Comparison of HPV Genotyping Assays and Hybrid Capture 2 for Detection of High-Risk HPV in Cervical Specimens

Transcription

1 48 Available online at Annals of Clinical & Laboratory Science, vol. 41, no. 1, 2011 Comparison of HPV Genotyping Assays and Hybrid Capture 2 for Detection of High-Risk HPV in Cervical Specimens Tae Hyun Um, 1 Eun Hee Lee, 2 Hyun-Sook Chi, 3 Jong-Won Kim, 4 Young-Joon Hong, 5 and Young Joo Cha 6 1 Department of Laboratory Medicine, College of Medicine, Inje University, Goyang; 2 Department of Laboratory Medicine, Green Cross Reference Laboratory, Yongin; 3 Department of Laboratory Medicine, Asan Medical Center, College of Medicine, University of Ulsan, Seoul; 4 Department of Laboratory Medicine and Genetics, Samsung Medical Center, College of Medicine, Sungkyunkwan University, Seoul; 5 Department of Laboratory Medicine, Korea Cancer Center Hospital, Korea Institute of Radiological and Medical Sciences, Seoul; 6 Department of Laboratory Medicine, College of Medicine, Chung-Ang University, Seoul; South Korea Abstract. High-risk types of human papillomavirus (HR-HPV) are among the primary causes of cervical cancers. Hybrid Capture 2 (HC-II) (Digene, Gaithersburg, MD), which detects 13 HR-HPVs as a group, is the only HPV assay approved to date by the United States Food and Drug Administration. In Korea, several HPV genotyping assays are commercially available, including HPV RFMP (GeneMatrix Co., Seoul), HPVDNACHIP (Biomedlab Co., Seoul), and MyHPV Chip (Mygene Co., Seoul). We compared the results of these assays with those of HC-II. Among 553 residual samples of liquid-based Pap tests, a total of 435 (78.7%) were available for HPV assays. They were classified into four cytologic categories: normal, atypical squamous cells of undetermined significance (ASCUS), low-grade cervical squamous intraepithelial lesions (LSIL), and high grade SIL or carcinoma (HSIL+). Among these samples, 23.0%, 40.6%, 82.5%, 93.8% were HR-HPV positive by HC-II, respectively; 6.6%, 18.1%, 44.4%, 84.4%, by HPV RFMP, respectively; 5.7%, 24.5%, 54.0%, 84.4%, by HPVDNACHIP, respectively; and 6.6%, 11.6%, 42.9%, 84.4%, by MyHPV, respectively. Compared with HC-II, the concordance rates and kappa values were 70.6% and for HPV RFMP; 75.4% and for HPVDNACHIP; and 67.8% and for MyHPV. The concordance rates and kappa values between genotyping assays were 85.1% and for HPV RFMP and HPVDNACHIP; 83.4% and for HPV RFMP and MyHPV Chip; and 82.8% and for HPVDNACHIP and MyHPV Chip. In conclusion, compared with HC-II test, the genotyping tests showed more than fair concordance but lower sensitivity in the detection of HR-HPVs, limiting their usefulness as HR-HPV screening tools. Introduction The main risk factor for transformation of precancerous lesions to invasive cervical cancer is sexually transmitted infection by the human papillomavirus (HPV). Of the >100 HPV genotypes described, approximately 40 have tropism specifically for the anogenital mucosa. Based on the epidemiological association with high-grade cervical intraepithelial neoplasia (CIN) and cervical cancer, these anogenital HPV genotypes have been divided into high-risk and low-risk types. High-risk HPV genotypes (HR-HPV) have been shown to be causal factors in development of cervical carcinomas and precursor lesions thereof. The overall frequency of the association between HPV DNA infection and cervical squamous cell carcinoma has been reported [1,2]. Most HPV infections, regardless of high- or low-risk status, are transient and asymptomatic, and are cleared by the host immune system without development of obvious cellular abnormalities [3,4]. However, the persistent infection with the same HR-HPV genotypes substantially increases the risk of developing high-grade dysplasia, and ultimately cervical cancer [5-7]. Moreover, the risks of persistence and carcinogenicity have been found to differ markedly by HPV genotype. Thus, HPV genotypes 16 and 18 have been shown to cause 60-70% of cervical cancers worldwide; virtually all of the remaining cervical cancers are caused by HPVs of other genotypes [8-10]. Because of the association between HR-HPV infection and cervical cancer, HPV testing is an important adjunct to Pap smear cytology in screening for cervical carcinoma [11-13]. Many molecular tests for detecting HPV infection are currently available, with the Hybrid Capture 2 assay (HC-II; Digene Corp., Gaithersburg, MD) being the only commercial HPV DNA test currently approved by the United States Food and Drug Administration. HC-II detects 13 types of HR-HPVs (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, Address correspondence to Young Joo Cha, M.D., Ph.D., Department of Laboratory Medicine, College of Medicine, Chung- Ang University, Heukseok-dong, Dongiak-gu, Seoul, , Korea; chayoung@cau.ac.kr /11/ $ by the Association of Clinical Scientists, Inc.

