Abstract
Background: The human papillomaviruses (HPVs) are known to induce cutaneous and mucosal proliferation of epithelial cells. HPVs are classified into low-risk (LR) and high-risk (HR) types, depending on their association with the development of benign, potentially malignant or malignant lesions. Oral verrucal-papillary lesions (OVPL) are a distinct spectrum of epithelial lesions affecting the oral mucosa from benign to malignant: some are thought to be associated with HPV. The HR-HPV types have been implicated as causative agents in the development of some head and neck squamous cell carcinomas (HNSCC) and cervical cancer. However, the role of HPVs in the pathogenesis of OVPLs and their transformation to oral squamous cell carcinoma (OSCC) is uncertain.
Objective: To determine the expression of HPV and other HPV-related proteins and quantify HPV nucleic acids (DNA) in OVPLs and OSCC using immunohistochemistry (IHC), in-situ hybridisation (ISH) and quantitative real time polymerase chain reaction (qRT-PCR) detection methods and to compare the sensitivity and specificity of the HPV detection techniques.
Methods: Forty-one archival formalin-fixed paraffin-embedded (FFPE) tissue samples were obtained and grouped into four specific histopathologically confirmed groups of OVPLs and related lesions. For pan-HPV and p16 IHC the groups were as follows; squamous papilloma (SP; n=10), verrucous hyperplasia (VH; n=15), verrucous carcinoma (VC; n=6) and oral squamous cell carcinoma (OSCC; n=10). Cervical carcinoma tissue samples were used as positive controls (n=2) for validation. Qualitative assessment of the distribution and pattern of staining for both pan-HPV and p16 antibodies were carried out using light microscopy at various magnifications. p16 quantitative assessment of digitised IHC images (400x magnification) using Image J software to examine the extent of positive reaction, intensity of staining and determine immune-reactive score for all experimental tissue samples was also undertaken. One-way ANOVA tests in the Prism GraphPad 7 for Mac OS X (GraphPad Software, Inc, USA) and a Sidak's multiple comparisons test also within the ANOVA, was used to analyse the difference in the expression between the experimental test groups. P values of <0.05 were considered to be statistically significance between the experimental groups.
A trial run using both HR and LR ISH probes with a Ventana ISH system in conjunction with an iView detection kit and a Leica ISH system for probing for HR-HPV DNA on the cervical cancer positive control was undertaken. For qRT-PCR, extraction of DNA was conducted on all 43 FFPE tissue blocks including the two positive control tissue samples using GeneJET FFPE DNA Purification Kit (#K0881, Thermo ScientificTM Life Technologies, New Zealand Limited). The quality of DNA, was measured by A260/A280 ratio, ideally should be >1.7. However, from the trial experiment, a quality threshold value of 1.6 was determined. Quantification and detection of high-risk HPV (HR-HPV) (12 types, 16 and 18) DNA was carried out using qRT-PCR TaqMan assays.
Result: HPV and other related proteins were differentially expressed amongst the OVPLs and OSCC. Pan-HPV proteins were highly expressed in all tissue samples pertaining to the OVPLS and OSCC experimental groups. p16 protein was underexpressed in the aforementioned groups and only one tissue sample pertaining to the VC group was p16-positive. A statistically significant proportion of score (PS) (p < 0.05) in a sample pertaining to the SP group when compared to OSCC was also noted. With the Ventana ISH system no positive results were obtained with the ‘positive’ control, despite several trouble-shooting exercises. Positive brown punctated staining for HR-HPV types 16,18,31,33 and 51 using Leica Biosystem (DNA) probes was observed within nuclei of some malignant cervical epithelial cells, but due to technical difficulties the full experimental run could not be completed. HR-HPV DNA (16, 18 and 12 types) was not detected in any of the OVPL FFPE tissue specimens using qRT-PCR. However, HR-HPV (12 types) DNA was detected in the internal controls as well as both positive control cervical tissue samples.
Conclusion: Pan-HPV proteins (1, 6, 11, 16, 18 and 31) were highly expressed in OVPLs and OSCC samples. The overexpression of pan-HPV proteins may be due to the fact that only LR-HPV are associated with these lesions. p16 protein was underexpressed amongst the OVPLs and OSCC samples. Hence, in this study, there was no evidence to suggest that p16 was a useful surrogate marker for HR-HPV infection in OVPLs and OSCC. Using qRT-PCR HR-HPV types 16 and 18 were not detected in any of the OVPL and OSCC experimental samples nor in the positive control samples used. There was no evidence to suggest that HPV types 16 and 18 play an oncogenic role in the pathogenesis of OSCC. The other 12 HR-HPV types (31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68) were quantified and detected in the cervical (positive control) tissue samples. The rest of the tested group samples showed negative results. qRT-PCR is more sensitive and specific when compared to IHC and must be considered to be the first initial HPV detection method.