Skip to main content

Alzheimer's Disease

Simple, Right?

So, name a class of prescription drugs that lots of people have been taking daily over a period of decades: how about statins? From a distance, the story is completely understandable: statins inhibit the enzyme HMG-CoA reductase, which shuts down a key step in cholesterol biosynthesis. That lowers the amount of cholesterol in the body, and that lowers the blood levels of low-density lipoprotein (LDL), and that leads to lower rates of cardiovascular mortality. There you have it.

But step up closer and look again. We’ve already had some puzzling results about the cardiovascular benefits of lowering LDL per se, not to mention several whalloping failures in seeing good effects by raising HDL, which is (from that same distance above) just as solid a hypothesis. This is why the current PCSK9 therapies are still involved in large outcomes studies. No one doubts that they lower LDL, but no one’s completely sure if this is going to be one of those lowers-LDL-and-does-good stories or lowers-LDL-and-something-else-happened stories. You’d think the former, but you’d think a lot of things (I know I do).

OK, come over and look at this story from that same closer distance, but from another vantage point. So what do statins actually do? There have been reports over the years of decreases in some forms of cancer in statin patients versus matched control populations, and reports of similar effects in Alzheimer’s (see this new paper for more on that). These studies are not easy to do, because they’re mostly observational, but the flip side is that you can potentially use observational data collected from tens of millions of people. Over the years, more and more people have come to think that statins as a class do more than just lower LDL.

One of these things is almost certainly to affect prenylation. That’s one of the many post-translational modifications to proteins, and its availability is also tied to the mevalonate pathway that’s affected by statin effects of HMG-CoA reductase. Messing around with prenylation can have a number of effects, including tumor progression and modulation of the immune system. Prenylation is thought to be the mechanism by which statin treatment makes fruit flies live significantly longer – interestingly, it appears that the effect works through improvements in their small fruit fly hearts, which might well be a mechanism by which statins improve cardiovascular mortality in humans completely outside the LDL pathway. There’s a lot of argument about that last point, but it’s more about how large these beneficial pleiotropic effects are than whether they exist.

But if you go ask a thousand people how statin drugs work, to the extent that you’ll get any answers at all, they’re almost certainly going to be “They lower your cholesterol”. Lots of people are surprised to find out that cholesterol is necessary for life, although that certainly doesn’t mean that having heaps of it in your body =  more life (although having extremely low levels is actually correlated with early mortality). The story is more complicated, and it’s more complicated at every level.

Take as another example that link above to the new paper on statin effects on the incidence of Alzheimer’s disease. It will be easy to write a headline (someone probably will) that Cholesterol Drugs Fight Alzheimer’s, but it’s tricky. The paper found varying results depending on which statin was used in which population (by ethnicity and gender). Even with the huge Medicare data set the authors used, it’s hard to say just what’s going on, and hard (as yet) to made any recommendations. Pharmacokinetics (including blood-brain barrier penetration), metabolic differences between different groups, and who knows how many other variables are affecting the data.

So that brings us back around to the opening paragraph: here are some of the most widely taken, most well-studied drugs in the industrialized world, and we’re still trying to figure out what they do and how they do it. This is worth a thought whenever someone tries to pretend that drug discovery and development isn’t ( or shouldn’t be) so complicated. It is, it is.



49 comments on “Simple, Right?”

  1. steve says:

    This is exactly why pharma is so crazy to insist on target engagement data, mechanism, etc. before moving forward with a drug candidate. The simple fact is that we don’t know how any of this stuff works in vivo. The idea that some bi-molecular interaction in a test tube actually tells you how a drug works in the body is insane. How many perfectly good drugs have been dropped by pharma because they thought the mechanism wasn’t understood? I know of several where the preclinical (and even early clinical) data were great but they were dropped because of mechanism questions only to find out years later that the target just hadn’t been discovered yet. If they’d only realize that we don’t really know the mechanism for anything maybe they’d begin to rethink their approach and actually develop new drugs.

    1. Me says:

      But what’s the alternative? en masse in vivo phenotypic screening?

      I’ve seen it done and it ain’t pretty….

