- Email: [email protected]
HPV testing in cervical screening
Best Practice & Research Clinical Obstetrics and Gynaecology Vol. 20, No. 2, pp. 253–266, 2006 doi:10.1016/j.bpobgyn.2005.10.009 available online at http://www.sciencedirect.com
3 HPV testing in cervical screening Antoinette A.T.P. Brink Peter J.F. Snijders Chris J.L.M. Meijer* Department of Pathology, VU University Medical Center, P.O. Box 7057,1007 MB Amsterdam, The Netherlands
Johannes Berkhof Department of Clinical Epidemiology and Biostatistics, VU University Medical Center, Amsterdam, The Netherlands
Rene´ H.M. Verheijen Department of Obstetrics and Gynaecology, VU University Medical Center, P.O. Box 7057,1007 MB Amsterdam, The Netherlands
High-risk human papillomavirus (hrHPV) bearing cervical intraepithelial neoplasia (CIN) is considered, as the real precursor lesion of cervical cancer and persistence of an hrHPV infection is necessary for the progression to cervical cancer. This knowledge warrants the use of hrHPV testing as an adjunct to cervical cytology in population-based screening programmes and for monitoring therapy efficacy of high-grade CIN lesions. Replacement of cytology by hrHPV testing altogether is considered, but for this to be (cost-) effective, accurate information about the specificity of the hrHPV test is required. Additional test systems that can be used to stratify women with a positive hrHPV test are HPV genotyping, viral load analysis and hrHPV mRNA analysis. The need for HPV genotyping of cervical smears is illustrated by the increased risk for high-grade cervical lesions associated with HPV types 16 and 18. In particular, for women who have normal but persistently (O1 year) HPV18-positive smears, endocervical curettage is suggested (evidently considering the age and possible future pregnancies of the respective woman) because HPV18 is associated with glandular lesions in the cervix, which are difficult to detect by cytology. Key words: cervical cancer; cervical intraepithelial neoplasia; cytology; HPV; human papillomavirus; population-based screening.
* Corresponding author. Tel.: C31 20 444 4070; fax: C31 20 444 2964. E-mail address: [email protected] (C.J.L.M. Meijer).
1521-6934/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.
254 A. A. T. P. Brink et al
High-risk human papillomavirus (hrHPV) bearing cervical intraepithelial neoplasia (CIN) is now considered as the real precursor lesion of cervical cancer and persistence of an hrHPV infection is necessary for the progression to cervical cancer.1 This wellestablished knowledge has led to the development of diagnostic applications for hrHPV testing that will be discussed in this chapter. In addition, we will elaborate on more recent findings in hrHPV research and possible improvements of hrHPV diagnostics derived from these.
HRHPV TESTS CURRENTLY IN USE The most widely used hrHPV testing methods include the commercially available Hybrid Capture 2 (hc2)2 and polymerase chain reaction (PCR)-based methods. Hc2 detects 13 genital hrHPV types by a mixture of full-length RNA probes. Hybridization of one or more of the probes to hrHPV DNA present in heat-alkaline-denatured clinical samples is detected by peroxidase-labelled antibodies that recognize the RNA/DNA hybrid and are visualized by chemiluminescence. Some cross-reactivity of the hc2 probes with HPV types not represented in the probe mix, including some nononcogenic HPVs, has been described.3 When cervical scrapes have been collected in a medium supplied by the hc2 manufacturer, hc2 can be applied directly. Cervical scrapes collected in cytological preservation media require some adaptation before the hc2 test. PCR-based methods frequently rely on the use of consensus or multiplex primers that amplify a broad spectrum of HPV types. Examples of the former are the MY09/114 and GP5C/6C5 systems, an example of the latter is SPF10.6 PCR products can be detected with a cocktail of type-specific probes, for example in an enzymeimmunoassay (EIA).7 HPV typing can be done by, for example, reverse hybridization systems such as reverse line blotting (RLB)8,4 or line-probe assay (LiPa).9 In addition, real-time PCR methods have been developed for quantitation of HPV DNA10,11, but these have a low multiplicity for different hrHPV types and are therefore, not suitable as a high-throughput primary screening tool. PCR methods can be applied to cervical scrapes collected in phosphate-buffered saline1, in cytological preservation media (generally after extraction of DNA) and even on archival smears.12
HRHPV TESTING AS AN ADJUNCT TO CERVICAL CYTOLOGY The current screening protocols for cervical cancer are based on the ability to cytologically detect precursor lesions of cervical cancer (the ‘Pap’ test13). When identified in good time, these lesions can be treated successfully and with only minor side effects. However, the sensitivity and specificity of cytology are not optimal, resulting, respectively in missed cases of high-grade CIN and over-referral to the gynaecologist with many redundant follow-up smears for low-grade CIN. Given the causal relation between a persistent hrHPV infection and the development of highgrade CIN and cervical cancer, hrHPV testing has been advocated in addition to cytology. hrHPV testing is thought to improve the screening algorithms based on the detection of abnormal cervical cells for cervical cancer14, the management of women with cytologically equivocal smears15,16 and the management of women treated for high-grade CIN.17
HPV testing in cervical screening 255
The peak incidence of hrHPV infections in women occurs between 20 and 25 years of age and most of these infections are cleared spontaneously.18 Because the time span between infection and the development of an invasive lesion is at least 8 years, screening seems warranted among women aged 30 years and over. Thus, it seems safe that, in the Dutch population-based screening programme, women aged 30–60 are invited to participate. However, if this peak prevalence of hrHPV is earlier in life, as is the case in Greenland19, then cervical screening should be started earlier. hrHPV testing as a quality control in primary screening Cervical cytology has a high rate of false negativity. For example, we reported that for a group of 48 women with cervical cancer and a preceding smear reported as normal, blind cytological revision of those smears showed 40 of them to be abnormal.20 Most of those smears were hrHPV positive. In fact, the far majority of archival smears taken from women up to 15 years before they developed cervical cancer were already hrHPV positive.21,22 Furthermore, in a routine screening setting, rescreening of hrHPV positive cervical smears previously read as ‘normal’ resulted in 5–7% reclassification to abnormal, as compared to 0.5–1% of hrHPV-negative normal smears.23 hrHPV-positive women with abnormal cytology are at particular risk of having or developing CIN3 or cervical carcinoma.23–25 These data indicate that rescreening of hrHPV-positive cytologically normal smears would lead to a reduction of false-negative results, and that the combination of hrHPV testing and cervical cytology in cervical cancer screening increases the screening sensitivity.14,26 In addition, adenocarcinoma (AdCa) and its precursors, which—just like squamous cell carcinomas—are nearly always hrHPV positive24, are frequently missed by cytology and hrHPV testing is very helpful to detect these lesions.23–25,27 Hence, for future screening algorithms we suggest yearly follow-up of women with hrHPV-positive, indisputably normal smears, because these women have a highly increased risk of developing CIN3 and cervical cancer compared with women with hrHPV-negative normal smears.28,29 In case cytological abnormalities occur, these women are referred for colposcopy. If smears remain cytologically normal but hrHPV positive for 2 years, we suggest colposcopy-directed biopsy. If then no abnormalities are seen in the transformation zone, endocervical curettage could be performed. The increased costs associated with this alternative-screening algorithm can be overcome by increasing the screening interval for hrHPV-negative women with normal cytology26 because their risk for CIN3 or worse is much less than that of women with normal cytology and an unknown hrHPV status.30 hrHPV testing in the management of women with cytologically equivocal smears In the Dutch and other European classification systems, borderline and mild dyskaryosis (BMD) is comparable to atypical squamous cells of undetermined significance (ASC-US), atypical squamous cells cannot exclude high-grade squamous intraepithelial lesion (LASC-H) and low-grade squamous intraepithelial neoplasm (LSIL) according to the 2001 Bethesda system.31,32 In population-based cervical cancer screening, triage of women with cytological readings of BMD is based on repeat smears at 6 and 18 months. Women who have hrHPV-positive BMD are at risk for underlying or incipient CIN3, whereas this risk is negligible for women with hrHPV-negative
256 A. A. T. P. Brink et al
BMD.15,33,34 Therefore, we propose that women with an hrHPV-negative BMD smear have their subsequent smear in the next screening round. By contrast, women with hrHPV-positive BMD should either be referred to the gynaecologist or have a repeat smear at 6 months.15,16 The main profit of such a regime would be a marked reduction of follow-up smears: 65% of the women with BMD are hrHPV negative and these women do not need any further follow-up until the next screening round, after 3–5 years in the different European countries. This approach is safe and cost-effective.35 hrHPV testing in the management of women treated for high-grade CIN In most European countries and the USA, the follow-up of women treated for highgrade CIN consists of repeat cytology at 6, 12 and 24 months post-treatment. When, after the last abnormal smear, three subsequent smears at 6, 12 and 24 months are cytologically normal, women are discharged from follow-up. The number of follow-up smears can be reduced by using a combination of hrHPV testing and cytology. This combination has a very high sensitivity and negative predictive value for CIN3 and cervical cancer, and has shown to be superior to other follow-up strategies. A safe algorithm would be to test women at 6 months and to repeat women with a double negative test only once for a final test at 24 months. Women found positive for either test should be kept under gynaecological surveillance until they are negative for both tests.36–39
HRHPV TYPE AND CERVICAL CANCER SCREENING In terms of HPV type, the test should ideally allow detection of the 15 most common cancer-associated hrHPV types (i.e. HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68 and 73).40 However, it has been suggested that the impact of adding additional hrHPV types to the thirteen hrHPV types that many HPV tests have in common in their cocktails, would be small and probably irrelevant for screening programmes.41 Moreover, adding a less frequent hrHPV type increases the sensitivity for detecting CIN3 or carcinoma at the costs of a substantial decrease in specificity. This would leave clinicians with the problem of follow-up of many more women with an hrHPV infection but the number of high-grade lesions detected would only be slightly increased.42 However, there is growing evidence that, in addition to hrHPV testing, further hrHPV genotyping is necessary because certain hrHPV types confer increased risks for highgrade CIN and cervical carcinomas. HPV type and the risk of high-grade cervical lesions Within the framework of the so-called POBASCAM study26, a detailed cross-sectional analysis was done of the hrHPV type distribution in cervical scrapes obtained at baseline in relation to cytology and underlying histologically confirmedRCIN2.43 Among women with a cytological reading of moderate dyskaryosis or worse and underlying CIN2 or worse, HPV types 16 and 33 were more prevalent than in women with normal cytology. A particular hrHPV type distribution was also seen in the carcinoma stage: cervical squamous cell carcinomas (SCC) showed a strongly increased prevalence of HPV types 16 and a slightly increased prevalence of HPV18 as compared to normal
HPV testing in cervical screening 257
