College of American Pathologists
Printable Version

  PAP/NGC Program Review





cap today

August 2004
Special Section

Detection of adenocarcinoma in situ of the cervix in Papanicolaou tests

Jonathan H. Hughes, MD, PhD

Several studies show no clear evidence that Papanicolaou test screening is an effective method of detecting adenocarcinoma in situ (AIS) of the cervix or that screening has reduced the mortality from adenocarcinoma. Nonetheless, the use of the interpretive category of adenocarcinoma in situ of the cervix is becoming increasingly common in cytologic material. The reasons for this phenomenon are multifactorial and include an apparent increase in the incidence of the disease, the development of better endocervical sampling devices, and the publication of numerous papers that describe specific morphologic criteria for recognizing adenocarcinoma in situ in cervicovaginal specimens. In 1998, the International Academy of Cytology Task Force recommended that cases of AIS be specifically diagnosed when possible, and, more recently, the Bethesda 2001 classification included AIS as a distinct diagnostic category.

In a recent article, Andrew A. Renshaw, MD, and colleagues compared pathologists’ ability to correctly identify and categorize adenocarcinoma in situ with their ability to identify and categorize other types of high-grade lesions (Arch Pathol Lab Med. 2004; 128:153-157). Using data compiled from the CAP Interlaboratory Comparison Program in Cervicovaginal Cytopathology, the authors calculated reviewer pathologists' false-negative rates and the exact reference diagnosis agreement rates for 213 reviews of adenocarcinoma in situ, 2,821 reviews of adenocarcinoma, 7,535 reviews of high-grade squamous intraepithelial lesion, and 1,886 reviews of squamous cell carcinoma. They demonstrated that the false-negative rate for adenocarcinoma in situ (11.7 percent) was significantly higher than that for high-grade squamous intraepithelial lesion (4.6 percent, P < .001) and squamous cell carcinoma (3.3 percent, P < .001), but not for adenocarcinoma (8.9 percent, P < .16). Only 46.5 percent of the adenocarcinoma in situ reviews were interpreted specifically as adenocarcinoma in situ, compared with 72.2 percent of adenocarcinoma, 73.2 percent of high-grade squamous intraepithelial lesion, and 75.1 percent of squamous cell carcinoma. In fact, no individual case of adenocarcinoma in situ was always specifically recognized as adenocarcinoma in situ.

All of the cases of adenocarcinoma in situ in the Renshaw, et al, study were expert-referenced, biopsy-proven cases. In spite of the fact that these cases were classic examples, they proved problematic and difficult to diagnose. The findings suggest that adenocarcinoma in situ is not as easily recognized or categorized as other serious diagnoses.

Although additional studies should be performed to confirm their findings, Dr. Renshaw and colleagues have provided compelling evidence that laboratories and clinicians cannot expect all biopsy-proven adenocarcinoma in situ cases to be specifically identified. Cytopathology professionals may need to educate health care providers about this important limitation of the Papanicolaou test. This study also has important implications in the medicolegal arena, because missed cases of adenocarcinoma in situ constitute a significant proportion of litigated Papanicolaou test cases in which a false-negative interpretation of the slide in question is alleged. Dr. Renshaw's findings indicate that it may not be practical to assert that a reasonable practitioner standard of care requires the detection of adenocarcinoma in situ in all cases.

Dr. Hughes, a member of the CAP Cytopathology Committee, is staff pathologist at Laboratory Medicine Consultants, Las Vegas.

Body fluids: good and bad actors in the nongynecologic cytology program
Michael A. Schulte, MD

In everyday cytology practice, body fluid examination can be extremely challenging. Body fluids can present as some of the most difficult of all nongynecological specimens. Whether they present as a screening challenge or as a diagnostic challenge, one must be aware of the spectrum of changes that can be seen in benign and malignant cells in fluids.

