Analytical evaluation of the PapilloCheck test, a new commercial DNA chip for detection and genotyping of human papillomavirus.
Résumé
Recently, a commercially available HPV DNA chip, the PapilloCheck test, developed by Greiner Bio-One, has become available for human papillomavirus (HPV) genotyping. The PapilloCheck test is a PCR-based test using a new consensus primer set targeting the E1 HPV gene. HPV oligoprobes immobilized on a DNA chip allow for the identification of 24 HPV types from the amplified product. In the present study, the analytical performance of the PapilloCheck test is compared to the Linear Array HPV genotyping test (Roche Diagnostics). Cervical specimens collected in PreservCyt (Cytyc) solution and obtained from women who presented abnormal cytological findings were tested primarily by the Hybrid Capture 2 High-Risk assay (HC2-HR, QIAGEN). A total of 144 samples were selected according to the signal intensity obtained with the HC2-HR test, expressed as RLU/CO value, and divided into 4 groups as follows: [0-1] RLU/CO (negative HC2-HR result, 34 samples); [1-5] RLU/CO (positive HC2-HR result, 30 samples); [5-40] RLU/CO (positive HC2-HR result, 40 samples); >40 RLU/CO (positive HC2-HR result, 40 samples). The concordance levels between the HC2-HR test and each of the genotyping assays was similar (88.8%) and the crude agreement between these assays was considered as "good". The detailed analysis of the discrepant results confirmed a possibly high rate of false positive results of HC2-HR test in the 1-5 RLU/CO grey zone. Genotype-specific comparison analysis was limited to the 23 HPV types detected by both genotyping assays (HPV types 6, 11, 16, 18, 31, 33, 35, 39, 40, 42, 45, 51, 52, 53, 55, 56, 58, 59, 66, 68, 70, 73 and 82). Of the 135 samples available for comparison, 91 (67.4%) showed absolute agreement between the assays (concordant genotype-specific results), 34 (25.1%) showed correspondence for some but not all genotypes detected by both assays (compatible genotype-specific results), and the remaining 10 (7.4%) samples did not show any similarity between the tests (discordant results). The majority of discordances were found in samples containing multiple HPV types and in samples harboring low amounts of HPV. For some HPV genotypes, there were slight differences in the detection rate between the two genotyping methods. The Linear Array test seemed to be more sensitive to detect HPV type 53 whereas PapilloCheck test seemed to be more sensitive to detect HPV type 56. For the other genotypes, including HPV types 16 and 18, the results obtained by the two methods did not differ significantly. In conclusion, this study shows that the PapilloCheck test and the Linear Array test give comparable results for detecting HPV in cervical specimens. However, these results also suggest that there is a need to standardize the type-specific sensitivity of genotyping methods and to evaluate their accuracy to detect multiple HPV infections. This would be a prerequisite for the use of genotyping assays in cervical cancer screening algorithms.