2 and 68) and is the most widely used screening assay, with proven sensitivity and reliability [14-16]. However, the importance of HPV genotyping has been increasingly recognized in clinical practice. Thus, in addition to using pooled probes to detect the presence of one of the HR-HPVs, it is important to distinguish among individual genotypes, in particular to detect the presence of HPV 16 and/or HPV 18. As cancer intervention strategies such as vaccine preparation are becoming practical, there is an increased need for exact HPV genotyping to predict disease progression and for optimal monitoring. Thus, HPV genotyping tests may become standard clinical practice for detection of specific types of HPV strongly linked to cancer risk [17]. In Korea, several HPV genotyping tests are commercially available and have been widely used for screening of patients with cervical HPV infection, along with HC-II (Table 1). These tests include a PCR-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) assay, which is termed a restriction fragment mass polymorphism (RFMP) method (GeneMatrix Co., Seoul, Korea); and two PCRbased microarray methods, HPVDNACHIP (Biomedlab Co., Seoul, Korea) and MyHPV (Mygene Co., Seoul, Korea). Although these PCR-based HPV tests may provide alternative and rapid means for detecting HR-HPV in clinical samples [18-23], little has been known regarding their performance relative to HC-II. We therefore compared the performance of these three genotyping assays with that of HC-II in screening for HPV. Materials and Methods Patients and specimens. A total of 435 samples (78.7%) were available for HPV assays among 553 residual Comparison of HPV genotyping assays and Hybrid Capture 2 49 samples from of liquid-based Pap tests, PreservCyt (Cytyc Corp., Boxborough, MA) after cytologic diagnostic categorization as normal, atypical squamous cells of undetermined significance (ASCUS), lowgrade cervical squamous intraepithelial lesions (LSIL), and high grade SIL or carcinoma (HSIL+). Relative to the four cytologic categories, the numbers (%) of samples were 122 (28.0%), 155 (35.6%), 126 (29.0%), and 32 (7.4%), respectively. All patients provided written informed consent for the use of their samples in HPV genotyping assays. The study protocol was approved by the Institutional Review Board for clinical research on human subjects at Chung-Ang University. Liquid-based cytology. Brush swabs of uterine cervix samples, stored in vials of PreservCyt solution (Cytyc), were placed in the Thin Prep (Cytyc) Processor. Each ThinPrep slide was fixed in ethanol and stained with Papanicolaou stain. The number of epithelial cells on each slide was estimated from the number of cells contained within computer-derived coordinates for 50 random fields in a 20-mm diameter circular area centered where cells were deposited. Cytological diagnosis of each sample using the new Bethesda system was performed by pathologists in the Green Cross Reference Laboratory. Residual sample preparation. Five ml of each residual PreservCyt sample was used for HC-II assays. Each remaining sample (about 5 ml) after HC-II was centrifuged for 5 min at 3,000 g and 2 ml of the pellet was resuspended. Two hundred µl of resuspended pellet was used for cellular and viral DNA extraction using QIAamp DNA Blood Mini Kits (Qiagen Sciences, Germantown, MD). DNA concentration was measured, and if it was <100 ng/µl, the sample was discarded and excluded from this study. DNA concentration was adjusted to 100 ng/µl and 50 µl aliquots were sent to each participating institution for HPV genotyping using the HPV RFMP, HPVDNACHIP, and MyHPV CHIP assays. HC-II assay was performed at the IlsanPaik Hospital, and HPV RFMP assay at Green Cross Reference Laboratory, HPVDNACHIP at Asan Medical Center and Korea Cancer Center Hospital, and MyHPV Chip at Korea Cancer Center Hospital. High-risk HPV screening test by HC-II. Four ml of each residual PreservCyt sample was centrifuged and resuspended in 150 µl of a mixture of specimen transport medium and denaturation buffer provided in the kit. A 75 µl aliquot of each resuspended sample was utilized for each HC-II assay, a sandwich capture molecular hybridization assay that employs chemiluminescence detection. All assays were performed according to the manufacturer s protocol, with the results reported as a ratio of