Cervical human papillomavirus screening by PCR - CiteSeerX

Article in press - uncorrected proof Clin Chem Lab Med 2005;43(11):1171–1177 2005 by Walter de Gruyter • Berlin • New York. DOI 10.1515/CCLM.2005.203

2005/241

Review

Cervical human papillomavirus screening by PCR: advantages of targeting the E6/E7 region

Brian J. Morris* Basic & Clinical Genomics Laboratory, School of Medical Sciences and Institute for Biomedical Research, The University of Sydney, Sydney, Australia

gies currently in use could lead to a false negative finding, with serious clinical implications inherent in such an outcome.

Viral integration

Abstract

Physical forms of HPV DNA

PCR is a promising method for detection of human papillomavirus (HPV), the high-risk forms of which are responsible for cervical cancer. PCR primers that target the L1 or E1 region can be unreliable and may miss more advanced disease, whereas those directed at the E6 or E7 regions, which encode oncogenic products, are preferable because 1) the LI/E1 regions, but never the E6/E7 regions, are lost during integration of viral DNA into host genomic DNA, a process that can represent an integral component of progression from infection to tumorigenesis; and 2) the E6/ E7 nucleotide sequence exhibits less nucleotide variation. The choice of region used for PCR has implications for HPV screening strategies in the clinical diagnosis and management of cervical cancer.

The HPV genome consists of an upstream regulatory (URR), or long control, region (LCR), an early gene (E1, E2, E6 and E7) region, and a late gene (L1 and L2) region (3). It should be noted, moreover, that adjacent regions overlap: e.g., E6, which is bigger than E7, overlaps a significant portion of E7. After infection of a cervical epithelial cell, the circular HPV DNA initially remains free inside the cell. Low-risk HPV (types 6 and 11) remain exclusively in this episomal form (4, 5). This also applies to highrisk types in cervical intrepithelial neoplasia 1 (CIN1), and frequently in CIN2 and CIN3 (6, 7). However, in cervical carcinoma cell lines, the viral DNA is usually integrated into the cellular genome (8–11). Integration is common, precedes cancer development, and could be involved in malignant transformation (2, 6, 12). Most importantly, as far as screening is concerned, integration leads to loss of large regions, most often involving E2 and E1, but also L1 and L2, and residual integrated HPV DNA could ultimately be the only form present in an infected cell. This poses a serious challenge to PCR approaches that involve primers directed at any region that is deleted.

Keywords: cervical cancer; cervical dysplasia; cervical intrepithelial neoplasia (CIN); diagnostic testing; oncogenes; open reading frames E6, E7, E1, E2 and L1; papillomavirus, human; PCR; viral integration.

Introduction Virtually 100% of cervical cancer is caused by highrisk types of human papillomavirus (HPV) (1, 2), meaning that reliable detection has considerable diagnostic and prognostic relevance. The most common methods for viral detection today utilize the polymerase chain reaction (PCR). However, PCR tests usually target only a 100–500-bp region for amplification, whereas the HPV genome is 7.9 kb, meaning that there is a wide choice of which region to amplify. Most importantly, differences in biological properties and sequence characteristics of different regions mean that choice of the target region can be crucial to a reliable outcome of a PCR for HPV. Here I review the options and explain why certain popular strate*Corresponding author: Brian J. Morris, Basic & Clinical Genomics Laboratory, School of Medical Sciences and Institute for Biomedical Research, Building F13, The University of Sydney, NSW 2006, Australia Phone: q61-2-9351 3688, Fax: q61-2-9351 2227, E-mail: [email protected]

Viral integration and cancer In HPV16-infected cancer cells, both episomal and integrated, head-to-tail arrays of HPV DNA have been seen (13). For cervical carcinomas containing HPV16, one study found 44% had integrated, 44% had episomal and 11% had both forms of HPV DNA (14). In another study, 8/34 invasive cancers had only the integrated form (15). Yet another found 11/25 to have only integrated HPV (16). A heavy load of HPV16 DNA has been associated with rapid progression of CIN lesions (17). In the case of HPV18, the HPV DNA was integrated and amplified in three cell lines (18), and in carcinomas a mixture of integrated and episomal HPV16 was observed (19). In 68 cervical cancer cases, 75% had HPV 16 or 18 DNA, this being episomal in 7% and each form in 18% (20). Early estimates were 36–71% for episomal, 22–39% for integrated, and 6–25% for both (21). A method involving real-time PCR has been used to screen for integration status by

