- Email: [email protected]
Comparison of the AMPLICOR Human Papillomavirus Test and the Hybrid Capture 2 Assay for detection of high-risk human papillomavirus in women with abnormal PAP smear
Available online at www.sciencedirect.com
Journal of Virological Methods 147 (2008) 10–17
Comparison of the AMPLICOR Human Papillomavirus Test and the Hybrid Capture 2 Assay for detection of high-risk human papillomavirus in women with abnormal PAP smear M.A. De Francesco a,∗,1 , F. Gargiulo a,1 , C. Schreiber b , G. Ciravolo b , F. Salinaro b , N. Manca a b
a Institute of Microbiology and Virology, Spedali Civili, University of Brescia, Brescia, Italy Department of Obstetrics and Ginecology, Spedali Civili, University of Brescia, Brescia, Italy
Received 7 April 2007; received in revised form 25 July 2007; accepted 30 July 2007 Available online 12 September 2007
Abstract Infection with human papillomavirus (HPV) is a necessary step in the progression to cervical cancer. Many methods for HPV testing are currently available, mostly developed to detect pools of HPV types. Hybrid Capture 2 (HC2) is one of the most widely used. A new PCR-based assay, the Roche AMPLICOR HPV test, has been recently developed. Both assays recognize a group of 13 HR HPV types contemporaneously. This study evaluated the performance of both methods for detecting high-grade cervical lesions as a part of management for abnormal PAP smears. The study population was composed of 213 women, all referred to colposcopy and histologic diagnosis following an abnormal PAP test. Biopsy-confirmed high-grade cervical intraepithelial neoplasia was used as a gold standard. Overall agreement was 84.9% with a kappa value of 0.6. When comparing the ability to detect moderate cervical intraepithelial neoplasia (CIN2+) and high-grade cervical intraepithelial neoplasia (CIN3+/cancer), AMPLICOR proved slightly more sensitive than HC2, a finding that is important when HPV testing is used in a triage of borderline smear results. Genotyping of discordant results showed a prevalence of LR-HPV types in HC2 positive/AMPLICOR negative samples, and a similar prevalence of HR- and LR-HPV types in AMPLICOR positive/HC2 negative samples. In conclusion, the study shows that the AMPLICOR assay is more sensitive than HC2, which makes it a valid alternative for routine clinical use. © 2007 Elsevier B.V. All rights reserved. Keywords: Typing; Commercial assays; Cervical cancer
1. Introduction Infection with human papillomavirus (HPV) is closely associated with the development of cancer and is recognized as a necessary step in the progression to neoplastic disease (Walboomers et al., 1999). Evidence from cohort studies also indicates that the risk of cervical neoplasia is greatest among women who develop persistent infections with high oncogenicrisk HPV genotypes (HR HPV) (Schecht et al., 2001; Kjaer et al., 2002).
∗ Corresponding author at: Institute of Microbiology, P. le Spedali Civili, 1, 25123 Brescia, Italy. Tel.: +39 030 3995860; fax: +39 030 3996071. E-mail address: [email protected] (M.A. De Francesco). 1 These authors contributed equally to this work.
0166-0934/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2007.07.023
HPV infection can lead to equivocal cytomorphologic changes referred to as atypical squamous cells (ASCs), definite cytologic signs of HPV infections termed low-grade squamous intraepithelial lesions (LSILs), or cytologic signs of a potential cancer precursor designated as high-grade squamous intra-epithelial lesions (HSILs). Cervical cytology has played a major role in cervical cancer screening for the past 50 years. Recently, however, HPV testing has been incorporated into screening guidelines (Saslow et al., 2002) and current recommendations include the option of dual cytology and HPV testing for women aged >30 years (Saslow et al., 2002). Several molecular methods are currently available for HPV testing (Hildesheim et al., 1994; Melchers et al., 1991; Roda Husman et al., 1995), although the Hybrid Capture 2 Assay is the most frequently used commercial test, based on liquid
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
hybridization and capable of detecting 13 HR HPV genotypes simultaneously (ALTS group, 2000). Recently a newly developed PCR-based technique, the Roche AMPLICOR HPV test, was introduced. This test is capable of detecting also 13 HR HPV genotypes, with simultaneous assessment of the presence of the human -globin gene as a positive control. Several studies have shown that HC2 has higher sensitivity and higher negative predictive value (NPV) than repeat PAP smear or immediate colposcopy for the detection of cervical intraepithelial neoplasia (CIN), thus reducing the diagnostic variability of equivocal PAP smears (Manos et al., 1999; Sherman et al., 2002; Clavel et al., 2001). On the other hand, due to its recent introduction, the performance of Roche AMPLICOR HPV test in predicting high-grade cervical intraepithelial neoplasia (CIN) has only been targeted by a few authors. The purpose of this study was to evaluate the performance of the AMPLICOR HPV test and the Digene HC2 assay in diagnosing high-grade disease in cervical samples from women attending a hospital for routine cervical screening after an abnormal PAP smear. The results obtained by both assays were compared with hystologic diagnosis. Genotyping was used as a confirmatory test for discordant samples. 2. Materials and methods 2.1. Study population Cervical scrape specimens were obtained from women aged 19–70 years (median age 35.6) who visited Brescia’s main hospital (Spedali Civili) for routine cervical screening between May 2005 and May 2006. An average of 3900 women were screened during this period. Of these, 3500 were considered eligible for the study if they fulfilled the following criteria: (a) were not currently pregnant and were at least 2 months postpartum, (b) had an intact uterus and no current referral for hysterectomy, (c) had never been treated for squamous intraepithelial lesions (SILs), and (d) had no history of chronic illness (e.g. renal failure, diabetes, cancer or gastrointestinal malabsorption). We included all patients with a PAP smear classified as abnormal. A total of 213 women out of 3500 (6.08%) met these criteria and were enrolled. Cytological diagnoses were reported according to TBS 1991 by scoring negative, atypical squamous cells of uncertain significance (ASCUS), atypical glandular cells of uncertain significance (AGUS), low squamous intraepithelial lesion (LSIL), high squamous intraepithelial lesion (HSIL), and invasive carcinoma. Specimens for the HC2 assay and AMPLICOR HPV test were collected from all women during the visit. They were all referred for colposcopic examination and histological diagnosis. All patients provided informed consent for the tests. 2.2. Colposcopy Colposcopic examination of the cervix was performed on all patients by experienced colposcopists, following a jointly agreed protocol. Lesions in the transformation zone (TZ) were
11
assessed by applying 5% acetic acid and iodine solution, under 8 × 12 magnification. If colposcopy proved unsatisfactory, further exploration of the endocervix was carried out under 20× magnification using a Koogan speculum. The international IFCPC nomenclature was used to classify the colposcopic pattern as either: normal; abnormal transformation zone (ATZ) with minor changes suggesting low-grade cervical intraepithelial neoplasia (CIN1); abnormal transformation zone with major changes suggesting cervical intraepithelial neoplasia (CIN 2-3); and cancer. 2.3. Biopsy and histology All 213 women underwent directed punch biopsy. Pathological diagnosis of cervical lesions was performed according to a five-grade framework: CIN I (grade I cervical intraepithelial neoplasia, characterized by condilomatous lesions and/or light dysplasia), CIN II (moderate dysplasia), CIN III (severe dysplasia), carcinoma in situ and invasive carcinoma reflecting the Bethesda report model. Histological diagnoses were used as the reference standard. 2.4. Sample collection Cervical samples were taken using a Cytobrush (Copan, Brescia). Specimens were collected in 1 ml of Specimen Transport medium (Digene) and transferred to the laboratory for HPV analysis. A 500 l aliquot of each sample was used for DNA extraction and Digene HPV DNA testing, while a 250 l aliquot of each sample was used for DNA extraction and subsequent testing by AMPLICOR HPV assay. 2.5. HC2 HPV assay HPV DNA testing by HC2 assay method was performed according to the manufacturer’s protocol, with signal amplification based on the production of DNA/RNA hybrids by a chemiluminescent reporter system. The results are given as relative light unit (RLU) ratio and are a semiquantitative estimate of viral load in the samples. The test is validated to detect approximately 4700 genome equivalents (or 1 pg/ml) of target HPV, represented by an RLU greater than or equal to the cutoff value calculated in each run by a series of standards. Measurements below the cutoff were scored as negative. The samples were analysed for the presence of HR HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68. Positive and negative controls were included in each run. 2.6. AMPLICOR HPV test The AMPLICOR HPV test (Roche Molecular Systems, Inc., Branchburg, NJ, USA) uses biotinylted primers to define a sequence of approximately 165 bp in length within the polymorphic L1 region of the HPV genome. The primers, pooled in the same PCR master mix are designed to amplify viral DNA from the same 13 types included in the HC2 assay. Using the
12
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
same mix, the -globin gene (268 bp amplicon) was amplified to test whether the DNA extracted was suitable for amplification. Amplification was performed following manufacturer’s instructions. Capture probes, representing regions internal to the amplified sequences, were used to identify viral or human DNA. Following PCR amplification, hybridization and detection were performed according to the manufacturer’s instructions.
