Premature Claims for Neurotrophic Factors

Scientific medicine is not easy. By this point we have largely picked the low hanging fruit, and continued improvements are mostly incremental and hard won. In order to get the most out of our limited research dollars, and optimize medical practice with the safest and most effective treatments, we need to use all available scientific evidence in the proper way. That is the essence of SBM.

There are those, however, that misuse or abuse the scientific evidence — whether to promote an ideology, out of innocent ignorance, or for nefarious purposes. In order to be truly science-based a medical intervention should be plausible, or at least not implausible, based upon basic science evidence, and it should actually be safe and effective when tested in people. Therefore, medical practices can fail to be scientific for one of two broad reasons: they can be scientifically implausible, or they can lack proper clinical evidence for safety and efficacy (or even have evidence for lack of efficacy). Some modalities (like homeopathy) fail on both counts.

The more pernicious medical claims are those that seem highly plausible, that can be extrapolated from basic science, but simply lack adequate clinical evidence. Stem cell clinics are an example — they can easily dazzle desperate patients with scientific descriptions of how stem cells work, and even cite published basic-science papers showing the potential of this technology. But what they cannot do is provide clinical evidence that the specific intervention they are offering is safe and effective for the specific disease or condition they are treating.

I recently came across another example of this type of failure — the promotion of neurotrophic factors for autism and other neurodevelopmental disorders. One center, the Neurorecovery Center Edelfo, specializes in this claim.

Neurotrophic factors are generally proteins that bind to cells and cause them to grow, proliferate, or just maintain their functionality. Some nervous system cells would die without a steady supply of the needed neurotrophic factors. The discovery of neurotrophic factors and their role in nervous system function was certainly very important to our understanding of neurological diseases. This knowledge has also led to a great deal of speculation about the potential of neurotrophic factors in treating neurological diseases.

After a couple decades of clinical research, however, the reality has not lived up to the promise. As is often the case, highly plausible ideas do not pan out in clinical research. Several neurotrophic factors, for example have been studied in diseases like ALS without any measurable clinical benefit. They are just another reminder that the body is complex, and even when we have a very compelling picture of why a treatment should work, that picture is almost certainly incomplete. We cannot know what the net effect will be of a treatment until we do rigorous clinical trials.

Another way to look at this is that the body has evolved complex mechanisms of self-regulation. It is not easy to push or pull on one piece of this complex system and predict the net effect. There are many reasons why a treatment that we predict should work, doesn’t.

Edelfo is doing what many similar clinics or supplement marketers do — making simplistic extrapolations from the basic science to clinical claims, without supporting those claims with clinical evidence. They cite basic science research to give the false impression they are science-based. For example, they write on their website:

BDNF has shown to have a neurotrophic effect in serotoninergic and dopaminergic neurons, but also in glutamatergic neurons of the hippocampus and the sensory regions like pain, touch and hearing. BDNF also contributes to the maturation of GABAergic neurons, so it can also be useful in bipolar disorders, schizophrenia, ADD, memory, neuronal deficits that affect audition, sensitivity, etc.
GDNF has shown a specific effect in dopaminergic neurons and secondly in motor neurons, so it can be very useful in ADD, where a hypo development of the dopaminergic circuit has been shown, although there are other types of ADD that are generated by alterations in other regions like the frontal lobe, basal ganglia, etc. GDNF can also be useful in Parkinson’s, where there is a severe loss of the dopaminergic neurons of the substantia nigra and in motor problems, where experimental studies performed in rodents with motor damage have shown that the GDNF favors the axonal growth of motor neurons and improves the functional recovery.

Edelfo uses a diagnostic method based upon EEG to then design a nutritional supplement for their specific customers. The claims above are an excellent example of the type of pseudoscientific deception I am describing.  There is indeed a great deal of interest in GDNF for Parkinson’s disease. Animals studies have shown promise, but human studies have been negative. The problem is probably due to getting the GDNF to the target tissue.

GDNF is a protein. If you eat a protein it will be broken down into amino acids and used as food — it will not get into the bloodstream unaltered. This is why protein-based drugs typically need to be given intravenously, to bypass the stomach. GDNF has even more difficulty, however, since it also needs to cross the blood brain barrier — a protective barrier around blood vessels in the brain evolved to specifically regulate what gets into the brain. To overcome this issue researchers have used one of two methods — they either use gene therapy to insert GDNF producing genes into brain cells, or they use a continuous infusion of GDNF into the cerebrospinal fluid (CSF).

There are rather extreme methods of introducing a drug to the body, but they are necessary. And even still, researchers speculate that the CSF infusions used in human trials were not sufficient to get GDNF to the cells in the brain that are the target of treatment.

