Objectives: High-risk human papilloma viruses (HPV) cause a subset of oropharyngeal cancers. HPV 16 in particular, is responsible for over 90% of HPV-associated head and neck cancers. There is currently limited data on the natural history of oral HPV infection itself.

Methods: It is known that p16 expression localises to the specialised epithelium of tonsillar crypts but it is undetermined if tonsillectomy affects the natural history of oral HPV infection. This is the first UK study to provide crucial information on the potential impact of tonsillectomy on oral HPV infection.

Participants: 85 healthy participants aged >14years undergoing tonsillectomy for non-cancer reasons on the multicentre OROMOUTH UK study.

Samples: Oral rinse before tonsillectomy and 3 – 6 months after tonsillectomy. Tonsils, tongue base & pharyngeal wall brushes, were also collected pre-operatively.

Laboratory tests: HPV DNA amplification was performed using a broad-spectrum short-PCR-fragment assay [SPF10] PCR-DNA Enzyme Immunoassay (DEIA) followed by a primer SPF10-based line probe assay (SPF10LiPA) and MPTS 123 (a novel E6-based multiplex type-specific system that uses the Luminex xMAP technology).

Statistical analysis: All demographic and laboratory test results were extracted on to Excel spreadsheets and analysed using STATA SE.

Ethical approval: Approval was secured for this study with OROMOUTH MREC No.: 11/WM/0283, UKCRN ID: 12344.

Results: Overall median age in this tonsillectomy cohort was 23years (1QR: 19 – 31) with 70/85 (82.4%) females. 15/85 (17.6%) were positive for any oral HPV DNA before tonsillectomy in any oral sample. In the positive group were 2 males and 13 females.  8/85 (9.4%) were high-risk HPV while HPV 16 was identified in 2/85 (2.4%) in this cohort. The identified oral HPV genotypes were high-risk (16, 39, 51, 53, 59, 66) and low-risk (11, untypable). Tissue distribution of oral HPV detection pre-op was in oral rinse (9/85), tongue base brush (4/85), tonsils (3/85), and pharyngeal wall brushes (0/85).

74 patients had paired oral rinse available both pre- and post-operatively for analysis. There was a reduction from 7/74 (9.5%) with oral HPV positivity pre-operatively to only 1/62 (1.6%) post-operatively. This is an 83% reduction in oral HPV infection following tonsillectomy. The persistent infection was in a female patient with multiple high-risk oral HPV infection pre and post-op. New oral HPV infections were detected in 5/67 (7.5%) of patients who were previously oral HPV negative before tonsillectomy. The observed trend in oral HPV reduction following tonsillectomy was not statistically significant albeit promising (McNemar’s test statistic 0, 2-tailed p-value 1.0, odds ratio 1.2).

Conclusion: This is the first UK study to provide crucial data on the potential impact of tonsillectomy on the natural course of oral HPV infection. Tonsillectomy resulted in 83% clearance of oral HPV infection in our healthy cohort although not statistically significant, as there were new infections in the cohort indicative of its natural history. Tonsillectomy shows a trend towards abrogation of oral HPV infection which could be a significant intervention in the prevention of HPV oropharyngeal cancer. Validation of our study findings in a larger population is encouraged.