Human Papillomavirus Testing Improves Follow-Up after Treatment for Cervical Intraepithelial Neoplasia

‘…our model showed that by adding HPV testing, the number of follow-up visits in the first 2 years after treatment may be safely reduced from three to two…’

Women with serious cervical abnormalities (high-grade cervical intraepithelial neoplasia [CIN]) are usually treated with cervix-preserving methods in order to minimize the effects on fertility. The most commonly used method is loop electrosurgical excision procedure. Although the currently available methods are generally effective in eradicating the lesion, treated women still have a 5–15% risk of recurrent CIN_[1,2]. Therefore, they should be followed more closely than women in the general population [3].

Follow-up protocols vary widely across countries and there is no consensus on the frequency and duration of follow-up. In the UK, the NHS guidelines (2004) state that women treated for high-grade CIN require 6- and 12-month follow-up cytology and annual cytology for the subsequent 9 years at least. Women who have a positive cytological result after treatment should receive a colposcopy within 12 months. After 10 years of follow-up, women are returned to routine screening and are invited for cytological testing once every 3 years. In the rest of Europe, a less aggressive follow-up is often advocated. For example, in The Netherlands, follow-up consists of only three cytological tests at 6, 12 and 24 months after treatment. After three negative cytological results, women are returned to routine screening. In the USA, guidelines are similar as in the UK, except that human papillomavirus (HPV) DNA testing is considered acceptable for surveillance during the first 6 months after treatment.

Although intensive follow-up protocols facilitate early detection of recurrent CIN, they also lead to patient discomfort and costs. To minimize the discomfort for women, it is desirable to keep the number of follow-up tests to a minimum. However, having a low number of follow-up tests is acceptable only if the chance of missing a recurrent lesion is low as well. We will advocate that these low-up.

It is well established that hgoals can be achieved by adding HPV DNA testing to current cytological foligh-risk HPV infection is a prerequisite for the development and progression of CIN lesions and cervical cancer [4]. HPV DNA can be detected in 95% of high-grade CIN and cervical cancer [5]. Testing for HPV DNA has therefore been suggested as a screening instrument in cervical cancer screening. Owing to its high sensivity, HPV DNA testing can also be used after treatment of a CIN lesion. In a meta-analysis of 11 studies in which HPV testing was done after conservative treatment for CIN, Paraskevaidis and colleagues showed that in four studies, all recurrent CIN lesions were preceded by an HPV-positive cervical smear [6]. They discovered that among women who did not develop recurrent CIN during the study period, only 16% had a positive post-treatment HPV test. Although the studies included in the meta-analyses demonstrated some heterogeneity, there was sufficient evidence to conclude that a positive HPV test may pick up recurrent CIN early and accurately, even in the presence of a normal cytological result. Four of the studies in the meta-analysis allowed for a direct comparison between cytology and HPV DNA testing. During the first year of follow-up, more women with recurrent CIN were identified by a positive HPV test than by a positive cytological result. Since some of the women with recurrent CIN presented with a negative HPV test, the authors advised intensive surveillance with cytology and HPV testing.

‘…a potentially adverse effect of combined cytology and HPV testing is an increase in costs and burden if the number of follow-up visits is not reduced.’

In our viewpoint, a potentially adverse effect of combined cytology and HPV testing is an increase in costs and burden if the number of follow-up visits is not reduced. Post-treatment HPV testing can only be successfully implemented when several post-treatment strategies have been evaluated with respect to a chance of detecting recurrent disease, patient burden and costs. Ideally, this should be done by carrying out a randomized, controlled trial.

‘…combining cytological follow-up with HPV testing led to reductions in the number of missed recurrent CIN lesions ranging from 32 to 77% as compared with current strategy.’

