What's new in autism research

HS director RFK, Jr. has said he wants science to find the 'root cause' of autism spectrum disorder, or ASD, by next March. Is this even possible? Here I will give some background and a brief review of what we know so far.

Are rates of autism increasing?

Autism diagnoses in the US have increased enormously in recent years. According to the CDC, one in 36 children aged 8 now have ASD. The possibility that the increase is an artifact of increased diagnosis can't be dismissed, as we've seen similar changes in ADHD, depression, and other neuropsychiatric disorders that follow the same pattern. The idea is that the increases are due to the medicalization of behavior, where normal behavior by boys is treated as a disease by teachers and school bureaucrats, who have misguided and stereotyped views of male behavior.

If the increases are real, there must be an environmental cause. To an epidemiologist, 'environmental' means something not inherited, which could be anything from a pollutant to diet to social changes, and maybe even, with only a slight stretch, a virus (or a vaccine, though this is highly controversial). The challenge here is that the prevailing theory is that ASD is caused by changes in the brain structure, which could be caused by any number of things alone or in combination.

A related problem is that ASD is diagnosed subjectively by observing behavior.^[1] This is very unreliable, so any attempt to identify its causes or even to know for sure whether the rates are increasing would have to wait for a convincing biomarker. Finding one could take decades.

Is autism a single disorder?

The answer seems to be no. Many genetic disorders, such as phenylketonuria, FOXP1 syndrome, and impaired function of ABCA12, ^[2] are correlated or ‘associated’ with autistic symptoms. A multitude of DNA mutations have been claimed for ASD, but so far there is no clear picture.

Studies of twins have been important in clarifying whether there's a genetic component. In general, more severe autism is associated with greater familial ‘loading.’^[3] There are debates in these studies about whether the patients were selected randomly enough, but it's clear that if autism were a single disorder, we would not see such wide differences in severity. That brings up the dreaded term “multifactorial”, which is regarded in science as either an admission of defeat or an impossible research project: if there are 100,000 genes, and some combination of half of them is necessary for the disease while some other combination protects, even a GWAS (genome-wide association study) won't answer the question.

Even so, genetic studies have discovered rare genetic defects in 20% of ASD patients. There are perhaps dozens of other genetic changes, mainly deletions, rearrangements, and duplications, for instance in 1q21.1, which is part of chromosome 1, that have been claimed for ASD. Other chromosomal changes are on 15q11.2, 16p11.2, and 22q11.2. It's easy to find genes in or near these areas of a chromosome that have something to do with the brain, but as with schizophrenia, there are too many gene changes to convince anybody. How many are real, and what do these genes do, nobody really knows.

The most robust finding seems to be that there is a structural difference in the brain in ASD. The general idea is that synaptic and neuronal pruning, which occur before and shortly after birth, is deficient in ASD, leading to excess numbers of neurons. Consistent with this, many structural MRI studies have found that ASD patients have more gray matter but less white matter in the dorso­lateral and medial frontal cortex and medial and lateral temporal cortex.^[4] Gray matter is neurons; white matter is myelinated (insulated) connections that carry signals from one brain region to another. So, the idea is that different parts of the brain are not signaling to each other or doing it incorrectly. Some studies showed that the total brain volume is also increased.

There are some MRI findings that suggest that the amygdala, which determines emotional valence, and the hippocampus, which processes spatial memory, are bigger in ASD. Some researchers think an enlarged amygdala might explain greater levels of anxiety.

The immune system

You might think having more brain cells would be desirable, but in ASD it suggests that there could be some problem with pruning. Synaptic pruning gets rid of weak or improperly connected synapses. It is done by the microglia, which are cells that make up the innate immune system in the brain. The most important factor here is TREM2, a protein that regulates microglia—and also neuro­inflam­mation. Pruning is a complicated process about which little is known, and one that would be very risky to treat.

There's also something called neuronal guidance signaling, which is important for making the correct wiring in the brain.^[5] There's a lot of research trying to find out how it occurs and what molecules are involved. If guidance signaling went wrong, it could cause problems in dealing with environmental stimuli. It might even cause pruning problems. But how it goes wrong and what triggers it are unknown.

A Big Clue is that schizophrenia, which typically starts in the late teens or early twenties, is now recognized as a neuro­develop­mental disorder. Here again, there is strong evidence that the immune system plays a role: some autoimmune disorders produce symptoms identical to those in schizo­phrenia, even leading to catatonia indis­tinguish­able from that seen in schizophrenia patients. And indeed, some patients who were diagnosed with schizophrenia were found to have an autoimmune disorder instead.

The possible role of the immune system might explain the repeated suggestion that vaccine scheduling in early life might play some role. But many doctors dismiss this idea because of the fact that humans have two immune systems: vaccines affect the adaptive immune system, which produces antibodies; and synaptic pruning is done, as far as anyone knows, exclusively by the innate immune system—which does not. They also point out that many attempts to replicate one influential study failed. This, of course, doesn't rule out the possibility of a more complicated factor—or even ideologically driven bias, which as we saw happened extensively in Covid.

And indeed, there's ample research that autoimmune disorders, which are mediated by the antibodies from the adaptive immune system, cause brain disorders that would have been classified in earlier times as mental illness. The best example is anti-NMDA receptor encephalitis. Cancer can also cause autoimmune disorders. But we need to be careful: the immune system is a hot topic in biology these days, so the answer might not be as simple as we'd like.

Toxicology studies are a mess

The problem we face in toxicology is the large body of low-quality science on the topic. One group in Korea, for instance, claimed that ASD was caused by microplastics.^[6] These researchers fed lab mice with a large dose of polyethylene broken down by a sonicator into small particles, then claimed to find microplastics in the brain at a level of 1.43 ± 0.16 mg/g. They then did standard behavioral studies on the mice and their offspring on the assumption that these tests could detect ASD in mice. And they did DNA analysis of the gut microbiome and claimed that it ‘resembled’ that of ASD patients.

There are so many problems with this that the study is practically useless. Just to name two: they didn't perfuse the mouse brains, which means the microplastics they observed were likely in the microvasculature, the tiny blood vessels throughout the brain. Even if they had tried, perfusion might not remove them. And secondly, there's no reason to assume that the tests, such as nest building and the Y-maze test, which is used to measure memory, measure anything relevant to ASD.

I don't mean to single out these authors; there's a huge volume of similar research in the toxicology literature, and the problems in relating mice to humans are well known.

Political and fashionable interference

Another fashionable approach is to study the microbiome,^[7] which is the collection of bacteria in the gut. Here the problem is separating cause and effect, as any behavioral difference is bound to be reflected in food preferences, which would change the microbiome.

Asperger's disorder, a supposed type of high-functioning ASD, is no longer considered real. The main reason is not some big scientific discovery but the discovery that Hans Asperger himself was somehow associated with Nazism. How this means that a disorder, which was once popular among computer programmers, is not real is anybody's guess.

If the rates of ASD are really increasing, finding the causes for ASD and other neurodevelopmental disorders is critical. Even discovering whether an environmental toxin is involved would be a breakthrough. But in most science journals, you can't even get a paper published in nine months. Getting preliminary data, writing a grant, and waiting for NIH to fund it takes at least an additional year. The chances that we'll solve this problem by next March, as RFK, Jr. hopes, are slim indeed.

may 24 2025, 9:30 am