A new study is making the rounds in the media, with unfortunate but typical misrepresentation. This is partly due to media hype, but I think some of the responsibility goes back to the study authors. The title is: “Propionic Acid Induces Gliosis and Neuro-inflammation through Modulation of PTEN/AKT Pathway in Autism Spectrum Disorder”. I cannot figure out, from reading the study, why those last four words are in the title. Sure, that is the speculation that derives from the data, but the data has nothing directly to do with autism, as we will see. The effect, however, was dramatic. Here are some headlines:

Forbes: “Spike In Autism May Be Linked To Preservative In Processed Foods, Study Suggests”

Medical News Today: “Could processed foods explain why autism is on the rise?”

UPI: “Processed foods may lead to autism, study says”

Business Standard: “Researchers draw link between processed foods and rise in autism”

None of those headlines are fair or accurate. After reading those headlines you may be surprised to learn that the study did not involve children with autism, the risk of autism, or really anything to do with autism. The study, rather, looked at the effects of propionic acid (PPA) on neurons (brain cells) cultured in a dish:

We investigated the effect of Propionic acid (PPA), a short-chain fatty acid (SCFA) and a product of dys-biotic ASD gut, on human neural stem cells (hNSCs) proliferation, differentiation and inflammation.

They found that when the cells were exposed to high levels of PPA this promoted neural stem cell growth, increased the ratio of glial cells to neurons, decreased neurite outgrowth, and increased neural inflammation. These are also features seen in the brains of children with autism. Of course, this is one study and it needs to be replicated to see if this effect is robust, but let’s assume for now that the methods and results are valid. The real question is – how to interpret the results.

As we often point out as SBM, it is a long way from an in-vitro study to a clinical implication, and this applies for either treatment effects or risks. A dish of cultured cells is not an organism. Organisms are complex dynamic homeostatic living things. You cannot extrapolate directly from the petri dish to the animal. At best these in-vitro studies suggest a possibility that would then need to be confirmed in animals. The authors do acknowledge this, and the need for follow up animal study to show if the possible effect they are seeing is relevant in a whole live animal.

But even then we are not done, because rats and mice are not people. We would then need to do epidemiological studies in people, for example studying whether the diet of pregnant women, specifically relating to the amount of PPA, affect the risk of autism in their children. We cannot expose pregnant women to PPA for ethical reasons, but you can also do an experimental study in which you randomize pregnant women, or women who are trying to become pregnant, to a diet low in PPA. Only then, really, will we know if there is a “link” between PPA in the diet and risk of autism. That won’t end the scientific questioning either, as we need to answer many subquestions about dose, other confounding factors, vulnerable populations, and mechanism of action. At best the current study shows a potential mechanism of action IF an actual link is discovered – but it does not establish a link.

There are many ways in which these in-vitro finding will not translate to actual risk in humans. First, PPA is normally found in the gut of humans. So, does the amount of extra PPA consumed in processed foods significantly affect the amount of PPA in the GI system? Does this translate into changes in blood levels, and the levels that cross the placenta, and therefore concentrations in the developing fetus? Are those differences, if any, sufficient to cause biological changes? If any link in this chain is broken, the effects seen in-vitro will not translate to people. Most of the time, the vast majority, effects seen in-vitro do not directly translate to people.

Of course, most people will read the headlines, and not the study, and would not appreciate the distinctions made above in any case.

As an aside, the reporting accepts without challenge the premise that autism rates are increasing. As we have discussed many times, the rise in autism diagnoses is at least partly due to increased surveillance and recognition. We are diagnosing it more, but that does not mean the true incidence is increasing. It’s hard to know exactly, but if we look at all the evidence it seems likely that true rates are stable. There is definitely an increase in surveillance, diagnostic substitution, and a broadening of the definition. If we compare different age cohorts, however, using the same diagnostic criteria, the rates seem stable. But I will grant this is still a bit of an open question, but at the very least much of the apparent rise is an artifact and not real.

Another aspect of the question addressed in this study is the broader question of the apparent link between the GI system and autism. In the last two decades there has been increased evidence for an actual link. For example, there does seem to be a subset of children with ASD who have gastrointestinal symptoms that are sensitive to diet. Some of these children have been found to have specific species of bacteria in their gut microbiome not seen in people without ASD. This does not establish the arrow of cause and effect, but it is intriguing and deserves further research.

That is basically where we are with the current study as well. This is interesting, it deserves further research, and it may fit in with the broader picture of a gut-brain connection in autism, which is also preliminary and deserves further research. In fact the broader issue of a gut-brain connection in general is interesting but needs further research. I am hopeful that in 20-30 years we will actually have interventions based on altering the gut microbiome for certain neurological diseases, but that is likely the time frame we are dealing with.

But for now we are in the preliminary scientific stage, where hype and misleading headlines rule, while firm scientific conclusions, and even less so, interventions, are not yet here.


Posted by Steven Novella

Founder and currently Executive Editor of Science-Based Medicine Steven Novella, MD is an academic clinical neurologist at the Yale University School of Medicine. He is also the host and producer of the popular weekly science podcast, The Skeptics’ Guide to the Universe, and the author of the NeuroLogicaBlog, a daily blog that covers news and issues in neuroscience, but also general science, scientific skepticism, philosophy of science, critical thinking, and the intersection of science with the media and society. Dr. Novella also has produced two courses with The Great Courses, and published a book on critical thinking - also called The Skeptics Guide to the Universe.