- Email: [email protected]
NucliSENS® EasyQ® HPV v1 test – Testing for oncogenic activity of human papillomaviruses
Journal of Clinical Virology 45, S1 (2009) S29 S37
Contents lists available at ScienceDirect
Journal of Clinical Virology j o u r n a l h o m e p a g e : www.elsevier.com/locate/jcv
NucliSENS® EasyQ® HPV v1 test human papillomaviruses
Testing for oncogenic activity of
Damien Jeanteta , Fritz Schwarzmanna , Jessica Trompb , Willem J.G. Melchersc , Anneke A.M. van der Wurffd , Tom Oosterlakenb , Marcel Jacobsb , Alain Troescha, * a
bioM´ erieux, Grenoble, France bioM´ erieux bv, Boxtel, The Netherlands Department of Medical Microbiology, Radboud University, Nijmegen, The Netherlands d St. Elisabeth Hospital, Tilburg, The Netherlands b c
ARTICLE
INFO
Keywords: Human papilloma virus E6/E7 mRNA NucliSENS® EasyQ® Nucleic acid sequence-based amplification NASBA Cervical carcinoma HPV diagnostics
ABSTRACT Background: Analytical sensitivity of DNA-based assays to detect infection with human papillomaviruses is very high, but clinical specificity for cervical cancer strongly depends on the age of the patient and case classification. To solve the dilemma between sensitivity and specificity, a new generation of assays focuses on the pathogenic factors that underlie the development of HPV-associated tumors: the expression of the viral oncogenes E6 and E7. Demonstration of persistent expression of these mRNAs or expression in the context of relevant clinical symptoms has a strong positive predictive value for the development of HPV-associated carcinomas and strongly warrants further diagnostic action. Objectives: The NucliSENS® EasyQ® HPV v1 test was designed to test cervical scrapes for the expression of the oncogenic E6/E7 mRNA from the five most common carcinogenic HPV types (16, 18, 31, 33 and 45). The test can be used for confirmation and risk stratification of individuals with proven infection with high risk papillomaviruses. Study design: In order to establish performance of the assay, sensitivity, specificity, repeatability, and reproducibility were determined with artificial and clinical specimens. Further, a total of 420 cervical scrapes were tested and the results directly compared to the CE-market device PreTect HPV-Proofer assay (NorChip, Klokkarstua, Norway). For arbitration of discordant clinical results, the specimens were rated according to Pap-classification and the presence of HPV DNA was determined. Results: The limit of detection for the five HPV types 16, 18, 31, 33, and 45 ranged from 2.3×102 to 3.0×104 copies/mL on a background of 5×103 HPV-negative HS67 cells. No crossreactivity for other viral, bacterial, or fungal agents known to be potentially present in cervical fluids was detected. Repeatability and reproducibility were shown by testing panels of HPV-spiked artificial and clinical samples. A comparative analysis of 420 cervical scrapes demonstrated an overall concordance of >90% between the NucliSENS EasyQ HPV test and the technologically related PreTect HPV-Proofer assay. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Human papillomaviruses (HPV) have been identified as the causative agents of a number of benign and malignant tumors. These viruses are strictly tissue-specific and are associated with malignant tumors of the skin, the head and neck, and anogenital region.1,2 Over 100 subtypes are described. * Corresponding author. bioM´ erieux, Centre Christophe M´ erieux, 5, Rue des Berges, F-38024 Grenoble Cedex 01, France. E-mail address: [email protected] (A. Troesch). 1590-8658/ $
see front matter © 2009 Elsevier B.V. All rights reserved.
In general, first diagnosis of infection with HPV is done indirectly by visual inspection of the mucosa and cytological examination for abnormal cells. For advanced and persistent infections with HPV, the demonstration of over-expressed cellular p16 protein in the infected tissues may support diagnosis.3,4 However, reliable confirmation of diagnosis is possible only by direct detection of the infectious agent’s nucleic acids, since expression of viral proteins varies between active, latent, and persistent infections with the chance of integrated viral genomes and reduction in gene expression.
S30
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
Immunology-based diagnosis is not available, since humoral immune responses against HPV are weak and unreliable. Although cellular immunity plays a pivotal role in the control of infected cells, no assay exists to date for the detection and monitoring of HPV-specific T-cells.5,6 Furthermore, differentiation between the HPV types is possible only by molecular genetics. Introduction of HPVdiagnostic tests in addition to cytology (Pap smear) significantly increased the sensitivity of screening for cervical cancer.7 Several assays exist, most of them based on nucleic acid amplification of viral DNA. Analytical sensitivity of these assays is very high, but clinical specificity for high-grade lesions depends strongly on the age of the patient and the cytological case classification. In this respect, DNA-based assays with a balanced sensitivity and specificity did prove suitable for initial screening and for the detection and prediction of high-grade lesions (CIN 3) in women 30 years and older.8 Due to the high prevalence of infection and the transient nature of infection with HPV, the specificity and the positive predictive value of DNA-based assays for the detection of highgrade lesions is not satisfactory in women younger than 30 years. Furthermore and most important, in the context of triage of patients with abnormal cytological results and/or clinical symptoms of infection, these assays show only low specificity and positive predictive value for high-grade lesions (CIN 3).9,10 The diagnostic approach of bioM´ erieux is targeted at the identification of those individuals who are at higher risk of developing HPV-associated carcinomas (Fig. 1). Accordingly, the NucliSENS® EasyQ® HPV v1 assay is intended for confirmation or triage of individuals with the diagnosis of active infection with HPV types of the high-risk group. Risk stratification, aimed to identify those women with a higher potential to develop a carcinoma, is based on the underlying molecular mechanisms of HPV-related tumor development and tumor progression. With respect to the development of cervical cancer, a number of strains/genotypes of HPV have been classified into high-risk and low-risk according to their oncogenic potential.11 Briefly, most HPV infections, belonging to both the high-risk and low-risk groups, are transient and the viruses are eliminated from the host within six to eighteen months after infection. With infection by viruses of the high-risk group, the risk to develop a persistent infection, which is a prerequisite for the development of a malignant tumor, increases significantly. In the early phases of infection,
generally only episomal viral DNA is detected. However, the chance of integration of the virus into the host genome increases with the duration of the infection. In cervical carcinomas mostly integrated viral DNA is detected.12,13 In rare cases, integration of HPV genomes results in the inactivation of the viral regulatory gene E2, which in turn results in deregulation of the viral tumor genes E6 and E7. Expression of these genes is accepted to be a prerequisite for tumor progression since activity of these genes results in increased cell proliferation, prevention of apoptosis, and an increase in mutagenesis. Thus, monitoring of the activity of E6 and E7 by detecting the corresponding mRNA seems to be a reasonable strategy to stratify individuals with proven infection with high-risk HPV.14 16 The NucliSENS EasyQ HPV v1 is a real-time nucleic acid amplification and multiplex detection assay for the qualitative determination of E6/E7 mRNAs of the five most commonly identified carcinogenic HPV types in cervical cancer worldwide (HPV 16, 18, 31, 33 and 45). These five HPV genotypes are detected in approximately 82% of all cervical carcinomas.17,18 The NucliSENS EasyQ HPV v1 assay utilizes the nucleic acid sequence-based amplification (NASBA) technology that has been used in a wide variety of diagnostic assays, detecting both viral and bacterial nucleic acids.19 23 NucliSENS EasyQ HPV version 1 was launched in 2007 and was based on the original PreTect Proofer assay developed by NorChip (Klokkarstua, Norway), except for the NucliSENS hardware platform and the software for NASBA measurements and data analysis (NucliSENS Director software). The latest NucliSENS EasyQ HPV assay version 1 has recently been improved in the following ways: taking into account the RNA nature of the HPV-specific targets, the five positive controls for the different subtypes and the U1A internal control were changed from singlestranded DNA to RNA; in order to facilitate handling and to increase reproducibility of the assay, the positive controls are no longer provided in liquid phase but in freeze-dried accuspheres; in order to improve and harmonize the sensitivity for the five HPV types covered by the assay, the concentrations of the primers and beacons for HPV 16 and the U1A internal control were re-adjusted; the extraction protocol on NucliSENS EasyMAG® system was optimized; the HPV assay protocols were updated and the NucliSENtral™ HPV software V1.1 introduced to further improve the convenience of operator use for results reporting. 2. Materials and methods 2.1. Laboratory equipment
Fig. 1 The stepwise development of cervical cancer. The present RNA-based NucliSENS® EasyQ® HPV v1 test is intended to stratify individuals with confirmed active infection with high-risk HPV types. Testing is based on the detection of the viral oncogenic E6 and E7 mRNAs of HPV types 16, 18, 31, 33, and 45.
The NucliSENS EasyQ HPV v1 assay is designed for the bioM´ erieux NucliSENS platform. Nucleic acid extraction protocols have been validated on two extraction systems, either the semi-automated NucliSENS miniMAG® or the automated NucliSENS easyMAG® (bioM´ erieux, Marcy l’Etoile, France). Amplification, real-time detection and fully automated interpretation of the test results are done on a NucliSens EasyQ analyser used for all NucliSENS assays. The analytical part of the assay is performed in strips of microliter reaction tubes. One run covers up to 30 patient samples at a time (Fig. 2).
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
Fig. 2 EasyQ® HPV v1 Workflow. Nucleic acids are isolated using the NucliSENS® miniMAG® or the NucliSENS easyMAG® . The NucliSENS EasyQ HPV v1 assay utilizes the Nucleic Acid Sequence Based Amplification (NASBA). The time required to complete the assay (sample preparation, extraction of nucleic acids, amplification and detection, and evaluation of results) is 4 hours for 24 samples.