      1. Dr CNS says:

        Not a surprise, as we know very little about human physiology, in spite of what some folks believe.

        So, here’s an alternative: slowing down the pace of technology-driven research, and use that energy to think about better questions we need to ask, and try to answer them. Aka, doing hypothesis-driven scientific research.

        An obvious example: because we can, we are doing genomic studies up the wazoo. Across tissues, targets, species, diseases… you name it. We seek to find statistical correlations between genes and diseases for which we already have 1000 candidate genes correlated to a disease. How is that helping us make better drugs, and help patients? It might be more helpful if we did proteomics studies instead…. but these are more difficult, so we pass.

        Here’s another: stop running “re-purposing” clinical trials with minimal hypothesis and just because if it works the drug is ready for market.

        Where is scientific creativity when you need it?

      2. Anon says:

        First do phenotypic screening in human cell cultures, then in 3D tissue samples, then in whole organs (ex vivo), and then finally in humans.

        1. HTSguy says:

          Yes, that fits the flavor of the month. BUT … what “phenotypic screen in human cell culture” will tell you about prolonging CV health? Coming up with physiologically relevant endpoints for cell culture phenotypic assays is anything but trivial.

          1. Me says:

            Thanks guys! That about sums it up. I love the idea of a multi-omics approach with cells. Just not sure we’re there yet, and not sure if the world needs more assays with questionable translational potential.

          2. Cellbio says:

            For certain areas, primary cell culture pharmacology is tough to make relevant. Maybe CV is the poster child for “systems biology”, thinking about the old school pharmacologists and their tissue baths.

            ME, not sure we will know if we are ‘there yet’ until 5-10 years of a committed effort. Failure rates will not be an order of magnitude different, which is a problem in moving large organizations to something new. First hiccup and the naysayers jump back in.

  2. bhip says:

    Best I can tell, there isn’t a solid MOA for acetaminophen re: pain/fever,,,

  3. Marilyn Mann says:

    You might be interested in the cognitive outcomes of the HOPE-3 trial, presented at the American Heart Association conference, but not yet published, so far as I know.

  4. luysii says:

    Not to worry guys, an editorial and paper in the current Nature [ vol. 540 pp. 207 – 208, 230 – 235 ’16 ] says that all you have to do to decrease Abeta peptide in the occipital cortex of aged mice is expose them to light flickering at 40 Hertz for an hour or so. In more complicated work in the same paper, the authors found that implanting channelrhodopsin in interneurons of the hippocampus and then stimulating them at 40 Hertz by fiberoptics did the same thing to hippocampal Abeta plaques in a genetically abnormal mouse destined to have lots of plaques. The mechanism is thought to be activation of microglia which then phagocytose and destroy the plaques. No drug at all.

    For details please see A commenter on that post noted that it was quite likely that people were already trying rTMS and/or rACS as we speak based on this work.

    1. Lane Simonian says:

      Here is a better explanation for why the flickering light approach may be helpful.

      “Nitric oxide (NO) and its derivatives peroxynitrite and S-nitrosothiols inhibit mitochondrial respiration by various means, but the mechanisms and/or the reversibility of such inhibitions are not clear. We find that the NO-induced inhibition of respiration in isolated mitochondria due to inhibition of cytochrome oxidase is acutely reversible by light. Light also acutely reversed the inhibition of respiration within iNOS-expressing macrophages, and this reversal was partly due to light-induced breakdown of NO, and partly due to reversal of the NO-induced inhibition of cytochrome oxidase.”

      1. schinderhannes says:

        Should I get my ancient flickering computer screen back down form the attic?

  5. Lane Simonian says:

    Part of the effects of statins does involve inhibition of prenylation which limits g protein signalling. G protein signalling plays an important role in the development of Alzheimer’s disease although both the studies regarding the effects of statins through the inhibition of this pathway and on reducing the risk of Alzheimer’s disease have been contradictory.

    Statins can have a number of negative effects in part through the reduction in levels of the antioxidant coenzyme Q10. This may include transient global amnesia, type 2 diabetes, and mitochondrial dysfunction.

