Azacitidine plus enasidenib improved complete and overall responses in newly diagnosed acute myelogenous leukemia with isocitrate dehydrogenase 2 gene mutations, compared with azacitidine alone, but it did not improve overall survival in an open-label, phase 2 trial reported at the virtual annual congress of the European Hematology Association.
“Given the very high cost of” enasidenib, and the lack of survival benefit, Gunnar Juliusson, MD, PhD, of Lund University, Sweden, who moderated the study presentation, wondered if it might make more sense to hold enasidenib in reserve until after progression on azacitidine.
“The challenge is going to be exactly” that, “trying to figure out [if] you use both things together” or in sequence. “You can look at it in both ways,” said lead investigator Courtney DiNardo, MD, associate professor in the department of leukemia at the University of Texas MD Anderson Cancer Center, Houston.
“We do know” that with enasidenib monotherapy, there’s “a decrement in the rates of remission and in the duration of response” and overall survival in the salvage setting, so there’s “a clear rationale to give it earlier rather than later,” but “I think this study in some ways provides a few more questions than it really answers,” she said at the meeting.
About 15% of AML patients have leukemogenic isocitrate dehydrogenase 2 (IDH2) mutations; enasidenib, an oral small molecule, inhibits the mutant enzyme. The older AML patients are, the more likely they are to have an IDH2 mutation, so the work “is relevant to our older chemotherapy ineligible population,” Dr. DiNardo said.
The trial was prompted by preclinical indications of synergy with azacitidine; alone, each agent has an overall response rate of about 30% in newly diagnosed AML, and a complete remission (CR) rate of about 20%, she explained.
Her team randomized 68 adults with newly diagnosed AML and an IDH2 mutation to enasidenib 100 mg daily on a 28-day cycle with subcutaneous azacitidine 75 mg/m2 for 7 days during the cycle, and 33 others to just the azacitidine alone.
Their subjects were ineligible for intensive chemotherapy and had intermediate to poor risk cytogenetics. The median age was 75 years, and Eastern Cooperative Oncology Group performance scores were 2 or less.
The overall response rate was 71% with the combination and 42% in the azacitidine alone arm (P = .0064). Fifty-three percent of combination patients, but 12% of azacitidine alone subjects, had complete remissions (P = .0001). The median duration of response with combination therapy was 24.1 months, versus 12.1 months.
Enasidenib plus azacitidine subjects also had greater drops in mutant IDH2 variant allele frequency (median 83.4% versus 17.7%, P < .01) and levels of the downstream oncometabolite 2-hydroxyglutarate (97.8% versus 54.3%; P < .01).
However, median OS was 22 months in both arms (HR 0.99, 95% CI 0.52, 1.87, P = .97). Although median event-free survival favored the combination – 17.2 months versus 10.8 – the results were not statistically significant (HR 0.59, 95% CI 0.30, 1.17, P = .13).
A possible reason for the lack of survival benefit, Dr. DiNardo said, was that seven azacitidine-alone patients (21%) went on to enasidenib after leaving the study, most commonly for disease progression, which occurred in 31% of combination patients versus 52% in the azacitidine-alone arm.
Combination subjects had a median of 10 treatment cycles, vs. 7 in the azacitidine-alone group. Grade 3-4 adverse events included thrombocytopenia (37% combination, 19% azacitidine-alone), neutropenia (35% vs. 22%), anemia (19% vs. 22%), and febrile neutropenia (15% vs. 16%). Grade 3-4 infections were more common with azacitidine monotherapy (31% vs. 18%).
Twelve enasidenib/azacitidine subjects (18%) developed isocitrate dehydrogenase differentiation syndrome, a complication that carries a black box warning in enasidenib’s label.
The work was funded by enasidenib marketer Celgene. Dr. DiNardo is an adviser to, and receives research funding from, the company. Dr. Juliusson’s disclosures, if any, were not reported.
SOURCE: DiNardo CD et al. EHA Congress, abstract S139.