Telomerase and human papillomavirus as diagnostic adjuncts for cervical dysplasia and carcinoma

Original Contributions Telomerase and Human Papillomavirus as Diagnostic Adjuncts for Cervical Dysplasia and Carcinoma ELKE A. JARBOE, MD, MS, L. CHESNEY THOMPSON, MD, DAVID HEINZ, BS, J. A. MCGREGOR, MD, AND KENNETH R. SHROYER, MD, PHD Telomerase and human papillomavirus (HPV) DNA were evaluated as potential markers of high-grade dysplasia in cervical cytological specimens. Cytology specimens were collected from patients at the time of colposcopic evaluation for management of a previous abnormal cytology test result. Telomerase activity was evaluated by the telomeric repeat amplification protocol (TRAP), and HPV DNA was detected by polymerase chain reaction with L1 consensus-sequence primers and filter hybridization genotyping. Telomerase was detected in 8 of 97 (8.2%) cases with normal cytology or benign cellular changes, in 7 of 98 (7.1%) cases of atypical squamous cells of undetermined significance (ASCUS), in 3 of 95 (3.2%) cases of low-grade squamous intraepithelial lesion (LSIL), and in 17 of 48 (35.4%) cases with high-grade squamous intraepithelial lesion (HSIL). High-risk HPVs were detected in 23 of 97 (23.7%) cases with normal/reactive cellular changes (RCC) cytology, in 28 of 98 (28.6%) cases of ASCUS, in 69 of 95 (72.6%) cases of LSIL, and in 35 of 48 (72.9%) cases of HSIL. Telomerase expression did not correlate with the detection of high-risk HPVs in any cytological diagnostic categories. Telomerase and HPV test results of cytological specimens were correlated with the histological diagnoses of concurrent cervical bi-

opsy specimens. Telomerase showed a sensitivity of 29.9% and a specificity of 94.0% for biopsy-confirmed cervical intraepithelial neoplasia (CIN) II/III. In contrast, high-risk HPVs were detected in 70.1% of cases with underlying CIN II/III, with a specificity of 62.5%. A relatively high proportion of normal/RCC or ASCUS cases with telomerase-positive test results had underlying high-grade dysplasia on cervical biopsy. Thus, technical and practical limitations of the TRAP assay in cervical cytology specimens limit the practical application of telomerase as a diagnostic adjunct in cervical cytopathology. HUM PATHOL 35:396-402. © 2004 Elsevier Inc. All rights reserved. Key words: telomerase, cervical dysplasia, cervical carcinoma, human papillomavirus, cervical cytology. Abbreviations: ASCUS, atypical squamous cells of undetermined significance; CIN, cervical intraepithelial neoplasia; HPV, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; ITAS, internal telomerase assay standard; LSIL, low-grade squamous intraepithelial lesion; PAGE, polyacrylamide gel electrophoresis; RCC, reactive cellular changes; SCC, squamous cell carcinoma; TNS, type not specified; TRAP, telomeric repeat amplification protocol.

In the United States and other industrialized nations, cervical cancer ranks about tenth among fatal malignancies of women, but in many developing countries, cervical cancer remains the most common cause of cancer death in women.1 Squamous cell carcinoma (SCC) is the most common histological type of cervical cancer, representing 75% to 77% of cases.2 Potential precursors of SCC are detected by cytological examination (ie, Pap test) and are classified by the federally mandated Bethesda system of cytological classification. Human papillomaviruses (HPVs) are the etiological agents for almost all cases of cervical SCC and high-grade cervical dysplasia. More than 100 different

HPV types have been classified, including approximately 40 known to infect the female genital tract.3 HPVs have been grouped into high-risk and low-risk types, depending on their respective frequency of association with cervical cancer and dysplasia versus condyloma acuminatum. Although high-risk HPV DNA can be detected in almost all cases of high-grade cervical dysplasia and carcinoma, it is not a specific marker of high-grade dysplasia or carcinoma, because a significant proportion of women with normal cervical mucosa also test positive for high-risk HPVs. The HPV viral genome is encoded on an approximately 8000-bp, circularized, double-stranded DNA that includes early and late open reading frames. The early open reading frames encode proteins involved in the regulation of DNA synthesis and cell cycle control. From the standpoint of their role in cellular transformation, E6 and E7 play the most significant roles through their respective interaction with and inactivation of the p53 and pRb cell cycle checkpoint pathways.3-5 Because of the clear association of SCC with underlying HPV infection, HPV testing has been recommended as a routine component of the management of those cytological specimens diagnosed as atypical squamous cells of undetermined significance (ASCUS).6,7

From the Department of Pathology and Department of Obstetrics and Gynecology, University of Colorado Health Sciences Center, Denver, CO. Accepted for publication August 20, 2003. Supported in part by grant CA78442-03 from the National Cancer Institute. Address correspondence and reprint requests to Kenneth R. Shroyer, MD, PHD, Professor, Department of Pathology, University of Colorado Health Sciences Center, 4200 East Ninth Ave., Denver, CO 80262. 0046-8177/$—see front matter © 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2003.08.028

