| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Departments of Family Medicine, and Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada; Father Sean O’Sullivan Research Centre, St. Joseph’s Hospital, Hamilton, Ontario, Canada; and Program for Appropriate Technology in Health, Seattle, Washington.
Address reprint requests to: John Sellors, MD, MSc, Program for Appropriate Technology in Health, 1455 NW Leary Way, Seattle, WA 98107-5136; E-mail: jsellors{at}path.org.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Women with abnormal cytology were referred for colposcopy, and a cervical swab or brush specimen was obtained for human papillomavirus testing. Sensitivities, specificities, and likelihood ratios of different relative light unit ratio cutoffs were calculated using a reference standard of colposcopy or biopsy of either CIN II+ (CIN II, III, or carcinoma), or CIN I+ (CIN I, CIN II+). The receiver operating characteristic curve was used to estimate optimal test-positive cutoff points for the hybrid capture II test.
RESULTS: CIN II+ was found in 18.7% (98 of 524) and CIN I in 10.5% (55 of 524) of the women. The optimal relative light unit ratio was 15.56, giving a sensitivity and specificity of 82.7% and 73.2% for CIN II+, and 74.2% and 77.8% for CIN I+. In a stratified analysis, a higher relative light unit cutoff (15.19) optimized sensitivity and specificity for CIN II+ (sensitivity 81.8%, specificity 51.5%) for women with low-grade squamous intraepithelial lesions cytology, whereas the optimal cutoff was 2.36 (sensitivity 100%, specificity 73.0%) for women with atypical squamous cells of undetermined significance, yielding referral rates of 53.3% and 28.7%, respectively. Use of a swab or brush was not associated with the level of human papillomavirus detected (P > .05).
CONCLUSION: Use of a higher cutoff for the relative light unit ratio (higher viral load) of the hybrid capture II test may improve the management of women, especially those with low-grade squamous intraepithelial lesions cytology.
Cervical infection with specific types of human papillomavirus (HPV) has been established as the main cause of cervical cancer and cervical intraepithelial neoplasia (CIN).1,2 Infection with carcinogenic types of HPV is common among women, with a population-based prevalence as high as 24% in 20–24 year olds in developed3 and developing4,5 countries and an incidence of up to 28% per year in young women.6,7 The HPV infection is often transiently detectable,8 especially among young women,9 and a small proportion of infected women will develop cervical carcinoma without preventive screening.10
Molecular techniques to detect carcinogenic HPV in cervical samples have recently been of interest for improving the detection of CIN and cervical cancer. Hybrid capture technology (Digene Corp., Gaithersburg, MD) detects high-risk types of HPV deoxyribonucleic acid (DNA) using signal amplification.11 The hybrid capture II assay is a second-generation semiquantitative assay that detects the presence of any of 13 carcinogenic HPV types as a group. The reaction is chemiluminescent, and the result is read by a luminometer. Positive specimens are those with relative light units equal to or greater than the mean of positive control specimens containing 1.0 pg/mL of HPV DNA, ie, a relative light unit/positive control ratio of 1.0 or greater. Although it is recognized that the relative light unit emitted by the hybrid capture II test is affected by the volume of material in a specimen, the relative light unit is proportional to the amount of DNA in the specimen,11 and hence is an estimate of viral load.