2.2. Assay description Nature of specimens and isolation of nucleic acids The NucliSENS EasyQ HPV v1 assay is intended to be used with cervical scrapings. Due to the nature of the mRNA targets and their inherent sensitivity to nucleases, non-stabilized specimens like dry swabs are not suitable for testing. Thus, for preservation of the E6- and E7-specific mRNA, the cervical samples must be stabilized immediately after collection in transport media such as PreservCyt (Hologic. Inc., Marlborough, MA). The material is stable in this buffer at room temperature for up to six weeks. The use of other stabilizing media is currently under investigation. Due to the nature of the specimen, which always shows irregular numbers of cells, the sample is mixed thoroughly before use. A 5 mL volume of sample is transferred into a fresh 10 mL tube for extraction and then centrifuged for 12 minutes at 1125×g. The supernatant is removed leaving approximately 1 mL PreservCyt solution. After the re-suspension of the specimen’s cells in the remaining 1 mL PreservCyt solution, the sample is transferred to another tube containing 2 mL NucliSENS Lysis Buffer (bioM´ erieux, Marcy l’Etoile, France). Any cellular material or viral particles present in the sample are disrupted, RNases and DNases are inactivated, and total nucleic acids are released. After cell lysis, magnetized silica dioxide particles are added to the lysate of cervical specimen to initiate the RNA isolation process.24 The nucleic acids present in the NucliSENS Lysis Buffer bind to the magnetic silica under the high salt conditions. After binding of the nucleic acids, the magnetic silica is washed several times in order to remove contaminating and potentially inhibitory cellular components. The nucleic acids are eluted from the solid phase thereby making them available for use in the amplification and detection procedures. The mRNA is eluted in 55 mL of elution buffer and stored at 20ºC if not further processed immediately after extraction. Multiplex real-time NASBA The NucliSENS EasyQ HPV v1 assay utilizes a combined process of nucleic acid amplification and simultaneous detection of the amplicon using molecular beacons.25,26 All primers and probes in the assay are designed to detect the full-length mRNA transcripts of U1A and the E6/E7 region from HPV types 16, 18, 31, 33 and 45. A full-length transcript is defined as any un-spliced transcript covering at least the E6/E7 region and may also be fused to human sequences. The molecular beacons enable a real-time and multiplex assay suitable for high-throughput applications with increased specificity, without compromising
S31
sensitivity. In contrast to the widely used polymerase chain reaction (PCR), which is based on the use of thermo-stable DNA polymerase and temperature cycling, NASBA is an enzymatic amplification process that amplifies in isothermal conditions (41ºC). Three enzymes are involved in the amplification of the nucleic acids (either DNA or RNA): avian myeloblastosis virus reverse transcriptase (AMV-RT), Escherichia coli RNase H, and T7 RNA polymerase. The NASBA reaction continues in a selfsustained manner, and is capable of achieving amplification of approximately 1013 1014 -fold within 90 minutes. Briefly, the NASBA amplification is based on a repeated process of primer annealing, formation of double-stranded DNA containing a T7 promoter site, and T7 RNA polymerasemediated transcription of multiple anti-sense copies of RNA target sequences (amplicons). The principal amplification product of NASBA is single-stranded RNA that subsequently can easily be identified and characterized. Figure 3 provides an overview of the main steps involved in the NASBA reaction.27 The NucliSENS EasyQ HPV v1 test uses multiple primer sets for amplification of the five HPV genotypes HPV 6, HPV 18, HPV 31, HPV 33, and HPV 45. In order to preclude false-negative test results due to insufficient cellular material, degradation of the mRNA, or inhibition of amplification, the quality of the extracted nucleic acid is monitored by testing for an internal control mRNA, human U1 small nuclear ribonucleoprotein specific protein A (U1A). The U1A protein is part of the housekeeping U1snRNP, which binds to the 5 exon of hnRNAs and initiates spliceosomal assembly. The U1A-specific mRNA, which is present in low copy numbers in the cell, has been described previously as a suitable internal control in molecular-genetic virological assays.28 30 To this aim, a primer pair targeting U1A mRNA is included in the assay. Besides the U1A control, five positive controls consisting of artificial E6/E7-target sequences specific for the HPV types HPV 16, HPV 18, HPV 31, HPV 33 and HPV 45 are included to monitor functionality of primers, probes and NASBA reagents. The internal U1A control and the five HPV-specific controls are provided in one freeze-dried accusphere, which will be diluted in the corresponding buffer and used directly in amplification. The detection of HPV-specific amplification products is accomplished in real-time by the use of target-specific molecular beacons, which fluoresce only upon hybridization with their target. To detect amplification of the specific mRNAs, six different molecular beacons are used in the NucliSENS EasyQ HPV v1 assay, one specific beacon for each of the HPV subtypespecific amplicons and one for the U1A amplicons. The use of two fluorescent dyes (6-FAM and 6-ROX) allows the detection of the HPV target RNAs in three duplex amplification reactions: U1A and HPV 16, HPV 18 and HPV 31, HPV 33 and HPV 45. Interpretation of results The HPV Client analysis software (NucliSENtral™ HPV software V1.1) has been designed to automate parts of the run workflow and to allow the automated determination of the run and sample validity and interpretation of sample results. As depicted in Fig. 3, the NASBA process is characterized by an early exponential growth of a double-stranded intermediate DNA amplicon, followed by a linear increase in RNA levels which translates into a linear increase of fluorescence. The NucliSENS EasyQ Director Software determines the fluorescence values for each individual amplification taking place and validates the measured values against a number of validation criteria (statistical), the quality of the curve fit, and the timing of the
S32
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
Fig. 3. Principle of real-time NASBA. A template-specific DNA intermediate is initially generated through a process that involves the binding of a primer (primer 1) to the RNA template. This primer, which contains a T7 RNA polymerase promoter site, is then extended by AMV-RT to form an RNA DNA duplex. Degradation of the RNA strand of the duplex by RNase H permits the binding of a second primer (primer 2) to the remaining DNA strand. The second primer encompasses a short sequence which is identical to a segment of the target RNA and is located upstream of the region where the first oligonucleotide can anneal. The second primer is then extended by AMV-RT to form the double-stranded DNA intermediate, which contains the T7 RNA polymerase promoter needed for transcription. Once transcription is initiated, the resulting RNA transcripts, which are anti-sense to the original RNA present in the original specimen, can serve as a template to start a new amplification process.
onset of the fluorescence increase, which (to a limited extent) corresponds to the Ct value in real-time PCR. Furthermore, the final qualitative test result will account for the validity of the negative and the three positive controls, the internal control, and the results of the five HPV subtype-specific reactions. If a test result has been qualified as valid, a qualitative result (positive/negative) is reported for each HPV target.
Table 1 List of microbial agents tested for analytical specificity Mycoplasma hominis Escherichia coli Chlamydia trachomatis Gardnerella vaginalis Enterococcus faecalis Ureaplasma urealyticum Staphylococcus aureus
Mycoplasma genitalium Neisseria gonorrhoeae Candida albicans Human immunodeficiency virus Hepatitis C virus Hepatitis B virus
2.3. Determination of assay analytical sensitivity The analytical sensitivity of the assay was determined by analyzing a dilution series of in vitro generated E6/E7derived RNA specific for each HPV type (length around 1000 nucleotides, depending on the HPV type) on a background of 5×103 HPV-negative human cells (HS67 cell line). The input of transcripts ranged from zero to 107 copies per extraction. The concentration of in vitro generated RNA was determined by absorption spectroscopy. Subsequently, nucleic acids were extracted using a NucliSENS easyMAG and NucliSENS miniMAG® device and tested in replicates for the presence of E6/E7 RNA. Analytical sensitivity defined as the 50% and 95% hit rate values were determined by statistical Probit analyses.
done using the BLAST algorithm (Genebank, National Center for Biotechnology Information, Bethesda, MD). No homology alignments were found with both primers and the molecular beacons within a range of approximately 1000 nucleotides, being the maximal size of a NASBA amplification product under normal reaction conditions. Additionally, a selection of microbial agents either known to cause genital tract infections or to be potentially present in cervical secretions (Table 1), were tested. Briefly, the agents were cultured in vitro and 20 mL of the cultures were added to 2 mL lysis buffer tubes for extraction of the nucleic acids in a total volume of 55 mL. Five mL were used as input for the HPV-specific NASBA assay. Analytical specificity was defined as the percentage of false positive results for HPV.
2.4. Determination of assay analytical specificity
2.5. Determination of assay repeatability and reproducibility
To determine the analytical specificity of the assay, in first instance, sequence homology alignment analyses of the primers and molecular beacons with different organisms were
Assay repeatability and reproducibility were tested both in an analytical and a clinical background (patient specimen). For
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
2.6. Performance evaluation of the NucliSENS EasyQ HPV v1 using clinical samples An evaluation of the assay performance was done in direct comparison of the NucliSENS EasyQ HPV v1 and the established CE-marked device PreTect HPV-Proofer by NorChip AS (Klokkarstua, Norway). A total of 420 cervical scrapes in PreservCyt buffer were collected at two sites from women participating in a routine cervical cancer screening program. Samples were collected and tested in accordance with the approval of the Medical Ethical Committee of St. Elizabeth Hospital, Tilburg and in accordance with the policies of Radboud University, Nijmegen, for the use of discard de-identified samples. For all samples, routine diagnostics based on cytology was performed. A total of 248 samples were classified as Pap 1 (normal cells), 88 samples were found to be Pap 2 (mild inflammation), 80 samples were Pap 3 (mild, moderate or severe dysplasia), three samples were Pap 4 (carcinoma in situ), and one sample was classified as Pap 5 (invasive carcinoma).
progress to cervical carcinoma. Thus, the analytical sensitivity of DNA- and RNA-assays for detecting HPV viruses cannot be compared directly. The concept of detecting E6/E7 mRNA of high-risk HPV genotypes is to reduce non-relevant positive results and thus to increase the positive predictive value for cervical carcinoma. 3.1. Assay analytical sensitivity In order to assess the analytical sensitivity for viral E6/E7 mRNA, the limit of detection was determined by analyzing dilution series of in vitro generated RNA for each of the five HPV types. To mimic clinical samples, 5×103 human cells from an HPV-negative human cell line (HS67) were lysed and the diluted RNA was added before extraction. The U1A mRNA, which is used as intrinsic control, was contained in the cellular RNA. No additional positive and negative run controls were included in the amplification run, since positive and negative controls were already included in the RNA dilution series, considering the sample containing no HPV RNA as a negative control and the sample containing the highest in vitro transcribed RNA input as the positive control. Each input of in vitro transcribed RNA was tested in twenty replicates and the percentage of replicates tested positive (hit rate) was determined. The calculated analytical sensitivity (defined as the estimated concentration with 95% hit rate) was shown to range from 2.3×102 to 3.0×104 copies/mL, showing the highest sensitivity for HPV 45 and the lowest for HPV 31. The results are summarized in Fig. 4 and Table 2. 120
Probit HPV 18 100 % positive results
analytical studies, E6/E7-specific in vitro generated RNA was spiked in PreservCyt buffer at input levels of 103 copies/ml and 105 copies/ml corresponding to estimated hit rates of 1% and 99% determined previously on the basis of Probit analysis. Analytical repeatability was determined by the concurrent testing of twenty replicates of each input in one laboratory, by the same operator and using the same equipment. Analytical reproducibility was determined by testing the same panel in two laboratories by two operators using different equipment. For statistical analysis, data were compared using Fisher’s exact test. For studies with clinical specimens, ten HPV-negative cervical scrapes, five HPV 16-positive cervical scrapes, and five HPV-negative cervical scrapes that were spiked with HPV 16 at 105 copies/ml corresponding to 99% hit rate, were used. Nucleic acids were extracted using easyMAG. Repeatability was determined by concurrently testing these specimens in triplicate. Reproducibility was assessed by a second operator in a different laboratory using different equipment. Statistical concordance was assessed by pairwise comparison of the data of one operator using the same equipment in the same laboratory (repeatability), and comparison of the data of two different operators using different equipment in different laboratories (reproducibility) by Cohen’s kappa (ú) statistic. A ú estimate of less than 0.4 represents poor agreement, a ú estimate between 0.4 and 0.75 is fair to good agreement, and a ú estimate of more than 0.75 is excellent agreement.
