[Diffusion of genetic testing in oncology: what criteria for regulation?]

Does gene testing indicate a switch from an histopathological to a molecular approach of human diseases ? Disease management in oncology is already improved by gene testing, at least for some specific cancers. It is however necessary to distinguish the analysis of genes specific to the tumour which gives clues about the fate of the tumours, from those unique to the patients, which gives clues about the future of the person. For the latter so-called germline mutations, wide scale gene-default screening would put pressure on resource allocation from the health care systems of developed countries. Currently the cost of detecting of 700 genes in the whole French population would exceed the whole health budget of the country for the next 10 years. Even if we can anticipate a dramatic decrease in the unit cost of these genetic tests in the future, their diffusion should not be controlled exclusively by technological and market forces. In this paper, we propose to discuss four main parameters for regulating these genetic tests, using as an archetypal example their application to cancer prevention and treatment: (1) which specific cancer disease is targeted by the test (prevalence, incidence, likelihood of cure with current therapeutics, number of years of life potentially saved...); (2) what are the characteristics of the genes tested and which level of evidence is required about the predictive value of the test; (3) what are the size and characteristics of the population who will be offered the test, and (4) which process and public control are necessary before market approval of the test and reimbursement of related expenditures by health care insurance schemes.