Commercially available assays for multiplex detection of alpha human papillomaviruses

Abstract

Five main groups of commercial assays for the multiplex detection of alpha human papillomaviruses (HPVs) are currently available. DNA-based screening assays, which test for the presence of 13–14 HPVs without determination of HPV type, have been the standard for HPV detection in the last decade. Assays that combine testing for 14 HPVs and HPV-16 and HPV-18 genotyping are a potential future standard for HPV detection. The clinical value of HPV genotyping assays has still not been finally determined. Recently, one of the mRNA-based assays showed equal clinical sensitivity but higher clinical specificity for CIN2+/CIN3+ in comparison with the validated DNA-based assay. In situ hybridization assays are too laborious and have insufficient clinical sensitivity to be used in routine screening. Automation, price reduction and improvement of clinical specificity are the main goals for the future development of HPV assays.

Keywords::

• cervical cancer
• diagnosis
• HPV
• human papillomaviruses
• PCR

Human papillomaviruses (HPVs) are a group of remarkably diverse DNA viruses from the Papillomaviridae family, which are causally involved in the etiology of various benign and malignant neoplastic lesions of mucosal and skin epithelium. Different HPVs have been traditionally referred to as ‘types’, a type being a cloned full-length HPV genome, whose L1 nucleotide sequence is at least 10% dissimilar from that of any other papillomavirus type [1]. The Reference Center for Human Papillomaviruses at the German Cancer Research Center in Heidelberg (Germany) has for the past 25 years been instrumental in confirming the nucleotide sequence of novel HPV types and assigning appropriate HPV numbers. Currently, 120 different HPV types are officially recognized, ranging from HPV-1 to HPV-124 (HPV-46, HPV-55, HPV-64 and HPV-79, which did not yet meet the criteria as unique HPVs are now classified as subtypes) [1]. In addition to the 120 official HPVs, several new types were finally characterized during 2009–2010, the last being HPV-152 [2].

All known HPV types are currently classified by the similarity of their genome into five genera (alpha, beta, gamma, mu and nu) and 33 species [1]. Approximately 40 different HPV types from the clinically most important HPV genus, genus alpha, are known to infect the mucosal epithelium, with a subset of 10–15 HPV types being associated with lesions that can progress to cancer. These cancer-associated HPVs are traditionally designated as high-risk HPV (hr-HPV) types and are the etiological agents of virtually all cervical carcinomas. The clinically most important hr-HPVs are HPV-16 and HPV-18, found in 50–65% and 7–20% of cases of cervical cancer, respectively [3]. In addition to cervical carcinoma, hr-HPV types, the most frequent being HPV-16, play the leading etiological role in the development of anal cancer and a substantial proportion of vaginal, penile, vulvar and oropharyngeal (mainly tonsillar) cancers [4–9]. The remaining alpha-HPVs are classified as probable high-risk (phr-HPV) or low-risk (lr-HPV), depending on the frequency with which they are found in human cancers and their precursors, mainly cervical carcinoma. However, the categorization of HPV types into high-risk/probable high-risk/low-risk groups is extremely challenging, especially for weakly carcinogenic and rare HPV types, and because multiple HPV types often co-exist within the cervical epithelium [10–12]. As a consequence, in the last 15 years certain HPV types have been frequently moved from one category to another, for example, the most recent classification of the WHO International Agency for Research on Cancer slightly reduced the number of HPVs recognized as Group 1 carcinogenic agents to the following 12 HPV types: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58 and HPV-59 [10,13]. However, the design of all currently available commercial assays for the multiplex detection of alpha-HPVs still follows the traditional classification of HPVs based on meta-analysis of data from 11 case–control studies conducted in nine countries on 1918 patients with histologically confirmed squamous-cell cervical cancer and 1928 controls [14,15].

Human papillomaviruses cannot be cultured in conventional cell cultures. Other classical direct virological diagnostic techniques, such as electron microscopy and immunohistochemistry, lack the sensitivity as well as specificity for the routine detection of HPV. Serological assays for the detection of anti-HPV antibodies have only limited analytical accuracy and their possible clinical utility is currently unresolved. Consequently, all HPV tests currently in diagnostic use (in-house or commercial) rely on the detection of HPV nucleic acids in clinical specimens. Several in-house HPV DNA detection methods have been successfully used in research laboratories worldwide for more than two decades and some of them also for diagnostic purposes in countries that allow the use of nonapproved tests for clinical purposes (reviewed in [16–19]). Although many of the in-house HPV methods are still indispensable in many laboratories (including that of the authors) and some of the in-house HPV methods have been used in key randomized controlled clinical trials that have proved the value of HPV testing and prophylactic HPV vaccines, such as PCR-based GP5+/GP6+ enzyme immunoassay [20] and GP5+/GP6+ RLB Assay [21], in this article we will focus exclusively on currently available commercial assays for multiplex detection of alpha-HPV.

Evidence for this article has been provided by a detailed review of Medline/PubMed, Embase, Google, the Cochrane Library and relevant guidelines. The initial search was performed on February 1 2010 and repeated again on July 9 2010. According to the Expert Reviews publication policy, only articles published in peer-review journals in the period 2005–2010 have been considered for this article, with only a few exceptions (historical articles). In addition, a few abstracts that, in our opinion, contain breakthrough data and were presented at the 26th International Papillomavirus Conference (Montreal, Canada, July 2010) have been included in this article. Although we have tried to do our best, due to the explosive development in the field of HPV commercial assays and limited journal space, we have probably not been able to cite all important articles published in the last 5 years. For the same reason, we have focused in this article primarily on pivotal evaluations and have preferentially chosen publications evaluating a particular assay in comparison with the Hybrid Capture 2 High-Risk HPV DNA Test (Qiagen, Hilden, Germany), which is at present considered to be the assay with the most proven clinical value in HPV testing [22–26]. Evaluations of an assay’s performance described in the manufacturer’s instructions but not published in peer-reviewed journals have not been considered for this article. As summarized in Box 1, the most important currently available commercial assays for the multiplex detection of alpha-HPVs have been divided for the purpose of this article into five main groups and several subgroups (Box 1).

hr-HPV-DNA-based screening assays

High-risk HPV-DNA-based screening assays represent a group of qualitative or semi-quantitative multiplex assays in which the DNA of the targeted HPV types is detected using mixtures of probes (probe cocktails) for several HPV types with similar clinical characteristics. None of the assays from this group allow the exact determination of HPV type(s) present in a clinical specimen, but rather express the results of the tested group of HPV types as positive or negative. Until recently, such an approach has been widely accepted by the HPV community as the best way to present the results of hr-HPV testing to clinicians involved in primary screening for cervical carcinoma and management of patients with cervical intraepithelial neoplasia (CIN).

Hybrid Capture 2 HPV DNA Test

The Hybrid Capture 2 (hc2) HPV DNA Test, originally developed by Digene Corporation (Gaithersburg, MD, USA) in 1997 and currently marketed by Qiagen, has been the most important HPV diagnostic assay for the last decade and is still the most frequently used diagnostic HPV test worldwide. hc2 has been used in the majority of key randomized controlled and other clinical trials that have proved the clinical value of HPV testing in general (summarized in [22–24,26]). These trials have included a total of several hundred thousand patients worldwide. It has therefore recently been recommended that new HPV assays should show that they possess similar clinical characteristics as hc2 in the process of so-called clinical validation of the hr-HPV test, before they can be used for cervical cancer screening purposes [26].

In the hc2 assay, exfoliated cells are first treated with alkali-denaturing reagent, and the processed samples are hybridized under high-stringency conditions with two mixtures of unlabeled single-stranded full-genomic-length RNA probes, one specific for 13 HPVs: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59 and HPV-68 (high-risk probe cocktail B); and one for five HPVs: HPV-6, HPV-11, HPV-42, HPV-43 and HPV-44 (low-risk probe cocktail A). Positive specimens are detected by binding the hybridization complexes onto the surface of a microplate well coated with monoclonal antibodies specific to RNA–DNA hybrids. Immobilized hybrids are detected by the addition of an alkaline phosphatase-conjugated antibody to RNA–DNA hybrids, followed by the addition of a chemiluminescent substrate. The emission of light is measured semi-quantitatively as relative light units in a luminometer. For manual hc2 performance, several smaller laboratory instruments are required, including Digene Microplate Luminometer DML 2000. For higher throughput, the hc2 testing semi-automated pipetting and microplate handling Rapid Capture System (Qiagen) has been available for the last few years. Using the Rapid Capture System, up to 352 patient specimens can be processed by one technologist in an 8 h shift, with 3.5 h of hands-free operation.

Hybrid Capture 2 was approved by the US FDA in 2003 for the routine detection of hr-HPV infection as the third HPV diagnostic assay after ViraPap test (ViraPap; Life Technologies, Silver Spring, MD, USA) and Hybrid Capture 1 (Digene, Silver Spring, MD, USA) [24]. The US version of hc2, containing the high-risk probe cocktail only, is currently approved by the FDA for triage in cases of equivocal cytology results showing the presence of atypical squamous cells of undetermined significance (ASC-US), to determine which patients should be referred for a colposcopy and as a screening test for use in addition to cytology screening for women 30 years of age and older. Although the use of the hc2 lr-HPV cocktail is not recommended in the USA owing to lack of FDA approval, the ‘Conformité Européene’ (CE) certified version of hc2, containing both probe cocktails, is currently used in many laboratories outside the USA, mainly for individuals with clinically suspected lr-HPV infection or as a reflex test in women with ASC-US who tested negative for hr-HPVs.

Since the key randomized controlled and other clinical trials that proved the great clinical value of hc2 were published before 2005 (outside the scope of this article), only the conclusions of recent systematic reviews and meta-analyses of the performance of hc2, currently being also translated into Cochrane reviews, have been briefly summarized here (reviewed in detail in [22,23,26–29]). However, comparative evaluations of hc2 with other novel HPV assays published in the last 5 years will be described in appropriate chapters. Recent meta-analyses showed that in the triage of women with ASC-US, hc2 has improved accuracy (higher sensitivity but similar specificity) for CIN2 or worse (CIN2+) compared with cytological surveillance [22,27]. Primary screening for hr-HPV types using hc2 is more sensitive but less specific for identifying underlying CIN2+ and CIN3+ compared with cytology [22,27,29]. Women who are hc2 hr-HPV negative at screening have a lower risk of developing CIN3+ lesions in the next 3–6 years compared with women with a negative cervical cytology result [22,27]. After treatment of a CIN lesion, hc2 hr-HPV testing appears to be more sensitive but slightly less specific than follow-up cytology in the short term (less than 2 years) for predicting recurrent and residual CIN [22,27].

The main problems of the current version of hc2 are: analytical inaccuracy due to the cross-reactivity of its probe cocktails with untargeted HPV types; and lack of internal control to evaluate specimen adequacy or the presence of potentially interfering substances. In 2002, our research group showed that the hc2 high-risk probe cocktail detects at least 15 other nontargeted HPV types [30] and later that the probability of hc2-false reactivity increased with the proximity to the cut-off value [31]. These results were subsequently confirmed by other research groups. It has been recognized that the hc2 high-risk probe cocktail detects, in addition to the 13 HPV types included in the high-risk probe cocktail, at least 28 other HPV types, many of them considered to be lr-HPV types. A genotyping study of 3179 specimens of women participating in an ALTS trial (a clinical trial to evaluate management strategies for women with ASC-US or low-grade squamous intraepithelial lesions) showed that 7.8% of all hc2-positive results in this population were false-positive due to the cross-reactivity of hc2 with untargeted, noncarcinogenic HPV types [32]. Some recent studies have also shown that hc2 has an additional 5% false-positive rate due to positive hr-HPV results when no HPV DNA is present in the clinical specimen based on the use of different usually highly sensitive and broad-range PCR tests [32,33]. The relatively high false-positivity rate of the current hc2 hr-HPV cocktail has important clinical consequences as the 2006 American Society for Colposcopy and Cervical Pathology (ASCCP) guidelines state that it is ‘unacceptable’ for an HPV assay to test for nononcogenic HPV types if used in cervical cancer screening programs because false-positive results may result in unnecessary colposcopy procedures [34]. Since hc2 hr-HPV-positive samples containing only lr-HPV (false-positives) are usually weakly positive, several groups have proposed an hc2 hr-HPV cut-off adjustment or retesting of samples with initial borderline results [31,35–38]. In 2006, we recommended the introduction of a hc2 grey zone and retesting of samples collected in Digene specimen transport medium (STM) with repeatedly borderline hc2 hr-HPV results, for example, samples with repeated relative light unit per cut-off (RLU/CO) values between 1.0 and 4.0, by an alternative hr-HPV assay with high analytical specificity [31]. In a recent study that compared the clinical sensitivity and specificity of six HPV assays for the detection of high-grade CIN in a population of 953 women referred for colposcopy because of abnormal cytology, hc2 had a sensitivity of 99.6%, specificity of 28.4% and positive predictive value (PPV) of 36.1% for the detection of CIN2+ lesions [37]. The clinical specificity of hc2 increased with cut-off value. The authors suggested that hc2 might benefit from adjusting the positivity cut-off values (to 2.0 RLU/CO), which would improve the clinical specificity and PPV, while having minimal effect on clinical sensitivity [37]. On the basis of these results, the manufacturer slightly changed the positivity interpretation criteria but only for samples collected in ThinPrep PreservCyt solution (Hologic, Madison, WI, USA) and not for samples collected in Digene STM.

