Autism and Genetics
by Chloe Silverman & Martha Herbert

Autism, a devastating childhood neurodevelopmental disorder, was first characterized by Leo Kanner in 1943. Kanner described eleven children with odd and disturbed language, poor human connectedness, and repetitive, disturbed behavior.

Hitherto considered “genetic,” autism is now being diagnosed in unprecedented numbers. Yet while parents, clinicians, schools, and a growing number of researchers confirm the flood of autistic children, this biological epidemic seems to have touched off not an epidemic of administrative or public concern, but a near-pathological denial of both the fact of increasing rates and the role of extra-genetic factors implicated in the upsurge. This denial threatens to precipitate an educational, financial, social and human disaster in coming years.

Epidemic Denial

Diagnoses of autism and pervasive developmental disorders (PDDs) or autism spectrum disorders (ASDs) have increased since they were first observed, but the increase has accelerated over the last decade. Where rates were once 3 in 10,000, they are now, depending where you look, 1 in 500, 1 in 150, or even more — between a three- and a ten-fold increase. A recent state-funded study conducted by the M.I.N.D. institute in Davis, California confirmed a 273 percent increase in autism from 1987 through 1998, originally reported in a March 1999 study conducted by the California Department of Developmental Services, and argued that it could not be blamed on more aggressive diagnosis, improved ascertainment, or immigration. This increase is likely to be nationwide. Although no other state has kept comparable statistics, the US Department of Education has recorded a nationwide average increase of 544 percent in autistic students from 1992-93 to 2000-01, and comparable rates have been found in a number of local studies. A CDC study released at the end of December showed a tenfold increase over the last decade.

There are several ways in which these numbers considerably underestimate the magnitude of impact. For one thing, autism affects boys three to four times as often as girls; this means that if the overall rate is, say, 1 in 250, it will be closer to 1 in 100 in boys. For another, the figures in California count only the most severe cases; when mildly but still significantly impaired children are included, the figures can be considerably higher. And many people consider other disorders, such as Attention Deficit Hyperactivity Disorder (ADHD) and many learning disabilities, to be on a spectrum with autism. Considered in its broadest terms, this epidemic currently may affect from a few million to as many as 20 percent of U.S. children.

Such drastic increases imply the influence of non-inheritable or environmental factors, since, of course, there is no such thing as a “genetic epidemic.” But research continues to focus almost exclusively on studies of brains, screening and genes, as well as on denying the increase or disproving the role of controversial environmental triggers, notably vaccines (see sidebar).

How Autism Became a Genetic Disease

What does it mean to describe a condition as genetic? Every disease, including viral and infectious ones, involves vulnerabilities or resistances that relate in some way to genetic influences. But for many autism researchers, genetics is not about modulation or vulnerability — it is about “cause” and “determination.” These researchers justify their hunt for autism genes by pointing to studies of twins: while anywhere from 36 to 96 percent of both identical twins have autistic features, this is true for only 0 to 33 percent of fraternal twins. The very spread of these figures shows that claims of full genetic causation are flimsy; and these same figures can equally be used to argue that environmental factors must play a role, since the concordance for identical twins is not 100%.

Nevertheless this approach to autism as “genetic” has secured upwards of $60 million in research funding. To date there have been as many as eight genome scans and dozens of genetic studies. As in so many other gene hunts, at least a few candidate regions have been located on nearly every chromosome, but they have not led to specific genes and generally have not been replicated by later investigations. Meanwhile, other research approaches languish.

Autism as Biological: Paradigms Break Down

Many established autism researchers justifiably pride themselves on overthrowing the invidious “refrigerator mother” theory, which held that autism was a behavioral response of children to mothers who failed to display affection. Findings in the 1980s of abnormalities in autistic brains freed parents of this blame and shame, and opened the way to treating autism as a biological disorder. These brain abnormalities appeared to be of prenatal origin, and this seemed to fit with the evidence for genetic causes and the lifelong, apparently incurable impairment of people with autism. Researchers concluded that autism was determined by genes, was hard-wired before birth, and was treatable only by behavior modification. This has set the current research agenda, which is dominated by genetics, neuroscience and psychology.