Although the manufacturer’s suggested relative light unit/positive control positive cutoff of greater than or equal to 1.0 at an HPV DNA detection threshold of 1.0 pg/mL provides good sensitivity for detecting histologically confirmed high-grade CIN among women with previous atypical squamous cells of undetermined significance (ASCUS) or low- grade squamous intraepithelial lesions (LSIL), it may result in relatively low specificity.12–16 The background prevalence of HPV positivity is relatively high among women with low-grade cytologic abnormalities (ASCUS, LSIL), most of whom do not have high-grade lesions on colposcopy and biopsy, thereby reducing the specificity of the HPV test for high-grade CIN.13,17
The HPV viral load has been found to be associated with high-grade CIN and squamous cell carcinoma and thus may be a useful triage method for women with low-grade cytologic abnormalities. Using the hybrid capture II assay, Nindl et al found that the relative light unit/positive control ratio was significantly higher in high-grade compared with low-grade CIN among women referred to colposcopy with abnormal cytology.18 Studies using polymerase chain reaction (PCR) detection of HPV DNA have also found similar associations.19–22
This study used cross-sectional data to examine the diagnostic performance of the hybrid capture II test if various relative light unit/positive control ratios were chosen as the positive cutoff for detecting the presence of histologically confirmed CIN in women with a spectrum of previous cytologic abnormalities before colposcopic examination.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the two trials, women were randomized to different management strategies. In one study, women were randomized to immediate HPV testing followed by colposcopy, or to repeat cytology in 6 months followed by colposcopy. In the second trial, women were randomized to repeat cytology and HPV testing (repeat Papanicolaou + HPV group) every 6 months for 2 years, or repeat cytology alone (repeat Papanicolaou group) every 6 months for 2 years. The end-point assessment was colposcopy for both trials. If high-grade squamous intra-epithelial lesions (HSIL), atypical glandular cells of undetermined significance, or carcinoma was found on any repeat cytology, the woman was referred immediately to colposcopy. If HSIL, atypical glandular cells of undetermined significance, carcinoma, or a positive HPV result was found in the repeat Papanicolaou + HPV group, the woman was referred immediately to colposcopy. Women in the two studies who were randomized to receive only repeat cytology management were also tested for HPV, as per the trial protocols, but only the cytology result and not the HPV result was used to direct the decision to refer to colposcopy.
The cervical specimen for HPV testing was taken by the patients’ family physician for those women in the two trials and by the colposcopy physician immediately before colposcopy for those women in the cross-sectional study. The area of the cervical os and the transformation zone was sampled with a Dacron swab for women in the two trials or a soft cervical brush for women in the cross-sectional study (Cervical Sampler, Digene Corp.), and placed into specimen transport medium (Digene Corp.) for hybrid capture II HPV testing. A cervical smear was obtained with a modified Ayre spatula and endocervical brush, as required, and immediately fixed for cytologic examination. Colposcopy was performed, and directed biopsy or endocervical curettage was performed as required.
Cervical cytology and histology specimens were read by the pathology department of the Henderson Hospital (Hamilton). For the histology specimens, blinded review by an external gynecologic pathologist was performed. Cases of disagreement were resolved by consensus with a second gynecologic pathologist. If multiple biopsies were taken, the highest grade of lesion diagnosed was used as the reference standard for hybrid capture II test performance assessment.
The hybrid capture II assay is a second-generation DNA probe test based on signal amplification. The procedure has been described previously.11 Hybrid capture II testing was performed at the Hamilton Regional Virology and Chlamydiology Laboratory for the women in the two trials and by Digene Corp. for the women in the cross-sectional study. Both testing sites were blinded to all other results. The DNA was denatured and hybridized with a cocktail of 13 ribonucleic acid probes to carcinogenic HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68). A chemiluminescent substrate was added, and the relative light units were measured in a luminometer. Because the relative light unit ratio is the ratio of the light emitted by the specimen to the light emitted from the mean relative light unit of triplicate positive control specimens containing 1 pg/mL of HPV DNA (5000 copies of HPV genome), it is a semiquantitative estimate of viral load in the specimen.
Two types of outcomes for abnormal cervical histology (high grade and any grade) were used in the assessment of the hybrid capture II assay. The CIN II+ outcome was defined as a histologic diagnosis of CIN II or III or squamous cell carcinoma. The CIN I+ outcome was defined as a histologic diagnosis of any grade of CIN or squamous cell carcinoma (CIN I+).
