Currently, human papillomavirus (HPV) DNA tests validated in large trials and epidemiological studies are the hybrid capture second-generation (HC2) HPV DNA assay and a variety of polymerase chain reaction (PCR) protocols employing degenerate or consensus primers. This article describes the currently available technology for HPV detection and discusses novel technologies and their potential for large-scale screening. Ideally, an HPV test should allow detection of multiple HPV types, identify individual types, and provide quantitative information about the viral load of each individual type found. Moreover, it should be easy to perform, be highly reproducible, with a high specificity and sensitivity, and amenable for high throughput analysis and automation. Because we do not yet fully understand the true value of viral load and the biological relevance of the different HPV types, any HPV test should be able to detect the clinically relevant high-risk types with a sufficient sensitivity of at least 10 000 genome copies per sample. To validate the different current and future test systems and to compare inter-laboratory performance we urgently need reference samples, validated reagents, and standardized protocols.

1. Chapter 12: Human Papillomavirus Technologies
Thomas Iftner, Luisa Lina Villa
is then hybridized in solution with each of the probe cocktails
Currently, human papillomavirus (HPV) DNA tests vali- allowing the formation of specific HPV DNA-RNA hybrids.
dated in large trials and epidemiological studies are the hy- These hybrids are then captured by antibodies bound to the wells
brid capture second-generation (HC2) HPV DNA assay and of a microtiter plate that recognize specifically RNA-DNA hy-
a variety of polymerase chain reaction (PCR) protocols em- brids. After removal of excess antibodies and nonhybridized
ploying degenerate or consensus primers. This article de- probes, the immobilized hybrids are detected by a series of re-
scribes the currently available technology for HPV detection actions that give rise to a luminescent product that can be mea-
and discusses novel technologies and their potential for sured in a luminometer. The intensity of emitted light, expressed
large-scale screening. Ideally, an HPV test should allow de- as relative light units, is proportional to the amount of target
tection of multiple HPV types, identify individual types, and DNA present in the specimen, providing a semiquantitative mea-
provide quantitative information about the viral load of each sure of the viral load. The HC2 is currently available in a 96-well
individual type found. Moreover, it should be easy to per- microplate format, is easy to perform in clinical settings, and is
form, be highly reproducible, with a high specificity and suitable for automation. Furthermore, HC2 does not require spe-
sensitivity, and amenable for high throughput analysis and cial facilities to avoid cross-contamination, because it does not
automation. Because we do not yet fully understand the true rely on target amplification to achieve high sensitivity, as do
value of viral load and the biological relevance of the differ- PCR protocols. Often only the high-risk cocktail is used; this
ent HPV types, any HPV test should be able to detect the reduces time and cost of the test. The FDA-recommended cut-
clinically relevant high-risk types with a sufficient sensitivity off value for test-positive results is 1.0 relative light units
of at least 10 000 genome copies per sample. To validate the (equivalent to 1 pg HPV DNA per 1 ml of sampling buffer).
different current and future test systems and to compare Several studies have noted that the HC-2’s high-risk probe cock-
inter-laboratory performance we urgently need reference tail cross-reacts with HPV types that are not represented in the
samples, validated reagents, and standardized protocols. probe mix (3,14,15). Peyton and colleagues (3) found that HC2
[J Natl Cancer Inst Monogr 2003;31:80–8] using the high-risk probe at a 1.0-pg/ml cut-off detected HPV
types 53, 66, 67, 73, as well as other undefined types, and raising
the cut-off to 10.0 pg/ml did not eliminate the cross reactivity to
CURRENT TECHNOLOGY FOR HUMAN PAPILLOMAVIRUS types 53 and 67. Cross-reactivity of HC2 high-risk probe to HPV
DNA DETECTION OF GENITAL INFECTIONS types that have a significant risk for cervical cancer may be
Currently, human papillomavirus (HPV) DNA tests validated considered beneficial, but cross-reaction with low-risk types
in large trials and epidemiological studies are the hybrid capture causes false positive results and may decrease the specificity of
second-generation (HC2) and polymerase chain reaction (PCR)- the test (16).
based methods employing either MY09/11 or GP5/6 consensus PCR-Based Assays
primers (Table 1). Current testing for the 13 high-risk types, 16,
18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68, which are HPV DNA can be selectively amplified by a series of reac-
represented by the high-risk hybridization mix (B) of HC2, is tions that lead to an exponential and reproducible increase in the
performed for cervical cancer screening, because they are the viral sequences present in the biological specimen. Analysis of
most prevalent types found in cervical cancer worldwide [(12); the amplified products can be done in different ways including
see chapters 13 and 14]. In addition, more recent studies have gel electrophoresis, dot blot or line strip hybridization, and ul-
found types 66, 73, and MM4 (a novel type related to HPV82) timately can be coupled to direct DNA sequencing. The sensi-
to be present in squamous cell carcinomas of the cervix (13,13a), tivity and specificity of PCR-based methods can vary, depending
which might have to be included in future HPV test systems. It mainly on the primer sets, the size of the PCR product, reaction
also should be noted, however, that the adaptation of the high- conditions and performance of the DNA polymerase used in the
risk-type panel to geographically different HPV type preva- reaction, the spectrum of HPV DNA amplified and ability to
lences might enhance the specificity of the test. It might be a detect multiple types. PCR can theoretically produce one million
proper solution for developing countries with bad or no surveil- copies from a single double stranded DNA molecule after 30
lance programs in place to limit the number of high-risk HPV cycles of amplification. Therefore, care must be taken to avoid
types screened for to the geographically most prevalent types false-positive results derived from cross-contaminated speci-
found in cervical cancer instead of using the complete HC2
high-risk probe.
Hybrid Capture HPV DNA Assay Affiliations of authors: T. Iftner, Medical Virology, Section Experimental
Virology, University Hospital of Tuebingen, Tuebingen, Germany; L.L. Villa,
HC2 is based on hybridization in solution of long synthetic Virology, Ludwig Institute for Cancer Research, Sao Paulo, Brazil.
RNA probes complementary to the genomic sequence of 13 Correspondence to: Thomas Iftner, Ph.D., Institut fuer Medizinische Virolo-
high-risk (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) gie, Sektion Experimentelle Virologie, Universitatetsklinikum Tuebingen, Elf-
and five low-risk (6, 11, 42, 43, 44) HPV types, which are used riede-Aulhorn Str. 6, 72076 Tuebingen (e-mail: tsiftner@med.uni-tuebingen.de).
to prepare high (B) and low (A) probe cocktails that are used in Journal of the National Cancer Institute Monographs No. 31, © Oxford
two separate reactions. DNA present in the biological specimen University Press 2003, all rights reserved.
