Skip to main content

Cardiovascular Disease

A Low-Cholesterol Vaccination?

We have dueling PCSK9 antibodies on the market – Praluent (alirocumab) and Repatha (evolocumab) – and they’re both out there lowering LDL levels as we speak. (Whether these have the desired long-term outcomes remains to be seen; there’s no other way to find out than to wait and watch). But there’s another approach to this same pathway, as detailed in this release: a vaccine.

A new cholesterol-lowering vaccine leads to reductions in ‘bad’ LDL cholesterol in mice and macaques, according to research published in Vaccine. The authors of the study, from the University of New Mexico and the National Institutes of health in the United States, say the vaccine has the potential to be a more powerful treatment than statins alone.

Well, yeah, but taking the approved drugs has the potential for that, too, and that’s a better comparison. What does the vaccine bring that the antibody therapies don’t? One the one hand, the vaccine might be a one-time-only treatment, and that’s good. On the other hand, the downside is that the vaccine might be a one-time-only treatment, and that’s bad: you can’t turn it off, the way that you can decide to stop taking the antibody.

The authors have thought about this, but their first reason for the desirability of a vaccine is the cost of the antibody therapies, followed by patient compliance, and followed by the development of an immune response to the antibodies themselves and corresponding dropoffs in efficacy.

The induction of anti-self antibody responses understandably raises some safety concerns and should be reserved for those targets where the consequences of antibody-mediated inhibition are well known and the benefits of vaccination outweigh potential risks. At this point, PCSK9 appears to meet these criteria. PCSK9 mAb clinical trials have not revealed any major safety concerns and humans with mutations that abolish PCSK9 expression are apparently healthy. Nevertheless, vaccination would presumably elicit longer lasting antibody responses to PCSK9 and, in contrast to mAb-based therapies, it would not be possible to rapidly decrease the serum Ab levels in response to adverse side effects.

They suggest that a vaccine route might be particularly useful in developing countries, but I wonder if you’re still a developing country in that sense if high LDL levels are a major public health concern. Overall, I’d say that if it turns out that modern humans are better off without PCSK9 function, then so be it – we should knock it down with whatever tools are at hand. But I think that we can afford to approach that idea slowly – Chesterton’s fence and all that. A billion years of evolution have done some funny things and made some funny connections, and it would be worth watching the antibody performance for a few years before going further.

22 comments on “A Low-Cholesterol Vaccination?”

  1. scientist says:

    Is there precedent for vaccination against human proteins?

    1. vetstat says:

      there is an anti-gnrf vaccine to control sex hormones in animals (improvac). Early versions I was familiar with had a tendency to wear off after a few years. I think there was some variation between animals. The vaccine would also work in humans but I don’t think they ever tried to go down that route.

    2. Cellbio says:

      In an unattended way….human proteins administered as therapeutics generate an antibody response. Rates are not high, especially for neutralizing antibodies. Things get worse if the therapeutic is aggregated. So, one could drive a response with adjuvant and antigen form, but with all the concerns for subsequent challenges mentioned by others.

  2. PharmaHeretic says:

    What could go wrong with a vaccine that elicits an immune response to an endogenous protein? After all, humans are as inbred as a batch of lab animals. It is not like different people produce antibodies of varying specificities in response to the same antigen. It is also not as if the extent and magnitude of the immune response in individuals exhibits considerable variation.

  3. Rule (of 5) Breaker says:

    I’m with PharmaHeretic. This seems like a bad idea. Also, If we were better off without PCSK9 function, guess what? We would have lost it by now. I am thinking that if this moves forward and a batch of people are actually vaccinated, then we are going to find out the full extent of PCSK9 function when things go terribly wrong.

    1. Barry says:

      We would have lost the expression of PCSK9 if it were purely deleterious to us before breeding age. Selective pressure after that is far weaker.

  4. Mark Thorson says:

    What if we could edit PSCK9 out of the human genome? Wouldn’t that be the best thing of all, now that we presumably know it has no good functions only bad ones?

  5. Chrispy says:

    You’d probably need a heck of a good adjuvant to get an immune response to a human protein — they used virus-like particles in the paper. And you’d definitely run the risk of autoimmunity through epitope spreading. Plus how would you keep the anti-PCSK9 levels stable over years? The whole approach has enough obvious flaws that I suspected the authors were really not serious about developing a real therapy until I read this:

    National regulators in the U.S. and Europe have approved this type of design for phase I/II trials of other auto-antibody-inducing vaccines [19], [43] and [44]. Thus, there is ample precedent for clinical trials of a vaccine targeting PCSK9.

    So VLPs expressing angiotensin II have been tried in the clinic, as has immunization with aggregated Abeta. Apparently the latter did not go so well:

    However, treatment had to be discontinued prematurely in this trial since 6% of the vaccinated patients developed aseptic meningo/-encephalitis. (Human Vaccines & Immunotherapeutics 10:4, 847–851; April 2014)

    Apparently it was self-reactive T cells that were the problem — this seems like it would be a fundamental issue with a vaccination approach that used self antigens.