relative light units (RLU). The analytical sensitivity of this assay is 1 pg HPV DNA, which is equal to approximately 100,000 HPV genome equivalents [24]. HPV genotyping by HPV RFMP. The HPV RFMP method, which uses PCR, enzyme restriction, and matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry (MALDI-TOF MS), can detect 14 high-risk HPV types (Table 1). HPV DNA was amplified with PGMY09/11, consisting of two non-degenerate pools of L1 consensus primers [20]. Each 18 µl of PCR amplification mixture contained 4 µl of cervical DNA, 20 mm Tris-HCl (ph 8.4), 50 mm KCl, 0.2 mm of each dntp, 2 mm MgSO 4, 0.4 µm of each primer, and 1 unit of Platinum Taq polymerase (Invitrogen). The amplification protocol consisted of 40 cycles of denaturation at 94 C for 15 sec,

3 50 Annals of Clinical & Laboratory Science, vol. 41, no. 1, 2011 Table 1. Characteristics of HPV DNA tests evaluated in this study. HPV tests Test principle Target genotypes HC-II Hybridization 13 HR; 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68 & chemiluminescence HPV RFMP PCR, enzyme restriction, 14 HR; 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 67, 82 mass analysis of restricted Others; 6, 11, 13, 26, 30, 34, 40, 42, 43, 44, 53, 54, 55, 57, 61, 62, 64, 66, products by mass spectrometry 70, 72, 74, 81, 83, 84, 89, 90 HPVDNACHIP PCR and chip hybridization 15 HR; 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68, 69 7 LR; 6, 11, 34, 40, 42, 43, 44 My HPV CHIP PCR and chip hybridization 16 HR; 16, 18, 31, 33, 35, 39, 45, 51, 52, 53, 54, 56, 58, 59, 66, 68 8 LR; 6, 11, 34, 40, 42, 43, 44, 70 Abbreviations: HR, high risk; LR, low risk. annealing at 60 C for 15 sec, and extension at 72 C for 30 sec. The second round of amplification utilized nested PCR primer pairs, consisting of a sense primer: (5 -GCMCAGGGHCAYAAGGATGAATGG-3 ; M=A/C, H=A/C/T, Y=C/T), and an antisense primer: (5 -GTACTDCKDGTRGTATCHACMACGG ATGTAACAAA-3 ; D=A/G/T, K=G/T, R=A/G). Each 25 µl of PCR amplification mixture was identical to the firstround mixture except for the primers and the addition of 2 µl of a 1/20 dilution of the first-round PCR product. The amplification protocol was identical to that of the first round, except that the annealing temperature was 45 C. A plasmid containing the entire HPV16 genome (ATCC 45113D) was used as the positive control and distilled water as the negative control. PCR products were digested with the restriction enzymes FokI and BtsCI by incubating each PCR reaction mixture with 10 µl of buffer containing 50 mm potassium acetate, 20 mm Tris-acetate, 10 mm magnesium acetate, 1 mm dithiothreitol, and 1 unit each of FokI and BtsCI at 37 C for 1 hr. The resulting digest was purified by vacuum filtration through Oasis elution plates (Waters Co., MA). Each well was equilibrated with 90 µl of 1 M triethylammonium acetate (ph 7.6). Desalted reaction mixtures were resuspended in matrix solution containing 15 mg/ml 3-hydroxypicolinic acid, M ammonium citrate, and 12% (V/V) acetonitrile; 3 µl of each sample was spotted onto a polished MTP AnchorChip (Bruker Daltonics, ME). Mass spectra were acquired on a Biflex IV linear MALDI-TOF MS (Bruker Daltonics) workstation with a 337 nm nitrogen laser and a nominal ion flight path length of 1.25 m. HPV genotyping by HPVDNA- CHIP. HPVDNACHIP (Biomedlab Co.) is a PCR-based DNA microarray system containing 22 type-specific HPV probes, 15 highrisk and 7 low-risk (Table 1). All assays were performed according to the manufacturer s protocol [21]. Briefly, target HPV DNA was PCRamplified using hpv1/2 primers and Cy5-dUTP (NEN Life Science, Boston, MA). As an internal control, β-globin DNA was simultaneously amplified. The resulting products were denatured with NaOH solution, neutralized with Tris/HCl and hydrochloric acid, and cooled for 5 min on ice. Samples were mixed with hybridization solution and applied to the DNA chip. The chip was hybridized at 40 C for 2 hr, washed with saline-sodium phosphate-edta, and air-dried at room temperature. Hybridized HPV DNA was visualized using a commercial DNA chip scanner (GenePix Pro 3.10; Axon Instruments, Union City, CA). HPV genotyping by MyHPV CHIP. MyHPV CHIP (MyGene Co.) is a PCR-based DNA microarray system containing 24 type-specific probes, 16 high-risk and 8 low-risk (Table 1). All assays were performed according to the manufacturer s