The HPV test result used in the analysis was the one taken by the family physician at the visit just before colposcopy referral (either at the last scheduled study visit if repeat cytology results were normal, ASCUS, or LSIL consistently, or immediately if HSIL, atypical glandular cells of undetermined significance, or carcinoma was found on repeat cytology) for the women in the two trials. For women in the cross-sectional study, only one HPV test was done, at the time of colposcopy. The cytology result used in the stratified receiver operating characteristic (ROC) curve analysis was the one taken at the time of colposcopy for all women.
The ROC curves and the area under the curves with 95% confidence intervals (CI) were calculated, using nonparametric methods for the calculation of standard error (SPSS 10.0.5, SPSS Inc., Chicago, IL). The ROC curve plots 1-specificity (false-positive rate) on the x-axis, versus sensitivity (true-positive rate) on the y-axis, using various relative light unit ratios as the putative cutoff for a positive test. This method allows a visual comparison of the relationship between sensitivity and specificity for the different cutoffs of relative light unit ratio and correspondingly viral load. Separate curves were also plotted for women 30 years old and younger and women older than 30 years and for women with ASCUS versus LSIL on cytology. The sensitivities and specificities for older women versus younger women and for women with ASCUS versus LSIL were compared using the normal approximation to the binomial distribution.23
An estimated optimal viral load cutoff was determined by finding the relative light unit ratio cutoff at which Youden’s index24 (sensitivity + specificity - 1) was maximized. As the value of the Youden index approaches one, this reflects improved diagnostic accuracy. A test with perfect diagnostic accuracy would have a Youden index of 1.0.
Logistic regression models were derived (SPSS 10.0.5) for both outcomes (CIN II+, CIN I+) using categoric relative light unit ratio data and adjusting for potential confounders. The relative light unit ratios were categorized, using a cutoff of 1.0 (a positive test as defined by the manufacturer) and then dividing the ratios equal to or greater than 1.0 into four equally prevalent categories of 1.0–9.37, 9.38–64.96, 64.97–281.20, and 281.21 and higher. Adjustment was made for the potentially confounding effects of age, marital status, smoking status, number of sexual partners in the past year, and use of oral contraceptives. Smoking status was dichotomized in the analysis, with current smokers compared with women who had never smoked or had smoked but quit. Number of lifetime sexual partners was a continuous variable.
The association between HPV results and sampling method (cervical swab versus brush) was also examined using the 
2 test for trend.23 Statistical significance was defined as a type I error rate (
) of less than 0.05 (two tailed).
The likelihood ratio or the odds that a given test result or relative light unit ratio will be found in a woman with disease as opposed to without is a useful way to express test performance because it can be compared across different settings and is unaffected by the prevalence of a condition.25 Likelihood ratios for the optimal relative light unit ratios and their 95% CIs were also calculated.26
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
shows the demographic characteristics of the women in the three studies.
|
). The repeat cytology results and the median relative light unit ratios in women with no biopsy done, normal results on colposcopy or biopsy, CIN I, and CIN II or III, or squamous cell carcinoma on biopsy are shown in Table 2.
|
. Sensitivity rose steadily as the relative light unit ratio (viral level) decreased. The increases in sensitivity were accompanied by decreases in specificity, especially when sensitivity increased to above 80%. The area under the ROC curve in Figure 1 was 0.82 (95% CI 0.78, 0.87, P < .001). The optimal cutoff occurred at a relative light unit ratio test-positive cutoff of greater than or equal to 15.56. At the manufacturer’s suggested test-positive cutoff of greater than or equal to 1.0, sensitivity was 93.9% (92 of 98) and specificity was 56.8% (242 of 426) (Table 3). At the test-positive cutoff of greater than or equal to 15.56, sensitivity was 82.7% (81 of 98), and specificity was 73.2% (312 of 426). Table 3 shows for both the 1.0 and 15.56 relative light unit ratio cutoffs, sensitivity is lower and specificity is higher for women older than 30 years compared with women 30 years and younger. The differences in specificity were statistically significant for both the 1.0 cutoff and the 15.56 cutoff (P < .001), but the differences in sensitivity were not statistically significant. In Figure 1, the curves for women older than 30 years and for younger women also indicate that for a given sensitivity, specificity is generally higher in older women.
