Multiple Human Papillomavirus Infections: The Exception or the Rule?

Abstract

In this issue of the Journal of Infectious Diseases, Chaturvedi et al [1] report on a large epidemiological study of human papillomavirus (HPV) infection focusing on a topic of increasing interest in this field. HPV is recognized as a necessary cause of cervical cancer, but there are multiple oncogenic HPV types that may be found together in the same woman. What do such multiple infections mean, and how can they be analyzed in the context of epidemiological studies? In the earliest studies of HPV, multiple infections were seldom detected [2]. This is likely to be due to the characteristics of the earliest diagnostic tests. Some, such as Hybrid Capture 2 (HC2) (Qiagen, Gaithersburg, MD), do not distinguish individual HPV types, and the first polymerase chain reaction (PCR)–based assays included relatively few types other than HPV-16 and HPV18. In the past decade, however, PCRassays have become available that can detect broader ranges of oncogenic and nononcogenic HPV types (n . 40) and, most notably, have much higher analytical sensitivity (ie, ability to identify correctly HPV-positive women) [3, 4]. In a recent cross-wise comparison of 4 widely used PCR assays among women who had previously tested positive by HC2, the rate of detection of HPV infection ranged from 62.6% to 91.4%, and the proportion of multiple infections among all HPV-positive tests ranged from 24.8% to 52.6% [5]. This shows that there can be substantial disagreement even between high-quality PCR assays. The concordance between tests greatly increased, however, with increasing viral load and when the presence of cervical intra-epithelial neoplasia 2 or worse (CIN21) rather than HPV infection was set as an end point (clinical sensitivity) [5], [6]. There is no doubt that high clinical sensitivity and specificity are the most important features for an HPV test to help determine follow-up and treatment of HPV-positive women. Conversely, the use of a test with very high analytical sensitivity may be undesirable in cervical cancer screening programs, because it could lead to overdetection of infections that have no clinical consequences [6]. The shift toward highly sensitive PCR assays has, however, affected HPV findings in the most recent epidemiological studies in a way that cannot be ignored. Notably, (1) HPV infections are apparently more widespread than had previously been suggested,(2) coinfection with multiple types is very common in sexually active women and even more common than infection with single types among immunosuppressed human immunodeficiency virus (HIV)–infected women [7], and (3) the fraction of multiple HPV infections found in invasive cervical carcinomas has greatly increased during the past 2 decades, from 4.0% to 15.7% [8], despite the fact that this is a monoclonal disease that should have been caused by 1 HPV type only. The increasing frequency of detection of multiple HPV infections is, at present, complicating the estimate of the benefits of tests and vaccines including more or less broad ranges of HPV types [8], [3], [9], but it has also raised great interest in the biological implications of coinfections [10–14]. Much of the research on multiple HPV infections addresses 2 distinct questions. The first question considers the effects of infections with distinct HPV types on each other, either in terms of concurrent infections or sequential infections. The second question concerns the impact of multiple-type infections on the risk of future cervical disease, which, in screened populations, means CIN21. Both questions are addressed by Chaturvedi et al [1] in a study of 2478 HPV-positive young women, of whom 43.2% were infected with multiple HPV types. Received 29 October 2010; accepted 1 November 2010. Potential conflicts of interest: none reported. Reprints or correspondence: Silvia Franceschi, MD, International Agency for Research on Cancer, 150 cours Albert Thomas, 69372 Lyon cedex 08, France (franceschiS @iarc.fr). The Journal of Infectious Diseases 2011;203:891–3 The Author 2011. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com 1537-6613/2011/2037-0001$15.00 DOI: 10.1093/infdis/jiq146