Abstract
I've often told students that we understand the aetiology of cervical cancer better than almost any other cancer. As Walboomers and colleagues (J Pathol 1999;189:12–9) concluded: human papillomavirus (HPV) is a necessary cause. It is therefore refreshing to be challenged to question whether there might in fact be two types of cervical cancer with distinct aetiologies (Seppä et al. BJOG 2015; doi: 10.1111/1471-0528.13754). It is accepted that squamous cell carcinoma of the vulva can develop in one of two ways, but that has not seriously been postulated for cervical cancer since the role of HPV was identified in the late 1980s. Nevertheless it is still possible that the cofactors that allow cervical cancer to develop in a young woman soon after HPV infection are distinct from those associated with cancer evolving slowly over several decades. Vulval, like cervical, cancer has an unusual age incidence curve, but it also has two distinct types (Trimble et al. Obstet Gynecol 1996;87:59–64). Cases in younger women tend to be associated with HPV infection (and not p53 mutations); basaloid or warty in appearance; and often develop within a field of vulval intraepithelial neoplasia (VIN). Cancers in older women (over 55 years) only rarely have HPV DNA present. These keratinising cancers are more likely to be associated with lichen sclerosis than with VIN and often have p53 mutations. There are no such differences between cervical cancer in younger and older women. The epidemiological risk factors distinguishing women with cervical HPV infection but no cancer from those with cancer are the same at all ages (e.g. Deacon et al. Br J Cancer 2000;83:1565–72). Nevertheless, we (Sasieni et al. Br J Cancer 2003;89:88–93) and others have shown that cervical cancer in young women is less susceptible to prevention by screening than it is in older women, which supports the idea of cervical cancer being heterogeneous. Analysis of age-specific trends in cervical cancer is complicated by three factors: To account for these effects I would have applied the two-component model to birth-cohorts rather than to periods and explicitly modelled the impact of screening. It is impressive that the model fits the data in Figure 1 of the linked article so well; however, it has five parameters, so it is not altogether surprising that it is rich enough to approximate closely a wide variety of unimodal and bimodal curves. Efforts to control cervical cancer should remain focused on HPV vaccination of adolescents and screening of unvaccinated women aged 25–64. A better understanding of the aetiology of cervical cancer beyond HPV infection is important for cervical cancer control in women who were not vaccinated before being infected and will be particularly important for prevention of other HPV-related cancers, such as oropharyngeal cancers, which seem to have an altogether different age-profile. Full disclosure of interests available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call