|
|
. The area under the ROC curve was also 0.82 (95% CI 0.78, 0.86, P < .0001). The optimal cutoff also occurred at a relative light unit ratio of 15.56. The sensitivity and specificity for detection of CIN I+ for the relative light unit ratio cutoffs of 1.0 and 15.56 in all women and in the two age groups are shown in Table 3
. Sensitivity was statistically significantly higher in women younger than 30 years of age (P = .02) for the 1.0 cutoff but not for the 15.56 cutoff (P = .51). Specificity was significantly higher among the women aged 30 years and over for both the 1.0 cutoff (P = .01) and the 15.56 cutoff (P = .02).
|
. Reflecting the improved specificity for HPV testing among older women, likelihood ratios tended to be higher among the older women. Likelihood ratios were statistically significantly higher for the relative light unit ratio cutoff of 15.56, compared with the cutoff of 1.0 in older women for CIN II+ and in all women for CIN I+ (P < .05). Separate analysis of women with ASCUS versus LSIL on repeat cytology revealed that the optimal cutoff for ASCUS occurred at a relative light unit ratio of 1.12 for CIN+ (sensitivity 100%, five of five, 95% CI 46.3, 100.0; specificity 66.9%, 83 of 124, 95% CI 57.8, 75.0) and at 2.36 for CIN II+ (sensitivity 100%, three of three, 95% CI 31.0, 100.0; specificity 73.0%, 92 of 126, 95% CI 64.3, 80.4). The optimal cutoff for LSIL occurred at a relative light unit ratio of 76.37 for CIN I+ (sensitivity 61.7%, 37 of 60, 95% CI 48.2, 73.7; specificity 77.2%, 71 of 92, 95% CI 67.0, 85.0) and at 15.19 for CIN II+ (sensitivity 81.8%, 18 of 22, 95% CI 59.0, 94.0; specificity 51.5%, 67 of 130, 95% CI 42.7, 60.3). Using the recommended cutoff of 1.0 gave similar results for ASCUS for CIN I+ (sensitivity 100%, five of five, 95% CI 46.3, 100.0; specificity 65.3%, 81 of 124, 95% CI 56.2, 73.5) and for CIN II+ (sensitivity 100%, three of three, 95% CI 31.0, 100.0; specificity 64.3%, 81 of 126, 95% CI 55.2, 72.5). In contrast, at the 1.0 cutoff, sensitivity and specificity among women with LSIL were 83.3% (50 of 60, 95% CI 71.0, 91.3) and 34.8% (32 of 92, 95% CI 25.9, 46.4) for CIN+, and 90.9% (20 of 22, 95% CI 69.4, 98.4) and 30.8% (40 of 130, 95% CI 23.2, 39.6) for CIN II+. Using the 1.0 cutoff, 37.2% (48 of 129, 95% CI 29.0, 46.2) of women with ASCUS would be referred to colposcopy, compared with 72.4% (110 of 152, 95% CI 64.4, 79.2) of women with LSIL. Using the optimal cutoffs for detecting CIN I+, 35.7% (46 of 129, 95% CI 27.6, 44.6) of women with ASCUS would be referred, compared with 38.2% (81 of 152, 95% CI 30.5, 46.4) of women with LSIL. Using the optimal cutoffs for detecting CIN II+, 28.7% (46 of 129, 95% CI 21.2, 37.4) of women with ASCUS and 53.3% (81 of 152, 95% CI 45.1, 61.4) of women with LSIL would be referred.
The hybrid capture II relative light unit ratio was not associated with whether a cervical brush or a swab was used for the HPV DNA specimen collection (2 test for trend, P = .72).
