The 2006 consensus guidelines for the management of women with abnormal cervical screening tests

Challenges remain

Abstract

The American Society for Colposcopy and Cervical Pathology (ASCCP) guidelines for the management of women with abnormal cervical screening test results is a powerful, evidence-based document representing an advance in the consistent treatment of women with cervical abnormalities. Nevertheless, the complexity of the guidelines is an outstanding issue, compliance with the guidelines is problematic in some areas, and there is a growing recognition of the deficiencies of colposcopy. In light of these challenges, the cytopathology community has an opportunity to play an active role in integrating data, defining risk, and recommending proper management for women screened for cervical cancer. Cancer (Cancer Cytopathol) 2010. © 2010 American Cancer Society.

In 2001, the American Society of Colposcopy and Cervical Pathology (ASCCP), under the leadership of Dr. Edward Wilkinson, spearheaded the development of consensus guidelines for the management of abnormal Papanicolaou (Pap) tests and cervical intraepithelial neoplasia (CIN) and adenocarcinoma in situ, commonly known as the ASCCP guidelines.1 These guidelines were re-examined and revised in 2006.2, 3 Four years have elapsed since the development and release of these 2006 consensus guidelines, and during this period the guidelines have been widely disseminated and moved into practice, and have led to considerable harmonization of management throughout North America.

Without doubt, the consensus guidelines have been a milestone in improving the management of women with abnormalities detected during cervical screening. Nevertheless, it is timely to reflect on the outstanding challenges that remain after the release of the second set of guidelines so that future research and educational efforts can be directed toward these areas. We can focus our attention on 3 major issues: complexity, compliance, and colposcopy.

The sheer complexity of the guidelines continues to be a challenge. To understand the guidelines, practitioners must understand basic differences between cytology and histology terms. An example of a common error is low-grade squamous intraepithelial lesion (LSIL), a cytologic interpretive category used in the guidelines, being confused with cervical intraepithelial neoplasia (CIN) type 1 (CIN1), a histopathologic/biopsy result, potentially leading to inappropriate management.

Even with knowledge of basic histologic definitions, the management of a single histologic diagnosis may vary. For example, the clinical significance and subsequent management of a CIN1 biopsy result varies according to the preceding index Pap test result. A single histologic diagnosis has different management pathways. Indeed, multiple parameters help determine the management of a patient with an abnormal cervical screening test. These parameters include age, symptoms, current Pap test result, Pap test history, biopsy result, and human papillomavirus (HPV) status. Some decision trees are long and intricate. Although facile use of the guidelines is possible for healthcare professionals from a variety of disciplines who have maintained a special interest in lower genital tract disease, many general gynecologists, nurse practitioners, and family physicians are challenged to maintain sufficient familiarity with the management of a common screening test. Previously, laboratories made management recommendations for common cytologic abnormalities to assist the generalist. Now, many laboratories have questioned this practice because recommendations cannot be based solely on a Pap test result. For example, the management of a woman with a Pap test result of LSIL or atypical squamous cells of undetermined significance (ASC-US) varies by age, high-risk HPV (HR-HPV) testing availability, and whether the woman is being actively managed. Faced with this situation, many laboratories now omit providing any recommendations and may cite management guideline references only. Thus, a useful guide for the nonspecialist practitioner is often no longer available.

Is compliance with the guidelines good or poor? In those cytologic interpretative categories in which the guidelines are similar to long-standing practice patterns, compliance will be high. For example, compliance with the management recommendation for a high-grade SIL (HSIL) Pap test result, colposcopic investigation, is likely high because this recommendation reflects decades of practice. However, studies of compliance with Pap test management guidelines in general are cause for concern. A 2005 study of guideline adherence for the follow-up of women with low-grade cytologic abnormalities found that the follow-up of >50% of women was not in accordance with established guidelines.4 A survey of primary care physicians identified a wide variation in adherence to Pap test guidelines despite awareness of the published guidelines.5 Two interpretive categories of the 2001 Bethesda terminology system and HPV testing deserve special attention.

