Long-term lipid lowering with evolocumab (Repatha) further reduces cardiovascular events, including CV death, without a safety signal, according to results from the FOURIER open-label extension (OLE) study.
In the parent FOURIER trial, treatment with the PCSK9 inhibitor over a median of 2.2 years reduced the primary efficacy endpoint by 15% but showed no CV mortality signal compared with placebo in patients with atherosclerotic disease on background statin therapy.
Now with follow-up out to 8.4 years — the longest to date in any PCSK9 study — cardiovascular mortality was cut by 23% in patients who remained on evolocumab compared with those originally assigned to placebo (3.32% vs 4.45%; hazard ratio [HR] 0.77; 95% CI, 0.60 – 0.99).
The Kaplan-Meier curves during FOURIER were “essentially superimposed and it was not until the open-label extension period had begun with longer-term follow up that the benefit in terms of cardiovascular mortality reduction became apparent,” principal investigator Michelle O’Donoghue, MD, MPH, of Brigham and Women’s Hospital, Boston, said.
Pivotal statin trials have median follow-up times of 4-5 years and demonstrated both a lag effect, meaning clinical benefit grew over time, and a legacy effect, where clinical benefit persisted in extended follow-up after the parent study, O’Donoghue observed.
With shorter follow-up in the parent FOURIER trial, there was evidence of a lag effect with the risk reduction in CV death, myocardial infarction (MI), and stroke increasing from 16% in the first year to 25% over time with evolocumab.
FOURIER-OLE enrolled 6635 patients (3355 randomly assigned to evolocumab and 3280 to placebo), who completed the parent study and self-injected evolocumab subcutaneously with the choice of 140 mg every 2 weeks or 420 mg monthly. Study visits were at week 12 and then every 24 weeks. Median follow-up was 5 years.
Their mean age was 62 years, three fourths were men, a third had diabetes. Three fourths were on a high-intensity statin at the time of enrollment in FOURIER, and median LDL-cholesterol (LDL-C) at randomization was 91 mg/dL (2.4 mmol/L).
At week 12, the median LDL-C was 30 mg/dL (0.78 mmol/L), and this was sustained throughout follow-up, O’Donoghue reported. Most patients achieved very low LDL-C levels, with 63.2% achieving levels less than 40 mg/dL (1.04 mmol/L) and 26.6% less than 20 mg/dL (0.52 mmol/L).
Patients randomly assigned in the parent trial to evolocumab vs placebo had a 15% lower risk of the primary outcome of CV death, MI, stroke, hospitalization for unstable angina, or coronary revascularization (15.4% vs 17.5%; HR, 0.85; 95% CI, 0.75 – 0.96).
Their risk of CV death, MI, or stroke was 20% lower (9.7% vs 11.9%; HR, 0.80; 95% CI, 0.68 – 0.93), and, as noted previously, 23% lower for CV death.
When major adverse cardiovascular events data were parsed out by year, the largest LDL reduction was in years 1 and 2 of the parent study (Δ 62 mg/dL between treatment arms), “highlighting that lag of benefit that continued to accrue with time,” O’Donoghue said.
“There was then carryover into the extension period, such that there was legacy effect from the LDL delta that was seen during the parent study,” she said. “This benefit was most apparent early on during open-label extension and then, as one might expect when all patients were being treated with the same therapy, it began to attenuate somewhat with time.”
Although early studies raised concerns that very low LDL cholesterol may be associated with an increased risk of hemorrhagic stroke and neurocognitive effects, the frequency of adverse events did not increase over time with evolocumab exposure.
Annualized incidence rates for patients initially randomized to evolocumab did not exceed those for placebo-treated patients for any of the following events of interest: serious safety events (10% vs 13%), hemorrhagic stroke (0.04% vs 0.05%), new-onset diabetes (1.2% vs 2.3%), muscle-related events (1.2% vs 1.9%), injection site reactions (0.4% vs 0.7%), and drug-related allergic reactions (0.6% vs 1.1%).
“Long-term use of evolocumab with a median follow-up of more than 7 years appears both safe and well-tolerated,” O’Donoghue said.
Taken together with the continued accrual of cardiovascular benefit, including CV mortality, “these findings argue for early initiation of a marked and sustained LDL-cholesterol reduction to maximize benefit,” she concluded.
Translating the Benefits
Professor Ulrich Laufs, MD, Leipzig University Hospital, Germany, and invited commentator for this session, said the trial addresses two key issues: (1) the long-term safety of low LDL-C lowering and (2) the long-term safety of inhibiting PCSK9, which is highly expressed not only in the liver but also in the brain, small intestine, and kidneys. Indeed, an LDL below 30 mg/dL is lower than the ESC treatment recommendation for very high risk patients and is, in fact, lower than most assays are reliable to interpret.
“So it is very important that we have these very clear data showing us that there were no adverse events, also including cataracts and hemorrhagic stroke, and these were on the level of placebo and did not increase over time,” he said.
The question of efficacy is triggered by observations of another PCSK9, the humanized monoclonal antibody bococizumab, which was associated in the SPIRE trial with an increase in LDL-C over time due to neutralizing antibodies. Reassuringly, there was “completely sustained LDL reduction” with no neutralizing antibodies with the fully human antibody evolocumab in FOURIER-OLE and in recent data from the OSLER-1 study, Laufs observed.
Acknowledging the potential for selection bias with an open-label extension program, Laufs said there are two important open questions: “Can the safety data observed for extracellular PCSK9 inhibition using an antibody be transferred to other mechanisms of PCSK9 inhibition? And obviously, from the perspective of patient care, how can we implement these important data into patient care and improve access to PCSK9 inhibitors?”
With regard to the latter point, he said physicians should be cautious in using the term “plaque regression,” opting instead for prevention and stabilization of atherosclerosis, and when using the term “legacy,” which may be misinterpreted by patients to imply there was cessation of therapy.
“From my perspective, [what] the open-label extension really shows is that earlier treatment is better,” Laufs said. “This should be our message.”
In a press conference prior to the presentation, ESC commentator Johann Bauersachs, MD, Hannover Medical School, Germany, said “this is extremely important data because it confirms that it’s safe, and the criticism of the FOURIER study that mortality, cardiovascular mortality, was not reduced is now also reduced.”
Bauersachs said it would have been unethical to wait 7 years for a placebo-controlled trial and questioned whether data are available and suggestive of a legacy effect among patients who did not participate in the open-label extension.
O’Donoghue said unfortunately those data aren’t available but that Kaplan-Meier curves for the primary endpoint in the parent trial continued to diverge over time and that there was somewhat of a lag in terms of that divergence. “So, a median follow-up of 2 years may have been insufficient, especially for the emerging cardiovascular mortality that took longer to appear.”
The study was funded by Amgen. Donoghue reports research grants from Amgen, AstraZeneca, Janssen, Intarcia, and Novartis, and consulting fees from Amgen, Novartis, AstraZeneca, and Janssen. Laufs reports honoraria/reimbursement for lecture, study participation, scientific cooperation with Saarland or Leipzig University, and relationships with multiple pharmaceutical and device makers.
Circulation. Published online August 29, 2022. Abstract