Table 4 shows the results of the logistic regression analyses. The odds ratios for having CIN I+ or CIN II+, by increasing the relative light unit ratio and adjusting for demographic factors, were calculated. Adjusted odds ratios for both outcomes increased with each increasing category of relative light unit ratio (viral level).
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our results showed similarities to the baseline, cross-sectional results of the ASCUS LSIL Triage Study.16 Our finding that the optimal cutoff occurred at 15.56 was likely driven by women with LSIL on cytology. Based on their cross-sectional results, the Triage Study group found that HPV testing was not useful in women with LSIL because of the overwhelming number of colposcopy referrals, even when a higher relative light unit ratio of 10.0 was used.16,28 The results of our stratified ROC curve analysis found that HPV testing at the identified optimal cutoff among women with LSIL had a sensitivity of 81.8%, whereas referring 53.3% of women to colposcopy. In contrast, for women with ASCUS cytology, the optimal cutoff occurred at a relative light unit ratio of only slightly higher than 1.0. The manufacturer’s recommended cutoff of 1.0 was useful for women with ASCUS, giving a more reasonable colposcopy referral rate (35.7%) compared with the referral rate for LSIL (72.4%) using the same cutoff. Similar to the Triage Study results,28 we found significantly improved specificity in older women; however, our smaller sample size precluded an age-stratified analysis within cytologic diagnosis.
Similar to previous studies using the PCR assay,20–22,29 the hybrid capture I,29 and the hybrid capture II assay,18,30 HPV viral level in the present study was found to be strongly associated with the risk of CIN or squamous cell carcinoma. Longitudinal follow-up studies have shown that persistence of high viral levels using PCR testing22,31–33 and the type of HPV present, particularly type 16,31,34–36 are predictive of progression from mild abnormalities to high-grade CIN. Also, HPV 16 has been found to be associated with greater persistence of infection and with progression of low-grade CIN to high-grade CIN,32 especially when it occurs in combination with other carcinogenic HPV types.20,37 The lower odds ratios for predicting any grade of CIN in the present study could be attributed to the presence of some CIN I associated with low-risk HPV types17 and to increased misclassification of histology at the low end of the histologic spectrum.38
The relative light unit ratio in the hybrid capture II test corresponds to the amount of HPV DNA in the sample, and therefore may be influenced by factors such as the adequacy of the sample and the size of the lesion. It has been suggested that viral quantitation should take into account the number of epithelial cells sampled39 because it has been reported that the sample cellular DNA content increases significantly with increasing level of cervical histologic abnormality.40 This may be a result of increased lesion size in higher grade CIN. Sun et al reported that the risks of both a large lesion and HSIL were associated with higher viral level,41 although the influence of viral level on lesion grade controlling for lesion size was not assessed. Although we did not assess sample DNA content, it did not make any practical difference as to which sampling device was used (brush or swab) in our analysis. In a previous population-based study of HPV prevalence, the order in which the samples for PCR and hybrid capture II testing were obtained by cervical soft brush had no effect on the prevalence results.3 The use of various cervical sampling devices with semiquantitative HPV assays, such as PCR,9,10 hybrid capture II,5,18,39 and the older hybrid capture I test,29,42 is supported by the present study.