0.0–0.02, 0.21–0.40, 0.41–0.60, 0.61–0.80, 0.81–0.99 and 1.0 indicate poor, slight, moderate, substantial, almost excellent, and excellent agreement, respectively (Fleiss, 1981). 3. Results 3.1. HPV prevalence by AMPLICOR and HC2 compared to that by cytology and histology
2.7. Linear Array HPV genotyping test The Linear Array HPV genotyping test (Roche Molecular Systems, Inc., Branchburg, NJ, USA) employs biotinylated primers to define a sequence of nucleotides within the polymporphic L1 region of the HPV genome, which is approximately 450 bp long. A pool of HPV primers is used to amplify HPV DNA from 37 HPV genotypes, including the same 13 highrisk genotypes detected by the AMPLICOR test. An additional primer pair targets the human -globin gene as a control for cell adequacy, extraction and amplification. Amplification was performed following the manufacturer’s instructions. Following PCR amplification, hybridization and detection were performed according to the manufacturer’s instructions. 2.8. Statistical analysis The sensitivity and specificity of cytology, of the AMPLICOR HPV test and the HC2 assay were determined against histological diagnosis. Performance indicators of cytology and HPV testing with both methods for detecting outcome variables (CIN1, CIN2, CIN3/cancer) were calculated using conventional contingency tables to calculate sensitivity, specificity, positive (PPV) and negative predictive value (NPV), with 95% confidence intervals. The absolute risk or positive predictive value for each independent test mode and its combinations in the histological diagnoses of CIN2+ and CIN3+ were evaluated. The respective 95% confidence intervals (95% CI) were also calculated. Agreement between the two assays was calculated by Cohen’s kappa statistic. In general, kappa values of
Table 1 shows the relationship between PAP smear, HPV detection with both assays, and histology. Most of the advanced lesions (CIN3/cancer) had a cytological diagnosis of high squamous intraepithelial lesion (38 out of 71, 53.5%), while 22.5% of samples classified as atypical squamous cells/atypical glandular cells of uncertain significance (ASCUS/AGUS). HPV positivity in the whole series was 73.3% for the Digene assay and 75.2% for AMPLICOR. HPV prevalence increased in parallel with lesion grade with both tests. Sixty-two out of 105 CIN 1 histological lesions were positive with both tests. Thirty-one (83.7%) out of 37 CIN2 were positive by HC2 and 34 (91.8%) were positive by AMPLICOR. Of 71 CIN3 positive samples, 63 (88.7%) were positive by HC2 and 64 (90.1%) were positive by AMPLICOR. The interrelationship between referral cytology, HPV detection and histology is reported in Table 2. The prevalence of HPV infection was 67.2% with AMPLICOR and 64% with HC2 in ASCUS/AGUS, while it was 74.1% with AMPLICOR and 69.8% with HC2 in low squamous intraepithelial lesion (LSIL) smears. About 10% (AMPLICOR) and 9.1% (HC2) of the PAP smears classified as ASCUS/AGUS and negative for HPV were diagnosed as CIN3/cancer. 3.2. Clinical sensitivity and specificity The performance traits of cytology (with LSIL cutoff), HC2 and AMPLICOR for the detection of high-grade cervical intraepithelial neoplasia (CIN2-3/cancer) are shown in Table 3. The highest sensitivity was obtained by AMPLICOR (90%).
Table 1 Cytologic abnormalities and HPV DNA detection rates related to biopsy results Histologic diagnosis CIN1 (N = 105)
CIN2 (N = 37)
CIN3/cancer (N = 71)
Total (N = 213) N
N
%
N
%
N
%
38 61 6
36 58 6
7 15 15
18 41 41
16 17 38
23 24 53
61 93 59
29 44 27
HPV DNA testing (Digene) Negative 43 Positive 62
41 59
6 31
16 84
8 63
11 89
57 156
27 73
HPV DNA testing (Roche) Negative 43 Positive 62
41 59
3 34
8 92
7 64
10 90
53 160
25 75
Cytologic diagnosis ASCUS/AGUS LSIL HSIL
%
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion.
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
13
Table 2 Interrelationship of referral cytology, HPV detection, and histology Cytology
HPV testing
Histologic diagnosis CIN1
CIN2
N
%
ASCUS
HPV − (Roche) HPV − (Digene) HPV + (Roche) HPV + (Digene)
17 17 21 21
85 77 50 53
LSIL
HPV − (Roche) HPV − (Digene) HPV + (Roche) HPV + (Digene)
22 23 39 38
HSIL
HPV − (Roche) HPV − (Digene) HPV + (Roche) HPV + (Digene)
4 3 2 3
N
CIN3/cancer
Total
%
N
%
N
1 3 6 4
5 14 15 10
2 2 14 14
10 9 35 37
20 22 41 39
92 82 56 58
0 1 15 14
0 3 22 22
2 4 15 13
8 15 22 20
24 28 69 65
44 42 4 6
2 2 13 13
22 29 26 25
3 2 35 36
34 29 70 69
9 7 50 52
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion. Table 3 The performance characteristics of PAP test (with LSIL cutoff), Roche and Digene HPV tests in detecting biopsy—confirmed high-grade CIN Diagnostic test
Sensitivity
Cytology HPV Digene test HPV Roche test
Specificity
PPV
NPV
%
95% CI
%
95% CI
%
95% CI
%
95% CI
78 87 90
71–85 81–93 85–95
36 41 41
27–45 32–50 32–50
56 60 61
49–63 53–67 54–68
63 75 75
51–75 64–86 64–86
CIN2, CIN3/cancer; CIN, cervical intraepithelial neoplasia; LSIL, low squamous intraepithelial lesion.
Both HPV DNA assays had higher sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) than cytology with LSIL cutoff. The overall risk of cervical intraepithelial neoplasia (CIN3+) among HPV-negative women was 13.2 % for AMPLICOR and 14% for HC2 but dropped to 10 and 9%, respectively, with the two HPV DNA tests among women with a cytological diagnosis of ASCUS (Table 4). However, the low absolute risk was largely due to the HPV-negativity, and in general, HPV-negative
test results predicted low risk for high-grade cervical intraepithelial neoplasia (CIN3+) regardless cytology results. At the other extreme, the overall risk among women with a high squamous intraepithelial lesion (HSIL) was 64.4 and 70% among HPV-positive women by AMPLICOR and 69.2% among those HPV-positive by HC2. Up to this point the addition of each test modality (cytology and HPV DNA testing) improved the predictive value in terms of CIN3+ outcome, particularly for HPV-positive women. As for the risk of CIN2+ diagnoses, the
Table 4 Risks (positive predictive values) for single and two-stage strategies with both assays for CIN3 + histological diagnosis Citology
HPV test negative
HPV test positive
N CIN3+/total
Risk (%)
95% CI
N CIN3+/total
Risk (%)
Roche ASCUS LSIL HSIL
16/61 17/93 38/59
26.2 18.2 64.4
15.2–37.2 11.2–29.3 58.4–70.4
2/20 0/24 3/9 7/53
10 0 33.3 13.2
Digene ASCUS LSIL HSIL
16/61 17/93 38/59
26.2 18.2 64.4
15.2–37.2 11.2–29.3 58.4–70.4
2/22 4/28 2/7 8/57
9 14.2 28.5 14
95% CI
N CIN3+/total
Risk (%)
95% CI
3.3–63.3 4.2–22.2
14/41 15/69 35/50 64/160
34.1 21.7 70 40
20–48 12.7–30.7 58–82 36–43
0–21 1.2–27.2 14.2–62.2 5–23
14/39 13/65 36/52 63/156
35.8 20 69.2 40.3
23.5–50.8 11–29 57.2–81.2 31.3–45.3
0–23
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion.
14
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
Table 5 Risks (positive predictive values) for single and two-stage strategies with both assays for CIN2 + histological diagnosis Citology
HPV test negative
HPV test positive
N CIN2+/total
Risk (%)
95% CI
N CIN2+/total
Risk (%)
95% CI
N CIN2+/total
Risk (%)
95% CI
Roche ASCUS LSIL HSIL
7/61 15/93 15/59
11.4 16.1 25.4
3.4–19.4 9.1–23.1 20.4–30.4
1/20 0/24 2/9 3/53
5 0 22 5.6
0–1.4 0–49 0–11.6
6/41 15/69 13/50 34/160
14.6 21.7 26 21.2
4.6–24.6 12.7–30.7 14–38 15.2–26.2
Digene ASCUS LSIL HSIL
7/61 15/93 15/59
11.4 16.1 25.4
3.4–19.4 9.1–23.1 20.4–30.4
3/22 1/28 2/7 6/57
13.6 3.5 28.5 10.5
0–27.6 0–9.5 0–61.5 6.5–14.5
4/39 14/65 13/52 31/156
10.2 21.5 25 19.8
1.2–19.2 11.6–31.4 14–36 13.8–25.8
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion.