396

DIAGNOSIS OF CERVICAL DYSPLASIA AND CARCINOMA (Jarboe et al)

The American Society for Colposcopy and Cervical Pathology was convened in September 2001 to define patient management guidelines for the management of cytological abnormalities and cervical cancer precursors (http://www.asccp.org). The resulting consensusbased guidelines from this meeting support the application of HPV testing as a diagnostic adjunct to triage patients with a cytological diagnosis of ASCUS, but do not support implementation of HPV testing in other diagnostic categories. Despite this recommendation, however, studies have failed to demonstrate a high level of specificity of HPV DNA for clinically significant cervical disease.6-8 Multiple studies have evaluated telomerase as a biomarker of dysplasia with potential utility as a diagnostic adjunct in cervical cytology. Telomerase is a ribonucleoprotein DNA polymerase containing an RNA component that directs the synthesis of telomeric DNA repeats.9 Telomerase is expressed in normal germ line cells and normal proliferative and secretory phase endometrium, but is found at low to undetectable levels in most other somatic tissues.10 The RNA component of human telomerase, designated as hTR, contains an 11-nucleotide sequence that is complementary to the telomeric DNA repeat, (TTAGGG)n.11 The major catalytic protein subunit of human telomerase is designated as hTERT; hTERT cDNA includes seven reverse-transcriptase motifs, as well as a single telomerase-specific motif, in a single continuous open reading frame.12 hTERT mRNA has been shown to be expressed in both primary tumors and cancer cell lines, but is not detectable in telomerase-negative cell lines or most normal somatic tissues, including heart, brain, placenta, liver, skeletal muscle, and prostate.13 Telomerase activity has also been found to be expressed in a significant proportion of both high-grade squamous intraepithelial lesions (HSILs) and cervical SCCs, suggesting that telomerase contributes to maintenance of the unlimited replicative phenotype of intramucosal and invasive neoplastic cervical epithelial cells.14 The malignant transformation of the cervical mucosa resulting from HPV infection may be accomplished largely through the activation of telomerase. Interaction of the E6 oncoprotein with the hTERT promoter proximal to the ATG initiation codon has been shown to result in telomerase activation.15 Other studies of cultured human keratinocytes have demonstrated that E6 oncoprotein of high-risk HPVs is capable of activating telomerase expression via a p53-independent mechanism.16,17 When E7 was also expressed in these cells, an even higher level of telomerase expression was achieved.15 Conversely, repression of either E6 or E7 in HeLa cells has been shown to inhibit telomerase activity.18 Using L1 consensus sequence polymerase chain reaction (PCR) Anderson et al17 found HPV 16 DNA in 3 and HPV 18 DNA in 4 of 10 telomerase-positive cases of cervical carcinoma. Yashima et al19 detected HPV type 16/18 in 11 of 13 HSIL cervical biopsy specimens using GP5/GP6 consensus sequence PCR, including 4 biopsy specimens that were positive for both HPV DNA and telomerase

activity. In cases of high-grade dysplasia and SCC, teleromase expression has been shown to correlate with the intensity of HPV 16/18 E6/E7 oncogene expression as determined by in situ hybridization.20 In the present study we tested the hypothesis that telomerase is a sensitive and specific marker of premalignant lesions of the cervical squamous mucosa in patients undergoing colposcopic evaluation for previous cytological abnormalities. Telomerase test results were correlated with the cytological diagnosis, colposcopic biopsy results, and detection of HPV by a sensitive PCR-based assay. MATERIALS AND METHODS Specimens Cervical cytology specimens were collected from 456 patients seen in the colposcopy clinics of the University of Colorado Health Sciences Center. All patients were considered “high-risk,” requiring colposcopic evaluation due to the detection of abnormal cervical cytology on a previous cytological examination, including repeat diagnosis of ASCUS or previous diagnosis of low-grade squamous intraepithelial lesions (LSIL) or greater. A comprehensive review of the diagnosis for each specimen, as well as for each corresponding biopsy, was performed, excluding two cases that were unavailable for subsequent review. Specimens were collected using the Accelon Combi Cervical Biosampler (Medscand USA, Hollywood, FL) cervical brush system, immediately after collection of the routine cervical smear and before the performance of colposcopic examination and cervical biopsy. The samples were transferred into sterile collection tubes containing 2.25 mL of wash buffer (10 mmol Hepes-KOH [pH 7.5], 1 mmol MgCl2, 10 mmol KCl, 1 mmol dithiothreitol) and stored on ice or at 4°C until delivery to the laboratory for preliminary characterization and processing. After releasing the cells from the sampling device by gentle vortexing, the squamous epithelial component of each sample was counted using a hemocytometer, and specimens that contain less than 6 ⫻ 104 squamous epithelial cells were excluded from further study. The cells were collected by centrifugation at 10,000 xg at 4°C for 10 minutes. The supernatant buffer was removed, and the cells were lysed in 60 ␮L of 0.5% 3-[(3-chloamidopropyl)-dimethyl-ammonio]-1-propanesulfonate (CHAPS) lysis buffer. The residual cell pellets were collected by centrifugation for subsequent HPV DNA studies. Adequate cellularity for inclusion in the study was obtained in 365 of 456 (80%) of cases. Amplifiable DNA, as determined by the detection of ␤-globin DNA on polyacrylamide gel electrophoresis of PCR amplification products, was confirmed in 338 of 365 (92.6%) of specimens.