In order to resolve the current problems of analytical inaccuracy and to improve hc2 throughput, a next-generation diagnostic system has been developed [39]. The QIAensemble benchtop analyzers (2000 and 400) utilize significantly re-engineered hc2 chemistry (NextGen or eHC chemistry) in order to maintain the established and clinically validated sensitivity of the original hc2 (1 pg/ml or 5000 copies per assay) but with significantly improved analytical specificity. A prototype of the assay showed no cross-reactivity of the system from any of the 13 lr-HPV types tested at a concentration of 10⁷ copies per assay [39]. According to the company’s statement, the final system will be able to process more than 2000 specimens in an 8-h shift, with fully continuous loading. The possibility of automatically initiating reflex testing for HPV-16/HPV-18/HPV-45 after an hc2 hr-HPV-positive result will be incorporated in the system.

Cervista HPV HR Test

The Cervista HPV HR Test (Cervista; Hologic) is another FDA-approved signal amplification-based qualitative test for the routine detection of 14 HPVs (see list of HPV types later) [40]. The assay was originally developed by Third Wave Technologies (Madison, WI, USA), which was acquired in 2008 by Hologic, the manufacturer of the liquid-based cytology ThinPrep Pap test.

In March 2009, the FDA approved Cervista for two indications: to screen patients with ASC-US cervical cytology results to determine the need for referral to colposcopy; and to be used adjunctively with cervical cytology to screen women 30 years of age and older to assess the presence or absence of hr-HPV types [40]. Cervista is currently FDA approved for use with cervical specimens collected in ThinPrep PreservCyt solution.

In contrast to hc2, Cervista is based on the Invader chemistry, a signal amplification method for detecting specific nucleic acid sequences, which was successfully used in a multinational research collaboration to develop a freely available haplotype map of the human genome [41]. The Invader technology is comprised of two isothermal reactions: a primary reaction that occurs on the targeted DNA sequence and a secondary reaction that produces a fluorescent signal. These two reactions occur simultaneously in the same reaction vessel. The technical details of the Invader technology are available elsewhere [40,41].

Cervista detects 14 HPV types using three separate oligonucleotide probe sets: A5/A6 (HPV-51, HPV-56 and HPV-66), A7 (HPV-18, HPV-39, HPV-45, HPV-59 and HPV-68) and A9 (HPV-16, HPV-31, HPV-33, HPV-35, HPV-52 and HPV-58) and the results are reported as positive or negative for each probe set [24,40]. Oligonucleotides that bind to the human histone 2 gene (HIST2H2BE) are also present in all three oligonucleotide mixtures. HIST2H2BE serves as an internal control, producing a semi-quantitative signal from cellular DNA present in the sample [40].

In contrast to hc2, literature data on analytical and clinical validation of Cervista are relatively scarce, especially on samples originating from primary screening cohorts. To the best of our knowledge, eight evaluations have been published to date: one analytical [40] and seven evaluations comparing the performance of different prototypic versions of the current assay with other HPV assays, mainly hc2 [33,42–47]. A study that characterized the analytical performance of an investigational use only version of Cervista showed that analytical sensitivity for the targeted 14 HPV types ranged from 1250 to 7500 copies per reaction, depending on the HPV type [40]. The assay showed no cross-reactivity with DNA from seven lr-HPV types and 17 different microorganisms at up to 10⁷ copies per reaction [40]. Interference or inhibition was not observed with any of the tested substances that may be present in cervical specimens [40]. In four studies, which included a total of 1248 samples, different prototypic versions of Cervista showed 81.6–90.8% agreement with hc2 [33,41–43,45]. After testing 12,490 specimens using Cervista assay prototype Invader V2.0, Harvey et al. recently assessed the clinical performance of this prototype assay by comparison of obtained HPV results with corresponding biopsy results for a total of 1931 cases [44]. CIN1, CIN2 and CIN3 were Cervista HPV positive in 87.1, 90.3 and 93.0% of cases, respectively; non-neoplastic/reactive biopsies were hr-HPV-positive in 56.5% of cases. There was insufficient data to allow any significant conclusion with regard to hr-HPV positivity in squamous cell carcinoma. The clinical sensitivity and negative predictive value (NPV) of Invader V2.0 for CIN3+ were 97.4 and 99.1%, respectively. The clinical specificity and PPV for CIN3+ in the tested population were 10.3 and 3.7%, respectively [44]. In a recent prospective multicenter clinical study (89 sites across 22 states) on 4000 residual liquid-based cytology specimens, clinical performance of Cervista was assessed on 1347 ASC-US subjects who had complete data sets (HPV status, cytology, colposcopy and histology) [47]. The clinical sensitivity and NPV of Cervista for CIN3+ were 100 and 100%, respectively, and the clinical specificity and PPV for CIN3+ were 43 and 2.9%, respectively [47]. Although both large Cervista ASC-US triage studies published to date [44,47] showed excellent clinical sensitivity and NPV of Cervista for CIN3+, clinical specificity and PPV for CIN3+ in the study performed by Harvey et al. [44] are quite poor and worrisome. Although the authors stated that it is likely that the PPV would be much higher if the statistical analysis was limited to women over 30 years of age [44], it is not clear why such analysis has not been performed. Age-stratified analysis of clinical performance of Cervista in the study performed by Einstein et al. [47] showed clear influence of age on Cervista specificity. Although specificity for CIN2+ significantly improved with age, this caused dramatic reduction of sensitivity for CIN2+ from 100% in age group 18–30 years to 76.9% in age group 30–39 years.

Despite FDA approval, to the best of our knowledge, no data exist in peer-reviewed literature concerning the clinical sensitivity and clinical specificity of Cervista on samples that originate from a population-based screening cohort. According to the manufacturers’ package insert, 18.5% of women aged 30 years and older with normal cytology had positive results by Cervista in the premarketing approval trial. This unacceptably high detection rate has been recently heavily criticized by the influent members of the HPV community [48].

The current version of Cervista is not well suited for high-throughput testing, since the performance of the assay includes many manual steps. In November 2009, the Tecan Group (Männedorf, Switzerland) announced a global agreement with Hologic that they will supply a fully automated solution for Cervista. The system (high-throughput automation) will be based on the Tecan Freedom EVO liquid handling platform with an integrated detection device. According to a company statement, a prototype of the Cervista automated solution has already been successfully tested, but no evaluation results are currently available in peer-reviewed journals.

Amplicor HPV Test

The Amplicor HPV Test (Amplicor; Roche Molecular Systems, Branchburg, NJ, USA), launched on the European market in 2004, is a qualitative PCR-based test designed to detect the same 13 HPV types as hc2: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59 and HPV-68. Similarly to hc2 and Cervista, Amplicor expresses the results of the tested group of hr-HPV types as positive or negative. Amplicor is based on standard PCR amplification and detection of PCR products on microwell plates. Briefly, after sample preparation, a 165 bp long part of the HPV L1 gene and fragments of the human β-globin gene are co-amplified with a mixture of biotin-labeled primers. Aliquots of denaturated amplicons are added to separate wells of microwell plates coated with either hr-HPV probes or β-globin-specific oligonucleotide probes. After a washing procedure, bound hybrids are detected with a biotin avidin-horseradish peroxidase assay [49]. The original Amplicor test was set up with a cut-off value for positivity of 0.2.

Several studies, which included more that 15,000 patients in total, have compared the analytical and clinical performance of Amplicor with hc2 or with other commercial HPV tests or clinical criteria [37,49–61]. Amplicor performed well in the prediction of CIN2+ lesions in women with abnormal cervical smears, using colposcopic biopsy and liquid-based cytology as the reference standards [60]. In two studies that compared Amplicor with standardized HPV genotyping tests, there was a 97.5 and 97.8% agreement (κ values of 0.93 and 0.94) between results obtained with Amplicor and those obtained with the INNO-LiPA HPV Genotyping Test (INNO-LiPA; Innogenetics, Gent, Belgium) and Linear Array HPV Genotyping Test (Roche), respectively [50,55]. In ten studies comparing the analytical and clinical performance of Amplicor with hc2 on different patient populations (mainly selected) and using different cut-off values, Amplicor showed 83.0–92.8% agreement with hc2, with κ values ranging from 0.46–0.92 [37,49,51,53–59]. All ten comparative studies clearly showed that Amplicor is analytically more specific than hc2 for detecting targeted hr-HPV types, mainly due to hc2 cross-reactivity with nontargeted lr-HPVs. However, because of the higher analytical sensitivity, the clinical specificity of Amplicor was significantly lower in comparison with hc2 at the original 0.2 cut-off. These findings led to the idea of an Amplicor cut-off adjustment, which was first retrospectively examined on a cohort of 12,527 women aged 32–38 years who attended invitational, population-based screening and who were then followed with comprehensive registry linkages for 4 years [56]. In this evaluation, a large proportion of the general population (5.7%) was weakly positive (optical density [OD] between 0.2 and 2.2) in the Amplicor test. The absolute risk of developing CIN2+ during follow-up was 13.9% among strongly (OD>2.2) Amplicor-positive women, but only 0.14% among weakly (OD 0.2–2.2) Amplicor-positive women, which was similar to the 0.13% CIN2+ risk among Amplicor-negative women, implying that weak Amplicor positivity is not of clinical relevance [56]. In a study that compared the clinical sensitivity and specificity of six HPV assays for the detection of high-grade CIN in a population of 953 women referred for colposcopy because of abnormal cytology, Amplicor had a sensitivity of 98.9%, specificity of 21.7% and PPV of 33.5% for the detection of CIN2+ lesions [37]. The clinical specificity of Amplicor increased with the cut-off value. The authors also suggested that Amplicor might benefit from adjusting the positivity cut-off values, which would improve the clinical specificity and PPV, while having minimal effect on clinical sensitivity [37]. These findings were finally confirmed in 2009 in the most extensive comparison to date between Amplicor and hc2, on 3277 samples originating from an ALTS study [59]. The study compared the ability to detect the 2-year cumulative incidence of CIN3+ and to detect 13 targeted hr-HPV types using three different Amplicor cut-off values. At the original cut-off value of 0.2, Amplicor had significantly higher sensitivity for CIN3+ compared with hc2 (95.8 vs 92.6%) but lower specificity (38.9 vs 50.6%). The specificity of Amplicor slightly increased at a 1.5 cut-off. The PPV of hc2 was higher at all three Amplicor cut-offs, whereas referral rates were significantly lower (53.2% for hc2 vs 64.1% at the Amplicor 0.2 cut-off and 56.0% at the Amplicor 1.5 cut-off). The authors concluded that the performance of Amplicor at the 1.5 cut-off closely mimics the performance of hc2 and called for re-analysis of previous studies using higher cut-off values [59]. As a result of these studies, many clinical laboratories have decided to use in practice alternative Amplicor cut-off values for referral to colposcopy, ranging from 1.0–2.0 instead of the original cut-off of 0.2. However, it should be stressed that such an approach is not officially recommended by the manufacturer, since an alternative cut-off value is not approved by regulatory agencies. Some authors consider that the original cut-off value for positivity of 0.2 can be useful in some clinical situations, for example, prediction of therapeutic outcome after treatment of CIN lesions (test of cure), in which the high analytical sensitivity of the HPV test is beneficial [49,55,57].

The current version of Amplicor is not well suited for high-throughput testing, since the performance of the assay includes many manual steps. The recommended manual extraction of DNA using the AmpliLute liquid media extraction kit is especially time-consuming, labor-intensive and prone to potential specimen cross-contamination, particularly when large numbers of specimens are being processed [52]. It has recently been shown that DNA suitable for Amplicor testing can be alternatively generated using automated extraction systems MagNA Pure LC (Roche) using DNA-I and Total Nucleic Acid kits (Roche) or using the easyMAG instrument (bioMérieux, Marcy l’Etoile, France) [52,62]. In our experience, DNA suitable for Amplicor testing can also be generated after appropriate validation using the EZ-1 automated extraction system (Qiagen).