low absolute risk was due to the HPV-negative AMPLICOR test (5.6%) as compared to HC2 (10.5%) (Table 5). Furthermore, the addition of the HPV DNA test to the cytology improved the overall predictive values for CIN2+ outcomes only with the AMPLICOR assay. HR HPV DNA was detected by HC2 in 156/213 (73.3%) cervical samples and by AMPLICOR in 160/213 (75.2%) cervical samples. Both assays gave positive results for 142 and negative results for 39 samples (Table 6). The AMPLICOR and HC2 assays thus yielded corroborative results for HR HPV DNA in 181/213 samples, demonstrating a concordance of 84.9% (Cohen’s kappa = 0.6). 3.3. Genotyping results of discordant samples For greater accuracy in determining the sensibility and specificity of the assays, discordant cases were resolved by Linear Array method. A total of 32 samples gave discordant results. Genotyping assay of the 14 cases positive by HC2 and negative by AMPLICOR identified genotypes in 11/14 samples (Table 7). In total, low-risk (LR) genotypes were observed in four cases, two of which showed coinfection with other LR genotypes. HPV genotypes 53, 66, 73, which are probably carcinogenic but not targeted by these two HPV DNA assays, were found in five cases: four of these showed coinfection with LR genotypes and one was a coinfection of two of these genotypes (HPV 66 and 73). Histology yielded a diagnosis of CIN 1 for most of these Table 6 Concordance between the results of the 213 samples analysed by the HC2 assay and the AMPLICOR HPV test HC2 assay result
Number of samples with AMPLICOR HPV test result Positive
Negative
Total
Positive Negative
142 18
14 39
156 57
Total
160
53
213
Cohen’s Kappa = 0.6.
14 patients, with one CIN 3 histology result and one patient diagnosed with an in situ carcinoma. Genotyping by Linear Array of the 18 samples negative with HC2 and positive with AMPLICOR is illustrated in Table 7. Genotypes were identified in 13/18 samples. Infection with only a single HR genotype was observed in one case; five samples had HR and LR coinfections; three samples had only single LR infections; one case had genotype 66 and three samples had coinfections with LR genotypes. In terms of histology results, nine patients had a diagnosis of CIN 1, four patients had a CIN2 result and five patients a CIN3. These results permitted the assessment of the sensitivity and specificity of both tests with the inclusion or exclusion of untyped specimens, on the assumption that all samples positive in both tests had a high-risk genotype, while all samples negative in both tests were HPV-negative or low risk. However, even assuming that untyped samples were false positives, a higher sensitivity and specificity was observed with AMPLICOR (100 and 81.5%, respectively) than with HC2 (95.9 and 78.4%, respectively). 4. Discussion This study compares the presence of HR HPV in cervical scrape specimens, using the AMPLICOR HPV assay and the HC2 method. Concordance between the two tests was 84.9%, with a Cohen’s kappa of 0.60, which indicates a substantial degree of correlation. The AMPLICOR had a slightly higher sensitivity than the HC2 assay (75.2 and 73.3%, respectively). This may be linked to the analytical sensitivity of the Roche AMPLICOR HPV test in detecting the 13 high-risk HPV types, which is higher (i.e. 480 copies/ml) than that of the HC2 assay (around 5000 copies/ml). Other studies have evaluated the performance of the AMPLICOR HPV test (Poljak et al., 2005; van Ham et al., 2005) compared to HC2 and another commercial PCR-based assay (SPF10-LiPA, Innogenetics), with an agreement of 85.9 and 97.5%, respectively. However, these studies do not correlate HPV testing with histological findings but only with cytological results. Two other recent studies (Sandri et al.,
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
15
Table 7 HPV genotyping of discordant samples Test results
Genotype(s)a
AMPLICOR test negative, HC2 assay positive
6, 53 42, 61, 67, 73 66, 73 62, 84 67 42 81, 11 Cp6108, 73, 67, 6 67, 73 Not typed Total
AMPLICOR test positive, HC2 assay negative
61, 51, 52, 16 73, 59 6 66 52, 53 42, 54 42 84, 66 16, 51, 66 16, 68 83, 84, 72, 40 39 84 Not typed Total
Number of specimens 1 1 1 1 1 1 1 1 1 5 14 1 1 1 1 1 1 1 1 1 1 1 1 1 5 18
Cytology results
Hystology results
LSIL LSIL LSIL LSIL LSIL LSIL ASCUS ASCUS HSIL HSIL (2), LSIL (2), ASCUS (1)
CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN3 CIN1 (4), CIS (1)
14
14
LSIL LSIL LSIL ASCUS LSIL LSIL LSIL LSIL LSIL ASCUS LSIL HSIL LSIL ASCUS (3), LSIL (2)
CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN2 CIN2 CIN3 CIN3 CIN3 CIN1 (4), CIN2 (1)
18
18
a
HR HPV genotypes are indicated in boldface type; CIN, cervical intraepithelial neoplasia; ASCUS, atypical squamous cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion CIS = carcinoma in situ.
2006; Carozzi et al., 2007) have compared AMPLICOR and HC2 performance: the former on samples from women attending a colposcopic clinic either for follow-up or for clinical purposes (with concordant results for 83% of samples) and the latter on archival HC2 positive samples collected over 2 years (89.2% concordance). Monsonego et al. (Monsonego et al., 2005) analysed the AMPLICOR performance both for the management of women with abnormal PAP tests and in opportunistic screening for cervical cancer. Their study (which did not involve another HPV DNA test) highlighted that the AMPLICOR HPV test, combined with abnormal colposcopy and HSIL cytology, is a powerful independent predictor of high-grade cervical intraepithelial neoplasia (CIN2-3). In the management of women with abnormal PAP smears, the diagnostic set-up usually starts with an abnormal PAP test leading to referral for colposcopy, biopsy taking and finally a histopathological examination of the biopsy (Syrj¨anen, 2000). This sequence of diagnostic steps can be complemented by adding HPV testing, as in this study. The accuracy of both HPV DNA assays was assessed for predicting high-grade cervical disease and observed that both methods showed higher sensitivity for detecting high-grade cervical intraepithelial lesions (CIN2+/CIN3+) than cytology, even if AMPLICOR had higher sensitivity compared with HC2 (91.8% versus 83.7% for CIN2+ lesions; 90.1% versus 88.7% for CIN3+/cancer lesions).
The clinical sensitivity of AMPLICOR may have been underestimated, because of the retrospective study design, which excluded patients whose samples were negative in cytology tests and were not referred for colposcopy but may still have included a small proportion of potential HPV DNA test positive CIN2+ lesions or higher lesions. Both tests missed about 10% of samples with a histological diagnosis of CIN3. However, neither the Hybrid Capture II assay nor the AMPLICOR HPV detect HPV 26, 53, 66, 73 and 82, which are types that were probably classified HR (26, 53, 66) or HR (73, 82) in a large epidemiological study of the oncogenic potential of HPV genotypes (Munoz et al., 2003) and some of these might be involved in the development of high-grade cervical lesions. The predictive positive value by CIN3+ outcome, a combination of the HPV DNA test (HC2 or AMPLICOR) and cytology surpassed that of an individual test; while for CIN2+, only the combination of AMPLICOR HPV and cytology surpassed that of an individual test. These results also underline that, in view of the highly negative predictive value of both HPV DNA assays, further testing is not likely to be needed within HPV-negative women (although follow-up and screening visits should continue according to current guidelines). It should be noted, however, that these results are not representative of the screened population at large. Risk estimates reported are higher than may be expected in the general pop-
16
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