80 Journal of the National Cancer Institute Monographs No. 31, 2003

2. Table 1. Established HPV test technologies and performance characteristics*
Test Analytical No. of HPV
Designation Reference Probes/primers reaction product sensitivity, fg types detectable
Hybridization HC2 HPV DNA assay (1) Mixture of complementary genome length DNA/RNA hybrids 25–75 13
RNA probes
PCR MY09/11 Dot blot (2,3) Degenerate primers 450 bp 0.1–100 39
PCR PGMY09/11 reverse LBA (4,5) Mixture of consensus primers 450 bp 0.1 27
PCR GP5+/GP6+ EIA (6,7,8) Consensus primers 150 bp 0.5–10 20
ELISA system
PCR GP5+/6+ reverse LBA (9) Consensus primers 150 bp 0.5–10 37
PCR SPF-PCR reverse LiPA (10,11) Mixture of consensus primers 65 bp 0.1–10 43
*HC2 hybrid capture second-generation, HPV human papillomavirus, LBA line blot assay, LiPA reverse hybridization line probe assay, PCR
polymerase chain reaction, SPF-PCR short PCR fragment amplification and detection system.
mens or reagents. Several procedures are available to avoid this genotypes (10). The performance characteristics of these assays
potential problem in using PCR protocols for HPV DNA detec- are presented in Table 1.
tion. Because of its versatility and very high sensitivity many
Performance and Reproducibility of HPV DNA Tests
PCR systems are available, but care must be taken for the lack
of validation and comparison with established protocols like Several studies have carefully compared the performance of
those described below. the available assays (Tables 2 and 3). In general, there is an
The most widely used protocols employ consensus primers excellent concordance with regard to sensitivities and specifici-
that are directed to a highly conserved region of the L1 gene, and ties obtained in a large series of screening trials performed all
are potentially capable of detecting all mucosal HPV types (17). over the world with the HC2 assay (Table 2). This underlines the
Among these are the single pair of consensus primers GP5/6 importance of the availability of a commercial manufactured
(6,18) and its extended version GP5+/6+ (8) and the MY09/11 quality-controlled robust assay. Good agreement rates were also
degenerate primers (2) and its modified version, PGMY09/11 observed between PCR-based tests as MY09/11 and GP5+/6+
(4,5). Full distinction of roughly 40 types can be achieved by systems (Table 3). However, there are several conditions, such
hybridization with type-specific probes (4–8,10), that can be as the DNA extraction procedures, differences in sampling
performed in different formats, including line strip assays and methods and sample transport/storage, and especially the use of
microtiter plates which are amenable to automation. Another different polymerases for the PCR reactions that can affect test
pair of consensus primers is available that amplifies a smaller performance, as recently reported by Castle et al. (27). There-
fragment (65 bp compared with 150 bp for the GP primers and fore, validated protocols, reagents, and reference samples need
450 bp for MY09/11) of the L1 gene therefore potentially in- to be further developed and more generally used. Recently, a
creasing the sensitivity of the assay. This SPF-PCR (10,11) is collaborative study has been launched by the World Health Or-
designed to discriminate a broad spectrum of HPVs by reverse ganization to provide a tool for standardization of HPV DNA
line blot hybridization. However, because of the small size of the assays both for epidemiological studies and vaccine efficacy
amplified fragment, one can anticipate reduced discrimination trials in different regions of the world. The goal is to prepare and
ability as compared with the other systems described, although characterize a panel of HPV DNA reference samples that would
the authors reported discrimination between 43 different HPV enable different laboratories to assess and compare the analytical
Table 2. Clinical performance data for HC2 HPV DNA assay compared with cervical cytology*
HC2 Cervical cytology
Disease Positive Sensitivity, Specificity, Positive Sensitivity, Specificity,
Reference Objective Country Population threshold threshold % % threshold % %
(19) ASCUS triage United 957 women with CIN2 1.0 pg/ml 89.2 64.1 ASCUS 76.2 N/A
States ASCUS cytology
(20) ASCUS triage United 3488 women with CIN3 1.0 pg/ml 96.3 N/A ASCUS 85.3 N/A
States ASCUS cytology
(21) Screening France 1518 women— CIN2 0.2 pg/ml 100 85.2 ASCUS 85.3 94.9
routine screening
(22) Screening United 1703 women aged CIN2 1.0 pg/ml 95.2 95.1 Borderline 85.7 96.9
Kingdom 35—routine
screening
(23) Screening South 1365 previously CIN2 1.0 pg/ml 83.9 84.5 ASCUS 67.9 87.9
Africa unscreened
women
(24) Screening Costa Rica 1119 specimens CIN2 1.0 pg/ml 88.4 89.0 ASCUS 77.7 94.2
(24a) Screening Germany 7908 women— CIN2 1.0 pg/ml 97.8 93.8 Borderline 43.5 98.0
routine screening
*ASCUS atypical squamous cell of undetermined significance, CIN2 cervical intra-epithelial lesion grade 2, HC2 hybrid capture second-generation,
HPV human papillomavirus.
Journal of the National Cancer Institute Monographs No. 31, 2003 81

3. Table 3. Comparison of technologies for detection of HPV in clinical samples*
Ref Population No. Sample type Method 1 Method 2 Agreement Kappa Comments
(3) Screening 208 Cervical brush MY09/11 HC2 (HR) 71.1 0.58 Analysis restricted to HPV types
1.0 pg detected by both assays
HPV
DNA
(25) Participants from other 211 Cervicovaginal MY09/11 GP5+/6+ 94.7 0.79 Dilution series analysis indicated
research studies lavage samples (Southern blot) (Southern blot) a 3-log lower amplification
efficiency for type 35 by
MY09/11 and for types 53 and
61 by GP5+/6+.
(5) N/A 247 Cervicovaginal PGMY09/11 + MY09/11 + line 87.7 0.83
lavage specimens line blot blot assay
assay
(10) Gynecology outpatients 534 Cervical scrapes SPF-PCR GP5+/6+ 70.6 0.65
(normal cytology EIA (Southern blot)
to severe
dyskaryosis and
carcinoma)
(11) Gynecology outpatients 766 Cervical scrapes (697 SPF-PCR GP5+/6+ 69.0 0.77 HPV types 34, 53, 70, 74 not
normal, 69 ASCUS LiPA EIA(?) represented in GP5+/6+
cytology)
(26) Participants from other 422 Cervicovaginal MY09/11 GP5+/6+ 96 0.8 GP5+/6+ detected by far less
research studies lavage (211 Dot blot Dot blot multiple infections. Differences
samples) in detection systems for types
35, 53, and 61.