CareHPV Test

With support from the Bill and Melinda Gates Foundation, a careHPV Test (Qiagen), based on simplified hc2 technology, has been recently developed to detect the 13 HPV types included in the original hc2 plus HPV-66, in approximately 3 h. Such rapid, simple and affordable HPV tests, whereby results can be given to a patient within the same visit, are anticipated to have the greatest impact in countries in which cervical cancer screening programs do not exist or in countries in which substantial loss to follow-up impairs the effectiveness of cervical cancer screening programs [63]. In a cross-sectional study that assessed the clinical accuracy of the careHPV Test as a rapid screening test in two county hospitals in rural China, the careHPV Test showed sensitivity and specificity for CIN2+ of 90.0 and 84.2%, respectively, while both parameters of standard hc2 were 97.1 and 85.6%, respectively [64]. This trial showed that the careHPV Test is promising as a primary screening method for prevention of cervical cancer in low-resource countries, due to the ability to obtain accurate HPV results in a few hours, allowing the treatment of high-grade CIN during the same visit (‘screen and treat strategy’).

HPV4A ACE Screening CE & HPV/STD4 ACE Screening CE assays

The HPV4A ACE Screening CE and HPV/STD4 ACE Screening CE assays (Seegene, Seoul, Korea) utilize multiplex PCR-amplification of less than 700 bp regions of the HPV genome and an internal control template (∼1000 bp) using proprietary dual priming oligonucleotide primer technology, followed by detection of the PCR product by an auto-capillary electrophoresis system. The HPV4A ACE Screening CE assay allows simultaneous screening for 17 HPVs: HPV-6/HPV-11, HPV-26, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-53, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, HPV-73 and HPV-82, and individual genotyping for HPV-16 and HPV-18. The HPV/STD4 ACE Screening CE allows simultaneous screening for 21 HPVs using two primer cocktails: one for 16 HPVs: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59, HPV-66, HPV-67, HPV-68 and HPV-70, and the second for five HPVs: HPV-6, HPV-11, HPV-42, HPV-43 and HPV-44, as well as for Chlamydia trachomatis and Neisseria gonorrhoeae. To the best of our knowledge, no data is currently available in peer-reviewed literature concerning the analytical and clinical performance of these assays.

hr-HPV-DNA-based screening assays with concurrent or reflex HPV-16 & HPV-18 genotyping

High-risk HPV-DNA-based screening assays with individual or pooled HPV-16 and HPV-18 genotyping are a group of novel HPV assays in which qualitative detection of 13–14 HPV types is combined with concurrent or reflex HPV-16 and HPV-18 genotyping. The design of such HPV assays for routine diagnostic purposes was based on the results of two large clinical studies conducted in Brazil and the USA [65,66]. In a study conducted in Brazil and published in 2001, the authors prospectively demonstrated an increased incidence of high-grade squamous intraepithelial lesion (HSIL) in women infected with HPV-16 and HPV-18 in comparison with other hr-HPV types, during a 72-month follow-up [65]. In a long-term HPV natural history study of 20,810 women aged 30 years or older conducted in the USA, the 10-year cumulative incidence rates of CIN3+ were 17.2% among HPV-16-positive women and 13.6% among those positive for HPV-18 [66]. By contrast, only 3.0% of women positive for hc2 but negative for HPV-16 and HPV-18 and only 0.8% of women initially negative for hc2 were found to have CIN3+ during a 10-year follow-up. The authors concluded that HPV screening that distinguishes HPV-16 and HPV-18 from other hr-HPV types may identify women at greatest risk of CIN3 and may permit less aggressive management of women with other hr-HPV infections [66]. Based on these data, which clearly demonstrated the exceptionally high oncogenic potential of HPV-16 and HPV-18 compared with other hr-HPV types, the 2006 ASCCP consensus guidelines for the management of women with abnormal cervical cancer screening tests included a recommendation that women could benefit from HPV-16/HPV-18 genotyping [34]. Because no FDA-approved HPV genotyping assay was available in 2006, this recommendation was made contingent on approval of a HPV-16/HPV-18 genotyping assay by the FDA. Following FDA approval of Cervista HPV 16/18 in March 2009, the ASCCP immediately released a guideline Clinical Update regarding the use of HPV genotyping tests [201]. In this Clinical Update, ASCCP recommends that cytology negative women aged 30 years and older who are positive for any of the 13 or 14 hr-HPVs and infected with HPV-16 or HPV-18 should be referred for immediate colposcopy, whereas women infected with other hr-HPV types could be followed-up with repeat cytology and hr-HPV testing in 12 months. In contrast to the recommendation for cytology negative women, the 2006 ASCCP guidelines do not recommend the use of HPV-16/HPV-18 genotyping in women with hr-HPV-positive ASC-US.

As summarized in Box 1, four commercially available assays have the potential to be used to separate cytology negative/hr-HPV-positive women at most risk for CIN3+ (HPV-16/HPV-18 positive) from those at less risk (HPV-16/HPV-18 negative), but after appropriate clinical validation. Currently, only one of the four available assays is FDA-approved for this indication. Two of the available assays allow concurrent detection of 14 HPVs and individual typing for HPV-16 and HPV-18, and two assays are designed to be used for HPV-16 and HPV-18 reflex testing after HPV positivity is determined by corresponding HPV DNA-based screening assays (Box 1).

hr-HPV-DNA-based screening assays with concurrent individual genotyping for HPV-16 & HPV-18

RealTime High Risk HPV test

The Abbott RealTime High Risk HPV test (RealTime; Abbott Molecular, Des Plaines, IL, USA) is a real-time PCR assay based on concurrent individual genotyping for HPV-16 and HPV-18 and pooled detection of 12 other HPVs: HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59, HPV-66 and HPV-68. Amplification of human β-globin is used as an internal control. The assay was launched on the European market in January 2009. RealTime is performed on the m2000rt real-time PCR instrument (Abbott Molecular) using a modified GP5+/GP6+ primer mix consisting of three forward and two reverse primers [67]. The assay uses four channels for the detection of fluorescent probes: one for the detection of the internal control (human β-globin); a second one for the detection of HPV-16; a third one for the detection of HPV-18; and a fourth one for the detection of the remaining 12 hr-HPV types. The assay turnaround time is 6–8 h for 96 samples, depending on the method used for DNA extraction [67]. The fully automated high-throughput instrument m2000sp (Abbott Molecular) or smaller m24sp instrument (Abbott Molecular) can be used for DNA extraction or, alternatively, DNA can be prepared manually. The assay is validated for use with cervical specimens collected with ThinPrep PreservCyt solution, SurePath Preservative Fluid or Cervi-Collect Specimen Collection Kit (Abbott) [67], but in our experience, cervical specimens collected in Digene STM (Qiagen) are also appropriate [68].

Probit analysis showed that the analytical sensitivity of RealTime is between 500 and 5000 copies of HPV DNA per assay, depending on the HPV type [67]. An analytical specificity study on a panel of 41 bacteria, viruses (including several lr-HPV types) and fungi that can be found in the female anogenital tract, revealed no cross-reactivity with any of the organisms tested [67]. The analytical specificity of the RealTime test was also evaluated on 37 samples with previously determined hc2 false-positive results due to cross-reactivity with untargeted lr-HPV genotypes; all 37 samples tested clearly negative for the 14 hr-HPV included in the RealTime test [68]. Interference or PCR inhibition was not observed with any of the tested substances that may be present in cervical specimens [67].

Literature data on clinical validation of RealTime are relatively scarce. To the best of our knowledge, six studies have evaluated the clinical performance of RealTime on different patient populations to date [68–73]. In the first study, the clinical sensitivity of the RealTime test was evaluated in 593 archived cervical specimens from Amsterdam (The Netherlands) and the presence of hr-HPV was detected in 97.2% (246 out of 253) and 98.5% (335 out of 340) of CIN3 and cervical cancer specimens, respectively [69]. Of 12 hr-HPV-negative specimens evaluated further, eight out of 12 specimens had invalid β-globin results by the Linear Array HPV Genotyping Test and four out of 12 specimens contained only lr-HPVs. In specimens obtained from women 30 years of age or older with normal cytology in a screening population, the HPV positivity rate was 6.5% (49 out of 757) [69]. In the second study, RealTime demonstrated similar clinical performance (sensitivity, specificity, PPV and NPV) for the detection of CIN2+ and CIN3+ when compared with hc2, in 702 patients with abnormal cytology referred to colposcopy [70]. The clinical sensitivity for detection of CIN2+ was shown to be 97.8% with RealTime and 95.6% with hc2, and for the detection of CIN3+ 100% with RealTime and 97% with hc2 [70]. In the third study, our laboratory evaluated the clinical sensitivity of RealTime in comparison with hc2 on 95 and 267 archived routine cervical specimens collected in STM from women with histologically confirmed cervical carcinoma and CIN3 lesions, respectively [68]. RealTime showed 94.7 and 93.6% agreement with hc2 on cervical cancer and CIN3 samples, respectively. The sensitivity of RealTime and hc2 for cervical cancer was 98.8 and 95.3%, respectively, and for CIN3 lesions 96.4 and 92.5%, respectively, in samples containing at least one targeted hr-HPV [68]. In the fourth study, the performance of RealTime and hc2 was compared in 108 samples collected from Croatian women in primary screening settings. The results showed that the two assays produced comparable results but some analytical and clinical discrepancies were noted [71]. In the fifth study, performed on 143 samples collected from French women referred for colposcopy, the agreement between RealTime and hc2 was 93% (κ value 0.85) [73]. Finally, in a study that compared the clinical sensitivity and specificity of seven HPV assays for the detection of high-grade CIN in a population of 858 women referred for colposcopy because of abnormal cytology, RealTime had a sensitivity of 98.9%, specificity of 31.5% and PPV of 27.8% for the detection of CIN2+ lesions [72].

In summary, available published data have shown that RealTime has similar analytical sensitivity but better analytical specificity than hc2 for detecting targeted hr-HPVs, mainly due to hc2 cross-reactivity with nontargeted lr-HPV types. The clinical sensitivity and NPV of RealTime for CIN2+ lesions in all published studies were at least comparable with hc2, if not better. Clinical specificity and PPV for either CIN2+ or CIN3+ were similar between RealTime and hc2 in one study [70] and RealTime performed slightly better in another study [72]. To the authors’ knowledge, more clinical specificity data on samples that originate from a population-based screening cohort are currently being generated; for example, data from a Slovenian national HPV prevalence study, which has already prospectively enrolled more than 4000 unselected women 20–64 years of age attending organized routine Pap screening, will be available in October 2010.

cobas 4800 HPV Test

The cobas 4800 HPV Test (Roche Molecular Diagnostics, Pleasanton, CA, USA) is a real-time PCR assay based on concurrent individual genotyping for HPV-16 and HPV-18 and pooled detection of 12 other HPVs: HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59, HPV-66 and HPV-68. Amplification of human β-globin is used as an internal control. The cobas 4800 HPV Test was launched on the European market in November 2009.

The cobas 4800 HPV Test is performed using the cobas 4800 System, which consists of two separate instruments. The cobas x480 instrument (Roche) is an automated multichannel pipetting instrument used to extract, purify and prepare target nucleic acid, which then automatically sets up the PCR in a microwell plate. The microwell plate with the PCR-ready samples is then manually unloaded, sealed and transferred to another instrument – cobas z 480 analyzer (Roche) – a rapid thermal block cycler with four detection channels for amplification and detection using real-time PCR. The cobas 4800 System has software that integrates sample preparation, amplification and detection, and result management. The software allows the user to select two different testing options for each sample; first, screening pooled testing for all 14 targeted hr-HPVs together or, second, pooled testing plus separate individual genotyping for HPV-16 and HPV-18. Assay throughput is up to 288 HPV tests in 8 h and the first 94 HPV results can be obtained in less than 5 h. The assay is currently validated for use with cervical specimens collected in cobas PCR Cell Collection Media (Roche), ThinPrep PreservCyt solution and SurePath Preservative Fluid.

