Introduction: We investigated clustering patterns of human papillomavirus (HPV) in a large study, the New Technologies in Cervical Cancer (NTCC) screening study.
Materials and methods: Women aged 25-60 years who attended cervical screening in eight different areas in Northern and Central Italy were tested for HPV infection with Hybrid Capture 2 (HC2). Genotyping of the HC2-positive samples was performed for 13 HPV types targeted by GP5+/GP6+ PCR, followed by Reverse Line Blot. Logistic regression was used to model type-specific HPV positivity, adjusted for age, study area, and specific HPV type prevalence. Subject-level random effects were added to represent unobservable risk factors common to all HPV types.
Results: A total of 36,877 women were included. Of 2833 HC2-positive women, 2108 were confirmed to be positive for any of the 13 specific HPV infections using the PCR assay, and amongst them 430 (20.4% of all PCR-positive women) were infected with multiple types. The observed-to-expected ratio for infection with ≥2 HPV types was 1.21 (95% Credible Interval: 1.13-1.30). Amongst the 78 combinations of specific HPV types, none of the pairs reached the chosen level of significance, p-value <0.01.
Conclusions: Multiple HPV infections occurred more frequently than predicted by chance. The excess of multiple infections was small, though not completely absent, after controlling for all sources of common correlation between HPV types. The present analysis of the NTCC screening study showed no evidence that specific HPV types have the tendency to be found more or less often than others in coinfections.
Copyright © 2011 Elsevier Ltd. All rights reserved.