The category of “benign-appearing endometrial cells” in women aged ≥40 years was established in the 2001 revision of the Bethesda terminology because it was recognized that there was a low prevalence of significant disease in this group of patients; no management recommendation were made.6 The ASCCP guidelines addressed this deficit by recommending that no further evaluation of asymptomatic, premenopausal women with benign-appearing endometrial cells was necessary.2 Nevertheless, studies identifying a significant risk of underlying disease in this group of patients are still reported, with the suggestion that such women do indeed “warrant” consideration of biopsy,7 and there continues to be a need for further prospective studies.8 Whether the result of thoughtful disagreement, or caution bred by medicolegal concerns, many women aged ≥40 years undergo endometrial biopsy/curettage and/or transvaginal ultrasound evaluation of endometrial thickness solely to investigate a Pap test finding of benign-appearing endometrial cells.8

The second Pap test interpretive category of particular note with respect to noncompliance is the finding of atypical glandular cells (AGC), which is often followed by inappropriate, delayed, or absent management because of oversight or misunderstanding.9 Although some improvements in appropriate management have been reported recently,10 there is still an inordinate percentage of women with an AGC Pap test result who may experience a delay in their final diagnosis and inappropriate management. Therefore, the overmanagement of women with benign-appearing endometrial cells and the undermanagement of women with AGC are cause for concern.

Continued requests for low-risk HPV (LR-HPV) testing and the inappropriate ordering of HR-HPV testing indicate that HPV testing guidelines are frequently ignored. The guidelines state that there is no valid clinical indication for the use of LR-HPV testing, but there clearly remains a clinical demand, as revealed in both laboratory and clinical surveys. Although HR-HPV testing was stressed in both the earlier 2001 and 2006 guidelines, nearly one-half of laboratories responding to a 2007 College of American Pathologists' questionnaire reported that they continue to offer LR-HPV testing in response to physician ordering.11 An analysis of 376 responses to a cervical cancer screening supplement of the 2006 National Ambulatory Medical Care Survey revealed that nearly one-third of providers across all specialties continued to use LR-HPV testing.12

There also continues to be widespread misuse of HR-HPV testing and inappropriate response to HR-HPV positivity. Both community and university physicians continue to use HR-HPV testing in adolescents.13 In another study from Chicago, 84% of adolescents (aged ≤20 years) with an ASC-US Pap test result were tested for HR-HPV inappropriately in 2008 and 2009, 2 years and 3 years, respectively, after the release of the guidelines.14 Nearly one-half of women aged ≥30 years whose combined Pap test and HR-HPV screen were negative for an intraepithelial lesion or malignancy and HR-HPV test was positive (negative for intraepithelial lesion or malignancy [NILM])/HPV positive) underwent Pap tests or biopsy in <1 year by obstetrician-gynecologists, although the guidelines recommend repeat HR-HPV and Pap testing at 1 year. Some have estimated that HR-HPV positivity has now replaced ASC-US as the most prevalent cause for colposcopic referral.15 Another abuse of HPV testing is exemplified when a woman aged >30 years has a negative co-test (NILM negative/HPV negative) and the co-test is repeated annually instead of lengthening the interval between screening as recommended. The consensus statement regarding HPV DNA test use by 8 societies is an attempt to refocus attention on the correct use of HPV testing.16 Others have concluded that it is unlikely that educational efforts alone can alter the widespread misuse of HPV testing, and that wide-reaching interventions are necessary.12

Technological advances resulting in the recent availability of type-specific HPV testing have contributed to the growing complexity of HPV testing. In March 2009, the US Food and Drug Administration (FDA) approved a test for the identification of HPV types 16 and 18 (Cervista HPV 16/18; Hologic, Marlborough, MA) for clinical use as an adjunctive tool with Cervista HR-HPV and Pap testing in women aged ≥30 years, and in reflex testing of patients with an ASC-US result. Only 1 of the uses of Cervista HPV that met FDA approval has been endorsed by the ASCCP: the reflex use of Cervista HPV in a woman aged ≥30 years who has had a screening result of NILM/HR-HPV positive.17 The ASCCP does not accept that Cervista HPV has any role in the reflex testing of patients with an HR-HPV-positive ASC-US result, because all these women should undergo colposcopy regardless of their HR-HPV subtype. The recent availability of type-specific HPV testing (more choices) and the discrepancy between the manufacturer's claims and ASCCP recommendations (mixed messages) superimposed on pre-existing widespread noncompliance in HPV testing suggest that improving compliance for HPV testing will become a greater challenge.

