麻豆传媒

The addition of ivosidenib (Tibsovo) to azacitidine led to a threefold survival improvement for newly diagnosed IDH1-mutant acute myeloid leukemia (AML) patients who were ineligible for intensive induction therapy, a phase III study showed.

In the so-called AGILE trial, median overall survival (OS) improved from 7.9 months with azacitidine plus placebo to 24 months with azacitidine plus the IDH1 inhibitor (HR 0.44, 95% CI 0.27-0.73, one-sided P=0.0005), reported Hartmut Döhner, MD, of Ulm University Hospital in Germany.

And the study met its primary endpoint, with a significant improvement in event-free survival (EFS) favoring the ivosidenib arm in the intent-to-treat population (HR 0.33, 95% CI 0.16-0.69), he said during a press briefing at the American Society of Hematology annual meeting.

"Overall survival tripled" with ivosidenib and "the tolerance is excellent," said co-investigator Stephane de Botton, MD, PhD, of Institut Gustave Roussy in Villejuif, France, during a Q&A session at the briefing.

"I'm really excited by the results," said press briefing moderator Mikkael Sekeres, MD, of the Sylvester Comprehensive Cancer Center at the University of Miami. He noted that the question now becomes whether to choose the AGILE regimen for these patients, or azacitidine plus venetoclax (Venclexta), the current standard of care for patients who cannot receive intensive chemotherapy in the inpatient setting regardless of their IDH1 mutational status.

"Azacitidine and venetoclax is not truly non-intensive therapy, and it's probably on a spectrum between non-intensive and intensive therapy, and probably closer to 7+3 than a lot of people recognize," Sekeres explained. "For a particularly older population -- particularly those with comorbidities who might not be able to tolerate the combination of azacitidine and venetoclax -- the combination of azacitidine and ivosidenib represents a very real option." He noted, however, that ivosidenib is not yet approved in the upfront setting.

De Botton agreed that venetoclax plus azacitidine is not always well tolerated, calling it "a semi-intensive treatment" and said ivosidenib-azacitidine should be the "preferred treatment" for these patients harboring IDH1 mutations.

The phase III AGILE study randomized 146 patients with newly diagnosed IDH1-mutant AML who were ineligible for standard intensive induction chemotherapy to azacitidine (75 mg/m²) plus either oral ivosidenib (500 mg daily; n=72) or placebo (n=74). The median patient age was 76 years and slightly more than half were men.

Döhner highlighted that the targeted combination led to significant improvements in clinical response. The overall response rate (ORR) was three times higher with ivosidenib versus placebo, a difference that extended to rates of complete response and complete response with partial hematologic recovery (CR/CRh) as well (P<0.0001 for all):

• ORR: 62.5% vs 18.9%, respectively
• CR: 47.2% vs 14.9%
• CR/CRh: 52.8% vs 17.6%

Median duration of response was 22.1 months in the azacitidine-ivosidenib arm versus 9.2 months in the azacitidine-placebo arm.

The EFS/OS benefit was also consistent across subgroups, said Döhner, including age, race, sex, de novo status, region, World Health Organization AML classification, as well as baseline Eastern Cooperative Oncology Group (ECOG) performance status, cytogenetic risk status, percentage of bone marrow blasts, and white blood cell count.

For patients who achieved a complete response at 24 weeks, the median EFS was not reached in the ivosidenib arm versus 17.8 months in the placebo arm.

Global health status/quality of life and functional subscales on the EORTC QLQ-C30 (performed at cycle 5 day 1) all favored the ivosidenib-treated patients, though no differences were significant. For symptom subscales on EORTC QLQ-C30, ivosidenib-treated patients had significantly less appetite loss and diarrhea.

Grade ≥3 treatment-emergent adverse events (TEAEs) occurred in 70.4% of patients on the ivosidenib arm compared to 64.4% of those on the placebo arm. The most common non-hematologic grade ≥3 TEAEs with ivosidenib-azacitidine were pneumonia (22.5%) and infections (21.1%), both lower than in the placebo arm (28.8% and 30.1%, respectively).

"The main thing we found was less infections across all grades," said de Botton (28.2% vs 49.3% in the placebo arm).

"Generally speaking, these results are in line with what we found previously in the phase I trial," he added. "You do not add any toxicity by adding [ivosidenib to azacitidine], so you may slightly increase the production of neutrophils ... leading to less infections."

TEAEs of special interest included grade ≥2 differentiation syndrome (14.1% in ivosidenib arm vs 8.2% in placebo arm) and grade ≥3 QT prolongation (9.9% vs 4.1%). There were no treatment-related deaths in either arm.

At enrollment, two-thirds of patients in the study had an ECOG performance status of 0-1, while the remaining had an ECOG score of 2. In the ivosidenib arm, 67% had intermediate-risk and 22% had poor-risk cytogenetics. In the placebo arm, these rates were 60% and 27%, respectively. Median bone barrow blast percentage was 54% for the ivosidenib arm and 48% for the placebo arm.

Comment