| |
How might HPV testing be integrated into cervical screening?
|
| |
| |
The Lancet Oncology, Early Online Publication, 15 December 2011
In The Lancet Oncology, final data from the POBASCAM trial1 show the effectiveness of HPV testing for cervical cancer screening. At baseline, HPV testing identified 79 additional cervical precancers (cervical intraepithelial neoplasia [CIN] grade 3) per 100 000 women and 30 additional cancers per 100 000 women compared with screening using cytology only. During the subsequent 5 years, the improved detection of CIN grade 3 at baseline led to the detection of 24 fewer precancers per 100 000 women per year and prevented 10 cancers per 100 000 women per year.
Moreover, with an assumption that all women in the intervention group who tested negative for HPV and had a normal Pap test returned in 5 years on average, we estimate a very low cancer risk of 2·2 cancers per 100 000 women per year, which shows that 5 year screening intervals are safe.
POBASCAM reinforces findings from cohort studies,2, 3 clinical trials,4-6 and routine clinical practice7 by providing overwhelming evidence of the benefits of inclusion of HPV testing in screening programmes. However, clinical trials do not assess primary HPV testing in isolation, but in the context of a full protocol that determines management of all HPV-positive and HPV-negative women. Detailed understanding of the protocol is important to assess the feasibility of the introduction of HPV testing in different settings and to formulate guidelines for implementation.
The POBASCAM finding that is most immediately translatable to screening programmes worldwide is the 5 year screening interval for women 30 years and older who test HPV-negative with normal cytology. The low estimated cancer risk for these women in POBASCAM is much the same as the estimate from an analysis of clinical practice in the USA-3·2 per 100 000 women per year.7 These extremely low risks are consistent with the fact that HPV infection is the cause of nearly all cervical cancer and that cancer usually takes decades to develop.8 We expect that almost every woman who tests negative for HPV, irrespective of country or screening protocol, has an extremely low risk of cancer over 3 or 5 years.
How to manage HPV-positive women is less clear. Most HPV infections will clear naturally and only a minority of CIN grade 2 or 3 will progress to cancer. Thus, an effective protocol that determines which HPV-positive women should be referred to colposcopy is crucial to prevent unnecessary colposcopies and excisional procedures.9 The most aggressive protocol refers all HPV-positive women to colposcopy, as was done in the New Technologies for Cervical Cancer screening (NTCC) trial5 of women 35 years and older. POBASCAM used a more conservative protocol, referring women to colposcopy only when they had moderate dyskaryosis or worse cytology (equivalent to high-grade squamous intraepithelial lesion) at 6 months or moderate dyskaryosis or worse or persistent HPV infection at 18 months. An aggressive protocol has the advantage of potentially preventing more cancer, whereas a conservative protocol has the benefit of fewer diagnoses of intermediate endpoints that might require unnecessary interventions.
In POBASCAM, ten cancers were prevented, with only 32 additional CIN grade 2 or 3 diagnosed (table). HPV-positive, Pap-negative women had eight of the ten excess cancers, but also 26 of the 32 additional CIN grade 2 or 3, confirming that HPV testing identifies Pap-negative women at risk of cancer who are challenging to manage. Conservative management of HPV-positive women in POBASCAM resulted in a smaller increase in CIN grade 2 detection, and a greater increase in CIN grade 3 to grade 2 ratio than in NTCC, probably because of regression of transient CIN grade 2 lesions not destined to progress to cancer. Despite the use of a more aggressive protocol, NTCC prevented fewer cancers than did POBASCAM (nine in total, but in more than twice the number of women as POBASCAM), because NTCC had a lower baseline cancer rate in the cytology arm than did POBASCAM (19 per 100 000 vs 30 per 100 000). When a disease is rare, fewer cases can be prevented, and prevention of them might need a very aggressive protocol to catch rare fast-progressing cancers, resulting in more overtreatment.10
The POBASCAM trial shows that 5 year screening intervals are safe, and that conservative management of HPV-positive women can control excess CIN grade 2 or 3 while preventing cervical cancer. However, how the POBASCAM protocol would perform in other populations that have different baseline cancer rates, compliance, and management infrastructure, is unclear. Clinical management based on a woman's individual risk of cervical cancer might account for population-specific factors that can have a substantial effect on the performance of screening protocols.11
The authors declare that they have no conflicts of interest. HAK and NW are supported by the Intramural Research Program of the National Cancer Institute, National Institute of Health. The views expressed do not represent the views of the US National Cancer Institute, the National Institutes of Health, the Department of Health and Human Services, or the US Government.
| |
| |
| |
|
|
|