However, there are too many potentially feasible strategies with combined cytology and HPV testing. Moreover, as recurrent CIN lesions are relatively rare, a very large study of women treated for high-grade CIN would be needed. An alternative approach is to evaluate the value of different follow-up strategies with a mathematical simulation model. Simulation modeling has become a popular tool to gain insight into the effects and costs of an intervention as it enables us to make a quick comparison between several strategies. For the problem under consideration, the simulation population would consist of women treated for high-grade CIN who may develop recurrent CIN lesions. The chance of recurrent CIN for an individual woman in this simulation population should be estimated from real patient data if available. The simulation population would then be subjected to different follow-up strategies with cytology and HPV DNA testing. The health effects of each strategy could be measured by the number of recurrent diseases, the number of colposcopies and repeat visits, and the costs.

At our department, we developed a simulation model to compare the current Dutch follow-up protocol of cytological testing at 6, 12 and 24 months with six plausible HPV-testing strategies. The HPV-testing strategies consisted of either two follow-up visits or only one follow-up visit instead of the current three visits. We found that combining cytological follow-up with HPV testing led to reductions in the number of missed recurrent CIN lesions ranging from 32 to 77% as compared with current strategy. The improved detection of recurrent CIN could be achieved in combination with a decrease in the number of follow-up visits and without an increase in colposcopical examinations. The low patient burden of the strategies with HPV testing was reflected in the costs, which were lower for HPV-testing strategies than for the current follow-up strategy.

Interestingly, we found that the number of unnecessary colposcopies, in other words, examinations that do not lead to detection of high-grade CIN, was lower if the 6-month follow-up visit consisted of HPV testing alone than if the 6-month follow-up visit consisted of both cytology and HPV testing. This probably results from the established difficulty in interpreting cervical smears in recently treated women [7]. Integrating this observation with results regarding detection of post-treatment CIN and costs, we considered a strategy consisting of HPV testing alone at 6 months and both HPV and cytological testing at 24 months particularly interesting.

Basically, our model showed that by adding HPV testing, the number of follow-up visits in the first 2 years after treatment may be safely reduced from three to two, that is, the 12-month follow-up visit can be omitted. For the Dutch situation, such an adapted follow-up protocol would lead to relatively small reductions in patient discomfort and costs, because current follow-up is already quite restrictive. However, for the UK the benefits are expected to be much larger. As our model predicted a very low percentage of missed CIN lesions for strategies with combined HPV and cytological testing, annual cytological follow-up after three negative cytological results in the UK could be replaced by postponing retesting to 3 years after double negative test results at 6 and 24 months. This would lower the number of follow-up visits in the 10 years after treatment from 12 to five.

To summarize the evidence, the value of adding HPV testing to current follow-up protocols based on cytology has been shown by several investigators [2,6,8].

‘…HPV genotyping before treatment may contribute to a better assessment of the risk of recurrent CIN.’

In addition to previous clinical evidence, the results of our modeling study demonstrate that HPV testing is also justified from an economic perspective: implementation can be achieved without increasing costs or patient burden by reducing the number of repeat tests from three to two, provided that only HPV testing is performed at the first follow-up visit.

To further improve the follow-up of women treated for CIN, it would be worthwhile to focus future research on the differentiation of treated women according to risk for recurrent CIN. If women can be classified accurately into high- and low-risk groups, follow-up can be more focused, in other words, more intensive follow-up for high-risk women and possibly less frequent follow-up for low-risk women. Some work on risk differentiation has already been done by Flanelly et al. [9]. Flanelly and colleagues found that women could be classified as low risk if they were younger than 50 years and if the excision of the CIN lesion was complete. They suggested less intensive follow-up for these women and more intensive follow-up for women not satisfying these criteria. Another possibility for risk differentiation is provided by HPV genotyping. HPV genotyping is relevant because HPV16 and HPV18 are more oncogenic than other high-risk HPV types [10]. Gök and colleagues showed that recurrent CIN lesions were more prevalent in women infected with HPV16 before treatment [11]. Therefore, HPV genotyping before treatment may contribute to a better assessment of the risk of recurrent CIN. Follow-up protocols may be adapted accordingly.

To conclude, follow-up after treatment for CIN should be woman-friendly and safe. This involves minimizing the amount of diagnostic procedures while guaranteeing only a minimal chance of developing recurrent CIN. We believe that including HPV testing in the follow-up brings us closer to this goal.