Literature data on analytical or clinical validation of the cobas 4800 HPV Test are scarce. To the best of our knowledge, there is only one paper in the peer-reviewed literature to date evaluating the prototype of the current cobas 4800 HPV Test [74]. At the time of evaluation, a validated clinical cut-off was not available and thus a conservative cut-off of 45 was used to determine positivity. The cobas 4800 HPV Test prototype was comparatively evaluated with the Linear Array HPV Genotyping Test on 531 convenience anonymized residual PreservCyt cervical samples and the overall agreement between the two assays was 94.7% [74]. However, to our knowledge, data on analytical and clinical validation of the cobas 4800 HPV Test are currently being generated; for example, VU University Medical Center (Amsterdam, The Netherlands) is finishing a comparative evaluation of the clinical performance of the cobas 4800 HPV Test and hc2 on samples originating from a population-based screening cohort using recently described evaluation criteria [19,26], and the first preliminary results of an ATHENA trial (an HPV screening study aimed at determining the effectiveness of HPV testing as part of a cervical cancer screening program, which has already recruited 46,867 women from 61 clinical sites across the USA) have been presented at the 26th Annual International Papillomavirus Conference (Montréal, Canada) in July 2010 [75,76]. Briefly, among 32,260 cytology-negative women older than 30 years, the prevalence of 14 HPV genotypes targeted by cobas 4800 HPV Test was 6.7% [75]. In the same population, absolute risk for CIN3+ among cobas 4800 HPV-negative women, those positive for 14 targeted HPV types and those positive for HPV-16/18 were 0.3, 4.0 and 9.8, respectively, and relative risk for CIN3+ of those positive for 14 targeted HPV types and those positive for HPV-16/18 in comparison with HPV-negative women was 14.53 and 35.02 [75]. In the population of women with ASC-US, the sensitivity, specificity, PPV and NPV for CIN3+ for cobas 4800 HPV Test were 93.5, 69.3, 8.4 and 99.7%, respectively, and for hc2 were 91.3, 70.0, 8.5 and 99.6%, respectively [76]. In the same population, relative risk for CIN3+ of those positive for 14 targeted HPV types and those positive for HPV-16/18 in comparison with HPV-negative women was 8.4 and 15.9 [76].

hr-HPV-DNA-based screening assays with reflex genotyping for HPV-16 & HPV-18

Cervista HPV 16/18 Test

The Cervista HPV 16/18 Test (Hologic) is a signal amplification qualitative test based on the Invader chemistry specifically designed to detect HPV-16 and HPV-18. The genotyping assay shares workflow and critical reagents with Cervista and is intended to be used as a reflex test after hr-HPV positivity is detected using Cervista. The reagents for this test are provided as two oligonucleotide mixtures, which separately detect HPV-16 and HPV-18. Oligonucleotides that bind to the human histone 2 gene are also present in the oligonucleotide mixtures and serve as an internal control, producing a signal from cellular DNA present in the sample.

The Cervista HPV 16/18 Test is currently the only FDA-approved HPV genotyping assay. The test was approved by the FDA in March 2009 for two clinical indications: first, in women 30 years of age and older the test may be used adjunctively with Cervista in combination with cervical cytology, to assess the presence or absence of specific hr-HPV types and, second, used adjunctively with Cervista in patients with ASC-US cervical cytology results, to assess the presence or absence of specific hr-HPV types. The results of this test are not intended to prevent women from proceeding to colposcopy.

To the best of our knowledge, the performance of this genotyping test has been evaluated to date in two studies published in peer-reviewed literature [42,47]. In a small study, the authors tested 100 ASC-US samples positive for hr-HPV by the Inv2 assay ASRs (prototype of Cervista), additionally using Invader HPV-16/18 ASRs (prototype of the Cervista HPV 16/18 Test) [42]. The authors noted that reflex testing of hr-HPV-positive samples for HPV-16 separated women into three groups: high, intermediate and low risk for CIN2+. Women with ASC-US who tested positive for HPV-16 had a relatively higher risk for CIN2+ on follow-up biopsy than women who either tested positive for non-HPV-16 types (p = 0.003) or were negative for hr-HPV (p < 0.001; 54, 9 and 0%, respectively) [41,42]. In a recent prospective multicenter clinical study (89 sites across 22 states) on 4000 residual liquid-based cytology specimens, clinical performance of Cervista was assessed on 1312 ASC-US subjects who had complete data sets (HPV status, cytology, colposcopy and histology) [47]. The clinical sensitivity and NPV of Cervista for CIN3+ were 77.3 and 99.0%, respectively, and the clinical specificity and PPV for CIN3+ were 67.3 and 6.6%, respectively [47].

HR-HPV 16/18/45 Probe Set Test

The HR-HPV 16/18/45 Probe Set Test (PST; Qiagen) is a signal amplification qualitative test based on standard hc2 chemistry specifically designed to detect HPV-16, HPV-18 and HPV-45. PST shares workflow and critical reagents with standard hc2 and is intended to be used as a reflex test after hr-HPV positivity is detected using standard hc2. Similarly to standard hc2, PST does not identify specific HPV types but expresses the test result as negative or positive for three targeted HPV types. To the best of our knowledge, only three relatively small pilot evaluations of PST have been published to date [77–79]. All three studies showed good efficiency of PST to detect the targeted HPV types. In the largest study, which prospectively recruited 586 women who have been followed up semi-annually, 77 CIN2+ lesions have already been confirmed and 85.7% of them were PST positive [78].

In parallel with the development of the next-generation hc2 assay, the manufacturer has recently significantly re-engineered PST in order to significantly improve analytical specificity and allow separate detection of three targeted HPV types, but maintaining the established hc2 clinically validated sensitivity (1 pg/ml or 5000 copies per assay) in the new product [80]. Although the proposed novel HPV-16/18/45 genotyping test still uses hc2 technology, the long transcribed RNA probes have been replaced with a mixture of approximately 150 unique oligoribonucleotides of 25 bases in length in each probe mix. A prototype of the novel HPV-16/18/45 genotyping assay showed no cross-reactivity from any tested nontargeted hr-HPV or lr-HPV types at a concentration of 10⁸ per assay [80]. Even homologous type HPV-18 and HPV-45 when tested with the assay did not show cross-reactivity against each other at 10⁷ copies per assay and only minimal cross-reactivity at 10⁸ copies per assay. According to the company statement, the possibility of automatically initiating reflex testing for HPV-16/HPV-18/HPV-45 after an hc2-positive result will be incorporated in the next-generation hc2 diagnostic systems QIAensemble 400 and 2000 [39].

HPV DNA-based genotyping assays

HPV DNA-based genotyping assays, which allow exact determination of several alpha-HPV types present in a clinical sample, are the largest group of currently available HPV commercial assays. In contrast to the previously described two groups of commercial HPV assays, the clinical value of HPV DNA-based genotyping assays has still not been finally determined [22,24,63,81]. Currently, HPV genotyping methods are indispensable as research tools for the study of the natural history, transmission, pathogenesis and prevention of HPV infection. However, it is highly likely that genotyping methods will also have a role in clinical management in the near future [63,82,83]. As the use of prophylactic HPV vaccines becomes more widespread, surveillance for population-level effectiveness will become an increasingly important activity, which will require the use of a HPV genotyping method [63,83]. If HPV genotyping is to be used for diagnostic applications and not just as a research tool, it will require standardized and validated methods for the specific detection and identification of a defined spectrum of HPV types [63]. When discussing type-specific performance comparisons between different HPV DNA-based genotyping tests (e.g., analytic sensitivity, specificity and reproducibility) in the present article, it is essential to appreciate that several factors additively contribute to observed performance differences which are in some cases quite significant: primer and probe differences, variation in the sampling method, the sample collection medium, the fraction of the sample tested, the DNA extraction method used, the PCR buffer and thermal cycling conditions, the probe hybridization conditions and the interpretation of results [63].

Although DNA sequencing is still considered to be the ‘gold standard’ for HPV genotyping, it is costly, time-consuming and difficult to apply in routine diagnostic settings. Thus, in daily practice, the majority of laboratories use nonsequencing based methods for HPV genotyping, which have been divided for the purpose of this article into four main groups according to the genotyping technology (Box 1).

Reverse line-blot hybridization-based HPV genotyping assays

The most frequently used HPV genotyping assays today utilize the principle of reverse line-blot hybridization. In these assays a fragment of the HPV genome is first PCR-amplified using biotinylated HPV-specific primers and the resulting amplicons are then denatured and hybridized with HPV-specific oligonucleotide probes immobilized as parallel lines on nylon or a nitrocellulose membrane strip. After hybridization, streptavidin-conjugated alkaline phosphatase or horseradish peroxidase is added, which binds to any biotinylated hybrid previously formed. Incubation with chromogenic substrates (e.g., BCIP/NBT for alkaline phosphatase) yields a colored precipitate at the probe positions where hybridization occurs. The genotyping strip is then read and interpreted visually by comparing the pattern of HPV-positive probes to the test reference guide for each of the targeted HPV types. In addition to the in-house GP5+/GP6+ RLB Genotyping Assay (RLB), which has been used in several important HPV trials [21,25], at least five commercial assays based on this technology are available today.

INNO-LiPA HPV Genotyping

INNO-LiPA HPV Genotyping is one of the most widely used HPV genotyping tests. Several versions of this assay have been developed. In all versions, amplification of a 65 bp region of the HPV L1 gene using biotinylated SPF10 primers is followed by hybridization of the resulting amplicons with 17–28 (depending on the assay version) HPV-specific oligonucleotide probes immobilized on a nitrocellulose strip. Most of the targeted HPV types are recognized by hybridization to a single type-specific probe, while a number of them hybridize to more than one probe and are recognized by a specific hybridization pattern. The prototype version of this assay, the INNO-LiPA₂₅ HPV Genotyping V1.0 (Labo Biomedical Products, Rijswijk, The Netherlands), allowed identification of 26 HPVs: HPV-6, HPV-11, HPV-16, HPV-18, HPV-31, HPV-33–35, HPV-39, HPV-40, HPV-42–45, HPV-51–54, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, HPV-70, HPV-73 and HPV-74 [84]. INNO-LiPA₂₅ does not distinguish between HPV-68 and HPV-73. Several comparative evaluations of INNO-LiPA₂₅ with other HPV assays have been published in peer-reviewed journals and, in addition, the assay has been frequently used in discordant analyses [50,82,85,86]. INNO-LiPA₂₅ has demonstrated higher analytical sensitivity than the GP5+/GP6+ based genotyping assays over a broad range of HPVs, especially when multiple HPV types were present [84,85], and has produced highly comparable genotyping results to the Linear Array [82,86]. In the largest comparative study, which included 5659 clinical samples, INNO-LiPA₂₅ demonstrated high concordance with Linear Array (κ value 0.82) [86]. Similarly, INNO-LiPA₂₅ demonstrated high concordance with Amplicor for the overall detection of the 13 assay-common hr-HPVs, with a κ value of 0.9327 [50]. The INNO-LiPA HPV Genotyping v2 and INNO-LiPA HPV Genotyping CE assays (Innogenetics), which followed the prototype test, allowed the identification of 24 and 17 individual HPV types, respectively [87,88]. The latest version of INNO-LiPA HPV Genotyping, INNO-LiPA HPV Genotyping Extra, provides a standardized amplification mix that contains an additional primer pair for the amplification of the human HLA-DPB1 gene (270 bp) to monitor PCR inhibition, sample quality and DNA extraction, as well as an anticontamination system based on uracil-N-glycosylase. INNO-LiPA Extra has a sensitivity of 20–70 viral copies per assay (estimations made for HPV-16, HPV-18, HPV-31, HPV-45 and HPV-52) and allows simultaneous identification of 28 different HPV types, including all HPV types included in INNO-LiPA₂₅ (listed above) except HPV-34 and HPV-42, and additionally HPV-26, HPV-69/HPV-71 and HPV-82 on a newly designed strip with 28 HPV-specific probes, two HPV universal probes (covering all detectable HPVs) and one human cellular target-specific probe. To the best of our knowledge, the only published evaluation of INNO-LiPA Extra, performed on 70 hc2-positive samples, showed comparable genotyping results to the digene HPV Genotyping RH Test RUO (digene RH Test; Qiagen) and a significantly higher sensitivity of INNO-LiPA Extra for the detection of multiple infections [89]. The performance of INNO-LiPA Extra requires standard laboratory equipment. Fully automated processing of the strips is possible using Auto-LiPA 48 (Innogenetics) or Autoblot (MedTec, Hillsborough, NC, USA) and automated and objective interpretation of the strips can be done with LiRAS software (Innogenetics).

Linear Array HPV Genotyping Test

The Linear Array HPV Genotyping Test (Linear Array) is one of the most commonly used HPV genotyping assays, which combines PCR amplification and reverse line-blot hybridization for the identification of 36 alpha-HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-26, HPV-31, HPV-33, HPV-35, HPV-39, HPV-40, HPV-42, HPV-44, HPV-45, HPV-51–54, HPV-56, HPV-58, HPV-59, HPV-61, HPV-62, HPV-64, HPV-66–73, HPV-81–84 and HPV-89, and one subtype (subHPV-82 or IS39). Linear Array is based on the co-amplification of a 450 bp region of the HPV L1 gene and a 268 bp region of the human β-globin gene, using biotinylated primer sets PGM09/PGMY11 [90] and PC04/GH20 [91], respectively, and subsequent genotyping of the resulting amplicons with a single-typing nylon strip coated with 37 HPV type-specific and two human β-globin-specific oligonucleotide probes. Owing to intellectual property rights for HPV-52 detection, this HPV type is identified using a cross-reactive probe that also hybridizes with HPV-33, HPV-35 and HPV-58. A sample is defined as HPV-52 positive if it reacts with the cross-reactive probe, but not with the HPV-33, HPV-35 and/or HPV-58 type-specific individual probes. Because of the problems in determining the HPV-52 status in patients infected with these three types, several HPV-52-type specific assays for confirmation/exclusion of HPV-52 in specimens positive with the Linear Array cross-reactive probe have been developed [92–95].

