Post hoc analyses of recent large, clinical outcomes studies of PCSK9 inhibitors have revealed two tantalizing and unexpected potential benefits from these drugs: an ability to substantially reduce the incidence or severity of venous thromboembolism and aortic stenosis.
Dr. Nicholas A. Marston
The evidence also suggests that these effects are linked to the ability of these drugs to reduce blood levels of Lp(a) lipoprotein by roughly a quarter, currently the biggest known effect on Lp(a) levels of any approved medication.
One study ran post hoc analyses of venous thromboembolism (VTE) events in the FOURIER pivotal trial of evolocumab (Repatha), with more than 27,500 randomized patients (N Engl J Med. 2017 May 4; 376[18]:1713-22), and in the ODYSSEY OUTCOMES pivotal trial of alirocumab (Praluent), with nearly 19,000 randomized patients (N Engl J Med. 2018 Nov 29;379[22]:2097-2107). The analyses showed that, with evolocumab treatment, the incidence of VTE events fell by a statistically significant 29%, compared with patients on placebo, while in ODYSSEY OUTCOMES patients treated with alirocumab had a 33% cut in VTE events, compared with placebo-treated patients, a difference that just missed statistical significance (Circulation. 2020 Mar 29. doi: 10.1161/CIRCULATIONAHA.120.046524) in analyses that were not prespecified before these trials started, Nicholas A. Marston, MD, said in a presentation of his research during the joint scientific sessions of the American College of Cardiology and the World Heart Federation, which was presented online this year. ACC organizers chose to present parts of the meeting virtually after COVID-19 concerns caused them to cancel the meeting.
ODYSSEY Outcomes: Alirocumab cut stroke, PAD, VTE
A combined analysis of 46,488 patients from both studies showed a 31% cut in VTE events with PCSK9 inhibitor treatment, a highly significant finding using VTE endpoints that were not specifically tallied nor adjudicated but collected as part of the serious adverse event reporting in the two pivotal trials, said Dr. Marston, a cardiologist at Brigham and Women’s Hospital in Boston. This is the first report of a statistically significant link between treatment with PCSK9-inhibiting agents and a reduction in VTE, he added. Researchers from the ODYSSEY OUTCOMES trial had reported a VTE analysis in 2019, and while data from that trial on its own showed a nominal 33% lower VTE rate with alirocumab treatment, it just missed statistical significance.
The VTE effect took about a year on treatment to start to manifest. During the first 12 months of FOURIER, the rate of VTE events among patients in the two treatment arms was virtually identical. But starting during months 13-18 on treatment, the event curves in the two arms began to increasingly diverge, and overall during the period from month 13 to the end of the study treatment with evolocumab was linked with a statistically significant 46% reduction in VTE events, compared with patients who received placebo. The results Dr. Marston reported were also published online (Circulation. 2020 Mar 29. doi: 10.1161/CIRCULATIONAHA.120.046397).
The suggestion that this association may be linked to the impact of PCSK9 inhibitors on Lp(a) came from an additional analysis that Dr. Marston presented, which looked at the link between evolocumab use and a change in VTE event rates, compared with placebo, depending on baseline lipoprotein levels. Evolocumab treatment was associated with a roughly similar, modest, and not statistically significant reduction in VTE events, compared with placebo regardless of whether patients had baseline levels of LDL cholesterol below the median or at or above the median. In contrast, when a similar analysis divided patients based on whether their Lp(a) level at baseline was below, or at or above, the median the results showed no discernible effect of evolocumab treatment, compared with on VTE events in patients with lower baseline Lp(a), but in those with higher levels treatment with evolocumab linked with a 48% cut in VTE events, compared with placebo, a statistically significant difference.
In FOURIER, treatment with evolocumab lowered baseline Lp(a) levels by a median of 27%, compared with placebo, among the 25,096 enrolled patients who had their baseline levels measured. As previously reported, prespecified analysis of FOURIER data also showed that the impact of evolocumab, compared with placebo, on the combined rate of coronary heart disease death, MI, or need for urgent coronary revascularization was enhanced among patients with elevated baseline Lp(a) and moderated in those who entered with lower levels. Among patients who entered FOURIER with Lp(a) levels at or below the median treatment with evolocumab cut the primary endpoint by 7%, compared with placebo, a difference that was not statistically significant. Among patients who began the study with Lp(a) levels above the median, evolocumab treatment cut the primary endpoint by 23%, compared with placebo, a statistically significant effect (Circulation. 2019 Mar 19;139[12]:1483-92).