Article in press - uncorrected proof 1172

Morris: Cervical HPV PCR strategies

determination of the E2/E6 ratio and may be suitable for monitoring disease progression (17). Viral integration and precancer In the case of dysplasias, an early report noted integration of HPV16 DNA in 86% of cases, i.e., occurring in the precancerous stage (22). In CIN samples, the integrated form was seen in three of the seven positive for HPV16 (19). The likelihood that integration is a characteristic of progression to malignancy was highlighted in a study that showed integrated HPV in 3% of 100 CIN biopsy specimens, but 81% of 69 carcinomas (6). HPV16 was most prevalent (58%) and, of these, 72% contained integrated and 27% exclusively episomal HPV DNA. Of those with the integrated form, in 80% this was the only form in which the viral DNA was present in the cell, with just 20% having both episomal and integrated. A recent study found integrated HPV (mostly HPV16) in 80% of CIN3, 63% of CIN2 and 36% of CIN1 (23). The absence of the episomal form (i.e., complete integration) was 36%, 25% and 7% at each respective stage. The ratio of integrated to episomal forms did not differ in each stage, nor did the viral load (23). Integration of HPV16 compared to HPV18 In a study referred to above (20), of 51 cervical cancers with HPV16, 71% had only the integrated form, 20% had both integrated and episomal, and 10% had episomal only, whereas all 17 HPV18-containing cancers had just the integrated form. The significance of this is accentuated by claims of the aggressive, rapidly progressing nature of HPV18-associated dysplasias, and is thus highly relevant to screening tests. Similarly, others found that in the case of 23 carcinomas with HPV18, which has greater transforming efficiency, all had integrated HPV DNA, with only one having episomal DNA as well (6). Genomic site for integration Integration involves breakage of the HPV DNA and loss of portions of the genome. Early observations of cervical carcinoma cell lines revealed deletion of 2–3 kb that included the E2 to L2 region (18). Importantly, E6–E7 transcripts could be detected in the cells. It was then found that open reading frames (ORFs) E1, E2, E4 and E5 were interrupted by flanking host cell DNA, pointing to these as sites of integration (24). HPV mRNAs present hybridized to probes for the entire E6 and E7 ORF and a minor part of the E1 ORF, meaning that these were the only portions of the HPV genome present. No hybridization to L1 and L2 ORFs could be detected, implying that these regions had either been deleted or were no longer being transcribed. Similarly, in a study of six cervical carcinoma samples, the E6/E7 region was found to be retained on integration, but the E2 region was lost (25). E2, as well as most of E1, was also deleted in a cervical carcinoma containing HPV16 DNA (26). Examination of a cell line showed that the integration that disrupted E2

and L2 was found to have occurred in the premalignant lesion from which the line was derived (27). In a study of four clones with integrated HPV16 DNA and deletions in the E1/E2 and L1/L2 regions, no site specific for integration is present in the viral sequence (28). Rather, during integration, viral sequences are opened within any of the ORFs, except the E6/E7 ORFs and locus control region. Integration of the viral genome into human chromosomal DNA of cancer cells usually disrupts or deletes the E2 ORF, which results in the loss of expression of the E2 gene, but is associated with the continuation of high expression of E6 and E7 (29). Genomic site for mutation Another issue in HPV screening by PCR is that mutations of HPV DNA take place in the L1 region of the HPV genome, but less frequently in the E6 region. This could affect hybridization fidelity of PCR primers and thus the ability to detect HPV present. For example, a HPV16-positive cervical carcinoma had only 3091 bp of the original 7905 bp of the viral genome left (30). This included the E6/E7 region. Importantly, whereas the E6 and E7 ORFs showed complete concordance with the published sequences (which also supports the role of E6/E7 in tumorigenesis), there were multiple mutations (transversions, transitions, small deletions, and small insertions) in the remaining integrated HPV16 DNA, which was composed of parts of the L1 and E1 ORFs. (The 59 end of L1 was missing, the integrated DNA beginning at nt 6334.) The mutations included: a single base deletion at nt 6387, insertion of the nucleotide sequence CAT at position 6901 wwhich has also been noted by others (26, 31, 32)x and deletion of a GAT sequence at nt 6949. Why does integration occur? Integration of HPV16 DNA leads to increased steadystate mRNA encoding the viral oncogenes E6 and E7 as a consequence of increased stability conferred by disruption of an AqU-rich sequence in the 39-UTR of E6 and E7 mRNAs and replacement with cellular sequences with lower AqU content (33). This would account, at least in part, for the higher concentration of E6 and E7 proteins in clonal populations with integrated HPV16 DNA compared with ones in which the DNA is present exclusively in an episomal form (29). The cells with integrated DNA therefore outgrow those with episomal HPV DNA only, i.e., integration provides a growth advantage (29). The protein product of E6 binds p53, an important negative regulator of the cell cycle, and in so doing inhibits its activity, thus leading to uncontrolled growth. This offers a biological basis for oncogenesis. Similarly, the E7 protein attaches to another crucial cell cycle regulator, the retinoblastoma binding protein. The E2 gene encodes a site-specific DNA-binding protein that regulates the HPV promoter and so directs E6 and E7 expression (34–36). As far as cervical cancer is concerned, what matters is the E6 and

Article in press - uncorrected proof Morris: Cervical HPV PCR strategies 1173

E7 regions. The rest of the genome may have had a role in transmission of the virus to a new host cell, but it is the proteins encoded by E6 and E7 that cause the cancer. Integration of HPV DNA occurs early in cancer development and is important as an activation mechanism for progression from precancer to cancer (2, 6, 12).

