Based on the central role of persistent cervical infections by carcinogenic human papillomavirus (HPV) genotypes in cervical carcinogenesis,1 HPV DNA testing has been introduced into cervical cancer screening. In 1999, the U.S. Food and Drug Administration (FDA) approved the first carcinogenic or high-risk HPV DNA test for triage of equivocal cervical cytology and in 2003 for cotesting of women 30 and older (Hybrid Capture 2; Qiagen Corporation, Gaithersburg, MD). The American Cancer Society (ACS) in 20022 and a workshop cosponsored by the National Institutes of Health–National Cancer Institute, the American Society of Colposcopy and Cervical Pathology, and the American Cancer Society in 20043 published guidelines for adjunctive HPV testing with cervical cytology (cotesting) for cervical cancer screening of women aged 30 years and older in the United States.
Kaiser Permanente Northern California, a health maintenance organization that serves more 3 million members living in Northern California, was one of the first to adopt HPV DNA testing into routine cervical cancer screening of women aged 30 and older starting in 2003. The size of its membership and time since adoption provides a unique opportunity to explore the relationships of screening results and histologic outcomes of cervical precancer and cancer. Of particular importance is describing the age-specific risks of cervical precancer and cancer for each cytologic interpretation of abnormality and HPV status. These data will serve as a baseline for understanding the future effect of HPV vaccination on the cervical cancer screening results. The objective of this study was to estimate the relationship of HPV detection and abnormal cytology with histologic diagnoses of cervical precancer and cancer.
MATERIALS AND METHODS
The Regional Laboratory of the Northern California Kaiser Permanente Medical Care Program provides cervical cytology services for approximately 75% of the Northern California Kaiser membership, a racially and ethnically diverse catchment of approximately 1 million women. Since the 1990s, the Pap and cervical biopsy correlation data generated as part of our quality assurance programs has been retained. The period before 2003 was one of rapid evolution of clinical practice, with increasing uniformity of diagnostic categories, the advent of the current Bethesda system,4 and the introduction of HPV testing by Hybrid Capture 2 for atypical squamous cells of undetermined significance (ASC-US) triage in 1999, and significant changes in clinical management protocols.5 In late 2002, a Cervical Cancer Screening Guideline was approved for Kaiser Permanente Northern California, which included the recommendation for general population screening of women age 30 and over with both Pap and HPV testing. The period from 2003 to 2008, in contrast, was marked by relative stability and uniformity of diagnostic categories and clinical practice recommendations.
During this latter period all of the cervical biopsy reports referable to abnormal screening tests in facilities served by the Regional Laboratory were manually reviewed and entered into the database by Regional Laboratory personnel under the direction of one of the authors (B.F.). A random sample of 9.9% of all of the biopsy reports (for all cytologic diagnoses and ages) was subsequently re-reviewed by the investigators to assess the accuracy of the data entry. Using cervical intraepithelial neoplasia grade 2 (CIN2), or cervical intraepithelial neoplasia grade 2 or more severe (CIN2+) compared with less than CIN2+ as the cut point, classifications of abnormal histology in the database were found worthy of revision in 1.6% of cases, mostly in relation to “CIN1–2” which was reclassified as less than CIN2 rather than CIN2+.
Data were extracted from all Pap tests performed during January 1, 2003, through December 31, 2008, regardless of whether biopsy correlation was available. Paps occurring during the period January 1, 2003, through December 31, 2008 for which histologic correlation was available were mapped to the worst subsequent histologic diagnosis occurring from January 1, 2003, through June 30, 2009. Women may have contributed multiple results to our analysis. Data from the Kaiser Permanente Northern California cancer registry were used to supplement the histology data for more complete ascertainment of cervical cancer incidence within the Kaiser Permanente Northern California membership. Of the 48.8% women who self-reported their race/ethnicity, 62.1% were Caucasian, 12.4% were Asian/Pacific Islander, 12.2% were Hispanic, and 8.4% were African-American.
