Journal of Clinical Virology 34 (2005) 76–80
Short communication
Quantitative analysis of human papillomavirus type 16 in
cervical neoplasm: A study in Chinese population
Keith Wing-Kit Lo a,∗ , Sze-Wan Yeung a , Tak-Hong Cheung a ,
Nelson Shing-Shun Siu a , Thomas Kahn b , Yick-Fu Wong a
a
Department of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Prince of
Wales Hospital, Shatin, Hong Kong, China
b Deutsche Bank AG, Expert Team Life Sciences, Frankfurt, Germany
Received 10 May 2005; received in revised form 20 May 2005; accepted 25 May 2005
Abstract
Background: Human papillomavirus (HPV) infection was recognized as a major causal factor for the development and progression of
squamous intraepithelial lesions (SIL). It is possible to use HPV test for the detection of cervical lesions as an adjunct to cervical cytology.
Objectives: To evaluate the relation between HPV 16 viral load and the severity of cervical lesions in a Chinese population.
Methods: Study population was recruited from the colposcopy and general outpatient clinic. The presence of HPV 16 E6 and E7 in cytological
specimens was detected using HPV 16 specific polymerase chain reaction (PCR). The viral load in the specimens that were positive for HPV
16 specific PCR, was quantified by using real-time PCR assay.
Results: The study recruited 394 women, in which 148 were high-grade SIL (HG-L), 121 were low-grade SIL (LG-L) and 125 were Normal.
Sufficient DNA integrity was proven in 347 samples. Among 121 positive cases for HPV 16, 70 were HG-L, 34 were LG-L and 17 were
Normal. Using quantitative real-time PCR, the percentages of samples with greater DNA copies were found to increase with the severity
of diseases. There was also a significant difference in DNA copies among the three groups (HG-L versus Normal, p < 0.001; HG-L versus
LG-L, p < 0.001). Area under receiver operating characteristic (ROC) curve of the HG-L versus LG-L and Normal was 0.836 indicating that
quantitative PCR had a good diagnostic value in differentiating HG-L from the LG-L and Normal groups.
Conclusions: Our data suggested HPV 16 viral load was significantly related to the severity of cervical lesions. Evaluation of viral burden
could be a potential clinical tool in management of cervical lesions.
© 2005 Elsevier B.V. All rights reserved.
Keywords: HPV; Squamous intraepithelial lesions; Viral load; Polymerase chain reaction
1. Introduction
Although cervical screening programmes have dramatically reduced the incidence of cervical cancer, 50% of invasive cervical cancers arise in women screened with existing
cytological methodologies due to some inherent limitations
Abbreviations: AUC, area under curves; HG-L, high-grade SILs; HPV,
human papillomavirus; LG-L, low-grade SILs; PCR, polymerase chain reaction; ROC, receiver operating characteristic; SIL, squamous intraepithelial
lesion
∗ Corresponding author. Tel.: +852 2632 2810; fax: +852 2636 0008.
E-mail address: [email protected] (K.W.-K. Lo).
1386-6532/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jcv.2005.05.006
(Cuzick et al., 1998). The sensitivity of cytology for highgrade squamous intraepithelial lesion (SIL) was only 40%
possibly due to sampling or interpretation errors (Cuzick et
al., 1995). On the other hand, over-reading of high-grade
SIL leads to unnecessary colposcopic examinations and overtreatment. Various researchers have suggested for additional
methods to improve the accuracy of smear, especially in
distinction between low-grade and high-grade disease. In
late 1970s, the hypothesis that human papillomavirus (HPV)
were related to cervical cancers was proposed and throughout 1980s gained rapid support. Those positive for HPV DNA
have a risk of developing cervical cancer 15–50 times higher
than those without HPV DNA (Lo et al., 2002). Similar
K.W.-K. Lo et al. / Journal of Clinical Virology 34 (2005) 76–80
correlation is found between HPV and SIL in epidemiological studies (Kulasingam et al., 2002). Therefore, it is possible
to use HPV test for the detection of cervical lesions as an
adjunct to cervical cytology. The present study was conducted
to establish a quantitative real-time polymerase chain reaction (PCR) assay for a precise determination of HPV 16 viral
load, and to investigate the role of viral load in the development of cervical precancerous lesions and invasive cancers.
2. Materials and methods
2.1. Specimens
Patients with normal and abnormal smears were recruited
with informed consents and ethical approval from the local
institute. All patients with abnormal smears were examined
colposcopically with biopsies when indicated. According to
WHO criteria, they were classified as high-grade SILs (HGL), low-grade SILs (LG-L) and Normal. Smears collected
by cervical brush were immersed into phosphate buffered
saline and centrifuged. The pellet at the bottom of the tube
was homogenized by TRIzol® Reagent for the extraction of
DNA.