But this picture is now unraveling. New evidence is emerging from both within and outside the dominant research domains that makes autism look more like an environmentally mediated illness. Many autistic children turn out to develop abnormally large brains — and do so after birth, in the first 2-3 years of life. Recent work suggests that other brain changes, previously thought to be prenatal, could occur postnatally. It also turns out that autistic children have substantial illness not only in their brains but in their bodies. While the mainstream research and clinical community considers physical symptoms to be “incidental” to the core autism, and pays little attention to them, subgroups of autistic children have common patterns of significant biomedical illness — notably immune system disturbances, disturbances in various biochemical pathways (including impaired detoxification, which may explain increased vulnerability to toxic exposure), and painful gastrointestinal disease. When treated biomedically, many autistic children have shown substantial behavioral improvement and improved receptiveness, suggesting that their behaviors aren’t totally “wired in.”

From the mainstream perspective, which defines autism in terms of unchangeable, genetically determined brain damage, it is incomprehensible that nutritional or metabolic interventions could have any effect. Therefore, such approaches are dismissed, usually without investigation, as “quackery.” But for those who see autism as an environmentally mediated illness, it makes perfect sense that the body as well as the brain should be affected. After all, why would toxins affect us only from the neck up? A growing social movement of parents and doctors who take environmental influences seriously feel that the genetic approach has betrayed autistic children by assuming that biomedical treatment can’t change anything. They argue that they look at their physically ill children and “believe what they see,” while the genetic determinists “see what they believe.”

Why the Vehemence?

The mainstream’s dismissal of new approaches to autism reveals a double standard about evidence. While parents and practitioners offering these methods are unable on their own to bear the staggering cost of double-blind controlled studies, neither have the accepted pharmacological and behavioral approaches been tested in this rigorous fashion. In truth, nobody in the field of autism treatment has much basis for invoking the legitimizing mantra of “Evidence-Based Medicine.” But the deck is stacked. In October, the FDA seized supplies of the amino acid taurine from Kirkman Labs, a nutrition company catering to autistic children. Their claim, that Kirkman Labs was making unsupported claims for this substance, came ironically within days of news stories exposing the agency’s lowest-ever rate of taking action against misleading advertisements by pharmaceutical companies.

Because the environmental triggers underlying the autism epidemic are presumably injuring not just the brain but the rest of the body as well, we need to test and refine biomedical approaches, rather than just aiming for new psychopharmacological drugs, behavioral treatments and gene identification. But advances in biomedical treatments will only occur once we move beyond a strictly gene-brain paradigm and allocate serious funding to physiological and toxicological autism research. Unfortunately, pharmaceutical companies on their own are unlikely to investigate biomedical approaches to autism, since many of the most promising nutritional interventions have little potential for patentability and therefore profitability.

Understanding the source of these twin denials — the denial of increasing incidence and the denial of non-genetic biological and environmental factors — is key both to addressing autism and to understanding the ideological role of genes and genetics research in contemporary America.

What if We Looked at Environmental Causes?

At a conference in October 2002 at Rutgers University, New Brunswick, entitled “Autism: Genes and the Environment,” leading researchers of the genetic persuasion shared the stage with toxicologists in what seemed to herald the cautious beginnings of a new synthesis. But we do not yet see a concerted shift to a research model that incorporates genes and environment.

There are probably several reasons for this. Certainly there are significant economic forces that stand to gain from the current direction of research. The notion that identifying genes and brain circuitry will lead to targeted drug development not only benefits the pharmaceutical industry, but also leads researchers down pathways that feel familiar and rational. In the belief system of many researchers, genes loom powerful, while environmental factors seem trivial and secondary — so that big effects just “naturally” are presumed to be “genetic.” This new line of investigation calls for a knowledge of toxicology and biochemistry much more detailed than most current autism researchers possess.