protocol [22]. Briefly, target HPV DNA was PCR-amplified using GP5d+/GP6d+ primers and Cy5- dutp (NEN Life Science), with β-globin DNA simultaneously amplified as an internal control. PCR products were denatured by heating for 5 min at 95 C, mixed with hybridization solution and applied to the DNA chip. The chip was hybridized at 43 C for 90 min, washed with saline-sodium phosphate-edta, and air-dried at room temperature. Hybridized HPV DNA was visualized using a DNA chip scanner (GenePix Pro 3.10). Statistical analysis. A positive result on genotyping assays was defined as the presence of one or more of the 13 high-risk HPV genotypes that are detected by HC-II. The Chisquare test was used to analyze

4 Comparison of HPV genotyping assays and Hybrid Capture 2 51 Table 2. HPV test results of genotyping assays. Result interpretation Number Genotyping assays of genotypes HPV RFMP HPVDNACHIP MyHPV HR-HPV Positive 1 1 HR 106 (24.4%) 100 (23.0%) 76 (17.5%) 2 2 HR 2 (0.5%) 11 (2.5%) 12 (2.8%) 1 HR + 1 LR 11 (2.5%) 15 (3.4%) 14 (3.2%) 3 3 HR 3 (0.7%) 2 (0.5%) 2 HR + 1 LR 5 (1.1%) 2 (0.5%) 1 HR + 2 LR 3 (0.7%) 1 (0.2%) 4 3 HR + 1 LR 1 (0.2%) 2 HR + 2 LR 1 (0.2%) 5 2 HR + 3 LR 1 (0.2%) Subtotal 119 (27.4%) 140 (32.2%) 107 (24.6%) HR-HPV Negative 0 No HPV 211 (48.5%) 246 (56.6%) 291 (66.9%) 1 1 LR 62 (14.3%) 24 (5.5%) 25 (5.7%) 2 2 LR 5 (1.1%) 3 (0.7%) 3 (0.7%) Other 38 (8.7%) 22 (5.1%) 9 (2.1%) Subtotal 316 (72.6%) 295 (67.8%) 328 (75.4%) Total 435 (100%) 435 (100%) 435 (100%) Abbreviations: See Table 1. differences in cytological proportions among initial residual cervical samples and samples available for HPV tests. The Chi-square test or Fisher s exact test was employed to analyze contingency tables comparing HPV positivity rates yielded by the various assays. Correlations between HPV assays were assessed using Cohen kappa index statistics, with values of 0.00 to 0.20 indicating poor agreement; 0.21 to 0.40 fair agreement; 0.41 to 0.60 moderate agreement; 0.61 to 0.80 substantial agreement; and 0.81 to 1.00 nearly perfect agreement. The statistical tests were performed using MedCalc v. 9.3 for Windows (MedCalc Software, Mariakerke, Belgium); p <0.05 was considered statistically significant. Results Specimens for the comparison study. Of the 553 cervical specimens available, 151 (27.3%), 200 (36.2%), 150 (27.1%), and 52 (9.4%) were categorized cytologically as normal, ASCUS, LSIL, and HSIL+, respectively. Because of sample quality, 435 of the 553 samples (78.7%) were available for comparative testing. The 435 samples were composed of 122 (28.0%), 155 (35.6%), 126 (29.0%), and 32 (7.4%), with cytological categories of normal, ASCUS, LSIL, and HSIL+, respectively. There was no significant difference in cytological categories between the initial cervical samples and the samples available for HPV testing (p = 0.67). Comparison of HPV genotypes detected by genotyping assays vs HC-II results. Genotyping assays could detect co-infections with as many as 5 different genotypes in one sample (Table 2). In genotyping assays, HPV 16, 58, 52, 56, 18 were frequently detected HR-HPVs, with HPV 16 being the most frequent detected one. But the exact orders of detected HR-HPVs were different between assays; HPV RFMP: 16, 52, 58, 39, 18, 33, 35, 56, 31, 45, 51, 68; for HPVDNACHIP: 16, 58, 52, 56, 51, 35, 39, 33, 68, 18, 31, 45, 59; for MyHPV: 16, 58, 18, 31, 33, 52, 56, 39, 68, 35. Considerable discrepancy was found in results between HC-II and genotyping assays, such that HC-II was negative when HR-HPV was detected in genotyping assays, or HC-II was positive but any HR- HPV was not detected in genotyping assays, and vice versa. Among the genotypes not included in the high risk HC-II panel, HPV 66 was the most frequent genotype to test positive. HPV positivity rates relative to cytologic category. Using HC-II, we found that the overall HPV positivity rate was 51.7% (225/435), similar to that of HPV RFMP (51.5%), but higher than those of HPVDNACHIP (43.4%) and MyHPV (33.1%) (Table 3). When results from all four HPV tests were combined, the HPV positivity rate increased in parallel with rising cytological grade. Of samples from the normal, ASCUS, LSIL, and HSIL+ groups, 23.0%, 40.6%, 82.5%, and 93.8%, respectively, were positive by the HC-II assay; 19.7%, 45.2%, 80.2%, and 90.6%

