B-261 Identification of HLA-DQ2.5 Haplotype Using a Real-Time PCR Method

Abstract

Abstract Background Celiac disease is an autoimmune pathology triggered by the ingestion of gluten in genetically susceptible patients. The disease presents a clinical heterogeneity that makes the combined use of laboratory and histopathological approaches necessary. The presence of the DQ2.5 haplotype (DQA1*0501/DQB1*0201) is associated with celiac disease. For example, in Caucasian patients, approximately 90% of patients with celiac disease carry the DQ2.5 haplotype. However, the DQ2.5 haplotype is present in only 15 to 30% of the general population. Due to this, the presence of the haplotype cannot be considered decisive for the diagnosis of celiac disease. On the other hand, the identification of the haplotype is useful in excluding the possibility of the disease, since the absence of the marker makes the hypothesis that a patient has celiac disease very unlikely. One of the main approaches for identifying the DQ2 haplotype is the use of PCR with specific primer sequences. Although this technique has great sensitivity and specificity, the process is time-consuming and costly. Thus, in order to improve the cost-effectiveness of the diagnosis we investigated if the use of SNP tag approach is able to correctly identify HLA-DQ2.5 genotype. Methods We selected 235 samples from patients genotyped for HLA-DQ2.5 by PCR followed by electrophoresis detection. 94 samples were positive for the HLA-DQ2.5 haplotype. The 235 samples were genotyped using TaqMan probe assay (ThermoFisher Scientific) for the SNP rs2187668 according with the manufacter instructions. The SNP rs2187668 has a C/T substitution. The homozygous C genotype is associated with the absence of the HLA-DQ2.5 haplotype, while the presence of at least one T allele is associated with the presence of the HLA-DQ2.5 haplotype. Results Of the 94 positive samples for the HLA-DQ2.5 haplotype, Real-Time PCR genotyping was able to correctly identify 74 (70 C/T samples and 4 T/T samples). The other 20 positive samples were identified with the C/C genotype. Among the negative samples, Real-Time PCR genotyping correctly identified the genotype of 139 of the 141 samples. These results indicate a sensitivity of 82.4% and a specificity of 98.6%. On the other hand, the positive predictive value is 97.4% and the negative predictive value is 89.7%. Conclusion The results observed suggest that the use of the SNP rs2187668 is effective for detecting the presence of the HLA-DQ2.5 haplotype, given the low rate of false-positive individuals and the high positive predictive value. On the other hand, the exclusion of the presence of the haplotype did not obtain the same effectiveness, since a high frequency of false-negative results was observed. One of the possibilities for this high frequency of false-negatives may be due to the high number of polymorphisms in the SNP region or the low frequency of the T allele in the population. Thus, other approaches are necessary to establish the diagnosis of the HLA-DQ2.5 genotype by Real-Time PCR, such as, for example, the adoption of other SNPs in the region.