HPV PCR detection methods So how does all of this affect decisions about how to approach HPV detection by PCR? L1 consensus PCR This approach targets so-called ‘‘consensus’’ primers to the highly conserved L1 region of the HPV genome. An extra step is then required, such as probing with radioactive- or biotin-labeled type-specific oligonucleotides in order to address the important question of type of HPV present. Consensus primer sets include those directed at various parts of the L1 region, viz. ones that have been termed by those who developed them MY09-MY11, GP5-GP6, GP5q-GP6q, and oli1b-oli-2I, and those directed at the E1 region, viz. CpICpIIG (37). The MY09/MY07 (MY-PCR) primer set (38) and the GP5q/GP6q (GPq-PCR) primer set (39) are frequently used for HPV detection in clinical samples. The latter are a modification (extension in length) of an earlier version, GP5/GP6, which were designed to amplify the nt 6624–6765 region of the HPV genome to yield a 140–150-bp product (40). The MY-PCR set amplifies the nt 6582–7033 region to yield a 450-bp product (38). The MY-PCR primer set is synthesized with several degenerate nucleotides in each primer and is thus a mixture of 25 primers capable of amplifying a wide spectrum of HPV types (38, 41). In contrast, there are only two primers in the GPq-PCR set and detection of a broad range of HPVs is achieved using a lowered annealing temperature during PCR (39). The MY09/11 primers preferentially amplify certain HPV types (1, 42), so missing other types that may be present. Nevertheless, although the MY09/11 and GP5q/GP6q systems yield a nearly identical prevalence of HPV in a set of clinical samples, the MY09/11 primers detect twice as many samples with multiple HPV types than do the GP5q/GP6q primers (43). The MY-PCR is used more in America and Asia, and the GPq-PCR in Europe, reflecting the geographical locality where each set was developed. Instead of random degeneracies, inosine, which matches with any nucleotide, can be incorporated into a series of primers directed at the same target region. This allows PCR at optimum, rather than low, annealing temperature. Examples are PGMY (42, 44) and SPF10 (45) primers, which yield amplicons of ;450 and 65 bp, respectively, and are technically superior to and lead to higher detection rates than MY09/11 (42, 44–46). A recent addition is the Roche Amplicor HPV test, which involves a pool of primers that give a ;165-bp amplicon from the L1 region of

13 high-risk HPV types. This test shows promise (47), but clinical trials are needed. There is also a PCR method that amplifies DNA within the conserved E1 region using two 21mers (48). E6 primer approach This PCR strategy targets the cancer-causing E6, or E7, portion of the virus (49–52), because this region shows the greatest differences in nucleotide sequence between different HPV types and is always retained by cells that are infected. Although sequence variations in the E6 region have been reported, these are almost exclusively in the N-terminal region (16), so that designing PCR primers in the C-terminal half would appear preferable. In this approach type- or clade-specific primers can be designed to not only determine if HPV is present, but at the same time distinguish high- from low-risk HPV types by, for example, generating PCR products that differ in size for low- and high-risk types or by using real-time PCR strategies. Quantification, as achieved by real-time PCR, can provide all-important information about viral load, which has prognostic implications (53). HPV groupings based on the sequence homology between the E6 regions of HPVs correlate with clinical significance (54), leading to support for the use of such riskgroup-specific primers. The value of this approach has been confirmed (55) and offers a direct means of obtaining the kind of information that is needed for clinical decision-making. All of the major HPV types are now known, so that suitable primers directed at the E6/E7 region of the various types, and capable of telling whether a woman is infected with a high-risk HPV, can be synthesized and readily incorporated into the same reaction tube. In practice there is a limit to the extent of multiplexing, so that a series of reaction tubes with different sets of primers may lead to greater reliability. The use of E6/E7 primers makes the chance of missing an infection very unlikely indeed. Costs are also lower, since fewer steps are needed. E6 testing could thus be more attractive in widespread screening than the L1 approach. In so doing, it has, moreover, been emphasized that any HPV assay should only detect high-risk HPV types (56). At least one commercial PCR kit directed at the E6/E7 region has been developed (52). Studies that have compared L1 and E6 primers The advantage of targeting the E1 or L1 region is that this has the potential to detect many, perhaps all, HPV types. The major disadvantage is that this has the potential not to occur if the virus is present in an exclusively integrated form, as is indeed the case for many advanced abnormalities. Thus, the target for PCR can be absent for regions other than E6 and E7. On the other hand, targeting the E6 or E7 regions will detect all samples infected with high-risk HPV, provided that primers for all relevant HPV types are used. That the method of detection strongly influences the rate of HPV detection was emphasized in a study of