ulation, most of whom are unlikely to have had an equivocal or mildly abnormal PAP test result, as indicated for this study population. Discordant results between the two tests were found in 15% of samples. Genotyping by the Linear Array method showed that most of the samples (9 out of 15, i.e. 64.7%) were positive with HC2 and negative using AMPLICOR had LR-HPV types—a finding consistent with previous reports of HC2 cross-reactions (Castle et al., 2002; Poljak et al., 2002; De Francesco et al., 2005). Among samples that were positive by AMPLICOR and negative by HC2, 6 out of 18 (33.3%) had HR HPV types, 5 out of 18 (27.7%) were not typed and 7 out of 18 (39%) had LR-HPV types. False negative results for HR HPV detection with HC2 have been reported (Cuzick et al., 1999; Kulmala et al., 2004). False positive results for HR HPV detection with AMPLICOR might be due both to the possibility that such samples contained a low copy number of an HR HPV genome not detected by the Linear Array method or to that genotypes other than those included in the panel may cross-react with AMPLICOR. The data suggest, therefore, that AMPLICOR is slightly more sensitive than the HC2 test and more accurate in detecting CIN2+ and CIN3+/cancer lesions—a finding confirmed by various previous studies (Sandri et al., 2006; Carozzi et al., 2007, 2005; Arbyn et al., 2004; Solomon and Schiffman, 2004). This slight difference may not be important in a screening setting, where high-grade lesions are rare events, but it becomes an important advantage in the triage of borderline smear results, where CIN2+ or CIN3+ lesions are more frequent and their correct identification is of key importance for the subsequent treatment of the patients. The use of HPV testing, with its higher sensitivity, is therefore a particularly suitable approach for patients with equivocal results (abnormal squamous cells of undetermined significance) in their cervicovaginal cytology tests, especially if associated with a genotyping assay. The risk of cancer development after HPV infection is of course type-specific and there are at least 45 HPV types capable of infecting the human genital tract (Munoz et al., 2003). There is yet no clear information on the contribution of intrinsic protumorigenic properties in cancer associated HPVs versus prevalence within the “at risk” population. Through HPV detection and typing it is therefore possible to discriminate between patients harbouring intermediate and highrisk types, who are advised to undergo more extensive/invasive procedures to rule out intraepithelial carcinoma, and patients negative for HPV or carrying low-grade HPV types, who can be spared such procedures and checked yearly by conventional cervical cytology instead. Acknowledgements This work was supported by Ministero dell’Universit`a e della Ricerca Scientifica e Tecnologica (MIUR) (40%). References Arbyn, M., Buntinx, F., van Ranst, M., Paraskevaidis, E., Martin-Hirsch, P., Dillner, J., 2004. Virologic versus cytologic triage of women with equivocal Pap
smears: a meta-analysis of the accuracy to detect high grade intraepithelial neoplasia. J. Natl. Cancer Inst. 96, 280–293. Carozzi, F., Bisanti, S., Sani, C., Zappa, M., Cecchini, S., Ciatto, S., Confortini, M., 2007. Agreement between the AMPLICOR human papillomavirus test and the Hybrid Capture 2 assay in detection of high-risk human papillomavirus and diagnosis of biopsy-confirmed high grade cervical disease. J. Clin. Microbiol. 45, 364–369. Carozzi, F.M., Confortini, M., Cecchini, S., Bisanzi, S., Cariaggi, M.P., Pontenani, G., Raspollini, M.R., Sani, C., Zappa, M., Ciatto, S., 2005. Triage with human papillomavirus testing of women with cytologic abnormalities prompting referral for colposcopic assessment. Cancer 105, 2–7. Castle, P.E., Lorincz, A.T., Mielzynska-Lohnas, I., Scott, D.R., Glass, A.G., Sherman, M.E., Schussler, J.E., Schiffman, M., 2002. Results of human papillomavirus DNA testing with the hybrid capture 2 assay are reproducible. J. Clin. Microbiol. 40, 1088–1090. Clavel, C., Masure, M., Bory, J.P., Putaud, I., Mangeonjean, C., Lorenzato, M., Nazeyrollas, P., Gabriel, R., Quereux, C., Birembaut, P., 2001. Human papillomavirus testing in primary screening for the detection of high grade cervical lesions a study of 7932 women. Br. J. Cancer 84, 1616–1623. Cuzick, J., Beverley, E., Ho, L., Terry, G., Sapper, H., Mielzynska, I., Lorincz, A., Chan, W.K., Krausz, T., Soutter, P., 1999. HPV testing in primary screening of older women. Br. J. Cancer 81, 554–558. De Francesco, M.A., Gargiulo, F., Schreiber, C., Ciravolo, G., Salinaro, F., Manca, N., 2005. Detection and genotyping of human papillomavirus in cervical samples of Italian patients. J. Med. Virol. 75, 588–592. Fleiss, J.L., 1981. Statistical Methods for Rates and Proportion, second ed. John Wiley & Sons Inc., New York. Hildesheim, A., Schiffman, M.H., Gravitt, P.E., Glass, A.G., Greer, C.E., Zhang, T., Scott, D.R., Rush, B.B., Lawler, P., Sherman, M.E., 1994. Persistence of type-specific human papillomavirus infection among cytologically normal women. J. Infect. Dis. 169, 235–240. Kjaer, S.K., van den Brule, A.J., Paull, G., Svare, E.I., Sherman, M.E., Thomsen, B.L., Suntum, M., Bock, J.E., Poll, P.A., Meijer, C.J., 2002. Type specific persistence of hig risk human papillomavirus (HPV) as indicator of high grade cervical squamous intraepithelial lesions in young women: population based prospective follow up study. BMJ 325, 572. Kulmala, S.M., Syrjanen, S., Shabalova, I., Petrovichev, N., Kozachenko, V., Podistov, J., Ivanchenko, O., Zakharenko, S., Nerovjna, R., Kljukina, L., Branovskaja, M., Grunberga, V., Juschenko, A., Tosi, P., Santopietro, R., Syrjanen, K., 2004. Human papillomavirus testing with the hybrid capture 2 assay and PCR as screening tools. J. Clin. Microbiol. 42, 2470– 2475. Manos, M.M., Kinney, W.K., Hurley, L.H., Sherman, M.E., Shieh-Ngai, J., Kurman, R.J., Ransley, J.E., Fetterman, B.J., Hartinger, J.S., McIntosh, K.H., Pawlick, G.F., Hiatt, R.A., 1999. Identifying women with cervical neoplasia using human papillomavirus DNA testing for equivocal Papanicolau results. JAMA 281, 1605. Melchers, W.J., Claas, H.C., Quint, W.G., 1991. Use of the polymerase chain reaction to study the relationship between human papillomavirus infections and cervical cancer. Eur. J. Clin. Microbiol. Infect. Dis. 10, 714–727. Monsonego, J., Bohot, J.M., Pollini, G., Krawec, C., Vincent, C., Merignargues, I., Haroun, F., Sednaoui, P., Monfort, L., Dachez, R., Syrjanen, K., 2005. Performance of the Roche AMPLICOR human papillomavirus (HPV) test in prediction of cervical intraepithelial neoplasia (CIN) in women with abnormal PAP smear. Gynecol. Oncol. 99, 160–168. Munoz, N., Bosch, F.X., de Sanjose, S., Herrero, R., Castellsague, X., shah, K.V., Snijders, P.J., Meijer, C.J., 2003. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N. Engl. J. Med. 348, 518–527. Poljak, M., Fujs, K., Seme, K., Kocjan, B.J., Vrtacnik-Bokal, E., 2005. Retrospective and prospective evaluation of the Amplicor HPV test for detection of 13 high risk human papillomavirus on 862 clinical samples. Acta Dermatovenerol. Alp. Panonica Adriat. 14, 147–152. Poljak, M., Marin, I.J., Seme, K., Vince, A., 2002. Hybrid Capture II HPV test detects at least 15 human papillomavirus genotypes not included in its current high-risk probe cocktail. J. Clin. Virol. 25 (Suppl. 3), S89–S97. Roda Husman, A.M., Walboomers, J.M., van den Brule, A.J., Meijer, C.J., Snijders, P.J., 1995. The use of general primers GP5 and GP6 elongated at
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17 their 3 ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J. Gen. Virol. 76, 1057–1062. Sandri, M.T., Lentati, P., Benini, E., Dell’Orto, P., Zorzino, L., Carozzi, F.M., Maisonneuve, P., Passerini, R., Salvatici, M., Casadio, C., Boveri, S., Sideri, M., 2006. Comparison of the Digene HC2 assay and the Roche AMPLICOR human papillomavirus (HPV) test for detection of high-risk HPV genotypes in cervical samples. J. Clin. Microbiol. 44, 2141–2146. Saslow, D., Runowicz, C.D., Solomon, D., Moscicki, A.B., Smith, R.A., Eyre, H.J., Cohen, C., 2002. American Cancer Society guideline for the early detection of cervical neoplasia and cancer. Cancer J. Clin. 52, 342– 362. Schecht, N.F., Kulaga, S., RobitailleJ, Ferreira, S., Santos, M., Miyamura, R.A., Duarte-Franco, E., Rohan, T.E., Ferenczy, A., Villa, L.L., Franco, E.L., 2001. Persistent human papillomavirus infetion as a predictor of cervical intraepithelial neoplasia. JAMA 286, 3106–3114. Sherman, M.E., Schiffman, M., Cox, J.T., 2002. Effect of age and human papilloma viral load on colposcopy triage: data from the randomised atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesion triage study (ALTS). J. Natl. Cancer Inst. 94, 102– 107.
17
Solomon, D., Schiffman, M., 2004. Have we resolved how to triage equivocal cervical cytology? J. Natl. Cancer Inst. 96, 250–251. Syrj¨anen, K.J., 2000. Management of abnormal PAP smears (MAPS): implications of terminology used in cytopathology. J. Low Genit. Tract Dis. 4, 217–223. The atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesions triage study (ALTS) group, 2000. Human papillomavirus testing for triage of women with cytologic evidence of low grade squamous intraepithelial lesions: baseline data from a randomised trial. J. Natl. Cancer Inst. 92, 397–402. van Ham, M.A.P.C., Bakkers, J.M.J.E., Harbers, G.K., Quint, W.G.V., Massuger, L.F.A.G., Melchers, W.J.G., 2005. Comparison of two commercial assay for detection of human papillomavirus (HPV) in cervical scrape specimens: validation of the Roche AMPLICOR HPV test as a means to screen for HPV genotypes associated with a higher risk of cervical disorders. J. Clin. Microbiol. 43, 2662–2667. Walboomers, J.M., Jacobs, M.V., Manos, M.M., Bosch, F.X., Kummer, J.A., Shah, K.V., Snijders, P.J., Peto, J., Meijer, C.J., Munoz, N., 1999. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J. Pathol. 189, 12–19.