cytology. By contrast, a strongly increased prevalence of HPV18 and a slightly increased prevalence of HPV16 were seen in AdCa and adenocarcinomas in situ (ACIS).44 Several explanations can be given for the observation that the prevalence of HPV18 was strongly increased in invasive carcinoma but not in moderate dyskaryosis and the underlying CIN2 or worse in the POBASCAM study. HPV18 infections can be associated with rapid progression of the lesion thus lacking an intermediate phase of BMD. Alternatively, HPV18 can be associated with a cytopathological effect high in the endocervical canal that is likely to be missed by cytology until at least CIN3/ACIS has developed. This latter hypothesis is supported by an 18-month prospective analysis of hrHPV-positive women in the POBASCAM study with normal cytology or BMD at baseline.45 Both HPV16 and HPV18 were associated with risk of high-grade CIN after normal cytology at baseline but, unlike HPV16, HPV18 was not associated with risk of high-grade CIN after BMD. Considering the short follow-up period, it is plausible that at least some of the HPV18-positive CIN lesions after a baseline normal smear were already developed but missed by cytology. This finding is consistent with an increased risk posed by HPV16 and HPV18 for cervical (pre)cancers of the squamous and adenohistotypes as shown in a prospective screening cohort of 20 810 women followed for up to 10 years46, and an increased risk for cervical precancer posed by HPV16 but not by HPV18 in a prospective trial of women with equivocal or mild cervical abnormalities.47 Several factors might explain the increased prevalence of certain hrHPV types in high-grade CIN and cervical carcinomas. Viral factors might play a role, including the innate ability to persist that is shown by the hrHPV type.48,49 Moreover, certain hrHPV types might more easily undergo a switch from a productive infection to a transformed state as a result of deregulated expression of the viral oncogenes E6 and E7.50 In this case, other non-HPV16 or HPV18 types, such as HPV45, HPV59, HPV66 and HPV68, would result in productive infections yielding cytopathological effects, i.e. CIN1 and partly CIN2, which are easily detected in a Pap smear, resulting in early referral to the gynaecologist for colposcopically directed biopsies. Host factors are likely to contribute as well. For example, certain alleles of the polymorph human histocompatibility leukocyte antigen (HLA) class II are associated with an increased susceptibility to certain hrHPV types51–53, which might even be reflected in an increased viral load.51,53 These findings strongly suggest that specific clinical measures should be taken with respect to certain hrHPV types. In particular, women who are positive for HPV16 or HPV18 should be monitored very closely, even if their smears are cytologically normal, because these two hrHPV types are associated with a strongly increased risk of highgrade CIN and are prevalent in cervical cancer. Even more important, HPV18 seems to be the main hrHPV type capable of causing severe lesions arising from columnar epithelial cells. Because these lesions reside in the endocervical part of the cervix, they are difficult to detect by cytology alone. Therefore, when the transformation zone is found to be normal by colposcopy endocervical curettage is increasingly advocated for women who remain HPV18 positive but cytologically normal for at least a year, to exclude the presence of putatively underlying ACIS or AdCa. HPV typing and CIN3 post-treatment monitoring In general, HPV typing of CIN3 and cervical cancer is not necessary because the treatment does not depend on the typing result. Hence, for clinical use an hrHPV test indicating presence or absence of hrHPV is sufficient. However, additional HPV typing
258 A. A. T. P. Brink et al
has some small advantage and perhaps some prognostic value. Demonstrating the same hrHPV type in the post-treatment specimen as in the primary CIN3 lesion might indicate a recurrent hrHPV infection, which suggests that the woman apparently cannot clear this particular hrHPV type51,53 and might require more intense follow up. In this context, differences in risk of CIN3/cervical carcinoma conferred by different hrHPV types seem important. The highest risk of CIN3/cervical carcinoma is posed by HPV16 and it would be interesting to evaluate whether HPV16—and also perhaps HPV18—are over-represented in recurrent or persistent CIN3 compared to primary CIN3.30,54 However, until now no systematic analysis of hrHPV types in primary and posttreatment CIN3 has been published.
HRHPV TESTING IN SELF-SAMPLED VAGINAL MATERIAL FROM NONRESPONDERS OF THE CERVICAL SCREENING PROGRAMME The compliance rate of current population-based cervical screening programmes is not optimal. In the Netherlands55, as well as in the UK and Scandinavian countries56–58, 30% of the invited women do not participate in the cervical screening programme; half of the subsequent cervical cancers are diagnosed in this group of women. A user-friendly, self-sampling method for collecting representative cervical cell material at home would lower the threshold for women to participate in the screening. Self-sampled vaginal material is not representative for the cytological state of the cervix but it is highly representative for the HPV status of the cervix as tested by PCR59 and hc2.60–63 Recently, we conducted a study among more than 2500 women who, even after a recall, had not responded to the invitation for screening.64 When offered the possibility for self-sampling, over 30% of these women responded actively by sending in a selfobtained sample to the test laboratory. Altogether, 1.6% of the women with a valid selfsample test had underlyingRCIN2. This prevalence is significantly higher than in the historical cohort of women participating in the POBASCAM trial, where 1% had an underlyingRCIN2 lesion.26 These findings strengthen the hypothesis that nonresponders of the cervical screening programme are a risk group59,61–63 and indicate that hrHPV testing on self-sampled vaginal material is a powerful means of increasing cervical cancer screening efficacy.
CAN HRHPV TESTING REPLACE CYTOLOGY? As described above, cytological analysis of self-sampled vaginal material is not feasible and hence this type of material requires hrHPV testing instead of cytology.61,65 However, even for samples on which cytological analysis is possible, preliminary data indicate that the results of an objective hrHPV test are much better than those based on a subjective reading by a cytopathologist.15 It is thought that a small percentage of high-grade CIN is missed using hrHPV testing only (see above). However, recent data indicate that the false-negative rate of hrHPV testing is lower than that of cytology.30,66 A remaining problem is the suboptimal positive predictive value of the current hrHPV tests for high-grade CIN lesions, but the positive predictive value can be boosted by stratifying hrHPV-positive women in several ways, as discussed below.
HPV testing in cervical screening 259
Nonetheless, for an accurate assessment of the effects of replacing cytology by hrHPV testing, we will have to wait until the results of ongoing large trials such as the TOMBOLA67, HART15 and POBASCAM26 studies become available. Considerations regarding hrHPV test sensitivity Replacement of cytology by hrHPV testing calls for an hrHPV test that has a sensitivity approaching 100% to detect high-grade CIN and cervical cancer. Unfortunately, the currently available hrHPV tests miss approximately 1% of such lesions.36,37 For example, most consensus primer PCR systems are based on amplification of a region within the hrHPV L1 gene4–6, which might be lost on viral integration. This would not be a problem for hc2, which uses whole-genome probes, but the lower analytical sensitivity of hc2 precludes detection of lesions with low viral loads. The ideal hrHPV test used in primary screening should combine the high analytical sensitivity of PCR assays with the possibility of detecting all configurations of the hrHPV genome, including integrated genomes. Because the intact E6/E7 region of the hrHPV genome is necessary for cellular transformation68 and is invariably retained upon integration69, a PCR system detecting this region would be the method of choice. Considerations regarding hrHPV test specificity It is self-evident that an hrHPV test should correctly identify all women who have highgrade lesions or are at risk for developing these, and preferably give negative test results in women who are not at risk. A previous critical review on HPV testing suggests that this demands a tight balance between the analytical sensitivity (i.e. the minimal amount of hrHPV DNA necessary to obtain a positive test result) and clinical specificity (i.e. the proportion of women who test positive and who are truly at risk of high-grade lesions) of a given HPV test.70 The most widely used test systems have a high analytical sensitivity so as not to miss any high-grade lesions, but they also detect clinically irrelevant transient hrHPV infections and, consequently, have limited clinical specificity. Hence, a complete replacement of cytology by a very sensitive hrHPV test method would result in a substantial increase in referrals and repeat smears. In the Dutch screening population, 3.4% of women have abnormal cytology and 5.0% are hrHPV positive. Because even women with hrHPV-positive normal cytology have a 210 times greater risk of developing CIN3 compared to women with hrHPV-negative normal cytology71, all 5.0% hrHPV-positive women need follow-up. However, 70% of the hrHPV-positive women have normal cytology and only 8% of those women with normal cytology will actually develop CIN3.28 This calls for an hrHPV test with increased clinical specificity, or for additional stratification of women who have a positive hrHPV test. In the past few years, several putative test systems have been studied. These are discussed below. Viral load analysis An accurate estimate of the viral genome copy number per cervical scrape can be made by real-time PCR.72–75 Several studies using such quantitative hrHPV detection have shown a positive correlation between the amount of hrHPV DNA (the ‘viral load’) in cervical smears and the risk of high-grade cervical lesions.72–74,76 This is true for most of the hrHPV types studied, but the odds ratios seem to be the highest for HPV type
260 A. A. T. P. Brink et al
16.75,76 This is in line with the above-mentioned finding that this particular hrHPV type confers a higher risk of high-grade cervical lesions than other hrHPV types. The relationship between viral load and the risk of high-grade cervical lesions holds for both women with normal cytology and women with BMD. However, only a relative risk can be calculated from these findings, whereas the large variation in viral loads between cervical smears does not permit clear-cut individual risk assessment. Considering HPV16 in women with BMD smears, we tentatively proposed referral to a gynaecologist for women with more than 4.3!106 HPV16 genome copies per smear because they are at increased risk of high-grade CIN, although a lower viral load does not necessarily exclude progressive CIN disease and would at least warrant a repeat smear.74 The advantage of such a policy would be a decrease in the number of repeat smears in the group of women with high viral loads, because they can be sent for colposcopy-directed biopsy. In addition to HPV16, we developed type-specific real-time PCRs for HPV types 18, 31 and 33 to define viral load cut-off points at the 25th, 33th and 50th percentiles. Although we observed a substantial overlap between HPV DNA load levels in GP5C /6CPCR-positive women with and without CIN3, all women with CIN3 had viral load levels above the 25th percentile threshold of women with normal cytology harbouring the respective hrHPV type. This indicates that, among women who test hrHPV positive by GP5C/6CPCR, viral load assessment will identify 25% of those without clinically relevant hrHPV infections (Hogewoning et al, unpublished data). Cut-off points defined by real-time PCR methods depend on the PCR method used and might vary between laboratories using the same PCR method. They might also depend on the screening population, as younger women may show increased viral loads in productive infections that will be cleared within a few months.18 As well as real-time PCR, viral load can be estimated from signals obtained in hc2 analysis by comparison of sample signals to positive control signals (‘relative light units’; RLU). This yields a semiquantitative result.77 Gravitt et al78 showed that the standard cut-off point of 1 pg viral DNA/ml sample (approximately 105 viral genome copies) differentiated cytologically normal women from women with LSIL or worse. However, among women with abnormal cytology, the viral load did not change as lesion severity increased. The predictive value for hc2 viral load measurement for high-grade CIN lesions is still a matter of debate, differences between studies probably being caused by different sampling methods and different study cohorts. At present, it is thought that high viral load as measured by hc2 is—if at all—only a short-term marker of the risk of CIN3 or worse.77,79,80 Differences in predictive value between hc2 and real-time PCR might, apart from differences in study design, be caused by the fact that real-time PCR is HPV type specific whereas hc2 detects a range of 13 hrHPV types simultaneously, giving rise to viral load signals that might be composites of multiple infections. In addition, real-time PCR—but not hc2—allows quantitation of the cells in the sample, thus enabling better standardization of the measurement. Although viral load measurement by hc2 apparently is not suitable (yet) for risk assessment, its specificity might be enhanced by increasing the cut-off point.81 hrHPV mRNA analysis hrHPV DNA detection methods such as PCR and hc2 are unlikely to distinguish active, persistent and progressive infections from regressive infections. Even in regressive infections, hrHPV DNA might be present in the cervical smear, for example contained within apoptotic cells or cellular debris. A probable hallmark of
HPV testing in cervical screening 261
an active, persistent or progressive hrHPV infection is transcription of the viral oncogenes E6 and E7, because these are essential for transformation of the infected cells.82 In transformed cells, the E6 and E7 proteins disrupt the cell cycle control by interacting with p53 and pRb, respectively.83,84 Hence, transcriptional analysis of E6/E7 might be useful for risk stratification of hrHPVinfected women. Indeed, only in a small proportion of hrHPV DNA-positive cervical smears E6/E7 mRNA can be detected.85,86 Several studies have addressed the clinical value of this finding. Molden et al have shown that, among women with BMD cytology, hrHPV mRNA positivity confers a relative risk of 70 for developing high-grade CIN as compared to hrHPV mRNA-negative women.86 Moreover, hrHPV mRNA positivity appeared to be related to persistence of the infection87, which is a prerequisite for the development of cervical cancer and its precursors. These data suggest that hrHPV mRNA analysis is a valuable tool in cervical cancer screening. At present, the only commercially available hrHPV mRNA detection system is a real-time nucleic acid sequence based amplification (NASBA) method detecting HPV types 16, 18, 31, 33 and 45 (NorChip AS, Norway). Whether this limited number of hrHPV types is sufficient to detect all women at risk for highgrade lesions remains to be investigated.
