RESUMO
BACKGROUND: Persistent infection with human papillomavirus (HPV) can lead to cervical cancer (CxCa). During the progression to CxCa, the expression of HPV oncogenes E6 and E7 is upregulated. In turn, cellular proteins such as p16INK4a are also modulated. The combined detection of HPV oncogenes and cellular biomarkers indicative for dysplasia could be informative and convey better specificity than the current HPV tests that cannot discriminate transient infection from dysplastic changes. METHODS: The QuantiGeneTM 2.0 Plex Assay platform was chosen for the effective multiplexing and quantitative detection of seven HPV-E7 mRNA targets (HPV6, 16, 18, 31, 45, 59, and 68) and the cellular mRNA of p16INK4a as a biomarker for HPV-induced transformation. Actin-beta (ACTB) and hypoxanthine-guanine phosphoribosyltransferase 1 (HPRT1) were included as reference markers. Sequences for the specific capture and detector probes were customized and developed by ThermoFisher and formulated as a QuantiGene proof-of-concept (QG-POC) plex-set. The crude lysates of the HPV-positive cervical cancer cell lines CaSki (HPV16), HeLa (HPV18), MRHI-215 (HPV45), Erin59 (HPV59), ME180 (HPV68), and the HPV-negative cell line C33A, as well as liquid-based cytology smear samples (n = 441) were analyzed. The study was a proof-of-concept evaluating the feasibility of the platform. Logistic regression and receiver operating characteristic (ROC) analyses were performed to test for the sensitivity and specificity of HPV detection and dysplastic stage discrimination. RESULTS: A QG-POC assay specifically and sensitively detects the HPV-E7 mRNA of seven different genotypes with an assay linearity between 20 and 13,000 cells. Cellular mRNA was detected from the crude lysates of cell lines and of cellular material from clinical liquid-based cytology smear samples. By combining HPV-E7 and p16INK4a expression normalized to ACTB, high-grade dysplasia (HCIN) and invasive cervical cancer (CxCa) were detectable, discriminable, and correlated to the biomarker expression strength. The ROC analysis from the multivariate logistic regression model including HPV-E7 and p16 INK4a resulted in an AUC of 0.74, at the optimal cut-off (sensitivity: 70.4%; specificity: 66.0%) for HCIN detection. CxCa was detected with an AUC of 0.77 (sensitivity: 81.8%, specificity: 77.4%). CONCLUSIONS: The QG-POC assay is sufficiently sensitive to detect and quantify HPV-E7 and cellular mRNA species. Multiplexing allows the specific detection of at least 10 analytes in a single reaction. Determining the abundance of E7 and p16INK4a transcripts when normalized to ACTB is informative about the presence of cervical dysplasia and potentially discriminates between low-grade and high-grade dysplasia and invasive cervical cancer. Further studies including more HPV genotypes and biomarkers are warranted.