anal carcinomas in which it was shown that the least sensitive method for HPV detection was PCR using L1 consensus primers (rate of 16%), whereas type-specific primers directed at the E6 region yielded a positivity of 46% (57). L1 PCR probably detects the majority of cases with low-risk HPV types 6 or 11, but could underestimate high-risk HPV 16 or 18 infections. Fragmentation of template DNA in formalin-fixed tissues has been considered (58), and consistency of amplification of HPV DNA from such specimens has been reported (57). Even though one L1 PCR gives amplicons of 450 bp, type-specific primers give 380–440-bp PCR products from formalin-fixed specimens. The more likely explanation was said to be loss of HPV genomic DNA upon integration, in particular deletion of L1 and L2 (57). Loss of transcripts from these regions has indeed been demonstrated in HPV16-infected cells (59). Although the rarity of positivity for viral capsid proteins in intra-epithelial and invasive anogenital neoplasia (18, 60) could suggest loss of L1, transcription of the L1 gene is usually restricted in undifferentiated epithelium of high-grade CIN and cervical cancer, due to trans-repression by E2 or unknown mechanism(s). In contrast to L1, for a given HPV type, the E6 and E7 genes are highly conserved (28, 30, 31, 59) and it has been stated that HPV infection should therefore be detectable in almost all cases where viral DNA is present (57). In a comprehensive study that addressed this issue, a range of consensus primer sets were tested, including those directed at various parts of the L1 region, viz. My09-My11, Gp5-Gp6, Gp5q-Gp6q, and oli-1boli-2I, and those directed at the E1 region, viz. CpICpIIG (37). By testing with all of these primer sets, as well as primers directed at the E6 or E7 region, 98% of 355 biopsy specimens from patients with invasive cervical carcinomas were found to be positive for HPV. However, use of just one set gave a much lower rate of detection. It is interesting to note that typespecific primers (for HPV 11, 16, 18, 31, 33 and 35) detected more HPV-infected patients than the most sensitive consensus primer set. In fact, it was necessary to use several consensus primer sets together (viz. the My, Gp/Gpq, and Cp sets) in order to detect a high number of HPV-positive patients. Moreover, based on results using consensus primers, L1 deletions were present in 23 of 56 (41%) samples. The data argue strongly against the reliability of using L1 consensus primers alone. Deletions mean that not all consensus primers will have hybridization targets in the HPV DNA, leading to the conclusion that a combination of consensus primers must be included in any PCR test to have any hope of detecting all or most HPV present in a specimen (35). In other studies, involving 635 w436 (61) and 199 (62)x cervical cancer samples from Spain, Columbia, and Brazil, HPV was detected in just 85% of samples using only one consensus primer set. An even lower rate (69%) emerged from a study involving the My L1 primers (63). Indeed, there appears to be agreement that use of either the MY-PCR or GPq-PCR methods

alone will underestimate the true prevalence of HPV in cervical samples (37, 64). Comparison of an E6/E7 consensus PCR test with the MY consensus primer L1 test found that 76 of 83 samples (92%) gave concordant results (52). Only one was positive by E6/E7, but negative by MY09/11. On the other hand, six were positive only by L1. This may be because the MY test can detect 25 HPVs, whereas the particular E6/E7 test used could only detect eight. Comparison of an E7 test with a test based on L1 modified nested primers (MY11/GP6q and GP5q/ GP6q) in 152 samples found the E7 test to be more sensitive and gave a higher detection rate (65). As mentioned in the present review, these authors also pointed out that almost all HPV18-positive cervical carcinomas contain only integrated HPV DNA (6, 20), stating that one of the underlying causes of false negativity was deletion or integration events affecting L1 and E1 sequences, and that disruption in the L1 region occurred more frequently for HPV18 than for other HPV types (1, 2). In a study of 69,290 samples, the lack of amplification of all HPV types when using the MY09/11 method led the authors to then subject those that tested negative to a subsequent E7 typespecific PCR (66). Of the types tested for, this showed that HPV51 was the most frequent type not detected by consensus primers (72%), followed by types 39 (12%), 18 and 35 (7%) and 66 (3%). As well as lack of specificity for certain HPV types (42), these authors cite loss of L1 due to integration (1) as another possibility, particularly for HPV18. Taken together, it would therefore seem that as long as primers directed at the E6/E7 region of all common high-risk HPV types are used, this PCR strategy should represent the most accurate diagnostic test.