Journal of Virological Methods 147 (2008) 10–17
Comparison of the AMPLICOR Human Papillomavirus Test and the Hybrid Capture 2 Assay for detection of high-risk human papillomavirus in women with abnormal PAP smear M.A. De Francesco a,∗,1 , F. Gargiulo a,1 , C. Schreiber b , G. Ciravolo b , F. Salinaro b , N. Manca a b
a Institute of Microbiology and Virology, Spedali Civili, University of Brescia, Brescia, Italy Department of Obstetrics and Ginecology, Spedali Civili, University of Brescia, Brescia, Italy
Received 7 April 2007; received in revised form 25 July 2007; accepted 30 July 2007 Available online 12 September 2007
Abstract Infection with human papillomavirus (HPV) is a necessary step in the progression to cervical cancer. Many methods for HPV testing are currently available, mostly developed to detect pools of HPV types. Hybrid Capture 2 (HC2) is one of the most widely used. A new PCR-based assay, the Roche AMPLICOR HPV test, has been recently developed. Both assays recognize a group of 13 HR HPV types contemporaneously. This study evaluated the performance of both methods for detecting high-grade cervical lesions as a part of management for abnormal PAP smears. The study population was composed of 213 women, all referred to colposcopy and histologic diagnosis following an abnormal PAP test. Biopsy-confirmed high-grade cervical intraepithelial neoplasia was used as a gold standard. Overall agreement was 84.9% with a kappa value of 0.6. When comparing the ability to detect moderate cervical intraepithelial neoplasia (CIN2+) and high-grade cervical intraepithelial neoplasia (CIN3+/cancer), AMPLICOR proved slightly more sensitive than HC2, a finding that is important when HPV testing is used in a triage of borderline smear results. Genotyping of discordant results showed a prevalence of LR-HPV types in HC2 positive/AMPLICOR negative samples, and a similar prevalence of HR- and LR-HPV types in AMPLICOR positive/HC2 negative samples. In conclusion, the study shows that the AMPLICOR assay is more sensitive than HC2, which makes it a valid alternative for routine clinical use. © 2007 Elsevier B.V. All rights reserved. Keywords: Typing; Commercial assays; Cervical cancer
1. Introduction Infection with human papillomavirus (HPV) is closely associated with the development of cancer and is recognized as a necessary step in the progression to neoplastic disease (Walboomers et al., 1999). Evidence from cohort studies also indicates that the risk of cervical neoplasia is greatest among women who develop persistent infections with high oncogenicrisk HPV genotypes (HR HPV) (Schecht et al., 2001; Kjaer et al., 2002).
∗ Corresponding author at: Institute of Microbiology, P. le Spedali Civili, 1, 25123 Brescia, Italy. Tel.: +39 030 3995860; fax: +39 030 3996071. E-mail address: [email protected] (M.A. De Francesco). 1 These authors contributed equally to this work.
0166-0934/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2007.07.023
HPV infection can lead to equivocal cytomorphologic changes referred to as atypical squamous cells (ASCs), definite cytologic signs of HPV infections termed low-grade squamous intraepithelial lesions (LSILs), or cytologic signs of a potential cancer precursor designated as high-grade squamous intra-epithelial lesions (HSILs). Cervical cytology has played a major role in cervical cancer screening for the past 50 years. Recently, however, HPV testing has been incorporated into screening guidelines (Saslow et al., 2002) and current recommendations include the option of dual cytology and HPV testing for women aged >30 years (Saslow et al., 2002). Several molecular methods are currently available for HPV testing (Hildesheim et al., 1994; Melchers et al., 1991; Roda Husman et al., 1995), although the Hybrid Capture 2 Assay is the most frequently used commercial test, based on liquid
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
hybridization and capable of detecting 13 HR HPV genotypes simultaneously (ALTS group, 2000). Recently a newly developed PCR-based technique, the Roche AMPLICOR HPV test, was introduced. This test is capable of detecting also 13 HR HPV genotypes, with simultaneous assessment of the presence of the human -globin gene as a positive control. Several studies have shown that HC2 has higher sensitivity and higher negative predictive value (NPV) than repeat PAP smear or immediate colposcopy for the detection of cervical intraepithelial neoplasia (CIN), thus reducing the diagnostic variability of equivocal PAP smears (Manos et al., 1999; Sherman et al., 2002; Clavel et al., 2001). On the other hand, due to its recent introduction, the performance of Roche AMPLICOR HPV test in predicting high-grade cervical intraepithelial neoplasia (CIN) has only been targeted by a few authors. The purpose of this study was to evaluate the performance of the AMPLICOR HPV test and the Digene HC2 assay in diagnosing high-grade disease in cervical samples from women attending a hospital for routine cervical screening after an abnormal PAP smear. The results obtained by both assays were compared with hystologic diagnosis. Genotyping was used as a confirmatory test for discordant samples. 2. Materials and methods 2.1. Study population Cervical scrape specimens were obtained from women aged 19–70 years (median age 35.6) who visited Brescia’s main hospital (Spedali Civili) for routine cervical screening between May 2005 and May 2006. An average of 3900 women were screened during this period. Of these, 3500 were considered eligible for the study if they fulfilled the following criteria: (a) were not currently pregnant and were at least 2 months postpartum, (b) had an intact uterus and no current referral for hysterectomy, (c) had never been treated for squamous intraepithelial lesions (SILs), and (d) had no history of chronic illness (e.g. renal failure, diabetes, cancer or gastrointestinal malabsorption). We included all patients with a PAP smear classified as abnormal. A total of 213 women out of 3500 (6.08%) met these criteria and were enrolled. Cytological diagnoses were reported according to TBS 1991 by scoring negative, atypical squamous cells of uncertain significance (ASCUS), atypical glandular cells of uncertain significance (AGUS), low squamous intraepithelial lesion (LSIL), high squamous intraepithelial lesion (HSIL), and invasive carcinoma. Specimens for the HC2 assay and AMPLICOR HPV test were collected from all women during the visit. They were all referred for colposcopic examination and histological diagnosis. All patients provided informed consent for the tests. 2.2. Colposcopy Colposcopic examination of the cervix was performed on all patients by experienced colposcopists, following a jointly agreed protocol. Lesions in the transformation zone (TZ) were
11
assessed by applying 5% acetic acid and iodine solution, under 8 × 12 magnification. If colposcopy proved unsatisfactory, further exploration of the endocervix was carried out under 20× magnification using a Koogan speculum. The international IFCPC nomenclature was used to classify the colposcopic pattern as either: normal; abnormal transformation zone (ATZ) with minor changes suggesting low-grade cervical intraepithelial neoplasia (CIN1); abnormal transformation zone with major changes suggesting cervical intraepithelial neoplasia (CIN 2-3); and cancer. 2.3. Biopsy and histology All 213 women underwent directed punch biopsy. Pathological diagnosis of cervical lesions was performed according to a five-grade framework: CIN I (grade I cervical intraepithelial neoplasia, characterized by condilomatous lesions and/or light dysplasia), CIN II (moderate dysplasia), CIN III (severe dysplasia), carcinoma in situ and invasive carcinoma reflecting the Bethesda report model. Histological diagnoses were used as the reference standard. 2.4. Sample collection Cervical samples were taken using a Cytobrush (Copan, Brescia). Specimens were collected in 1 ml of Specimen Transport medium (Digene) and transferred to the laboratory for HPV analysis. A 500 l aliquot of each sample was used for DNA extraction and Digene HPV DNA testing, while a 250 l aliquot of each sample was used for DNA extraction and subsequent testing by AMPLICOR HPV assay. 2.5. HC2 HPV assay HPV DNA testing by HC2 assay method was performed according to the manufacturer’s protocol, with signal amplification based on the production of DNA/RNA hybrids by a chemiluminescent reporter system. The results are given as relative light unit (RLU) ratio and are a semiquantitative estimate of viral load in the samples. The test is validated to detect approximately 4700 genome equivalents (or 1 pg/ml) of target HPV, represented by an RLU greater than or equal to the cutoff value calculated in each run by a series of standards. Measurements below the cutoff were scored as negative. The samples were analysed for the presence of HR HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68. Positive and negative controls were included in each run. 2.6. AMPLICOR HPV test The AMPLICOR HPV test (Roche Molecular Systems, Inc., Branchburg, NJ, USA) uses biotinylted primers to define a sequence of approximately 165 bp in length within the polymorphic L1 region of the HPV genome. The primers, pooled in the same PCR master mix are designed to amplify viral DNA from the same 13 types included in the HC2 assay. Using the
12
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
same mix, the -globin gene (268 bp amplicon) was amplified to test whether the DNA extracted was suitable for amplification. Amplification was performed following manufacturer’s instructions. Capture probes, representing regions internal to the amplified sequences, were used to identify viral or human DNA. Following PCR amplification, hybridization and detection were performed according to the manufacturer’s instructions.
