Skip to main content

Cardiovascular Disease

PCSK9 In the Details

The PCSK9 inhibitors are a class of drugs from which great things are expected. The first ones hitting the market are antibodies, and blocking this enzyme, which is involved in cholesterol homeostasis, clearly has major effects on circulating LDL levels. There are a lot of approaches to going after this target, in the lab, the clinic, and even the courtroom, because the marketing projections are huge for the entire area.

Because of this, people with experience in drug development have been nervously watching the skies, alert for tiny clouds. One such was the FDA’s request that companies monitor cognitive functions during the trials, after some concerns were raised that this mechanism might have an effect in that direction. And just recently, there have been some worries about cardiovascular events (which, to be sure, is one of the problems that such inhibitors are supposed to be preventing). Here’s some correspondence in the New England Journal of Medicine on the topic.

The first letter wonders about the statin-treatment group in a recently reported trial, and notes that they don’t seem to have been treated with appropriate doses by which to make a real-world comparison. The second calls attention to some “methodologic particularities” in a second trial: the safety data look reasonable, but the patients therein appear to be from the subset of those who reported no adverse events in parent studies, which makes it harder to say what the adverse event rate will be in a general population.

In both cases, the authors of the original studies respond. The first point is not denied, but the authors mention that there’s an ongoing trial (ODYSSEY Outcomes) that is specifically comparing a PCSK9 antibody versus high-dose statins in a high-risk population that has already had one cardiovascular event. So that gap should presumably be filled in. The second response also agrees that “data on safety and side-effect profiles in our study come from a cohort of patients who had all successfully received injections and many of whom had received evolocumab for at least 12 weeks”. But they also note that if you pool all the parent trials, the adverse events across them actually look slightly higher in the control group than the PCSK9 group. And in this case, too, there’s a massive outcomes trial (FOURIER), with 27,500 patients (yikes) that should settle this question as well.

That’s what’s worth keeping in mind, though: for cardiovascular drugs, it can take years for the real benefits and risks to become apparent. I think that approving the PCSK9 drugs based on the data in hand is a reasonable decision, but their story is still being written.

8 comments on “PCSK9 In the Details”

  1. Bryan says:

    This makes me think about the recent paper in Nature showing that the phenotypes of genetic knockout animals are often less severe than the phenotype from gene knockdown experiments (doi:10.1038/nature14580). It turns out that animals which have had a gene knocked out from the embryo stage induce compensatory mechanisms that mitigate the negative effects of the knockout, and these compensatory mechanism do not take place (at least in the short term) when gene function is knocked down in adult animals. So, while individuals with PCSK9 LOF may display no ill effects, it’s certainly possible that this is due to compensatory mechanisms that may not occur with pharmalogical knockdown of PCSK9.

    1. Frank N. Stein says:

      Great point, Bryan, that zebrafish paper has also been swimming around my head in the past week. Thanks for your insight.

  2. PharmaHeretic says:

    Let us see how far they can go by using proxy markers for therapeutic efficacy in a unique subset of the patient population. Then again, compensation for upper levels of management will remain the same whether these drugs ultimately succeed or fail.

    On another note- have you noticed that the average number of comments per post has dropped since you migrated to this site (and format style).

  3. sgcox says:

    I noticed that too but it may be an August effect – many people are on holidays.

  4. Bryan says:

    @PharmaHeretic, the average number of posts through the first twenty days of August in 2014 on the Corante site was 20.3. The average number of posts in Aug on the STM site (excluding the first post and troubleshooting post) is 19.5. Probably within the margin of error, especially given the comment numbers from the Corante site are inflated by the double posting errors. There are issues with the new site (e.g. having to click read more for every article), but they do not seem to be discouraging commenters. This past week, however, has been fairly light on commenting.

  5. AvidReader says:

    That read more click… painful.

  6. johnnyboy says:

    aarrrgh…. my finger ! my finger !!!!

  7. Dylan says:

    If you come to the posts via RSS then there’s no read more link. I had to go to the front page just to see what you guys were talking about. I do miss the links to go forward or back a post though.

Comments are closed.