Expert opinion Although the evidence suggests that integration may be an important step in the progression to malignancy, it is the very event of integration that could on occasions pose a challenge to PCR approaches that involve primers directed at the L1 region. This raises the specter of missing HPV during testing by PCR if the primers are directed at a mutated or variable region, even if this region is not deleted. That is, both deletions and mutations raise serious concerns when targeting the L1 region using a consensus PCR approach. This is much less likely to happen for the E6 or E7 region, since any mutation could have functional implications, most likely deleterious to viral oncogenic activity, so that the ramifications of failure to detect are likely to be inconsequential. And sites harboring mutations known to increase oncogenic potential should be avoided when designing primers. In reality, for a given HPV type, there is ‘‘extraordinary conservation of the E6/E7 DNA sequence’’ (30). This is significant, because it means that specific primers directed at the E6 region have a very much greater likelihood of annealing than ‘‘consensus’’ primers tar-


geting the L1 region. The implication of these findings is that whereas screening in women with CIN will probably pick up most HPV infections when L1 primers are used for PCR, some women with more advanced lesions may be missed because of the loss of L1 during integration. And it is these women in particular who must be correctly diagnosed, since their need for treatment is greater and more immediate. A more common problem is that the degenerate primers used to amplify L1 sequences are not 100% reliable, owing to the greater sequence variation in this region of the HPV genome, as compared to E6, and thus may fail to hybridize or lead to mispriming. On the other hand, the uniqueness of E6 between different HPV types means that detection of HPV by E6 primers requires a separate PCR for each HPV type, so adding to workload and costs. The problems associated with L1 consensus primers have tended to lead to large sets of type-specific primers anyway. Given the potential for loss of L1, it would appear far more sensible to use primers directed at the E6/E7 region in the first place. Thus, a continuum of HPV DNA forms are present. The episomal form tends to represent early HPV infection or CIN1, and, with progression to cancer, the prevalence of integrated HPV increases. Determination of the proportion of each may represent a rough guide to the stage of HPV-based cervical disease. Moreover, since the E6/E7 regions are retained and the L1/E2 regions tend to be deleted as disease progresses, measurement of the relative amounts of E6 and/or E7 expression and L1 and/or E2 expression may help to provide: • An assessment of the stage of HPV-based disease; • An assessment of the risk of HPV-based disease in an infected woman; and • A way of categorizing or staging women with HPV infection, but without detectable HPV-based disease, into those at risk for progression to disease and those not at risk of progression to disease. Nevertheless, since 37–71% of cervical cancers have episomal HPV DNA, detection of persistent HPV infection may be more important than detection of HPV DNA integration for the assessment of risk or stage of HPV-based disease in an infected woman.

Outlook Detection of HPV is widely accepted as being more reliable than cytological testing (Papanicolou smear) in cervical screening. Since PCR came ‘‘off-patent’’’ on 28 March 2005, as will patent coverage of the HPV E6 region in the USA and Europe in 2008, and the entire HPV genome in Australia in the same year, the use of PCR in clinical diagnostic testing procedures is set to escalate. Speed, accuracy, reliability, acceptability, familiarity and low cost all point to HPV PCR becoming a primary means of routine cervical cancer screening within a few years. Indeed, sensitivity and predictive value appear identical with the gold stan-

dard, colposcopy (47). To this end it is important that the most clinically reliable PCR strategy be adopted.

Highlights HPV PCR strategies directed at the E6/E7 region may be preferable to L1 because: • E6 and E7 are the oncogenic regions of HPV; • E6 and E7 are retained after infection, whereas E2 and L1 can be deleted; • E6 and E7 exhibit strong sequence conservation, unlike other regions such as L1; • The uniformity of the E6 sequence for each HPV type means less likelihood of primer mismatch during PCR; • The E6/E7 sequence differs between high- and lowrisk HPV types, thus providing diagnostic discrimination using PCR for testing; and • Overall selection of E6/E7 may provide the optimum choice for PCR.