Moreover, examining environmental causes opens up a Pandora’s box of vulnerabilities. Some are personal: it is hard to think about toxic effects in autism without also wondering about one’s own health and the health of one’s family. That human actions, rather than genes, might be responsible for compromising the health of a significant proportion of a whole generation is so painful as to be, for many, unthinkable. And if there are environmental causes, then there may be liability and corporate accountability. If our carelessness with chemicals can harm children so profoundly, we may be called upon to fundamentally restructure the way we make things, the precautions we take, and the way we live – something industry and regulatory agencies go to great lengths to prevent.

Social Costs

Meanwhile, the determination of many researchers, regulators, legislators, funders, and even some parent groups to explain away the increased incidence of autism will have grave social effects. School districts are feeling the burden of increased numbers of autistic children. These children are often unable to function within a mainstream classroom, whether because of violent or self-injurious behaviors, lack of toilet training, or inability to communicate their needs. The cost of providing them with individualized behavioral therapy, which requires up to forty hours a week for maximum effectiveness, can run from $30,000 to $60,000 per year, per child. As a result, already underfunded public school districts will do almost anything to avoid providing these services, and parents without ample financial means are left with few options for their children, since Medicaid will not cover the optimum amount of behavioral therapy. Community treatment centers are faced with too many disabled children to accommodate, and are increasingly forced to turn away all but the most severe cases — although milder forms of pervasive developmental disorders, or high- functioning forms of autism such as Asperger’s syndrome, are also potentially devastating conditions.

Furthermore, full-spectrum autism, even if treated early and intensively, continues to have a poor prognosis. As this generation of children ages and their parents are no longer able to provide full-time care for them, residential institutions will be unable to provide facilities for even a fraction of these autistic adults. Estimates of the lifetime costs of care for a child diagnosed with autism today range widely — from conservative predictions of $2 million, to published figures as high as $12.4 million, depending on the extent of therapies, care, and support services figured into the equation. Over the next decade, the autism epidemic is likely to cost the U.S. economy hundreds of millions of dollars.

Conclusion

A children’s epidemic by all rights ought to open floodgates of concern and set off an urgent search for creative responses. While discoveries that pathological events may be occurring after birth should provoke a search for environmental triggers, it also opens up hope for prevention and for treatment. Moreover, the improvements experienced by growing numbers of children from biomedical interventions should fuel an intensive study of the mechanisms by which these treatments work, and a search for more finely targeted approaches.

Instead, we are treated all too often to methodological quibbles about epidemiological research; patronizing remarks by researchers about “hysterical parents” who “can’t accept their child’s genetic fate;” highly publicized, but methodologically weak, studies that claim to definitively refute any role for various vaccines in the increased rates of autism but raise no alarms about the increased rates themselves — and a press blackout on subsequent critiques and refutations. And we still get press conferences triumphantly announcing that “scientists are closing in” on the genetics of autism. The response of genetics researchers to the failure of their efforts so far is simply to increase the number of genes they expect to find — it’s now up to twenty or more, where it used to be as few as three or four -– rather than attempting, for instance, to discern increased vulnerability in small subgroups. What we don’t get is a frank admission that a purely genetic model is inadequate.

To cling to a genetic explanation for autism, to insist that the epidemic is a consequence of methodological rather than toxicological effects, is thus a desperate attempt to maintain the illusion that one lives in a comfortable and rational world where all is basically well, new chemicals and technologies always mean progress, experts are always objective and thorough, and authorities can be trusted. This form of genetic reductionism is a roadblock to developing the forceful science and social policy called for by the epidemic, it sustains taboos within the scientific community against potentially controversial ideas about environmental factors, and it distracts governments from addressing the financial and social demands that this epidemic creates. In short, it weakens our response to a disaster that has already begun.