*EIA enzyme immunoassay, HPV human papillomavirus, HR high risk, LiPA reverse hybridization line probe assay, SPF-PCR short PCR
fragment.
sensitivity and specificity of the assay in use. The results are still Viral load determination. Viral load determination became
not available. a methodological challenge since it has been suggested that high
Samples with multiple HPV types. Studying the role of copy numbers are correlated with increased risk of development
multiple infections in the development of high-grade cervical of HPV-associated cervical lesions. HPV DNA quantification in
disease or cervical cancer has been hampered by the fact that the the biological sample can be achieved by PCR-based methods
most frequently used HPV DNA test system, HC2, does not or by HC2 assay in a rather semiquantitative way (33–38). Es-
discriminate between different types in a multiple infection. timates of viral copy numbers depend directly on the total input
Moreover, not all PCR-based methods perform equally well in of cells, and ultimately of DNA, in the test. Therefore, adjust-
detecting multiple infections because of limitations in the num- ment for cell load is an absolute requirement that is frequently
ber of HPV types detectable and assay performance. It has been not fulfilled, as is the case for HC2 and some PCR-based pro-
shown for example that the GP5+/6+ system detects only 47% of tocols. Currently, there is no consensus about the best method to
samples with multiple HPV types in comparison to 90% de- quantify HPV in biological specimens. One of the most accurate
tected by MY09/11 PCR (26). and controlled assay available is Real-Time PCR (see below);
Another problem is the difference in sensitivity for distinctive but like any PCR-based methodology, it is subject to variations
HPV types between different test systems used and the repro- according to primer sequences, target DNA, detection method,
ducibility of different HPV tests for determining the exact HPV and so on. Moreover, it requires expensive equipment and re-
type in the sample (28,29). agents. Nevertheless, its use in epidemiological studies seems
In general, it seems that PCR systems using multiple primers to be warranted by the information provided. An alternative is a
such as PGMY09/11 and SPF-PCR are more robust to detect labor-intensive, Low Stringent-PCR with consensus primers
multiple infections than systems using single consensus primers (39) that has been shown to provide reliable information in
such as GP5+/6+. This may especially be true in cases of mixed epidemiological studies (40). It has been discussed that such
infections where one type is present in large amounts. The ki- methods with lower analytical sensitivities, including HC2,
netics of the PCR reaction when using single primer pairs is then could be able to detect only “high” enough viral loads, which
unfavorable for types present in smaller amounts. would be those clinically relevant (41,42). Together with the
Given that more accurate tools have now been developed for studies performed with Real-Time PCR, these studies provide
identifying multiple infections such as the reverse line blot as- an indication that high viral loads, at least for HPV 16, may
says, it would be worth establishing whether the presence of provide a mechanism to distinguish between clinically rele-
multiple infections/lesions would be a useful marker for persis- vant infections and those that are unlikely to progress. How-
tent infection and disease onset or progression. Interestingly, Ho ever, these results need to be extended and confirmed in fur-
et al. (30), investigating the natural history of cervicovaginal ther studies, including the dissection of differences in viral
papillomavirus infections in young women, defined an odds ra- load by genotype. Unfortunately, the true clinical relevance of
tio of 4.1 (2.7–6.3 95% CI) associated with the presence of viral load measurement may continue to be clouded by sampling
multiple types over a 6-month period for persistent HPV infec- error in the collection of cervical specimens. Even with normal-
tions. Two other studies show, on the contrary, that persistence ization to a cellular control, the lesion size and proportion
of HPV infection was independent of co-infection with other of infected to normal cells will be extremely difficult to con-
HPV types (31, 32). trol for.
82 Journal of the National Cancer Institute Monographs No. 31, 2003

4. Real-time PCR. Recently, PCR protocols based on a 5 -exo- Buffers and other transport media. Currently, there is a
nuclease assay and real-time detection of the accumulation of great interest in the development of collection media that could
fluorescence were developed. The release of fluorescence at preserve cells and macromolecules including DNA, RNA, and
each amplification cycle is directly proportional to the amount of proteins. The main reason is to be able to perform both mor-
amplicon generated and, therefore, it is considered to be an phological analysis and molecular tests from the very same
accurate method of estimating viral load. Moreover, the assay is specimen. Several methanol-based collection solutions are avail-
designed to keep contamination to a minimum (43,44). The Taq- able with the advantage that each collection bottle has approxi-
man quantitative PCR system has been reported for assessing mately 10–15 ml residual fluid after monolayer processing for
HPV viral load, while controlling for variation in the cellular cytological diagnosis. This amount of residual fluid should be
content of the sample by quantification of the nuclear gene for sufficient for several molecular tests involving DNA, RNA, or
beta-actin. Ylitalo and colleagues (45), using a nested case– immunohistochemical analysis of proteins performed in the re-
control design, found that cases had consistently higher viral sidual cells. In fact, in some settings, reflex DNA test (HPV test
loads for HPV 16 than controls and that high viral loads could be from specimens obtained for cytology) from liquid cytology
detected up to 13 years before the diagnosis of cervical cancer. cervical specimens is already in use.
As such, women with high viral loads for HPV 16 had a 30 times Lin et al. (51) have shown that the ThinPrep methanol fixa-
greater relative risk compared with women who were HPV nega- tion fluid-based collection of cervical cytological specimens pre-
tive, and this increased risk was consistent over time. Impor- serves cellular contents that can be used shortly after collection
tantly, this also applies to women under the age of 25, an age for detailed molecular analysis of DNA (HPV detection by PCR
group that has a particularly high prevalence of HPV infection and typing by DNA sequencing), RNA (expression of p53 and
and in which a method to distinguish clinically relevant infec- GAPDH by RT-PCR), and proteins (immunohistochemical
tions would be particularly valuable. A second study performed analysis of CEA antigen). Most of the assays were conducted
in the same population showed that 20% of the women with the
with samples stored for up to 30 days at room temperature, but
highest viral loads for HPV 16 had a 60 times higher risk of
it is possible that longer storage periods will still preserve the
developing carcinoma in situ than women who were HPV nega-
cellular contents. Tarkowski et al. (52) evaluated the recovery
tive (43). In addition, a recent publication by Woodman et al.
and detection of limiting amounts of high-risk HPV RNA from
(46) investigating the natural history of cervical HPV infection
cells fixed in liquid-based cytology media (PreservCyt-fixed).
in young women found that abnormal smears were significantly
Even after 1 year of storage at -20°C, RNA extracted from these
less likely to be associated with low-viral-load samples.
cells was suitable in RT–PCR assay for HPV-16 E6-E7 onco-
Recent developments exploit a multiplex format, including
genic transcripts.
more than one HPV type as well as targeting a cellular gene,
The new Universal collection medium (UCM) from Digene
which controls for the amount of input DNA (47). Recently, a
(Gaithersburg, MD) is a medium that is capable of preserving
variation of real-time PCR that uses self-probing amplicons
DNA, RNA, protein, as well as cell morphology. It is formulated
known as Scorpions has been suggested for detection, typing and
to be directly compatible with molecular testing with minimal or
quantification of more than 40 different HPV types in clinical
no processing. It is not inhibitory for hybrid capture when placed
samples (48). Although promising, the clinical application of
directly into the reaction. For PCR there is a need to purify DNA
methods based on real-time remains to be demonstrated.
in a spin column before the thermocycling. Preliminary date
Biological Specimen: Collection, Storage, and Processing indicate that it is capable of preserving RNA for several weeks
Sample quantity, quality and storage conditions, as well as at room temperature, but further studies are required to demon-
DNA preparation procedures can affect the performance of dif- strate its utility. Ambion (Austin, TX) has a product named
ferent HPV test systems. Generally, tests that use no primary RNAlater Tissue collection: RNA stabilization solution that is
amplification step, like HC2, are less affected by most of these an aqueous, nontoxic collection reagent that stabilizes and pro-
variables, whereas PCR-based procedures tolerate less well im- tects cellular RNA in intact, unfrozen tissue and cell samples.