POSSIBLE ALGORITHMS FOR REPLACEMENT OF CYTOLOGY BY HRHPV TESTING If cytology as a primary screening method is to be replaced by hrHPV testing, the screening algorithm should be sufficiently sensitive to identify all women at risk for high-grade cervical lesions, yet sufficiently specific to avoid unnecessary referral and repeat smears. At present, no hrHPV test is available that fulfils both these conditions. However, we are obtaining more and more data that show that hrHPV testing is a better screening tool than cytology. Hence, it seems logical to subject all women participating in cervical screening to a sensitive hrHPV test first to make sure no high-grade lesions are missed. All women who test negative need no further follow-up until the next screening round (5 years or maybe more; see above), because their chance of having CIN3 or worse is very low.30 Because a sensitive hrHPV detection method will classify women with a transient hrHPV infection to be at risk for high-grade lesions, the specificity of the screening algorithm should be enhanced by an additional test before making decisions about referral and/or repeat smears. Reflex cytological analysis of all hrHPV-positive smears is a logical next step, because all women who have hrHPV-positive cytology classified as borderline dyskaryosis or worse are at an increased risk of high-grade CIN and should be referred for colposcopy. Women with cytologically normal hrHPV-positive smears are subjected to additional tests. First, their smears should be genotyped. This is necessary for additional tests such as viral load analysis by real-time PCR or hrHPV mRNA analysis. Genotyping will distinguish HPV16 and HPV18 infections that require a status aparte and seem to warrant more aggressive clinical management than other hrHPV types. Women who are infected by either of these two types could be referred to the gynaecologist if after 1 year, hrHPV is still present and their cytology is normal. Women with low viral loads and/or absence of viral mRNA could probably be referred to the next screening round, but these strategies require further confirmation.
262 A. A. T. P. Brink et al
Practice points † hrHPV testing is a useful tool in population-based cervical cancer screening— as a quality control for cervical cytology and for the triage of women with cytologically equivocal smears—and for monitoring therapy efficacy of highgrade CIN lesions † among hrHPV-positive women, those infected with HPV16 or HPV18 will need extra follow-up because they are at an increased risk of cervical cancer † endocervical curettage is warranted for women who have had an HPV18positive but cytologically normal smear for more than 1 year, because they are at an increased risk for high-grade glandular lesions of the cervix, which might be missed by cytology
Research agenda † † † † †
prospective trials for the clinical validation of hrHPV mRNA analysis hPV viral load analysis self-sampling feasibility of displacement of cytology by hrHPV testing
REFERENCES 1. Nobbenhuis MA, Walboomers JM, Helmerhorst TJ et al. Relation of human papillomavirus status to cervical lesions and consequences for cervical-cancer screening: a prospective study. Lancet 1999; 354: 20–25. 2. Clavel C, Masure M, Putaud I et al. Hybrid capture II, a new sensitive test for human papillomavirus detection. Comparison with hybrid capture I and PCR results in cervical lesions. Journal of Clinical Pathology 1998; 51: 737–740. 3. Castle PE, Schiffman M, Burk RD et al. Restricted cross-reactivity of hybrid capture 2 with nononcogenic human papillomavirus types. Cancer Epidemiology, Biomarkers & Prevention 2002; 11: 1394–1399. 4. Gravitt PE, Peyton CL, Apple RJ & Wheeler CM. Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by a single-hybridization, reverse line blot detection method. Journal of Clinical Microbiology 1998; 36: 3020–3027. 5. de Roda Husman AM, Walboomers JM, van den Brule AJ et al. The use of general primers GP5 and GP6 elongated at their 3 0 ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. The Journal of General Virology 1995; 76(Pt 4): 1057–1062. 6. Kleter B, van Doorn LJ, ter Schegget J et al. Novel short-fragment PCR assay for highly sensitive broadspectrum detection of anogenital human papillomaviruses. The American Journal of Pathology 1998;. 7. Jacobs MV, Snijders PJ, van den Brule AJ et al. A general primer GP5C/GP6(C)-mediated PCR-enzyme immunoassay method for rapid detection of 14 high-risk and 6 low-risk human papillomavirus genotypes in cervical scrapings. Journal of Clinical Microbiology 1997; 35: 791–795. 8. van den Brule AJ, Pol R, Fransen-Daalmeijer N et al. GP5C/6C PCR followed by reverse line blot analysis enables rapid and high-throughput identification of human papillomavirus genotypes. Journal of Clinical Microbiology 2002; 40: 779–787. 9. Kleter B, van Doorn LJ, Schrauwen L et al. Development and clinical evaluation of a highly sensitive PCRreverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. Journal of Clinical Microbiology 1999; 37: 2508–2517.
HPV testing in cervical screening 263 10. Josefsson A, Livak K & Gyllensten U. Detection and quantitation of human papillomavirus by using the fluorescent 5 0 exonuclease assay. Journal of Clinical Microbiology 1999; 37: 490–496. 11. Szuhai K, Sandhaus E, Kolkman-Uljee SM et al. A novel strategy for HPV detection and genotyping with SybrGreen and Molecular Beacon PCR. The American Journal of Pathology 2001; 159: 1651–1660. 12. Roda Husman AM, Snijders PJ, Stel HVet al. Processing of long-stored archival cervical smears for human papillomavirus detection by the polymerase chain reaction. British Journal of Cancer 1995; 72: 412–417. 13. Papanicolao GN & Traut HF. The diagnostic value of vaginal smears in carcinoma of the uterus. American Journal of Obstetrics and Gynecology 1941; 42: 193–206. 14. Lorincz AT & Richart RM. Human papillomavirus DNA testing as an adjunct to cytology in cervical screening programs. Archives of Pathology & Laboratory Medicine 2003; 127: 959–968. 15. Cuzick J, Szarewski A, Cubie H et al. Management of women who test positive for high-risk types of human papillomavirus: the HART study. Lancet 2003; 362: 1871–1876. 16. Bais AG, Rebolj M, Snijders PJ et al. Triage using HPV-testing in persistent borderline and mildly dyskaryotic smears: proposal for new guidelines. International Journal of Cancer 2005; 116: 122–129. 17. Zielinski GD, Bais AG, Helmerhorst TJ et al. HPV testing and monitoring of women after treatment of CIN 3: review of the literature and meta-analysis. Obstetrical & Gynecological Survey 2004; 59: 543–553. 18. Evander M, Edlund K, Gustafsson A et al. Human papillomavirus infection is transient in young women: a population-based cohort study. The Journal of Infectious Diseases 1995; 171: 1026–1030. 19. Svare EI, Kjaer SK, Worm AM et al. Risk factors for HPV infection in women from sexually transmitted disease clinics: comparison between two areas with different cervical cancer incidence. International Journal of Cancer 1998; 75: 1–8. 20. Zielinski GD, Snijders PJ, Rozendaal L et al. HPV presence precedes abnormal cytology in women developing cervical cancer and signals false negative smears. British Journal of Cancer 2001; 85: 398–404. 21. Walboomers JM, Husman AM, Snijders PJ et al. Human papillomavirus in false negative archival cervical smears: implications for screening for cervical cancer. Journal of Clinical Pathology 1995; 48: 728–732. 22. Wallin KL, Wiklund F, Angstrom T et al. Type-specific persistence of human papillomavirus DNA before the development of invasive cervical cancer. The New England Journal of Medicine 1999; 341: 1633–1638. 23. Meijer CJ, Rozendaal L, van der Linden HC et al. Human papillomavirus testing for primary cervical cancer screening 1997;338–347. 24. Zielinski GD, Snijders PJ, Rozendaal L et al. The presence of high-risk HPV combined with specific p53 and p16INK4a expression patterns points to high-risk HPV as the main causative agent for adenocarcinoma in situ and adenocarcinoma of the cervix. The Journal of Pathology 2003; 201: 535–543. 25. Krane JF, Granter SR, Trask CE et al. Papanicolaou smear sensitivity for the detection of adenocarcinoma of the cervix: a study of 49 cases. Cancer 2001; 93: 8–15. 26. Bulkmans NWJ, Rozendaal L, Snijders PJ et al. POBASCAM, a population-based randomised controlled trial for implementation of high-risk HPV testing in cervical screening; design, methods and baseline data of 44,102 women. International Journal of Cancer 2003; 110: 94–101. 27. Pirog EC, Kleter B, Olgac S et al. Prevalence of human papillomavirus DNA in different histological subtypes of cervical adenocarcinoma. The American Journal of Pathology 2000; 157: 1055–1062. 28. Rozendaal L, Walboomers JM, van der Linden JC et al. PCR-based high-risk HPV test in cervical cancer screening gives objective risk assessment of women with cytomorphologically normal cervical smears. International Journal of Cancer 1996; 68: 766–769. 29. Bosch FX, Lorincz A, Munoz N et al. The causal relation between human papillomavirus and cervical cancer. Journal of Clinical Pathology 2002; 55: 244–265. 30. Bulkmans NW, Rozendaal L, Voorhorst FJ et al. Long-term protective effect of high-risk human papillomavirus testing in population-based cervical screening. British Journal of Cancer 2005; 92: 1800– 1802. 31. Bulk S, van Kemenade FJ, Rozendaal L & Meijer CJ. The Dutch CISOE—a framework for cytology reporting increases efficacy of screening upon standardisation since 1996. Journal of Clinical Pathology 2004; 57: 388–393. 32. Solomon D, Davey D, Kurman R et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. Journal of the American Medical Association 2002; 287: 2114–2119. 33. Zielinski GD, Snijders PJ, Rozendaal L et al. High-risk HPV testing in women with borderline and mild dyskaryosis: long-term follow-up data and clinical relevance. The Journal of Pathology 2001; 195: 300–306.
