Background
Phase 3, randomized, open-label trial. 454 patients with newly diagnosed higher-risk MDS (n=324, 71%), higher-risk CMML (n=27, 6%), or AML with 20–30% blasts (n=103, 23%). Randomized 1:1. Pevonedistat inhibits NEDD8-activating enzyme (NAE), disrupting ubiquitin-mediated protein degradation and inducing preferential apoptosis of malignant cells. Conducted at 90 sites globally.
Interventions and follow up
Arm A: Pevonedistat 20 mg/m² IV days 1, 3, 5 + azacitidine 75 mg/m² days 1–5 of each 28-day cycle (n=227)
Arm B: Azacitidine 75 mg/m² days 1–5 of each 28-day cycle (n=227)
Primary endpoint: Event-free survival (EFS)
mFollow up: Not separately stated
Arm B: Azacitidine 75 mg/m² days 1–5 of each 28-day cycle (n=227)
Primary endpoint: Event-free survival (EFS)
mFollow up: Not separately stated
Results
EFS (ITT): Median 17.7 vs 15.7 months, HR 0.968 (95% CI 0.757–1.238), P=.557 (not significant)
EFS (HR-MDS cohort): 19.2 vs 15.6 months, HR 0.887, P=.431 (not significant)
OS (HR-MDS cohort): 21.6 vs 17.5 months, HR 0.785, P=.092 (not significant)
OS (AML 20–30% blasts cohort): 14.5 vs 14.7 months, HR 1.107, P=.664 (not significant)
Post hoc — OS (HR-MDS, >3 cycles): 23.8 vs 20.6 months, P=.021
Post hoc — OS (HR-MDS, >6 cycles): 27.1 vs 22.5 months, P=.008
EFS (HR-MDS cohort): 19.2 vs 15.6 months, HR 0.887, P=.431 (not significant)
OS (HR-MDS cohort): 21.6 vs 17.5 months, HR 0.785, P=.092 (not significant)
OS (AML 20–30% blasts cohort): 14.5 vs 14.7 months, HR 1.107, P=.664 (not significant)
Post hoc — OS (HR-MDS, >3 cycles): 23.8 vs 20.6 months, P=.021
Post hoc — OS (HR-MDS, >6 cycles): 27.1 vs 22.5 months, P=.008
Adverse events
Hematologic: Grade ≥3 anemia 33% vs 34%, neutropenia 31% vs 33%, thrombocytopenia 30% vs 30% (similar between arms)
Other: Azacitidine dose intensity was maintained; no new safety signals with the combination
Other: Azacitidine dose intensity was maintained; no new safety signals with the combination
Conclusions
Pevonedistat + azacitidine did not significantly improve EFS or OS versus azacitidine alone in higher-risk MDS or AML with low blast count. Post-hoc analyses suggested a potential OS benefit in patients receiving >3 cycles, but these exploratory findings require prospective validation.
Key Limitations
The primary EFS endpoint was not met in any prespecified population. The heterogeneous study population (MDS, CMML, and low-blast AML combined) may have diluted any disease-specific signal. Post-hoc survival analyses in patients receiving >3 or >6 cycles are heavily confounded by selection bias (patients who tolerate more cycles have better prognosis). EFS as a primary endpoint in HR-MDS encompasses heterogeneous events that may obscure OS benefit. The 5-day azacitidine schedule differs from the standard AZA-001 schedule used in practice.
Clinical Context
PANTHER is a negative phase 3 trial. Pevonedistat development for HR-MDS was discontinued following these results. The trial reinforces the difficulty of improving on HMA monotherapy in the upfront HR-MDS setting and highlights the challenges of EFS as a primary endpoint when its component events are heterogeneous. Azacitidine monotherapy remains standard of care in transplant-ineligible HR-MDS per ELN/ESMO guidance, pending results of ongoing combination trials (venetoclax + azacitidine).
References