77
0.5 unit of AmpliTaq Gold Polymerase (Applied Biosystems, NJ, USA) and the corresponding concentration of
primers. All reactions were performed in GeneAmp® PCR
System 2700 (Applied Biosystems, Foster City, USA). The
PCR products were analyzed by 2% agarose gel. Primer and
probe sequences for HPV 16 were selected using Primer
Express software package. Extracted CaSki DNA (American Type Culture Collection, Rockville, MD, USA) was
amplified by conventional PCR for HPV 16 E6 and E7
genes. The PCR products were recovered by gel extraction (DNA Purification Kit, Qiagen, USA) and the purified
DNA concentration was evaluated by spectrophotometer.
The purified DNA was used as the source for the standard
curve. The copy number of the DNA was calculated as the
formula: (bp size of double-stranded product) × (330 dalton
(Da) × 2 nt bp)/(Avogadro’s no.), where Da = g/mol and Avagadro’s no. = 6.023 × 1023 . Those smears with positive HPV
16 specific conventional PCR were subjected to the fluorogenic 5 exonuclease assay. After amplification, they were
analyzed by ABI Prism 7900HT Sequence Detection System
(Applied Biosystems, Foster City, USA). The fluorescence
collected by the system was interpreted by Sequence Detection System software.
2.2. PCR and real-time quantitative PCR
3. Results
DNA was tested for two genes, ␤-actin and HPV 16 E6/7,
by PCR. PCR was performed in 25 ␮l volumes containing
50 ng of DNA template, 1.5 mM MgCl2 , 200 ␮M deoxynucleotide triphosphates (Roche, Germany), 1× PCR buffer,
3.1. Grading of smears
The study design is summarized in Fig. 1. There were
269 cervical smears collected in colposcopic clinic. After
Fig. 1. Flow chart showing the study design and methodology.
78
K.W.-K. Lo et al. / Journal of Clinical Virology 34 (2005) 76–80
Table 1
HPV 16 E6/7 DNA copy number in HPV 16 positive cervical samples by real-time PCR analysis
Lesion groups
Copy no.a
0
1–10
10–102
102 –103
LG–Normal
%
30
58.8
10
19.6
3
5.9
5
9.8
2
3.9
0
–
1
2.0
51
–
HG-L
%
8
11.4
7
10
6
8.6
10
14.3
16
22.9
12
17.1
11
15.7
70
–
103 –104
104 –105
>105
No. of samplesb
LG–Normal: low-grade lesion group and Normal; HG-L: high-grade lesion group; %: percentage of samples in each copy number groups.
a Number of E6/7 DNA copies per 50 ng DNA. Bold numbers represent the largest percentage in each copy number category except zero.
b Total number of samples in each lesion groups.
colposcopic examination with biopsies, 148 cases with cervical cancer, CIN 2 or CIN 3 were classified as HG-L; 121
with CIN 1, HPV or inflammation were LG-L. Another 125
smears collected from women with normal cytology at outpatient clinic served as control.
3.2. Incidence of HPV 16
All the smears underwent DNA extraction and subsequently screened with PCR for ␤-actin and HPV 16 specific
primers. There were 347 positive cases for ␤-actin PCR reaction and 47 cases were excluded due to insufficient DNA
integrity. Among these ␤-actin positive cases, 130 were HGL, 100 cases were LG-L and 117 were Normal. These specimens then underwent HPV 16 specific PCR reaction, in which
121 cases were positive for HPV 16 E6/7 genes. Among these
cases, 70 were HG-L, 34 were LG-L and 17 were Normal.
The detection rates of HPV 16 in HG-L, LG-L and Normal
were 53.8%, 34.0% and 14.5%, respectively.
Fig. 2. Scatter plots of the HPV 16 DNA copy numbers of cervical specimens
in different lesion groups.
3.3. Quantification by real-time PCR
All specimens that were positive in HPV 16 specific PCR
were being quantified by quantitative real-time but some
cases showed no fluorescence signal as the no template control. Positive cases for HPV 16 E6/7 were 62 (88.6%) HG-L,
20 (58.8%) LG-L and 1 (5.9%) Normal. Table 1 shows
the HPV 16 E6/7 DNA copy number for different groups.
Fig. 2 shows the scatter plots of the three groups with corresponding HPV 16 E6/7 copy numbers per 50 ng of DNA
indicating that there was an increase trend from Normal to
HG-L progressively. Differences among these three groups
were determined by the Kruskal–Wallis test at 95% confidence interval which showed at least one difference between
these three groups (p < 0.001). Difference between two independent groups was then determined by the Mann–Whitney
Rank Test at 95% confidence interval that gave three combinations: HG-L versus Normal (p < 0.001), LG-L versus Normal (p = 0.001) and HG-L versus LG-L (p < 0.001). Receiver
operating characteristic (ROC) curves of HG-L versus Normal and HG-L versus LG-L and Normal are shown in Fig. 3.