5 52 Annals of Clinical & Laboratory Science, vol. 41, no. 1, 2011 Table 3. HPV positivity relative to cytologic grade. HPV test N (%) of HPV positive and high risk HPV positive specimens/ Proportion of high risk HPV positives among HPV positive specimens Normal (122) ASCUS (155) LSIL (126) HSIL+ (32) Total (435) HC-II 28 (23.0%), 28 (23.0%), (100%) 63 (40.6%), 63 (40.6%), (100%) 104 (82.5%), 104 (82.5%), (100%) 30 (93.8%), 30 (93.8%), (100%) 225 (51.7%), 225 (51.7%), (100%) HPV RFMP 24 (19.7%), 8 (6.6%), (33.3%) 70 (45.2%), 28 (18.1%), (40.0%) 101 (80.2%), 56 (44.4%), (55.4%) 29 (90.6%), 27 (84.4%), (93.1%) 224 (51.5%), 119 (27.4%), (53.1%) HPVDNACHIP 15 (12.3%), 7 (5.7%), (46.7%) 55 (35.5%), 38 (24.5%), (69.1%) 91 (72.2%), 68 (54.0%), (74.7%) 28 (87.5%), 27 (84.4%), (96.4%) 189 (43.4%), 140 (32.2%), (74.1%) MyHPV 9 (7.4%), 8 (6.6%), (88.9%) 33 (21.3%), 18 (11.6%), (54.5%) 75 (59.5%), 54 (42.9%), (72.0%) 27 (84.4%), 27 (84.4%), (100%) 144 (33.1%), 107 (24.6%), (74.3%) p (HC-II vs HPV RFMP) , , NA , <0.0001, NA , <0.0001, NA , , NA , <0.0001, NA p (HC-II vs HPVDNACHIP) , , NA , , NA , <0.0001, NA , , NA , <0.0001, NA p (HC-II vs MyHPV) , , NA , <0.0001, NA , <0.0001, NA , , NA < , <0.0001, NA p (HPV RFMP vs , , , , , , , , , , < HPVDNACHIP) p (HPV RFMP vs MyHPV) , , <0.0001, , , , , , <0.0001, , p (HPVDNACHIP vs , , , , , , , , , , MyHPV) The p values were calculated using Chi-square test or Fisher s exact test according to sample size. Table 4. Concordance rate for HR-HPV detection between HPV genotyping tests and genotyping with HC-II. HPV Tests Pos-Pos Neg-Neg Pos-Neg Neg-Pos Agreement kappa (95% CI) HC-II vs HPV RFMP % (307/435) ( ) HC-II vs HPVDNACHIP % (328/435) ( ) HC-II vs MyHPV % (295/435) ( ) HPV RFMP vs HPVDNACHIP % (370/435) ( ) HPV RFMP vs MyHPV % (363/435) ( ) HPVDNACHIP vs MyHPV % (360/435) ( ) Pos=Positive; Neg=Negative

6 were positive by HPV RFMP; 12.3%, 35.5%, 72.2%, and 87.5% were positive by HPVDNACHIP; and 7.4%, 21.3%, 59.5%, and 84.4% were positive by MyHPV. HR-HPV positivity rates relative to cytologic category. The overall positivity rate of HC-II for the 13 HR-HPV types (51.7%) was significantly higher than the positivity rates of HPV RFMP (27.4%), HPVDNACHIP (32.2%), and MyHPV (24.6%) (Table 3). In each of these HPV tests, the HR- HPV positivity rate increased in parallel with rising cytologic category. Of samples from the normal, ASCUS, LSIL, and HSIL+ groups, 6.6%, 18.1%, 44.4%, and 84.4%, respectively, were HR-HPV positive by HPV RFMP assay; 5.7%, 24.5%, 54.0%, and 84.4% were HR-HPV positive by HPVDNACHIP assay; and 6.6%, 11.6%, 42.9%, and 84.4% were HR-HPV positive by MyHPV assay. Rates of HR-HPV positivity relative to HPV positivity. In every HPV genotyping test except MyHPV, the proportions of HR-HPV detection relative to detection of all HPVs increased with rising cytological grade (Table 3). In the normal, ASCUS, LSIL, and HSIL+ groups, 33.3%, 40.0%, 55.4% and 93.1%, respectively, of HPVs were HR- HPV by HPV RFMP assay; 46.7%, 69.1%, 74.7% and 96.4% were HR- HPV by HPVDNACHIP assay; and 88.9%, 54.5%, 72.0% and 100% were HR-HPV by MyHPV assay. Correlations of genotyping assay results with HC-II and between genotyping assays. Compared to HC-II, the correlation of each genotyping test for the 13 HR- HPVs was fair-to-moderate. The kappa values of HPV RFMP, HPVDNACHIP, and MyHPV assays were (95% confidence interval [CI] = to 0.505), (95% CI = to 0.594), and (95% CI = to 0.454), respectively (Table 4). We found that the concordance rates between genotyping tests for the 13 HR-HPVs were moderate-tosubstantial, thus higher than those between genotyping tests and HC- II (Table 4). The kappa values were (95% CI = to 0.724) for HPV RFMP vs HPVDNACHIP, (95% CI = to 0.664) for HPV RFMP vs MyHPV, and (95% CI = to 0.666) for HPVDNACHIP vs MyHPV. Discussion Comparison of HPV genotyping assays and Hybrid Capture 2 53 In assessing women with ASCUS smears, HPV testing has been found to be a cost-effective secondary triage test, after cervical cytology [11]. Many different methods of HPV testing, of varying sensitivities and specificities, have been used to detect the presence of HR-HPV in cervical samples. To date, the HC-II system is the only test approved by the United States Food and Drug Administration for diagnosis of HPV infection. The HC-II system is very sensitive and has a high negative predictive value, thus