Analysis of Telomerase Expression Telomerase activity was detected through a modification of the original protocol described by Kim et al,10 using a commercially available system (TRAPeze Telomerase Detection Kit; Intergen, Purchase, NY). The TS primer (5'-AATCCGTCGAGCAGAGTT-3') was end-labeled by 0.5 U polynucleotide kinase (Life Technologies, Baltimore, MD), using gamma [32P]dATP (10 ␮Ci/␮L, 3,000 Ci/mmol). The 32P-TS primer was incubated with 5 ␮L aliquots of cellular lysate standardized to a maximum of 5.0 ␮g protein/reaction in a reaction buffer containing 20 mmol Tris-HCl (pH 8.3), 1.5 mmol MgCl2, 63 mmol KCl, 0.05% Tween 20, 1.0 mmol

397

HUMAN PATHOLOGY

Volume 35, No. 4 (April 2004)

EGTA, 1 ␮L 50 ␮mol deoxynucleoside triphosphates, 1 ␮L telomeric repeat amplification protocol (TRAP) primer mix (RP, K1, and TSK1 template), 5.0 ␮g bovine serum albumin, and 2 U Taq DNA polymerase (Life Technologies, Baltimore, MD), in a total volume of 50 ␮L. TS template extension was performed at 30°C for 30 minutes, followed by 30 cycles of PCR with a thermal profile of 94°C for 30 seconds and 60°C for 30 seconds. An internal standard (ITAS) was amplified by both the TS primer and by its own specific return primer (K1). Lysates from HeLa cells (American Type Culture Collection, Rockville, MD) at 103, 102, and 10 cell equivalents, were used as quantitation controls in each assay. Telomerasepositive test results were confirmed by repeat TRAP assay, with heat pretreatment (85°C for 10 minutes) of cellular lysates to monitor for false-positive results. Lysis buffer reagent controls were included in each to monitor for the possibility of reagent contamination. The TRAP extension and ITAS products were detected by autoradiography, after electrophoresis on a nondenaturing 12% polyacrylamide gel. Samples that produced the characteristic 6-bp telomerase product ladder were scored as positive for telomerase activity. Samples that exhibited an anomalous band pattern were further evaluated by repeat analysis with heat denaturation control and by TRAP enzyme-linked immunosorbent assay (Telo TAGGG Telomerase PCR ELISAplus; Roche Molecular Biochemicals, Indianapolis, IN). Samples that exhibited amplification of ITAS but were lacking the 6-bp TRAP ladder were scored as negative for telomerase. PCR amplification products were quantitated using a phosphorimager (Phosphorimager SI, Molecular Dynamics, Sunnyvale, CA) and Image Quant for Windows NT (Molecular Dynamics), after exposures of 45 minutes and 16 hours.

Detection of HPV DNA Evaluation for HPV DNA in the residual cell pellets was done using consensus sequence PCR followed by polyacrylamide gel electrophoresis and also by filter hybridization linear array genotyping,21,22 using reagents provided by Roche Molecular Systems (Alameda, CA). The filter hybridization genotyping method is capable of amplifying and identifying a wide range of HPVs, including low-risk or uncharacterized types 6, 11, 40, 42, 54, 55, 57, MM4, MM7, MM8, and MM9 and high-risk types 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, and 68.23 Cell pellets were digested overnight with proteinase K (0.2 mg/mL) at 37°C. HPV DNA was then amplified using HPV L1 consensus sequence primers (PMY09/PMY11) following modifications of the original protocol, described by Ting and Manos.24,25 PCR was done in a 50 ␮L reaction mixture containing 5 ␮L of 10 ⫻ Buffer II (Roche Molecular Systems), 4 mmol MgCl2, 20 mmol of each deoxynucleoside triphosphate (60 mmol of dUTP), 12.5 pmol of each biotin-labeled HPV primer pool (PGMY09/ PGMY11), and 3.75 U of Taq polymerase. A total of 40 cycles were performed with a thermal profile of 95°C for 1 minute, 55°C for 1 minute, and 72°C for 1 minute. ␤-globin primers (B_PC04 and B_GH20) were included in each amplification to evaluate the presence of amplifiable DNA. HPV-positive cases were identified by the detection of a discrete band that migrated at about 450 bp, and ␤-globin amplification was confirmed by the detection of a band at about 270 bp on polyacrylamide gel electrophoresis. Positive test results were classified as HPV, type not specified (TNS). The amplified products from matched PCR reactions were denatured with 50 ␮L of 1.6% w/w NaOH, followed by filter strip hybridization at 53°C for 30 minutes (hybridization solution contains 4 ⫻ SSPE (1 ⫻ SSPE is 0.18 mol NaCl, 10 mmol NaPO4, and 1

