Loops in the natural history of cervical cancer

Few human cancers are as well understood as cancer of the uterine cervix. When Harald zur Hausen received the 2008 Nobel Prize in Physiology or Medicine in honour of his research on the causal role of human papillomavirus (HPV) in the genesis of this tumour, the natural history of cervical cancer seemed to be deciphered, apart from small pieces of the puzzle. As a result, cervical cancer research shifted towards next-generation vaccines, therapies, and best screening strategies, and basic researchers had a new focus on the role of HPV in other malignancies. In The Lancet Oncology, Xavier Castellsague and coworkers 1 report on one of the missing pieces of the puzzle of cervical cancer epidemiology, the role of intrauterine devices (IUDs). Their surprising results should spark new research and might lead to revision of our current understanding of the genesis of cervical cancer. The analysis of ten case–control studies of cervical cancer and 16 HPV prevalence surveys from four continents was carefully done, and fi ndings were corrected for confounding eff ects such as number of Papanicolaou (Pap) smears. The study found that IUD use did not aff ect risk of HPV infection, but was associated with a signifi cantly lower risk for cervical cancer. At fi rst glance, the main implication of this fi nding is that it provides a high level of evidence to contradict a widespread assumption that IUD use increases the risk of cervical cancer. However, restored confi dence in the safety of IUDs is really just a secondary result. The protective eff ect of IUD use challenges some key elements in the current model of the natural history of cervical cancer. The type of IUD used was not documented in this analysis, but if the protective eff ect was related to gestagens, copper, or infl ammation, there should be a positive correlation between protective eff ect and duration of use. Instead, the risk reduction was similar for IUD use of a few months up to 9 years. The best explanation for this observation is that the process of loop insertion, rather than the type of loop or duration of IUD use, is associated with a reduced risk of cervical cancer. Castellsague and co-workers 1 propose that complete destruction of precancerous lesions by loop insertion is a theoretical explanation for the eff ect. But in clinical practice, even low-grade cervical intraepithelial neoplasia (CIN) show a complex and often multifocal distribution within individually shaped transformation zones. To remove CIN lesions with clear margins, usually the complete transformation zone needs to be excised with tissue loss measured in cubic centimetres. The superfi cial trauma associated with loop insertion might occasionally result in complete destruction of tiny CIN lesions located in the external ostium of the cervical canal, but as a colposcopist, I doubt that this mechanical eff ect can explain the observed signifi cant reduction in cancer risk. Instead, I postulate that the tissue trauma associated with loop insertion induces a cellular immune response that might fi nally clear persistent HPV infections and preinvasive lesions. Small cohort studies have already shown that taking cervical biopsies increases the spontaneous regression rates of CIN, and that the residual dysplastic tissue remaining after surgical treatment is cleared in many cases. Also, the importance of the cellular immune response in determining the clinical course of CIN has been well known for decades. 2