Genes: tests that save

Between 2001 and 2007, the number of genetic tests has tripled. For less than 650 €, start-ups sell online genetic kits to know our predispositions to twenty diseases. One hundred to one hundred and fifty genes are routinely tested in official health networks or on the Web. All this makes you dizzy! And yet, the order of genetic tests via the Internet abroad remains illegal in France and punishable by penalties (fine, most often). All this seems far from your practices. Not so sure...

You may not know it, but your baby has been tested for genetic diagnosis at the maternity clinic, during a routine neonatal screening with a simple heel prick. Tests that have saved many thousands of children from deep mental retardation. If necessary, a treatment is quickly put in place: a specific diet for phenylketonuria, for example, or thyroid hormones for hypothyroidism. These genetic diseases are detected and prevented by simple predictive tests based on biomarkers (proteins or enzymes), indirect reflections of the activity of failed genes. This is the magic and wonderful side of predictive medicine.

Thanks to the considerable progress of genetics and microarrays, we can go even further and probe the genes themselves. To the point of reading the medical future of everyone in the genetic book? This is where things need to be cleared

Genes: Predictions to Relate

In concrete terms, what is a gene and what are genetic tests based on? " A gene is information that gives instructions to the cells of our body. Like any information, it can be good or bad, "says Ségolène Aymé, geneticist epidemiologist, director of research at Inserm and head of the platform on rare diseases Orphanet. In each of our cells, there are 25,000 genes, they are the ones that are explored by chips and DNA probes. When variations of a single gene cause an inherited genetic disease - such as for cystic fibrosis or hemophilia - a high predictive genetic test is obtained.

But this predictive value becomes very low for diseases that are dependent on several genes, influenced by the environment - by far the most numerous (diabetes, cardiovascular disorders, non-familial cancers, Alzheimer's disease ...). For these multifactorial diseases, the genetic part is most often less than 30%, and then the question of the usefulness of these tests arises.

" What good is it to know that, compared to the total population, we have 5% or even 20% additional risk of developing Alzheimer's disease? "Ignites the professor Jean-Claude Ameisen, president of the committee of ethics of Inserm. This means that in 80 to 95% of cases, the risk is the same as for the rest of the population. Suppose that because of a mutation found on one of your genes, you are diagnosed with a risk 70 or 80 times higher than that obtained in the population, you will be frightened. Wrong! Indeed, you need to know the real risk of illness in the whole population, and if it is ridiculous (style 1 case out of 2,000, or 0.05%), your real risk will be just as ridiculous. 70 or 80 multiplied by 0.05%, this gives at most a risk of 4% to contract the disease: what relativize your fears.

Rodent studies have shown that if we introduce the gene deficient Alzheimer's disease in their genetic code (genome), some mice lose memory, but others, on the contrary, would offset the deficit acquired in function of their place of life. In other words, environmental factors affect the activation of a gene, to "turn on" or "extinguish" it, and for the moment, including in humans, the best predictive test of this gene. Degenerative pathology of the brain is the social level and standard of living.

Thus the disease is not inscribed "only" in our genes but in our environment and our way of life. So the belief that our medical future and that of our descendants is definitely inscribed in our genes is a big mistake, that the test merchants exploit for their benefit. Or again, as the generalist dreads in the " Bulletin of the Order of Doctors" (No. 4, March-April 2009): "I fear that we will replace the illusion of a medicine who heals everything by a medicine that predicts everything . "

Ségolène Aymé does not mince words: "All these tests sold in the United States and on the Internet are wind, business! Of course, we manage to read the genetic book of each but often we do not know if the variants of the detected genes will have consequences or not, especially since, for the same genetic alteration, a person will suffer serious symptoms while another will remain unscathed, the failure having been compensated by the genes present, opposite to the other chromosome of the same pair, or by genes located elsewhere. And even if we detect with a high probability a disease that will occur several decades later, what is the service rendered to the person if we know nothing to prevent the disease from occurring? "