With this in mind the authors of a recent article published in Archives of Pathology & Laboratory Medicine (Moriarty, et al. 2004;128:513-518) reviewed data from the CAP Interlaboratory Comparison Program in Nongynecologic Cytopathology to identify cellular characteristics in body fluids that place them at the opposite ends of the diagnostic spectrum. They examined the features of individual body cavity fluid slides that demonstrated good performance characteristics and compared them with slides that were poor performers.

Cases were selected from a data bank of 10,396 laboratory responses obtained from 1997 through 2001. Both benign and malignant cases were included. Cases that performed well or performed poorly relative to the reference diagnoses were identified using a cumulative slide history. All slide diagnoses were confirmed by consensus of four CAP Cytopathology Committee members. Observations and characteristics of good and bad performers in each category (adenocarcinoma, squamous cell carcinoma, small cell carcinoma, mesothelioma, melanoma, lymphoma, and negative/reactive) were recorded and summarized. Concordance of the poor performers ranged from zero percent to 58 percent. The good performers showed a high concordance of laboratory diagnoses in each reference category of greater than 80 percent.

A number of patterns emerged. The poorly performing cases of adenocarcinoma consisted of slides with rare tumor cells, hypercellular malignant cases without two cell populations, and cases with single cells. Poor performance in confirmed squamous cell carcinoma cases related to rare cells without keratinization. Small cell carcinoma and melanoma cases performed poorly when the slides contained few malignant cells. Lymphoma cases demonstrated poor performance when there was abundant pleomorphic lymphoid cells or when rare Reed-Sternberg-like cells were present. Reactive or negative slides performed best with a polymorphous population; poor performers were those with a predominant lymphocyte population mistaken for a hematopoietic neoplasm.

The authors concluded that close attention to classic cytologic criteria and careful examination of slides may enhance the educational experience of participants and the performance characteristics of body cavity fluid specimens in the CAP NGC program. Lessons from bad actors in the CAP NGC program may increase awareness of potential diagnostic problems in daily practice or help identify areas for laboratory quality improvement.

Dr. Schulte, a member of the CAP Cytopathology Committee, is in the Department of Pathology, St. Michael Hospital, Milwaukee, Wis.

Defining the precision of the Pap test
Emily E. Volk, MD

Many studies have addressed the accuracy and the positive predictive value of cervicovaginal smears, but the precision or reproducibility of the test had not been well studied until recently. Andrew A. Renshaw, MD, and colleagues evaluated data from the CAP Interlaboratory Comparison Program in Cervicovaginal Cytopathology, or Pap program, to better define the precision of the Pap test (Renshaw, et al. Arch Pathol Lab Med. 2003;127:1413-1420).

The group analyzed the pathologists' interpretations for both validated (25,745 responses) and educational, conventional (14,353 responses) slides in the 2001 program. Responses for liquid-based cytology were not considered in this study. To be considered "validated," each slide must achieve at least 90 percent agreement with the current selection series, with a minimum of 20 correct responses. The standard of error must not exceed five percent. Slides that have been reviewed by the CAP Cytopathology Committee and have begun circulating among participating laboratories but have not yet obtained validation status are designated educational. A correct response by a participant in the program requires only that the slide be placed in the correct selection series (000—unsatisfactory, 100—negative for intraepithelial lesion (NIL), or 200—squamous intraepithelial lesion and carcinoma). An exact match to the specific target interpretation is not required. Thus, the exact match rate is less than the number of correct responses by participants.

The study found that interpretations of negative, Candida, Trichomonas, herpes, and low-grade squamous intraepithelial lesion (LSIL) had a high degree of exact matching. In contrast, slides with the target interpretation of repair, high-grade squamous intraepithelial lesion (HSIL), adenocarcinoma, and squamous cell carcinoma frequently did not show exact matches between the pathologists' responses and the target interpretation. The group also found that when the exact match rate of HSIL is compared directly with that of LSIL, the difference of poor match rates (less than 50 percent exact match) and moderate match rates (less than 100 percent exact match) is statistically significant (P<0.001) for validated slides.