References 1. Walboomers JM, Jacobs MV, Manos MM, Bosch F, Kummer JA, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189:12–9. 2. Bosch FX, Manos MM, Munoz M, Sherman M, Jansen AM, Peto J, et al. Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. J Natl Cancer Inst 1995;87:796–802. 3. Turek LP. The structure, function, and regulation of papillomaviral genes in infection and cervical cancer. Adv Virus Res 1994;44:305–56. 4. Dowhanick JJ, McBride AA, Lowly PM. Suppression of cellular proliferation by the papillomavirus E2 protein. J Virol 1995;69:7791–9. 5. Kobayashi Y, Yoshinouchi M, Tianqi G, Nakamura K, Hongo A, Kamimura S, et al. Presence of human papillomavirus DNA in pelvic lymph nodes can predict unexpected recurrence of cervical cancer in patients with histologically negative lymph nodes. Clin Cancer Res 1998;4:979–83. 6. Cullen AP, Reid R, Campion M, Lorincz A. Analysis of the physical state of different human papillomavirus DNAs in intraepithelial and invasive cervical neoplasms. J Virol 1991;65:606–12. 7. Das BC, Sharma JK, Gopalakrishna V, Luthra UK. Analysis by polymerase chain reaction of the physical state of human papillomavirus type 16 DNA in cervical preneoplastic and neoplastic lesions. J Gen Virol 1992; 73:2327–36. 8. Boshart M, Gissmann L, Ikenberg H, Kleinheinz A, Scheurlen W, zur Hausen H. A new type of papillomavirus DNA, its presence in genital cancer biopsies and in cell lines derived from cervical cancer. EMBO J 1984;3:1151–7. 9. Howley PM. Presence and expression of human papillomavirus sequences in human cervical carcinoma cell lines. Am J Pathol 1985;119:361–6. 10. Yee C, Krishnan-Hewlett I, Backer CC, Schlegal R, Howley PM. Presence and expression of human papillomavirus sequences in human cervical carcinoma cell lines. Am J Pathol 1985;119:361–6.



11. Tsunokawa Y, Takebe N, Nozawa S, Kasamatsu T, Gissmann L, zur Hausen H, et al. Presence of human papillomavirus type-16 and type-18 DNA sequences and their expression in cervical cancers and cell lines from Japanese patients. Int J Cancer 1986;37:499–503. 12. Vernon DD, Unger ER, Miller DL, Lee DR, Reeves WC. Association of human papillomavirus type 16 integration in the E2 gene with poor disease-free survival from cervical cancer. Int J Cancer 1997;74:50–6. 13. Du¨rst M, Kleinheinz A, Hotz M, Gissman L. The physical state of human papillomavirus type 16 DNA in benign and malignant genital tumours. J Gen Virol 1985;66: 1515–22. 14. Yiu KC, Huang DP, Chan MK, Foo W. The physical state of human papillomavirus type 16 DNA in cervical carcinomas of Hong Kong Chinese. Oncogene 1991;6: 1339–42. 15. Matsukura T, Koi S, Sugase M. Both episomal and integrated forms of human papillomavirus type 16 are involved in invasive cervical cancers. Virology 1989; 172:63–72. 16. Watts KJ, Thompson CH, Cossart YE, Rose BR. Sequence variation and physical state of human papillomavirus type 16 cervical cancer isolates from Australia and New Caledonia. Int J Cancer 2002;97:868–74. 17. Peitsaro P, Johansson B, Syrjanen S. Integrated human papillomavirus type 16 is frequently found in cervical cancer precursors as demonstrated by a novel quantitative real-time PCR technique. J Clin Microbiol 2002; 40:886–91. 18. Schwarz E, Freese U, Gissmann L, Mayer W, Roggenbuck B, Stremlau A, et al. Structure and transcription of human papillomavirus sequences in cervical carcinoma cells. Nature 1985;314:111–14. 19. Fukushima M, Yamakawa Y, Shimano S, Hashimoto M, Sawada Y, Fujinaga K. The physical state of human papillomavirus 16 DNA in cervical carcinoma and cervical intraepithelial neoplasia. Cancer 1990;66:2155–61. 20. Park JS, Hwang ES, Park SN, Ahn HK, Um SJ, Kim CJ, et al. Physical status and expression of HPV genes in cervical cancers. Gynecol Oncol 1997;65:121–9. 21. Pfister H, Fuchs PG. Relation of papillomaviruses to anogenital cancer. Dermatol Clin 1991;9:267–76. 22. Shirasawa H, Tomita Y, Kubota K, Kasai T, Sekiya S, Takamizawa H, et al. Detection of human papillomavirus type 16 DNA and evidence for integration into the cell DNA in cervical dysplasia. J Gen Virol 1986;67:2011–5. 23. Andersson S, Safari H, Mints M, Lewensohn-Fuchs I, Gyllensten U, Johansson B. Type distribution, viral load and integration status of high-risk human papillomaviruses in pre-stages of cervical cancer (CIN). Br J Cancer 2005;92:2195–200. 24. Shirasawa H, Tomita Y, Sekiya S, Takamizawa H, Simizu B. Integration and transcription of human papillomavirus type 16 and 18 sequences in cell lines derived from cervical carcinomas. J Gen Virol 1987;68:583–91. 25. Choo KV, 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;161:259–61. 26. Matsukura T, Kanda T, Furuno A, Yoshikawa H, Kawana T, Yoshiike K. Cloning of monomeric human papillomavirus type 16 DNA integrated within cell DNA from a cervical carcinoma. J Virol 1986;58:979–82. 27. Schneider-Maunoury S, Croissant O, Orth G. Integration of human papillomavirus type 16 DNA sequences: a possible early event in the progression of genital tumours. J Virol 1987;61:3295–8. 28. Wagatsuma M, Hashimoto K, Matsukura T. Analysis of integrated human papillomavirus type 16 DNA in cervi-