  6. Barry says:

    before everyone had to have an HMGCoA reductase inhibitor (statin) we explored blocking cholesterol biosynthesis at other points. I remember that blocking squalene oxidase caused squalene to back up in big fatty globules in hepatocytes. HMGCoA reductase was deemed to be the “first committed step” in cholesterol synthesis, and therefore the ideal place to intercede. But this ignored protein prenylation; HMGCoA reduction is upstream of more than just cholesterol

    1. Mol Biologist says:

      Ironically, you was much close to juvenile hormone than you think. Since you have been working with squalene. There is an alternative way of condensation for two molecules of farnesol “head to head” – leads to squalene (l = 6) that can undergo an oxidative cyclization to form cholesterol (l = 6) and other steroids.
      The oxidation of farnesol to farnesoic acid is a key step in insect juvenile hormone (JH) biosynthesis.
      I assume that blocking of farnesol oxidation will also lead to accumulation of farnesol.

  7. Sad Face says:

    What a mess! Should I take my statin tonight?

    1. Ted says:

      Clearly, the evidence supports your taking two.

      Especially if you forget…


  8. Sad Face says:

    I remember reading somewhere that the Therapeutic Index of statins in general is 1000, meaning that a thousand patients have to take the drug in order to prevent a single CV event. And I don’t know if these are patients with a previous CV event or not. I asked my doc once and he said to me: take it, it doesn’t hurt… Obviously not his liver…

    1. Barry says:

      that is certainly not what “therapeutic index” is usually understood to mean

    2. Handles says:

      Better late than never with this comment I suppose.

      The NNT for statins is much better than 1000 for those with known heart disease.
      For 5 years of statin treatment of patients with known heart disease:
      1 in 83 were helped (life saved)
      1 in 39 were helped (preventing non-fatal heart attack)
      1 in 125 were helped (preventing stroke)
      1 in 50 were harmed (develop diabetes)
      1 in 10 were harmed (muscle damage)

      These numbers (both those helped and harmed) are greater than for those treated with aspirin to prevent a second heart attack/stroke. If you do not have known heart disease, both statin therapy and daily aspirin therapy do more harm than good.
      For those without known heart disease, switching to a Mediterranean diet for 5 years will help 1 in 61 (avoiding a stroke, heart attack or death).

      All data and conclusions taken from here ( a great resource):

  9. Marcin Stasiak says:

    I have just seen a lecture stating that it is really LP(a) or lipoprotein a, and NOT CHOLESTEROL (good or bad) that is the predictor and culprit of atherosclerosis and stroke.

    But obviously cholesterol is as easy a biomarker as it comes…

  10. Sad Face says:

    Barry, you’re right. After a quick search on the net, this is called “number needed to treat” (NNT). Thanks for pointing this out.

    1. ignorant about NNT says:

      Sad face,

      Yes… after 25 years in the industry, I think NNT may be one of pharma’s best kept secrets.
      I am curious how many folks in this blog are aware of this parameter…

      Thanks to those who respond.

      1. LiqC says:

        But if you consider what a CV event would entail – hospitalization and death – one in a thousand events averted is well worth it, economically and otherwise. Vaccines NNT numbers are in the millions.

        1. Li Zhi says:

          “Worth it” for who? The 1 who benefits or the 999 who must allocate resources and suffer risks with zero benefit? Anyone believe that if vaccination required daily dosing, it would be feasible?

          1. Janex says:

            The problem is that no one knows which one of the 1000 patients is “the one”. It’s like buying fire insurance for your house. Say 1 house in 1000 will ever burn down. But if that is your house you lose everything you own. So the impact is severe. But there is a cost. for the 999 people whose how didn’t burn down they have to pay $$$$ to the insurance company that they could otherwise have spent on say remodeling the kitchen. But because the cost is so severe if you are “the one”, people spend that money on insurance. It is the same here, for most patients (not all) the costs of statins are far far lower than the cost of a cardiac event.

  11. JimM says:

    I haven’t had a chance to look at the 40Hz photic stimulation stuff beyond headlines and short paragraphs, but I presume an ultimate source of inspiration was the fact that Coherent 40-Hz oscillation characterizes dream state in humans.

