Pitfalls in the epidemiologic classification of human papillomavirus types associated with cervical cancer using polymerase chain reaction: driver and passenger

Abstract

Cervical cancer is a common malignancy in women worldwide, and it has now been established that the human papillomavirus (HPV) is both necessary and causal for these lesions. HPV itself is both ubiquitous and markedly heterogeneous but can nevertheless be classified as either a high-risk type or a low-risk type based upon its frequency of detection in cervical cancer. Given that the association between HPV and cervical cancer is causal, the classification of this virus has been strengthened by large-scale epidemiologic studies and is widely accepted across many disciplines. It is evident, however, that cervical cancer is frequently associated with multiple HPV types. Therefore, it is crucial to distinguish causal types of HPV (drivers) from noncausal types (passengers) in cervical lesions. In this review, we highlight the current pitfalls of using polymerase chain reaction methods instead of Southern blot hybridization for detecting HPV and discuss the distinction between driver and passenger HPVs with regard to the viral type, the length of the viral genome, and the levels of viral DNA associated with cervical cancer. Finally, we newly propose three categories of HPV instead of two risk groups, based on similarities between viral genes