BACKGROUND & OBJECTIVES: Second generation antipsychotics (SGAP) used to treat severe mental illnesses (SMI) are known to induce metabolic side effects which, over time, can lead to type 2 diabetes (T2D)1. An important interindividual variability in the susceptibility to T2D is observed during SGAP treatment, explained in part by genetic factors, emphasizing the importance to identify patients at risk2. The aim of this study is to find genetic markers that can help to predict the risk to develop T2D as side effect of SGAP to improve prognosis of patients affected by SMI.
METHODS: Data were collected from an ongoing observational study monitoring metabolic traits during SGAP treatment. Diagnosis of T2D was established on the presence of an antidiabetic drug and/or on two consecutive abnormal fasting blood glucose levels (≥ 7mmol/l). To evaluate the contribution of genetic markers to risk prediction of T2D, two different logistic models were used: one with clinical data only and a second one including clinical data and single nucleotide polymorphisms (SNPs) previously associated with T2D and obesity in genome wide asso- ciation studies. The area under the curve (AUC) of the receiver operating characteristics analysis of the different models was compared.
RESULTS: 274 patients (median age (IQR)= 49 (30-71) years) were included out of which 31 were diagnosed as being diabetic. Among clinical variables, age and waist circumference were significantly associated with T2D. AUC of the final model including SNPs was higher than the model including only clinical variables ((AUCfinal(95th)=90.5(86.5-94.4), (AUCclinical(95 th)=78.0(69.9-86.1), P-value=0.0003). Nevertheless accuracy, specificity and sensitivity were not statistically improved after adding genetic variants.
DISCUSSION & CONCLUSIONS: These results demonstrate that some known SNPs may help to predict T2D during SGAP treatment. However the clinical relevance of using genetic markers for T2D prediction can presently not be established. Replication studies, including many other genetic variations and larger number of patients, are needed.