  12. Palo says:

    Perhaps the most peculiar fact in the statins and fruit fly lifespan paper is this disclaimer:

    “Funding: The research was supported by Alva LLC, a private, basic research foundation. Alva has no web site. Its funder, a high net worth individual, prefers to remain anonymous. The funder has explicitly asked (several times) that nothing commercial be entangled in any way in the funds from ALVA LLC.”

  13. Mol Biologist says:

    Derek, your numerous readers should not to be confused with biological information which you posted here.
    IMO to understand more broad statin function you must go back to the fundamentals of evolutionary cooperation. Vitellinogen (VG) is a protein present in the blood, from which the substance of egg yolk is derived. VG belongs to a family of several lipid transport proteins. Vitellogenin and juvenile hormone (JH) only JH messing around with prenylation) work antagonistically in the honey bee to regulate the honey bees development and behavior. There is a interplay and suppression of one leads to high titers of the other.
    So, if for some reasons you have suppressed pathway which is exhibited by high level of LDL. You definitely may get a second chance by messing around with several players including statins to reanimate the dead.

    1. Barry says:

      what is the relevance of Juvenile Hormone (prenylated or unprenylated) to mammalian physiology?

    2. Me says:

      And chickens derive egg yolk from bee blood? Awesome! Presumably the yolk is transferred quantum mechanically at 40Hz oscillation. Or is it via ‘G-protein signalling’ (whatever that means).

      1. Mol Biologist says:

        Oh you forget to add Nitric oxide (NO) and its derivatives peroxynitrite and S-nitrosothiols and Longevity, Can’t be ignored ever.

        1. Me says:

          But you were there to remind me….the power of collaboration!

  14. Mol Biologist says:

    If you look at Vitellogenin there are at least few proteins related to mammalian physiology. My favorite is Von Willebrand or vWF have important role in predicting outcome in subjects with acute coronary syndromes (ACS), stroke, and perhaps atrial fibrillation.

    I assume it should be same for JH (prenylated or unprenylated) but need to be studied.

  15. Barry says:

    Is von Willebrand factor prenylated? If not, why should it be responsive to HMGCoA reductase inhibition?

  16. Mol Biologist says:

    VG product is classified as a glycolipoprotein, having properties of a sugar, fat and protein and can not be prenylated. There are several components of VG one of them vWF domain.
    So I assume to serve in mammalian physiology it was spited to several proteins.

    You may start looking for any proteins which can be prenylated. May be one of them is what you asking for?

  17. Barney Featherson says:

    Garlic-nuff said

  18. Dolph says:

    Reading these comments here it makes we asking myself once again how it is possible that the most ridiculous reservations, rumors and FUD always surround the absolutely best available therapeutic options. It’s always vaccinations, statins, amalgam fillings and so on. The best options we have. Crazy.

    1. statin skeptic and mercury amalgam filling hater says:

      Good point with respect to vaccine misinformation, but must disagree on both statins and amalgam fillings. The latter are really good at cracking teeth (and look awful to boot) and so are not the best option these days while the former provide little therapeutic benefit (see NNT comments above) at the cost of tremendous cumulative economic burden on the healthcare system along with significant risk of myopathy.

      1. Eric says:

        I must disagree with the statin comment. They’ve been shown to be quite effective and substantially reduce deaths and for most people there is no myopathy. Yes, some patients experience it, but they can discontinue the drug. Cases of severe rhabdomyolysis do occur but at very low rates which the FDA has judged to be acceptable relative to the benefits of the treatment. One statin (cerivastatin) was pulled off the market because it had higher rates which resulted in less favorable risk/benefit ratio. No drug (and no food for that matter) is perfectly safe and it’s up to the patient and physician to decide if the risk is worth it.

        I also disagree with the value of the quoted NNT from above. NNT is a useless measure unless you also specify the duration of the trial. In any rare event (stroke, heart attack, hip fracture) there will be very few incidences during a short trial. If you only studied heart attacks for one week you would need tens (hundreds?) of thousands of patients to even observe a handful of heart attacks. Even if the drug prevented all of them, the NNT would still be very, very large. Over the course of 20-30 years of elderly life the number of heart attacks goes up dramatically and the NNT will drop. It’s much like wearing a seatbelt. The odds of me getting in a crash on my way home from work today are exceptionally low. Over the course of my lifetime however, I’m more likely to be in a crash than not. Should I stop wearing a seat belt because it’s a rare event over the short term?