  6. johnnyboy says:

    I would concur with the other commenters. Beyond the potential loss of unknown non-LDL functions of PCSK9 (which would probably have been hinted at in clinical studies of the currently approved antibodies), I would worry about a sustained life-long immune reaction. Since PCSK9 is continuously produced and secreted, the constant production of antibodies and possible antibody-dependent cell cytoxicity would set the body up for a lifetime of immune-complex injuries (renal, joint, vascular, among others) and direct cytoxicity of cells expressing PCSK9 (liver, intestine). Any semi-intelligent regulatory body should ask for long-term (1 year +) primate studies before allowing this into phase 1.

  7. NJBiologist says:

    People have been proposing immunocontraception for something like 30 years now, and that’s failed to move forward because of some of the concerns above. For example, much of the preclinical work uses purified zona pellucida antigen; however, this has been linked to oophoritis–likely an immune attack on the oocyte–in clinical trials (see S Lekhwani et al 2014 Adv Biomed Res vol. 3 p. 247 for a review).

  8. PharmaHeretic says:

    So I hear that Roche is going to lay off more than a thousand people and close down four of its sites. Any comments?

  9. tally ho says:

    In case you didn’t already read it, here’s a good overview of human PCSK9 loss-of-function, both hetero and homozygous:

    The genetic/phenotype findings are quite compelling. However, to the concerns raised above, I wonder whether healthy individuals with homozygous PCSK9 loss-of-function may have secondary genetic differences that compensate for loss of PCSK9, yielding a healthy phenotype (non-LDL function). In other words, is PCSK9 loss-of-function in a naïve genetic population healthy (as questioned above)? Does anyone know how well the bioinformatics experts double-checked the genetic background of healthy PCSK9 KO phenotypes to see if there were associated (compensating) genetic differences?

  10. Two things says:

    1. Reminds me of Elan’s self-vaccine against Abeta, which caused deadly encephalitis.
    2. Evolution doesn’t usually spend hundreds of millions of years just to make proteins that are harmful to the organism. That’s not how natural selection works. There is usually some important function for every native protein, so inhibition should always be modulated.

  11. dearieme says:

    And all because maybe, perhaps,conceivably, it might extend your life by a couple of weeks in the dementia ward. Ain’t science wonderful?

  12. SP says:

    “We would have lost it by now”
    “Evolution doesn’t usually spend hundreds of millions of years just to make proteins that are harmful to the organism. ”
    The time in which the selective pressures of lipid metabolism have changed are on the order of thousands of years, not millions- think agricultural revolution time frame.

    1. a. nonymaus says:

      On the other hand, thousands of years is the timeframe over which we have evolved things like lactose tolerance (and thrice at that). However, the sort of caloric abundance where we need to worry about this enzyme has become a selection pressure only on a scale of a century or so rather than thousands of years.

      1. Anon says:

        … so these people should be fine as long as they don’t ever skip breakfast?

  13. Anchor says:

    @ PharmaHeretic-Do you think suppressing PCSK9 with antibodies could also result in suppressing the useful cellular role for cholesterol, like bio-conjugation? I think this vaccination issue could complicate the life for those individuals whose cholesterol level is marginally high, no?

  14. JK says:

    I agree with the need for caution, but in regards to evolved function it’s definitely worth thinking about life history theory. It’s possible that, for example, PCSK9 is a net positive below the age of 40 and a net negative above. That would be completely compatible with evolution, and make the case for vaccination later in life. I make the point completely as an abstract possibility – I haven’t looked at any evidence.

  15. PCSK9 vax says:

    I have worked on an anti-PCSK9 vaccine at a major pharma, so I might be able to offer some insight here.

    1. Lots of immune responses look *great* in mice, then weak in non-human primates, then near-zero in humans. We’ve seen this many times in vaccine development. Potent adjuvants can overcome this, but not always. A poor immunogen just can’t be rescued with any adjuvant that’s reasonably free of adverse events.

    2. Anti-self proteins are particularly poor immunogens, especially something as ubiquitous as PCSK9.

    3. The concerns about persistent autoimmunity are very real, and these are hard to eliminate in the development process, so most of these types of projects get killed long before clinical trials, even if the data look good up till then.

    4. Given that an anti-PCSK9 antibody therapy exists, the reasons for a vaccine are not compelling. Ethically, cost can’t be factor. Compliance is not a problem for a long-lasting human mAb. Then you are left with the cold-chain problem. I would presume the same persons at risk due to sedentary lifestyles and changing diets are also reachable by cold trucks.

    In short, these studies are interesting but not developable.

  16. tangent says:

    @PKCS9vax, that leaves me scratching my head, is management knowingly doing noncommercial pure research? Or do they think they’ve got a worthwhile lottery ticket in this?

  17. metaphysician says:


    Almost certainly that latter. Presumably they are gambling on the idea that the protein really isn’t *that* critical, and a vaccine will be a one-shot cure for heart disease, or whatnot.

Comments are closed.