It is with these types of issues that pseudoscientific treatments often fail — bioavailability and pharmacokinetics. Does the substance get into the body, and what happens to it once it does? Basing claims for a nutritional supplement on the activity of BDNF is highly deceptive. It certainly can sound compelling to a non-expert, but the claims utterly fail when closely examined. The same, by the way, can be said for BDNF, which is also a protein.

Edelfo’s use of EEG (electroencephalogram), visual and auditory evoked potentials (VEP, AEP) is also questionable. These modalities are all legitimate methods of research and diagnosis, but they need to be specifically validated for each specific clinical use. They are therefore ripe for exploitation — they can be used inappropriately to give the impression of high-tech diagnostic procedure without the evidence to support the specific application.

The published research in this area is complex and mixed. In autism, for example, studies do show that there is a statistical difference in AEPs between autistic and typical children. This does not mean, however, that the test has specificity or sensitivity in the diagnosis of an individual. This is a common misuse of scientific evidence: using group comparisons and applying them to the individual. A different kind of study is necessary to see what predictive value the test has when applied to an individual. There are therefore many diagnostic modalities that are useful for research, but are not used clinically for this reason.

To further bolster their claims Edelfo does publish a few in-house studies. The quality of these studies is so low, however, that they are useless for anything but marketing. In a study of another neurotrophic factor, FGF2, in autism, the study was published in Defeat Autism Now (DAN), which is a dubious site dedicated to biological treatments for autism and notorious for making pseudoscientific claims. That aside, the study itself is fatally flawed in numerous ways, including using very low subject numbers and not describing any blinding. These types of studies are almost guaranteed to generate false positive results — again, useful for marketing, but not doing good science.


The public needs to be aware of the types of misleading claims that are made by clinics, centers, and marketers of health products that can seem very scientific when they are not. The use of pre-clinical evidence to make clinical claims, the use of research methods for clinical diagnosis, and the use of supplements without any consideration of bioavailability are chief among them. Patients and their families are cautioned to be skeptical of any clinic or center that claims to offer a unique service. Further, any clinical claims being made should be backed by well-designed clinical trials of the specific claims published in respected peer-reviewed journals.

Anything less than this is likely to be selling false hope, not a legitimate treatment.

Posted in: Herbs & Supplements, Neuroscience/Mental Health, Science and Medicine

Leave a Comment (9) ↓

9 thoughts on “Premature Claims for Neurotrophic Factors

  1. Ed Whitney says:

    The October 10 issue of Archives of Internal Medicine (Prasad et al, Arch Intern Med 2011;171:1675-6) had a very short research letter on medical reversal, defined as occurring when a methodologically sound study goes against current clinical practice. The authors looked at 124 articles in the New England Journal from 2009 which made some conclusion regarding a medical practice. Of these 124, 16 articles, or about one in eight, published a conclusion that contradicts current practice. Some of the practices were medical, some were invasive, and some were screening tests.

    The interesting thing (which deserves much greater discussion) was the column labeled “Why We Got It Wrong Initially.” The most common reason was “confidence that the pathophysiologic concepts underlying the practice were rational.”

    A worthy research project would be taking a larger sample of published articles from a larger sample of journals, and discussing in greater detail why the earlier practices were supported and where their conclusions went too far or went astray altogether. The relevance of this research letter to the current thread is clear. The full text is behind a paywall, but maybe as this thread develops, more detail can be posted.

  2. “Confidence that the pathophysiologic concepts underlying the practice were rational.” Love that. Thanks, Ed.

    In the manual therapy world, I sure hear a lot of highly overconfident speculating about the clinical relevance of scraps of physiology. That was the inspiration for my article criticizing the alleged importance of fascia. Too much is made of much too little, and the same is true of a long list of “tissues of interest” that are supposedly worth trying to therapize.

  3. I always seem to forget about the bioavailability issue, which is odd, because it seems like an extremely important one, eclipsing many other concerns. I must make a point of making that point more often.

  4. Quill says:

    Premature claims never seem to rule out mature profit-making enterprises using them to boost their ledgers regardless of any actual demonstrated human efficacy.

    Intellectually, I don’t mind reading highly plausible extrapolations and claims but become immediately suspicious when the group publishing such things just happens to have something for sale resting (however shakily) on them. (That’s one of the ways SBM trumps most of its alleged opponents: knowledge is usually freely given, is expected to be rationally considered, tested and verified, and thus discussion and understanding moved forward.)

  5. Angora Rabbit says:

    Oh, yes. Parents definitely want to dose their autistic children willy-nilly with neurotrophins given this article in this week’s JAMA:

    “Neuron Number and Size in Prefrontal Cortex of Children with Autism” by Courchesne et al, JAMA (2011) 306(18) 2001-2010.

    Not. Definitely not.

  6. daedalus2u says:

    I agree with Dr Novella. neurotrophic factors are not a panacea. I will comment more later.

    I am going to the Society for Neuroscience meeting in DC and have a poster that bears directly on this topic (and also involved nitric oxide).

