Human papillomavirus and other biomarkers in neck masses of unknown origin and in head and neck cancer

Abstract

Background. Human papillomavirus (HPV) is, in addition to smoking and alcohol, a risk factor for developing oropharyngeal squamous cell carcinoma (OPSCC), and more specifically tonsillar- and base of tongue squamous cell carcinoma (TSCC & BOTSCC respectively). Interestingly, HPV-positive (HPV+) TSCC/BOTSCC has a remarkably better clinical outcome than HPV-negative (HPV-) TSCC/BOTSCC and head and neck cancer in general. However, the role of HPV and other biomarkers in cancer of unknown primary in the head and neck region (HNCUP), i.e. where only a lymph node metastasis and no primary tumour is found, is not well understood. Moreover, the genetic landscape of HPV-positive HNCUP and the prognostic implications of certain mutations in TSCC/BOTSCC has not been studied extensively. Furthermore, using fine-needle aspiration cytology (FNAC) from neck masses for HPV-detection and potentially for predicting an HPV-positive TSCC/BOTSCC as the final diagnosis would be highly clinically useful, but has not been studied prospectively in a cohort consisting of both malignant and benign neck masses.

Aims. To investigate HPV DNA and mRNA and other prognostic biomarkers (p16, p53, CD8+ tumour infiltrating lymphocytes (TILs), HLA Class I expression) in HNCUP in relation to clinical outcome. Moreover, to study mutations in HPV+ HNCUP and TSCC/BOTSCC in relation to prognosis and to analyse whether HPV-detection in FNAC from neck masses is reliable and if a finding of HPV could predict an HPV+ TSCC/BOTSCC as the final diagnosis.

Results. In Paper I, we show that HPV is a favourable prognostic factor in HNCUP, and in Paper II, we validate this finding in a separate cohort (3-year overall survival 86% vs. 54% for HPV+ and HPV- HNCUP respectively, in the combined cohorts). In Paper II, we also demonstrate that HPV mRNA is expressed in the vast majority of HPV DNA+ HNCUP. Moreover, in Paper I we find that high p53-expression correlates to a poor prognosis, and in Paper II that a low number of CD8+ TILs are potentially related to a poor outcome, while HLA Class I expression does not appear to be a prognostic factor in HNCUP. In Paper III, we show that TP53, CDKN2A and PIK3CA are the most commonly mutated genes in HPV+ HNCUP, and that having a mutation in FGFR3 correlated to a poor prognosis in HPV+ TSCC/BOTSCC. In Paper IV, we demonstrate that HPV DNA detection in FNAC from neck masses of unknown origin is reliable and could be used prospectively to predict an HPV+ TSCC/BOTSCC as the final diagnosis. HPV DNA was not found in malignant conditions other than HPV+ TSCC/BOTSCC or in any benign conditions, including branchial cleft cysts.

Conclusions. HPV appears to have a similar causative and prognostic role in HNCUP as in TSCC/BOTSCC. Investigation of HPV-status should therefore be part of the diagnostic work-up of an HNCUP. Examination of HPV DNA-status using FNAC is useful in the clinical investigation of patients with neck masses of unknown origins, including patients eventually diagnosed with HNCUP or TSCC/BOTSCC.