264 A. A. T. P. Brink et al 34. Fait G, Kupferminc MJ, Daniel Y et al. Contribution of human papillomavirus testing by hybrid capture in the triage of women with repeated abnormal pap smears before colposcopy referral. Gynecologic Oncology 2000; 79: 177–180. 35. Berkhof J, de Bruijne MC, Zielinski GD et al. Evaluation of cervical screening strategies with adjunct highrisk human papillomavirus testing for women with borderline or mild dyskaryosis. International Journal of Cancer. In press. 36. Zielinski GD, Rozendaal L, Voorhorst FJ et al. HPV testing can reduce the number of follow-up visits in women treated for cervical intraepithelial neoplasia grade 3. Gynecologic Oncology 2003; 91: 67–73. 37. Nobbenhuis MA, Meijer CJ, van den Brule AJ et al. Addition of high-risk HPV testing improves the current guidelines on follow-up after treatment for cervical intraepithelial neoplasia. British Journal of Cancer 2001; 84: 796–801. 38. Kim JJ, Wright TC & Goldie SJ. Cost-effectiveness of human papillomavirus DNA testing in the United Kingdom, The Netherlands, France, and Italy. Journal of the National Cancer Institute 2005; 97: 888– 895. 39. Paraskevaidis E, Arbyn M, Sotiriadis A et al. The role of HPV DNA testing in the follow-up period after treatment for CIN: a systematic review of the literature. Cancer Treatment Reviews 2004; 30: 205–211. 40. Munoz N, Bosch FX, de Sanjose S et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. The New England Journal of Medicine 2003; 348: 518–527. 41. Munoz N, Bosch FX, Castellsague X et al. Against which human papillomavirus types shall we vaccinate and screen? The international perspective. International Journal of Cancer 2004; 111: 278–285. 42. Schiffman M, Khan MJ, Solomon D et al. A study of the impact of adding HPV types to cervical cancer screening and triage tests. Journal of the National Cancer Institute 2005; 97: 147–150. 43. Bulkmans NW, Bleeker MC, Berkhof J et al. Prevalence of types 16 and 33 is increased in high-risk human papillomavirus positive women with cervical intraepithelial neoplasia grade 2 or worse. International Journal of Cancer 2005;. 44. Bulk S, Bulkmans NW, Zielinski GD et al. Cervical adenocarcinoma in situ and invasive adenocarcinoma have an increased prevalence of HPV18 only, whereas squamous cell carcinoma is associated with both HPV16 and HPV18. Submitted for publication. 45. Berkhof J, de Bruijne MC, Zielinski GD et al. Evaluation of cervical screening strategies with adjunct highrisk human papillomavirus testing for women with borderline or mild dyskaryosis. International Journal of Cancer 2005; October 10 [Epub ahead of print]. 46. Khan MJ, Castle PE, Lorincz AT et al. The elevated 10-year risk of cervical precancer and cancer in women with human papillomavirus (HPV) type 16 or 18 and the possible utility of type-specific HPV testing in clinical practice. Journal of the National Cancer Institute 2005; 97: 1072–1079. 47. Castle PE, Solomon D, Schiffman M & Wheeler CM. Human papillomavirus type 16 infections and 2-year absolute risk of cervical precancer in women with equivocal or mild cytologic abnormalities. Journal of the National Cancer Institute 2005; 97: 1066–1071. 48. Schiffman M, Herrero R, Desalle R et al. The carcinogenicity of human papillomavirus types reflects viral evolution. Virology 2005; 337: 76–84. 49. Richardson H, Kelsall G, Tellier P et al. The natural history of type-specific human papillomavirus infections in female university students. Cancer Epidemiology, Biomarkers & Prevention 2003; 12: 485–490. 50. Steenbergen RD, de Wilde J, Wilting SM et al. HPV-mediated transformation of the anogenital tract. Journal of Clinical Virology 2005; 32(supplement 1): S25–S33. 51. Beskow AH, Moberg M & Gyllensten UB. HLA class II allele control of HPV load in carcinoma in situ of the cervix uteri. International Journal of Cancer 2005;. 52. Beskow AH, Josefsson AM & Gyllensten UB. HLA class II alleles associated with infection by HPV16 in cervical cancer in situ. International Journal of Cancer 2001; 93: 817–822. 53. Beskow AH & Gyllensten UB. Host genetic control of HPV 16 titer in carcinoma in situ of the cervix uteri. International Journal of Cancer 2002; 101: 526–531. 54. Clifford GM, Smith JS, Aguado T & Franceschi S. Comparison of HPV type distribution in high-grade cervical lesions and cervical cancer: a meta-analysis. British Journal of Cancer 2003; 89: 101–105. 55. van Ballegooijen M, Rebolj M, Meerding WJ et al. De praktijk van het bevolkingsonderzoek naar baarmoederhalskanker in Nederland in 2001. 2003.
HPV testing in cervical screening 265 56. Peto J, Gilham C, Fletcher O & Matthews FE. The cervical cancer epidemic that screening has prevented in the UK. Lancet 2004; 364: 249–256. 57. Sawaya GF & Grimes DA. New technologies in cervical cytology screening: a word of caution. Obstetrics and Gynecology 1999; 94: 307–310. 58. Sasieni PD, Cuzick J & Lynch-Farmery E. Estimating the efficacy of screening by auditing smear histories of women with and without cervical cancer. The national co-ordinating network for cervical screening working group. British Journal of Cancer 1996; 73: 1001–1005. 59. Nobbenhuis MA, Helmerhorst TJ, van den Brule AJ et al. Primary screening for high risk HPV by home obtained cervicovaginal lavage is an alternative screening tool for unscreened women. Journal of Clinical Pathology 2002; 55: 435–439. 60. Sellors JW, Lorincz AT, Mahony JB et al. Comparison of self-collected vaginal, vulvar and urine samples with physician-collected cervical samples for human papillomavirus testing to detect high-grade squamous intraepithelial lesions. Canadian Medical Association Journal 2000; 163: 513–518. 61. Hillemanns P, Kimmig R, Huttemann U et al. Screening for cervical neoplasia by self-assessment for human papillomavirus DNA. Lancet 1999; 354: 1970. 62. Dannecker C, Siebert U, Thaler CJ et al. Primary cervical cancer screening by self-sampling of human papillomavirus DNA in internal medicine outpatient clinics. Annals of Oncology 2004; 15: 863–869. 63. Wright Jr. TC, Denny L, Kuhn L et al. HPV DNA testing of self-collected vaginal samples compared with cytologic screening to detect cervical cancer. Journal of American Medical Association 2000; 283: 81–86. 64. Bais AG, Rebolj M, Snijders PJ et al. Triage using HPV-testing in persistent borderline and mildly dyskaryotic smears: proposal for new guidelines. International Journal of Cancer 2005; 116(1): 122–129. 65. Brink AATP, Meijer CJ, Wiegerinck MAHM et al. High concordance in HPV test results between cervical samples taken by a novel self-sampler and a cytobrush. In preparation. 66. Solomon D, Schiffman M & Tarone R. Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: baseline results from a randomized trial. Journal of the National Cancer Institute 2001; 93: 293–299. 67. Rebello G, Hallam N, Smart G et al. Human papillomavirus testing and the management of women with mildly abnormal cervical smears: an observational study. British Medical Journal 2001; 322: 893–894. 68. Hawley-Nelson P, Vousden KH, Hubbert NL et al. HPV16 E6 and E7 proteins cooperate to immortalize human foreskin keratinocytes. EMBO Journal 1989; 8: 3905–3910. 69. Choo KB, Pan CC & Han SH. Integration of human papillomavirus type 16 into cellular DNA of cervical carcinoma: preferential deletion of the E2 gene and invariable retention of the long control region and the E6/E7 open reading frames. Virology 1987; 161: 259–261. 70. Snijders PJ, van den Brule AJ & Meijer CJ. The clinical relevance of human papillomavirus testing: relationship between analytical and clinical sensitivity. The Journal of Pathology 2003; 201: 1–6. 71. Rozendaal L, Westerga J, van der Linden JC et al. PCR based high risk HPV testing is superior to neural network based screening for predicting incident CIN III in women with normal cytology and borderline changes. Journal of Clinical Pathology 2000; 53: 606–611. 72. Ylitalo N, Sorensen P, Josefsson AM et al. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: a nested case-control study. Lancet 2000; 355: 2194–2198. 73. Josefsson AM, Magnusson PK, Ylitalo N et al. Viral load of human papilloma virus 16 as a determinant for development of cervical carcinoma in situ: a nested case-control study. Lancet 2000; 355: 2189–2193. 74. van Duin M, Snijders PJ, Schrijnemakers HF et al. Human papillomavirus 16 load in normal and abnormal cervical scrapes: an indicator of CIN II/III and viral clearance. International Journal of Cancer 2002; 98: 590– 595. 75. Moberg M, Gustavsson I & Gyllensten U. Real-time PCR-based system for simultaneous quantification of human papillomavirus types associated with high risk of cervical cancer. Journal of Clinical Microbiology 2003; 41: 3221–3228. 76. Moberg M, Gustavsson I & Gyllensten U. Type-specific associations of human papillomavirus load with risk of developing cervical carcinoma in situ. International Journal of Cancer 2004; 112: 854–859. 77. Lorincz AT, Castle PE, Sherman ME et al. Viral load of human papillomavirus and risk of CIN3 or cervical cancer. Lancet 2002; 360: 228–229.
266 A. A. T. P. Brink et al 78. Gravitt PE, Burk RD, Lorincz A et al. A comparison between real-time polymerase chain reaction and hybrid capture 2 for human papillomavirus DNA quantitation. Cancer Epidemiology, Biomarkers & Prevention 2003; 12: 477–484. 79. Dalstein V, Riethmuller D, Pretet JL et al. Persistence and load of high-risk HPV are predictors for development of high-grade cervical lesions: a longitudinal French cohort study. International Journal of Cancer 2003; 106: 396–403. 80. Castle PE, Schiffman M & Wheeler CM. Hybrid capture 2 viral load and the 2-year cumulative risk of cervical intraepithelial neoplasia grade 3 or cancer. American Journal of Obstetrics and Gynecology 2004; 191: 1590–1597. 81. Howard M, Sellors J & Kaczorowski J. Optimizing the hybrid capture II human papillomavirus test to detect cervical intraepithelial neoplasia. Obstetrics and Gynecology 2002; 100: 972–980. 82. Munger K, Phelps WC, Bubb V et al. The E6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes. Journal of Virology 1989; 63: 4417–4421. 83. Scheffner M, Werness BA, Huibregtse JM et al. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 1990; 63: 1129–1136. 84. Dyson N, Howley PM, Munger K & Harlow E. The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science 1989; 243: 934–937. 85. Molden T, Kraus I, Karlsen F et al. Comparison of human papillomavirus messenger RNA and DNA detection: a cross-sectional study of 4,136 women O30 years of age with a 2-year follow-up of high-grade squamous intraepithelial lesion. Cancer Epidemiology, Biomarkers & Prevention 2005; 14: 367–372. 86. Molden T, Nygard JF, Kraus I et al. Predicting CIN2C when detecting HPV mRNA and DNA by PreTect HPV-proofer and consensus PCR: a 2-year follow-up of women with ASCUS or LSIL Pap smear. International Journal of Cancer 2005; 114: 973–976. 87. Cuschieri KS, Whitley MJ & Cubie HA. Human papillomavirus type specific DNA and RNA persistenceimplications for cervical disease progression and monitoring. Journal of Medical Virology 2004; 73: 65–70.