Area under curves (AUC) indicates that quantitative PCR
has good diagnostic value in differentiating HG-L from the
Normal (AUC = 0.934) and HG-L from LG-L and Normal
(AUC = 0.836).
4. Discussion
The accumulated molecular and clinical evidences have
left no doubt that HPV directly influences the pathogenesis of
cervical neoplasia (Alani and Munger, 1998). Several markers have been proposed for the use in prediction of disease
severity of SILs. The presence of HPV DNA, the integration
state of HPV genome and the viral load were three major
areas currently under investigations.
A study showed that 75.4% of patients with abnormal
cytology were HPV-positive, a much higher proportion than
among those with normal cytology (Woodman et al., 2001).
It was considered that a positive oncogenic HPV test selects
a high-risk population for high-grade SIL, while a negative
test has a very good negative predictive value. However, this
approach was not comprehensive enough to define which
K.W.-K. Lo et al. / Journal of Clinical Virology 34 (2005) 76–80
Fig. 3. Receiver operating characteristic (ROC) curve of: (a) HG-L and (b)
HG-L vs. LG-L and Normal.
patient was in a higher risk, especially among the younger
population. WHO has estimated that 300 million people are
infected with HPV yearly around the world (de Villiers et
al., 1992). Despite this high prevalence, the incidence of SIL
and invasive cervical carcinoma in women infected with HPV
is relatively low. Indeed, in over 90% of those infected, the
79
virus is eliminated from the body as a result of the host’s
immune response. Therefore, the presence of HPV has a low
predictive value.
Apart from the type of HPV infection, the physical state is
a major factor contributing to malignant transformation and
disease progression, that is, integration of genomic sequences
of oncogenic HPV types into the cellular host genome (Park
et al., 1997). Integration of viral genomic sequences can cause
deletions and/or a disruption of E2 gene and result in a loss of
its function as a regulator of viral gene expression. This event
is often followed by up-regulation of E6/7 gene transcription. E6 and E7 proteins deregulate cell-cycle control through
interaction with different cell proteins thereby initiating the
transformation and immortalization of HPV-infected cells.
Various studies have demonstrated the presence of integrated
HPV 16 and 18 genomes in the vast majority of cancers and
in cell lines isolated from cervical malignancies (Park et al.,
1997; Pirami et al., 1997). Some studies also suggested that
the integration of HPV 16 already occurs in the precancerous
stage (Yoshinouchi et al., 1999; Peitsaro et al., 2002). Nevertheless, studies concerning the integration of HPV DNA in the
preneoplastic and neoplastic cervical lesions have produced
contrasting results by detecting a number of cervical malignancies predominately containing episomal forms of HPV 16
genomic sequences (Fuchs et al., 1989; Das et al., 1992).
Recently, the route for the exploration has shifted to the
quantity study on HPV. The viral load could be an alternative
to determine the risk of cancer development because high
viral load may result from an active viral replication that
might support viral persistence (Ylitalo et al., 2000). Study
has suggested that the amount of HPV could be an important
factor for progression to cervical cancer (Terry et al., 1993).
With increasing amount of HPV, the risk for development of
CIN as well as the severity of cervical disease is likely to
increase. Heavy viral load in CIN lesions has recently been
shown to constitute an at least 60-fold increased risk of CIN
III (Josefsson et al., 2000).
Although studies in Europe and North America had been
published (Cuzick et al., 1992; Josefsson et al., 2000), the
correlation between viral load and lesion severity is not yet
clear. In order to study the relationship of HPV viral load
and SIL of Chinese women in Hong Kong, we conducted the
study to investigate whether the load of HPV 16 was varied
with respect to the grade of SIL positively, and whether the
viral load of different grades of cervical disease was distinguishable. Our results showed that different levels of HPV
16 viral load were observed in different grades of cervical
disease; additionally, the viral load was changed simultaneously with the grade of the disease where the high viral load
was detected in HG-L and the low viral load was found in
Normal group.
