Role of High-Throughput Whole Genome Sequencing for the Diagnosis and Care of Atypical Diabetes

The prevalence of diabetes is 7.4% in France among people aged 20 to 79 years in 2015. We must also consider "pre-diabetes" (subjects with glucose intolerance), whose prevalence is equivalent to that of diabetes (2012 estimate). The incidence of diabetes is exploding both for type 2 diabetes, which represents 85% of diabetes, and for type 1 diabetes, which represents 10% of cases and starts one out of two times before the age of 20. Diabetes typing is essential to guide therapeutic choices, particularly the use of insulin. This typing is based on the pathophysiology of the disease, distinguishing insulinopenia from autoimmune causes in type 1 diabetes, monogenic diabetes, secondary or atypical diabetes and type 2 diabetes, where insulinopenia and insulin resistance coexist. Thus, while a formal biological diagnosis is possible for some forms of atypical diabetes and for type 1 diabetes, no biological parameter is currently available for type 2 diabetes, which remains a diagnosis of exclusion. As a result, diabetes represents a source of diagnostic and therapeutic erraticism, amplified by the clinical heterogeneity of type 2 diabetes, which is obvious and underestimated, and by a clinical phenotyping of patients that is often defective. The economic consequences are important because the health costs are very different depending on whether or not patients are treated with insulin. Type 1 and type 2 diabetes are examples of chronic, non-transmissible, multigenic, multifactorial diseases. However, less than 10% of the heritability of type 2 diabetes is currently explained by the associated genetic variants. And although genetic tests exist to diagnose certain monogenic diabetes, this diagnosis is made in less than 20% of cases, mainly in the presence of an atypical clinical presentation of diabetes. Moreover, there is no reason to rule out the hypothesis of paucigenic forms, at the interface of monogenic diabetes and multigenic forms as usually envisaged, as has been observed in chronic pancreatitis, which is also accompanied by diabetes.

The study will be conducted according to a randomized trial design comparing two diagnostic strategies defined as follows:

• Control strategy: in silico analysis of a panel of validated genes (ISApanel - Diabetome 1). Patients recruited along the control procedure will stay in their group using current genetic diagnosis practices and standard of care that may differ from one center to another.

• Intervention strategy: whole genome sequencing coupled with multidisciplinary conciliation meeting.

We plan to randomize one patient in the control group for two in the intervention group.

The main objective of the study is to assess the contribution of whole genome sequencing (WGS) coupled with a multidisciplinary conciliation meeting (MCM) on diagnosis of atypical forms of diabetes compared to an in-silico analysis of a panel of validated genes (ISApanel), corresponding to current practice.

The target population is 1020 adults with atypical diabetes for whom it is possible to obtain a blood sample.