3 HPV testing in cervical screening Antoinette A.T.P. Brink Peter J.F. Snijders Chris J.L.M. Meijer* Department of Pathology, VU University Medical Center, P.O. Box 7057,1007 MB Amsterdam, The Netherlands
Johannes Berkhof Department of Clinical Epidemiology and Biostatistics, VU University Medical Center, Amsterdam, The Netherlands
Rene´ H.M. Verheijen Department of Obstetrics and Gynaecology, VU University Medical Center, P.O. Box 7057,1007 MB Amsterdam, The Netherlands
High-risk human papillomavirus (hrHPV) bearing cervical intraepithelial neoplasia (CIN) is considered, as the real precursor lesion of cervical cancer and persistence of an hrHPV infection is necessary for the progression to cervical cancer. This knowledge warrants the use of hrHPV testing as an adjunct to cervical cytology in population-based screening programmes and for monitoring therapy efficacy of high-grade CIN lesions. Replacement of cytology by hrHPV testing altogether is considered, but for this to be (cost-) effective, accurate information about the specificity of the hrHPV test is required. Additional test systems that can be used to stratify women with a positive hrHPV test are HPV genotyping, viral load analysis and hrHPV mRNA analysis. The need for HPV genotyping of cervical smears is illustrated by the increased risk for high-grade cervical lesions associated with HPV types 16 and 18. In particular, for women who have normal but persistently (O1 year) HPV18-positive smears, endocervical curettage is suggested (evidently considering the age and possible future pregnancies of the respective woman) because HPV18 is associated with glandular lesions in the cervix, which are difficult to detect by cytology. Key words: cervical cancer; cervical intraepithelial neoplasia; cytology; HPV; human papillomavirus; population-based screening.
* Corresponding author. Tel.: C31 20 444 4070; fax: C31 20 444 2964. E-mail address: [email protected] (C.J.L.M. Meijer).
1521-6934/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.
254 A. A. T. P. Brink et al
High-risk human papillomavirus (hrHPV) bearing cervical intraepithelial neoplasia (CIN) is now considered as the real precursor lesion of cervical cancer and persistence of an hrHPV infection is necessary for the progression to cervical cancer.1 This wellestablished knowledge has led to the development of diagnostic applications for hrHPV testing that will be discussed in this chapter. In addition, we will elaborate on more recent findings in hrHPV research and possible improvements of hrHPV diagnostics derived from these.
HRHPV TESTS CURRENTLY IN USE The most widely used hrHPV testing methods include the commercially available Hybrid Capture 2 (hc2)2 and polymerase chain reaction (PCR)-based methods. Hc2 detects 13 genital hrHPV types by a mixture of full-length RNA probes. Hybridization of one or more of the probes to hrHPV DNA present in heat-alkaline-denatured clinical samples is detected by peroxidase-labelled antibodies that recognize the RNA/DNA hybrid and are visualized by chemiluminescence. Some cross-reactivity of the hc2 probes with HPV types not represented in the probe mix, including some nononcogenic HPVs, has been described.3 When cervical scrapes have been collected in a medium supplied by the hc2 manufacturer, hc2 can be applied directly. Cervical scrapes collected in cytological preservation media require some adaptation before the hc2 test. PCR-based methods frequently rely on the use of consensus or multiplex primers that amplify a broad spectrum of HPV types. Examples of the former are the MY09/114 and GP5C/6C5 systems, an example of the latter is SPF10.6 PCR products can be detected with a cocktail of type-specific probes, for example in an enzymeimmunoassay (EIA).7 HPV typing can be done by, for example, reverse hybridization systems such as reverse line blotting (RLB)8,4 or line-probe assay (LiPa).9 In addition, real-time PCR methods have been developed for quantitation of HPV DNA10,11, but these have a low multiplicity for different hrHPV types and are therefore, not suitable as a high-throughput primary screening tool. PCR methods can be applied to cervical scrapes collected in phosphate-buffered saline1, in cytological preservation media (generally after extraction of DNA) and even on archival smears.12
HRHPV TESTING AS AN ADJUNCT TO CERVICAL CYTOLOGY The current screening protocols for cervical cancer are based on the ability to cytologically detect precursor lesions of cervical cancer (the ‘Pap’ test13). When identified in good time, these lesions can be treated successfully and with only minor side effects. However, the sensitivity and specificity of cytology are not optimal, resulting, respectively in missed cases of high-grade CIN and over-referral to the gynaecologist with many redundant follow-up smears for low-grade CIN. Given the causal relation between a persistent hrHPV infection and the development of highgrade CIN and cervical cancer, hrHPV testing has been advocated in addition to cytology. hrHPV testing is thought to improve the screening algorithms based on the detection of abnormal cervical cells for cervical cancer14, the management of women with cytologically equivocal smears15,16 and the management of women treated for high-grade CIN.17
HPV testing in cervical screening 255
The peak incidence of hrHPV infections in women occurs between 20 and 25 years of age and most of these infections are cleared spontaneously.18 Because the time span between infection and the development of an invasive lesion is at least 8 years, screening seems warranted among women aged 30 years and over. Thus, it seems safe that, in the Dutch population-based screening programme, women aged 30–60 are invited to participate. However, if this peak prevalence of hrHPV is earlier in life, as is the case in Greenland19, then cervical screening should be started earlier. hrHPV testing as a quality control in primary screening Cervical cytology has a high rate of false negativity. For example, we reported that for a group of 48 women with cervical cancer and a preceding smear reported as normal, blind cytological revision of those smears showed 40 of them to be abnormal.20 Most of those smears were hrHPV positive. In fact, the far majority of archival smears taken from women up to 15 years before they developed cervical cancer were already hrHPV positive.21,22 Furthermore, in a routine screening setting, rescreening of hrHPV positive cervical smears previously read as ‘normal’ resulted in 5–7% reclassification to abnormal, as compared to 0.5–1% of hrHPV-negative normal smears.23 hrHPV-positive women with abnormal cytology are at particular risk of having or developing CIN3 or cervical carcinoma.23–25 These data indicate that rescreening of hrHPV-positive cytologically normal smears would lead to a reduction of false-negative results, and that the combination of hrHPV testing and cervical cytology in cervical cancer screening increases the screening sensitivity.14,26 In addition, adenocarcinoma (AdCa) and its precursors, which—just like squamous cell carcinomas—are nearly always hrHPV positive24, are frequently missed by cytology and hrHPV testing is very helpful to detect these lesions.23–25,27 Hence, for future screening algorithms we suggest yearly follow-up of women with hrHPV-positive, indisputably normal smears, because these women have a highly increased risk of developing CIN3 and cervical cancer compared with women with hrHPV-negative normal smears.28,29 In case cytological abnormalities occur, these women are referred for colposcopy. If smears remain cytologically normal but hrHPV positive for 2 years, we suggest colposcopy-directed biopsy. If then no abnormalities are seen in the transformation zone, endocervical curettage could be performed. The increased costs associated with this alternative-screening algorithm can be overcome by increasing the screening interval for hrHPV-negative women with normal cytology26 because their risk for CIN3 or worse is much less than that of women with normal cytology and an unknown hrHPV status.30 hrHPV testing in the management of women with cytologically equivocal smears In the Dutch and other European classification systems, borderline and mild dyskaryosis (BMD) is comparable to atypical squamous cells of undetermined significance (ASC-US), atypical squamous cells cannot exclude high-grade squamous intraepithelial lesion (LASC-H) and low-grade squamous intraepithelial neoplasm (LSIL) according to the 2001 Bethesda system.31,32 In population-based cervical cancer screening, triage of women with cytological readings of BMD is based on repeat smears at 6 and 18 months. Women who have hrHPV-positive BMD are at risk for underlying or incipient CIN3, whereas this risk is negligible for women with hrHPV-negative
256 A. A. T. P. Brink et al
BMD.15,33,34 Therefore, we propose that women with an hrHPV-negative BMD smear have their subsequent smear in the next screening round. By contrast, women with hrHPV-positive BMD should either be referred to the gynaecologist or have a repeat smear at 6 months.15,16 The main profit of such a regime would be a marked reduction of follow-up smears: 65% of the women with BMD are hrHPV negative and these women do not need any further follow-up until the next screening round, after 3–5 years in the different European countries. This approach is safe and cost-effective.35 hrHPV testing in the management of women treated for high-grade CIN In most European countries and the USA, the follow-up of women treated for highgrade CIN consists of repeat cytology at 6, 12 and 24 months post-treatment. When, after the last abnormal smear, three subsequent smears at 6, 12 and 24 months are cytologically normal, women are discharged from follow-up. The number of follow-up smears can be reduced by using a combination of hrHPV testing and cytology. This combination has a very high sensitivity and negative predictive value for CIN3 and cervical cancer, and has shown to be superior to other follow-up strategies. A safe algorithm would be to test women at 6 months and to repeat women with a double negative test only once for a final test at 24 months. Women found positive for either test should be kept under gynaecological surveillance until they are negative for both tests.36–39
HRHPV TYPE AND CERVICAL CANCER SCREENING In terms of HPV type, the test should ideally allow detection of the 15 most common cancer-associated hrHPV types (i.e. HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68 and 73).40 However, it has been suggested that the impact of adding additional hrHPV types to the thirteen hrHPV types that many HPV tests have in common in their cocktails, would be small and probably irrelevant for screening programmes.41 Moreover, adding a less frequent hrHPV type increases the sensitivity for detecting CIN3 or carcinoma at the costs of a substantial decrease in specificity. This would leave clinicians with the problem of follow-up of many more women with an hrHPV infection but the number of high-grade lesions detected would only be slightly increased.42 However, there is growing evidence that, in addition to hrHPV testing, further hrHPV genotyping is necessary because certain hrHPV types confer increased risks for highgrade CIN and cervical carcinomas. HPV type and the risk of high-grade cervical lesions Within the framework of the so-called POBASCAM study26, a detailed cross-sectional analysis was done of the hrHPV type distribution in cervical scrapes obtained at baseline in relation to cytology and underlying histologically confirmedRCIN2.43 Among women with a cytological reading of moderate dyskaryosis or worse and underlying CIN2 or worse, HPV types 16 and 33 were more prevalent than in women with normal cytology. A particular hrHPV type distribution was also seen in the carcinoma stage: cervical squamous cell carcinomas (SCC) showed a strongly increased prevalence of HPV types 16 and a slightly increased prevalence of HPV18 as compared to normal
HPV testing in cervical screening 257
cytology. By contrast, a strongly increased prevalence of HPV18 and a slightly increased prevalence of HPV16 were seen in AdCa and adenocarcinomas in situ (ACIS).44 Several explanations can be given for the observation that the prevalence of HPV18 was strongly increased in invasive carcinoma but not in moderate dyskaryosis and the underlying CIN2 or worse in the POBASCAM study. HPV18 infections can be associated with rapid progression of the lesion thus lacking an intermediate phase of BMD. Alternatively, HPV18 can be associated with a cytopathological effect high in the endocervical canal that is likely to be missed by cytology until at least CIN3/ACIS has developed. This latter hypothesis is supported by an 18-month prospective analysis of hrHPV-positive women in the POBASCAM study with normal cytology or BMD at baseline.45 Both HPV16 and HPV18 were associated with risk of high-grade CIN after normal cytology at baseline but, unlike HPV16, HPV18 was not associated with risk of high-grade CIN after BMD. Considering the short follow-up period, it is plausible that at least some of the HPV18-positive CIN lesions after a baseline normal smear were already developed but missed by cytology. This finding is consistent with an increased risk posed by HPV16 and HPV18 for cervical (pre)cancers of the squamous and adenohistotypes as shown in a prospective screening cohort of 20 810 women followed for up to 10 years46, and an increased risk for cervical precancer posed by HPV16 but not by HPV18 in a prospective trial of women with equivocal or mild cervical abnormalities.47 Several factors might explain the increased prevalence of certain hrHPV types in high-grade CIN and cervical carcinomas. Viral factors might play a role, including the innate ability to persist that is shown by the hrHPV type.48,49 Moreover, certain hrHPV types might more easily undergo a switch from a productive infection to a transformed state as a result of deregulated expression of the viral oncogenes E6 and E7.50 In this case, other non-HPV16 or HPV18 types, such as HPV45, HPV59, HPV66 and HPV68, would result in productive infections yielding cytopathological effects, i.e. CIN1 and partly CIN2, which are easily detected in a Pap smear, resulting in early referral to the gynaecologist for colposcopically directed biopsies. Host factors are likely to contribute as well. For example, certain alleles of the polymorph human histocompatibility leukocyte antigen (HLA) class II are associated with an increased susceptibility to certain hrHPV types51–53, which might even be reflected in an increased viral load.51,53 These findings strongly suggest that specific clinical measures should be taken with respect to certain hrHPV types. In particular, women who are positive for HPV16 or HPV18 should be monitored very closely, even if their smears are cytologically normal, because these two hrHPV types are associated with a strongly increased risk of highgrade CIN and are prevalent in cervical cancer. Even more important, HPV18 seems to be the main hrHPV type capable of causing severe lesions arising from columnar epithelial cells. Because these lesions reside in the endocervical part of the cervix, they are difficult to detect by cytology alone. Therefore, when the transformation zone is found to be normal by colposcopy endocervical curettage is increasingly advocated for women who remain HPV18 positive but cytologically normal for at least a year, to exclude the presence of putatively underlying ACIS or AdCa. HPV typing and CIN3 post-treatment monitoring In general, HPV typing of CIN3 and cervical cancer is not necessary because the treatment does not depend on the typing result. Hence, for clinical use an hrHPV test indicating presence or absence of hrHPV is sufficient. However, additional HPV typing
258 A. A. T. P. Brink et al
has some small advantage and perhaps some prognostic value. Demonstrating the same hrHPV type in the post-treatment specimen as in the primary CIN3 lesion might indicate a recurrent hrHPV infection, which suggests that the woman apparently cannot clear this particular hrHPV type51,53 and might require more intense follow up. In this context, differences in risk of CIN3/cervical carcinoma conferred by different hrHPV types seem important. The highest risk of CIN3/cervical carcinoma is posed by HPV16 and it would be interesting to evaluate whether HPV16—and also perhaps HPV18—are over-represented in recurrent or persistent CIN3 compared to primary CIN3.30,54 However, until now no systematic analysis of hrHPV types in primary and posttreatment CIN3 has been published.