0.0–0.02, 0.21–0.40, 0.41–0.60, 0.61–0.80, 0.81–0.99 and 1.0 indicate poor, slight, moderate, substantial, almost excellent, and excellent agreement, respectively (Fleiss, 1981). 3. Results 3.1. HPV prevalence by AMPLICOR and HC2 compared to that by cytology and histology
2.7. Linear Array HPV genotyping test The Linear Array HPV genotyping test (Roche Molecular Systems, Inc., Branchburg, NJ, USA) employs biotinylated primers to define a sequence of nucleotides within the polymporphic L1 region of the HPV genome, which is approximately 450 bp long. A pool of HPV primers is used to amplify HPV DNA from 37 HPV genotypes, including the same 13 highrisk genotypes detected by the AMPLICOR test. An additional primer pair targets the human -globin gene as a control for cell adequacy, extraction and amplification. Amplification was performed following the manufacturer’s instructions. Following PCR amplification, hybridization and detection were performed according to the manufacturer’s instructions. 2.8. Statistical analysis The sensitivity and specificity of cytology, of the AMPLICOR HPV test and the HC2 assay were determined against histological diagnosis. Performance indicators of cytology and HPV testing with both methods for detecting outcome variables (CIN1, CIN2, CIN3/cancer) were calculated using conventional contingency tables to calculate sensitivity, specificity, positive (PPV) and negative predictive value (NPV), with 95% confidence intervals. The absolute risk or positive predictive value for each independent test mode and its combinations in the histological diagnoses of CIN2+ and CIN3+ were evaluated. The respective 95% confidence intervals (95% CI) were also calculated. Agreement between the two assays was calculated by Cohen’s kappa statistic. In general, kappa values of
Table 1 shows the relationship between PAP smear, HPV detection with both assays, and histology. Most of the advanced lesions (CIN3/cancer) had a cytological diagnosis of high squamous intraepithelial lesion (38 out of 71, 53.5%), while 22.5% of samples classified as atypical squamous cells/atypical glandular cells of uncertain significance (ASCUS/AGUS). HPV positivity in the whole series was 73.3% for the Digene assay and 75.2% for AMPLICOR. HPV prevalence increased in parallel with lesion grade with both tests. Sixty-two out of 105 CIN 1 histological lesions were positive with both tests. Thirty-one (83.7%) out of 37 CIN2 were positive by HC2 and 34 (91.8%) were positive by AMPLICOR. Of 71 CIN3 positive samples, 63 (88.7%) were positive by HC2 and 64 (90.1%) were positive by AMPLICOR. The interrelationship between referral cytology, HPV detection and histology is reported in Table 2. The prevalence of HPV infection was 67.2% with AMPLICOR and 64% with HC2 in ASCUS/AGUS, while it was 74.1% with AMPLICOR and 69.8% with HC2 in low squamous intraepithelial lesion (LSIL) smears. About 10% (AMPLICOR) and 9.1% (HC2) of the PAP smears classified as ASCUS/AGUS and negative for HPV were diagnosed as CIN3/cancer. 3.2. Clinical sensitivity and specificity The performance traits of cytology (with LSIL cutoff), HC2 and AMPLICOR for the detection of high-grade cervical intraepithelial neoplasia (CIN2-3/cancer) are shown in Table 3. The highest sensitivity was obtained by AMPLICOR (90%).
Table 1 Cytologic abnormalities and HPV DNA detection rates related to biopsy results Histologic diagnosis CIN1 (N = 105)
CIN2 (N = 37)
CIN3/cancer (N = 71)
Total (N = 213) N
N
%
N
%
N
%
38 61 6
36 58 6
7 15 15
18 41 41
16 17 38
23 24 53
61 93 59
29 44 27
HPV DNA testing (Digene) Negative 43 Positive 62
41 59
6 31
16 84
8 63
11 89
57 156
27 73
HPV DNA testing (Roche) Negative 43 Positive 62
41 59
3 34
8 92
7 64
10 90
53 160
25 75
Cytologic diagnosis ASCUS/AGUS LSIL HSIL
%
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion.
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
13
Table 2 Interrelationship of referral cytology, HPV detection, and histology Cytology
HPV testing
Histologic diagnosis CIN1
CIN2
N
%
ASCUS
HPV − (Roche) HPV − (Digene) HPV + (Roche) HPV + (Digene)
17 17 21 21
85 77 50 53
LSIL
HPV − (Roche) HPV − (Digene) HPV + (Roche) HPV + (Digene)
22 23 39 38
HSIL
HPV − (Roche) HPV − (Digene) HPV + (Roche) HPV + (Digene)
4 3 2 3
N
CIN3/cancer
Total
%
N
%
N
1 3 6 4
5 14 15 10
2 2 14 14
10 9 35 37
20 22 41 39
92 82 56 58
0 1 15 14
0 3 22 22
2 4 15 13
8 15 22 20
24 28 69 65
44 42 4 6
2 2 13 13
22 29 26 25
3 2 35 36
34 29 70 69
9 7 50 52
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion. Table 3 The performance characteristics of PAP test (with LSIL cutoff), Roche and Digene HPV tests in detecting biopsy—confirmed high-grade CIN Diagnostic test
Sensitivity
Cytology HPV Digene test HPV Roche test
Specificity
PPV
NPV
%
95% CI
%
95% CI
%
95% CI
%
95% CI
78 87 90
71–85 81–93 85–95
36 41 41
27–45 32–50 32–50
56 60 61
49–63 53–67 54–68
63 75 75
51–75 64–86 64–86
CIN2, CIN3/cancer; CIN, cervical intraepithelial neoplasia; LSIL, low squamous intraepithelial lesion.
Both HPV DNA assays had higher sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) than cytology with LSIL cutoff. The overall risk of cervical intraepithelial neoplasia (CIN3+) among HPV-negative women was 13.2 % for AMPLICOR and 14% for HC2 but dropped to 10 and 9%, respectively, with the two HPV DNA tests among women with a cytological diagnosis of ASCUS (Table 4). However, the low absolute risk was largely due to the HPV-negativity, and in general, HPV-negative
test results predicted low risk for high-grade cervical intraepithelial neoplasia (CIN3+) regardless cytology results. At the other extreme, the overall risk among women with a high squamous intraepithelial lesion (HSIL) was 64.4 and 70% among HPV-positive women by AMPLICOR and 69.2% among those HPV-positive by HC2. Up to this point the addition of each test modality (cytology and HPV DNA testing) improved the predictive value in terms of CIN3+ outcome, particularly for HPV-positive women. As for the risk of CIN2+ diagnoses, the
Table 4 Risks (positive predictive values) for single and two-stage strategies with both assays for CIN3 + histological diagnosis Citology
HPV test negative
HPV test positive
N CIN3+/total
Risk (%)
95% CI
N CIN3+/total
Risk (%)
Roche ASCUS LSIL HSIL
16/61 17/93 38/59
26.2 18.2 64.4
15.2–37.2 11.2–29.3 58.4–70.4
2/20 0/24 3/9 7/53
10 0 33.3 13.2
Digene ASCUS LSIL HSIL
16/61 17/93 38/59
26.2 18.2 64.4
15.2–37.2 11.2–29.3 58.4–70.4
2/22 4/28 2/7 8/57
9 14.2 28.5 14
95% CI
N CIN3+/total
Risk (%)
95% CI
3.3–63.3 4.2–22.2
14/41 15/69 35/50 64/160
34.1 21.7 70 40
20–48 12.7–30.7 58–82 36–43
0–21 1.2–27.2 14.2–62.2 5–23
14/39 13/65 36/52 63/156
35.8 20 69.2 40.3
23.5–50.8 11–29 57.2–81.2 31.3–45.3
0–23
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion.
14
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
Table 5 Risks (positive predictive values) for single and two-stage strategies with both assays for CIN2 + histological diagnosis Citology
HPV test negative
HPV test positive
N CIN2+/total
Risk (%)
95% CI
N CIN2+/total
Risk (%)
95% CI
N CIN2+/total
Risk (%)
95% CI
Roche ASCUS LSIL HSIL
7/61 15/93 15/59
11.4 16.1 25.4
3.4–19.4 9.1–23.1 20.4–30.4
1/20 0/24 2/9 3/53
5 0 22 5.6
0–1.4 0–49 0–11.6
6/41 15/69 13/50 34/160
14.6 21.7 26 21.2
4.6–24.6 12.7–30.7 14–38 15.2–26.2
Digene ASCUS LSIL HSIL
7/61 15/93 15/59
11.4 16.1 25.4
3.4–19.4 9.1–23.1 20.4–30.4
3/22 1/28 2/7 6/57
13.6 3.5 28.5 10.5
0–27.6 0–9.5 0–61.5 6.5–14.5
4/39 14/65 13/52 31/156
10.2 21.5 25 19.8
1.2–19.2 11.6–31.4 14–36 13.8–25.8
CIN, cervical intraepithelial neoplasia; ASCUS/AGUS, atypical squamous cells of uncertain significance/atypical glandular cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion.
low absolute risk was due to the HPV-negative AMPLICOR test (5.6%) as compared to HC2 (10.5%) (Table 5). Furthermore, the addition of the HPV DNA test to the cytology improved the overall predictive values for CIN2+ outcomes only with the AMPLICOR assay. HR HPV DNA was detected by HC2 in 156/213 (73.3%) cervical samples and by AMPLICOR in 160/213 (75.2%) cervical samples. Both assays gave positive results for 142 and negative results for 39 samples (Table 6). The AMPLICOR and HC2 assays thus yielded corroborative results for HR HPV DNA in 181/213 samples, demonstrating a concordance of 84.9% (Cohen’s kappa = 0.6). 3.3. Genotyping results of discordant samples For greater accuracy in determining the sensibility and specificity of the assays, discordant cases were resolved by Linear Array method. A total of 32 samples gave discordant results. Genotyping assay of the 14 cases positive by HC2 and negative by AMPLICOR identified genotypes in 11/14 samples (Table 7). In total, low-risk (LR) genotypes were observed in four cases, two of which showed coinfection with other LR genotypes. HPV genotypes 53, 66, 73, which are probably carcinogenic but not targeted by these two HPV DNA assays, were found in five cases: four of these showed coinfection with LR genotypes and one was a coinfection of two of these genotypes (HPV 66 and 73). Histology yielded a diagnosis of CIN 1 for most of these Table 6 Concordance between the results of the 213 samples analysed by the HC2 assay and the AMPLICOR HPV test HC2 assay result
Number of samples with AMPLICOR HPV test result Positive
Negative
Total
Positive Negative
142 18
14 39
156 57
Total
160
53
213
Cohen’s Kappa = 0.6.
