Development of Diagnostic Tests and Autism

In a paper published in Molecular Psychiatry (July 2005), Philippi et al. from Integragen reported on a gene polymorphism coding for a kinase, which is associated with autism. In their conclusion, the authors suggest that dysregulation of this kinase’s activity might be involved in autism’s etiology. Integragen, building on these findings, announced on its website the development of a diagnostic test for autism. However, no scientific details or information are provided to support this test. The press has also picked up on this announcement.

Genetic tests hold great potential in that they can inform each individual about their predisposition to developing a given condition. However, performing genetic tests in human healthcare requires a prior, independent scientific evaluation, especially regarding the test’s analytical and clinical validity as well as its usefulness in diagnosing the associated pathology.

The announcement of a genetic diagnostic test for autism by Integragen through the press raises several scientific and ethical issues. Medically, autism is a neurodevelopmental disorder with an extremely heterogeneous clinical presentation. Autism is associated with known genetic diseases in 10–25% of cases, such as tuberous sclerosis, fragile X syndrome, Angelman syndrome, certain metabolic diseases, chromosomal abnormalities, or infections like congenital rubella. However, in most cases, autism's etiology remains unknown, making early diagnosis difficult.

Family aggregation and the high concordance rate among monozygotic twins compared to dizygotic twins demonstrate the strong role of genetic factors in autism. In fact, autism stands out as the psychiatric syndrome with the strongest genetic component, with heritability over 90% and a sibling recurrence risk estimated at 45%. Consequently, the potential to develop a diagnostic test for autism based on genetic characteristics has generated significant hope.

A diagnostic test for autism could enable early intervention and tailored therapeutic strategies. Numerous epidemiological genetic studies have excluded monogenic Mendelian transmission with full penetrance, instead suggesting the involvement of multiple interacting genes. Autism is further considered a genetically heterogeneous condition, where susceptibility genes can vary from individual to individual.

Systematic genome studies in samples from families with multiple cases have identified several chromosomal regions containing autism susceptibility genes. However, the lack of strongly linked regions in these genome screenings suggests the existence of multiple loci (perhaps ≥15) interacting with each other, none of which plays a major role in autism susceptibility. Therefore, many specialists believe that continued research in this area would better define the disease's nosography.

Integragen’s publication of a gene polymorphism on chromosome 16 associated with autism adds to our understanding of the disease's etiology but does not constitute a significant breakthrough in its diagnosis. Integragen’s announcement of an autism diagnostic test, without publishing the scientific basis of this test in specialized scientific journals, does not allow for an objective assessment of its validity and usefulness.

This announcement through the press highlights the need for an evaluation procedure for genetic tests before their commercialization. In the United States, the CDC created an independent scientific committee to verify the validity of genetic tests. With the rise of genetic information in various diseases, the creation of a similar body at the national or European level is essential. This body would be responsible for ensuring that predefined criteria have been met before using a genetic test in clinical practice or its commercialization.

Furthermore, the French National Ethics Committee (CCNE), in its opinion on the commercialization of self-tests for HIV screening and genetic disease diagnostics (No. 86), warned against “the simplification and sometimes media exaggeration of genetic discoveries, leading, for instance, to the notion of discovering ‘the gene for…’.” Most genetic diseases have complex, multifactorial components. The associations identified between a condition and genetic markers certainly help us better understand disease mechanisms and gene-environment interactions, but they have limited clinical application, even in diagnostics. This general observation applies particularly to autism, where families’ anxiety about having an autistic sibling must be supported by appropriate genetic counseling.

Advances in genetics and the proliferation of data across various pathologies, especially mental disorders, bring up the question of “normality.” In The Normal and the Pathological (Paris, Presses Universitaires de France, 1966), Canguilhem argued that the individual is the subject of their norms, meaning there is no inherent vital and social normality. In the vital realm, there are no ultimate norms, a scientific standard beyond individuals that can determine their relationship to normality. While normal and pathological states do not blur together, their distinction only holds for an individual in a specific state within their environment. The pathological state is primarily defined by the individual themselves.

Canguilhem’s analysis, further developed by Ricoeur in The Just 2 (Paris, Éditions Esprit, 2001), is notable in that it relativizes science’s power, relating life events to individuals' biographies and qualifying life norms by the unique perspectives that people bring based on their personal histories. This approach also considers the social representation of disease. Medicine should draw from new knowledge—such as genetics—to improve its effectiveness without reducing itself solely to that knowledge.