FIGURE 1. Analysis of telomerase expression in cervical cytology samples. Serial dilutions of HeLa cells, 10, 102, and 103 cell equivalents; lanes A, B, C, and D, HSIL; lanes E, F, G, and H, LSIL. Arrow, internal telomerase assay standard. Phosphorimager analysis. Lanes A, C, and E reveal a progressive 6-bp ladder, consistent with telomerase activity. Heat pretreatment of cell lysates (⫺) blocked telomerase extension of the TS primers. Lanes B, D, F, G, and H show nonspecific heat-resistant amplification products that are not diagnostic of telomerase activity.

mmol EDTA [pH 7.7]) and 1.0% sodium dodecyl sulfate). The filter strips were then washed with wash buffer for 15 minutes at 53°C, then incubated for 30 minutes at room temperature in a 0.3% streptavidin (SA)– horseradish peroxidase conjugate solution. After three 10-minute washes at room temperature, 3 mL of color development solution (Roche Molecular Systems) was applied to each strip and incubated on a shaking platform for 2 to 4 minutes at room temperature. Samples were evaluated by comparing the location of the color precipitate with the known location of the HPV probe sequences on the filter strip. Cases that demonstrated a signal for either a single specific high-risk HPV type or multiple HPVs, including at least one high-risk HPV type by filter hybridization, were scored as high-risk HPV positive.

Statistical Analysis Fisher’s exact test or the ␹2 test was used, where appropriate, to examine for a potential association between study variables of interest.26,27

RESULTS Cases demonstrating amplification of a progressive 6-bp ladder that comigrated with the primary amplification products from the HeLa cell internal standards were defined as telomerase positive, and cases with PCR products that did not show the characteristic 6-bp pattern of amplification were scored as telomerase negative (Fig 1). The intensity of telomerase activity was estimated by comparing the telomerase/ITAS ratio from positive cases with the HeLa cell control curve. In positive cases, the intensity of the signal ranged from 100 to fewer than 10 HeLa cell equivalents. Telomerase activity was confirmed by repeat TRAP assay of positive cases, with and without heat pretreatment (85°C for 10

398

DIAGNOSIS OF CERVICAL DYSPLASIA AND CARCINOMA (Jarboe et al)

FIGURE 2. HPV DNA filter hybridization genotyping. Genomic DNA was extracted from residual cell pellets by proteinase K digestions, after CHAPS lysis for telomerase analysis. PCR was performed using biotinylated L1 consensus sequence primers over 40 cycles, and the products were evaluated by filter hybridization genotyping strips that contain immobilized probes for a broad range of HPV types (Roche Molecular Systems). Lane A, types 18 and MM9; lane B, type 16; lane C, type 58; lane D, HPV negative; lane E, type 16; lane F, HPV negative; lane G, type 39; lane H, HPV negative; lane I, types 16 and 18; lane J, types 53 and 59; lane K, type 58. The arrow indicates ␤-globin high-level and low-level internal amplification controls.

minutes) of the cellular extracts, as an internal control to monitor for the possibility of PCR amplicon sample contamination or nonspecific amplification. Cases that exhibited amplification of a 450-bp band by polyacrylimide gel electrophoresis (PAGE) were scored as positive for HPV, TNS (Fig 2). A broad range of HPVs was detected by filter hybridization genotyping (Fig 3). In most cases, filter hybridization demonstrated a single HPV signal, but hybridization with multiple HPV probes was detected in some specimens. These cases were interpreted as evidence of infection by multiple HPV types, rather than as a reflection of nonspecific hybridization. Telomerase and HPV test results were compared with the colposcopic Pap test diagnosis in 338 specimens (Table 1). Telomerase activity was detected in 35 of 338 (10.4%) of informative cases, including 8 of 97 (8.2%) of cases with normal cytology or reactive cellular changes (RCCs), 7 of 98 (7.1%) of ASCUS, 3 of 95 (3.2%) of LSIL, and 17 of 48 (35.4%) of cases with a cytological diagnosis of HSIL. HPV, TNS was detected in 220 of 338 (65.1%) of all cases, including 35 of 97 (36.1%) specimens with normal/RCC cytology, 54 of 98 (55.1%) cases of ASCUS, 86 of 95 (90.5%) cases of LSIL, and 45 of 48 (93.8%) cases of HSIL (Table 1).