decreasing costs associated with unnecessary colposcopic referrals. However, the HC-II assay does not yield any genotype-specific information, cannot detect multiple infections, and does not distinguish between persistent infections with the same HR-HPV type and those in which the genotype has changed between tests. Although it is important to detect the presence of one or more HR-HPV genotypes using pooled probes, it is also necessary to distinguish among individual genotypes, especially to detect the presence of HPV 16 and/or HPV 18. The importance of HPV genotyping has been increasingly recognized in clinical practice. The most frequently detected HR-HPV genotypes are known to be different according to countries. HPV types 16, 18, 31 and 45 are reported to be dominant in Europe and USA, whereas genotypes 16, 58, 52, 56, 51 are dominant in Korea as observed in this study. Identification of HPV genotype can be used to predict the biological course of HPV type-specific cervical lesions, thus allowing appropriate patient care to be targeted according to HPV genotype and possibly yielding information that may lead to the development of vaccines for cervical cancer [8]. As cancer intervention strategies such as vaccine preparation are becoming practical, there is an increased need for exact HPV genotyping to predict disease progression and for optimal monitoring. Thus, HPV genotyping tests may become standard clinical practice for detection of specific types of HPV strongly linked to cancer risk [18]. Women infected with HR-HPV have higher rates of CIN progression and cervical cancer than do those infected with low-risk HPV. These findings suggest that HPV genotyping assays are effective, especially in diagnosing infections with multiple HPV genotypes. Several HPV genotyping tests, based on chip microarrays or PCR- RFMP, have been developed in Korea and are commercially available, primarily for the purpose of HPV screening. We evaluated the performance of these genotyping assays in comparison with HC-II, for detection of HPV and HR-HPV. We found that the overall HPV positivity rates were higher with HC-II than by the genotyping assays, and the differences were statistically significant for two genotyping assays, HPVDNACHIP and MyHPV, but not for HPV RFMP. Moreover, the HR-HPV positivity rates of all three genotyping assays (24.6% to 32.2%) were much lower than that of HC-II (51.7%). This may be attributable to low sensitivities of the genotyping

7 54 Annals of Clinical & Laboratory Science, vol. 41, no. 1, 2011 assays, because the HPV-positivity rates of the genotyping tests were lower than that of HC-II, which is similar to those reported elsewhere [13,18,19,25]. The sensitivity of each HPV detection method is based on the threshold value of viral load/viral concentration. In general, amplification assays, such as PCR-based techniques, are more sensitive than liquid hybridization tests, such as the HC-II. In contrast, we found that, after sorting by cytologic group, HR-HPV positivity rates were significantly lower when assessed by genotyping tests than when measured by HC-II, except in the HSIL+ group. The interpretation of genotyping tests employing chip technology is somewhat subjective because the hybridized PCRamplified products are scanned and interpreted by the human assayer. The positive predictive value of HPVDNACHIP has been reported to be 53.8%, similar to 56.8% for HC-II [21]. Although highly sensitive HPV tests are preferable to less-sensitive tests for epidemiological purposes, such tests are not necessarily optimal for clinical purposes. For example, there may be a viral load threshold below which an HPV infection is not clinically relevant. Definitions of analytical or clinical sensitivity and specificity have been used to distinguish clinically irrelevant from relevant HPV- positive result [26]. Analytical sensitivity is defined as the proportion of HPV-positive women correctly identified by a given test, regardless of clinical importance, whereas clinical sensitivity is defined as the proportion of women with disease correctly identified by a positive HPV test. HPV infection is an indicator of the risk of having or developing a cervical lesion but is not equivalent to a morphological disorder. All assays evaluated in this study showed a tendency toward increased HR-HPV- positive