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

cal cancers: amplification of viral sequences together with cellular flanking sequences. J Virol 1990;64:813–21. Jeon S, Allen-Hoffmann BL, Lambert PF. Integration of human papillomavirus type 16 into the human genome correlates with a selective growth advantage of cells. J Virol 1995;69:2989–97. Cone RW, Minson AC, Smith MR, McDougall JK. Conservation of HPV-16 E6/E7 ORF sequences in a cervical carcinoma. J Med Virol 1992;37:99–107. Baker CC, Phelps WC, Lindgren V, Braun MJ, Gonda MA, Howley PM. Structural and transcriptional analysis of human papillomavirus type 16 sequences in cervical carcinoma cell lines. J Virol 1987;61:962–71. Choo KB, Lee HH, Pan CC, Wu SM, Liew LN, Cheung WF, et al. Sequence duplication and internal deletion in the integrated human papillomavirus type 16 genome cloned from a cervical carcinoma. J Virol 1988;62: 1659–66. Jeon S, Lambert PF. Integration of human papillomavirus type 16 DNA into the human genome leads to increased stability of E6 and E7 mRNAs: implications for cervical carcinogenesis. Proc Natl Acad Sci USA 1995;92:1654–8. Romanczuk H, Thierry F, Howley PM. Mutational analysis of cis elements involved in E2 modulation of human papillomavirus type 16 P99 and type 18 P105 promoters. J Virol 1990;64:5240–9. Thierry F, Howley PM. Functional analysis of E2 mediated repression of the HPV 18 P105 promoter. New Biol 1991;3:90–100. Bernard BA, Bailly C, Lenoir MC, Darmon M, Thierry F, Yaniv M. The human papillomavirus type 18 (HPV18) E2 gene product is a repressor of the HPV 18 regulatory region in human keratinocytes. J Virol 1989;63:4317–24. Karlsen F, Kalantari M, Jenkins A, Pettersen E, Kristensen G, Holm R, et al. Use of multiple PCR primer sets for optimal detection of human papillomavirus. J Clin Microbiol 1996;34:2095–100. Manos MM, Ting Y, Wright DK, Lewis AJ, Broker TR, Wolinsky SM. The use of polymerase chain reaction amplification for the detection of genital human papillomaviruses. Cancer Cells 1989;7:209–14. de Roda Husman AM, Walboomers JM, van den Brule AJ, Meijer CJ, Snijders PJ. The use of general primers GP5 and GP6 elongated at their 39 ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J Gen Virol 1995;76:1057–62. Snijders PJ, van den Brule AJ, Schrijnemakers HF, Snow G, Meijer CJ, Walboomers JM. The use of general primers in the polymerase chain reaction permits the detection of a broad spectrum of human papillomavirus genotypes. J Gen Virol 1990;71:173–81. Hildesheim A, Schiffman MH, Gravitt PE, Glass AG, Greer CE, Zhang T, et al. Persistence of type-specific human papillomavirus infection among cytologically normal women. J Inf Dis 1994;169:235–40. Gravitt PE, Peyton CL, Alessi TQ, Wheeler CM, Coutlee F, Hildesheim A, et al. Improved amplification of genital human papillomaviruses. J Clin Microbiol 2000;38: 357–61. Qu W, Jiang G, Cruz Y, Chang CJ, Ho GY, Klein RS, et al. PCR detection of human papillomavirus: comparison between MY09/MY11 and GP5q/GP6q primer systems. J Clin Microbiol 1997;35:1304–10. Coutle´e F, Gravitt P, Kornegay J, Hankins C, Richardson H, Lapointe N, et al. Use of PGMY primers in L1 consensus PCR improves detection of human papillomavirus DNA in genital samples. J Clin Microbiol 2002;40:902–7. Kleter B, van Doorn LJ, ter Schegget J, Schrauwen L, van Krimpen K, Burger M, et al. Novel short-fragment PCR assay for highly sensitive broad-spectrum detection of