purities because of their enzymatic nature. Therefore, it is de- Cells or tissues can be collected into RNAlater and stored at
sirable to use sample devices that allow the collection of a large room temperature for up to 1 week, at 4°C for up to 1 month, or
cell sample, like a cytobrush, and storage/transport media that at −20°C indefinitely. In our experience, it works very well for
not only preserve cell morphology but also stabilize DNA as tissues, not only for RNA studies but also in histopathological
well as RNA. diagnosis (better tissue preservation than formalin fixed). How-
Collection devices. Although a large variety of sample in- ever, there are no reports of the use of this medium to extract
struments for taking cervical swabs is available, preference RNA from smears or tissue culture cells. Moreover, this medium
should be given to devices that sample ecto- and endocervix at is not suitable for protein studies.
the same time, resulting in sufficient numbers of cells that are Importantly, there is a clear need for intensified public fund-
not sequestered by the instrument itself, as is the case for the ing to support validation studies of different transport mediums
cotton-tipped applicator. and collection devices instead of relying on information ob-
Further development of devices for the self-collection of tained and delivered by industry as it is mostly the case at
vaginal samples is warranted and field studies, especially in present.
older women that do not comply with existing screening pro- Archival specimens. Archival specimens, including fixed
grams, in developed countries have to be performed (49,50). smears and paraffin-embedded tissues, constitute an important
In addition, we need the development of special devices for source of materials for epidemiological retrospective studies.
sample taking on men. Studies have to be performed to define However, it is well known that time of fixation and the type of
the sensitivity threshold when testing samples from males for fixative used can considerably affect the quality of the extracted
HPV DNA. nucleic acids. Degradation of DNA and RNA is the most com-
Journal of the National Cancer Institute Monographs No. 31, 2003 83

5. mon type of damage. Formation of adducts and crosslinks with tives. Among the 247 samples examined, there was an HPV
proteins can also occur, which directly interfere with any enzy- prevalence of 62.8% using the PGMY09/11 compared with a
matic or hybridization reaction. Two HPV detection methods prevalence of 55.1% with the MY09/11 primer pair. There was
can be considered suitable for archival specimens: in situ hy- also a profound increase in the detection of multiple infections
bridization and PCR targeting a short region of the genes of using the PGMY09/11 primers, and in the amplification of cer-
interest. The sensitivities of these methods differ widely but, in tain HPV types that are inefficiently detected using MY09/11
general, PCR is more sensitive and reproducible. It is very dif- primers. This phenomenon has been observed in other studies
ficult to compare the results obtained in different series, because where certain HPV genotypes are amplified with higher effi-
of the diversity of protocols employed, both for DNA recovery ciency using PGMY09/11 compared to the MY09/11 (57). HPV
and amplification reaction, and specimen preservation. The typing is performed using a reverse hybridization line blot assay,
method of recovering DNA from the smear is an important which is based on the hybridization of the amplicon to specific
determinant of successful amplification of HPV DNA (53). In DNA probes that have been immobilized on nitrocellulose or
addition, procedures such as laser microdissection of target cells nylon strips. To provide type discrimination, probes for specific
from archival smears and tissues are currently widely used and HPV types are bound to the strip in individual parallel lines at
strongly recommended to increase the test sensitivity. defined positions and amplicon hybridization at a particular po-
A few studies have detected HPV in archival smears from sition thereby identifies the type. To detect hybridization, PCR
epidemiological studies, mostly by using a consensus PCR. De- amplification must be undertaken with primers having an at-
tection of high-risk HPV in archival smears was suggested as a tached biotin molecule, which becomes incorporated into the
means to reduce the rate of false negatives by cytology (54). amplicon. The amplicon is applied to the strips under conditions
McGoogan et al. (55) compared the performance of high-risk- that allow for specific hybridization with the immobilized probe.
HC2 assay with a consensus PCR in archival smears, and con- Once this has occurred, the presence of the amplicon is detected
cluded that both methods could be used, but that PCR was more with an alkaline phosphatase-labeled streptavidin conjugate,
sensitive. The utility of archival material for longitudinal studies whereby the streptavidin binds to the biotin on the amplicon and
of HPV presence was further corroborated by studies performed immobilizes the alkaline phosphatase in the specific region
in Greenland and Denmark (56) and in a prospective study con- where hybridization has occurred. The alkaline phosphatase
ducted in Sweden (45). The findings show the value of using catalyzes color formation upon addition of a substrate and a
archival specimens in epidemiological studies provided that op- colored line develops where amplicon has hybridized to a spe-
timal conditions for morphological classification, DNA extrac- cific probe. Measuring the position of the colored line relative to
tion, and gene amplification are established. In our experience an established base line allows the type to be identified.
this is not an easy task to accomplish.
GP5+/GP6+ System
NOVEL HPV TECHNOLOGIES FOR GENITAL SAMPLES Recently, a reverse line blot typing assay for the GP5+/6+
system capable of typing 37 mucosotropic HPVs has been de-
PGMY09/11 System
veloped, allowing for high-throughput testing both in epidemio-
The PGMY09/11 primer system was developed by Gravitt logical and clinical research (8).
et al. (4) to address some limitations in the traditional MY09/11 SPF-PCR System (Innogenetics)
degenerate primer system. The demonstrated sensitivity for the
PGMY09/11 primer system is 10 HPV genomes per PCR am- Kleter and colleagues (10) developed a short PCR fragment
plification for all representative genotypes. PGMY09/11 is com- amplification and detection system (SPF-PCR). A novel set of
prised of two non-degenerate pools of oligonucleotide primers primers (SPF1/2) was designed to target highly conserved re-
designed to amplify the same 450 bp region of the L1 gene as the gions of the HPV L1 gene and thereby allow for amplification of
original MY09/11 primers. Members of the primer pools were a broad spectrum of HPV types. Given that PCR amplification
chosen using sequence alignments of all known genital HPV efficiency is, in general, inversely related to the size of the
types and minimizing any potential mismatches while simulta- region amplified, the SPF1/2 primers target a small region of
neously minimizing the number of oligonucleotides in each only 65 bp, which should enhance sensitivity. However, because
pool. The upstream PGMY11 primer pool is composed of five of the small size of the fragment, type discrimination is more
oligonucleotides, whereas the PGMY09 pool contains 13. The complex.