Several comparative evaluations of Linear Array with other HPV assays have been published in peer-reviewed journals and, in addition, the assay has been frequently used in discordant analyses [37,55,68,86,96–101]. In the largest comparative study, which included 5659 clinical samples, Linear Array demonstrated high concordance with INNO-LiPA₂₅ for the overall detection of 15 carcinogenic HPVs, with a κ value of 0.82; however, in comparisons of individual HPV types, Linear Array detected significantly more HPV-16, HPV-18, HPV-39, HPV-40, HPV-42, HPV-54, HPV-58, HPV-59, HPV-66, HPV-70 and HPV-68/HPV-73 and fewer HPV-11, HPV-31 and HPV-52 than INNO-LiPA₂₅[86]. In addition, Linear Array was able to detect more multiple HPV infections and a greater number of HPV types per multiple infection [86]. In another study, which included 1676 samples, Linear Array demonstrated a high concordance level with Amplicor for the overall detection of the 13 assay-common hr-HPVs, with a κ value of 0.944 [55]. In the study, which compared the clinical sensitivity and specificity of six different HPV assays for the detection of high-grade CIN in a population of 953 women referred to colposcopy because of abnormal cytology, Linear Array had a sensitivity of 98.2%, specificity of 32.8% and PPV of 37.7% for the detection of CIN2+ lesions [37].

The performance of the Linear Array test requires standard laboratory equipment. It has recently been shown that DNA suitable for Linear Array testing can be generated using the automated extraction systems MagNA Pure LC using DNA-I and Total Nucleic Acid kits or using an easyMAG instrument [52,62].

digene HPV Genotyping RH Test RUO

The digene RH Test is a recently developed reverse line-blot assay designed for the detection and identification of 18 HPV types: HPV-16, HPV-18, HPV-26, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51–53, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, HPV-73 and HPV-82. The assay is based on the PCR amplification of a 150 bp region of the HPV L1 gene using biotinylated GP5+ and GP6+ primers, followed by genotyping of the resulting amplicons with a single-typing nitrocellulose strip containing 18 HPV type-specific probe lines [102]. The performance of the digene RH Test requires standard laboratory equipment.

To the best of our knowledge, only two evaluations of this assay have been published in peer-reviewed journals to date [89,102]. Evaluation on 70 hc2-positive clinical samples showed comparable genotyping results to INNO-LiPA Extra, although INNO-LiPA Extra identified significantly more samples with multiple HPV types [89]. Evaluation on 493 hc2-positive samples demonstrated high concordance between digene RH Test and the in-house GP5+/GP6+ RLB Assay [21] for the overall detection of the 18 targeted HPV types (κ value 0.886) and typing of individual types (κ value 0.951; values for individual types ranging from 0.77 to 1.00) [102]. The latest version of the digene RH Test (V1.0), which was launched in November 2009, provides a standardized amplification mix containing an additional primer pair for human β-globin gene amplification, serving as an internal control for PCR inhibition and adequate sample taking and DNA purification.

EasyChip HPV Blot Kit

The EasyChip HPV Blot Kit (HPV Blot Kit; King Car, Taiwan) is a reverse dot-blot assay manufactured under class III Good Manufacturing Practices in Taiwan and designed for the identification of 37 alpha-HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-26, HPV-31–33, HPV-35, HPV-37, HPV-39, HPV-42–45, HPV-51–54, HPV-56, HPV-58, HPV-59, HPV-61, HPV-62, HPV-66–72, HPV-74 and HPV-81–85, and one HPV subtype (subHPV-44 or HPV-55). The biotinylated PCR amplicons of several HPV general primer sets or their combinations can be used for genotyping, including MY11/MY09, PGMY11/PGMY09, GP5+/GP6+, MY11/GP6+, MY11/GP6+ nested with GP5+/GP6+, and SPF1/GP6+ [103–106]. The latest version of the HPV Blot Kit (V1.0) provides, in addition to the genotyping panel, two standardized PCR amplification mixes: in the first, a modified MY11/GP6+ primer set is used to amplify a 190 bp region of the HPV L1 gene; in the second, the quality of extracted DNA is validated by amplification of a 136 bp region of the human GAPDH gene. The resulting amplicons from an individual DNA sample are genotyped together on a single-typing nylon membrane (12 mm × 15 mm) containing 39 HPV type-specific and one GAPHD-specific oligonucleotide probe spotted in duplicate. The overall analytical sensitivity of HPV Blot Kit V1.0 is 1–50 copies of HPV genome equivalent and no cross-reactivity with nontargeted HPVs has been recorded [105]. In a comparative evaluation of HPV Blot Kit and hc2 on 354 samples, 80.8% agreement (κ value 0.68) was recorded [104], and in a study on 433 samples 96.8% agreement (κ value 0.93) was recorded between the HPV Blot Kit and 20 different type-specific PCRs [106]. The HPV Blot Kit has been successfully used in two large population-based studies in Taiwan [107,108].

REBA-HPV-ID

REBA-HPV-ID (Catch by Gene, Gangwon-do, Korea) is a recently launched reverse line-blot assay for the identification of up to 25 alpha-HPVs: HPV-6, HPV-11, HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-42–45, HPV-51–53, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, HPV-70, HPV-72, HPV-81, HPV-84 and HPV-87. REBA-HPV-ID does not distinguish between HPV-59 and HPV-68 and between HPV-81 and HPV-87. REBA-HPV-ID is based on one-step nested PCR with biotinylated MY09/MY11 and GP5+/GP6+ primers and subsequent genotyping of the resulting amplicons with a single-typing strip containing 23 HPV type-specific probes and one HPV universal probe. To the best of our knowledge, no data is currently available in peer-reviewed literature concerning the analytical and clinical performance of this assay.

Microarray-based HPV genotyping assays

Similar to reverse line-blot assays, microarray-based HPV genotyping assays also employ the principle of reverse hybridization. Following PCR amplification of a fragment of a viral genome with HPV-specific primers, the resulting amplicons are denatured and hybridized with a number of HPV-specific oligonucleotide probes attached on the surface of an insoluble supporter or DNA chip (also known as microchip, biochip and gene chip). DNA chips can consist of one to several DNA microarrays, thus enabling simultaneous analysis of multiple samples in a single experiment. After hybridization, fluorescence light from the bound PCR amplicon is detected by excitation with monochromatic light. Currently, laser scanners are used for the highly sensitive fluorescence microarray readout systems. The fluorescence label of the hybridizing amplicons can be introduced during PCR (e.g., by incorporating Cy5-dUTP) and/or during the hybridization step. Some of the microarray-based HPV genotyping tests utilize the principle of chromogenic precipitation instead of fluorescence detection.

PapilloCheck HPV-Screening Test

The PapilloCheck HPV-Screening Test (PapilloCheck; Greiner Bio-One GmbH, Frickenhausen, Germany) is currently one of the two most frequently used PCR-microarray-based assays. The assay allows identification of 24 alpha-HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-40, HPV-42–45, HPV-51–53, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, HPV-70, HPV-73 and HPV-82. In this assay, a 350 bp region of the HPV E1 gene is amplified using a consensus primer set, together with an internal control template to monitor PCR inhibitions, and the human ADAT1 gene to ensure adequate sample quality and DNA extraction. The resulting PCR amplicons are hybridized on a DNA chip comprising 12 microarray wells, each containing five replicate oligonucleotide probes for each of the targeted HPVs, as well as a cellular and internal control template; the fluorescence label of the hybridizing amplicons is introduced during the PCR and the hybridization step. PapilloCheck detects from 30 to 750 viral copies per assay, depending on the targeted HPV genotype [96]. The performance of the PapilloCheck test requires standard laboratory equipment and CheckScanner instrument for DNA-chip readings and interpretation of the hybridization results.

Several comparative evaluations of PapilloCheck with other HPV assays have been published recently in peer-reviewed journals [96–98,109,110]. Three comparative evaluations of PapilloCheck with Linear Array on a total of 1200 samples demonstrated highly concordant genotyping results for the 23 HPVs targeted by both assays [96–98]. In a comparative study of PapilloCheck with hc2 and GP5+/GP6+ EIA on 878 samples, no statistically significant difference between the performance of each assay was observed when hr-HPV-positive samples were linked with cytology and histology grade [110]. In a study that included 1437 samples from women 40–60 years of age with normal cytological findings and without evidence of CIN2+ (controls) and 192 women with CIN3+ (cases), PapilloCheck demonstrated excellent concordance with clinically validated in-house GP5+/GP6+ EIA and RLB assays for the detection (overall κ value 0.93) and genotyping (κ values for individual HPV types ranging from 0.60 to 0.95) of 13 hr-HPVs and HPV-66 [109]. The study showed that, in relation to 14 HPV assay-common types, the PapilloCheck assay is clinically compatible with GP5+/GP6+ EIA for the detection of CIN3+ lesions [19,109].

Clart HPV 2 – papillomavirus clinical arrays

Clart HPV 2 – papillomavirus clinical arrays (Clart HPV 2; Genomica, Coslada, Spain) combine PCR amplification and an oligonucleotide microarray-based detection system for the identification of 35 alpha-HPVs: HPV-6, HPV-11, HPV-16, HPV-18, HPV-26, HPV-31, HPV-33, HPV-35, HPV-39, HPV-40, HPV-42–45, HPV-51–54, HPV-56, HPV-58, HPV-59, HPV-61, HPV-62, HPV-66, HPV-68, HPV-70–73, HPV-81–85 and HPV-89. The assay consists of pre-aliquoted, ready to use amplification mix tubes and allows amplification of a 450 bp region of the HPV L1 gene together with an internal control template and the human CFTR gene using three different sets of biotinylated primers. The resulting amplicons are genotyped in a single-typing 2 ml tube or eight-well strip with a bottom-fixed low-density microarray containing triplicate oligonucleotide probes for each of the targeted HPVs, as well as a cellular and internal control template. The performance of Clart HPV 2 requires standard laboratory equipment and the Clinical Arrays Processor instrument (Genomica) for fully automated microarray processing and genotyping analysis; alternatively, readings and interpretation of the hybridization results can be performed with a less expensive Clinical Arrays Reader instrument (Genomica).

A comparative evaluation of Clart HPV 2 with Linear Array and INNO-LiPA on 78 hc2-positive samples showed highly comparable genotyping results among the three methods for the 21 assay-common HPVs [99]. Fairly good concordance was recorded in a comparative evaluation on 367 samples between Clart HPV 2 and hc2 (κ value 0.62) [111]. In a study that compared the clinical sensitivity and specificity of six different HPV assays for the detection of high-grade CIN in a population of 953 women referred to colposcopy because of abnormal cytology, Clart HPV 2 had a sensitivity of 80.9%, specificity of 37.1% and PPV of 33.0% for the detection of CIN2+ lesions [37]. Clart HPV 2 has been successfully used in a large population-based study in Spain [112].

HPV GenoArray Test Kit

The HPV GenoArray Test Kit (GenoArray; HybriBio Limited, Hong Kong) allows simultaneous identification of 21 alpha-HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-42–45, HPV-51–53, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68 and HPV-81. After PCR amplification of a small fragment of the HPV L1 gene, together with an artificial internal control template, using two different sets of biotinylated primers, the resulting amplicons are genotyped by a ‘flow-through hybridization’ procedure on a DNA chip comprising 15 macroarray wells, each containing 21 HPV type-specific and internal control-specific oligonucleotide probes. The performance of the GenoArray requires standard laboratory equipment and a HybriMax instrument (HybriBio).

A comparative evaluation of GenoArray and Amplicor showed excellent agreement between the two methods (κ value 0.98) for the overall detection of the 13 assay-common hr-HPVs [61]. Similarly, GenoArray demonstrated high concordance with Linear Array for the overall detection (κ value 0.87) and genotyping of the 15 assay-common HPVs (κ values for individual HPV types ranging from 0.45 to 1.00) [100]. GenoArray has been successfully used in several large population-based studies in China [113,114].

GeneTrack HPV DNA Chip

The GeneTrack HPV DNA Chip (GeneTrack; Genomic Tree, Daejeon, South Korea) allows identification of up to 28 alpha-HPVs: HPV-6, HPV-11, HPV-16, HPV-18, HPV-31, HPV-33–35, HPV-39, HPV-40, HPV-42–45, HPV-51, HPV-52, HPV-54, HPV-56, HPV-58, HPV-59, HPV-62 and HPV-66–72. After concurrent PCR amplification of a 450 bp region of the HPV L1 gene using MY09/MY11/HMB01 primers and a portion of the human interferon-2 gene (internal control), the resulting amplicons are genotyped on a DNA chip consisting of eight microarray hybridization chambers, each containing a duplicate of 28 HPV type-specific probes, duplicate of five HPV universal probes and tenplicate of a human cellular target-specific probe. Cy5-dUTP is incorporated during amplification. The performance of the GeneTrack prototype was evaluated in HPV-positive cell lines and a small series of normal and tumor biopsies from patients with tonsillar carcinoma [115]. To the best of our knowledge, no additional data about the clinical performance of GeneTrack is available in the peer-reviewed literature.

