[Predictive genetic screening for type-1 diabetes in the Hungarian population]

In the modern genetic era the principles of genetic screening are being changed. In addition to diagnostic screening for rare monogenic diseases, predictive screening for common polygenic conditions, like type 1 diabetes, will be more widely implemented. The majority of the genetic background of type 1 diabetes is encoded by the HLA DQ and DR genes, selected variants of which could be used as screening markers. However, risk conferred by various HLA genotypes shows considerable ethnic variation, therefore population-specific screening markers need to be established.

The aim of this study was to describe a screening strategy based on risk-defining HLA DRB1-DQA1-DQB1 markers to identify individuals at risk for type 1 diabetes in the Hungarian population.

HLA genotypes of 149 consecutively diagnosed children with type 1 diabetes (age at diagnosis 0-14, mean 8.8 +/- 4.2 years) and 177 randomly selected healthy schoolchildren were studied. Allele-specific polymerase chain reaction method was used for HLA typing. The diagnostic sensitivity, specificity and predictive value of diabetes associated DRB1-DQA1-DQB1 alleles were analysed in a step-wise strategy.

The highest diagnostic sensitivity was detected when DQB1 typing was complemented by DQA1 typing on DQB1*0201 positive samples with additional DRB1*04 subtyping in DQB1*0302 carriers. The combination of the following markers gave a relative risk of 28.9 (95% confidence interval: 15.9-52.7, p = 10(-6): DQB1*0201/0302-DQA1*0301,0501-b (b not equal to DRB1* 0403), DQB1*0302/x-DRB1*0401,0402 (not equal to DQB1*0201, 0301,0602,0603), DQB1*0301/0302-DRB1*0401,0404, DQB1*0304/s (s = any DQB1 alleles), DQB1*0201/y-DQA1*0501/a (y not equal to DQB1*0301,0302,0602, 0603,0604, a not equal to DQA1*0201). The diagnostic sensitivity, specificity and positive predictive values for this marker combination were 79.2%, 88.7%, and 1.1%, respectively.

Using HLA DRB1-DQA1-DQB1 markers predictive genetic screening for type 1 diabetes is feasible in the Hungarian population with high diagnostic sensitivity and specificity. At present, such a screening for individuals at risk for type 1 diabetes in the general population is recommended only as part of prospective studies on the natural history or prevention of disease. To increase the positive predictive value of the model, pancreas beta-cell autoantibodies need to be measured and followed in the high-risk cohort.