At Kaiser Permanente Northern California, continued screening is recommended for women aged 65 years and older only if they have 1) not had three or more documented, consecutive, technically satisfactory normal/negative cervical cytology tests within the 10-year period before age of 65, or 2) if either the Pap or the HPV test was positive at their last cotest. Continued screening is also left to the discretion of the clinician if a woman 65 or older tests HPV positive. These guidelines are consistent with national screening guidelines that recommend discontinuing cervical cancer screening above the age of 65 (http://www.ahrq.gov/) or 702 only when these women have had adequate recent screening with normal Pap test and are not otherwise at high risk of cervical cancer. We therefore did not present data on women aged 65 and older as any screening in this age group is likely biased by the inclusion of higher-risk women.
Women aged 30 and older were screened by cotesting using both HPV testing and conventional Pap tests or could opt for annual Pap test screening. More than 90% women underwent cotesting. Human papillomavirus DNA testing for 13 carcinogenic HPV genotypes (HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 68) was performed (Hybrid Capture 2) according to the manufacturer's instructions. Conventional Pap tests rather than liquid-based cytology were used, and read using the BD FocalPoint GS (guided screener) Imaging System, a location-guided screening system from BD Diagnostics (Burlington, NC). Recent data from both meta-analysis and randomized trials have not documented a significant difference in the performance of conventional Pap tests and liquid-based cytology for detection of cervical precancer and cancer.6,7
Women with abnormal cytology or ASC-US with a positive HPV test were referred to colposcopy. Women with HPV-negative ASC-US were generally rescreened in 6 or 12 months and did not undergo colposcopy. A small, potentially biased subset of women with HPV-negative ASC-US was referred to colposcopy, and therefore these histological results are not presented in this analysis. The management of women with a negative Pap and a positive HPV test evolved during this time period, from no specific recommendations to watch-and-wait to referral to colposcopy after a repeat HPV positive, Pap test negative result (or more severe) according to current guidelines.8 Due to these changing management strategies, we do not present the histologic outcomes of women with HPV-positive, Pap-negative results. Women with a histological diagnosis of CIN2/3 were treated by excision (eg, loop electrosurgical excision procedure) according to U.S. standards of practice.
Cytologic results were categorized as high-grade squamous intraepithelial lesion (HSIL), atypical squamous cells, cannot exclude HSIL (ASC-H), atypical glandular cells (AGC), low-grade squamous intraepithelial lesion (LSIL), ASC-US, and negative. Histological results were classified as less than CIN2, CIN2 (n=2,057), CIN3/adenocarcinoma in situ (n=1,592), and cervical cancer (squamous cell carcinoma and adenocarcinoma) (n=167). Clinical-relevant histologic outcomes were categorized as CIN2+, CIN3+, and cancer.
Binomial 95% confidence intervals were calculated as needed. Mantel-Haenzel tests of trends9 were used to test for statistically significant (P<.05) age-group trends.
The Kaiser Permanente Northern California institutional review board approved the use of the data, and the National Institutes of Health Office of Human Subjects Research deemed it was exempt from institutional review board review.
The percentages of women 30 and older with abnormal cervical cytology who tested HPV positive were: HSIL 92.2% (95% confidence interval [CI] 90.8–93.3%), ASC-H 65.8% (95% CI 63.6–67.9%), AGC 21.3% (95% CI 19.7–22.9%), LSIL 83.5% (95% CI 82.6–84.3%), and ASCUS 34.5% (95% CI 34.0–35.0%). Only 4.2% (95% CI 4.1–4.2%) women with negative cytology tested HPV positive (Table 1). There was a notable decline in the percentage of HPV-positive women in all categories of cytology with age (Ptrend<.001).