PGMY primer system was evaluated using a reverse line-blot Amplicor MWP (Roche Molecular Diagnostics)
assay (5), which includes probes for 27 different HPV genotypes
(16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 55, 56, 58, 59, 68, MM4 This assay is based on a nondegenerate pool of primers to
[Novel HPV types: MM4 related to HPV82; MM7 to HPV83; amplify a short fragment of the L1 gene of 13 high-risk geno-
MM8 to HPV84; MM9 to HPV73], MM7, MM9 and 6, 11, 40, types (170 bp, compared with the 450 bp obtained with
42, 53, 54, 57, 66, MM8) along with two human beta-globin PGMY09/11). The amplicon is immobilized using a pool of
control lines. This reverse line-blot system is currently being capture molecules bound to the wells of a microtiter well plate
expanded to include 12 additional genital HPV genotypes (61, (MWP) and visualized by colorimetric detection by Roche Am-
62, 64, 67, 69, 70, 71, 72, CP6108 [Novel HPV types: CP6108 plicor chemistry. Moreover, the new test was developed to em-
related to HPV91; CP8304 related to HPV62; IS39 related to ploy the TaqGold DNA polymerase, which minimizes the
HPV82] CP8304, IS39) described by Peyton et al (15). In the amount of nonspecific amplification and increases the sensitivity
preliminary comparison of PGMY09/11 with the MY09/11 of the test. Because it amplifies a shorter fragment it is consid-
primer system, there was an overall agreement of 91.5%; how- ered to be more sensitive and amenable for less-preserved speci-
ever, the PGMY system picked up significantly more HPV posi- mens; it was reported that these primers detect about 13% more
84 Journal of the National Cancer Institute Monographs No. 31, 2003

6. HPV in cervical smears than the PGMY09/11 primers (Janet HPV SEROLOGICAL ASSAY
Kornegay, personal communication). However, because the new
Overall, frequency and titer of several types of serum anti-
primers were designed for high risk types, this test is not truly
bodies generated against HPV show a great variability that is
generic. An automated system, completely hands-off probed, is
dependent on the HPV type specificity, on the recognized epi-
under development with a capacity of performing 96 samples
topes, on the type of samples, and on the sensitivity of the assay.
a day.
Anti-HPV humoral immune responses are generally measured
Hybrid Capture-3 (Digene Company) by enzyme-linked immunoabsorbent assay (ELISA) with HPV
type-specific virus-like particles (VLPs) adsorbed in the plates.
The newly developed HC-3 assay (58) uses RNA probes as in Recombinant HPV proteins and peptides are also used in ELISA
HC-2, but in combination with biotinylated capture oligonucle- tests (59). Currently, several assays are available that exploit
otides that are directed to unique sequence regions within the different formats including ‘sandwich’ ELISA and radioimmu-
desired target. These oligonucleotides are only used for captur- noassays (60,61). These assays have been developed to detect
ing the desired target sequence into streptavidine-coated wells of with high specificity antibodies against the early proteins E6 and
a microtiter plate. Signals are generated by RNA probes that E7. Assay conditions may vary considerably and most published
hybridize to other regions of the captured sequences as the DNA studies do not provide information about standardization and
capture oligonucleotides. Moreover, the assay has been further quality control. The mere definition of cut-off values can hinder
developed to reduce unspecific hybridization by using “blocker comparison between different epidemiological surveys. How-
oligonucleotides” (unlabeled DNA molecules that are comple- ever, even an optimized HPV VLP-specific ELISA (62,63) will
mentary to the biotinylated capture oligonucleotides), aiming at still rely on the availability of an expensive and unstable antigen
eliminate cross-reactivity while maintaining specificity. The (the native recombinant virus-like particle or VLP). Therefore, it
same assay can in principle be used either for DNA or for RNA is highly desirable to develop alternative antigens for HPV se-
targets. Using this technique will also make it possible to test for rology as well as to validate positive and negative sera controls.
molecular variants of certain HPV types because even targets Several studies have shown that serological diagnosis of HPV
with single nucleotide exchanges can specifically be detected. infection using genetically engineered HPV capsids (also known
as virus-like particles or VLP) correlate well with HPV DNA
Rapid Capture System (Digene Company) presence in cervical smears. The antibodies invoked recognize
type-specific conformational epitopes present on VLPs, and the
For high-volume laboratory testing, Digene has described the
humoral response (IgG) against HPVs is stable over time. More-
development of an automated robotic platform for hybrid cap-
over, HPV VLP ELISAs show sensitivities between 50 and
ture called the Rapid Capture System (RCS; 58) that allows
60%, very high specificities (>90%), and good interlaboratory
robotic handling of 96-well microplates. This robot station per-
agreement (64). Therefore, VLP serology has been used as a
forms incubations, shakings, and washings. However, the dena-
marker of cumulative exposure to HPV and of sexual behavior.
turation of specimens in the sample device tubes still has to be
However, it has been shown that seroconversion may be delayed
performed by hand. This automatic station shall allow a single
or never occur in a subset of women testing positive for HPV
user to test 450 specimens in an 8-hour shift.
DNA. These data originate from several studies including epi-
HPV Genotyping Chips (BioMedlab Company) demiological surveys [reviewed in (59)].
The modest antibody responses measured in several studies
Biomedlab Company (Seoul, Korea) has developed an HPV may reflect lack of sensitivity of the assay or a deficient immune
Oligonucleotide Microarray-based detection system of HPV response to HPV, particularly in the case of cancer series where
types that currently allows the detection of 22 HPV types, by integration of HPV genomes impairs the expression of capsid
immobilizing HPV type-specific oligonucleotide probes and a antigens. Moreover, a weak immune response is often observed
control (beta-globin probe) on an aldehyde-derivatized slide in HPV infections because these viruses do not cause viremia.
glass. Target DNA is submitted to a standard PCR in the pres- As a consequence, one would expect misclassification as an
ence of fluoresceinated nucleotides (Cy5 or Cy3) employing important problem when considering VLP serology to estimate
primers for both the beta-globin (PC03/04) and for the L1 region the cumulative prevalence of HPV infection. Nevertheless,
(modified Gp5/6 primers) of several HPV types. Randomly la- stronger positive associations have been described for tumors of
beled PCR products are then hybridized onto the chip, which is the anogenital region when compared with seropositivity in pa-
then scanned by laser fluorescence. In the case of multiple in- tients with epithelial cancers in other anatomical locations,
fections, multiple hybridization signals can be seen. The utility which is consistent with the HPV DNA evidence (65,66). More-
and practicability of this method remains to be demonstrated, over, high titers of antibodies have been measured in women
although preliminary data (T. Ifter, personal communication) who have received HPV VLP vaccines (67,68). Another source
demonstrated high sensitivity and a high detection rate for of misclassification could originate from different viral loads
samples with multiple HPV types. present in lesions (66) or a differential immune response ac-
Ideally, a larger number of HPV type-specific oligonucleo- cording to anatomical site (69). In patients with head and neck
tides could be spotted easily on the Chip. In principle, such a tumors, it was found that among those seropositive all but one
system also allows semiquantitative analysis enabling at the had antibodies against HPV16 early proteins (70).
same time viral load analysis and HPV typing. However, signal Concerning type specificity, it has been demonstrated that
detection in microarrays is subject to variation, requires some VLPs of each HPV type induce serum antibody response that is
sort of input DNA normalization, and in its present format re- genotype-specific, with the exception of HPV types 6 and 11,
quires expensive equipment for signal detection, but it certainly which are cross-reactive, and HPV 31 and 45, which show low
has a large potential for further development. levels of cross-reactive antibodies against HPV 33 and 18, re-
Journal of the National Cancer Institute Monographs No. 31, 2003 85

7. spectively (71). Likewise, variants of HPV 16 have been dem- mavirus types associated with cervical cancer. N Engl J Med 2003;348:
onstrated to belong to the same serotype (72). Most HPV sero- 518–27.