    I will be presenting at 3 to 4 on Saturday November 11.

    Poster 35.03/C6

    I hope to see a lot of SBM regulars there.

  7. weing says:

    It was recently surprised by a report that nasally introduced insulin was found beneficial for Alzheimer’s.
    This may be a way of introducing proteins like GDNF to the target tissues. This is pure speculation on my part as I am not involved in research.

  8. weing says:

    It = I

  9. daedalus2u says:

    I had a chance to look more carefully at the Eldefo site and it is very misleading. The reasoning is very simplistic, so simplistic as to certainly be wrong.

    Neurotrophic factors are signaling molecules that signal cells to do things. The way that happens, the peptide activates a receptor which then activates a whole signaling cascade inside the cell. Many of these neurotrophic factors activate nitric oxide synthase and make a little bit of NO. That NO adds to the other NO in the vicinity and that new NO level reflects (in part) the physiological effect of the neurotrophic factor. Some of these neurotrophic factors are out at the ends of axons, and they release NO as they are carried back to the cell body. This NO, released all along the length of the axon, is likely to be an important factor in maintaining that specific axon. All axons need to control their environment such that they have sufficient nutritive blood supply to satisfy their physiologic needs. I think that the NO released this way is the mechanism by which this is done. A slightly higher NO level does trigger angiogenic factors that slightly increase local blood flow. Many (essentially all) blood flow control pathways use NO as a signaling molecule. In the brain, this is observed in the fMRI BOLD technique which monitors blood flow in the brain, real time with a sub-second and sub-0.1 mm resolution. There is not a separate mechanism to regulate nutritive blood flow.

    All neurodegenerative disorders are characterized by reduced blood flow. My hypothesis is that this reduced blood flow is secondary to a reduced basal NO level which shifts the operating point of the blood flow neuronal activation control system to place where there is insufficient nutritive blood flow and the cells slowly degenerate because physiology prioritizes short-term continued function of the CNS (staying conscious) over long term degeneration. If you are running from a bear, consciousness has to have higher priority than preventing neurodegeneration (that is what all that cognitive reserve is for). I go into this in excruciating detail in my poster.

    I suspect this is the mechanism behind some of the seemingly beneficial effects in some animal models. The animal models for neurodegenerative disorders are not very good. For the most part they don’t mimic the “true” cause of these disorders in humans (which is unknown), so a positive result in an animal model may simply reflect artifacts in the animal model compared to the human disorder. Using neurotrophic factors to raise NO levels is a non-physiologic way to raise NO levels and is not a good way.

    A reason that nerves work this way is because if a nerve is not receiving signals, then it needs to be removed because it is taking up space, is wasting resources and might do something bad if it is allowed to stay there (like trigger a seizure at a bad time). For most of an animal’s lifetime, this is a good trade-off, a trade-off of improved stability for reduced function. It is only after all the cognitive reserve has been used up in this way that the trade-off becomes detrimental. Organisms don’t have the ability to change their evolved nervous system control paradigms when they become detrimental. Evolution balanced the sum of deaths and non-deaths (before reproduction) due to each evolved “feature”. You can’t change a billion years of evolution when those traits catch up with you.

    They are called neurotrophic factors because in cell culture they do seem to “attract” growing neurons up a gradient of the growth factor (from low to high). In vivo, these factors are made and released by single cells and so the signal they provide in the brain has single-cell resolution in time, space and concentration. The signaling by these compounds probably has automatic gain control, that is the cells “tune” their responses such that the whole neural network works “in sync”. This “tuning” is thought to be one of the mechanisms by which antidepressants work, the synapses “tune” themselves slightly differently in response to the presence of the drug, which is why they often take weeks to be effective; the “tuning” requires modifying the receptor number and affinity which can mean synthesis of new receptors (to a different setpoint) and then moving them out to the tippy end of the axons.

    There are two levels that are most important and they are going to be different for different cells. The first is the level that causes attraction of new connections, and the second is the level that prevents cell death. It is conceivable that adding enough neurotrophic factor to do the first will also program the cells in the vicinity to require more simply to stay alive.

    Any administration of these compounds will be non-physiologic. It might cause non-physiologic growth of neurons in the desired target, but it could also (simultaneously and unavoidably) cause cell death in other cells by modifying the levels needed by those cells to stay alive. This may occur even with perfect administration of exactly the right amount at exactly the right target cells. That perfect administration cannot be achieved in practice. It is wishful thinking to expect these things to do what the Eldefo clinic wants them to do (which seems to be different depending on what disorder they are trying to treat).

    These factors do many different things in many different cells and only a tiny fraction of those things are understood and none of them are understood well.

    Degenerative disorders are whole-tissue-compartment disorders. It is the whole brain or a brain region that is degenerating, not single cells. Degenerative disorders are very likely due to signaling problems between cells such that they prioritize continued function over non-degeneration.

Comments are closed.