Relationship between CK19 expression, deregulation of normal keratinocyte differentiation pattern and high risk-human papilloma virus infection in oral and oropharyngeal squamous cell carcinoma

BACKGROUND: Simple epithelial keratins appear early during embryonic development and are expressed in non-stratified, ductal and pseudo-stratified epithelial tissues. CK19, the lowest molecular weight keratin, is also expressed in basal layer of squamous epithelia of mucosal surfaces. Previous studies have shown that High Risk-Human Papilloma Virus (HR-HPV) epithelial infection induces cell immortalization via E6 and E7 viral proteins and this, in turn, impairs cytokeratin expression in cancerous cells lines derived from uterine cervix. Here, we demonstrate the possible relationship between HR-HPV(+) oral/oropharyngeal cancer and the high levels of CK19 expression. METHODS: We analyzed 38 cases of Oral Squamous Cell Carcinomas/ Oro-Pharyngeal Squamous Cell Carcinomas (OSCCs/OPSCCs) by Immunohistochemistry (IHC) using specific antibody (Ab) detecting CK19, by In Situ Hybridization (ISH) and Polymerase Chain Reaction (PCR) based methods in order to define the HPV infectious status. We also evaluated the variation of CK19 expression in UPCI-SCC-131 (HPV(−)) and UPCI-SCC-154 (HPV(+)) cell lines by immunocytochemistry (ICC) and flow cytometry. RESULTS: CK19 OSCC/OPSCC score has been identified multiplying percentage of cancer expressing cells to staining intensity. CK19 expression score in OSCCs/OPSCCs was very different between HPV(+) (mean: 288.0 ± 24.3) and HPV(−) cancers (mean: 66.2 ± 96.9). This difference was statistically significant (p < 0.001) with a strong evidence of correlation (p < 0.001; Spearman’s R: +0.72). ROC curve analysis was performed on CK19 expression index related to HPV positivity. Heterogeneous areas of immunoreactivity varying in percentage value, intensity and/or localization were observed in normal epithelium, both perilesional and distant from the tumor with important differences between HR-HPV(+) and HR-HPV(−) carcinomas. By ICC and flow cytometry, the two analyzed cell lines were both CK19 positive but showed a different level of expression, in particular it should be noted that the UPCI-SCC-154 (HPV(+)) cell line had a higher expression than UPCI-SCC-131 (HPV(−)). CONCLUSIONS: In this study we demonstrated, for the first time, strong association between CK19 up-regulation and HR-HPV(+) OSCCs/OPSCCs. This test has a good accuracy. We identified ROC curve with a cut-off > 195 for HR-HPV positive results (Sensitivity: 92.3 %; Specificity: 89.3 %). Furthermore, in OSCC/OPSCC, the CK19 test may be useful in identifying HR-HPV infection, the latter being related to HPV E7 potential to disrupt normal cytokeratin expression pattern.