Evaluation of performances of three DNA enzyme immunoassays for detection of Helicobacter pylori PCR products from biopsy specimens.

PCR is recognized as a promising method for the detection of Helicobacter pylori in gastric biopsy specimens. However, detection of PCR products by gel electrophoresis is difficult to implement in routine clinical laboratories. The aim of this study was to compare three new DNA enzyme immunoassays with the standard method in their ability to detect PCR products. The three assays were based on the amplification of a fragment of the ureC gene of H. pylori and a colorimetric hybridization assay. The first assay (GEN-ETI-K DNA enzyme immunoassay; Sorin, Sallugia, Italy) was based on the hybridization of amplified DNA with a probe bound in microtiter wells and detected with labelled anti-DNA antibody. The second assay (Pylori-prob; Biocode, Sclessin, Belgium) comprised a solid-phase sandwich hybridization system with a specific biotinylated probe being used for detection. Finally, the third assay (PCR enzyme-linked immunosorbent assay; Boehringer, Mannheim, Germany) was based on the hybridization of amplified DNA labelled with digoxigenin as a probe (used as a coating in microtiter wells) and detected with antidigoxigenin-peroxidase as conjugate. The sensitivity of the colorimetric assay was evaluated by using amplification products from PCR assays performed on several 10-fold dilutions of DNA from H. pylori CIP 101260, and the specificity was assessed with different urease-positive bacteria. Biopsy specimens from 199 patients were tested; 106 were classified as H. pylori positive, and 93 were classified as H. pylori negative by culture and/or histological examination as the "gold standard." The receiving operating characteristic curve was used to determine the best cutoff point for each assay. The detection of PCR products by colorimetric hybridization increases the sensitivity up to 100-fold compared to that with gel electrophoresis. The results are rapid (4 h) and easy to interpret and can be automated.