14 patients, with one CIN 3 histology result and one patient diagnosed with an in situ carcinoma. Genotyping by Linear Array of the 18 samples negative with HC2 and positive with AMPLICOR is illustrated in Table 7. Genotypes were identified in 13/18 samples. Infection with only a single HR genotype was observed in one case; five samples had HR and LR coinfections; three samples had only single LR infections; one case had genotype 66 and three samples had coinfections with LR genotypes. In terms of histology results, nine patients had a diagnosis of CIN 1, four patients had a CIN2 result and five patients a CIN3. These results permitted the assessment of the sensitivity and specificity of both tests with the inclusion or exclusion of untyped specimens, on the assumption that all samples positive in both tests had a high-risk genotype, while all samples negative in both tests were HPV-negative or low risk. However, even assuming that untyped samples were false positives, a higher sensitivity and specificity was observed with AMPLICOR (100 and 81.5%, respectively) than with HC2 (95.9 and 78.4%, respectively). 4. Discussion This study compares the presence of HR HPV in cervical scrape specimens, using the AMPLICOR HPV assay and the HC2 method. Concordance between the two tests was 84.9%, with a Cohen’s kappa of 0.60, which indicates a substantial degree of correlation. The AMPLICOR had a slightly higher sensitivity than the HC2 assay (75.2 and 73.3%, respectively). This may be linked to the analytical sensitivity of the Roche AMPLICOR HPV test in detecting the 13 high-risk HPV types, which is higher (i.e. 480 copies/ml) than that of the HC2 assay (around 5000 copies/ml). Other studies have evaluated the performance of the AMPLICOR HPV test (Poljak et al., 2005; van Ham et al., 2005) compared to HC2 and another commercial PCR-based assay (SPF10-LiPA, Innogenetics), with an agreement of 85.9 and 97.5%, respectively. However, these studies do not correlate HPV testing with histological findings but only with cytological results. Two other recent studies (Sandri et al.,
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
15
Table 7 HPV genotyping of discordant samples Test results
Genotype(s)a
AMPLICOR test negative, HC2 assay positive
6, 53 42, 61, 67, 73 66, 73 62, 84 67 42 81, 11 Cp6108, 73, 67, 6 67, 73 Not typed Total
AMPLICOR test positive, HC2 assay negative
61, 51, 52, 16 73, 59 6 66 52, 53 42, 54 42 84, 66 16, 51, 66 16, 68 83, 84, 72, 40 39 84 Not typed Total
Number of specimens 1 1 1 1 1 1 1 1 1 5 14 1 1 1 1 1 1 1 1 1 1 1 1 1 5 18
Cytology results
Hystology results
LSIL LSIL LSIL LSIL LSIL LSIL ASCUS ASCUS HSIL HSIL (2), LSIL (2), ASCUS (1)
CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN3 CIN1 (4), CIS (1)
14
14
LSIL LSIL LSIL ASCUS LSIL LSIL LSIL LSIL LSIL ASCUS LSIL HSIL LSIL ASCUS (3), LSIL (2)
CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN1 CIN2 CIN2 CIN3 CIN3 CIN3 CIN1 (4), CIN2 (1)
18
18
a
HR HPV genotypes are indicated in boldface type; CIN, cervical intraepithelial neoplasia; ASCUS, atypical squamous cells of uncertain significance; LSIL, low squamous intraepithelial lesion; HSIL, high squamous intraepithelial lesion CIS = carcinoma in situ.
2006; Carozzi et al., 2007) have compared AMPLICOR and HC2 performance: the former on samples from women attending a colposcopic clinic either for follow-up or for clinical purposes (with concordant results for 83% of samples) and the latter on archival HC2 positive samples collected over 2 years (89.2% concordance). Monsonego et al. (Monsonego et al., 2005) analysed the AMPLICOR performance both for the management of women with abnormal PAP tests and in opportunistic screening for cervical cancer. Their study (which did not involve another HPV DNA test) highlighted that the AMPLICOR HPV test, combined with abnormal colposcopy and HSIL cytology, is a powerful independent predictor of high-grade cervical intraepithelial neoplasia (CIN2-3). In the management of women with abnormal PAP smears, the diagnostic set-up usually starts with an abnormal PAP test leading to referral for colposcopy, biopsy taking and finally a histopathological examination of the biopsy (Syrj¨anen, 2000). This sequence of diagnostic steps can be complemented by adding HPV testing, as in this study. The accuracy of both HPV DNA assays was assessed for predicting high-grade cervical disease and observed that both methods showed higher sensitivity for detecting high-grade cervical intraepithelial lesions (CIN2+/CIN3+) than cytology, even if AMPLICOR had higher sensitivity compared with HC2 (91.8% versus 83.7% for CIN2+ lesions; 90.1% versus 88.7% for CIN3+/cancer lesions).
The clinical sensitivity of AMPLICOR may have been underestimated, because of the retrospective study design, which excluded patients whose samples were negative in cytology tests and were not referred for colposcopy but may still have included a small proportion of potential HPV DNA test positive CIN2+ lesions or higher lesions. Both tests missed about 10% of samples with a histological diagnosis of CIN3. However, neither the Hybrid Capture II assay nor the AMPLICOR HPV detect HPV 26, 53, 66, 73 and 82, which are types that were probably classified HR (26, 53, 66) or HR (73, 82) in a large epidemiological study of the oncogenic potential of HPV genotypes (Munoz et al., 2003) and some of these might be involved in the development of high-grade cervical lesions. The predictive positive value by CIN3+ outcome, a combination of the HPV DNA test (HC2 or AMPLICOR) and cytology surpassed that of an individual test; while for CIN2+, only the combination of AMPLICOR HPV and cytology surpassed that of an individual test. These results also underline that, in view of the highly negative predictive value of both HPV DNA assays, further testing is not likely to be needed within HPV-negative women (although follow-up and screening visits should continue according to current guidelines). It should be noted, however, that these results are not representative of the screened population at large. Risk estimates reported are higher than may be expected in the general pop-
16
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17
ulation, most of whom are unlikely to have had an equivocal or mildly abnormal PAP test result, as indicated for this study population. Discordant results between the two tests were found in 15% of samples. Genotyping by the Linear Array method showed that most of the samples (9 out of 15, i.e. 64.7%) were positive with HC2 and negative using AMPLICOR had LR-HPV types—a finding consistent with previous reports of HC2 cross-reactions (Castle et al., 2002; Poljak et al., 2002; De Francesco et al., 2005). Among samples that were positive by AMPLICOR and negative by HC2, 6 out of 18 (33.3%) had HR HPV types, 5 out of 18 (27.7%) were not typed and 7 out of 18 (39%) had LR-HPV types. False negative results for HR HPV detection with HC2 have been reported (Cuzick et al., 1999; Kulmala et al., 2004). False positive results for HR HPV detection with AMPLICOR might be due both to the possibility that such samples contained a low copy number of an HR HPV genome not detected by the Linear Array method or to that genotypes other than those included in the panel may cross-react with AMPLICOR. The data suggest, therefore, that AMPLICOR is slightly more sensitive than the HC2 test and more accurate in detecting CIN2+ and CIN3+/cancer lesions—a finding confirmed by various previous studies (Sandri et al., 2006; Carozzi et al., 2007, 2005; Arbyn et al., 2004; Solomon and Schiffman, 2004). This slight difference may not be important in a screening setting, where high-grade lesions are rare events, but it becomes an important advantage in the triage of borderline smear results, where CIN2+ or CIN3+ lesions are more frequent and their correct identification is of key importance for the subsequent treatment of the patients. The use of HPV testing, with its higher sensitivity, is therefore a particularly suitable approach for patients with equivocal results (abnormal squamous cells of undetermined significance) in their cervicovaginal cytology tests, especially if associated with a genotyping assay. The risk of cancer development after HPV infection is of course type-specific and there are at least 45 HPV types capable of infecting the human genital tract (Munoz et al., 2003). There is yet no clear information on the contribution of intrinsic protumorigenic properties in cancer associated HPVs versus prevalence within the “at risk” population. Through HPV detection and typing it is therefore possible to discriminate between patients harbouring intermediate and highrisk types, who are advised to undergo more extensive/invasive procedures to rule out intraepithelial carcinoma, and patients negative for HPV or carrying low-grade HPV types, who can be spared such procedures and checked yearly by conventional cervical cytology instead. Acknowledgements This work was supported by Ministero dell’Universit`a e della Ricerca Scientifica e Tecnologica (MIUR) (40%). References Arbyn, M., Buntinx, F., van Ranst, M., Paraskevaidis, E., Martin-Hirsch, P., Dillner, J., 2004. Virologic versus cytologic triage of women with equivocal Pap