S33
80 60 40
Probit 95% lower limit
20
95% upper limit measured
0 1
2
3
4
5
log input
120
Probit HPV 33
3. Results DNA-based assays for HPV show a very high sensitivity and specificity for the presence of human papillomaviruses. Thus, these assays significantly enhance the sensitivity of cervical cancer screening. However, due to the high prevalence of HPV and the transient nature of most infections, specificity and the positive predictive value of these tests for high-grade lesions and progression to cervical carcinoma strongly depend on the age of the women and the clinical presentation at the time of diagnosis. In contrast, the negative predictive value is high and the cancer risk is very low when the test is negative. In this respect, the present assay, based on the detection of the viral oncogenic E6 and E7 mRNA, is intended to detect specifically those HPV infections which bear a potentially higher risk to
% positive results
100 80 60 40 Probit 95% lower limit
20
95% upper limit measured
0 1
1,5
2
2.5 3 log input
3.5
4
Fig. 4 Probit analysis for HPV 18 and HPV 33 to determine the analytical sensitivity.
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
S34
Table 2 Analytical sensitivity as defined as the 50% and 95% hit rate values of in vitro RNA dilution series calculated by using Probit analysis Hit rate
In vitro HPV RNA (copies/mL) HPV 16
50% 95 %
HPV 18 1
4.0×10 2.6×10
2
HPV 31 2
8.8×10
3
6.4×10
HPV 33 3
4.3×10
4
3.0×10
HPV 45 2
3.5×101
3
2.3×102
5.6×10 3.0×10
3.2. Assay analytical specificity The specificity of the NucliSENS EasyQ HPV v1 assay is based upon careful selection of primers and probes taking into account genetically related viruses and other organisms that are known to occur in cervical scrapes. Cross-reactivity of various micro-organisms was assessed by testing directly nucleic acids from these organisms in the NucliSENS EasyQ HPV v1 assay. Briefly, pure bacterial or viral cultures were diluted one hundred fold in lysis buffer for extraction of nucleic acids and 5 mL of the eluate was used for the NASBA. No cross-reactivity was found with the organisms listed in Table 1. The absence of cross-reactivity for other HR HPV types or other LR HPV types has been demonstrated by theoretical sequence homology alignment analyses and by some preliminary experiments on a limited set of clinical samples found positive for HPV 6, 52 or 58 using InnoLipa (Gent, Belgium) (data not shown). More extensive experimental studies will be performed to further guarantee the specificity of the assay. 3.3. Assay repeatability and reproducibility Repeatability and reproducibility of the NucliSENS EasyQ HPV v1 assay was assessed by testing panels of artificial and clinical specimens. Briefly, repeatability was determined by successive testing of sample panels at one site by one operator. Reproducibility implied testing of the panels at a second site by at least one additional operator. Two types of panels were employed for these studies, either based on in vitro generated HPV 16-derived E6/E7-specific RNA spiked in PreservCyt buffer (artificial specimens) or based on clinical specimens. Since the NucliSENS EasyQ HPV v1 is a qualitative assay, statistical analysis was performed on the number of positive or negative results, and not on the underlying quantitative fluorescent values. Analysing the artificial samples with spiked E6/E7-specific RNA, with the high positive samples (105 copies/mL; 99% hit rate) no statistical analysis was performed, since all the replicates were tested 100% positive. With the low input samples (103 copies/mL; 1% hit rate) we observed hit rates of 10% (2/20), 0% (0/20), and 10% (2/20) for three consecutive tests. Statistical analysis using Fisher’s exact test yielded a p-value of 0.53 for repeatability and a p-value of 1.0 for reproducibility, indicating that no statistically significant differences between the individual tests were observed. During testing of the clinical specimens, three invalid results were obtained with HPV-negative samples due to negative results for the U1A internal control. Thus, the total number of clinical samples was reduced from 20 to 17. In terms of repeatability, 100% (10/10) of each of the clinical samples tested positive for all three replicates. In total, 30/30 positive samples were tested positive. Of the 7 (valid) negative samples, 100% (21/21) of the replicates were tested negative.
With respect to repeatability, one of the ten positive samples tested negative in one of two duplicates. For the negative samples, again, three invalid results were obtained with HPV-negative samples due to negative results for the U1A internal control. Thus, the total number of samples in the study was reduced from 20 to 17. Of the 7 (valid) negative samples, 100% of the duplicates (14/14) tested negative. This resulted in 0.88 concordance and demonstrated sufficient clinical repeatability and reproducibility. 3.4. Comparison of the NucliSENS EasyQ HPV v1 to the PreTect HPV-Proofer In order to evaluate the clinical performance of the NucliSENS EasyQ HPV v1 assay, a study with a total of 420 cervical scrapes was conducted. Specimens were collected at two study sites from women participating in a routine cervical cancer screening program or under their gynecologists’ care for abnormal smears. In addition, all samples were analyzed in parallel with the established CE-marked device PreTect HPV-Proofer by NorChip AS (Klokkarstua, Norway). The results obtained in this study are summarized in Table 3. For the calculation of the percentage of concordance and for the calculation of Cohen’s kappa, invalid results (due to invalid controls) and undetermined results of the PreTect HPV-Proofer assay (due to values of fluorescence in the grey zone of the assay, i.e. between 1.4 and 1.7 Relative Fluorescence Units) were rejected.31 Ninety percent concordance and a Cohen’s kappa of 0.706 demonstrated an equivalent performance between the NucliSENS EasyQ HPV v1 and the PreTect HPVProofer assay. Interestingly, the concordance for the individual subtypes was significantly higher, showing 96%, 97%, 95% and 98% for HPV 16, HPV 18, HPV 31 and HPV 45, respectively. This difference in overall and subtype-specific concordance was due to mixed infections with several HPV types. (The results for HPV 33 were retrospectively withdrawn from this part of the study, due to the suspicion of contamination.) Table 4 summarizes the discordant results obtained in this part of the study. Samples which yielded invalid results with one of the two tests were excluded from this analysis. For arbitration of the results, cytological classification is shown. Herein, sixty-four percent (16/25) of the discordant results were obtained with samples classified as Pap 1. Two and seven discordant results were derived from Pap 2 and Pap 3 samples (Table 5). Interestingly, for two of the three Pap 3 specimens, Table 3 Overall concordance of E6/E7 mRNA test results of 420 cervical scrapings in pos/neg classification between the NuclSENS and the PreTect HPV-Proofer test PreTect HPV-Proofer Invalid
a
NucliSENS Invalid
Negative
Positive
Total
61
57
13
131
Negative
22
171
20
213
Positive
6
4
40
50
Indeterminate a
0
12
14
26
89
244
87
420
Total a b
invalid = tests with at least one invalid control. indeterminate = tests with fluorescence values in the grey zone (between 1.4 and 1.7 relative fluorescence units) of the PreTect HPV-Proofer assay.
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
S35
Table 4 Discordant results between NucliSENS® EasyQ® HPV v1 and the established CE-marked PreTect HPV-Proofer ID
Pap
LiPA
PreTect HPV-Proofer
EasyQ Results U1A
16
18
31
45
33
U1A
16
18
31
45
33
Site 1 2
1
31
+
3
1
16
+
19
1
22
1
16
+
69
1
33
+
71
1
31
+
72
1
33
+
76
1
83
3a1
85
1
87
3a2
96
1
122
1
136
1
146
1
+
+/
+
+
+ + +
+ + +
+ 18
+
+
+
+
+
+ +
I
+
+
I
+
+
+
+
+
+ 45
I +
16 31
+/
+
+
+
+
+/
Site 2 102b
2
+
280
1
+
305b
3a
+
396
1
404
1
+
547b
3a
+
596b
2
+
605b
3a
+
+
882b
3a
+
+
+
972b
3a
+
+
+
+
+
+
+ +
+
+
+ + +
+
+
+
+
+
+
+
+ +/
The upper and lower parts of the table show the results obtained with samples collected at two different study sites. The data on cytological classification according Pap are shown in the column Pap. For samples selected at Site 1, data on HPV DNA were available as well; these are listed in the LiPA column. See Table 5 for the correlation of discordant results with Pap classification.