prevalence studies have measured serum antibodies against a (14) Vernon et al., 2000 Vernon SD, Unger ER, Williams D. Comparison of
human papillomavirus detection and typing by cycle sequencing, line blot-
single HPV type, mostly against HPV-16 VLPs. Others have
ting, and hybrid capture. J Clin Microbiol 2000;38:651–5.
tested sera from both healthy women and cancer patients for (15) Peyton CL, Gravitt PE, Hunt WC, Hundley RS, Zhao M, Apple RJ, et al.
anti-VLP antibodies for HPV types 16, 18, 31, and 58 (73) or 16, Determinants of genital human papillomavirus detection in a US popula-
18, 31, 33, 39, 58, and 59 (74). Cross-reactivity was observed tion. J Infect Dis 2001;183:1554–64.
between HPV responses against 16 and 31, and between 58 and (16) Castle PE, Schiffman M, Burk RD, Wacholder S, Hildesheim A, Herrero R,
18, 45 and 59. Bratti MC, Sherman ME, Lorincz A. Restricted cross-reactivity of hybrid
It has been suggested that a search of immunoglobulins in capture 2 with nononcogenic human papillomavirus types. Cancer Epide-
genital secretions would provide a better indicator of HPV in- miol Biomarkers Prev 2002;11:1394–9.
(17) Bernard H, Chan S, Manos MM, Ong C, Villa LL, Delius H, et al. Iden-
fection, but reproducibility and standardization of the detection
tification and assessment of known and novel human papillomavirus by
methods are unresolved (75,76). polymerase chain reaction amplification, restriction fragment polymor-
phisms, nucleotide sequence, and phylogeny algorithms. J Infect Dis 1994;
REFERENCES 170:1077–85.
(18) Van den Brule AJ, Meijer CJ, Bakels V, Kenemans P, Walboomers JM.
(1) Lorincz A. Hybrid capture method for detection of human papillomavirus
¨ Rapid detection of human papillomavirus in cervical scrapes by combined
DNA in clinical specimens. Papillomavirus Rep 1996;7–15. general-primer mediated and type-specific polymerase chain reaction.
(2) Manos MM, Ting Y, Wright DK, Lewis AJ, Broker TR, Wolinsky SM. Use J Clin Microbiol 1990;28:2739–43.
of polymerase chain reaction amplification for detection of genital papil- (19) Manos MM, Kinney WK, Hurley LB, Sherman ME, Shieh-Ngai J, Kurman
lomavirus. Cancer Cells 1989;29:20–7. RJ, Ransley JE, Fetterman BJ, Hartinger JS, McIntosh KM, Pawlick GF,
(3) Peyton CL, Schiffman M, Lorincz AT, Hunt WC, Mielzynska I, Bratti C, Hiatt RA. Identifying women with cervical neoplasia: using human papil-
et al. Comparison of PCR- and hybrid capture-based human papillomavirus lomavirus DNA testing for equivocal Papanicolaou results. JAMA 1999;
detection systems using multiple cervical specimen collection strategies. 281:1605–10.
J Clin Microbiol 1998;36:3248–54. (20) Solomon D, Schiffman M, Tarone R. Comparison of three management
(4) Gravitt PE, Peyton CL, Apple RJ, Wheeler CM. Genotyping of 27 human strategies for patients with atypical squamous cells of undetermined sig-
papillomavirus types by using L1 consensus PCR products by a single- nificance: baseline results from a randomized trial. J Natl Cancer Inst 2001;
hybridization, reverse line blot detection method. J Clin Microbiol 1998; 93:293–9.
36:3020–7. (21) Clavel C, Masure M, Bory JP, Putaud I, Mangeonjean C, Lorenzato M,
(5) Gravitt PE, Peyton CL, Alessi TQ, Wheeler CM, Coutlee F, Hildesheim A, et al. Hybrid capture II-based human papillomavirus detection, a sensitive
et al. Improved amplification of genital human papillomaviruses. J Clin test to detect in routine high-grade cervical lesions: a preliminary study on
Microbiol 2000;38:357–61. 1518 women. Br J Cancer 1999;80:1306–11.
(6) De Roda Husman AM, Walboomers JM, van den Brule AJ, Meijer CJ, (22) Cuzick J, Beverley E, Ho L, Terry G, Sapper H, Mielzynska I, et al. HPV
Snijders PJ. The use of general primers GP5 and GP6 elongated at their 3 testing in primary screening of older women. Br J Cancer 1999;81:
ends with adjacent highly conserved sequences improves human papillo- 554–8.
mavirus detection by PCR. J Gen Virol 1995;76:1057–62. (23) Wright TC, Jr, Denny L, Kuhn L, Pollack A, Lorincz A. HPV DNA testing
(7) Jacobs MV, de Roda Husman AM, van den Brule AJ, Helmerhorst TJ, of self-collected vaginal samples compared with cytologic screening to
Sjniders PJ, Meijer CJ, Walboomers JM. Group-specific differentiation detect cervical cancer. JAMA 2000;283:81–6.
between high- and low-risk human papillomavirus genotypes by general (24) Schiffman M, Herrero R, Hildesheim A, Sherman ME, Bratti M, Wac-
primer-mediated PCR and two cocktails of oligonucleotide probes. J Clin holder S, et al. HPV DNA testing in cervical cancer screening: results from
Microbiol 1995;33:901–5. women in a high risk province of Costa Rica. JAMA 2000;283:87–93.
(8) Jacobs MV, Snijders PJ, van den Brule AJ, Helmerhorst TJ, Meijer CJ, (24a) Petry KU, Menton S, Menton M, van Loenen-Frosch F, de Carvalho
Walboomers JM. A general primer GP5+/GP6(+)-mediated PCR-enzyme Gomes H, Holz B, et al. Inclusion of HPV-testing in routine cervical cancer
immunoassay method for rapid detection of 14 high-risk and 6 low-risk screening for women above 29 years in Germany: results for 8466 patients.
human papillomavirus genotypes in cervical scrapings. J Clin Microbiol Br J Cancer. In press 2003.
1997;35:791–5. (25) Swan D, Tucker RA, Tortolero-Luna G, Follen MM, Wideroff, L, Unger
(9) Van den Brule AJ, Pol R, Fransen-Daalmeijer N, Schouls LM, Meijer CJ, ER, et al. Human papillomavirus (HPV) DNA copy number is dependent
Snijders PJ. GP5+/GP6+ PCR followed by reverse line blot analysis en- on grade of cervical disease and HPV type. J Clin Microbiol 1999;37:
ables rapid and high-throughput identification of human papillomavirus 1030–34.
genotypes. J Clin Microbiol 2002;40:779–87. (26) Qu W, Jiang G, Cruz Y, Chang CJ, Ho GY, Klein RS, et al. PCR detection
(10) Kleter B, van Doorn LJ, ter Schegget J, Schrauwen L, van Krimpen K, of human papillomavirus: comparison between MY09/MY11 and GP5+/
Burger M, et al. Novel short-fragment PCR assay for highly sensitive GP6+ primer systems. J Clin Microbiol 1997;35:1304–10.
broad-spectrum detection of anogenital human papillomaviruses. Am (27) Castle PE, Schiffman M, Gravitt PE, Kendall H, Fishman S, Dong H, et al.