HRHPV TESTING IN SELF-SAMPLED VAGINAL MATERIAL FROM NONRESPONDERS OF THE CERVICAL SCREENING PROGRAMME The compliance rate of current population-based cervical screening programmes is not optimal. In the Netherlands55, as well as in the UK and Scandinavian countries56–58, 30% of the invited women do not participate in the cervical screening programme; half of the subsequent cervical cancers are diagnosed in this group of women. A user-friendly, self-sampling method for collecting representative cervical cell material at home would lower the threshold for women to participate in the screening. Self-sampled vaginal material is not representative for the cytological state of the cervix but it is highly representative for the HPV status of the cervix as tested by PCR59 and hc2.60–63 Recently, we conducted a study among more than 2500 women who, even after a recall, had not responded to the invitation for screening.64 When offered the possibility for self-sampling, over 30% of these women responded actively by sending in a selfobtained sample to the test laboratory. Altogether, 1.6% of the women with a valid selfsample test had underlyingRCIN2. This prevalence is significantly higher than in the historical cohort of women participating in the POBASCAM trial, where 1% had an underlyingRCIN2 lesion.26 These findings strengthen the hypothesis that nonresponders of the cervical screening programme are a risk group59,61–63 and indicate that hrHPV testing on self-sampled vaginal material is a powerful means of increasing cervical cancer screening efficacy.
CAN HRHPV TESTING REPLACE CYTOLOGY? As described above, cytological analysis of self-sampled vaginal material is not feasible and hence this type of material requires hrHPV testing instead of cytology.61,65 However, even for samples on which cytological analysis is possible, preliminary data indicate that the results of an objective hrHPV test are much better than those based on a subjective reading by a cytopathologist.15 It is thought that a small percentage of high-grade CIN is missed using hrHPV testing only (see above). However, recent data indicate that the false-negative rate of hrHPV testing is lower than that of cytology.30,66 A remaining problem is the suboptimal positive predictive value of the current hrHPV tests for high-grade CIN lesions, but the positive predictive value can be boosted by stratifying hrHPV-positive women in several ways, as discussed below.
HPV testing in cervical screening 259
Nonetheless, for an accurate assessment of the effects of replacing cytology by hrHPV testing, we will have to wait until the results of ongoing large trials such as the TOMBOLA67, HART15 and POBASCAM26 studies become available. Considerations regarding hrHPV test sensitivity Replacement of cytology by hrHPV testing calls for an hrHPV test that has a sensitivity approaching 100% to detect high-grade CIN and cervical cancer. Unfortunately, the currently available hrHPV tests miss approximately 1% of such lesions.36,37 For example, most consensus primer PCR systems are based on amplification of a region within the hrHPV L1 gene4–6, which might be lost on viral integration. This would not be a problem for hc2, which uses whole-genome probes, but the lower analytical sensitivity of hc2 precludes detection of lesions with low viral loads. The ideal hrHPV test used in primary screening should combine the high analytical sensitivity of PCR assays with the possibility of detecting all configurations of the hrHPV genome, including integrated genomes. Because the intact E6/E7 region of the hrHPV genome is necessary for cellular transformation68 and is invariably retained upon integration69, a PCR system detecting this region would be the method of choice. Considerations regarding hrHPV test specificity It is self-evident that an hrHPV test should correctly identify all women who have highgrade lesions or are at risk for developing these, and preferably give negative test results in women who are not at risk. A previous critical review on HPV testing suggests that this demands a tight balance between the analytical sensitivity (i.e. the minimal amount of hrHPV DNA necessary to obtain a positive test result) and clinical specificity (i.e. the proportion of women who test positive and who are truly at risk of high-grade lesions) of a given HPV test.70 The most widely used test systems have a high analytical sensitivity so as not to miss any high-grade lesions, but they also detect clinically irrelevant transient hrHPV infections and, consequently, have limited clinical specificity. Hence, a complete replacement of cytology by a very sensitive hrHPV test method would result in a substantial increase in referrals and repeat smears. In the Dutch screening population, 3.4% of women have abnormal cytology and 5.0% are hrHPV positive. Because even women with hrHPV-positive normal cytology have a 210 times greater risk of developing CIN3 compared to women with hrHPV-negative normal cytology71, all 5.0% hrHPV-positive women need follow-up. However, 70% of the hrHPV-positive women have normal cytology and only 8% of those women with normal cytology will actually develop CIN3.28 This calls for an hrHPV test with increased clinical specificity, or for additional stratification of women who have a positive hrHPV test. In the past few years, several putative test systems have been studied. These are discussed below. Viral load analysis An accurate estimate of the viral genome copy number per cervical scrape can be made by real-time PCR.72–75 Several studies using such quantitative hrHPV detection have shown a positive correlation between the amount of hrHPV DNA (the ‘viral load’) in cervical smears and the risk of high-grade cervical lesions.72–74,76 This is true for most of the hrHPV types studied, but the odds ratios seem to be the highest for HPV type
260 A. A. T. P. Brink et al
16.75,76 This is in line with the above-mentioned finding that this particular hrHPV type confers a higher risk of high-grade cervical lesions than other hrHPV types. The relationship between viral load and the risk of high-grade cervical lesions holds for both women with normal cytology and women with BMD. However, only a relative risk can be calculated from these findings, whereas the large variation in viral loads between cervical smears does not permit clear-cut individual risk assessment. Considering HPV16 in women with BMD smears, we tentatively proposed referral to a gynaecologist for women with more than 4.3!106 HPV16 genome copies per smear because they are at increased risk of high-grade CIN, although a lower viral load does not necessarily exclude progressive CIN disease and would at least warrant a repeat smear.74 The advantage of such a policy would be a decrease in the number of repeat smears in the group of women with high viral loads, because they can be sent for colposcopy-directed biopsy. In addition to HPV16, we developed type-specific real-time PCRs for HPV types 18, 31 and 33 to define viral load cut-off points at the 25th, 33th and 50th percentiles. Although we observed a substantial overlap between HPV DNA load levels in GP5C /6CPCR-positive women with and without CIN3, all women with CIN3 had viral load levels above the 25th percentile threshold of women with normal cytology harbouring the respective hrHPV type. This indicates that, among women who test hrHPV positive by GP5C/6CPCR, viral load assessment will identify 25% of those without clinically relevant hrHPV infections (Hogewoning et al, unpublished data). Cut-off points defined by real-time PCR methods depend on the PCR method used and might vary between laboratories using the same PCR method. They might also depend on the screening population, as younger women may show increased viral loads in productive infections that will be cleared within a few months.18 As well as real-time PCR, viral load can be estimated from signals obtained in hc2 analysis by comparison of sample signals to positive control signals (‘relative light units’; RLU). This yields a semiquantitative result.77 Gravitt et al78 showed that the standard cut-off point of 1 pg viral DNA/ml sample (approximately 105 viral genome copies) differentiated cytologically normal women from women with LSIL or worse. However, among women with abnormal cytology, the viral load did not change as lesion severity increased. The predictive value for hc2 viral load measurement for high-grade CIN lesions is still a matter of debate, differences between studies probably being caused by different sampling methods and different study cohorts. At present, it is thought that high viral load as measured by hc2 is—if at all—only a short-term marker of the risk of CIN3 or worse.77,79,80 Differences in predictive value between hc2 and real-time PCR might, apart from differences in study design, be caused by the fact that real-time PCR is HPV type specific whereas hc2 detects a range of 13 hrHPV types simultaneously, giving rise to viral load signals that might be composites of multiple infections. In addition, real-time PCR—but not hc2—allows quantitation of the cells in the sample, thus enabling better standardization of the measurement. Although viral load measurement by hc2 apparently is not suitable (yet) for risk assessment, its specificity might be enhanced by increasing the cut-off point.81 hrHPV mRNA analysis hrHPV DNA detection methods such as PCR and hc2 are unlikely to distinguish active, persistent and progressive infections from regressive infections. Even in regressive infections, hrHPV DNA might be present in the cervical smear, for example contained within apoptotic cells or cellular debris. A probable hallmark of
HPV testing in cervical screening 261
an active, persistent or progressive hrHPV infection is transcription of the viral oncogenes E6 and E7, because these are essential for transformation of the infected cells.82 In transformed cells, the E6 and E7 proteins disrupt the cell cycle control by interacting with p53 and pRb, respectively.83,84 Hence, transcriptional analysis of E6/E7 might be useful for risk stratification of hrHPVinfected women. Indeed, only in a small proportion of hrHPV DNA-positive cervical smears E6/E7 mRNA can be detected.85,86 Several studies have addressed the clinical value of this finding. Molden et al have shown that, among women with BMD cytology, hrHPV mRNA positivity confers a relative risk of 70 for developing high-grade CIN as compared to hrHPV mRNA-negative women.86 Moreover, hrHPV mRNA positivity appeared to be related to persistence of the infection87, which is a prerequisite for the development of cervical cancer and its precursors. These data suggest that hrHPV mRNA analysis is a valuable tool in cervical cancer screening. At present, the only commercially available hrHPV mRNA detection system is a real-time nucleic acid sequence based amplification (NASBA) method detecting HPV types 16, 18, 31, 33 and 45 (NorChip AS, Norway). Whether this limited number of hrHPV types is sufficient to detect all women at risk for highgrade lesions remains to be investigated.
