Prognostic value of post-treatment HPV-specific antibodies in patients with oropharyngeal tumours

Published: 2 April 2019| Version 3 | DOI: 10.17632/77trkhy93c.3
Ruth Tachezy,


The presence of HPV-specific antibodies in patients with head and neck cancer at enrolment has prognostic significance. In cervical carcinoma patients, the decrease of HPV E6/E7-specific antibodies appears to be associated with a better prognosis. In this study, we analysed sera of 93 patients taken a year after the end of treatment and sera from 58 of these patients taken up to 14 years after treatment. HPV-specific antibodies in patients with oropharyngeal cancer reflect the tumour removal and their level declines with time from the treatment. The decline of HPV16 E7-specific antibodies is statistically steeper in non-recurrent in comparison to recurrent patients. However, since the decline occurs both in recurrent and non-recurrent patients these antibodies cannot be used as a marker for stratification of patients to the risk groups. While the presence of HPV-specific antibodies at enrolment is predictive of better survival, their level at enrolment is not. However, the changes in the level of HPV16 E6 antibodies on the one-year follow-up after the treatment predict better overall as well as disease-specific survival.


Steps to reproduce

In the present study, we analyzed sera of patients from our prospective study and long-term follow-up cohort. These patients had primary squamous cell cancer of the oral cavity or oropharynx. The enrolment sera were taken before treatment. In this study, we evaluated sera taken from 133 patients diagnosed with oropharyngeal cancer and positive at enrolment for HPV16 E6 and/or E7-specific antibodies. One-year follow-up sera were available from 93 of these 133 patients. Furthermore, we have invited these 93 patients for an additional check-up and blood sampling in 2015. Altogether, 58 patients presented for additional blood sampling. The presence of antibodies to antigens derived from HPV-specific proteins was tested using an in-house enzyme-linked immunosorbent assay (ELISA). L1 virus-like particles (VLP), mimicking HPV 16, 18, 31, and 33 capsids were prepared in a baculovirus expression system and used as antigens to detect antibodies to viral capsids. The presence of antibodies to HPV16 E6 and E7 oncoproteins, overexpressed in bacteria as fusion proteins with glutathione-S-transferase (GST), was assessed using the GST capture system. To estimate the change in mean HPV-specific antibody levels over time, the generalized estimating equations (GEE) with a working exchangeable correlation structure were used. The Cox proportional hazards model was used for the disease-specific survival (DSS) and overall survival (OS) analysis. The models included indicators of a decrease in specific antibodies levels and age of a patient. The assumption of proportionality was confirmed by Grambsch and Therneau test. Graphical comparison of survival rates for the patients with decreasing and nondecreasing antibodies levels was based on Kaplan-Meier plots. The Wilcoxon rank sum test was used for the test of difference in the OD index at enrolment between patients with and without recurrences. The changes in antibody levels were evaluated in patients positive for the respective antibody at enrolment.


Viruses, Head and Neck Cancer, Infectious Disease Serology