The aortic stenosis connection
A second study reported in the online scientific sessions (Abstract 914-08) used only FOURIER data, and showed that patients treated with evolocumab had a roughly similar response pattern in their incidence of aortic stenosis (AS) events as they did for VTE events.
During the first year of the study, the incidence of AS events was virtually identical among patients treated with evolocumab and those who received placebo. But after the first 12 months and through the study’s end, patients on evolocumab showed a statistically significant 52% relative reduction in AS events, compared with control patients, said Brian A. Bergmark, MD. For the entire study duration, treatment with evolocumab linked with a 34% relative reduction in AS events, compared with placebo, a difference that did not reach statistical significance, added Dr. Bergmark, an interventional cardiologist also at Brigham and Women’s Hospital. The observed halving in total AS events that linked with evolocumab treatment after the first year of the study included a similar-magnitude reduction specifically in the incidence of aortic valve replacement procedures in the evolocumab-treated patients.
Further analysis of both total AS events and aortic valve replacements in FOURIER patients showed that they occurred at a significantly elevated rate in patients who entered the study with higher baseline Lp(a) levels in a multivariate analysis, but a similar analysis showed no significant association between the incidence of these AS-related events and baseline levels of LDL cholesterol, he said.
The AS analysis carried the same important limitations as the VTE analysis: It ran on a post hoc basis and focused on events that were relatively uncommon and not adjudicated, Dr. Bergmark cautioned. Nonetheless, other investigators saw important potential implications from both the VTE and AS observations, with the huge caveat that they need replication in prospective studies designed to specifically address the validity of these findings.
What it could mean
These observed associations between PCSK9 inhibitor treatment and apparent reductions in the rate of both VTE and AS events “represent a tremendous clinical breakthrough,” commented Michelle L. O’Donoghue, MD, a cardiologist at Brigham and Women’s Hospital who is a FOURIER coinvestigator and has led some of the Lp(a) analyses run from that study.
Bruce Jancin/MDedge News
Dr. Michelle L. O'Donoghue
“To date, we have not identified any therapies that slow progression of AS. Other classes of lipid-lowering therapies, such as statins, have been tested and not demonstrated a significant effect,” Dr. O’Donoghue said in an interview.
“For AS, the results are very intriguing. If confirmed, it could be groundbreaking. AS is the most common valve disease in the developed world, and no medical therapy exists. The potential is immense,” commented George Thanassoulis, MD, director of preventive and genomic cardiology at McGill University, Montreal. “Having a medical treatment that could slow AS progression would completely change the disease. It’s conceivable to slow the disease enough that patients may never require valve replacement.” But an interview he cautioned that, “although the results are exciting, the analysis has many limitations. What we need is a dedicated, randomized trial for AS. I hope this stimulates that.”
“For VTE, it’s an interesting finding, but I don’t think it will have clinical utility because we have good treatment for VTE,” added Dr. Thanassoulis, but others saw more opportunity from what could be a new way to reduce VTE risk.
Dr. Gregory Piazza
“Given that many patients have difficulty with the bleeding risk from anticoagulants, this option [a PCSK9 inhibitor] may be quite welcome for preventing VTE,” commented Gregory Piazza, MD, a cardiologist and VTE specialist at Brigham and Women’s Hospital who was not involved in any of the PCSK9 inhibitor studies.
“At this time we would not suggest that PCSK9 inhibitors replace an anticoagulant for patients with an established clot or at high risk for a recurrent clot, but if patients have an indication for a PCSK9 inhibitor, the further reduction in venous clot can be viewed as an additional benefit of this therapy,” said Dr. O’Donoghue.