FIGURE 3. Cytological correlation with telomerase and HPV test results. (A) Benign reactive cellular changes, telomerase negative, HPV negative; biopsy demonstrated normal cervical mucosa. (B) ASCUS Pap, telomerase negative, HPV negative; biopsy exhibited only atypical squamous metaplasia. (C) LSIL Pap, telomerase negative, high-risk HPV positive; biopsy demonstrated CIN I. (D) HSIL Pap, telomerase positive, high-risk HPV positive; biopsy revealed CIN III.

399

HUMAN PATHOLOGY

Volume 35, No. 4 (April 2004)

TABLE 1. Cytology Correlation (n ⫽ 338) Colposcopic Pap diagnosis Normal/RCC ASCUS LSIL HSIL Total

HPV-positive TNS*

High-risk HPV positive†

TABLE 3. Correlation of Telomerase with High-Risk HPV

Telomerase positive

35/97 (36.1%) 23/97 (23.7%) 8/97 (8.2%) 54/98 (55.1%) 28/98 (28.6%) 7/98 (7.1%) 86/95 (90.5%) 69/95 (72.6%) 3/95 (3.2%) 45/48 (93.8%) 35/48 (72.9%) 17/48 (35.4%) 220/338 (65.1%) 155/338 (45.9%) 35/338 (10.4%)

*TNS, type not specified. †High-risk HPVs 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, and 68.

High-risk HPVs were detected by filter hybridization genotyping in 155 of 338 (45.9%) of informative cases, including 23 of 97 (23.7%) cases with normal/RCC cytology, 28 of 98 (28.6%) cases of ASCUS, 69 of 95 (72.6%) cases of LSIL, and 35 of 48 (72.9%) of cases of HSIL (Table 1). The differences in the proportion of cases positive for telomerase, HPV TNS, or high-risk HPVs in the detection of HSIL were all statistically significant (P ⬍0.001). Coinfection by multiple HPV types was detected by filter hybridization genotyping in 58 of 220 (26.4%) cases that were HPV positive by PAGE. The most common HPV detected was type 16, which was identified in 53 of 220 (24.1%) of all HPV-positive cases and in 53 of 338 (15.7%) of all cases (normal/RCC, ASCUS, LSIL, and HSIL). Other prevalent HPV types included type 53 (21 cases), type 52 (20 cases), type 66 (16 cases), type 18 (14 cases), type 59 (14 cases) and type 45 (13 cases). HPV types 6 and 11 were each identified in only 3 cases, and the other HPV types were found in 1 to 12 cases. Cervical biopsies were performed in 299 of 338 cases with a colposcopic cytological diagnosis of normal/RCC, ASCUS, LSIL, or HSIL (Table 2). In the remaining cases, a lesion was not identified at colposcopic examination or insufficient squamous mucosa was present to make a histological diagnosis. Telomerase activity was detected in 35 of 299 (11.7%) of cases that had colposcopic biopsies, including 11 of 139 (7.9%) of cases with biopsies that were negative for cervical intraepithelial neoplasia (CIN), 4 of 93 (4.3%) of cases with a biopsy diagnosis of CIN I, and 20 of 67 (29.9%) of cases with biopsy-confirmed CIN II/III. High-risk HPVs were detected in 134 of 299 (44.8%) of cases with colposcopic biopsies, including 39 of 139 (28.1%) of cases that were negative for dysplasia, 48 of 93 (51.6%) of cases with a biopsy diagnosis of CIN I, and 46 of 67 (70.1%) of cases that had biopsy confirmed CIN II/III. The differences in the proportion of

Any cytological diagnosis (n ⫽ 338) Telomerase positive Telomerase negative Total HSIL cytological diagnosis (n ⫽ 47) Telomerase positive Telomerase negative Total