The patterns of abnormal cytologic interpretation by age group are shown in Figure 1. There was age-specific decline in any cytologic abnormality for all women (Ptrend<.001) and HPV-positive women (Ptrend<.001) (Fig. 1A). Among all women with any cytologic abnormality (Fig. 1B), there was a shift from more certain (HSIL and LSIL) to less certain (ASC-US, ASC-H, and AGC) cytologic interpretations with increasing age. Specifically, there was a marked decline in the percentage of the LSIL among all women with abnormal Pap test results with increasing age (Ptrend<.001), from 21.7% in women aged 30–34 years to 8.8% women aged 60–64 years. There was a similar trend of decreasing occurrences of HSIL cytology among all women with abnormal Pap test results with increasing age (Ptrend<.001). Conversely, there was an increase the percentage of the ASC-US among all women with abnormal Pap test results with age (Ptrend<.001), from 67.2% in women aged 30–34 years to 77.9% in women aged 60–64 years. There was similar trend of an increasing percentage of ASC-H (Ptrend<.001) and AGC cytology (Ptrend<.001) among all women with abnormal Pap test results with increasing age. We noted similar age patterns observed among HPV-positive women (Fig. 1C). Among HPV-positive women with abnormal cytology, 57% had ASC-US, 27% had LSIL, 2% had AGC, 6% had ASC-H, and 7% had HSIL. Among HPV-negative women with abnormal cytology, 86% had ASC-US, 4% had LSIL, 7% had AGC, 2% had ASC-H, and <1% had HSIL.
The risks of CIN2+ after each abnormal cervical cytology interpretation were HSIL (75.3%), ASC-H (37.9%), AGC (11.9%), LSIL (15.9%), and HPV-positive ASC-US (14.8%) (Table 2). The risks of CIN3+ after each abnormal cytology interpretation were HSIL (46.4%), ASC-H (21.2%), AGC (8.2%), LSIL (4.9%), and HPV-positive ASC-US (5.3%). The risks of cervical cancer for each abnormal cytologic interpretation were HSIL (6.0%), ASC-H (2.2%), AGC (1.7%), LSIL (0.2%), and HPV-positive ASC-US (0.2%). Most cases of CIN2+ (93.6%), CIN3+ (94.7%), and cancer (92.6%) diagnosed in women with HSIL, ASC-H, AGC, or LSIL cytology tested HPV positive (HPV-negative women with ASC-US did not routinely undergo colposcopy). There was a decrease in CIN2 (Ptrend<.001) and CIN3 (Ptrend<.001), and an increase in cervical cancer (Ptrend=.03), with increasing age (data not shown).
Human papillomavirus–positive women remained at an elevated risk of CIN2+ and CIN3+ at all ages compared with HPV-negative women in all cytologic categories. The absolute risks of CIN3+ were greatest among women with HSIL, ASC-H, and AGC cytology, whereas the absolute risks of CIN2 were greatest among women with LSIL and HPV-positive ASC-US cytology.
Human papillomvirus–negative women with HSIL or ASC-H cytology were at a reduced but significant risk of CIN2+ (43.4% and 10.6%, respectively), CIN3+ (18.9% and 4.3%, respectively), and cancer (0.9% and 1.0%, respectively) compared with HPV-positive women with HSIL or ASC-H cytology. Human papillomvirus–negative women with LSIL were only at a 4.0% risk of CIN2+, 1.8% risk of CIN3+, and 0.2% risk of cervical cancer. Among HPV-negative women with LSIL who are 50 years and older, the risks of CIN2+ and CIN3+ were 0.5% and 0.2%, respectively, and no cancers were observed (n=688) (95% CI 0.0–0.5%).
The fraction of histologic diagnoses of CIN2+, CIN3+, and cancer with each cytologic interpretation is shown in Figure 2. Human papillomavirus–positive ASC-US (35%) and HSIL (26%) were the most common abnormal cytologic interpretation preceding a CIN2+ diagnosis. High-grade squamous intraepithelial lesion (35%) and HPV-positive ASCUS (28%) were the most common abnormal cytologic interpretation preceding a CIN3+ diagnosis. High-grade squamous intraepithelial lesion (47%) and AGC (18%) were the most common cytologic interpretation preceding cervical cancer. Similar patterns were observed in HPV-positive women, women aged less than 50 years of age, and women aged 50 years and older (data not shown).