***

SIDEBAR: Thimerosal and Autism: An Emerging Scandal

by Sallie Bernard

The Homeland Security Act, signed into law by President Bush on November 25, contained a provision shielding a small group of pharmaceutical companies from liability for harm to children caused by thimerosal, a mercury-based vaccine preservative. The rider was inserted just prior to the full House vote, and few in Congress were aware of it.

The liability shield has nothing to do with homeland security. Thimerosal is not found in vaccines for smallpox or any other possible bioterror agent. Until recently, however, it was used in most routine infant vaccines — and many parents, along with a growing number of scientists, are linking it to childhood autism and related disorders.

The rider effectively ends hundreds of lawsuits filed, or about to be filed, by parents of autistic children. The main beneficiary is Eli Lilly & Company, the developer of thimerosal. From 1990 to 2002, Lilly made political contributions amounting to almost $6 million — nearly a quarter of which was spent on November’s mid-term elections — and has substantial ties to the Bush Administration. Mitchell Daniels, the White House budget director, is a former Eli Lilly executive. The company’s current CEO, Sidney Taurel, is on the Homeland Security Advisory Council. George Bush Sr. was a member of Eli Lilly’s Board of Directors during the 1970’s.

The White House denies responsibility for the rider, but Representative Dan Burton (R-Indiana), who has a grandson with autism and has held hearings on autism and vaccines, lays the blame at their door. Bill Frist, the new Senate majority leader, has been linked to the provision.

Adding to the intrigue, the Department of Justice asked the special vaccine injury court to seal the records relating to more than one thousand autism-thimerosal cases now pending. This action would have effectively blocked public scrutiny of any vaccine manufacturer’s documents to emerge from the proceedings. Although the record-sealing was eventually rescinded, parents of autistic children were left wondering exactly what in those documents needed to be hidden — and, ultimately, why Eli Lilly and the other manufacturers are so scared.

Apparently, there is plenty to worry about. Although autism was once thought to be a purely genetic disorder, the steep rise in number of cases during the last decade shows that environmental causes are also at work. The huge increase in autism rates coincides with the addition, in the early 1990s, of two new thimerosal vaccines to the infant immunization schedule. In 1999, the FDA disclosed that the amount of mercury in vaccines exceeded EPA safety guidelines. Manufacturers were asked to remove thimerosal, although existing stocks were left on the shelves. Parent groups like Safe Minds demonstrated that the symptoms of mercury poisoning matched the abnormalities they saw in their children. Scientists are now showing that vaccine-levels of thimerosal can cause neuronal apoptosis, immune imbalances, and autistic-like brain lesions and behaviors.

Documents obtained by lawyers show that Lilly and others knew about the dangers of thimerosal as far back as the 1940s. In fact, as the FDA has admitted, safety testing of thimerosal has never been conducted. A recent attempt by vaccine researchers to absolve thimerosal of toxicity through a poorly designed — but well promoted — study that used only 33 infants (Pichichero et al., The Lancet, November 2002) was derided by parent advocates as inadequate and overreaching.

It is not surprising that Lilly and the vaccine industry would fight so hard to dismiss concerns over thimerosal and, failing that, to have the concerns forcibly dismissed by law. Not only are lawsuits a threat, but the presence of an untested toxin in infant vaccines raises the ugly question of whether vaccines are being properly evaluated before being unleashed on the public. A lot is at stake. Vaccines are considered one of the few bright spots for pharmaceuticals in the future. Hundreds of vaccines are in various stages of development. Revenues are expected to reach into the hundreds of billions of dollars.

Parents of autistic children are making substantial inroads, but the pharmaceutical giants are fighting back with a vengeance. Truth, however, has a way of becoming undeniable.

***

Martha Herbert, MD, PhD, is a pediatric neurologist and brain development researcher at Massachusetts General Hospital. Chloe Silverman is a graduate student in History and Sociology of Science at the University of Pennsylvania. Sally Bernard is Executive Director of Safe Minds.