The primary investigative tool for the management of patients with an abnormal Pap test has been colposcopy. Studies published since 2006 have raised serious doubts as to whether colposcopy will retain this central determinative role. The guidelines recognize that a single colposcopic examination fails to identify approximately one-third of prevalent CIN2 and CIN3, but additional data raise even greater concerns regarding the reliability of colposcopic impression and the selection of biopsy site in the fully observed transformation zone.15 In the immediate colposcopy arm of the ASCUS/LSIL Triage Study (ALTS), slightly greater than 50% of women with a histopathologic diagnosis of CIN3 within the 2-year follow-up period were found to have a CIN2-positive biopsy at the time of study enrolment. In other words, in >40% of cases, enrolment colposcopy failed to detect incipient or current CIN3. The risk of CIN3 among women with a low-grade Pap test finding and an enrolment colposcopy that did not detect CIN3 was found to be equivalent among women with colposcopic results of CIN1, those with negative biopsy results, and those with normal colposcopy results and no biopsy. The CIN3 lesions in the ALTS that were missed by colposcopy were determined to be small.18 Increasing the number of biopsies was found to improve the sensitivity of the colposcopic procedure regardless of the experience of the colposcopist. Other studies have demonstrated that random biopsy specimens taken from areas without observed colposcopic abnormalities often demonstrate positivity for CIN2, albeit on a smaller scale than in colposcopically observed lesions.19-23 The poor sensitivity of colposcopy has led to recommendations that increased numbers of directed or random biopsies be routinely adopted,24 and for “rigorous, formal study” of colposcopy to improve its reliability.15 Collaboration among the National Cancer Institute, the National Library of Medicine, and the ASCCP using digitalized images is currently underway in an attempt to improve colposcopic outcomes. However, it would be surprising if, after 80 years of use and development, even an organized research program led to a sudden remarkable breakthrough in the sensitivity of manual, non-computer–assisted colposcopy. Indeed, in the postvaccine era, further deterioration in colposcopic skills and the sensitivity of CIN3 can be anticipated as colposcopic volumes decrease and lesion size shrinks.

It therefore appears reasonable to speculate that colposcopy may not always play a key central role in the investigation and management of patients with an abnormal Pap test, and that alternative investigative tools such as markers of local neoplastic transformation could be superior, or assist, in identifying women with existing or incipient CIN3. Even in the 2006 ASCCP guidelines, the central role of colposcopy in assessing 1 of the most significant abnormal Pap test results, HSIL, was supplanted by the recognition that immediate loop electrosurgical excision in nonpregnant women aged ≥21 years without colposcopic assessment (“see and treat”) is acceptable.

In summary, adherence to the current management algorithms continues to be a challenge, particularly for the nonspecialist, because of the complexity of these algorithms. Future management algorithms are likely to be more complex to incorporate type-specific HPV testing and HPV immunization history, leading to further deteriorations in compliance. At the current time, ordering the HPV test does not conform to the guidelines and, with new HPV tests now available, it is probable that this noncompliance will persist and most likely increase. The sensitivity of colposcopy is only fair and unlikely to improve drastically, thereby providing an impetus for the development of innovative molecular tests to improve our detection of cervical disease.

The current challenges are to improve test ordering, enhance adherence to the ASCCP guidelines, and embrace new and emerging tests. These challenges are opportunities for the cytopathology laboratory. Cervical screening and management decisions can be made by establishing the risk estimate for cervical disease using all available clinical and laboratory data.25 Could the laboratory assume a key role in ensuring or directing appropriate test ordering; integrating all laboratory data; establishing a risk estimate for cervical disease; and, yes, even recommending management in some situations? Recently, Austin et al reported just such a risk assessment tool, the Pittsburgh Cervical Cancer Screening Model, which uses a Bayesian network of 19 variables including historical data, Pap test results, HR-HPV testing, procedure data, and biopsy results to determine risk.26 There may be increased opportunities for data integration as the electronic medical record grows. Could we envision the day when laboratories report a risk of 0.005 for current or incipient (<5 years) CIN3 based on current and past cytology results, a negative HR-HPV test, age, and biopsy interpretation? The laboratory should not just generate data and provide Pap testing, but also integrate information and provide comprehensive risk estimate(s) for women: a new vision for the cytopathology laboratory.

CONFLICT OF INTEREST DISCLOSURES

The author made no disclosures.