The Kaiser Permanente Northern California database on cervical cancer screening results provided a unique opportunity to examine the age-specific relationships of HPV testing and cytology results and their relationship with clinically relevant histopathologic outcomes. The real-world clinical setting of Kaiser Permanente Northern California proved to be both the strength and the limitation of this analysis. One of the strengths of our analysis was the very large sample size, the availability of approximately 1 million screening results, which cannot be practically replicated in a clinical trial. Thus, we are able to examine the age-specific relationship between screening results and outcomes in finer detail and to study rarer cytologic categories such as AGC and ASC-H. Importantly, women with LSIL, AGC, ASC-H, and HSIL cytology were referred to colposcopy irrespective of the HPV testing results, removing any concerns regarding referral bias.
However, real-world clinical practice (compared with a controlled clinical trial) has a few irregularities, including changing of clinical guidelines over time. As mentioned, the management of HPV-positive women with negative cytology evolved over the 5 years since the introduction of HPV testing in women 30 and older at Kaiser Permanente Northern California. Thus, we were not able to report accurately on the point estimates for precancer and cancer risk after a HPV positive, cytology-negative result during this 5-year time period.
These data can provide important insights into the clinical utility of HPV testing in conjunction with cervical cytology. One example is the use of HPV testing as a triage of abnormal cytology to decide who needs immediate colposcopic evaluation, that is, HPV-positive women are referred to colposcopy while HPV-negative women are rescreened in follow-up. Human papillomavirus testing for triage of ASC-US cytology is now widely accepted in the United States and Europe. There has been some evidence to suggest that HPV-based triage of ASC-H10 and LSIL11 may be useful by reducing the number of unnecessary colposcopic evaluations in low-risk, HPV-negative women.
There are two main criteria for deciding whether HPV-based triage is clinically useful. First, does the triage test (HPV testing) identify a significant proportion of women who need or do not need referral to expensive colposcopy and biopsy procedures that justify the cost of the triage test? That is, if HPV testing is almost always negative or positive, there is little value in using it. As an example of the latter, the HPV testing arm for women referred into the ASCUS-LSIL Triage Study (ALTS) with LSIL cytology was terminated before completion because most tested HPV positive.12 The second criterion is whether HPV-negative women can be safely followed rather than sent to colposcopy.
By these criteria, HPV triage of women 45 years and older or 50 and older with LSIL may be rational because of the lower percentage of those testing HPV positive (less than 80%) and the very low risk of CIN3 and cancer among the HPV-negative women. This is consistent with findings from a large clinical trial in Europe.11 The discrepancy between these observations and ALTS is that ALTS enrolled predominately younger women (median age 24 years).13
Conversely, while there is a notable decrease in the percentage of HPV-positive women with ASC-H cytology, the risk of CIN2+ and CIN3+ remains high in the HPV-negative women, rendering HPV triage unsafe for this cytologic category. This is perhaps to be expected given the a priori high risk of CIN3+ among women with ASC-H, and all tests and procedures are imperfect.14
As we previously reported,15 HPV test results for women with AGC cytology may not be useful for triage but may guide clinicians to which anatomical site, cervix or endometrium, a problem may exist: women with AGC cytology who test HPV positive have exceptional high risks of cervical precancer and cancer while those who test HPV negative have a high risk of endometrial cancer.
As expected, HPV-positive ASC-US and LSIL has very similar risks of CIN2+ and CIN3+ at all ages, again clearly demonstrating that HPV-positive ASC-US is LSIL, that is, HPV testing distinguishes between true HPV-related morphologic changes and “look-alike” changes that carry little risk of significant cervical disease.16–18 Likewise, as noted above, HPV testing in older women with LSIL cytology may be useful to make a similar distinction between true infection and false abnormal cytology. It is also noteworthy that the disease found in HPV-positive ASC-US and LSIL is primarily CIN2, an equivocal diagnosis of cervical precancer,19 and virtually no cancer was found, especially in women aged 50 years and older. Low-grade squamous intraepithelial lesion and its equivalent (HPV-positive ASC-US), are considered cytologic manifestations of HPV infection, providing further indirect evidence that a significant portion of CIN2 diagnoses are probably nothing more than HPV infection. Distinguishing between HPV infection and true cervical precancer following a colposcopically-directed biopsy diagnosis of CIN2 should continue to be a research priority to reduce unnecessary treatment, which has negative consequences on reproductive outcomes,20 relevant to those women 30 and older still considering pregnancies.