smears: a meta-analysis of the accuracy to detect high grade intraepithelial neoplasia. J. Natl. Cancer Inst. 96, 280–293. Carozzi, F., Bisanti, S., Sani, C., Zappa, M., Cecchini, S., Ciatto, S., Confortini, M., 2007. Agreement between the AMPLICOR human papillomavirus test and the Hybrid Capture 2 assay in detection of high-risk human papillomavirus and diagnosis of biopsy-confirmed high grade cervical disease. J. Clin. Microbiol. 45, 364–369. Carozzi, F.M., Confortini, M., Cecchini, S., Bisanzi, S., Cariaggi, M.P., Pontenani, G., Raspollini, M.R., Sani, C., Zappa, M., Ciatto, S., 2005. Triage with human papillomavirus testing of women with cytologic abnormalities prompting referral for colposcopic assessment. Cancer 105, 2–7. Castle, P.E., Lorincz, A.T., Mielzynska-Lohnas, I., Scott, D.R., Glass, A.G., Sherman, M.E., Schussler, J.E., Schiffman, M., 2002. Results of human papillomavirus DNA testing with the hybrid capture 2 assay are reproducible. J. Clin. Microbiol. 40, 1088–1090. Clavel, C., Masure, M., Bory, J.P., Putaud, I., Mangeonjean, C., Lorenzato, M., Nazeyrollas, P., Gabriel, R., Quereux, C., Birembaut, P., 2001. Human papillomavirus testing in primary screening for the detection of high grade cervical lesions a study of 7932 women. Br. J. Cancer 84, 1616–1623. Cuzick, J., Beverley, E., Ho, L., Terry, G., Sapper, H., Mielzynska, I., Lorincz, A., Chan, W.K., Krausz, T., Soutter, P., 1999. HPV testing in primary screening of older women. Br. J. Cancer 81, 554–558. De Francesco, M.A., Gargiulo, F., Schreiber, C., Ciravolo, G., Salinaro, F., Manca, N., 2005. Detection and genotyping of human papillomavirus in cervical samples of Italian patients. J. Med. Virol. 75, 588–592. Fleiss, J.L., 1981. Statistical Methods for Rates and Proportion, second ed. John Wiley & Sons Inc., New York. Hildesheim, A., Schiffman, M.H., Gravitt, P.E., Glass, A.G., Greer, C.E., Zhang, T., Scott, D.R., Rush, B.B., Lawler, P., Sherman, M.E., 1994. Persistence of type-specific human papillomavirus infection among cytologically normal women. J. Infect. Dis. 169, 235–240. Kjaer, S.K., van den Brule, A.J., Paull, G., Svare, E.I., Sherman, M.E., Thomsen, B.L., Suntum, M., Bock, J.E., Poll, P.A., Meijer, C.J., 2002. Type specific persistence of hig risk human papillomavirus (HPV) as indicator of high grade cervical squamous intraepithelial lesions in young women: population based prospective follow up study. BMJ 325, 572. Kulmala, S.M., Syrjanen, S., Shabalova, I., Petrovichev, N., Kozachenko, V., Podistov, J., Ivanchenko, O., Zakharenko, S., Nerovjna, R., Kljukina, L., Branovskaja, M., Grunberga, V., Juschenko, A., Tosi, P., Santopietro, R., Syrjanen, K., 2004. Human papillomavirus testing with the hybrid capture 2 assay and PCR as screening tools. J. Clin. Microbiol. 42, 2470– 2475. Manos, M.M., Kinney, W.K., Hurley, L.H., Sherman, M.E., Shieh-Ngai, J., Kurman, R.J., Ransley, J.E., Fetterman, B.J., Hartinger, J.S., McIntosh, K.H., Pawlick, G.F., Hiatt, R.A., 1999. Identifying women with cervical neoplasia using human papillomavirus DNA testing for equivocal Papanicolau results. JAMA 281, 1605. Melchers, W.J., Claas, H.C., Quint, W.G., 1991. Use of the polymerase chain reaction to study the relationship between human papillomavirus infections and cervical cancer. Eur. J. Clin. Microbiol. Infect. Dis. 10, 714–727. Monsonego, J., Bohot, J.M., Pollini, G., Krawec, C., Vincent, C., Merignargues, I., Haroun, F., Sednaoui, P., Monfort, L., Dachez, R., Syrjanen, K., 2005. Performance of the Roche AMPLICOR human papillomavirus (HPV) test in prediction of cervical intraepithelial neoplasia (CIN) in women with abnormal PAP smear. Gynecol. Oncol. 99, 160–168. Munoz, N., Bosch, F.X., de Sanjose, S., Herrero, R., Castellsague, X., shah, K.V., Snijders, P.J., Meijer, C.J., 2003. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N. Engl. J. Med. 348, 518–527. Poljak, M., Fujs, K., Seme, K., Kocjan, B.J., Vrtacnik-Bokal, E., 2005. Retrospective and prospective evaluation of the Amplicor HPV test for detection of 13 high risk human papillomavirus on 862 clinical samples. Acta Dermatovenerol. Alp. Panonica Adriat. 14, 147–152. Poljak, M., Marin, I.J., Seme, K., Vince, A., 2002. Hybrid Capture II HPV test detects at least 15 human papillomavirus genotypes not included in its current high-risk probe cocktail. J. Clin. Virol. 25 (Suppl. 3), S89–S97. Roda Husman, A.M., Walboomers, J.M., van den Brule, A.J., Meijer, C.J., Snijders, P.J., 1995. The use of general primers GP5 and GP6 elongated at
M.A. De Francesco et al. / Journal of Virological Methods 147 (2008) 10–17 their 3 ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J. Gen. Virol. 76, 1057–1062. Sandri, M.T., Lentati, P., Benini, E., Dell’Orto, P., Zorzino, L., Carozzi, F.M., Maisonneuve, P., Passerini, R., Salvatici, M., Casadio, C., Boveri, S., Sideri, M., 2006. Comparison of the Digene HC2 assay and the Roche AMPLICOR human papillomavirus (HPV) test for detection of high-risk HPV genotypes in cervical samples. J. Clin. Microbiol. 44, 2141–2146. Saslow, D., Runowicz, C.D., Solomon, D., Moscicki, A.B., Smith, R.A., Eyre, H.J., Cohen, C., 2002. American Cancer Society guideline for the early detection of cervical neoplasia and cancer. Cancer J. Clin. 52, 342– 362. Schecht, N.F., Kulaga, S., RobitailleJ, Ferreira, S., Santos, M., Miyamura, R.A., Duarte-Franco, E., Rohan, T.E., Ferenczy, A., Villa, L.L., Franco, E.L., 2001. Persistent human papillomavirus infetion as a predictor of cervical intraepithelial neoplasia. JAMA 286, 3106–3114. Sherman, M.E., Schiffman, M., Cox, J.T., 2002. Effect of age and human papilloma viral load on colposcopy triage: data from the randomised atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesion triage study (ALTS). J. Natl. Cancer Inst. 94, 102– 107.
17
Solomon, D., Schiffman, M., 2004. Have we resolved how to triage equivocal cervical cytology? J. Natl. Cancer Inst. 96, 250–251. Syrj¨anen, K.J., 2000. Management of abnormal PAP smears (MAPS): implications of terminology used in cytopathology. J. Low Genit. Tract Dis. 4, 217–223. The atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesions triage study (ALTS) group, 2000. Human papillomavirus testing for triage of women with cytologic evidence of low grade squamous intraepithelial lesions: baseline data from a randomised trial. J. Natl. Cancer Inst. 92, 397–402. van Ham, M.A.P.C., Bakkers, J.M.J.E., Harbers, G.K., Quint, W.G.V., Massuger, L.F.A.G., Melchers, W.J.G., 2005. Comparison of two commercial assay for detection of human papillomavirus (HPV) in cervical scrape specimens: validation of the Roche AMPLICOR HPV test as a means to screen for HPV genotypes associated with a higher risk of cervical disorders. J. Clin. Microbiol. 43, 2662–2667. Walboomers, J.M., Jacobs, M.V., Manos, M.M., Bosch, F.X., Kummer, J.A., Shah, K.V., Snijders, P.J., Peto, J., Meijer, C.J., Munoz, N., 1999. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J. Pathol. 189, 12–19.