DUBLIN – Proprotein convertase subtilisin–kexin type 9 (PCSK9) inhibitors are as good as but no better than statins in reducing the risk of cardiovascular events in patients with cardiovascular disease, according to a large-scale Mendelian randomization study published this week in the New England Journal of Medicine. The study also found an association between PCSK9 inhibition and an increased diabetes risk in patients with impaired fasting glucose, a form of pre-diabetes characterized by elevated blood glucose levels during fasting.

"The diabetes findings were slightly unexpected," corresponding author, Brian Ference, of the School of Medicine at Detroit-based Wayne State University, told BioWorld Today. Nevertheless, the risk does not outweigh the benefit of PCSK9 inhibition. "The clinical benefit of cardiovascular risk reduction is likely to far exceed the clinical relevance of any increased risk of diabetes," he said.

The study's key finding is that the magnitude of cardiovascular (CV) risk reduction is driven by the absolute reduction in low-density lipoprotein cholesterol (LDL-C), not the percentage reduction. "There is still a persistent expectation that PCSK9 inhibitors will reduce risk by 50 percent," Ference said. His analysis suggests the risk reduction is closer to 25 percent. The effect is additive to the benefit offered by statin therapy, as has already been seen in numerous clinical trials to date.

The study anticipates the findings of two large-scale CV outcomes trials involving the approved PCSK9-targeting antibodies, Repatha (evolocumab) and Praluent (alirocumab), marketed by Amgen Inc., of Thousand Oaks, Calif., and by Paris-based Sanofi SA and Tarrytown, N.Y.-based Regeneron Pharmaceuticals Inc., respectively.

Although each gained FDA approval last year on the basis of their dramatic effects on lowering cholesterol, the extent to which that reduction translates into clinical benefit is still not clear. The outcomes data are key to underwriting the commercial potential of each drug. Neither has made much headway since their respective launches last year. Repatha reached $40 million in sales in the third quarter this year, while Praluent took in $38 million during the same period.

Top-line data from Amgen's Fourier trial, which has recruited 27,500 patients, are expected in the first quarter of next year. A recent look at interim data from Sanofi's and Regeneron's Odyssey Outcomes trial, which has recruited more than 18,000 patients, failed to convince an independent data monitoring committee that the trial should be stopped early on efficacy grounds. Top-line data are due in late 2017.

Ference already has a picture of what the data will look like. "Because the baseline LDL-C among patients enrolled in the ongoing PCSK9 CV outcome trials is around 90 mg/dL and because a 50-60 percent reduction in LDL-C will translate into a 45-50 mg/dL absolute reduction in LDL-C, the PCSK9 outcome trials are very likely to report a 22-25 percent risk reduction, not the 'irrationally exuberant' 50 percent risk reduction that many are expecting," he stated.

Mendelian randomization employs what Ference terms "genetic instruments" to introduce a randomization scheme into an observational trial. Ference and colleagues previously applied the same approach to correctly predict the outcome of the Improve-It trial, which assessed the benefit of adding Merck & Co. Inc.'s Vytorin (ezetimibe) to statin therapy.

In the current study, the group compared the effects on the risk of both cardiovascular disease and diabetes of independently inherited genetic variants in the gene encoding PCSK9 with variants in the gene encoding 3-hydroxy-3-methylglutaryl–co-enzyme A reductase, the target of statins. The method is particularly apt when applied to PCSK9 inhibition, as the drug class was conceived as means way of recapitulating in a drug therapy the favorable phenotypes observed in individuals with loss-of-function mutations in the PCSK9 gene.

The study involved a total of 112,772 subjects, drawn from 14 clinical trials. The dataset included 14,120 cardiovascular events and 10,635 cases of diabetes. Seven variants were included in the PCSK9 genetic score, while six variants were included in the HMGCR genetic score. Subjects were "randomized" into roughly equal arms based on whether their genetic scores for PCSK9 inhibition were above or below the median. A dose-response relationship was assessed by assigning participants to one of four groups on the basis of their quartile genetic scores. The same procedure was followed for the HMGCR, and the separate and additive effect of the two scores was assessed in a 2x2 factorial analysis.

Because the genetic variations that mimicked PCSK9 inhibition and statin therapy had similar effects on lowering CV risk and on raising the diabetes risk, the group posited a common underlying mechanism. That led to an analysis of LDL receptor-mediated pathways playing a role in increasing the diabetes risk. A genetic score based on variants in the gene encoding the LDL receptor mirrored the findings for PCSK9 and HMGCR variation. "Whether it actually causes diabetes is very hard to say," Ference said. The association will motivate further research in the area, he added.

Ference's study appears in the Dec. 1, 2016, issue of the New England Journal of Medicine, under the title "Variation in PCSK9 and HMGCR and Risk of Cardiovascular Disease and Diabetes." It adds further evidence to the debate on the benefits of PCSK9 inhibitors – but for those with genetically inherited conditions that cause high LDL-C levels, such as familial hypercholesterolemia, the therapy remains "a transformation of care," he said.

Meanwhile, New York-based Pfizer Inc. has abandoned plans for a €400 million (US$432 million) expansion at its Grange Castle biologics manufacturing campus near Dublin, following its recent decision to terminate phase III development of bococizumab. (See BioWorld Today, Nov. 2, 2016.)