High-risk HPV positive

High-risk HPV negative

Total

19 136 155

16 167 183

35 303 338

12 23 35

5 7 12

17 30 47

cases that were positive for telomerase, HPV TNS, or high-risk HPVs in the detection of CIN II/III were each significant (P ⬍0.001). The corresponding cervical biopsy specimens were reviewed in cases with normal/RCC or ASCUS cytology that tested positive for telomerase. CIN II/III was diagnosed on concurrent cervical biopsy specimens from 8 of 15 (53.3%) cases, including 3 of 8 specimens with normal/RCC cytology and 5 of 7 cases of ASCUS. Among the remaining cases, CIN I was diagnosed in 3 of 15 (20%) cases, including 2 of 8 normal/RCC specimens and 1 of 7 cases of ASCUS. The corresponding cervical biopsy specimens were similarly evaluated in cases with normal/RCC or ASCUS cytology that tested positive for high-risk HPVs. CIN II/III was diagnosed on concurrent cervical biopsy specimens from 9 of 51 (17.6%) cases, including 5 of 23 specimens with normal/RCC cytology and 4 of 28 cases of ASCUS. Among the remaining cases, CIN I was diagnosed in 13 of 51 (25.5%) cases, including 5 of 23 normal/RCC specimens and 8 of 28 cases of ASCUS. The results of telomerase detection were compared with the detection of HPVs (Table 3). Among all cases entered into the study (n ⫽ 338), telomerase activity did not correlate with the detection of either HPV TNS or high-risk HPVs (P ⬎0.05). Furthermore, no correlation was observed between the detection of telomerase and either HPV TNS or high-risk HPVs in the subsets of cases with cytological diagnoses of normal/RCC, ASCUS, LSIL, or HSIL (P ⬎0.05). DISCUSSION This study has demonstrated that the detection of telomerase in cytological specimens is a moderately

TABLE 2. Biopsy Correlation (n ⫽ 299) Biopsy diagnosis

HPV-positive TNS

High-risk HPV positive

Telomerase positive

Negative for dysplasia CIN I CIN II/III Total

67/139 (48.2%) 65/93 (69.9%) 57/67 (85.1%) 189/299 (63.2%)

39/139 (28.1%) 48/93 (51.6%) 47/67 (70.1%) 134/299 (44.8%)

11/139 (7.9%) 4/93 (4.3%) 20/67 (29.9%) 35/299 (11.7%)

400

DIAGNOSIS OF CERVICAL DYSPLASIA AND CARCINOMA (Jarboe et al)

sensitive, but not specific, marker of HSIL or underlying CIN II/III on concurrent cervical biopsy. Although telomerase was detected in more than one-third of cases that had a cytological diagnosis of HSIL, telomerase activity was also detected in more than 6% of specimens that were negative for high-grade cytological abnormalities. In contrast, the detection of HPV TNS or high-risk HPVs were much more sensitive but relatively nonspecific for HSIL or underlying CIN II/III on concurrent cervical biopsy. Telomerase expression was not significantly correlated with the detection of HPV TNS or high-risk HPVs in either the total study population or cases within specific cytological diagnostic categories. Although telomerase was generally less sensitive than HPV as a marker of CIN II/III, more than half of all cases with normal/RCC or ASCUS cytology that tested positive for telomerase had CIN II/III on concurrent cervical biopsy. HPV testing has been extensively evaluated as a diagnostic adjunct to identify patients with equivocal or low-grade cytological abnormalities that are at highest risk for underlying high-grade dysplasia. Data from the ALTS (ASCUS/LSIL Triage Study) trial demonstrated that HPV testing in cytology specimens was useful in the triage of patients with ASCUS but not those with LSIL.6-8 Although ASCUS Paps that were negative for high-risk HPVs rarely had underlying CIN II/III on subsequent cervical biopsy, a positive test result was not predictive of underlying CIN II/III in patients with a cytological diagnosis of LSIL.6-8 Our current study found that the detection of high-risk HPVs had a specificity for biopsy-confirmed CIN II/III of 62.5%, similar to the specificity of 53% calculated for HPV detection as a marker of underlying CIN III in the ALTS trial.28 Thus, although HPV DNA testing can be used to identify patients at low risk for clinically significant lesions of the cervical mucosa in ASCUS cytology, its clinical utility in other diagnostic cytological categories is limited by relatively low specificity. Both the current study and the ALTS study involved the analysis of cervical cytology specimens classified according to the original 1988 Bethesda system of cytological classification, before implementation of the revised 2001 Bethesda system. Some cases of ASCUS from the current study were subclassified as “ASCUS, favor reactive”; in the Bethesda 2001 system, these cases would have been classified as “reactive cellular changes.” Furthermore, in the Bethesda 2001 system, cases of ASCUS with features suspicious of but not diagnostic for HSIL would have been placed in a separate diagnostic category, ASC-H. Thus the current study did not evaluate the specific association of telomerase or HPV with ASCUS versus ASC-H diagnostic subcategories. Previous studies have evaluated the clinical utility of telomerase expression in both cervical biopsy specimens and cervical cytology specimens.19,20,29-37 Work from our laboratory demonstrated that telomerase can be detected in almost all cases of CIN high-grade squamous dysplasia or cervical SCC.29 In contrast, the proportion of cases of HSIL that tested positive for telom-