The high prevalence and incidence of carcinogenic HPV in young women may limit the usefulness of a simple dichotomous HPV test in improving or substituting for cytologic screening programs.3,9,22,35,43 Optimizing the relative light unit ratio cutoff to reduce false-positive tests and unnecessary referrals is one strategy that is desirable in settings where colposcopy resources are low.5,30 It has been suggested that in settings with no cytologic screening programs, a one-time HPV test with maximum sensitivity, in women in their 30s could have a significant impact on the morbidity and mortality of cervical cancer patients.43 In more highly screened populations such as in the United States, any reduction in sensitivity may be unacceptable to patients or clinicians, for fear of missed disease or liability. The use of the higher viral load cutoff for women with LSIL cytology resulted in a reduction in sensitivity for CIN II+ to just over 80% in the present study, similar to the recently reported sensitivity for thin-layer cytopathology.16 In some settings, particularly those in which routine cytologic screening is not done, or where women are unlikely to comply with repeat screening, reductions in test sensitivity for high-grade lesions may have serious ramifications. Studies of the natural history of cervical neoplasia have shown that 12–70% of CIN III progress to invasive cervical cancer, whereas 60% of CIN I lesions regress, and a minority will progress to CIN III.44
The ROC curve analysis is useful to set the acceptable level of specificity and sensitivity of the hybrid capture II test for a given population, depending on the burden of disease and the colposcopy resources available. Recent findings in the United States have suggested that HPV testing may not be as useful in the triage of women with LSIL compared with those with ASCUS.16,28 We found that a higher cutoff for the relative light unit ratio of the hybrid capture II test determined by ROC curve methods significantly improved the specificity of the test for detecting CIN II or III for women with LSIL cytology and among older women. The relatively large gains in specificity, compared with the corresponding smaller decreases in sensitivity, suggest that information on the amount of oncogenic HPV DNA in a cervical specimen may improve the clinical utility of the hybrid capture II test in a triage setting.
|
Footnotes |
|---|
The authors thank Digene Corp. (Gaithersburg, MD) for performing the hybrid capture II testing of a portion of the samples free of charge, and Sylvia Chong for laboratory work.
Received December 28, 2001. Received in revised form May 3, 2002. Accepted May 16, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Franco EL. Cancer causes revisited: Human papillomavirus and cervical neoplasia. J Natl Cancer Inst 1995;87: 779–80.
3. Sellors JW, Mahony JB, Kaczorowski J, Lytwyn A, Bangura H, Chong S, et al. Prevalence and predictors of human papillomavirus infection in women in Ontario, Canada. CMAJ 2000;163:503–8.
4. Herrero R, Hildesheim A, Bratti C, Sherman ME, Hutchinson M, Morales J, et al. Population-based study of human papillomavirus infection and cervical neoplasia in rural Costa Rica. J Natl Cancer Inst 2000;92:464–74.
5. Schiffman M, Herrero R, Hildesheim A, Sherman ME, Bratti M, Wacholder S, et al. HPV DNA testing in cervical cancer screening; results from women in a high-risk province of Costa Rica. JAMA 2000;238:87–93.
6. Thomas KK, Hughes JP, Kuypers JM, Kiviat NB, Lee SK, Adam DE, et al. Concurrent and sequential acquisition of different genital human papillomavirus types. J Infect Dis 2000;182:1097–102.[Medline]
7. Woodman CBJ, Collins S, Winter H, Bailey A, Ellis J, Prior P, et al. Natural history of cervical human papillomavirus infection in young women: A longitudinal cohort study. Lancet 2001;357:1831–6.[Medline]
8. Franco EL, Villa LL, Sobrinho JP, Prado JM, Rousseau MC, Désy M, et al. Epidemiology of acquisition and clearance of cervical human papillomavirus infection in women from a high-risk area for cervical cancer. J Infect Dis 1999;180:1415–23.[Medline]
9. Ho GY, Bierman R, Beardsley L, Chang C, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998;338:423–8.
10. Lancaster EJ, Banach L, Lekalakala T, Mandiwana I. Carcinoma of the uterine cervix: Results of Ka-Ngwane screening programme and comparison between the results obtained from urban and other unscreened rural communities. East Afr Med J 1999;76:101–4.[Medline]
11. Lorincz AT. Molecular methods for the detection of human papillomavirus infection. Obstet Gynecol Clin North Am 1996;23:707–29.[Medline]
12. Manos MM, Kinney WK, Hurley LB, Sherman ME, Shieh-Ngai J, Kurman RJ, et al. Identifying women with cervical neoplasia using human papillomavirus DNA testing for equivocal Papanicolaou results. JAMA 1999;281: 1605–10.