GeneSQUARE HPV Microarray

The GeneSQUARE HPV Microarray (Kurabo Industries, Osaka, Japan) allows identification of up to 23 alpha-HPVs: HPV-6, HPV-11, HPV-16, HPV-18, HPV-30, HPV-31, HPV-33–35, HPV-39, HPV-40, HPV-42, HPV-45, HPV-51–54, HPV-56, HPV-58, HPV-59, HPV-61, HPV-66 and HPV-68. After concurrent PCR amplification of 130–423 bp fragments of the LCR, E6, E7, E1, E5, L2 or L1 genomic regions, two E. coli derived internal control templates and the human G3PDH gene, the resulting amplicons are genotyped on a DNA chip comprising 24 macroarray wells, each containing duplicate oligonucleotide probes for each of the targeted HPVs, as well as cellular and internal control templates. Cy3 dye is incorporated during amplification. The detection limit of the assay is estimated at 10²–10³ viral copies per assay, depending on the targeted HPV type. To the best of our knowledge, no data is currently available in peer-reviewed literature concerning the analytical and clinical performance of this assay.

Infiniti HPV assays

The Infiniti (AutoGenomics, Carlsbad, CA, USA) family currently consists of three genotyping assays: Infiniti HPV Genotyping Assay, Infiniti HPV-HR Quad Assay and Infiniti HPV-Quad Assay. These assays are based on the multiplex amplification of less than 300 bp regions of the HPV E1 gene and human β-globin gene, followed by automated processing of PCR amplicons on a film-based microarray BioFilmChip. The assays are performed using an Infiniti Analyser. The Infiniti HPV Genotyping Assay allows identification of 26 HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-26, HPV-30, HPV-31, HPV-33–35, HPV-39, HPV-45, HPV-51–53, HPV-56, HPV-58, HPV-59, HPV-66–70, HPV-73, HPV-82 and HPV-85. The Infiniti HPV-HR Quad Assay allows identification of 14 HPVs: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59, HPV-66 and HPV-68. The Infiniti HPV-Quad Assay allows identification of five individual HPV types (HPV-16, HPV-18, HPV-31, HPV-33 and HPV-45) and five combinations of HPV types: HPV-6/HPV-11, HPV-35/HPV-68, HPV-39/HPV-56, HPV-58/HPV-52 and HPV-59/HPV-51. The analytical sensitivity of the HPV-Quad Assay for HPV-16 and HPV-18 is 300 and 3000 viral copies per PCR, respectively [116]. The single published study on 197 specimens demonstrated an overall agreement of 83% (κ value 0.65) between the HPV-Quad Assay and hc2 for the detection of 13 hr-HPVs [116].

PANArray HPV Genotyping Chip

The PANArray HPV Genotyping Chip (PANArray; PANAGENE, Daejeon, Korea) is a microarray-based HPV genotyping panel designed for the identification of 31 HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-26, HPV-31–35, HPV-39, HPV-40, HPV-42–45, HPV-51–54, HPV-56, HPV-58, HPV-59, HPV-62, HPV-66, HPV-68–70, HPV-73, HPV-81 and HPV-83, and one subtype (subHPV-44 or HPV-55). MY09/MY11 PCR products nested with GP5+/biotinylated-GP6+ primers are used for HPV genotyping [117]. In addition to the genotyping panel, the latest version of PANArray (#PHP-1001) contains two standardized PCR amplification mixes: in the first, a 150–450 bp region of the HPV L1 gene is amplified using MY09/MY11 and GP5+/GP6+ primers; in the second, the quality of extracted DNA is validated by amplification of a 130 bp region of the human β-globin gene. The resulting amplicons from an individual DNA sample are genotyped together on a DNA chip comprising eight microarray wells, each containing 32 HPV type-specific and one human β-globin gene-specific peptide nucleic acid oligomer spotted in duplicate. Peptide nucleic acid is a DNA analogue that is very chemically and biologically stable (even at room temperature) and has a strong binding affinity to its complementary DNA sequences, which results in faster hybridization. Labeling of the hybridized biotinylated amplicons is done using streptavidin conjugated with Cy5 dye, serving as a reporter fluorophore. The detection limit of PANArray is estimated at 10 viral copies per assay [117]. An evaluation on 72 clinical samples showed excellent genotyping agreement between the PANArray and sequencing and type-specific PCR [117].

HPVDNAChip

HPVDNAChip (Biomedlab, Seoul, South Korea) allows simultaneous identification of 22 HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-31, HPV-33–35, HPV-39, HPV-40, HPV-42–45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, and HPV-69. After concurrent PCR amplification of HPV DNA using hpv1/hpv2 primers and human β-globin gene using bg1/bg2 primers, the resulting amplicons are genotyped on a DNA chip with four microarrays wells, each containing a duplicate oligonucleotide probe for each targeted HPV type, as well as the cellular template. The fluorescence label is introduced during the PCR step utilizing Cy5-dCTP or Cy5-dUTP. The performance of HPVDNAChip requires additional PCR reagents except primers, standard laboratory equipment and a commercial DNA-chip scanner. In a study that compared the performance of four HPV genotyping assays on a panel of 824 clinical samples, HPVDNAChip detected HPV in 62.6% of samples, whereas the HPV detection rate for other assays ranged from 72.1 to 91.4% [118]. In the same study, performed on a subset of 324 samples with known histological results, HPVDNAChip yielded the lowest clinical sensitivity (84.1%) and the highest specificity (44.1%) for the detection of CIN2+ [118]. HPVDNAChip was successfully used in a large population-based study in Korea [119].

GG HPVCHIP

GG HPVCHIP (GoodGene, Seoul, Korea) allows the identification of 41 individual alpha-HPV types: HPV-6, HPV-7, HPV-10, HPV-11, HPV-16, HPV-18, HPV-26, HPV-27, HPV-30–35, HPV-39, HPV-40, HPV-42, HPV-44, HPV-45, HPV-51–54, HPV-56–59, HPV-61, HPV-62, HPV-66–70, HPV-72, HPV-73, HPV-81–84 and HPV-91, and one subtype (subHPV-44 or HPV-55). The assay utilizes PCR amplification of the HPV E6–E7 (250 bp) and L1 (182 bp) genomic regions and human β-globin gene (110 bp) using Cy5-labeled primers. The resulting amplicons are genotyped on a DNA chip with eight microarray wells, each containing multiple oligonucleotide probes for targeted HPVs and human DNA [120]. The performance of the GG HPVCHIP was evaluated in comparison with PCR sequencing on 100 archival cervical cancer specimens; hr-HPV types were found in 98% of the samples tested [120].

Suspension array-based HPV genotyping assays

The suspension array that uses bead-based xMAP or Luminex technology is based on the use of polystyrene beads with a diameter of 5.6 µm, which are internally dyed with various ratios of two spectrally distinct fluorophores (red and infrared). Different bead sets with specific absorption spectra can be mixed and used for the multiplexed detection of different analytes; currently, an array of 100 bead sets (bead mix) can be generated. For HPV genotyping purposes, each bead set in the bead mix is usually coupled to a single oligonucleotide probe specific for one HPV type. HPV genotyping is done by reverse hybridization technique using biotinylated PCR amplicons. After denaturation and hybridization of target HPV sequences to the bead-bound probes, labeling of the hybridized biotinylated amplicons is done using R-phycoerythrin-labeled streptavidin, serving as a reporter fluorophore. The bead sets are then read and analyzed on a Luminex analyzer; the HPV types are discerned according to the unique bead signature, whereas the presence of specific PCR amplicons is determined by R-phycoerythrin fluorescence. The Luminex readouts are expressed as the median fluorescent intensity of the reporter fluorescence for each HPV type. Positivity for the relevant types is calculated from the median fluorescent intensity over a defined threshold level, and can provide a semi-quantitative numerical output. Several in-house genotyping protocols based on xMAP technology have been developed in the last 5 years [121–127]; some of them are considered to be the most sensitive HPV detection methods currently available [123]. In addition, at least two commercial assays based on this technology are available at present.

Multiplex HPV Genotyping Kit

The Multiplex HPV Genotyping Kit V1.0 (Progen/Multimetrix, Heidelberg, Germany) is a research use only assay developed according to the Schmitt et al. protocol [121]. The assay allows detection and identification of 24 HPV types: HPV-6, HPV-11, HPV-16, HPV-18, HPV-26, HPV-31, HPV-33, HPV-35, HPV-39, HPV-42–45, HPV-51–53, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, HPV-70, HPV-73 and HPV-82. In the assay, the 150 bp region of the HPV L1 gene is PCR amplified using biotinylated GP5+/GP6+ primers concurrently with the human β-globin gene, serving as an internal control. The resulting amplicons are then genotyped using the Luminex IS System (Luminex Corporation, Austin, TX, USA) and 26 distinct fluorescence-labeled polystyrene bead populations: 24 HPV type-specific, one β-globin-specific and one hybridization control-specific.

digene HPV Genotyping LQ Test RUO

The digene HPV Genotyping LQ Test RUO (digene LQ Test; Qiagen) allows detection and identification of 18 HPVs: HPV-16, HPV-18, HPV-26, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51–53, HPV-56, HPV-58, HPV-59, HPV-66, HPV-68, HPV-73 and HPV-82. The assay is based on the amplification of a 150 bp region of the HPV L1 gene using biotinylated GP5+/GP6+ primers and subsequent genotyping of the resulting amplicons with an array of HPV type-specific bead populations using the LiquiChip 200 Workstation (Qiagen) or Luminex IS System [128]. The single published comparative evaluation of the digene LQ Test, performed on 434 hc2-positive samples, demonstrated high agreement with in-house GP5+/GP6+ RLB for overall detection (κ value 0.884) and type-specific identification (κ value 0.958) of the 18 targeted HPV types [128]. The latest version of the digene LQ Test (V1.0), launched in November 2009, provides a standardized amplification mix containing an additional primer pair for human β-globin gene amplification and detects from 5 to 10,000 viral copies per assay, depending on the targeted HPV type.

Gel electrophoresis-based HPV genotyping assays

Gel electrophoresis-based HPV genotyping assays utilize general primer-mediated PCR, type-specific or group-specific PCR, to screen for a broad spectrum of HPVs, followed by agarose gel electrophoresis. Restriction fragment length polymorphism (RFLP) is consequently applied to identify HPV type-specific restriction patterns. In addition to several in-house genotyping protocols based on RFLP, one commercial assay based on this technology is currently available.

BIOTYPAP Kit

BIOTYPAP Kit (Biotools, Nave, Spain) allows simultaneous detection and identification of 31 HPV types: HPV-6, HPV-11, HPV-13, HPV-16, HPV-18, HPV-30–35, HPV-39, HPV-40, HPV-42–44, HPV-51–54, HPV-56–59, HPV-61, HPV-62, HPV-64 and HPV-66–69, and one subtype (subHPV-44 or HPV-55). The assay utilizes multiplex PCR amplification with two pairs of primers: one amplifies the 450 bp region of the HPV L1 or L2 genomic region of all tested HPVs and the second amplifies the 250 bp region of the HPV E6 or E7 genomic region of hr-HPVs only. In order to determine the HPV types present in a sample, the resulting amplicons are digested using five different restriction endonucleases and analyzed by agarose gel electrophoresis. The BIOTYPAP Kit has been recently used for genotyping of HPV in four children with recurrent respiratory papillomatosis in whom HPVs other than HPV-6 or HPV-11 were found [129]. To the best of our knowledge, no other data is currently available in peer-reviewed literature concerning the analytical and clinical performance of this assay.

hr-HPV E6/E7 mRNA-based screening assays

Several recent studies have clearly shown that testing for HPV mRNA instead of HPV DNA can be clinically useful (reviewed in [63,130,131]). The most relevant transcripts for diagnostic purposes are those encoding viral oncoproteins E6 and E7. The detection of viral mRNA can be done by reverse transcriptase PCR or by nucleic acid sequence-based amplification (NASBA). For the latter, three commercially available assays that detect E6/E7 transcripts are currently available.

PreTect HPV-Proofer

PreTect HPV-Proofer (HPV-Proofer; NorChip, Klokkarstua, Norway) is an assay based on NASBA technology, which allows qualitative determination of E6/E7 mRNA transcripts of the five most frequently identified hr-HPV types in cervical cancer worldwide: HPV-16, HPV-18, HPV-31, HPV-33 and HPV-45. The assay utilizes multiple primer sets to amplify hr-HPV mRNAs together with mRNA of human U1 small nuclear ribonucleoprotein specific protein A (U1A) to monitor the sample quality, mRNA extraction and integrity, and amplification inhibition. Real-time detection of amplicons is performed using six target-specific molecular beacons; two fluorescent dyes allow the detection and identification of cellular and HPV target RNAs in three separate duplex amplification reactions: U1A and HPV-16, HPV-18 and HPV-31, and HPV-33 and HPV-45. In addition to the U1A control, five positive controls, consisting of artificial E6/E7-target sequences specific for each targeted HPV type, are included in each assay run to monitor the functionality of the assay [132]. The performance of HPV-Proofer requires standard laboratory equipment, the Lambda FL600 fluorescence reader/NASBA platform (Bio-Tek, Winooski, VT, USA) and PreTect Analysis software (NorChip) for analysis of the experimental data.