Table 5 Summary of correlation of discordant results with Pap classification Pap classification
1
2
3
4
5
Number of discordant results
16
2
7
0
0
a result from DNA testing was available which confirmed the positive result of the NucliSENS EasyQ HPV assay. For 15 samples there was additional information on HPV DNA available (InnoLipa, Gent, Belgium). The detection of HPV DNA was in agreement with the NucliSENS EasyQ HPV assay for 5 samples containing HPV types 16, 18, 31 or 33, all of which were not detected by the PreTect HPV-Proofer assay. The NucliSENS EasyQ HPV assay detected HPV mRNA (type 16) in 2 samples, negative by both the PreTect HPV-Proofer assay and InnoLipa. Five samples contained HPV DNA (types 16, 31, 33, or 45) that were not detected by the NucliSENS EasyQ HPV assay. The PreTect HPV-Proofer assay detected HPV mRNA (types 16 and 33) in 2 samples negative by the NucliSENS EasyQ HPV assay and positive by InnoLipa and in 3 samples (types 16, 18 and 33) that were negative by both other assays. Five
samples contained HPV DNA (types 16, 18, 31, or 33) that was not detected by the PreTect HPV-Proofer assay. Three samples HPV DNA positive had a different HPV mRNA type detected by the PreTect HPV-Proofer assay (16 vs 33, 31 vs 33, and 45 vs 33, respectively). The InnoLipa results confirmed 5 of the 7 positive results obtained with the NucliSENS EasyQ HPV assay. 4. Conclusions and summary Introduction of molecular genetic testing for detecting HPV infections has significantly increased the sensitivity and efficacy of cervical cancer screening previously based solely on cytology and Pap testing. However, new testing for HPV E6/E7 mRNA expression has focused on detecting not just HPV infection but the detection of relevant viral (oncogenic or pre-cancerous) activity. Thus, these assays are intended to be used as a tool for the identification of clinically relevant HPV disease and/or for triage and risk stratification of patients with abnormal cytology and/or clinical symptoms. To this aim, the NucliSENS EasyQ HPV v1 test was designed to detect E6/E7 oncogenic activity from the five most
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
S36
carcinogenic HPV types found most frequently in cervical cancer worldwide. Demonstration of persistent expression of these mRNAs or expression in the context of relevant clinical symptoms has a strong positive predictive value for the development of HPV-associated carcinomas and strongly warrants further diagnostic action improving patient management. This has been shown impressively in a recent publication by Szarewski et al.9 presenting the results of a comparative study that determined the sensitivity, specificity and positive predictive value for the detection of CIN 2 and CIN 3 of seven commercially available clinical diagnostic tests and diagnostic approaches: detection of oncogenic HPV-derived RNA, demonstration of HPV DNA, or detection of cellular (HPVunrelated) markers for transformation. In this context cervical scrapes from 953 women with abnormal cervical smears were analysed and the results were compared to the gold standard of histopathology. The NucliSENS EasyQ HPV test was not included in this study in contrast to the technologically related PreTect HPV-Proofer assay. In this study, the best results in terms of specificity were obtained with the E6/E7 mRNA-based assay (PreTect HPV-Proofer) (73.1% and 70.4% for CIN 2+ and CIN 3+, respectively), closely followed by detection over-expression of p16INK4A (68.7% and 65.8% for CIN 2+ and CIN 3+, respectively). However, sensitivity of both the E6/E7 mRNA-based assay (73.6% and 82.2% for CIN 2+ and CIN 3+, respectively) and the p16INK4A -based assay (83.0% and 92.7% for CIN 2+ and CIN 3+, respectively) were significantly lower than that of some of the DNA-based assays (range 80.9 99.6%).9 Due to the high prevalence of infection and relatively low incidence of HPV-related tumours, a balanced sensitivity and specificity is of utmost importance for the diagnosis of high-grade lesions.32 This is true both for DNA-based assays and for RNA-based approaches. In this respect, another RNA amplification assay, the APTIMA® HPV Assay (GenProbe, Inc, San Diego, CA), targeting E6/E7 mRNA of 14 high-risk HPV types showed sensitivity similar to DNA assays (95.2%), better specificity (42.2%) than DNA assays (range 21.7% to 37.1%), but lower specificity than other RNA-based assays for high-grade disease.9 Owing to the lower sensitivity of the NucliSENS EasyQ HPV v1 test compared to DNA-based assays, at present, the NucliSENS EasyQ HPV v1 test is intended only for the confirmation and the risk stratification of individuals with proven infection with highrisk HPV. Ongoing improvement of the assay in the future, with respect to the number of HPV types covered, will comply with the broad change in the perception of viral mRNA detection for cervical cancer screening. In any case, comprehensive clinical studies will be essential to evaluate the performance and predictive values of any HPV assay with respect to the intended use as a tool for screening or for triage and risk stratification of patients with abnormal cytology and/or clinical symptoms. Acknowledgements: We are grateful to V´ eronique BaronWunderl´ e and Bernard Fleury (bioM´ erieux) for their critical review of the manuscript. Competing interests: DJ, FS, JT, TO, AT are employees of bioM´ erieux. AMvdW, WJGM: none declared. MJ: none declared. References 1. Munoz N, Castellsague X, de Gonzalez AB, Gissmann L. Chapter 1: HPV in the etiology of human cancer. Vaccine 2006;24(Suppl 3):S3-1 10.
2. zur Hausen H. Papillomaviruses in the causation of human cancers a brief historical account. Virology 2009;384:260 5. 3. Wentzensen N, Bergeron C, Cas F, Eschenbach D, Vinokurova S, von Knebel Doeberitz M. Evaluation of a nuclear score for p16INK4a-stained cervical squamous cells in liquid-based cytology samples. Cancer 2005;105:461 7. 4. Wentzensen N, Bergeron C, Cas F, Vinokurova S, von Knebel Doeberitz M. Triage of women with ASCUS and LSIL cytology: use of qualitative assessment of p16INK4a positive cells to identify patients with high-grade cervical intraepithelial neoplasia. Cancer 2007;111:58 66. 5. Carter JJ, Koutsky LA, Hughes JP, Lee SK, Kuypers J, Kiviat N, et al. Comparison of human papillomavirus types 16, 18, and 6 capsid antibody responses following incident infection. J Infect Dis 2000;181:1911 9. 6. Frazer IH. Interaction of human papillomaviruses with the host immune system: a well evolved relationship. Virology 2009;384:410 4. 7. Wright TC, Schiffman M, Solomon D, Cox JT, Garcia F, Goldie S, et al. Interim guidance for the use of human papillomavirus DNA testing as an adjunct to cervical cytology for screening. Obstet Gynecol 2004;103:304 9. 8. Meijer CJ, Berkhof J, Castle PE, Hesselink AT, Franco EL, Ronco G, et al. Guidelines for human papillomavirus DNA test requirements for primary cervical cancer screening in women 30 years and older. Int J Cancer 2009;124:516 20. 9. Szarewski A, Ambroisine L, Cadman L, Austin J, Ho L, Terry G, et al. Comparison of predictors for high-grade cervical intraepithelial neoplasia in women with abnormal smears. Cancer Epidemiol Biomarkers Prev 2008;17:3033 42. 10. van Ham MA, Bakkers JM, Harbers GK, Quint WG, Massuger LF, Melchers WJ. Comparison of two commercial assays 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 2005;43:2662 7. 11. International Agency for Research on Cancer. IARC monographs on the evaluation of carcinogenic risks to humans. Human papillomaviruses 2006;90. 12. 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. 13. Wettstein FO, Stevens JG. Variable-sized free episomes of Shope papilloma virus DNA are present in all non-virus-producing neoplasms and integrated episomes are detected in some. Proc Natl Acad Sci USA 1982;79:790 4. 14. Falcinelli C, Claas E, Kleter B, Quint WG. Detection of the human papilloma virus type 16 mRNA-transcripts in cytological abnormal scrapings. J Med Virol 1992;37:93 8. 15. Hsu EM, McNicol PJ, Guijon FB, Paraskevas M. Quantification of HPV-16 E6 E7 transcription in cervical intraepithelial neoplasia by reverse transcriptase polymerase chain reaction. Int J Cancer 1993;55:397 401. 16. Sotlar K, Selinka HC, Menton M, Kandolf R, Bultmann B. Detection of human papillomavirus type 16 E6/E7 oncogene transcripts in dysplastic and nondysplastic cervical scrapes by nested RT-PCR. Gynecol Oncol 1998;69:114 21. 17. Munoz N, Bosch FX, Castellsague X, Diaz M, de SS, Hammouda D, et al. Against which human papillomavirus types shall we vaccinate and screen? The international perspective. Int J Cancer 2004;111:278 85. 18. Smith JS, Lindsay L, Hoots B, Keys J, Franceschi S, Winer R, et al. Human papillomavirus type distribution in invasive cervical cancer and high-grade cervical lesions: a meta-analysis update. Int J Cancer 2007;121:621 32. 19. Bruisten S, van Gemen B, Koppelman M, Rasch M, van Strijp D, Schukkink R, et al. Detection of HIV-1 distribution in different blood fractions by two nucleic acid amplification assays. AIDS Res Hum Retroviruses 1993;9:259 65. 20. el-Galil KH, el-Sokkary MA, Kheira SM, Salazar AM, Yates MV, Chen W, et al. Real-time nucleic acid sequence-based amplification assay for detection of hepatitis A virus. Appl Environ Microbiol 2005;71:7113 6. 21. Greijer AE, Adriaanse HM, Dekkers CA, Middeldorp JM. Multiplex realtime NASBA for monitoring expression dynamics of human cytomegalovirus encoded IE1 and pp67 RNA. J Clin Virol 2002;24:57 66. 22. Lanciotti RS, Kerst AJ. Nucleic acid sequence-based amplification assays for rapid detection of West Nile and St. Louis encephalitis viruses. J Clin Microbiol 2001;39:4506 13. 23. Loens K, Ieven M, Pattyn S, Sillekens P, Goossens H. Sensitivity of detection of rhinoviruses in spiked clinical samples by nucleic acid sequence-based amplification in the presence of an internal control. J Microbiol Methods 2006;66:73 8. 24. Boom R, Sol CJ, Salimans MM, Jansen CL, Wertheim van Dillen PM, van der Noordaa J. Rapid and simple method for purification of nucleic acids. J Clin Microbiol 1990;28:495 503. 25. Tyagi S, Kramer FR. Molecular beacons: probes that fluoresce upon hybridization. Nat Biotechnol 1996;14:303 8. 26. Tyagi S, Bratu DP, Kramer FR. Multicolor molecular beacons for allele discrimination. Nat Biotechnol 1998;16:49 53.
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
27. Deiman B, van AP, Sillekens P. Characteristics and applications of nucleic acid sequence-based amplification (NASBA). Mol Biotechnol 2002;20:163 79. 28. Brink AA, Vervoort MB, Middeldorp JM, Meijer CJ, van den Brule AJ. Nucleic acid sequence-based amplification, a new method for analysis of spliced and unspliced Epstein Barr virus latent transcripts, and its comparison with reverse transcriptase PCR. J Clin Microbiol 1998;36:3164 9. 29. Greijer AE, Dekkers CA, Middeldorp JM. Human cytomegalovirus virions differentially incorporate viral and host cell RNA during the assembly process. J Virol 2000;74:9078 82. 30. Hayes DP, Brink AA, Vervoort MB, Middeldorp JM, Meijer CJ, van den Brule AJ.
S37
Expression of Epstein Barr virus (EBV) transcripts encoding homologues to important human proteins in diverse EBV associated diseases. Mol Pathol 1999;52:97 103. 31. Molden T, Kraus I, Skomedal H, Nordstrøm T, Karlsen F. PreTect™ HPV-Proofer: Real-time detection and typing of E6/E7 mRNA from carcinogenic human papillomaviruses. J Virol Methods 2007;142:204 12. 32. van Hamont D, van Ham MA, Bakkers JM, Massuger LF, Melchers WJ. Evaluation of the SPF10-INNO LiPA human papillomavirus (HPV) genotyping test and the Roche linear array HPV genotyping test. J Clin Microbiol 2006;44:3122 9.