We studied a single oncogenic type, HPV-16, as it is the
most common HPV type in the locality. While HPV-16 load
shows a clear relationship with cancer risk, other HPV types
appear to have weaker relationships between viral load and
risk of SIL. Viral load estimates based on summary mea-
80
K.W.-K. Lo et al. / Journal of Clinical Virology 34 (2005) 76–80
surements of HPV, such as those generated by the Hybrid
Capture II assay, may be blurred by not distinguishing HPV
types. On the other hand, given that there is a positive relation
between viral load and cancer risk for other oncogenic HPV
types, use of multiple assays of other HPV types associated
with cancer may substantially increase the positive predictive value. It is therefore important to select those HPV types
according to local prevalence. We chose the E6–E7 region
of the HPV genome since this region is highly conserved
among the oncogenic HPV types and is intact in both the
episomal and integrated forms of infectious HPV. The viral
loads measured were the representative of the viral persistence which could be sustained by viral integration and/or
maintenance of high viral loads, and evidence was given that
the viral persistence of oncogenic HPV types is essential for
the development of cervical lesions (Ho et al., 1995). As a
consequence, the higher the HPV load presents in the women,
the higher the risk for the development of cervical lesions.
This has already been proved in the current study that higher
HPV 16 viral load is found in high-grade CIN and cervical
carcinoma.
According to our results, HPV 16 viral load was highly
related to the clinical lesion grade. HPV 16 viral load detected
was higher in high-grade CIN and cancer cases (HG-L), lower
in low-grade CIN (LG-L) and the lowest in Normal cases.
There were also significant differences between these three
experimental groups (Kruskal–Wallis, p < 0.001). These findings encourage the establishment and development of diagnosing tool in clinical setting besides the current cytological
examinations. Therefore, evaluation of viral burden in conjunction with cytology may improve the detection of SILs in
cervical samples.
References
Alani RM, Munger K. Human papillomaviruses and associated malignancies. J Clin Oncol 1998;16:330–7.
Cuzick J, Meijer CJLM, Walboomers JMM. Screening for cervical cancer.
Lancet 1998;351:1439–40.
Cuzick J, Szarewski A, Terry G, Ho L, Hanby A, Maddox P, et al.
Human papillomavirus testing in primary cervical screening. Lancet
1995;345:1533–6.
Cuzick J, Terry G, Ho L, Hollingworth T, Anderson M. Human papillomavirus type 16 DNA in cervical smears as predictor of high-grade
cervical cancer. Lancet 1992;339:959–60.
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.
de Villiers EM, Wagner D, Schneider A, Wesch H, Munz F, Miklaw
H, et al. Human papillomavirus DNA in women without and with
cytological abnormalities: results of a 5 year follow-up study. Gynecol
Oncol 1992;44:33–9.
Fuchs PG, Girarid F, Pfister H. Human papillomavirus 16 DNA in cervical cancers and in lymph nodes of cervical cancer patients: a
diagnostic marker for early metastases? Int J Cancer 1989;43:41–
4.
Ho GY, Burk RD, Klein S, Kadish AS, Chang CJ, Palan P, et al.
Persistent genital human papillomavirus infection as a risk factor
for persistent cervical dysplasia. J Natl Cancer Inst 1995;87:1365–
71.
Josefsson AM, Magnusson PKE, Ylitalo N, Sorensen P, Qwarforth-Tubbin
P, Andersen PK, et al. Viral load of human papilloma virus 16 as a
determinant for development of cervical carcinoma in situ: a nested
case-control study. Lancet 2000;355:2189–93.
Kulasingam SL, Hughes JP, Kiviat NB, Mao C, Weiss NS, Kuypers J, et
al. Evaluation of human papillomavirus testing in primary screening
for cervical abnormalities: comparison of sensitivity, specificity, and
frequency of referral. JAMA 2002;288:1749–57.
Lo KWK, Wong YF, Chan MKM, Li JCB, Poon JS, Wang VW, et al.
Prevalence of human papillomavirus in cervical cancer: a multi-center
study in China. Int J Cancer 2002;100:327–31.
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.
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.
Pirami L, Giache V, Becciolini A. Analysis of HPV 16, 18, 31, and 35
DNA in pre-invasive and invasive lesions of the uterine cervix. J Clin
Pathol 1997;50:600–4.
Terry G, Ho L, Jenkins D, Hills M, Singer A, Mansell B, et al. Definition
of human papillomavirus type 16 DNA levels in low and high grade
cervical lesions by a simple polymerase chain reaction technique.
Arch Virol 1993;128:123–33.
Woodman CB, Collins S, Winter H, Bailey A, Ellis J, Prior P, et al.
Natural history of cervical human papillomavirus infection in young
women: a longitudinal cohort study. Lancet 2001;357:1831–6.
Ylitalo N, Sorensen P, Josefsson AM, Magnusson PKE, Andersen PK,
Ponten J, et al. Consistent high viral load of human papillomavirus
16 and risk of cervical carcinoma in situ: a nested case-control study.
Lancet 2000;355:2194–8.
Yoshinouchi M, Hongo A, Nakamura K, Kodama J, Itoh S, Sakai H,
et al. Analysis by multiplex PCR of the physical status of human
papillomavirus type 16 DNA in cervical cancers. J Clin Microbiol
1999;37:3514–37.