13. Lytwyn A, Sellors JW, Mahony JB, Daya D, Chapman W, Ellis N, et al. Comparison of human papillomavirus DNA testing and repeat Papanicolaou test in women with low-grade cervical cytologic abnormalities: A randomized trial. CMAJ 2000;163:701–7.
14. Sellors JW, Lorincz AT, Mahony JB, Mielzynska I, Lytwyn A, Roth P, et al. Comparison of self-collected vaginal, vulvar and urine samples with physician-collected cervical samples for human papillomavirus testing to detect high-grade squamous intraepithelial lesions. CMAJ 2000;163: 513–8.
15. Shlay JC, Dunn T, Byers T, Baron AE, Douglas JM. Prediction of cervical intraepithelial neoplasia grade 2-3 using risk assessment and human papillomavirus testing in women with atypia on Papanicolaou smears. Obstet Gynecol 2000;96:410–6.
16. Solomon D, Schiffman M, Tarone R. Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: Baselines results from a randomized trial. J Natl Cancer Inst 2001;93:293–9.
17. Kiviat NB, Koutsky LA, Critchlow CW, Lorincz AT, Cullen AP, Brockway J, et al. Prevalence and cytologic manifestations of human papilloma virus (HPV) types 6, 11, 16, 18, 31, 33, 35, 42, 43, 44, 45, 51, 52, and 56 among 500 consecutive women. Int J Gynecol Pathol 1992;11: 197–203.[Medline]
18. Nindl I, Lorincz A, Mielzynska I, Petry U, Baur S, Kirchmayr R, et al. Human papillomavirus detection in cervical intraepithelial neoplasia by the second-generation hybrid capture microplate test, comparing two different cervical specimen collection methods. Clin Diagn Virol 1998;10: 49–56.[Medline]
19. Cuzick J, Terry G, Ho L, Hollingworth T, Anderson M. Type-specific human papillomavirus DNA in abnormal smears as a predictor of high-grade cervical intraepithelial neoplasia. Br J Cancer 1994;69:167–71.[Medline]
20. Ho GYF, Palan PR, Basu J, Romney SL, Kadish AS, Mikhail M, et al. Viral characteristics of human papillomavirus infection and antioxidant levels as risk factors for cervical dysplasia. Int J Cancer 1998;78:594–9.[Medline]
21. Josefsson AM, Magnusson PKE, Ylitalo N, Sørensen P, Qwarforth-Tubbin P, Andersen PK, et al. Viral load of human papilloma virus 16 as a determinant for development of cervical carcinoma in situ: A nested case-control study. Lancet 2000;355:2189–93.[Medline]
22. Yitalo N, Sørensen P, Josefsson AM, Magnusson PKE, Andersen PK, Pontén J, et al. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: A nested case-control study. Lancet 2000;355: 2194–8.[Medline]
23. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York: John Wiley & Sons, 1981.
24. Youden JW. Index for rating diagnostic tests. Cancer 1950;3:32–5.[Medline]
25. Sackett DL, Haynes RB, Tugwell P. Clinical epidemiology: A basic science for clinical medicine. Toronto: Little, Brown and Company, 1985.
26. Kraemer HC. Evaluating medical tests: Objective and quantitative guidelines. Newbury Park, California: Sage Publications, 1992.
27. Cox JT. HPV testing: Is it useful in triage of minor Pap abnormalities? J Fam Pract 1998;46:121–4.[Medline]
28. Sherman ME, Schiffman M, Cox JT. Effects of age and human papillomavirus viral load on colposcopy triage: Data from the randomized atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesion triage study (ALTS). J Natl Cancer Inst 2002;94:102–7.