Contents lists available at ScienceDirect
Journal of Clinical Virology j o u r n a l h o m e p a g e : www.elsevier.com/locate/jcv
NucliSENS® EasyQ® HPV v1 test human papillomaviruses
Testing for oncogenic activity of
Damien Jeanteta , Fritz Schwarzmanna , Jessica Trompb , Willem J.G. Melchersc , Anneke A.M. van der Wurffd , Tom Oosterlakenb , Marcel Jacobsb , Alain Troescha, * a
bioM´ erieux, Grenoble, France bioM´ erieux bv, Boxtel, The Netherlands Department of Medical Microbiology, Radboud University, Nijmegen, The Netherlands d St. Elisabeth Hospital, Tilburg, The Netherlands b c
ARTICLE
INFO
Keywords: Human papilloma virus E6/E7 mRNA NucliSENS® EasyQ® Nucleic acid sequence-based amplification NASBA Cervical carcinoma HPV diagnostics
ABSTRACT Background: Analytical sensitivity of DNA-based assays to detect infection with human papillomaviruses is very high, but clinical specificity for cervical cancer strongly depends on the age of the patient and case classification. To solve the dilemma between sensitivity and specificity, a new generation of assays focuses on the pathogenic factors that underlie the development of HPV-associated tumors: the expression of the viral oncogenes E6 and E7. Demonstration of persistent expression of these mRNAs or expression in the context of relevant clinical symptoms has a strong positive predictive value for the development of HPV-associated carcinomas and strongly warrants further diagnostic action. Objectives: The NucliSENS® EasyQ® HPV v1 test was designed to test cervical scrapes for the expression of the oncogenic E6/E7 mRNA from the five most common carcinogenic HPV types (16, 18, 31, 33 and 45). The test can be used for confirmation and risk stratification of individuals with proven infection with high risk papillomaviruses. Study design: In order to establish performance of the assay, sensitivity, specificity, repeatability, and reproducibility were determined with artificial and clinical specimens. Further, a total of 420 cervical scrapes were tested and the results directly compared to the CE-market device PreTect HPV-Proofer assay (NorChip, Klokkarstua, Norway). For arbitration of discordant clinical results, the specimens were rated according to Pap-classification and the presence of HPV DNA was determined. Results: The limit of detection for the five HPV types 16, 18, 31, 33, and 45 ranged from 2.3×102 to 3.0×104 copies/mL on a background of 5×103 HPV-negative HS67 cells. No crossreactivity for other viral, bacterial, or fungal agents known to be potentially present in cervical fluids was detected. Repeatability and reproducibility were shown by testing panels of HPV-spiked artificial and clinical samples. A comparative analysis of 420 cervical scrapes demonstrated an overall concordance of >90% between the NucliSENS EasyQ HPV test and the technologically related PreTect HPV-Proofer assay. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Human papillomaviruses (HPV) have been identified as the causative agents of a number of benign and malignant tumors. These viruses are strictly tissue-specific and are associated with malignant tumors of the skin, the head and neck, and anogenital region.1,2 Over 100 subtypes are described. * Corresponding author. bioM´ erieux, Centre Christophe M´ erieux, 5, Rue des Berges, F-38024 Grenoble Cedex 01, France. E-mail address: [email protected] (A. Troesch). 1590-8658/ $
see front matter © 2009 Elsevier B.V. All rights reserved.
In general, first diagnosis of infection with HPV is done indirectly by visual inspection of the mucosa and cytological examination for abnormal cells. For advanced and persistent infections with HPV, the demonstration of over-expressed cellular p16 protein in the infected tissues may support diagnosis.3,4 However, reliable confirmation of diagnosis is possible only by direct detection of the infectious agent’s nucleic acids, since expression of viral proteins varies between active, latent, and persistent infections with the chance of integrated viral genomes and reduction in gene expression.
S30
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
Immunology-based diagnosis is not available, since humoral immune responses against HPV are weak and unreliable. Although cellular immunity plays a pivotal role in the control of infected cells, no assay exists to date for the detection and monitoring of HPV-specific T-cells.5,6 Furthermore, differentiation between the HPV types is possible only by molecular genetics. Introduction of HPVdiagnostic tests in addition to cytology (Pap smear) significantly increased the sensitivity of screening for cervical cancer.7 Several assays exist, most of them based on nucleic acid amplification of viral DNA. Analytical sensitivity of these assays is very high, but clinical specificity for high-grade lesions depends strongly on the age of the patient and the cytological case classification. In this respect, DNA-based assays with a balanced sensitivity and specificity did prove suitable for initial screening and for the detection and prediction of high-grade lesions (CIN 3) in women 30 years and older.8 Due to the high prevalence of infection and the transient nature of infection with HPV, the specificity and the positive predictive value of DNA-based assays for the detection of highgrade lesions is not satisfactory in women younger than 30 years. Furthermore and most important, in the context of triage of patients with abnormal cytological results and/or clinical symptoms of infection, these assays show only low specificity and positive predictive value for high-grade lesions (CIN 3).9,10 The diagnostic approach of bioM´ erieux is targeted at the identification of those individuals who are at higher risk of developing HPV-associated carcinomas (Fig. 1). Accordingly, the NucliSENS® EasyQ® HPV v1 assay is intended for confirmation or triage of individuals with the diagnosis of active infection with HPV types of the high-risk group. Risk stratification, aimed to identify those women with a higher potential to develop a carcinoma, is based on the underlying molecular mechanisms of HPV-related tumor development and tumor progression. With respect to the development of cervical cancer, a number of strains/genotypes of HPV have been classified into high-risk and low-risk according to their oncogenic potential.11 Briefly, most HPV infections, belonging to both the high-risk and low-risk groups, are transient and the viruses are eliminated from the host within six to eighteen months after infection. With infection by viruses of the high-risk group, the risk to develop a persistent infection, which is a prerequisite for the development of a malignant tumor, increases significantly. In the early phases of infection,
generally only episomal viral DNA is detected. However, the chance of integration of the virus into the host genome increases with the duration of the infection. In cervical carcinomas mostly integrated viral DNA is detected.12,13 In rare cases, integration of HPV genomes results in the inactivation of the viral regulatory gene E2, which in turn results in deregulation of the viral tumor genes E6 and E7. Expression of these genes is accepted to be a prerequisite for tumor progression since activity of these genes results in increased cell proliferation, prevention of apoptosis, and an increase in mutagenesis. Thus, monitoring of the activity of E6 and E7 by detecting the corresponding mRNA seems to be a reasonable strategy to stratify individuals with proven infection with high-risk HPV.14 16 The NucliSENS EasyQ HPV v1 is a real-time nucleic acid amplification and multiplex detection assay for the qualitative determination of E6/E7 mRNAs of the five most commonly identified carcinogenic HPV types in cervical cancer worldwide (HPV 16, 18, 31, 33 and 45). These five HPV genotypes are detected in approximately 82% of all cervical carcinomas.17,18 The NucliSENS EasyQ HPV v1 assay utilizes the nucleic acid sequence-based amplification (NASBA) technology that has been used in a wide variety of diagnostic assays, detecting both viral and bacterial nucleic acids.19 23 NucliSENS EasyQ HPV version 1 was launched in 2007 and was based on the original PreTect Proofer assay developed by NorChip (Klokkarstua, Norway), except for the NucliSENS hardware platform and the software for NASBA measurements and data analysis (NucliSENS Director software). The latest NucliSENS EasyQ HPV assay version 1 has recently been improved in the following ways: taking into account the RNA nature of the HPV-specific targets, the five positive controls for the different subtypes and the U1A internal control were changed from singlestranded DNA to RNA; in order to facilitate handling and to increase reproducibility of the assay, the positive controls are no longer provided in liquid phase but in freeze-dried accuspheres; in order to improve and harmonize the sensitivity for the five HPV types covered by the assay, the concentrations of the primers and beacons for HPV 16 and the U1A internal control were re-adjusted; the extraction protocol on NucliSENS EasyMAG® system was optimized; the HPV assay protocols were updated and the NucliSENtral™ HPV software V1.1 introduced to further improve the convenience of operator use for results reporting. 2. Materials and methods 2.1. Laboratory equipment
Fig. 1 The stepwise development of cervical cancer. The present RNA-based NucliSENS® EasyQ® HPV v1 test is intended to stratify individuals with confirmed active infection with high-risk HPV types. Testing is based on the detection of the viral oncogenic E6 and E7 mRNAs of HPV types 16, 18, 31, 33, and 45.
The NucliSENS EasyQ HPV v1 assay is designed for the bioM´ erieux NucliSENS platform. Nucleic acid extraction protocols have been validated on two extraction systems, either the semi-automated NucliSENS miniMAG® or the automated NucliSENS easyMAG® (bioM´ erieux, Marcy l’Etoile, France). Amplification, real-time detection and fully automated interpretation of the test results are done on a NucliSens EasyQ analyser used for all NucliSENS assays. The analytical part of the assay is performed in strips of microliter reaction tubes. One run covers up to 30 patient samples at a time (Fig. 2).
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
Fig. 2 EasyQ® HPV v1 Workflow. Nucleic acids are isolated using the NucliSENS® miniMAG® or the NucliSENS easyMAG® . The NucliSENS EasyQ HPV v1 assay utilizes the Nucleic Acid Sequence Based Amplification (NASBA). The time required to complete the assay (sample preparation, extraction of nucleic acids, amplification and detection, and evaluation of results) is 4 hours for 24 samples.
