Considering its epidemic-like development worldwide, associated with modifications in
lifestyle, as well as its enormous social and economic weight, the prevention of type II
diabetes is certain to be a central concern of health systems within the developed countries
in the decades to come. However, while simple obesity concerns the entire population, type 2
diabetes affects only one sub-population at high genetic risk. To be effective and realistic
in economic terms, efforts at prevention must be thus targeted towards these subjects at high
risk. The key issue involves identifying such subjects early enough so that a strategy of
effective prevention can be organized in good time.
Until now, efforts have been concentrated on individuals at risk for diabetes readily
identifiable within the general population, typically subjects in the second half of
adulthood, presenting abdominal obesity and mild abnormalities of blood sugar. Preventive
lifestyle and dietary measures are proposed but are constrictive and difficult to maintain
over time, and the results, although they may be significant, remain disappointing, with mere
postponement of an outcome which at this stage appears inevitable. The reason is ascribable
to excessively tardy intervention, when the pathogenic process has already been ongoing for
some ten years and the endocrine function of the pancreas is probably already irreparably
The alternative thus is earlier intervention, in childhood, adolescence or early adulthood.
The problem is to identify individuals at high risk of becoming diabetic at a time when they
are presenting no simple clinical or laboratory abnormalities allowing easy diagnosis. The
familial character of type 2 diabetes is now well established, and future diabetic subjects
are themselves above all the children of diabetic subjects. However, the prevalence of the
disease among the descendants of type 2 diabetic subjects is around 20-30% and predictive
tools are needed to combat diabetes in these high-risk families.
We propose to create a risk equation using an algorithm to reliably predict children most
likely to develop diabetes later in life.
The algorithm will include 3 classes of data:
- The genotype stemming from the genetic characterization of individuals and those their
- Environmental data concerning childhood, especially eating habits and physical activity;
- Data of the mother who was eventually diabetic during pregnancy.
From a methodological standpoint, it would be rather difficult to take blood samples from
children and wait some 50 years to determine whether or not they develop diabetes. To
circumvent this difficulty, we will recruit subjects in families with a history of type II
- Parents alive, including at least one type 2 diabetic subject
- Adult children (aged over 35 years), some of whom are already presenting type II
diabetes, and healthy brothers and sisters, who form the control population. Test will
be done to determine whether healthy subjects are really safe from the risk of diabetes
(HbA1c measurement and glucose load test).
The Descendence study will include 500 families at risk involving about 3000 subjects (1000
subjects with diabetes and 2000 healthy subjects). It is expected to answer the following
question: for a child born in such families at risk, what is the probability of developing
diabetes later in life, so that early preventive action may be taken