46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

anogenital human papillomaviruses. Am J Pathol 1998;153:1731–9. Perrons C, Kleter B, Jelley R, Jalal H, Quint W, Tedder R. Detection and genotyping of human papillomavirus DNA by SPF10 and MY09/11 primers in cervical cells taken from women attending a colposcopy clinic. J Med Virol 2002;67:246–52. Monsonego J, Bohbot JM, Pollini G, Krawec C, Vincent C, Merignargues I, et al. Performance of the Roche AMPLICOR Human papillomavirus (HPV) test in prediction of cervical intraepithelial neoplasia (CIN) in women with abnormal PAP smear. Gynecol Oncol 2005; July. Gregoire L, Arella M, Campione-Piccardo J, Lancaster WD. Amplification of human papillomavirus DNA sequences by using conserved primers. J Clin Microbiol 1989;27:2660–5. Morris BJ, Flanagan JL, McKinnon KJ, Nightingale BN. Papillomavirus screening of cervical lavages by polymerase chain reaction. Lancet 1988;ii:1368. Dallas PB, Flanagan JL, Nightingale BN, Morris BJ. Polymerase chain reaction for fast, nonradioactive detection of high- and low-risk papillomavirus types in routine cervical specimens and in biopsies. J Med Virol 1989; 27:105–11. Morris BJ, Rose BR, Flanagan JL, McKinnon KJ, Loo CY, Thompson CH, et al. Automated polymerase chain reaction for papillomavirus screening of cervicovaginal lavages: comparison with dot-blot hybridization in a sexually transmitted diseases clinic population. J Med Virol 1990;32:22–30. Venturoli S, Bonvicini F, Cricca M, Gallinella G, Giosa F, Farinazzo F, et al. Evaluation of commercial kits for the detection and typing of human papillomavirus in cervical swabs. J Virol Methods 2002;105:49–56. Schlecht NF, Trevisan A, Duarte-Franco E, Rohan TE, Ferenczy A, Villa LL, et al. Viral load as a predictor of the risk of cervical intraepithelial neoplasia. Int J Cancer 2003;103:519–24. Lorincz AT, Reid R, Jenson AB, Greenberg MD, Lancaster W, Kurman RJ. Human papillomavirus infection of the cervix: relative risk associations of 15 common anogenital types. Obstet Gynecol 1992;79:328–37. Fujinaga Y, Shimada M, Okazawa K, Fukushima M, Kato I, Fujinaga K. Simultaneous detection and typing of genital human papillomavirus DNA using the polymerase chain reaction. J Gen Virol 1991;72:1029–44.

56. Molijn A, Kleter B, Quint W, van Doorn LJ. Molecular diagnosis of human papillomavirus (HPV) infections. J Clin Virol 2005;32(Suppl 1):S43–S51. 57. Noffsinger AE, Suzuk L, Hui YZ, Gal AA, Fenoglio-Preiser CM. Differential sensitivities of E6 type-specific and L1 consensus primers in the detection of human papillomavirus in anal carcinoma. Mod Pathol 1995;8:509–14. 58. Park JS, Leake JF, Sharma BK, Toki T, Kessis TD, Ambros RA, et al. Variability in b-globin and HPV DNA amplification by PCR from fixed tissues. Mod Pathol 1991;4:667–70. 59. Stoler MH, Rhodes CR, Whitbeck A, Wolinsky SM, Chow LT, Broker TR. Human papillomavirus type 16 and 18 gene expression in cervical neoplasias. Human Pathol 1992;23:117–28. 60. Du¨rst M, Gallahan D, Jay G, Rhim JS. Glucocorticoidenhanced neoplastic transformation of human keratinocytes by human papillomavirus type 16 and an activated ras oncogene. Virology 1989;173:767–71. 61. Munoz N, Bosch FX, de Sanjose S, Tafur L, Izarzugaza I, Gili M, et al. The causal link between human papillomavirus and invasive cervical cancer: a population-based case-control study in Colombia and Spain. Int J Cancer 1992;52:743–9. 62. Eluf-Neto J, Booth M, Munoz N, Bosch FX, Meijer CJ, Walboomers JM. Human papillomavirus and invasive cervical cancer in Brazil. Br J Cancer 1994;69:114–9. 63. Guerrero E, Daniel RW, Bosch FX, Castellsague X, Munoz N, Gili M, et al. Comparison of ViraPap, Southern hybridization, and polymerase chain reaction methods for human papillomavirus identification in an epidemiological investigation of cervical cancer. J Clin Microbiol 1992;30:2951–9. 64. Smits HL, Bollen LJ, Tjong-A-Hung SP, Vonk J, Van Der Velden J, Ten Kate FJ, et al. Intermethod variation in detection of human papillomavirus DNA in cervical smears. J Clin Microbiol 1995;33:2631–6. 65. Huang L-W, Chao S-L, Chen P-H, Chou H-P. Multiple HPV genotypes in cervical carcinomas: improved DNA detection and typing in archival tissues. J Clin Virol 2004;29:271–6. 66. Depuydt CE, Vereecken AJ, Salembier GM, Vanbrabant AS, Boels LA, van Herck E, et al. Thin-layer liquid-based cervical cytology and PCR for detecting and typing human papillomavirus DNA in Flemish women. Br J Cancer 2003;88:560–6. Received June 29, 2005, accepted September 5, 2005