2.2. Assay description Nature of specimens and isolation of nucleic acids The NucliSENS EasyQ HPV v1 assay is intended to be used with cervical scrapings. Due to the nature of the mRNA targets and their inherent sensitivity to nucleases, non-stabilized specimens like dry swabs are not suitable for testing. Thus, for preservation of the E6- and E7-specific mRNA, the cervical samples must be stabilized immediately after collection in transport media such as PreservCyt (Hologic. Inc., Marlborough, MA). The material is stable in this buffer at room temperature for up to six weeks. The use of other stabilizing media is currently under investigation. Due to the nature of the specimen, which always shows irregular numbers of cells, the sample is mixed thoroughly before use. A 5 mL volume of sample is transferred into a fresh 10 mL tube for extraction and then centrifuged for 12 minutes at 1125×g. The supernatant is removed leaving approximately 1 mL PreservCyt solution. After the re-suspension of the specimen’s cells in the remaining 1 mL PreservCyt solution, the sample is transferred to another tube containing 2 mL NucliSENS Lysis Buffer (bioM´ erieux, Marcy l’Etoile, France). Any cellular material or viral particles present in the sample are disrupted, RNases and DNases are inactivated, and total nucleic acids are released. After cell lysis, magnetized silica dioxide particles are added to the lysate of cervical specimen to initiate the RNA isolation process.24 The nucleic acids present in the NucliSENS Lysis Buffer bind to the magnetic silica under the high salt conditions. After binding of the nucleic acids, the magnetic silica is washed several times in order to remove contaminating and potentially inhibitory cellular components. The nucleic acids are eluted from the solid phase thereby making them available for use in the amplification and detection procedures. The mRNA is eluted in 55 mL of elution buffer and stored at 20ºC if not further processed immediately after extraction. Multiplex real-time NASBA The NucliSENS EasyQ HPV v1 assay utilizes a combined process of nucleic acid amplification and simultaneous detection of the amplicon using molecular beacons.25,26 All primers and probes in the assay are designed to detect the full-length mRNA transcripts of U1A and the E6/E7 region from HPV types 16, 18, 31, 33 and 45. A full-length transcript is defined as any un-spliced transcript covering at least the E6/E7 region and may also be fused to human sequences. The molecular beacons enable a real-time and multiplex assay suitable for high-throughput applications with increased specificity, without compromising
S31
sensitivity. In contrast to the widely used polymerase chain reaction (PCR), which is based on the use of thermo-stable DNA polymerase and temperature cycling, NASBA is an enzymatic amplification process that amplifies in isothermal conditions (41ºC). Three enzymes are involved in the amplification of the nucleic acids (either DNA or RNA): avian myeloblastosis virus reverse transcriptase (AMV-RT), Escherichia coli RNase H, and T7 RNA polymerase. The NASBA reaction continues in a selfsustained manner, and is capable of achieving amplification of approximately 1013 1014 -fold within 90 minutes. Briefly, the NASBA amplification is based on a repeated process of primer annealing, formation of double-stranded DNA containing a T7 promoter site, and T7 RNA polymerasemediated transcription of multiple anti-sense copies of RNA target sequences (amplicons). The principal amplification product of NASBA is single-stranded RNA that subsequently can easily be identified and characterized. Figure 3 provides an overview of the main steps involved in the NASBA reaction.27 The NucliSENS EasyQ HPV v1 test uses multiple primer sets for amplification of the five HPV genotypes HPV 6, HPV 18, HPV 31, HPV 33, and HPV 45. In order to preclude false-negative test results due to insufficient cellular material, degradation of the mRNA, or inhibition of amplification, the quality of the extracted nucleic acid is monitored by testing for an internal control mRNA, human U1 small nuclear ribonucleoprotein specific protein A (U1A). The U1A protein is part of the housekeeping U1snRNP, which binds to the 5 exon of hnRNAs and initiates spliceosomal assembly. The U1A-specific mRNA, which is present in low copy numbers in the cell, has been described previously as a suitable internal control in molecular-genetic virological assays.28 30 To this aim, a primer pair targeting U1A mRNA is included in the assay. Besides the U1A control, five positive controls consisting of artificial E6/E7-target sequences specific for the HPV types HPV 16, HPV 18, HPV 31, HPV 33 and HPV 45 are included to monitor functionality of primers, probes and NASBA reagents. The internal U1A control and the five HPV-specific controls are provided in one freeze-dried accusphere, which will be diluted in the corresponding buffer and used directly in amplification. The detection of HPV-specific amplification products is accomplished in real-time by the use of target-specific molecular beacons, which fluoresce only upon hybridization with their target. To detect amplification of the specific mRNAs, six different molecular beacons are used in the NucliSENS EasyQ HPV v1 assay, one specific beacon for each of the HPV subtypespecific amplicons and one for the U1A amplicons. The use of two fluorescent dyes (6-FAM and 6-ROX) allows the detection of the HPV target RNAs in three duplex amplification reactions: U1A and HPV 16, HPV 18 and HPV 31, HPV 33 and HPV 45. Interpretation of results The HPV Client analysis software (NucliSENtral™ HPV software V1.1) has been designed to automate parts of the run workflow and to allow the automated determination of the run and sample validity and interpretation of sample results. As depicted in Fig. 3, the NASBA process is characterized by an early exponential growth of a double-stranded intermediate DNA amplicon, followed by a linear increase in RNA levels which translates into a linear increase of fluorescence. The NucliSENS EasyQ Director Software determines the fluorescence values for each individual amplification taking place and validates the measured values against a number of validation criteria (statistical), the quality of the curve fit, and the timing of the
S32
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
Fig. 3. Principle of real-time NASBA. A template-specific DNA intermediate is initially generated through a process that involves the binding of a primer (primer 1) to the RNA template. This primer, which contains a T7 RNA polymerase promoter site, is then extended by AMV-RT to form an RNA DNA duplex. Degradation of the RNA strand of the duplex by RNase H permits the binding of a second primer (primer 2) to the remaining DNA strand. The second primer encompasses a short sequence which is identical to a segment of the target RNA and is located upstream of the region where the first oligonucleotide can anneal. The second primer is then extended by AMV-RT to form the double-stranded DNA intermediate, which contains the T7 RNA polymerase promoter needed for transcription. Once transcription is initiated, the resulting RNA transcripts, which are anti-sense to the original RNA present in the original specimen, can serve as a template to start a new amplification process.
onset of the fluorescence increase, which (to a limited extent) corresponds to the Ct value in real-time PCR. Furthermore, the final qualitative test result will account for the validity of the negative and the three positive controls, the internal control, and the results of the five HPV subtype-specific reactions. If a test result has been qualified as valid, a qualitative result (positive/negative) is reported for each HPV target.
Table 1 List of microbial agents tested for analytical specificity Mycoplasma hominis Escherichia coli Chlamydia trachomatis Gardnerella vaginalis Enterococcus faecalis Ureaplasma urealyticum Staphylococcus aureus
Mycoplasma genitalium Neisseria gonorrhoeae Candida albicans Human immunodeficiency virus Hepatitis C virus Hepatitis B virus
2.3. Determination of assay analytical sensitivity The analytical sensitivity of the assay was determined by analyzing a dilution series of in vitro generated E6/E7derived RNA specific for each HPV type (length around 1000 nucleotides, depending on the HPV type) on a background of 5×103 HPV-negative human cells (HS67 cell line). The input of transcripts ranged from zero to 107 copies per extraction. The concentration of in vitro generated RNA was determined by absorption spectroscopy. Subsequently, nucleic acids were extracted using a NucliSENS easyMAG and NucliSENS miniMAG® device and tested in replicates for the presence of E6/E7 RNA. Analytical sensitivity defined as the 50% and 95% hit rate values were determined by statistical Probit analyses.
done using the BLAST algorithm (Genebank, National Center for Biotechnology Information, Bethesda, MD). No homology alignments were found with both primers and the molecular beacons within a range of approximately 1000 nucleotides, being the maximal size of a NASBA amplification product under normal reaction conditions. Additionally, a selection of microbial agents either known to cause genital tract infections or to be potentially present in cervical secretions (Table 1), were tested. Briefly, the agents were cultured in vitro and 20 mL of the cultures were added to 2 mL lysis buffer tubes for extraction of the nucleic acids in a total volume of 55 mL. Five mL were used as input for the HPV-specific NASBA assay. Analytical specificity was defined as the percentage of false positive results for HPV.
2.4. Determination of assay analytical specificity
2.5. Determination of assay repeatability and reproducibility
To determine the analytical specificity of the assay, in first instance, sequence homology alignment analyses of the primers and molecular beacons with different organisms were
Assay repeatability and reproducibility were tested both in an analytical and a clinical background (patient specimen). For
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
2.6. Performance evaluation of the NucliSENS EasyQ HPV v1 using clinical samples An evaluation of the assay performance was done in direct comparison of the NucliSENS EasyQ HPV v1 and the established CE-marked device PreTect HPV-Proofer by NorChip AS (Klokkarstua, Norway). A total of 420 cervical scrapes in PreservCyt buffer were collected at two sites from women participating in a routine cervical cancer screening program. Samples were collected and tested in accordance with the approval of the Medical Ethical Committee of St. Elizabeth Hospital, Tilburg and in accordance with the policies of Radboud University, Nijmegen, for the use of discard de-identified samples. For all samples, routine diagnostics based on cytology was performed. A total of 248 samples were classified as Pap 1 (normal cells), 88 samples were found to be Pap 2 (mild inflammation), 80 samples were Pap 3 (mild, moderate or severe dysplasia), three samples were Pap 4 (carcinoma in situ), and one sample was classified as Pap 5 (invasive carcinoma).
progress to cervical carcinoma. Thus, the analytical sensitivity of DNA- and RNA-assays for detecting HPV viruses cannot be compared directly. The concept of detecting E6/E7 mRNA of high-risk HPV genotypes is to reduce non-relevant positive results and thus to increase the positive predictive value for cervical carcinoma. 3.1. Assay analytical sensitivity In order to assess the analytical sensitivity for viral E6/E7 mRNA, the limit of detection was determined by analyzing dilution series of in vitro generated RNA for each of the five HPV types. To mimic clinical samples, 5×103 human cells from an HPV-negative human cell line (HS67) were lysed and the diluted RNA was added before extraction. The U1A mRNA, which is used as intrinsic control, was contained in the cellular RNA. No additional positive and negative run controls were included in the amplification run, since positive and negative controls were already included in the RNA dilution series, considering the sample containing no HPV RNA as a negative control and the sample containing the highest in vitro transcribed RNA input as the positive control. Each input of in vitro transcribed RNA was tested in twenty replicates and the percentage of replicates tested positive (hit rate) was determined. The calculated analytical sensitivity (defined as the estimated concentration with 95% hit rate) was shown to range from 2.3×102 to 3.0×104 copies/mL, showing the highest sensitivity for HPV 45 and the lowest for HPV 31. The results are summarized in Fig. 4 and Table 2. 120
Probit HPV 18 100 % positive results
analytical studies, E6/E7-specific in vitro generated RNA was spiked in PreservCyt buffer at input levels of 103 copies/ml and 105 copies/ml corresponding to estimated hit rates of 1% and 99% determined previously on the basis of Probit analysis. Analytical repeatability was determined by the concurrent testing of twenty replicates of each input in one laboratory, by the same operator and using the same equipment. Analytical reproducibility was determined by testing the same panel in two laboratories by two operators using different equipment. For statistical analysis, data were compared using Fisher’s exact test. For studies with clinical specimens, ten HPV-negative cervical scrapes, five HPV 16-positive cervical scrapes, and five HPV-negative cervical scrapes that were spiked with HPV 16 at 105 copies/ml corresponding to 99% hit rate, were used. Nucleic acids were extracted using easyMAG. Repeatability was determined by concurrently testing these specimens in triplicate. Reproducibility was assessed by a second operator in a different laboratory using different equipment. Statistical concordance was assessed by pairwise comparison of the data of one operator using the same equipment in the same laboratory (repeatability), and comparison of the data of two different operators using different equipment in different laboratories (reproducibility) by Cohen’s kappa (ú) statistic. A ú estimate of less than 0.4 represents poor agreement, a ú estimate between 0.4 and 0.75 is fair to good agreement, and a ú estimate of more than 0.75 is excellent agreement.