Finally, we observed a profound shift in the cytologic interpretations found in younger women compared with those found in older women. With increasing age, there was a greater percentage of equivocal cytologic interpretations, for example, ASC-US, ASC-H, and AGC, as well as lower percentages of HPV positives in these equivocal cytologic interpretations and in LSIL. The most likely explanation for these patterns is age-related atrophy of the cervical epithelium, due to lower estrogen in older women, which may reduce the ability of cervical cells to support a highly productive viral state as represented by LSIL and may also cause the appearance of mimics.
In HPV16/18-vaccinated populations, we expect that there will be fewer interpretations of HSIL cytology and less disease associated with any abnormal cytologic result because HPV16 preferentially causes HSIL cytology21,22 and HPV16 and 18 are the two most carcinogenic HPV genotypes.23 Partial protection by HPV vaccines against the next most carcinogenic HPV genotypes,24–26 especially HPV31 and HPV45, may further reduce the occurrence of HSIL cytology. Thus, there will be a profound shift toward equivocal cytology in HPV-vaccinated populations. Human papillomavirus testing will become increasingly important to distinguish women at high and low risks of cervical precancer and cancer among HPV-vaccinated women with equivocal and mildly abnormal cytology.
1. Wright TC Jr, Schiffman M. Adding a test for human papillomavirus DNA to cervical-cancer screening. N Engl J Med 2003;348:489–90.
2. Saslow D, Runowicz CD, Solomon D, Moscicki AB, Smith RA, Eyre HJ, et al. American Cancer Society guideline for the early detection of cervical neoplasia and cancer. CA Cancer J Clin 2002;52:342–62.
3. Wright TC Jr, Schiffman M, Solomon D, Cox JT, Garcia F, Goldie S, et al. Interim guidance for the use of human papillomavirus DNA testing as an adjunct to cervical cytology for screening. Obstet Gynecol 2004;103:304–9.
4. Solomon D, Davey D, Kurman R, Moriarty A, O'Connor D, Prey M, et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA 2002;287:2114–9.
5. Wright TC Jr, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ, ASCCP-Sponsored Consensus Conference. 2001 Consensus Guidelines for the management of women with cervical cytological abnormalities. JAMA 2002;287:2120–9.
6. Arbyn M, Bergeron C, Klinkhamer P, Martin-Hirsch P, Siebers AG, Bulten J. Liquid compared with conventional cervical cytology: a systematic review and meta-analysis. Obstet Gynecol 2008;111:167–77.
7. Siebers AG, Klinkhamer PJ, Grefte JM, Massuger LF, Vedder JE, Beijers-Broos A, et al. Comparison of liquid-based cytology with conventional cytology for detection of cervical cancer precursors: a randomized controlled trial. JAMA 2009;302:1757–64.
8. Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D, et al. 2006 consensus guidelines for the management of women with abnormal cervical screening tests. J Low Genit Tract Dis 2007;11:201–22.
9. Schlesselman JJ. Case-control studies: design, conduct, analysis. New York (NY): Oxford University Press; 1982.
10. Sherman ME, Castle PE, Solomon D. Cervical cytology of atypical squamous cells-cannot exclude high-grade squamous intraepithelial lesion (ASC-H): characteristics and histologic outcomes. Cancer 2006;108:298–305.
11. Ronco G, Cuzick J, Segnan N, Brezzi S, Carozzi F, Folicaldi S, et al. HPV triage for low grade (L-SIL) cytology is appropriate for women over 35 in mass cervical cancer screening using liquid based cytology. Eur J Cancer 2007;43:476–80.