Several studies (mainly cross-sectional) have comparatively evaluated HPV-Proofer with hc2 or different in-house or commercial PCR-based HPV DNA assays [37,130–138]. All studies have shown that HPV-Proofer has a lower clinical sensitivity for the detection of CIN2+ lesions than DNA-based assays, but a significantly higher clinical specificity. It is believed that the lower sensitivity of HPV-Proofer is mainly due to the detection of only five hr-HPVs, rather than the 13–14 types detected by DNA-based screening assays [139]. In the largest cross-sectional study to have compared HPV-Proofer and a DNA-based assay (GP5+/6+ PCR) on a total of 4136 women, 2.4 and 9.3% of cytologically normal women over 30 years were mRNA and DNA HPV positive, respectively [136]. The authors found that the detection rate of HPV DNA was significantly higher than the detection rate of HPV mRNA in all grades of cytology except for HSIL; of the HSIL cases that were confirmed to be CIN2+, 13 of 14 tested mRNA positive compared with 14 of 14 that were DNA positive [136]. In a study that compared the clinical sensitivity and specificity of six different HPV assays for the detection of high-grade CIN in a population of 953 women referred to colposcopy because of abnormal cytology, HPV-Proofer had a sensitivity of 73.6%, specificity of 73.1% and PPV of 52.0% for the detection of CIN2+ lesions [37].

NucliSENS EasyQ HPV

The NucliSENS EasyQ HPV V1 assay (NucliSENS; bioMérieux) was launched in 2007 and is based on the original HPV-Proofer assay, except for the proprietary hardware platform and the software for NASBA measurements and data analysis [140]. The latest NucliSENS version has been improved in several ways: positive controls have been changed from ssDNA to RNA, positive controls are freeze-dried, the concentrations of the primers and the beacons for HPV-16 and the U1A internal control have been re-adjusted, the extraction protocol on NucliSENS easyMAG has been optimized, assay protocols have been updated and new software (NucliSENtral HPV software V1.1) has been introduced [140]. The performance of the assay requires standard laboratory equipment and the NucliSENS amplification/detection platform; the recommended RNA extraction systems are NucliSENS miniMAG or NucliSENS easyMAG. Analytical evaluation of NucliSENS has shown that the limit of detection for the targeted HPV types ranged from 2.3 × 10² to 3.0 × 10⁴ copies/ml on a background of 5 × 10³ HPV-negative HS67 cells [140]. No cross-reactivity for other viral, bacterial or fungal agents known to be potentially present in cervical fluids was detected [140]. A recent analytical study using several HPV plasmids and HPV-infected cell lines showed that NucliSENS not only amplified HPV mRNA but also HPV dsDNA [141]. Based on their results, the authors postulated that the differences in the diagnostic performance between NASBA-based and DNA-based HPV assays do not seem attributable to the more specific amplification of viral mRNA, but to the limited type range and lower analytical sensitivity for HPV DNA [141].

A comparative analysis of NucliSENS and HPV-Proofer on 420 PreservCyt cervical samples collected from women participating in a routine cervical cancer screening program, showed an overall concordance of greater than 90% for the detection of targeted hr-HPVs [140]. In a comparative study of three HPV assays performed on 140 women referred for colposcopy and histology, NucliSENS showed significantly lower clinical sensitivity for the detection of CIN2+ or HSIL in comparison with hc2 and Linear Array (76 vs 95% and 92%, respectively), but had a significantly higher clinical specificity (63 vs 49% and 45%, respectively) [101].

APTIMA HPV Assay

APTIMA HPV Assay (APTIMA; Gen-Probe, San Diego, CA, USA) is a transcription-mediated amplification-based assay, which allows the detection of E6/E7 mRNA transcripts of 14 HPV types: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59, HPV-66 and HPV-68. The assay generates a qualitative result for the presence/absence of 14 targeted HPVs and does not allow the exact determination of HPV type(s) present in a clinical specimen [139]. APTIMA is a single tube test and is based on three main steps:

• • Capture of the target E6/E7 mRNAs using HPV-specific capture oligomers linked to magnetic microparticles;

• • Target E7 mRNA amplification using transcription-mediated amplification;

• • Detection of the resulting amplicons using chemiluminescent-labeled probes in hybridization protection assay.

An internal control transcript is added to each reaction to verify the performance of each step of the assay. APTIMA can currently be performed on the semi-automated DTS system or on the fully automated TIGRIS DTS system (Gen-Probe) [139]. The manufacturer recently announced that the company is developing another fully automated instrument, tentatively called the Panther system. The Panther system is designed to be used in low- and mid-volume laboratories, will have random access, STAT test prioritization and is intended to run several assays in a fully automated from sample to result fashion.

Analytical evaluation estimated that there was a 95% detection limit of APTIMA at 38–488 mRNA copies/reaction for the DTS system and at 17–275 HPV mRNA copies/reaction for the TIGRIS DTS system [139]. APTIMA did not cross-react with any tested lr-HPVs, nor with normal flora and opportunistic organisms that may be found in cervical samples [139]. Similarly as previously described for NucliSENS [141], a recent analytical study using samples spiked with various quantities of different plasmid vectors or purified SiHa cells showed that APTIMA detected not only HPV mRNA but also HPV dsDNA [142]. However, APTIMA had substantially higher sensitivity (2–4 log₁₀) for HPV-16 mRNA than for HPV-16 DNA and substantially lower (3 log₁₀) analytical sensitivity for HPV-16 DNA compared with hc2 [142].

In the first cross-sectional comparative evaluation of the APTIMA prototype, with a DNA line-blot assay using 531 cytology samples (with histology confirmation), two different cut-off values for APTIMA positivity were tested in relation to clinical findings [143]. In a study that compared the clinical sensitivity and specificity of six different HPV assays for the detection of high-grade CIN in a population of 953 women referred to colposcopy because of abnormal cytology, APTIMA performed well and had a sensitivity of 95.2%, specificity of 42.2% and PPV of 39.9% for the detection of CIN2+ lesions [37]. APTIMA yielded similar sensitivity for CIN2+ compared with hc2, Amplicor and Linear Array (95.2 vs 99.6%, 98.9% and 98.2%, respectively), but a significantly higher specificity (42.2 vs 28.4%, 21.7% and 32.8%, respectively) [37]. In a comparative evaluation of APTIMA and hc2 on PreservCyt specimens collected from 800 women referred to colposcopy, APTIMA showed comparable sensitivity to hc2 for the detection of CIN2+ (91 vs 95%), as well as CIN3+ (98 vs 99%), but had higher clinical specificity (>55 vs 47% for CIN2+; 53 vs 44% for CIN3+) [144]. In a recent study that included 1008 women attending two colposcopy clinics in the UK, in an interim analysis APTIMA showed equal sensitivity and comparable specificity to hc2 for the detection of CIN2+ (sensitivity 93.8 vs 94.2%, specificity 49.9 vs 44.7%) [145]. In a large French multicenter comparison between APTIMA and hc2, which was performed on 4429 samples obtained from 20–65-year old women undergoing routine screening, in an interim analysis hc2 was more sensitive for CIN2+ than APTIMA (96.7 vs 92.0%), but APTIMA showed better specificity (91.8 vs 86.4%) [146]. At the CIN3+ end point, both assays were equally sensitive (95%) but APTIMA was more specific than hc2 (90.3 vs 84.9%). Both assays performed better among women over the age of 30 years, but the difference in the specificity in favor of APTIMA remained unchanged [146].

In situ hybridization

In situ hybridization (ISH) is the only molecular method allowing reliable detection and identification of HPVs in topographical relation to their pathological lesions. Unlike other molecular methods, in ISH the whole HPV detection procedure occurs within the nuclei of infected cells and not on solid supports or in solutions. The result of the hybridization reaction is evaluated microscopically and the appearance of an appropriate precipitate within the nuclei of epithelial cells is indicative of the presence of HPVs in the specimen being tested. In addition, the physical state of the virus can be evaluated by the presence of punctuate signals for integrated virus and diffuse signals for episomal virus. Although commercially available HPV assays based on ISH have been validated technically, they are insufficiently clinically validated. In addition, current ISH-based assays are considered by many experts in the field to be too laborious and to have insufficient clinical sensitivity to be used in routine screening.

INFORM HPV

INFORM HPV (Ventana, Tucson, AZ, USA) represents several generations of commercially available ISH probe cocktails that can be used for HPV DNA testing in both cytological and histological specimens. INFORM HPV III is the latest generation of probe cocktails that utilizes a stacked antibody approach to enhance its sensitivity. Briefly, the primary antibody is directed against the dinitro-phenol hapten attached to the probes. Signal amplifications are generated through antibody stacks consisting of a secondary antibody and a biotinylated tertiary antibody. Finally, a streptavidin-alkaline phosphatase conjugate is added to generate color precipitates. Several probe cocktails are available for detection of a range of lr-HPV and hr-HPV, for example, the INFORM HPV III Family 16 Probe B and C contains a cocktail of dinitro-phenol-labeled HPV probes that target 12 HPVs: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58 and HPV-66. The INFORM HPV probe cocktails are usually used in conjunction with ISH iVIEW™ Blue Plus Detection Kit (Ventana) on formalin-fixed, paraffin-embedded tissue sections (probe cocktails B) or ThinPrep cervical cytological specimens (probe cocktails C) on the BenchMark Series (Ventana) automated slide stainers. The detection level of INFORM HPV III appears to be 10–50 viral copies per cell when using formalin-fixed tissue samples.

In a study performed on 137 CIN1–CIN3 and cervical cancer tissue specimens, INFORM HPV III showed similar sensitivity in comparison with GP5+/6+ PCR in detecting targeted HPVs across all categories of CIN lesions, but was significantly less sensitive in detecting HPV in cervical cancer specimens [147]. In a small study that compared the performance of INFORM HPV III with Cervista, hc2 and a commercial PCR (Access Genetics, Minneapolis, MI, USA) on PreservCyt specimens, the four assays appeared to be comparable with one another [46].

GenPoint HPV Biotinylated DNA Probe

GenPoint HPV Biotinylated DNA Probe (GenPoint) is the latest generation of Dako (Glostrup, Denmark) commercially available ISH probe cocktails that can be used for HPV testing in both cytological and histological specimens. This ISH probe cocktail targets 13 HPVs: HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59 and HPV-68, and contains viral genomic DNA in the form of double-stranded fragments of 500 bp or less (biotinylated and unlabeled) and multiple biotinylated oligonucleotides from 25 to 40 bases in length. It is optimized for amplified ISH using the GenPoint Tyramide Signal Amplification System (Dako). The GenPoint ISH assay can be performed automatically on Dako Hybridizer and Autostainer Plus. In addition to GenPoint, several other probe cocktails are available from the manufacturer for detection of a range of lr-HPV and hr-HPV. In a comparative evaluation of three chromogenic ISH assays, INFORM II, INFORM III and GenPoint, in conjunction with p16 immunohistochemistry and HPV PCR on 58 tissue specimens, the sensitivity of GenPoint was similar to INFORM II but lower than that of INFORM III. All three ISH assays demonstrated 100% specificity [148].

ZytoFast HPV Probes

ZytoFast HPV Probes (ZytoVision, Bremerhaven, Germany) are a range of ISH probe cocktails for the detection of a wide range of lr- and hr-HPVs, or their combinations, in either cytological or histological specimens. ZytoFast HPV Probes are directed against DNA and mRNA sequences of the HPV E6, E7 and L1 proteins and are labeled with either digoxigenin or biotin. When bonded to the target, they are detected directly using antidigoxigenin or antibiotin enzyme-conjugated antibody or indirectly using unconjugated primary antibody and secondary enzyme-conjugated antibody. In a recent study, the ZytoFast HPV Probes have been used successfully for simultaneous immunohistochemical detection of p16^INK4A protein and HPV on the same cervical carcinoma sections [149]. To the best of our knowledge, no other data is currently available in peer-reviewed literature concerning the analytical and clinical performance of these probe cocktails.