S33
80 60 40
Probit 95% lower limit
20
95% upper limit measured
0 1
2
3
4
5
log input
120
Probit HPV 33
3. Results DNA-based assays for HPV show a very high sensitivity and specificity for the presence of human papillomaviruses. Thus, these assays significantly enhance the sensitivity of cervical cancer screening. However, due to the high prevalence of HPV and the transient nature of most infections, specificity and the positive predictive value of these tests for high-grade lesions and progression to cervical carcinoma strongly depend on the age of the women and the clinical presentation at the time of diagnosis. In contrast, the negative predictive value is high and the cancer risk is very low when the test is negative. In this respect, the present assay, based on the detection of the viral oncogenic E6 and E7 mRNA, is intended to detect specifically those HPV infections which bear a potentially higher risk to
% positive results
100 80 60 40 Probit 95% lower limit
20
95% upper limit measured
0 1
1,5
2
2.5 3 log input
3.5
4
Fig. 4 Probit analysis for HPV 18 and HPV 33 to determine the analytical sensitivity.
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
S34
Table 2 Analytical sensitivity as defined as the 50% and 95% hit rate values of in vitro RNA dilution series calculated by using Probit analysis Hit rate
In vitro HPV RNA (copies/mL) HPV 16
50% 95 %
HPV 18 1
4.0×10 2.6×10
2
HPV 31 2
8.8×10
3
6.4×10
HPV 33 3
4.3×10
4
3.0×10
HPV 45 2
3.5×101
3
2.3×102
5.6×10 3.0×10
3.2. Assay analytical specificity The specificity of the NucliSENS EasyQ HPV v1 assay is based upon careful selection of primers and probes taking into account genetically related viruses and other organisms that are known to occur in cervical scrapes. Cross-reactivity of various micro-organisms was assessed by testing directly nucleic acids from these organisms in the NucliSENS EasyQ HPV v1 assay. Briefly, pure bacterial or viral cultures were diluted one hundred fold in lysis buffer for extraction of nucleic acids and 5 mL of the eluate was used for the NASBA. No cross-reactivity was found with the organisms listed in Table 1. The absence of cross-reactivity for other HR HPV types or other LR HPV types has been demonstrated by theoretical sequence homology alignment analyses and by some preliminary experiments on a limited set of clinical samples found positive for HPV 6, 52 or 58 using InnoLipa (Gent, Belgium) (data not shown). More extensive experimental studies will be performed to further guarantee the specificity of the assay. 3.3. Assay repeatability and reproducibility Repeatability and reproducibility of the NucliSENS EasyQ HPV v1 assay was assessed by testing panels of artificial and clinical specimens. Briefly, repeatability was determined by successive testing of sample panels at one site by one operator. Reproducibility implied testing of the panels at a second site by at least one additional operator. Two types of panels were employed for these studies, either based on in vitro generated HPV 16-derived E6/E7-specific RNA spiked in PreservCyt buffer (artificial specimens) or based on clinical specimens. Since the NucliSENS EasyQ HPV v1 is a qualitative assay, statistical analysis was performed on the number of positive or negative results, and not on the underlying quantitative fluorescent values. Analysing the artificial samples with spiked E6/E7-specific RNA, with the high positive samples (105 copies/mL; 99% hit rate) no statistical analysis was performed, since all the replicates were tested 100% positive. With the low input samples (103 copies/mL; 1% hit rate) we observed hit rates of 10% (2/20), 0% (0/20), and 10% (2/20) for three consecutive tests. Statistical analysis using Fisher’s exact test yielded a p-value of 0.53 for repeatability and a p-value of 1.0 for reproducibility, indicating that no statistically significant differences between the individual tests were observed. During testing of the clinical specimens, three invalid results were obtained with HPV-negative samples due to negative results for the U1A internal control. Thus, the total number of clinical samples was reduced from 20 to 17. In terms of repeatability, 100% (10/10) of each of the clinical samples tested positive for all three replicates. In total, 30/30 positive samples were tested positive. Of the 7 (valid) negative samples, 100% (21/21) of the replicates were tested negative.
With respect to repeatability, one of the ten positive samples tested negative in one of two duplicates. For the negative samples, again, three invalid results were obtained with HPV-negative samples due to negative results for the U1A internal control. Thus, the total number of samples in the study was reduced from 20 to 17. Of the 7 (valid) negative samples, 100% of the duplicates (14/14) tested negative. This resulted in 0.88 concordance and demonstrated sufficient clinical repeatability and reproducibility. 3.4. Comparison of the NucliSENS EasyQ HPV v1 to the PreTect HPV-Proofer In order to evaluate the clinical performance of the NucliSENS EasyQ HPV v1 assay, a study with a total of 420 cervical scrapes was conducted. Specimens were collected at two study sites from women participating in a routine cervical cancer screening program or under their gynecologists’ care for abnormal smears. In addition, all samples were analyzed in parallel with the established CE-marked device PreTect HPV-Proofer by NorChip AS (Klokkarstua, Norway). The results obtained in this study are summarized in Table 3. For the calculation of the percentage of concordance and for the calculation of Cohen’s kappa, invalid results (due to invalid controls) and undetermined results of the PreTect HPV-Proofer assay (due to values of fluorescence in the grey zone of the assay, i.e. between 1.4 and 1.7 Relative Fluorescence Units) were rejected.31 Ninety percent concordance and a Cohen’s kappa of 0.706 demonstrated an equivalent performance between the NucliSENS EasyQ HPV v1 and the PreTect HPVProofer assay. Interestingly, the concordance for the individual subtypes was significantly higher, showing 96%, 97%, 95% and 98% for HPV 16, HPV 18, HPV 31 and HPV 45, respectively. This difference in overall and subtype-specific concordance was due to mixed infections with several HPV types. (The results for HPV 33 were retrospectively withdrawn from this part of the study, due to the suspicion of contamination.) Table 4 summarizes the discordant results obtained in this part of the study. Samples which yielded invalid results with one of the two tests were excluded from this analysis. For arbitration of the results, cytological classification is shown. Herein, sixty-four percent (16/25) of the discordant results were obtained with samples classified as Pap 1. Two and seven discordant results were derived from Pap 2 and Pap 3 samples (Table 5). Interestingly, for two of the three Pap 3 specimens, Table 3 Overall concordance of E6/E7 mRNA test results of 420 cervical scrapings in pos/neg classification between the NuclSENS and the PreTect HPV-Proofer test PreTect HPV-Proofer Invalid
a
NucliSENS Invalid
Negative
Positive
Total
61
57
13
131
Negative
22
171
20
213
Positive
6
4
40
50
Indeterminate a
0
12
14
26
89
244
87
420
Total a b
invalid = tests with at least one invalid control. indeterminate = tests with fluorescence values in the grey zone (between 1.4 and 1.7 relative fluorescence units) of the PreTect HPV-Proofer assay.
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
S35
Table 4 Discordant results between NucliSENS® EasyQ® HPV v1 and the established CE-marked PreTect HPV-Proofer ID
Pap
LiPA
PreTect HPV-Proofer
EasyQ Results U1A
16
18
31
45
33
U1A
16
18
31
45
33
Site 1 2
1
31
+
3
1
16
+
19
1
22
1
16
+
69
1
33
+
71
1
31
+
72
1
33
+
76
1
83
3a1
85
1
87
3a2
96
1
122
1
136
1
146
1
+
+/
+
+
+ + +
+ + +
+ 18
+
+
+
+
+
+ +
I
+
+
I
+
+
+
+
+
+ 45
I +
16 31
+/
+
+
+
+
+/
Site 2 102b
2
+
280
1
+
305b
3a
+
396
1
404
1
+
547b
3a
+
596b
2
+
605b
3a
+
+
882b
3a
+
+
+
972b
3a
+
+
+
+
+
+
+ +
+
+
+ + +
+
+
+
+
+
+
+
+ +/
The upper and lower parts of the table show the results obtained with samples collected at two different study sites. The data on cytological classification according Pap are shown in the column Pap. For samples selected at Site 1, data on HPV DNA were available as well; these are listed in the LiPA column. See Table 5 for the correlation of discordant results with Pap classification.