erase was much lower in the current study of cervical cytology specimens. The basis for the apparent discrepancy between the high level of detection in cervical biopsy specimens but relatively low sensitivity in cervical cytology specimens may reflect false-negative test results. Thus, although almost all cases of CIN II/III are likely to express telomerase, most HSIL cytology cases produce false-negative telomerase test results. The application of the TRAP assay to cervical cytology specimens is subject to numerous theoretical limitations, including variability in the total number and proportion of HSIL cells present and variability in the relative proportion of benign cellular elements, inflammatory cells, red blood cells, mucin, and microbiological agents. Although the current study attempted to standardize the total number of squamous epithelial cells that were included in each assay, there was no standardization of the number of endocervical cells or other cellular elements. There was also variability (from 1 hour to up to 24 hours) in the time that elapsed between the collection of the cytological specimens and initial processing in the research laboratory. However, all Of the cases were uniformly collected into a standardized wash buffer and stored on ice or at 4°C until delivery to the laboratory. Pilot studies in our laboratory demonstrated that telomerase activity in clinical samples stored under these conditions is stable for at least 5 days (unpublished observations). Thus telomerase degradation is an unlikely source of false-negative test results in the current study. Although telomerase can be detected in most cases of cervical carcinoma, the level of telomerase expression at the cellular level has not been defined in dysplastic cells. It is likely, however, that telomerase is a low abundance protein in light of data, which indicated that on average, there is less than one full-length hTERT mRNA molecule per cell in a wide range of primary cell lines.38 Consistent with the hypothesis that telomerase is present in relatively low abundance, relatively weak signals for the telomerase hTERT protein and hTR RNA subunit were detected in histological sections of cervical dysplasia and carcinoma.39 Thus false-negative test results in the current study may reflect a relatively low level of expression in dysplastic cells at the cellular level, rather than at the tissue level. In summary, this study has shown that telomerase is unlikely to play a useful role as a diagnostic adjunct in cervical cytopathology. Despite the fact that a high proportion of normal/RCC or ASCUS cases with telomerase-positive test results had underlying high-grade dysplasia on cervical biopsy, the TRAP assay demonstrated an overall low sensitivity for the detection of clinically significant lesions. Although telomerase is expressed in most cases of high-grade cervical dysplasia and carcinoma, technical and practical limitations result in a high rate of false-negative test results in cervical cytology specimens. Future technical refinements in telomerase assay methods for cytological specimens could theoretically facilitate the use of telomerase as a diagnostic marker of cervical dysplasia; regardless, however, the low abundance of telomerase expression in

401

HUMAN PATHOLOGY

Volume 35, No. 4 (April 2004)

dysplastic cells may fundamentally limit the practical utility of telomerase as a cytological marker to use in the triage of cases with low-grade abnormalities or to decrease the false-negative rate in the detection of highgrade dysplastic lesions. Acknowledgments. The authors thank Dr. Janet Kornegay of Roche Molecular Systems, Alameda, CA, for kindly providing the reagents for HPV DNA filter hybridization genotyping. The authors appreciate the support of Dr. A. L. Shroyer, who performed the statistical analysis of the data.

REFERENCES 1. American Cancer Society: Cancer Facts & Figures. Atlanta, GA, American Cancer Society, 2003 2. Wright T, Kurman R, Ferenczy A: Carcinoma and other tumors of the cervix, in Kurman R (ed): Blaustein’s Pathology of the Female Genital Tract (ed 4). New York, Springer-Verlag, 1994, pp 287-289 3. zur Hausen H: Papillomavirus infections: A major cause of human cancers. Biochem Biophys Acta 1288:F55-78, 1996 4. Wolf JK, Ramiez PT: The molecular biology of cervical cancer. Cancer Invest 19:621-629, 2001 5. zur Hausen H: Papillomaviruses in human cancers. Proc Assoc Am Physicians 111:581-587, 1999 6. ALTS Group: Human papillomavirus testing for triage of women with cytologic evidence of low-grade squamous intraepithelial lesions: Baseline data from a randomized trial. J Natl Cancer Inst 92:397-402, 2000 7. Solomon D, Schiffman M, Tarone R, & for the ALTS Group: Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: Baseline results from a randomized trial. J Natl Cancer Inst 93:293-299, 2001 8. Herbst AL, Pickett KE, Folleen M, et al: The management of ASCUS cervical cytologic abnormalities and HPV testing: A cautionary note. Obstet Gynecol 98:849-851, 2001 9. Feng J, Funk WD, Wang S-S, et al: The RNA component of human telomerase. Science 269:1236-1241, 1995 10. Kim NW, Piatyszek MA, Prowse KR, et al: Specific association of human telomerase activity with immortal cells and cancer. Science 266:2011-2015, 1994 11. Nakamura TM, Morin GB, Chapman KB, et al: Telomerase catalytic subunit homologs from fission yeast and human. Science 277:955-959, 1997 12. Meyerson M, Counter CM, Eaton EN, et al: hest2, the putative human telomerase catalytic subunit gene, is upregulated in tumor cells and during immortalization. Cell 90:785-795, 1997 13. Jarboe EA, Liaw K-L, Thompson C, et al: Analysis of telomerase as a diagnostic biomarker of cervical dysplasia and carcinoma. Oncogene 21:664-673, 2002 14. Oh ST, Kyo S, Laimins LA: Telomerase activation by human papillomavirus type 16 E6 protein: Induction of human telomerase reverse transcriptase expression through Myc and GC-rich Sp1 binding sites. J Virol 75:5559-5566, 2001 15. Klingelhutz AJ, Foster AJ, McDougall JK: Telomerase activation by the E6 gene product of human papillomavirus type 16. Nature 380:79-82, 1996 16. Stoppler H, Hartman DP, Sherman L, et al: The human papillomavirus type 16 E6 and E7 oncoproteins dissociate cellular telomerase activity from the maintenance of telomere length. J Biol Chem 272:13332-13337, 1997 17. Anderson S, Shera K, Ihle J, et al: Telomerase activation in cervical cancer. Am J Pathol 151:25-31, 1997

