 28.5 days. The median time to IDH-DS resolution was 11.5 days.
The 60-day mortality rates were 7% in the combination arm and 3% in the azacitidine-only arm. Most deaths were related to progression of disease. Two deaths were considered likely or possibly due to IDH differentiation syndrome.
The historical median OS with azacitidine alone is about 10 months, noted DiNardo. The median OS in either arm was 22 months. In patients treated with the combination who achieved a CR, the median OS was not reached, with an estimated 1-year OS >90%.
“We saw deep reductions in 2-HG concentrations with the combination, which is indicative of on-target activity,” DiNardo said. “The combination led to impressive reductions in the IDH2 variant allele frequency.”