rates as cytologic grade increased. This finding is consistent with results showing that the percentage of women with multiple infections decreased whereas the percentage of women with HR-HPVs increased as cytologic severity rose [19]. HPV genotyping tests, using various technologies, have been employed primarily for epidemiological investigations. Performances are variable in resolution and in the number of identifiable genotypes. The three HPV genotyping tests we assessed consisted of two DNA array chip assays and one PCR- RFMP procedure. For the detection of 13 HR-HPV types, these three genotyping assays showed concordance rates of 67.8~75.4% and kappa values of 0.367~0.514, compared with HC-II test. The concordance rate between HPV RFMP and HC- II was 70.6%, although that has been 88% in another report [20]. Of the samples positive by HC-II, about half tested negative in genotyping tests. The discrepancy could be attributed to high falsepositive rate of HC-II for low, intermediate, or unassigned risk genotypes and/or high false-negative rates of genotyping tests. These results were not attributable solely to high false-positive rates of HC-II. Rather, the findings may be explained by the relatively low sensitivity of the genotyping tests, in consideration of an earlier study reporting only 10.3% (32/312) of HC-II positive samples were positive because of cross-reactivity with HPV genotypes not included in the high-risk probe cocktail of HC-II [27]. These discrepancies may be resolved by basic evaluations such as determination of limits of detection (LOD). It was noteworthy that the order of frequently detected HR- HPVs was quite different between genotyping assays, although HPV 16 was the most frequent HR-HPV in every genotyping test. They were 16, 52, 58, 39, 18, 33, 35, 56, 31, 45, 51, 68 in HPV RFMP; 16, 58, 52, 56, 51, 35, 39, 33, 68, 18, 31, 45, 59 in HPVDNACHIP; and 16, 58, 18, 31, 33, 52, 56, 39, 68, 35 in MyHPV. These results suggest that the differences in HR-HPV positive rates between assays may be attributable to differences in each genotype s detection performance. In conclusion, when compared with HC-II, the genotyping tests assessed showed more than fair concordance but lower sensitivity in detection of HR-HPV. The lower sensitivities and poorer objectivity in result interpretation limit the current usefulness of genotyping assays as HPV screening tools. Acknowledgement This study was supported by a grant 9A084507) of the Korea Healthcare Technology R&D Project, Ministry of Health and Welfare, Republic of Korea. References 1. Munõz N, Bosch FX, Sanjosé Sd, Herrero R, Castellsagué X, Shah KV, Snijders PJF, Meijer CJLM. Epidemiologic classification of human papillomavirus types associated with cervical cancer. NEJM 2003;348: Kleter B, van Doorn LJ, ter Schegget J, Schrauwen L, van Krimpen K, Burger M, ter Harmsel B, Quint W. Novel short-fragment PCR assay for highly sensitive broad-spectrum detection of anogenital human papilloma-viruses. Am J Pathol 1998;153: Elfgren K, Jacobs M, Walboomers JM, Meijer CJ, Dillner J. Rate of human papillomavirus clearance after treatment of cervical intraepithelial neoplasia. Obstet Gynec 2002;100: Richardson H, Kelsall G, Tellier P, Voyer H, Abrahamowicz M, Ferenczy A, Coutlee F, Franco EL. The natural history of type-specific human papillomavirus infections in female university students. Cancer

8 Epidemiol Biomark Prev 2003;12: Kjaer SK, van den Brule AJ, Paull G, Svare EI, Sherman ME, Thomsen BL, Suntum M, Bock JE, Poll PA, Meijer CJ. Type specific persistence of high risk human papillomavirus (HPV) as indicator of high grade cervical squamous intraepithelial lesions in young women: population based prospective follow up study. BMJ 2002;325: Wallin K-L, Wiklund F, Ångström T, Bergman F, Stendahl U, Wadell G, Hallmans G, Dillner J. Typespecific persistence of human papillomavirus DNA before the development of invasive cervical cancer. NEJM 1999;341: Clifford GM, Smith JS, Plummer M, Munoz N, Franceschi S. Human papillomavirus types in invasive cervical cancer worldwide: a metaanalysis. Br J Cancer 2003; 88: Khan MJ, Castle PE, Lorincz AT, Wacholder S, Sherman M, Scott DR, Rush BB, Glass AG, Schiff-man M. The elevated 10-year risk of cervical precancer and cancer in women with human papilloma-virus (HPV) type 16 or 18 and the possible utility