J Pathol 1998;153:1731–9. Comparisons of HPV DNA detection by MY09/11 PCR methods. J Med
(11) Kleter B, van Doorn LJ, Schrauwen L, Molijn A, Sastrowijoto S, ter Scheg- Virol 2002;68:417–23.
get J, et al. Development and clinical evaluation of a highly sensi- (28) Jacobs MV, Snijders PJ, Voorhorst FJ, Dillner J, Forslund O, Johansson B,
tive PCR-reverse hybridization line probe assay for detection and identifi- et al. Reliable high risk HPV DNA testing by polymerase chain reaction:
cation of anogenital human papillomavirus. J Clin Microbiol 1999;37: an intermethod and intramethod comparison. J Clin Pathol 1999;52:498–
2508–17. 503.
(12) Bosch FX, Manos MM, Muñoz N, Sherman M, Jansen AM, Peto J, et al. (29) Konya J, Veress G, Juhasz A, Szarka K, Sapy T, Hernadi Z, et al. Addi-
Prevalence of human papillomavirus in cervical cancer: a worldwide per- tional human papillomavirus types detected by the hybrid capture tube test
spective. J Natl Cancer Inst 1995;87:796–802. among samples from women with cytological and colposcopical atypia.
(13) Manos MM, Waldman J, Zhang TY, Greer CE, Eichinger G, Schiffman J Clin Microbiol 2000;38:408–11.
MH, Wheeler CM. Epidemiology and partial nucleotide sequence of four (30) Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD. Natural history of
novel genital human papillomaviruses. J Infect Dis 1994;170:1096–9. cervicovaginal papillomavirus infection in young women. N Engl J Med
(13a) Munoz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X, Shah 1998; 338:423–8.
KV, et al. International Agency for Research on Cancer Multicenter Cer- (31) Rousseau MC, Pereira JS, Prado JCM, Villa LL, Rohan TE, Franco EL.
vical Cancer Study Group. Epidemiologic classification of human papillo- Cervical co-infection with human papillomavirus (HPV) types as a predic-
86 Journal of the National Cancer Institute Monographs No. 31, 2003

8. tor of acquisition and persistence of HPV infection. J Infect Dis 2001;184: (51) Lin WM, Ashfaq R, Michalopulos EA, Maitra A, Gazdar AF, Muller CY.
1508–17. Molecular Papanicolaou tests in the twenty first century: molecular analy-
(32) Liaw KL, Hildesheim A, Burk RD, Gravitt P, Wacholder S, Manos MM, ses with fluid-based Papanicolaou technology. Am J Obstet Gynecol 2000;
et al. A prospective study of human papillomavirus (HPV) type 16 DNA 183:39–45.
detection by polymerase chain reaction and its association with acquisition (52) Tarkowski TA, Rajeevan MS, Lee DR, Unger ER. Improved detection of
and persistence of other HPV types. J Infect Dis 2001;183:8–15. viral RNA isolated from liquid-based cytology samples. Mol Diagn 2001;
(33) Cuzick J, Terry G, Ho L, Hollingworth T, Anderson M. Type-specific 6:125–30.
human papillomavirus DNA in abnormal smears as a predictor of high- (53) Jacobs MV, Zielinski D, Meijer CJ, Pol RP, Voorhorst FJ, de Schipper
grade cervical intraepithelial neoplasia. Br J Cancer 1994;69:167–71. FA, et al. A simplified and reliable HPV testing of archival Papanico-
(34) Sherman ME, Schiffman M, Cox JT. Effects of age and human papil- laou-stained cervical smears: application to cervical smears from cancer
loma viral load on colposcopy triage: data from the randomized atypi- patients starting with cytologically normal smears. Br J Cancer 2000;82:
cal squamous cells of undetermined significance/low-grade squamous 1421–6.
intraepithelial lesion triage study (ALTS). J Natl Cancer Inst 2002;94: (54) Walboomers JM, Husman AM, Snijders PJ, Stel HV, Risse EK, Helmer-
102–7. horst TJ, et al. Human papillomavirus in false negative archival cervical
(35) Sun CA, Liu JF, Wu DM, Nieh S, Yu CP, Chu TY. Viral load of high-risk smears: implications for screening for cervical cancer. J Clin Pathol 1995;
human papillomavirus in cervical squamous intraepithelial lesions. Int 48:728–32.
J Gynaecol Obstet 2002;76:41–7. (55) McGoogan E, Seager AL, Cubie HA. Detection of high-risk human pap-
(36) Swan D, Tucker RA, Tortolero-Luna G, Follen Mitchell M, Wideroff, L, illomavirus nucleic acid in archival cervical smears. Acta Cytol 1998;42:
Unger ER, et al. Human papillomavirus (HPV) DNA copy number is 1079–83.
dependent on grade of cervical disease and HPV type. J Clin Microbiol (56) Svare EI, Kjaer SK, Smits HL, Poll P, Tjong-A-Hung SP, ter Schegget J.
1999;37:1030–4. Risk factors for HPV detection in archival Pap smears. A population-based
(37) Van Duin M, Snijders PJ, Schrijnemakers HF, Voorhorst FJ, Rozendaal L, study from Greenland and Denmark. Eur J Cancer 1998;34:1230–4.
Nobbenhuis MA, et al. Human papillomavirus 16 load in normal and ab- (57) Kornegay JR, Shepard AP, Hankins C, Lapointe N, Richardson H, Coutlee ´
normal cervical scrapes: an indicator of CIN II/III and viral clearance. Int F, et al. Non-isotopic detection of human papillomavirus DNA in clin-
J Cancer 2002;98:590–5. ical specimens using a consensus PCR and a generic probe mix in an
(38) Lorincz AT, Castle PE, Sherman ME, Scott DR, Glass AG, Wacholder S, enzyme-linked immunosorbent assay format. J Clin Microbiol 2001;39:
Rush BB, Gravitt PE, Schussler JE, Schiffman M. Viral load of human 3530–6.
papillomavirus and risk of CIN3 or cervical cancer. Lancet 2002;360: (58) Lorincz A, Anthony J. Advances in HPV detection by hybrid capture.
¨
228–9. Papillomavirus Rep. 2001;145–54.
(39) Caballero OL, Villa LL, Simpson AJG. Low stringency PCR (LS-PCR) (59) Konya J, Dillner J. Immunity to oncogenic human papillomaviruses. Adv
allows entirely internally standardized DNA quantitation. Nucleic Acids Cancer Res 2001;82:205–28.