29. Cox JT, Lorincz AT, Schiffman MH, Sherman ME, Cullen A, Kurman RJ. Human papillomavirus testing by hybrid capture appears to be useful in triaging women with cytologic diagnosis of atypical squamous cells of undetermined significance. Am J Obstet Gynecol 1995;172: 946–54.[Medline]
30. Kuhn L, Denny L, Pollack A, Lorincz A, Richard RM, Wright TC. Human papillomavirus DNA testing for cervical cancer screening in low-resource settings. J Natl Cancer Inst 2000;92:818–25.
31. Cuzick J, Terry G, Ho L, Hollingworth T, Anderson M. Human papillomavirus type 16 DNA in cervical smears as predictor of high-grade cervical cancer. Lancet 1992;339: 959–60.[Medline]
32. Londesborough P, Ho L, Terry G, Cuzick J, Wheeler C, Singer A. Human papillomavirus genotype as a predictor of persistence and development of high-grade lesions in women with minor cervical abnormalities. Int J Cancer (Pred Oncol) 1996;69:364–8.[Medline]
33. Cuzick J, Terry G, Ho L, Hollingworth T, Anderson M. Human papillomavirus type 16 in cervical smears as predictor of high-grade cervical intraepithelial neoplasia. Lancet 1992;339:959–60.
34. Koutsky LA, Holmes KK, Critchlow CW, Stevens CE, Paavonen J, Beckmann AM, et al. A cohort study of the risk of cervical intraepithelial neoplasia grade 2 of 3 in relation to papillomavirus infection. N Engl J Med 1992; 327:1272–8.[Abstract]
35. Terry G, Ho L, Jenkins D, Hills M, Singer A, Mansell B, et al. Definition of human papillomavirus type 16 DNA levels in low and high grade cervical lesions by a simple polymerase chain reaction technique. Arch Virol 1993; 128:123–33.[Medline]
36. Schlecht N, Kugala S, Robitaille J, Ferreirea S, Santos M, Miyamuar RA, et al. Persistent human papillomavirus infection as a predictor of cervical intraepithelial neoplasia. JAMA 2001;286:3106–14.
37. Ho GYF, Burk RD, Klein S, Kadish AS, Chang CJ, Palan P, et al. Persistent genital human papillomavirus infection as a risk factor for persistent cervical dysplasia. J Natl Cancer Inst 1995;87:1365–71.
38. Stoler MH, Schiffman M. Interobserver reproducibility of cervical cytologic and histologic interpretations. Realistic estimates from the ASCUS-LSIL triage study. JAMA 2001;285:1500–5.
39. Clavel C, Masure M, Bory J-P, Putaud I, Mangeonjean C, Lorenzato M, et al. Human papillomavirus testing in primary screening for the detection of high-grade cervical lesions: A study of 7932 women. Br J Cancer 2001;89: 1616–23.
40. Swan DC, Tucker RA, Tortolero-Luna G, Follen Mitchell M, Wideroff L, Unger ER, et al. Human papillomavirus (HPV) DNA copy number is dependent on grade of cervical disease and HVP type. J Clin Microbiol 1999;37: 1030–4.
41. Sun C-A, Lai H-C, Chang C-C, Neih S, Yu C-P, Chu T-Y. The significance of human papillomavirus viral load in prediction of histologic severity and size of squamous intraepithelial lesions of the uterine cervix. Gynecol Oncol 2001;83:95–9.[Medline]
42. Ferris DG, Wright TC Jr, Litaker MS, Richart RM, Lorincz AT, Sun XW, et al. Triage of women with ASCUS and LSIL on Pap smear reports: Management by repeat Pap smear, HPV DNA testing, or colposcopy? J Fam Pract 1998;46:125–34.[Medline]
43. Meijer CJLM, Snijders PJF, van den Brule AJC. Screening for cervical cancer: Should we test for infection with high-risk HPV? CMAJ 2000;163:535–8.
44. Kiviat N. Natural history of cervical neoplasia: Overview and update. Am J Obstet Gynecol 1996;175:1099–104.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