12. Human papillomavirus testing for triage of women with cytologic evidence of low-grade squamous intraepithelial lesions: baseline data from a randomized trial. The Atypical Squamous Cells of Undetermined Significance/Low-Grade Squamous Intraepithelial Lesions Triage Study (ALTS) Group. J Natl Cancer Inst 2000;92:397–402.
13. Schiffman M, Adrianza ME. ASCUS-LSIL Triage Study. Design, methods and characteristics of trial participants. Acta Cytol 2000;44:726–42.
14. Castle PE, Rodriguez AC, Burk RD, Herrero R, Hildesheim A, Solomon D, et al. Neither one-time negative screening tests nor negative colposcopy provides absolute reassurance against cervical cancer. Int J Cancer 2009;125:1649–56.
15. Castle PE, Fetterman B, Poitras N, Lorey T, Shaber R, Kinney W. Relationship of atypical glandular cell cytology, age, and human papillomavirus detection to cervical and endometrial cancer risks. Obstet Gynecol 2010;115:243–8.
16. Solomon D, Schiffman M, Tarone R. Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: baseline results from a randomized trial. J Natl Cancer Inst 2001;93:293–9.
17. ASCUS-LSIL Triage Study (ALTS) Group. Results of a randomized trial on the management of cytology interpretations of atypical squamous cells of undetermined significance. Am J Obstet Gynecol 2003;188:1383–92.
18. Safaeian M, Solomon D, Wacholder S, Schiffman M, Castle P. Risk of precancer and follow-up management strategies for women with human papillomavirus-negative atypical squamous cells of undetermined significance. Obstet Gynecol 2007;109:1325–31.
19. Castle PE, Stoler MH, Solomon D, Schiffman M. The relationship of community biopsy-diagnosed cervical intraepithelial neoplasia grade 2 to the quality control pathology-reviewed diagnoses: an ALTS Report. Am J Clin Pathol 2007;127:805–15.
20. Arbyn M, Kyrgiou M, Simoens C, Raifu AO, Koliopoulos G, Martin-Hirsch P, et al. Perinatal mortality and other severe adverse pregnancy outcomes associated with treatment of cervical intraepithelial neoplasia: meta-analysis. BMJ 2008;337:a1284.
21. Kovacic MB, Castle PE, Herrero R, Schiffman M, Sherman ME, Wacholder S, et al. Relationships of human papillomavirus type, qualitative viral load, and age with cytologic abnormality. Cancer Res 2006;66:10112–9.
22. Safaeian M, Schiffman M, Gage J, Solomon D, Wheeler CM, Castle PE. Detection of precancerous cervical lesions is differential by human papillomavirus type. Cancer Res 2009;69:3262–6.
23. Smith JS, Lindsay L, Hoots B, Keys J, Franceschi S, Winer R, et al. Human papillomavirus type distribution in invasive cervical cancer and high-grade cervical lesions: a meta-analysis update. Int J Cancer 2007;121:621–32.
24. Wheeler CM, Kjaer SK, Sigurdsson K, Iversen OE, Hernandez-Avila M, Perez G, et al. The impact of quadrivalent human papillomavirus (HPV; types 6, 11, 16, and 18) L1 virus-like particle vaccine on infection and disease due to oncogenic nonvaccine HPV types in sexually active women aged 16–26 years. J Infect Dis 2009;199:936–44.
25. Brown DR, Kjaer SK, Sigurdsson K, Iversen OE, Hernandez-Avila M, Wheeler CM, et al. The impact of quadrivalent human papillomavirus (HPV; types 6, 11, 16, and 18) L1 virus-like particle vaccine on infection and disease due to oncogenic nonvaccine HPV types in generally HPV-naive women aged 16–26 years. J Infect Dis 2009;199:926–35.
26. Paavonen J, Naud P, Salmeron J, Wheeler CM, Chow SN, Apter D, et al. Efficacy of human papillomavirus (HPV)-16/18 AS04-adjuvanted vaccine against cervical infection and precancer caused by oncogenic HPV types (PATRICIA): final analysis of a double-blind, randomised study in young women. Lancet 2009;374:301–14.