HN 03-05P16INK4A EXPRESSION PATTERNS PREDICT CLINICAL OUTCOME OF PATIENTS WITH ORAL DYSPLASIA IRRESPECTIVE OF HPV INFECTION STATUS

26. Oral HPV infection
E.S. Prigge 1, T. Lubpairee 2, L. Zhang 3, M. Von Knebel Doeberitz 1, M. Reuschenbach 1, M. Rosin 2.
1Department of Applied Tumor Biology, University Hospital Heidelberg and Clinical Cooperation Unit Applied Tumor Biology, German Cancer Research Center (DKFZ), Heidelberg (Germany), 2British Columbia Oral Cancer Prevention Program, BC Cancer Research Center, Vancouver, British Columbia (Canada), 3Faculty of Dentistry, University of British Columbia, Vancouver, British Columbia (Canada)

Background / Objectives

Diffuse p16INK4a overexpression is a surrogate biomarker of transforming HPV infections in (pre-)cancerous lesions of the anogenital tract and in cancers of the head and neck. p16INK4a overexpression has also been reported in a fraction of oral premalignant lesions (OPL). Likewise, HPV DNA has been observed in a small proportion of OPL. However, the significance of p16INK4a as a marker of a transforming HPV infection in these precancerous oral lesions has not been determined to date. Moreover, it is not clear how p16INK4a expression status may impact on the lesion progression risk of OPL patients.


Methods

Formalin-fixed, paraffin-embedded (FFPE) tissue of mild, moderate and severe OPL from the “Oral Cancer Prediction Longitudinal study” (OCPL), British Columbia, Canada, were analyzed for HPV DNA, genotype and semi-quantitative viral load applying Luminex technology. p16INK4a/Ki-67 immunohistochemistry was performed on FFPE sections. p16INK4a expression patterns were classified as ‘negative’ (no p16INK4a-positive cells), ‘focal/patchy’ (scattered p16INK4a-positive cells only) or ‘diffuse’ (p16INK4a-expressing cells in a clonal distribution). The results were further correlated to lesion progression risk in the respective patients.


Results

HPV DNA from oncogenic types (16, 18 and 52) was detected in 15/241 OPL samples (6.2%). Among 237 analyzable samples, 5.5% demonstrated a diffuse p16INK4a expression pattern, 62.9% showed focal expression and in 31.6% no p16INK4a expression was observed. A significantly higher association was observed between HPV DNA-positivity and diffusely p16INK4a-stained lesions in comparison to samples with a patchy expression pattern or no expression at all (p=0.04). However, diffuse p16INK4a expression was also observed in a significant proportion of HPV DNA-negative cases. Patients with focal p16INK4a expression had a significantly lower risk of lesion progression over a total observation period of more than 12 years than OPL patients in which no p16INK4a overexpression was observed at all (p=0.0001). Patients with diffuse p16INK4a overexpression demonstrated an intermediate risk of lesion progression.


Conclusion

In contrast to the situation in the anogenital tract and in invasive head and neck cancers, diffuse p16INK4a overexpression does not represent a specific biomarker of HPV-related precancerous lesions in the oral cavity. A focal p16INK4a expression pattern in OPL provides valuable potential to identify cases with a low progression risk.


References