HPV OncoTect Test Kit

The HPV OncoTect Test Kit (OncoTect; IncellDx, El Camino Real Menlo Park, CA, USA; Invirion Diagnostics, Oak Brook, IL, USA) is a cell-based assay that utilizes FISH for the detection of hr-HPV E6/E7 mRNA transcripts in intact human cells, and has been designed to be used in conjunction with other HPV assays as an indicator of disease activity. OncoTect provides quantitative information on two levels: the quantity of hr-HPV E6/E7 mRNA present inside each tested cell; and the percentage of tested cells that overexpress E6/E7 mRNAs. The assay is based on five main steps: cell preparation, cell fixation and permeabilization, hybridization with IncellDx proprietary oligonucleotide probes, post-hybridization washes, and detection of the tagged (HPV positive) cells using flow or image cytometry. OncoTect can be used in any single cell suspension appropriate for flow cytometry or image analysis; types of specimen may include liquid-based cytology specimens, cell lines, dissociated tissues and body fluids. Using different colored dyes, the OncoTect assay can be run on all commercially available flow cytometers. In a single peer-reviewed publication, OncoTect demonstrated 83.3% sensitivity and 91.3% specificity for HSIL compared with the Pap test in 231 liquid-based cytology samples [150].

Expert commentary

In view of the fact that persistent infection with hr-HPVs is a necessary etiological factor in the development of cervical carcinoma, HPV testing has become an important part of cervical carcinoma screening and detection algorithms in several countries (for an excellent historical review see [24]). The four main clinical applications of HPV testing using clinically validated molecular assays at present are: triage of women with equivocal cytology results showing the presence of ASC-US, in order to determine which patients should be referred to colposcopy; follow-up of women with abnormal screening cytology results who are negative at initial colposcopy/biopsy; prediction of the therapeutic outcome after treatment of CIN2+; and primary screening of women aged 30 years and more in combination with Pap smear to detect cervical cancer precursors [22,24,151].

As reviewed in this article, several commercial assays for multiplex detection of alpha-HPVs are currently available (Box 1). However, in our opinion, the majority of HPV assays currently on the market are not very useful for the aforementioned clinical applications and especially not for primary screening. Thus, in line with other recent opinions [26,63,152], we consider that commercially available and in-house HPV assays that have not been fully clinically validated and shown to be reliable, reproducible and accurate should not be used in the clinical management of women with CIN nor for primary screening. Unfortunately, several such clinically unvalidated HPV assays are used worldwide in daily practice, due to lack of regulation. In order to facilitate the acceptance of novel HPV assays, mainly for cervical screening purposes, several recommendations have recently been published [26,63,152]. Meijer et al. recently proposed that before a new HPV assay can be used for cervical screening purposes, it should demonstrate at least similar if not better clinical characteristics (sensitivity, specificity, reproducibility, and so on) for the detection of CIN2+ as hc2 [26]. Other experts believe that large-scale clinical trials, with an assessment of prospective disease outcomes, are required to validate any proposed HPV screening test and that cross-sectional comparisons of new HPV assay to hc2 using several hundred specimens are not an acceptable form of assay validation [63]. Stoler et al. recently proposed that any novel HPV assay aiming to be used for cervical screening should have a clinical sensitivity of 92% ± 3% for CIN3+ to render a high NPV or the capacity to predict the future detection of a CIN3+ outcome that might occur during a recommended screening interval [152]. The HPV assay aiming to be used for cervical screening should also have a clinical specificity of at least 85% to achieve an adequate PPV for CIN3 [152]. The common idea behind all proposed recommendations is that a clinically useful HPV assay should achieve an optimal balance between clinical sensitivity and clinical specificity for detection of CIN2+/CIN3+ in order to minimize redundant or excessive follow-up procedures for hr-HPV-positive women with transient hr-HPV infections and/or without cervical lesions [26,63,152]. Thus, as an example, a HPV assay with very high analytical sensitivity can yield a large number of clinically insignificant positive results, which will cause unnecessary clinical follow-up, unnecessary diagnostic procedures and unnecessary treatment of healthy women [48,81].

Five-year view

In the last decade, hr-HPV-DNA-based screening assays that test for the presence of 13–14 hr-HPVs without an exact determination of HPV type have been widely accepted as the best hr-HPV testing strategy for clinicians involved in primary screening for cervical carcinoma and management of patients with CIN lesions [22–24,26,27,63,151,153,154]. Testing negative for hr-HPVs with these assays provides more reassurance against cervical precancerous lesions and cancer than cervical cytology, and therefore safely permits longer intervals between screens [155]. However, among hr-HPV-positive women, only a few will have a concurrent clinically-relevant disease, creating a clinical dilemma on how to identify the subset of women that require further immediate clinical attention, such as colposcopy [155,156]. Several approaches have been suggested to improve the PPV for disease among hr-HPV-positive women. One approach is the use of limited HPV genotyping for identification of the two most risky HPV types: HPV-16 and HPV-18. A new generation of commercially available assays that test for the presence of 14 hr-HPVs and have, in addition, the potential to separate cytology negative/hr-HPV-positive women at the most risk for CIN3+ (HPV-16/HPV-18 positive) from those at lesser risk (HPV-16/HPV-18 negative), could represent a standard of HPV molecular diagnostics in the immediate future, provided current clinical validations show satisfactory results. Another approach to improving the PPV for disease among hr-HPV-positive women is to test for HPV mRNA instead of HPV DNA. If the preliminary results of some evaluations of mRNA assays, which have shown equal clinical sensitivity but significantly higher clinical specificity for CIN2+/CIN3+ in comparison to clinically validated DNA-based assays, are confirmed in further multicenter trials, mRNA-based HPV assays will also play an important role in the next few years.

It is highly likely that within the next 5 years, five to eight currently available commercial HPV assays will be fully clinically validated and that laboratories will face a huge dilemma in relation to which assay to choose for routine HPV testing. In addition, several problems have to be resolved in the next 5 years and the text below can also be read as a wish list. First of all, with increasing concern over the escalating cost of medicine, we urgently need more competitively priced HPV assays, especially for resource-poor countries. We hope that the predicted availability of five to eight clinically validated commercial HPV assays on the market and their full automation in the next few years will lead to a significant price reduction due to competition and a shortening of hands-on time. A final consensus is urgently needed on how many and which hr-HPV types should be included in an HPV screening assay in order to achieve the best balance between clinical sensitivity and clinical specificity. When designing a HPV screening assay, ultimate clinical sensitivity for the detection of precancerous lesions by inclusion of HPV types that are rarely associated with cancer must be carefully weighed against the potentially dramatic loss of specificity when a particular HPV type (e.g., HPV-53 and HPV-66) is frequent in low-grade disease or in women without disease [11]. There is also an urgent need for the development of international standards for HPV types other that HPV-16 and HPV-18 [157] and the development of international quality control panels that can be used in analytical and clinical evaluations of current and future HPV assays. We need urgent evaluation of the performance of commercial HPV assays on alternative clinical specimens, for example, on self-collected cervicovaginal lavage specimens to be used for women who do not attend cervical screening [158]. The first HPV-based screening assays designed for use in resource-poor areas are being developed and show encouraging preliminary results [64]. However, to achieve real success, closer cooperation and collaboration among governments, external donors, healthcare providers, nongovernmental organizations, social workers and the community is needed [63,159].

An additional increase in clinical specificity and PPV are the main goal for further improvement of HPV-based cervical screening assays. Research on the mechanisms and corresponding novel biomarkers associated with neoplastic progression remains an important priority, since results from these studies may identify novel HPV-based or HPV-non-based biomarkers of CIN3/cervical cancer that can be adapted as screening and triage targets [63,160]. A detailed review of some of the most promising new screening biomarkers and technologies (p16, p16/Ki67, methylation markers, chromosomal abnormalities, HPV viral load, HPV integration, HPV genetic variants, and so on) has recently been published [63]. However, the main problem with novel biomarker candidates will be how to evaluate their clinical performance, due to the imperfection of CIN3, and especially CIN2, as surrogate end points for cervical cancer. As recently indicated by Castle [161], ironically, as biomarkers become more specific for cervical carcinogenesis, assays for their detection will appear to underperform compared with hr-HPV DNA assays, because there will be an increasing number of apparent (‘false’) negative results associated with CIN2 and perhaps even with CIN3, since not all CIN3 lesions are truly precancerous (i.e., having the potential to invade) [162]. Finally, because of the high efficacy of prophylactic HPV vaccines, the total burden of precancerous lesions will significantly decrease in vaccinated cohorts. As a consequence, it is predicted that current cytology-based screening programs will be less and less efficient in the next 10–15 years and will potentially increase the rates of overdiagnosis and overtreatment [22,24,163]. Novel screening programs, for example, HPV-based screening with cytology or other novel markers as a triage test(s), have great potential to redefine cervical screening, including a later age at starting and less frequent screening visits, without losing efficiency or compromising safety [163].

Box 1. Most important currently available commercial assays for the multiplex detection of alpha human papillomaviruses.

hr-HPV-DNA-based screening assays

• – Hybrid Capture 2 HPV DNA Test^†

  – Cervista HPV HR Test^†

  – Amplicor HPV Test

  – CareHPV Test

  – HPV4A ACE Screening CE and HPV/STD4 ACE Screening CE assays

hr-HPV-DNA-based screening assays with concurrent or reflex HPV-16 and HPV-18 genotyping

• • hr-HPV-DNA-based screening assays with concurrent individual genotyping for HPV-16 and HPV-18

  – RealTime High Risk HPV test

  – cobas 4800 HPV Test

• • hr-HPV-DNA-based screening assays with reflex genotyping for HPV-16 and HPV-18

– Cervista HPV 16/18 Test^†

– HR-HPV 16/18/45 Probe Set Test

HPV DNA-based genotyping assays

• • Reverse line-blot hybridization-based HPV genotyping assays

  – INNO-LiPA HPV Genotyping

  – Linear Array HPV Genotyping Test

  – digene HPV Genotyping RH Test RUO

  – EasyChip HPV Blot Kit

  – REBA-HPV-ID

• • Microarray-based HPV genotyping assays

  – PapilloCheck HPV-Screening Test

  – Clart HPV 2 – papillomavirus clinical arrays

  – HPV GenoArray Test Kit

  – GeneTrack HPV DNA Chip

  – GeneSQUARE HPV Microarray

  – Infiniti HPV Assays

  – PANArray HPV Genotyping Chip

  – HPVDNAChip

  – GG HPVCHIP

• • Suspension array (xMAP, Luminex) based HPV genotyping assays

  – Multiplex HPV Genotyping Kit

  – digene HPV Genotyping LQ Test RUO

• • Gel electrophoresis-based HPV genotyping assays

  – BIOTYPAP Kit

hr-HPV E6/E7 mRNA-based screening assays

• – PreTect HPV-Proofer

  – NucliSENS EasyQ HPV

  – APTIMA HPV Assay

In situ hybridization

• – INFORM HPV

  – GenPoint HPV Biotinylated DNA Probe

  – ZytoFast HPV Probes

  – HPV OncoTect Test Kit

Key issues

• • High-risk human papillomaviruses (HPVs) are necessary although not sufficient etiological agents of cervical carcinoma and their DNA can be detected in more than 99% of cervical carcinomas.

• • Testing for high-risk HPVs has four main clinical applications: triage of women with atypical squamous cells of undetermined significance; follow-up of women with abnormal screening cytology results who are negative at initial colposcopy/biopsy; prediction of the outcome after treatment of CIN2+; and primary screening of women aged 30 years and more in combination with Pap smear.

• • Several commercial assays for the multiplex detection of alpha-HPVs are currently available and can be divided into five main groups and several subgroups.

• • DNA-based screening assays that test for the presence of 13–14 HPVs without an exact determination of HPV type have been the standard for HPV detection in the last decade.

• • A new generation of assays, in which testing for the presence of HPV DNA of 14 HPVs is combined with concurrent or reflex HPV-16 and HPV-18 genotyping, could potentially represent a novel standard for HPV detection in the immediate future if current clinical validations show satisfactory results.

• • The clinical value of HPV DNA-based genotyping assays, which allow exact determination of several alpha-HPV types and are the largest group of currently available HPV commercial assays, has still not been determined.

• • Preliminary evaluations of one of the mRNA high-risk HPV assays have shown equal clinical sensitivity but significantly higher clinical specificity for CIN2+/CIN3+ in comparison with clinically validated DNA-based assays. If this is confirmed, mRNA-based assays will also play an important role in the next few years.

• • Further automation, significant price reduction and improvement of clinical specificity and positive predictive value are the main goals for the future development of HPV assays.

Financial & competing interests disclosure

The institution at which Mario Poljak and Boštjan J Kocjan have undertaken evaluations of various HPV tests has received research funding and/or free-of-charge kits from Abbott, Digene, Innogenetics, Roche and Seegene to support comparative studies of different HPV tests discussed in this article. Mario Poljak has also received travel grants and honoraria from Abbott, Digene, Innogenetics and Roche for speaking and for participation at scientific conferences and sitting on advisory boards. Boštjan J Kocjan has received a consultation fee from Abbott. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

Notes

^†US FDA-approved assays.

HPV: Human papillomavirus; hr: High-risk.