POSSIBLE ALGORITHMS FOR REPLACEMENT OF CYTOLOGY BY HRHPV TESTING If cytology as a primary screening method is to be replaced by hrHPV testing, the screening algorithm should be sufficiently sensitive to identify all women at risk for high-grade cervical lesions, yet sufficiently specific to avoid unnecessary referral and repeat smears. At present, no hrHPV test is available that fulfils both these conditions. However, we are obtaining more and more data that show that hrHPV testing is a better screening tool than cytology. Hence, it seems logical to subject all women participating in cervical screening to a sensitive hrHPV test first to make sure no high-grade lesions are missed. All women who test negative need no further follow-up until the next screening round (5 years or maybe more; see above), because their chance of having CIN3 or worse is very low.30 Because a sensitive hrHPV detection method will classify women with a transient hrHPV infection to be at risk for high-grade lesions, the specificity of the screening algorithm should be enhanced by an additional test before making decisions about referral and/or repeat smears. Reflex cytological analysis of all hrHPV-positive smears is a logical next step, because all women who have hrHPV-positive cytology classified as borderline dyskaryosis or worse are at an increased risk of high-grade CIN and should be referred for colposcopy. Women with cytologically normal hrHPV-positive smears are subjected to additional tests. First, their smears should be genotyped. This is necessary for additional tests such as viral load analysis by real-time PCR or hrHPV mRNA analysis. Genotyping will distinguish HPV16 and HPV18 infections that require a status aparte and seem to warrant more aggressive clinical management than other hrHPV types. Women who are infected by either of these two types could be referred to the gynaecologist if after 1 year, hrHPV is still present and their cytology is normal. Women with low viral loads and/or absence of viral mRNA could probably be referred to the next screening round, but these strategies require further confirmation.
262 A. A. T. P. Brink et al
Practice points † hrHPV testing is a useful tool in population-based cervical cancer screening— as a quality control for cervical cytology and for the triage of women with cytologically equivocal smears—and for monitoring therapy efficacy of highgrade CIN lesions † among hrHPV-positive women, those infected with HPV16 or HPV18 will need extra follow-up because they are at an increased risk of cervical cancer † endocervical curettage is warranted for women who have had an HPV18positive but cytologically normal smear for more than 1 year, because they are at an increased risk for high-grade glandular lesions of the cervix, which might be missed by cytology
Research agenda † † † † †
prospective trials for the clinical validation of hrHPV mRNA analysis hPV viral load analysis self-sampling feasibility of displacement of cytology by hrHPV testing
REFERENCES 1. Nobbenhuis MA, Walboomers JM, Helmerhorst TJ et al. Relation of human papillomavirus status to cervical lesions and consequences for cervical-cancer screening: a prospective study. Lancet 1999; 354: 20–25. 2. Clavel C, Masure M, Putaud I et al. Hybrid capture II, a new sensitive test for human papillomavirus detection. Comparison with hybrid capture I and PCR results in cervical lesions. Journal of Clinical Pathology 1998; 51: 737–740. 3. Castle PE, Schiffman M, Burk RD et al. Restricted cross-reactivity of hybrid capture 2 with nononcogenic human papillomavirus types. Cancer Epidemiology, Biomarkers & Prevention 2002; 11: 1394–1399. 4. Gravitt PE, Peyton CL, Apple RJ & Wheeler CM. Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by a single-hybridization, reverse line blot detection method. Journal of Clinical Microbiology 1998; 36: 3020–3027. 5. de Roda Husman AM, Walboomers JM, van den Brule AJ et al. The use of general primers GP5 and GP6 elongated at their 3 0 ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. The Journal of General Virology 1995; 76(Pt 4): 1057–1062. 6. Kleter B, van Doorn LJ, ter Schegget J et al. Novel short-fragment PCR assay for highly sensitive broadspectrum detection of anogenital human papillomaviruses. The American Journal of Pathology 1998;. 7. Jacobs MV, Snijders PJ, van den Brule AJ et al. A general primer GP5C/GP6(C)-mediated PCR-enzyme immunoassay method for rapid detection of 14 high-risk and 6 low-risk human papillomavirus genotypes in cervical scrapings. Journal of Clinical Microbiology 1997; 35: 791–795. 8. van den Brule AJ, Pol R, Fransen-Daalmeijer N et al. GP5C/6C PCR followed by reverse line blot analysis enables rapid and high-throughput identification of human papillomavirus genotypes. Journal of Clinical Microbiology 2002; 40: 779–787. 9. Kleter B, van Doorn LJ, Schrauwen L et al. Development and clinical evaluation of a highly sensitive PCRreverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. Journal of Clinical Microbiology 1999; 37: 2508–2517.
HPV testing in cervical screening 263 10. Josefsson A, Livak K & Gyllensten U. Detection and quantitation of human papillomavirus by using the fluorescent 5 0 exonuclease assay. Journal of Clinical Microbiology 1999; 37: 490–496. 11. Szuhai K, Sandhaus E, Kolkman-Uljee SM et al. A novel strategy for HPV detection and genotyping with SybrGreen and Molecular Beacon PCR. The American Journal of Pathology 2001; 159: 1651–1660. 12. Roda Husman AM, Snijders PJ, Stel HVet al. Processing of long-stored archival cervical smears for human papillomavirus detection by the polymerase chain reaction. British Journal of Cancer 1995; 72: 412–417. 13. Papanicolao GN & Traut HF. The diagnostic value of vaginal smears in carcinoma of the uterus. American Journal of Obstetrics and Gynecology 1941; 42: 193–206. 14. Lorincz AT & Richart RM. Human papillomavirus DNA testing as an adjunct to cytology in cervical screening programs. Archives of Pathology & Laboratory Medicine 2003; 127: 959–968. 15. Cuzick J, Szarewski A, Cubie H et al. Management of women who test positive for high-risk types of human papillomavirus: the HART study. Lancet 2003; 362: 1871–1876. 16. Bais AG, Rebolj M, Snijders PJ et al. Triage using HPV-testing in persistent borderline and mildly dyskaryotic smears: proposal for new guidelines. International Journal of Cancer 2005; 116: 122–129. 17. Zielinski GD, Bais AG, Helmerhorst TJ et al. HPV testing and monitoring of women after treatment of CIN 3: review of the literature and meta-analysis. Obstetrical & Gynecological Survey 2004; 59: 543–553. 18. Evander M, Edlund K, Gustafsson A et al. Human papillomavirus infection is transient in young women: a population-based cohort study. The Journal of Infectious Diseases 1995; 171: 1026–1030. 19. Svare EI, Kjaer SK, Worm AM et al. Risk factors for HPV infection in women from sexually transmitted disease clinics: comparison between two areas with different cervical cancer incidence. International Journal of Cancer 1998; 75: 1–8. 20. Zielinski GD, Snijders PJ, Rozendaal L et al. HPV presence precedes abnormal cytology in women developing cervical cancer and signals false negative smears. British Journal of Cancer 2001; 85: 398–404. 21. Walboomers JM, Husman AM, Snijders PJ et al. Human papillomavirus in false negative archival cervical smears: implications for screening for cervical cancer. Journal of Clinical Pathology 1995; 48: 728–732. 22. Wallin KL, Wiklund F, Angstrom T et al. Type-specific persistence of human papillomavirus DNA before the development of invasive cervical cancer. The New England Journal of Medicine 1999; 341: 1633–1638. 23. Meijer CJ, Rozendaal L, van der Linden HC et al. Human papillomavirus testing for primary cervical cancer screening 1997;338–347. 24. Zielinski GD, Snijders PJ, Rozendaal L et al. The presence of high-risk HPV combined with specific p53 and p16INK4a expression patterns points to high-risk HPV as the main causative agent for adenocarcinoma in situ and adenocarcinoma of the cervix. The Journal of Pathology 2003; 201: 535–543. 25. Krane JF, Granter SR, Trask CE et al. Papanicolaou smear sensitivity for the detection of adenocarcinoma of the cervix: a study of 49 cases. Cancer 2001; 93: 8–15. 26. Bulkmans NWJ, Rozendaal L, Snijders PJ et al. POBASCAM, a population-based randomised controlled trial for implementation of high-risk HPV testing in cervical screening; design, methods and baseline data of 44,102 women. International Journal of Cancer 2003; 110: 94–101. 27. Pirog EC, Kleter B, Olgac S et al. Prevalence of human papillomavirus DNA in different histological subtypes of cervical adenocarcinoma. The American Journal of Pathology 2000; 157: 1055–1062. 28. Rozendaal L, Walboomers JM, van der Linden JC et al. PCR-based high-risk HPV test in cervical cancer screening gives objective risk assessment of women with cytomorphologically normal cervical smears. International Journal of Cancer 1996; 68: 766–769. 29. Bosch FX, Lorincz A, Munoz N et al. The causal relation between human papillomavirus and cervical cancer. Journal of Clinical Pathology 2002; 55: 244–265. 30. Bulkmans NW, Rozendaal L, Voorhorst FJ et al. Long-term protective effect of high-risk human papillomavirus testing in population-based cervical screening. British Journal of Cancer 2005; 92: 1800– 1802. 31. Bulk S, van Kemenade FJ, Rozendaal L & Meijer CJ. The Dutch CISOE—a framework for cytology reporting increases efficacy of screening upon standardisation since 1996. Journal of Clinical Pathology 2004; 57: 388–393. 32. Solomon D, Davey D, Kurman R et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. Journal of the American Medical Association 2002; 287: 2114–2119. 33. Zielinski GD, Snijders PJ, Rozendaal L et al. High-risk HPV testing in women with borderline and mild dyskaryosis: long-term follow-up data and clinical relevance. The Journal of Pathology 2001; 195: 300–306.