Table 5 Summary of correlation of discordant results with Pap classification Pap classification
1
2
3
4
5
Number of discordant results
16
2
7
0
0
a result from DNA testing was available which confirmed the positive result of the NucliSENS EasyQ HPV assay. For 15 samples there was additional information on HPV DNA available (InnoLipa, Gent, Belgium). The detection of HPV DNA was in agreement with the NucliSENS EasyQ HPV assay for 5 samples containing HPV types 16, 18, 31 or 33, all of which were not detected by the PreTect HPV-Proofer assay. The NucliSENS EasyQ HPV assay detected HPV mRNA (type 16) in 2 samples, negative by both the PreTect HPV-Proofer assay and InnoLipa. Five samples contained HPV DNA (types 16, 31, 33, or 45) that were not detected by the NucliSENS EasyQ HPV assay. The PreTect HPV-Proofer assay detected HPV mRNA (types 16 and 33) in 2 samples negative by the NucliSENS EasyQ HPV assay and positive by InnoLipa and in 3 samples (types 16, 18 and 33) that were negative by both other assays. Five
samples contained HPV DNA (types 16, 18, 31, or 33) that was not detected by the PreTect HPV-Proofer assay. Three samples HPV DNA positive had a different HPV mRNA type detected by the PreTect HPV-Proofer assay (16 vs 33, 31 vs 33, and 45 vs 33, respectively). The InnoLipa results confirmed 5 of the 7 positive results obtained with the NucliSENS EasyQ HPV assay. 4. Conclusions and summary Introduction of molecular genetic testing for detecting HPV infections has significantly increased the sensitivity and efficacy of cervical cancer screening previously based solely on cytology and Pap testing. However, new testing for HPV E6/E7 mRNA expression has focused on detecting not just HPV infection but the detection of relevant viral (oncogenic or pre-cancerous) activity. Thus, these assays are intended to be used as a tool for the identification of clinically relevant HPV disease and/or for triage and risk stratification of patients with abnormal cytology and/or clinical symptoms. To this aim, the NucliSENS EasyQ HPV v1 test was designed to detect E6/E7 oncogenic activity from the five most
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
S36
carcinogenic HPV types found most frequently in cervical cancer worldwide. Demonstration of persistent expression of these mRNAs or expression in the context of relevant clinical symptoms has a strong positive predictive value for the development of HPV-associated carcinomas and strongly warrants further diagnostic action improving patient management. This has been shown impressively in a recent publication by Szarewski et al.9 presenting the results of a comparative study that determined the sensitivity, specificity and positive predictive value for the detection of CIN 2 and CIN 3 of seven commercially available clinical diagnostic tests and diagnostic approaches: detection of oncogenic HPV-derived RNA, demonstration of HPV DNA, or detection of cellular (HPVunrelated) markers for transformation. In this context cervical scrapes from 953 women with abnormal cervical smears were analysed and the results were compared to the gold standard of histopathology. The NucliSENS EasyQ HPV test was not included in this study in contrast to the technologically related PreTect HPV-Proofer assay. In this study, the best results in terms of specificity were obtained with the E6/E7 mRNA-based assay (PreTect HPV-Proofer) (73.1% and 70.4% for CIN 2+ and CIN 3+, respectively), closely followed by detection over-expression of p16INK4A (68.7% and 65.8% for CIN 2+ and CIN 3+, respectively). However, sensitivity of both the E6/E7 mRNA-based assay (73.6% and 82.2% for CIN 2+ and CIN 3+, respectively) and the p16INK4A -based assay (83.0% and 92.7% for CIN 2+ and CIN 3+, respectively) were significantly lower than that of some of the DNA-based assays (range 80.9 99.6%).9 Due to the high prevalence of infection and relatively low incidence of HPV-related tumours, a balanced sensitivity and specificity is of utmost importance for the diagnosis of high-grade lesions.32 This is true both for DNA-based assays and for RNA-based approaches. In this respect, another RNA amplification assay, the APTIMA® HPV Assay (GenProbe, Inc, San Diego, CA), targeting E6/E7 mRNA of 14 high-risk HPV types showed sensitivity similar to DNA assays (95.2%), better specificity (42.2%) than DNA assays (range 21.7% to 37.1%), but lower specificity than other RNA-based assays for high-grade disease.9 Owing to the lower sensitivity of the NucliSENS EasyQ HPV v1 test compared to DNA-based assays, at present, the NucliSENS EasyQ HPV v1 test is intended only for the confirmation and the risk stratification of individuals with proven infection with highrisk HPV. Ongoing improvement of the assay in the future, with respect to the number of HPV types covered, will comply with the broad change in the perception of viral mRNA detection for cervical cancer screening. In any case, comprehensive clinical studies will be essential to evaluate the performance and predictive values of any HPV assay with respect to the intended use as a tool for screening or for triage and risk stratification of patients with abnormal cytology and/or clinical symptoms. Acknowledgements: We are grateful to V´ eronique BaronWunderl´ e and Bernard Fleury (bioM´ erieux) for their critical review of the manuscript. Competing interests: DJ, FS, JT, TO, AT are employees of bioM´ erieux. AMvdW, WJGM: none declared. MJ: none declared. References 1. Munoz N, Castellsague X, de Gonzalez AB, Gissmann L. Chapter 1: HPV in the etiology of human cancer. Vaccine 2006;24(Suppl 3):S3-1 10.
2. zur Hausen H. Papillomaviruses in the causation of human cancers a brief historical account. Virology 2009;384:260 5. 3. Wentzensen N, Bergeron C, Cas F, Eschenbach D, Vinokurova S, von Knebel Doeberitz M. Evaluation of a nuclear score for p16INK4a-stained cervical squamous cells in liquid-based cytology samples. Cancer 2005;105:461 7. 4. Wentzensen N, Bergeron C, Cas F, Vinokurova S, von Knebel Doeberitz M. Triage of women with ASCUS and LSIL cytology: use of qualitative assessment of p16INK4a positive cells to identify patients with high-grade cervical intraepithelial neoplasia. Cancer 2007;111:58 66. 5. Carter JJ, Koutsky LA, Hughes JP, Lee SK, Kuypers J, Kiviat N, et al. Comparison of human papillomavirus types 16, 18, and 6 capsid antibody responses following incident infection. J Infect Dis 2000;181:1911 9. 6. Frazer IH. Interaction of human papillomaviruses with the host immune system: a well evolved relationship. Virology 2009;384:410 4. 7. Wright TC, Schiffman M, Solomon D, Cox JT, Garcia F, Goldie S, et al. Interim guidance for the use of human papillomavirus DNA testing as an adjunct to cervical cytology for screening. Obstet Gynecol 2004;103:304 9. 8. Meijer CJ, Berkhof J, Castle PE, Hesselink AT, Franco EL, Ronco G, et al. Guidelines for human papillomavirus DNA test requirements for primary cervical cancer screening in women 30 years and older. Int J Cancer 2009;124:516 20. 9. Szarewski A, Ambroisine L, Cadman L, Austin J, Ho L, Terry G, et al. Comparison of predictors for high-grade cervical intraepithelial neoplasia in women with abnormal smears. Cancer Epidemiol Biomarkers Prev 2008;17:3033 42. 10. van Ham MA, Bakkers JM, Harbers GK, Quint WG, Massuger LF, Melchers WJ. Comparison of two commercial assays 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 2005;43:2662 7. 11. International Agency for Research on Cancer. IARC monographs on the evaluation of carcinogenic risks to humans. Human papillomaviruses 2006;90. 12. 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. 13. Wettstein FO, Stevens JG. Variable-sized free episomes of Shope papilloma virus DNA are present in all non-virus-producing neoplasms and integrated episomes are detected in some. Proc Natl Acad Sci USA 1982;79:790 4. 14. Falcinelli C, Claas E, Kleter B, Quint WG. Detection of the human papilloma virus type 16 mRNA-transcripts in cytological abnormal scrapings. J Med Virol 1992;37:93 8. 15. Hsu EM, McNicol PJ, Guijon FB, Paraskevas M. Quantification of HPV-16 E6 E7 transcription in cervical intraepithelial neoplasia by reverse transcriptase polymerase chain reaction. Int J Cancer 1993;55:397 401. 16. Sotlar K, Selinka HC, Menton M, Kandolf R, Bultmann B. Detection of human papillomavirus type 16 E6/E7 oncogene transcripts in dysplastic and nondysplastic cervical scrapes by nested RT-PCR. Gynecol Oncol 1998;69:114 21. 17. Munoz N, Bosch FX, Castellsague X, Diaz M, de SS, Hammouda D, et al. Against which human papillomavirus types shall we vaccinate and screen? The international perspective. Int J Cancer 2004;111:278 85. 18. Smith JS, Lindsay L, Hoots B, Keys J, Franceschi S, Winer R, et al. Human papillomavirus type distribution in invasive cervical cancer and high-grade cervical lesions: a meta-analysis update. Int J Cancer 2007;121:621 32. 19. Bruisten S, van Gemen B, Koppelman M, Rasch M, van Strijp D, Schukkink R, et al. Detection of HIV-1 distribution in different blood fractions by two nucleic acid amplification assays. AIDS Res Hum Retroviruses 1993;9:259 65. 20. el-Galil KH, el-Sokkary MA, Kheira SM, Salazar AM, Yates MV, Chen W, et al. Real-time nucleic acid sequence-based amplification assay for detection of hepatitis A virus. Appl Environ Microbiol 2005;71:7113 6. 21. Greijer AE, Adriaanse HM, Dekkers CA, Middeldorp JM. Multiplex realtime NASBA for monitoring expression dynamics of human cytomegalovirus encoded IE1 and pp67 RNA. J Clin Virol 2002;24:57 66. 22. Lanciotti RS, Kerst AJ. Nucleic acid sequence-based amplification assays for rapid detection of West Nile and St. Louis encephalitis viruses. J Clin Microbiol 2001;39:4506 13. 23. Loens K, Ieven M, Pattyn S, Sillekens P, Goossens H. Sensitivity of detection of rhinoviruses in spiked clinical samples by nucleic acid sequence-based amplification in the presence of an internal control. J Microbiol Methods 2006;66:73 8. 24. Boom R, Sol CJ, Salimans MM, Jansen CL, Wertheim van Dillen PM, van der Noordaa J. Rapid and simple method for purification of nucleic acids. J Clin Microbiol 1990;28:495 503. 25. Tyagi S, Kramer FR. Molecular beacons: probes that fluoresce upon hybridization. Nat Biotechnol 1996;14:303 8. 26. Tyagi S, Bratu DP, Kramer FR. Multicolor molecular beacons for allele discrimination. Nat Biotechnol 1998;16:49 53.
D. Jeantet et al. / Journal of Clinical Virology 45 (2009) S29 S37
27. Deiman B, van AP, Sillekens P. Characteristics and applications of nucleic acid sequence-based amplification (NASBA). Mol Biotechnol 2002;20:163 79. 28. Brink AA, Vervoort MB, Middeldorp JM, Meijer CJ, van den Brule AJ. Nucleic acid sequence-based amplification, a new method for analysis of spliced and unspliced Epstein Barr virus latent transcripts, and its comparison with reverse transcriptase PCR. J Clin Microbiol 1998;36:3164 9. 29. Greijer AE, Dekkers CA, Middeldorp JM. Human cytomegalovirus virions differentially incorporate viral and host cell RNA during the assembly process. J Virol 2000;74:9078 82. 30. Hayes DP, Brink AA, Vervoort MB, Middeldorp JM, Meijer CJ, van den Brule AJ.
S37
Expression of Epstein Barr virus (EBV) transcripts encoding homologues to important human proteins in diverse EBV associated diseases. Mol Pathol 1999;52:97 103. 31. Molden T, Kraus I, Skomedal H, Nordstrøm T, Karlsen F. PreTect™ HPV-Proofer: Real-time detection and typing of E6/E7 mRNA from carcinogenic human papillomaviruses. J Virol Methods 2007;142:204 12. 32. van Hamont D, van Ham MA, Bakkers JM, Massuger LF, Melchers WJ. Evaluation of the SPF10-INNO LiPA human papillomavirus (HPV) genotyping test and the Roche linear array HPV genotyping test. J Clin Microbiol 2006;44:3122 9.