        1. Still a station skeptic says:

          Sorry, but I must disagree with your disagreement:

          1. Eric says:

            This paper is an odd choice to refute my point. From the conclusion of the paper:

            “We believe that statins should be prescribed according to the prevailing guidelines. Statins are usually inexpensive and safe, at least in a clinical trial setting,[20] and the benefit in terms of mortality or non-fatal cardiovascular outcomes cannot reasonably be challenged.”

  19. Lane Simonian says:

    Statins may have a generally positive effect by inhibiting oxidation (in part perhaps by inhibiting prenylation).

    “Treatment with statins lowers plasma nitrite and CRP [c-reactive protein] levels in patients with PAD [peripheral artery disease]. Our data support the effects of statins in vivo that have been demonstrated on the endothelium ex vivo, suggesting a beneficial effect by acting on the initial processes that trigger the disease, reducing oxidative stress (increase in the bioavailability of nitric oxide as peroxynitrite levels decrease) and curtailing the inflammatory processes which perpetuate the disease.”

    But for Alzheimer’s disease how does one explain the following?:

    Statins cause intracellular accumulation of amyloid precursor protein, beta-secretase-cleaved fragments, and amyloid beta-peptide via an isoprenoid-dependent mechanism.

    “As reported previously, low cellular cholesterol levels favored the alpha-secretase pathway and decreased A beta secretion presumably within the endocytic pathway. In contrast, low isoprenoid levels resulted in the accumulation of APP, amyloidogenic fragments, and A beta likely within biosynthetic compartments.”

    The prenylation of flavonoids may have positive effects; whether this is offset by the negative effects of prenylation in Alzheimer’s disease is another matter:

    “Therefore, S. flavescens and its prenylated flavonoids, possessing low molecular weights and lipophilic moieties may be potent preventive and therapeutic candidates for Alzheimer’s disease.”

    Another double edged sword for statins: they (like metformin and possibly bisphosphonate drugs) activate the AMP-activated protein kinase which may under conditions of low oxidative stress by productive but not under conditions of high oxidative stress.

    “Evidently, AMPK signaling can repress and delay the appearance of AD pathology but later on, with increasing neuronal stress, it can trigger detrimental effects that augment AD pathogenesis.”

    This is especially a problem if you are lowering levels of coenzyme Q10 with statin use.

    The memory problems sometimes associated with statin use have resolved themselves once the statin use discontinued. But as far as a strategy for reducing the risk for Alzheimer’s disease there are far better alternatives (a Mediterranean diet for instance).

  20. Forever a statin skeptic says:

    @ Eric

    Conclusions Statin treatment results in a surprisingly small average gain in overall survival within the trials’ running time. For patients whose life expectancy is limited or who have adverse effects of treatment, withholding statin therapy should be considered.

    My argument is that the small benefit demonstrated for statin therapy is simply not worth the collective economic cost incurred. Add to the unfavorable total exomomics the under-appreciated, under-reported side effects and these drugs are best left in the dust bin of history in my view. Baby aspirin instead!

    1. steve says:

      The side effects of statins are actually less than aspirin, which can cause GI bleeds and strokes. The cost of statins is minor now that they are off-patent. A recent Lancet systematic review concludes that the side-effects of statins are overblown and that no one should avoid them because of these reports unless they fall into the tiny percent of patients that can’t tolerate them.

      1. still a statin skeptic says:

        As Upton Sinclair said so very well, “It is dissicult to get a man to understand something when his salary depends on his not understanding it.” I question the incentives behind that Lancet analysis, truly. And see this for a thoughtful response:

        I’ve personally known multiple people who couldn’t tolerate extended statin therapy particularly at high doses.

        1. steve says:

          Well, knowing a few people certainly clinches the case against statins.

          1. Me says:

            Oooooh y’bitch!

Comments are closed.