264 A. A. T. P. Brink et al 34. Fait G, Kupferminc MJ, Daniel Y et al. Contribution of human papillomavirus testing by hybrid capture in the triage of women with repeated abnormal pap smears before colposcopy referral. Gynecologic Oncology 2000; 79: 177–180. 35. Berkhof J, de Bruijne MC, Zielinski GD et al. Evaluation of cervical screening strategies with adjunct highrisk human papillomavirus testing for women with borderline or mild dyskaryosis. International Journal of Cancer. In press. 36. Zielinski GD, Rozendaal L, Voorhorst FJ et al. HPV testing can reduce the number of follow-up visits in women treated for cervical intraepithelial neoplasia grade 3. Gynecologic Oncology 2003; 91: 67–73. 37. Nobbenhuis MA, Meijer CJ, van den Brule AJ et al. Addition of high-risk HPV testing improves the current guidelines on follow-up after treatment for cervical intraepithelial neoplasia. British Journal of Cancer 2001; 84: 796–801. 38. Kim JJ, Wright TC & Goldie SJ. Cost-effectiveness of human papillomavirus DNA testing in the United Kingdom, The Netherlands, France, and Italy. Journal of the National Cancer Institute 2005; 97: 888– 895. 39. Paraskevaidis E, Arbyn M, Sotiriadis A et al. The role of HPV DNA testing in the follow-up period after treatment for CIN: a systematic review of the literature. Cancer Treatment Reviews 2004; 30: 205–211. 40. Munoz N, Bosch FX, de Sanjose S et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. The New England Journal of Medicine 2003; 348: 518–527. 41. Munoz N, Bosch FX, Castellsague X et al. Against which human papillomavirus types shall we vaccinate and screen? The international perspective. International Journal of Cancer 2004; 111: 278–285. 42. Schiffman M, Khan MJ, Solomon D et al. A study of the impact of adding HPV types to cervical cancer screening and triage tests. Journal of the National Cancer Institute 2005; 97: 147–150. 43. Bulkmans NW, Bleeker MC, Berkhof J et al. Prevalence of types 16 and 33 is increased in high-risk human papillomavirus positive women with cervical intraepithelial neoplasia grade 2 or worse. International Journal of Cancer 2005;. 44. Bulk S, Bulkmans NW, Zielinski GD et al. Cervical adenocarcinoma in situ and invasive adenocarcinoma have an increased prevalence of HPV18 only, whereas squamous cell carcinoma is associated with both HPV16 and HPV18. Submitted for publication. 45. Berkhof J, de Bruijne MC, Zielinski GD et al. Evaluation of cervical screening strategies with adjunct highrisk human papillomavirus testing for women with borderline or mild dyskaryosis. International Journal of Cancer 2005; October 10 [Epub ahead of print]. 46. Khan MJ, Castle PE, Lorincz AT et al. The elevated 10-year risk of cervical precancer and cancer in women with human papillomavirus (HPV) type 16 or 18 and the possible utility of type-specific HPV testing in clinical practice. Journal of the National Cancer Institute 2005; 97: 1072–1079. 47. Castle PE, Solomon D, Schiffman M & Wheeler CM. Human papillomavirus type 16 infections and 2-year absolute risk of cervical precancer in women with equivocal or mild cytologic abnormalities. Journal of the National Cancer Institute 2005; 97: 1066–1071. 48. Schiffman M, Herrero R, Desalle R et al. The carcinogenicity of human papillomavirus types reflects viral evolution. Virology 2005; 337: 76–84. 49. Richardson H, Kelsall G, Tellier P et al. The natural history of type-specific human papillomavirus infections in female university students. Cancer Epidemiology, Biomarkers & Prevention 2003; 12: 485–490. 50. Steenbergen RD, de Wilde J, Wilting SM et al. HPV-mediated transformation of the anogenital tract. Journal of Clinical Virology 2005; 32(supplement 1): S25–S33. 51. Beskow AH, Moberg M & Gyllensten UB. HLA class II allele control of HPV load in carcinoma in situ of the cervix uteri. International Journal of Cancer 2005;. 52. Beskow AH, Josefsson AM & Gyllensten UB. HLA class II alleles associated with infection by HPV16 in cervical cancer in situ. International Journal of Cancer 2001; 93: 817–822. 53. Beskow AH & Gyllensten UB. Host genetic control of HPV 16 titer in carcinoma in situ of the cervix uteri. International Journal of Cancer 2002; 101: 526–531. 54. Clifford GM, Smith JS, Aguado T & Franceschi S. Comparison of HPV type distribution in high-grade cervical lesions and cervical cancer: a meta-analysis. British Journal of Cancer 2003; 89: 101–105. 55. van Ballegooijen M, Rebolj M, Meerding WJ et al. De praktijk van het bevolkingsonderzoek naar baarmoederhalskanker in Nederland in 2001. 2003.
HPV testing in cervical screening 265 56. Peto J, Gilham C, Fletcher O & Matthews FE. The cervical cancer epidemic that screening has prevented in the UK. Lancet 2004; 364: 249–256. 57. Sawaya GF & Grimes DA. New technologies in cervical cytology screening: a word of caution. Obstetrics and Gynecology 1999; 94: 307–310. 58. Sasieni PD, Cuzick J & Lynch-Farmery E. Estimating the efficacy of screening by auditing smear histories of women with and without cervical cancer. The national co-ordinating network for cervical screening working group. British Journal of Cancer 1996; 73: 1001–1005. 59. Nobbenhuis MA, Helmerhorst TJ, van den Brule AJ et al. Primary screening for high risk HPV by home obtained cervicovaginal lavage is an alternative screening tool for unscreened women. Journal of Clinical Pathology 2002; 55: 435–439. 60. Sellors JW, Lorincz AT, Mahony JB et al. Comparison of self-collected vaginal, vulvar and urine samples with physician-collected cervical samples for human papillomavirus testing to detect high-grade squamous intraepithelial lesions. Canadian Medical Association Journal 2000; 163: 513–518. 61. Hillemanns P, Kimmig R, Huttemann U et al. Screening for cervical neoplasia by self-assessment for human papillomavirus DNA. Lancet 1999; 354: 1970. 62. Dannecker C, Siebert U, Thaler CJ et al. Primary cervical cancer screening by self-sampling of human papillomavirus DNA in internal medicine outpatient clinics. Annals of Oncology 2004; 15: 863–869. 63. Wright Jr. TC, Denny L, Kuhn L et al. HPV DNA testing of self-collected vaginal samples compared with cytologic screening to detect cervical cancer. Journal of American Medical Association 2000; 283: 81–86. 64. Bais AG, Rebolj M, Snijders PJ et al. Triage using HPV-testing in persistent borderline and mildly dyskaryotic smears: proposal for new guidelines. International Journal of Cancer 2005; 116(1): 122–129. 65. Brink AATP, Meijer CJ, Wiegerinck MAHM et al. High concordance in HPV test results between cervical samples taken by a novel self-sampler and a cytobrush. In preparation. 66. Solomon D, Schiffman M & Tarone R. Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: baseline results from a randomized trial. Journal of the National Cancer Institute 2001; 93: 293–299. 67. Rebello G, Hallam N, Smart G et al. Human papillomavirus testing and the management of women with mildly abnormal cervical smears: an observational study. British Medical Journal 2001; 322: 893–894. 68. Hawley-Nelson P, Vousden KH, Hubbert NL et al. HPV16 E6 and E7 proteins cooperate to immortalize human foreskin keratinocytes. EMBO Journal 1989; 8: 3905–3910. 69. Choo KB, Pan CC & Han SH. Integration of human papillomavirus type 16 into cellular DNA of cervical carcinoma: preferential deletion of the E2 gene and invariable retention of the long control region and the E6/E7 open reading frames. Virology 1987; 161: 259–261. 70. Snijders PJ, van den Brule AJ & Meijer CJ. The clinical relevance of human papillomavirus testing: relationship between analytical and clinical sensitivity. The Journal of Pathology 2003; 201: 1–6. 71. Rozendaal L, Westerga J, van der Linden JC et al. PCR based high risk HPV testing is superior to neural network based screening for predicting incident CIN III in women with normal cytology and borderline changes. Journal of Clinical Pathology 2000; 53: 606–611. 72. Ylitalo N, Sorensen P, Josefsson AM et al. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: a nested case-control study. Lancet 2000; 355: 2194–2198. 73. Josefsson AM, Magnusson PK, Ylitalo N et al. Viral load of human papilloma virus 16 as a determinant for development of cervical carcinoma in situ: a nested case-control study. Lancet 2000; 355: 2189–2193. 74. van Duin M, Snijders PJ, Schrijnemakers HF et al. Human papillomavirus 16 load in normal and abnormal cervical scrapes: an indicator of CIN II/III and viral clearance. International Journal of Cancer 2002; 98: 590– 595. 75. Moberg M, Gustavsson I & Gyllensten U. Real-time PCR-based system for simultaneous quantification of human papillomavirus types associated with high risk of cervical cancer. Journal of Clinical Microbiology 2003; 41: 3221–3228. 76. Moberg M, Gustavsson I & Gyllensten U. Type-specific associations of human papillomavirus load with risk of developing cervical carcinoma in situ. International Journal of Cancer 2004; 112: 854–859. 77. Lorincz AT, Castle PE, Sherman ME et al. Viral load of human papillomavirus and risk of CIN3 or cervical cancer. Lancet 2002; 360: 228–229.
266 A. A. T. P. Brink et al 78. Gravitt PE, Burk RD, Lorincz A et al. A comparison between real-time polymerase chain reaction and hybrid capture 2 for human papillomavirus DNA quantitation. Cancer Epidemiology, Biomarkers & Prevention 2003; 12: 477–484. 79. Dalstein V, Riethmuller D, Pretet JL et al. Persistence and load of high-risk HPV are predictors for development of high-grade cervical lesions: a longitudinal French cohort study. International Journal of Cancer 2003; 106: 396–403. 80. Castle PE, Schiffman M & Wheeler CM. Hybrid capture 2 viral load and the 2-year cumulative risk of cervical intraepithelial neoplasia grade 3 or cancer. American Journal of Obstetrics and Gynecology 2004; 191: 1590–1597. 81. Howard M, Sellors J & Kaczorowski J. Optimizing the hybrid capture II human papillomavirus test to detect cervical intraepithelial neoplasia. Obstetrics and Gynecology 2002; 100: 972–980. 82. Munger K, Phelps WC, Bubb V et al. The E6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes. Journal of Virology 1989; 63: 4417–4421. 83. Scheffner M, Werness BA, Huibregtse JM et al. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 1990; 63: 1129–1136. 84. Dyson N, Howley PM, Munger K & Harlow E. The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science 1989; 243: 934–937. 85. Molden T, Kraus I, Karlsen F et al. Comparison of human papillomavirus messenger RNA and DNA detection: a cross-sectional study of 4,136 women O30 years of age with a 2-year follow-up of high-grade squamous intraepithelial lesion. Cancer Epidemiology, Biomarkers & Prevention 2005; 14: 367–372. 86. Molden T, Nygard JF, Kraus I et al. Predicting CIN2C when detecting HPV mRNA and DNA by PreTect HPV-proofer and consensus PCR: a 2-year follow-up of women with ASCUS or LSIL Pap smear. International Journal of Cancer 2005; 114: 973–976. 87. Cuschieri KS, Whitley MJ & Cubie HA. Human papillomavirus type specific DNA and RNA persistenceimplications for cervical disease progression and monitoring. Journal of Medical Virology 2004; 73: 65–70.