Evaluation of human papillomavirus DNA detection in samples obtained for routine Chlamydia trachomatis screening.

BACKGROUND

The costs and logistics involved in obtaining samples is a bottleneck in large-scale studies of the circulation of human papillomavirus (HPV), which are useful for monitoring and optimisation of HPV-vaccination programs. Residual samples obtained after screening for Chlamydia trachomatis could constitute a convenient, low-cost solution.

OBJECTIVES

We evaluated HPV DNA detection and typing using (i) the residual samples routinely taken for C. trachomatis screening or (ii) the sample types used in large-scale phase III HPV vaccination trials (cervical, vulvar, labial, perineal, perianal, scrotal and penile shaft samples).

STUDY DESIGN

Samples from 127 men and 110 women attending two sexual health clinics were analysed using PCR for HPV DNA, with typing using mass spectrometry.

RESULTS

The HPV DNA prevalence was 7.1% in male urine samples, but 57.3% in female urine/vaginal samples, which was even higher than the HPV prevalence found in cervical samples (54.1%). The sensitivity for HPV DNA detection in the urine/vaginal samples was 7.9% (95% CI 3.0-16.4) for men and 78.9% (95% CI 67.6-87.7) for women, using detection in any one of the reference samples as reference. With cervical samples as reference, the sensitivity was 89.3 % (95% CI 78.1-95.9).

CONCLUSIONS

Among men, low sensitivity of urine for HPV detection suggests limited usefulness. Among women, the high sensitivity of urine/vaginal samples for HPV detection suggests a useful low-cost solution for the study of HPV epidemiology.