18. DeFillippis R, Goodwin E, Lingling W, et al: Endogenous human papillomavirus E6 and E7 proteins differentially regulate proliferation, senescence, and apoptosis in HeLa cervical carcinoma cells. J Virol 77:1551-1563, 2003 19. Yashima K, Ashfaq R, Nowak J, et al: Telomerase activity and expression of its RNA component in cervical lesions. Cancer 82:13191327, 1998 20. Riethdorf S, Riethdorf L, Schulz G, et al: Relationship between telomerase activation and HPV 16/18 oncogene expression in squamous intraepithelial lesions and squamous cell carcinomas of the uterine cervix. Int J Gynecol 20:177-185, 2001 21. Gravitt P, Peyton C, Apple R, et al: Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by a singlehybridization, reverse line blot detection method. J Clin Microbiol 36:3020-3027, 1998 22. Gravitt P, Peyton C, Alessi T, et al: Improved amplification of genital human papillomaviruses. J Clin Microbiol 38:357-361, 2000 23. Munoz N, Bosch X, de Sanjose S, et al: Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 348:518-527, 2003 24. Ting Y, Manos MM: Detection and typing of genital human papillomaviruses, in Innis MA (ed): PCR Protocols: A Guide to Methods and Applications. Orlando, FL, Academic Press, 1990, pp 356-367 25. Resnick RM, Cornelissen MTE, Wright DK, et al: Detection and typing of human papillomavirus in archival cervical cancer specimens by DNA amplification with consensus primers. J Natl Cancer Inst 82:1477-1484, 1990 26. Rosner B: Fundamentals of Biostatistics, ed 3. Belmont, CA, Duxbury Press, 1990, pp 336-341 27. Rosner B: Fundamentals of Biostatistics, ed 3. Belmont, CA, Duxbury Press, 1990, pp 325-332 28. Harris JE: Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: Baseline results from a randomized trial. J Natl Cancer Inst 93:950, 1990 29. Shroyer KR, Thompson C, Enomoto T, et al: Telomerase expression in normal epithelium, dysplasia and squamous cell carcinoma of uterine cervix. Am J Clin Pathol 109:153-162, 1998 30. Wisman GB, Hollema H, de Jong S, et al: Telomerase activity as a biomarker for (pre)neoplastic cervical disease in scrapings and frozen sections from patients with abnormal cervical smear. J Clin Oncol 16:2238-2245, 1998 31. Yokoyama Y, Takahashi Y, Shinohara A, et al: Telomerase activity in the female reproductive tract and neoplasms. Gynecol Oncol 68:145-149, 1998 32. Zhang DK, Ngan HY, Cheng AN, et al: Clinical significance of telomerase activation and telomeric restriction fragment (TRF) in cervical cancer. Eur J Cancer 35:154-160, 1999 33. Gorham H, Yoshida K, Sugino T, et al: Telomerase activity in human gynecological malignancies. J Clin Pathol 50:501-504, 1997 34. Zheng P-S, Iwasaka T, Yokoyama M, et al: Telomerase activation in in vitro and in vivo cervical carcinogenesis. Gynecol Oncol 66:222-226, 1997 35. Reddy VG, Khann N, Jain SK, et al: Telomerase: A molecular marker for cervical cancer screening. Int J Cancer 11:100-106, 2001 36. Ngan H, Cheung A, Liu S, et al: Telomerase assay and HPV 16/18 typing as adjunct to conventional cytological cervical cancer screening. Tumour Biol 23:87-92, 2002 37. Reesink-Peters N, Helder M, Wisman G, et al: Detection of telomerase, its components, and human papillomavirus in cervical scrapings as a tool for triage in women with cervical dysplasia. J Clin Pathol 56:31-35, 2003 38. Yi X, Shay JW, Wright WE: Quantitation of telomerase components and hTERT mRNA splicing patterns in immortal human cells. Nucleic Acids Res 29:4818-4825, 2003 39. Frost M, Bobak J, Gianani R, et al: Localization of telomerase hTERT protein and hTR in benign mucosa, dysplasia, and squamous cell carcinoma of the cervix. Am J Clin Pathol 114:726-734, 2000

402