Moreover geneticists keep repeating that we are all healthy carriers of a good fifty genes of inherited diseases, that's what to relativize. They also say: "Genetics proposes, the environment disposes. "

Genes: prevention and bioethics

Mothers have the breasts of their barely pubescent girls removed because they carry the famous mutations on the BRCA1 or BRCA2 genes (1). Scandalous? Certainly not ! As long as you have not lived with such a sword of Damocles over your head, you can hardly become a judge. Thus, in the case of familial breast cancer , the presence of a mutation on the BRCA1 gene leads to a risk of developing a tumor in the breast of approximately 45% before the age of 50, as against 2% in women without the mutated gene. A woman who knows that she is affected can opt for close monitoring of the breasts and ovaries (potentially endangered too), she can also choose to undergo a preventive removal of the breasts or even the ovaries, which reduces to less than 2% the risk of mortality.

In the United States, other parents, dwarf or deaf mute, have been able to use preimplantation genetic diagnosis (PGD) after in vitro fertilization to be able to give birth to a baby with dwarfism or deaf-mutism - you read correctly: to convey their disability and not to avoid it! According to an article published in the specialized press (2), 3% of the American clinics practicing DPI used this one to transmit a handicap, at the request of the parents who wished that their children resemble them. In addition, 42% of these same establishments practiced this IPR to select the sex of the child without a medical reason, and 24% did so to give birth to a "medicinal" child, close to the immunity level. a sister or brother who is seriously ill. To date, in France, these practices (except the last) are not allowed, but are debated with the revision of the bioethics law, which should be completed in 2010.

Genes: When are genetic tests really useful?

The tests allow either to be reassured that one is free from a hereditary family disease, or to adopt adequate surveillance and appropriate treatment. In the person carrying the hemochromatosis gene, a disease due to the accumulation of iron in the blood, early diagnosis through the predictive test allows bleeding to prevent cirrhosis and liver cancer; in cases of familial polyposis polyps are removed to prevent colon cancer. But sometimes there will be no solution, and the test only serves to know if one is carrying or not the mutation. The most painful illustration of this phenomenon is Huntington's disease, a serious and incurable neurological condition that can be detected decades in advance, while the person is still in perfect health. "Where is the benefit of such information for the person? wonders Arnold Munnich (3). Of course, it's a hell of living with doubt about his status. But it is also doubt and the unknown that make us move forward ... The risk of prediction is precisely that the future is confused with the present! The challenge for teams in charge of presymptomatic tests (which precede the signs of the disease) is to determine whether genetic testing is good or bad for the person. And to evaluate if the arguments invoked to justify the test (marriage, child project ...) weigh quite heavy considering the earthquake that represents the fact of leaving a probabilistic situation to live in the certainty of carrying the gene of the disease. "

In France, 10% of people at risk for Huntington's disease ask to benefit from a predictive test, the others prefer to ignore the possible sanction.

At home, these tests are usually performed in view of a family history already heavy medical history. It is the attending physician or the pediatrician who sends his patient to the specialist for a genetic counseling consultation supported by Social Security. However, there is nothing stopping anyone from coming to laboratories that have an Internet presence, or even private genetic laboratories that offer the perfectly legal (but not reimbursed) possibility of testing a dozen diseases. Potential customers: couples who want children and who fear certain diseases.

Genes: Are genetic tests reliable?

" The tests sold on the Internet are out of control and can be fancy, says Ségolène Aymé . It's a little Russian roulette, without psychological counseling or medical advice in case of bad news. Another reason for concern is the perverse use of these tests. In Switzerland or the Netherlands, insurers require proof of good health to take out life insurance of a high amount. There is also a fear of using these tests for hiring. Most European states forbid these practices, but until when? And what will happen, asks Professor Jean-François Mattéi in "Questions of Biomedical Ethics " (Flammarion ed.), When your neighbor decides to use his genome vis-à-vis an employer to take advantage of it, illustrating a new form of discrimination?