of type-specific HPV testing in clinical practice. JNCI 2005;97: Schiffman M, Herrero R, Desalle R, Hildesheim A, Wacholder S, Rodriguez AC, Bratti MC, Sherman ME, Morales J, Guillen D, Alfaro M, Hutchinson M, Wright TC, Solomon D, Chen Z, Schussler J, Castle PE, Burk RD. The carcinogenicity of human papillomavirus types reflects viral evolution. Virology 2005;337: Schiffman M, Wheeler CM, Dasgupta A, Solomon D, Castle PE. A comparison of a prototype PCR assay and hybrid capture 2 for detection of carcinogenic human papillomavirus DNA in women with equivocal or mildly abnormal Papanicolaou smears. Am J Clin Pathol 2005;124: Cuschieri KS, Cubie HA. The role of human papillomavirus testing in cervical screening. J Clin Virol 2005;32:S Masumoto N, Fujii T, Ishikawa M, Mukai M, Saito M, Iwata T, Fukuchi Comparison of HPV genotyping assays and Hybrid Capture 2 55 T, Kubushiro K, Tsukazaki K, Nozawa S. Papanicolaou tests and molecular analyses using new fluidbased specimen collection technology in 3000 Japanese women. Br J Cancer 2003;88: Sarode VR, Werner C, Gander R, Foster B, Fulmer A, Saboorian MH, Ashfaq R. Reflex human papillomavirus DNA testing on residual liquid-based (TPPT) cervical samples: focus on agestratified clinical performance. Cancer 2003;99: Poljak M, Brencic A, Seme K, Vince A, Marin IJ. Comparative evaluation of first- and second-generation digene hybrid capture assays for detection of human papillomaviruses associated with high or intermediate risk for cervical cancer. J Clin Microbiol 1999;37: Castle PE, Wheeler CM, Solomon D, Schiffman M, Peyton CL. Interlaboratory reliability of Hybrid Capture 2. Am J Clin Pathol 2004; 122: Kulmala SM, Syrjanen S, Shabalova I, Petrovichev N, Kozachenko V, Podistov J, Ivanchenko O, Zakharenko S, Nerovjna R, Kljukina L, Branovskaja M, Grunberga V, Juschenko A, Tosi P, Santopietro R, Syrjanen K. Human papillo-mavirus testing with the hybrid capture 2 assay and PCR as screening tools. J Clin Microbiol 2004;42: Nobbenhuis MA, Walboomers JM, Helmerhorst TJ, Rozendaal L, Remmink AJ, Risse EK, van der Linden HC, Voorhorst FJ, Kenemans P, Meijer CJ. Relation of human papillomavirus status to cervical lesions and consequences for cervicalcancer screening: a prospective study. Lancet 1999; 354: An HJ, Cho NH, Lee SY, Kim IH, Lee C, Kim SJ, Mun MS, Kim SH, Jeong JK. Correlation of cervical carcinoma and precancerous lesions with human papillomavirus (HPV) genotypes detected with the HPV DNA chip microarray method. Cancer 2003;97: Hwang HS, Park M, Lee SY, Kwon KH, Pang MG. Distribution and prevalence of human papillo-mavirus genotypes in routine pap smear of 2,470 korean women determined by DNA chip. Cancer Epidemiol Biomark Prev 2004;13: Lee EH, Chung HJ, Oh HB, Chi HS, Jee MS, Park SN, Hong SP, Yoo W, Kim SO. Human papilloma virus genotyping assay using restriction fragment mass polymorphism analysis, and its comparison with sequencing and hybrid capture assays. Korean J Lab Med 2007;27: Lee GY, Kim SM, Rim SY, Choi HS, Park CS, Nam JH. Human papillomavirus (HPV) genotyping by HPV DNA chip in cervical cancer and precancerous lesions. Int J Gynecol Cancer 2005;15: Choi YD, Jung WW, Nam JH, Choi HS, Park CS. Detection of HPV genotypes in cervical lesions by the HPV DNA chip and sequencing. Gynecol Oncology 2005;98: Lee JK, Kim MK, Song SH, Hong JH, Min KJ, Kim JH, Song ES, Lee J, Lee JM, Hur SY. Comparison of human papillomavirus detection and typing by hybrid capture 2, linear array, DNA chip, and cycle sequencing in cervical swab samples. Int J Gynecol Cancer 2009;19: Davies P, Kornegay J, Iftner T. Current methods of testing for human papillomavirus. Best Pract Res Clin Obstet Gynaecol 2001; 15: Atypical Squamous Cells of Undetermined Significance/Low- Grade Squamous Intraepithelial Lesions Triage Study (ALTS) Group. Human papillomavirus testing for triage of women with cytologic evidence of low-grade squamous intraepithelial lesions: Baseline data from a randomized trial. JNCI 2000;92: Snijders PJ, Van den Brule AJ, Meijer CJ. The clinical relevance of human papillomavirus testing: relationship between analytical and clinical sensitivity. J Pathol 2003;201: Poljak M, Marin IJ, Seme K, Vince A. Hybrid Capture II HPV test detects at least 15 human papillomavirus genotypes not included in its current high-risk probe cocktail. J Clin Virol 2002;25:S89-97.