Res 1995;23:192–3. (60) Sehr P, Muller M, Hopfl R, Widschwendter A, Pawlita M. HPV antibody
(40) Schlecht N, Trevisan A, Duarte-Franco E, Rohan TE, Ferenczy A, Villa detection by ELISA with capsid protein L1 fused to glutathione S-trans-
LL, et al. Viral load as a predictor of the risk of cervical intraepithelial ferase. J Virol Methods 2002;106:61–70.
neoplasia. Int J Cancer 2003;103:519–24. (61) Palker TJ, Monteiro JM, Martin MM, Kakareka C, Smith JF, Cook JC,
(41) Nobbenhuis MA, Helmerhorst TJ, van den Brule AJ, Rozendaal L, Voor- et al. Antibody, cytokine and cytotoxic T lymphocyte responses in chim-
horst FZ, Bezemer PD, et al. Cytological regression and clearance of high- panzees immunized with papillomavirus-like particles. Vaccine 2001;19:
risk human papillomavirus in women with an abnormal cervical smear. 3733–43.
Lancet 2001;358:1782–3. (62) Karem KL, Poon AC, Bierl C, Nisenbaum R, Unger E. Optimization of a
(42) Zielinski GD, Snijders PJ, Rozendaal L, Voorhorst FJ, van der Linden HC, human papillomavirus-specific enzyme-linked immunosorbent assay. Clin
Runsink AP, et al. HPV presence precedes abnormal cytology in women Diagn Lab Immunol 2002;9:577–82.
developing cervical cancer and signals false negative smears. Br J Cancer (63) Studentsov YY, Schiffman M, Strickler HD, Ho GYF, Pang YYS, Schiller
2001;85:398–404. J, et al. Enhanced enzyme-linked immunosorbent assay for detection of
(43) Josefsson A, Livak K, Gyllensten U. Detection and quantitation of human antibodies to virus-like parfticles of human papillomavirfus. J Clin Micro-
papillomavirus by using the fluorescent 5 exonuclease assay. J Clin Mi- biol 2002;40:1755–60.
crobiol 1999;37:490–6. (64) Strickler HD, Schiffman MH, Shah KV, Rabkin CS, Schiller JT, Wac-
(44) Swan DC, Tucker RA, Holloway BP, Icenogle JP. A sensitive, type- holder S, et al. A survey of human papillomavirus 16 antibodies in patients
specific, fluorogenic probe assay for detection of human papillomavirus with epithelial cancers. Eur J Cancer Prev 1998;7:305–13.
DNA. J Clin Microbiol 1997;35:886–91 (65) Bouwes Bavinck JN, Stark S, Petridis AK, Marugg ME, Ter Schegget J,
(45) Ylitalo N, Joseffson A, Melbye M, Sorensern P, Frisch M, Andersen PK, Westendorp RG, et al. The presence of antibodies against virus-like par-
et al. A prospective study showing long-term infection with human papil- ticles of epidermodysplasia verruciformis-associated human papillomavir-
lomavirus 16 before the development of cervical carcinoma in situ. Cancer fus type 8 in patients with actinic keratoses. Br J Dermatol 2000;142:
Res 2000;60:6027–32. 103–9.
(46) Woodman CB, Collins S, Winter H, Bailey A, Ellis J, Prior P, et al. Natural (66) Viscidi RP, Kotloff KL, Claymaan B, Russ K, Shapiro S, Shah KV. Preva-
history of cervical human papillomavirus infection in young women: a lence of antibodies to human papillomavirus (HPV) type 16 virus-like
longitudinal cohort study. Lancet 2001;357:1831–6. particles in relation to cervical infection among college women. Clin Diagn
(47) Tucker RA, Unger ER, Holloway BP, Swan DC. Real-time PCR-based Lab Immunol 1997;4:122–6.
fluorescent assay for quantitation of human papillomavirus types 6, 11, 16, (67) Brown DR, Bryan JT, Schroeder JM, Robinson, TS, Fife KH, Wheeler CM,
and 18. Mol Diagn 2001;6:39–47. et al. Neutralization of human papillomavirus type 11 (HPV-11) by serum
(48) Hart KW, Williams OM, Thelwell N, Fiander AN, Brown T, Borysiewicz from women vaccinated with yeast-derived HPV-11 L1 virus-like particles:
LK, et al. Novel method of detection, typing, and quantification of hu- correlation with competitive radioimmunoassay titer. J Infect Dis 2001;
man papillomaviruses in clinical samples. J Clin Microbiol 2001;39: 184:1183–6.
3204–12. (68) Koutsky LA, Ault KA, Wheeler CM, Brown DR, Barr E, Alvarez FB, et al.
(49) Hillemanns P, Kimmig R, Huttemann U, Dannecker C, Thaler CJ. Screen- A controlled trial of a human papillomavirus type 16 Vaccine. N Engl
ing for cervical neoplasia by self-assessment for human papillomavirus J Med 2002;347:1645–51.
DNA. Lancet 1999;354:1970. (69) Carter JA, Madeleine MM, Shera K, SchwartzSM, Cushing-Haugen KL,
(50) Gravitt PE, Lacey JV, Jr, Brinton LA, Barnes WA, Kornegay JR, Green- Wipf GC, et al. human papillomavirus 16 and 18 L1 serology compared
berg MD, et al. Evaluation of self-collected cervicovaginal cell samples for across anogenital sites. Cancer Res 2001;61:1934–40.
human papillomavirus testing by polymerase chain reaction. Cancer Epi- (70) Zumbach K, Hoffman M, Kahn T, Bosch F, Gottschlich S, Gorogh T, et al.
demiol Biomarkers Prev 2001;10:95–100. Antibodies against oncoproteins E6 and E7 of human papillomavirus types
Journal of the National Cancer Institute Monographs No. 31, 2003 87

9. 16 and 18 in patients with head and neck squamous-cell carcinoma. Int (74) Combita AL, Bravo MM, Touze AA, Orozco O, Coursaget P. Serologic
´
J Cancer 2000;85:815–8. response to human oncogenic papillomavirus types 16, 18, 31, 33, 39, 58
(71) Giroglou T, Sapp M, Lane C, Fligge C, Christensen ND, Streeck R, et al. and 59 virus-like particles in Colombian women with invasive cervical
Immunological analyses of human papillomavirus capsids. Vaccine 2001; cancer. Int J Cancer 2002;97:796–803.
19:1783–93. (75) Hildesheim A, McShane LM, Schiffman M, Bratti MC, Rodriguez AC,
(72) Pastrana D, Vass WC, Lowy DR, Schiller JT. NHPV 16 VLP vaccine Herrero R, et al. Cytokine and immunoglobin concentrations in cervical
induces human antibodies that neutralize divergent variants of HPV 16. secretions: reproducibility of the Weck-cell collection instrument and cor-
Virology 2001;279:361–9. relates of immune measures. J Immunol Methods 1999;225:131–43
(73) Touze A, de Sanjose S, Coursaget P, Almirall MR, Palacio V, Meijer (76) Snowhite IV, Jones WE, Dumestgre J, Dunlap K, Braly PS, Hagensee ME.
CJLM, et al. Prevalence of anti-human papillomavirus type 16, 18, 31 and Comparative analysis of methods for collection and measurement of cyto-
58 virus-like particles in women in the general population and in prosti- kines and immunoglobulins in cervical and vaginal secretions of HIV and
tutes. J Clin Microbiol 2001;39:4344–48 HPV infected women. J Immunol Methods 2002;263:85–95.
88 Journal of the National Cancer Institute Monographs No. 31, 2003