How it might work
A possible mechanism underlying a VTE effect is unclear. Results from the JUPITER trial more than a decade ago had shown a significant association between treatment with 20 mg/day of rosuvastatin and a cut in VTE episodes, compared with placebo, in a prespecified, secondary analysis of the trial with nearly 18,000 patients selected for having a relatively high level of high-sensitivity C-reactive protein (N Engl J Med. 2009 Apr 30;360[18]:1851-61). But a meta-analysis of 29 controlled statin trials that used a variety of statin types and dosages (and included the JUPITER results) failed to confirm a statistically significant change in VTE rates from statins, though they produced a small, nominal reduction (PLoS Med. 2012 Sep 18. doi: 10.1371/journal.pmed.1001310).
Lp(a) “has long been linked to thrombosis, in particular arterial thrombosis,” so the link observed in the PCSK9 inhibitor trials “is not surprising,” said Dr. Piazza. Dr. O’Donoghue agreed that prior evidence had “suggested a prothrombotic role for Lp(a).”
Dr. Thanassoulis was more skeptical of a Lp(a) connection to VTE. “There has always been controversy regarding the prothrombotic effects of Lp(a) and whether it’s clinically relevant,” he said. “The genetic data, from Mendelian randomization studies, is not consistent” with a Lp(a) and VTE link.
The association of AS and Lp(a) may be stronger. “Our team showed that people with genetic variants that predispose to high Lp(a) have a much higher incidence of AS,” Dr. Thanassoulis noted. “We and others have also demonstrated that both Lp(a) and LDL are likely causal mediators of aortic valve calcification and stenosis.”
Dr. O’Donoghue also cited observational genetic data that linked elevated Lp(a) with AS. “Mendelian randomization studies have demonstrated that Lp(a) is a causal contributer to AS, and evolocumab reduced Lp(a) by 25%-30%, raising the possibility that Lp(a) lowering with these drugs may be the mechanism,” she said.
The future of Lp(a) lowering
This last point from Dr. O’Donoghue, that PCSK9 inhibitors cut Lp(a) levels by about 25%-30%, means that they are the most potent Lp(a)-lowering agents currently available, but it also leaves lots of room for other agents to do even better in cutting Lp(a).
“There are now drugs in development that block production of the Lp(a) protein and dramatically reduce its concentration, by about 80%,” Dr. O’Donoghue noted. “It will be of interest to study whether these novel therapies, now in phase 2 and phase 3 studies, have any effect on the risk for VTE and AS.”
“Several drugs in development, including antisense RNA and RNA-interfering molecules, are much more potent and lower Lp(a) by 80%-90%. Because of this potency they can completely normalize Lp(a) in most patients. For Lp(a) lowering, the future is in these new molecules. Randomized trials have started, and we will hopefully have some results in about 5 years,” said Dr. Thanassoulis.
Until then, the prospect of possibly soon documenting benefits from PCSK9 inhibitors beyond their impact on cutting LDL cholesterol raises some hope to get more bang for the considerable buck these drugs cost. But Dr. Thanassoulis was skeptical it would move the cost-benefit ratio much. “VTE and AS are relatively rare, compared with atherosclerotic cardiovascular events, and therefore the added value at the population level would be small,” he predicted. But if treatment with a drug could help patients avoid surgical or percutaneous valve interventions “that could be really interesting from a cost-benefit perspective.”
FOURIER was funded by Amgen, the company that markets evolocumab (Repatha). ODYSSEY OUTCOMES was funded by Sanofi and Regeneron, the companies that developed and market alirocumab (Praluent). Dr. Marston had no disclosures. Dr. Bergmark has been a consultant to Daiichi Sankyo, Janssen, Quark, and Servier and has received research funding from Abbott Vascular, AstraZeneca, and MedImmune. Dr. O’Donoghue has been a consultant to and has received research funding from Amgen; has been a consultant to Janssen and Novartis; and has received research funding from AstraZeneca, Eisai, GlaxoSmithKline, Janssen, Medimmune, Merck, and The Medicines Company. Dr. Thanassoulis has been an adviser to and speaker for Amgen; an adviser to Ionis and Sanofi/Regeneron; a speaker on behalf of Boehringer Ingelheim, Sanofi, and Servier; and has received research funding from Ionis and Servier. Dr. Piazza has been a consultant to Optum, Pfizer, and Thrombolex and he has received research funding from Bayer, Bristol-Myers Squibb, Daiichi Sankyo, Ekos, Janssen, and Portola.