Abemaciclib (Verzenio) plus fulvestrant (Faslodex) numerically extended overall survival (OS) without reaching statistical significance versuss placebo plus fulvestrant in
At a median follow-up of approximately 37 months, the median OS was 32.1 months (95% CI, 27.2-38.1) with abemaciclib plus fulvestrant (n = 182) vs 24.6 months (95% CI, 22.5-28.8) with placebo plus fulvestrant (n = 186; HR, 0.80; 95% CI, 0.60-1.06; P = .11). The prespecified final analysis was triggered at 200 OS events (90 vs 110), and the result did not cross the boundary for statistical significance.
In the simultaneous Annals of Oncology publication, investigators attributed early crossing of the OS curves to more deaths in the abemaciclib arm during the first 6 months (17 vs 8), mostly from disease progression; the curves separated after 12 months.2
"In totality, these data continue to support the consideration of abemaciclib-based treatments after progression on a prior CDK4/6 inhibitor," Kevin Kalinsky, MD, MS, FASCO, of Winship Cancer Institute, Emory University, said in a presentation of the data.
The global, randomized, double-blind, placebo-controlled trial enrolled patients whose disease progressed on a CDK4/6 inhibitor plus an aromatase inhibitor as initial therapy for advanced disease, or recurred on or after adjuvant CDK4/6 inhibitor plus endocrine therapy.2
Between March 2022 and June 2023, 368 patients at 96 centers in 16 countries were randomly assigned 1:1 to abemaciclib at 150 mg or matched placebo twice daily, each with fulvestrant at 500 mg intramuscularly on days 1 and 15 of cycle 1 and every 4 weeks thereafter.
The primary end point was investigator-assessed PFS. Secondary end points included OS and PFS by blinded independent central review (BICR), and exploratory end points included time to second progression (PFS2) and time to chemotherapy (TTC). The final OS analysis was prespecified at 200 events, with a one-sided P boundary of .023.
Median ages were 58 years in the abemaciclib arm and 61 years in the placebo arm; visceral metastases were present in 61.5% and 58.6% of patients, respectively. Across both arms (n = 368), prior CDK4/6 inhibitors included palbociclib (Ibrance; 59.0%), ribociclib (Kisqali; 33.2%), and abemaciclib (7.6%).
Primary analysis results, published in the Journal of Clinical Oncology in December 2024, showed that abemaciclib plus fulvestrant improved investigator-assessed PFS vs placebo plus fulvestrant (HR, 0.73; 95% CI, 0.57-0.95; nominal P = .017), with median PFS of 6.0 months (95% CI, 5.6-8.6) vs 5.3 months (95% CI, 3.7-5.6) and 6-month PFS rates of 50% vs 37%, respectively.3 OS data were immature at that time.1
At the updated analysis, median investigator-assessed PFS was 5.9 months (95% CI, 5.5-8.6) with abemaciclib vs 5.3 months (95% CI, 3.7-5.6) with placebo (HR, 0.78; 95% CI, 0.62-0.99; nominal P = .039). Median BICR-assessed PFS was 13.8 months (95% CI, 9.5-not estimable) vs 5.6 months (95% CI, 4.6-9.2; HR, 0.60; 95% CI, 0.44-0.83; nominal P = .0019). Kalinsky noted that discordance between the 2 PFS assessments was likely affected by informative censoring.
Median PFS2 was 13.9 months (95% CI, 11.8-17.2) vs 12.3 months (95% CI, 10.7-13.8; HR, 0.76; 95% CI, 0.59-0.97; nominal P = .029), and median TTC was 13.6 months (95% CI, 9.7-17.2) vs 9.7 months (95% CI, 7.7-11.5; HR, 0.76; 95% CI, 0.58-0.99; nominal P = .039). Among patients with measurable disease (n = 131; n = 127), the investigator-assessed objective response rate was 18.3% vs 7.1% (nominal P = .006).2
OS results were consistent across prespecified subgroups, with no significant treatment-by-subgroup interaction, including by prior palbociclib (HR, 0.809; 95% CI, 0.561-1.167) or ribociclib (HR, 0.985; 95% CI, 0.620-1.565).1 Among patients who discontinued study treatment (n = 171; n = 178), 77% vs 83% received subsequent systemic therapy, and 51% vs 54% received chemotherapy or an antibody-drug conjugate as first subsequent therapy.
In the safety population (abemaciclib, n = 181; placebo, n = 185), any-grade treatment-emergent AEs (TEAEs) occurred in 97.2% vs 82.2% of patients, grade 3 or higher TEAEs in 57.5% vs 23.2%, and serious AEs in 27% vs 12%.2
The most common any-grade TEAEs with abemaciclib were diarrhea (74.6%), neutropenia (41.4%), anemia (38.1%), and fatigue (37.0%); grade 3 or higher diarrhea and neutropenia occurred in 3.9% and 25.4% of patients, respectively. AEs led to dose reductions in 32.6% vs 4.3% of patients and to discontinuation in 6.1% vs 1.1%; pneumonia was the most frequent AE leading to discontinuation of abemaciclib (n = 3).
The investigators noted that postMONARCH was powered for PFS, with limited power to detect an OS difference, and that balanced access to later-line therapies likely diluted the OS effect of a single line of treatment. They cited the phase 3 EMBER-3 trial (NCT04975308), which included a CDK4/6 inhibitor–pretreated subgroup randomly assigned to imlunestrant (Inluriyo) plus abemaciclib, as converging evidence for continued CDK4/6 inhibition.
The
Disclosures: Kalinsky reported receiving institutional research funding from Novartis, Genentech/Roche, Eli Lilly and Company, Seattle Genetics, AstraZeneca, Daiichi Sankyo, and Ascentage; serving as a consultant for Eli Lilly and Company, Genentech/Roche, Gilead, Novartis, Seattle Genetics, AstraZeneca, Daiichi Sankyo, Puma Biotechnology, Mersana, Menarini Silicon Biosystems, Myovant Sciences, Takeda, Prelude Therapeutics, eFFECTOR Therapeutics, Regor Therapeutics, RayzeBio, Relay Therapeutics, and Merck; and having other relationships with EQRX and ADC Therapeutics.