The data from this study strongly suggest that some cytological interpretations are less precise than other cytological interpretations. In particular, the study identifies groups of cytological interpretations, including NIL, Candida, Trichomonas, herpes. and LSIL, that could be diagnosed more reproducibly than other groups of cytological interpretations, including repair, HSIL, SCC, and adenocarcinoma. These data do not include either the atypical squamous cell or atypical glandular cell categories. These categories are not included in the Pap program because the imprecision of the cytological interpretations is already known.

Of important note is the relative precision of LSIL versus HSIL identified in this study. HSIL is a more clinically significant lesion with a much more significant risk of progression to SCC than LSIL. The interpretation of HSIL will lead to colposcopic investigation and subsequent therapy. In contrast, LSIL often will regress without treatment and is often merely a reflection of a transient human papillomavirus infection.

The interpretation of LSIL may be a more appropriate measure of quality assurance than HSIL because it is more reproducible. The study finds that because the interpretation of HSIL is less reproducible, performance evaluations that include its identification as a quality parameter will have more testing failures than those that use more reproducible diagnoses.

The group acknowledges that these findings have implications regarding the selection of cytologic categories for quality management programs. Further implications of the study's findings reach into the medicolegal arena. Because this study reiterates the difficulty of the diagnosis of HSIL and the interobserver variability in the interpretation of some categories of Pap tests, it calls into question the validity of a single expert witness opinion in the interpretation of a disputed Pap test.

Dr. Volk, a member of the CAP Cytopathology Committee, is in the Department of Pathology and Laboratory Medicine, William Beaumont Hospital, Troy, Mich.

The Bethesda System for Reporting Cervical Cytology (2nd edition) and Web site
Nancy A. Young, MD

A joint American Society of Cytopathology and National Cancer Institute task force* has completed a two-year effort to revise the Bethesda System atlas and develop a complementary Web-based collection of cervical cytology images. The group’s efforts have resulted in a second edition of the first Bethesda "Blue Book," The Bethesda System for Reporting Cervical Cytology—Definitions, Criteria, and Explanatory Notes, and a companion Web site.

The atlas, co-edited by Ritu Nayar, MD, and Diane Solomon, MD, has 30 contributors and incorporates liquid-based morphologic criteria and images. The atlas is thicker than the original edition but still has the same blue cover and handheld dimensions. The second atlas has maintained its easy-to-read style with 186 corresponding illustrations that are stunning and represent the spectrum of changes seen on both conventional and liquid-based preparations. The final illustrations were chosen from an original pool of more than 1,000 images through a rigorous multistep approval process described in detail in the introduction section of the atlas. Some images represent classic examples of an entity. Others were selected to illustrate interpretive dilemmas, which all cytologists may not interpret in the same way.

The section on adequacy provides an in-depth account of how to determine cellular adequacy on conventional and liquid-based preparations. The section on non-neoplastic findings greatly expands the illustration of non-neoplastic conditions found in cervical cytology. New chapters on educational notes, reporting of ancillary testing, computer-assisted interpretation, and anal cytology are also included. The price of the atlas, only $34.95 (U.S.), was kept as low as possible for the benefit of the cytology community as a result of the authors not collecting any royalties or honoraria.

The companion Web site to the atlas can be found on the Bethesda educational Web site at and allows global access to diagnostic images and individual self-assessment. Before publication, about half of the atlas images (indicated by an asterisk in the atlas legend) were posted as unknowns on a Web site open to the cytopathology community as part of the Bethesda Interobserver Reproducibility Project, or BIRP. Hundreds of participants submitted their answers online, to provide a more realistic gauge of interpretive reproducibility. The resulting histograms of participants' interpretations of the images can be viewed on the Web site. The site, which can be updated continually, also has additional examples of Bethesda System interpretations and pertinent explanatory notes that are not practical to fit into the atlas.

*ASC-NCI Working Group for the Second Edition Bethesda Atlas and Web site:
ASC Bethesda 2001 Task Force: Ritu Nayar (chair), Diane Solomon (co-chair, NCI), George Birdsong (adequacy), Jamie Covell (glandular lesions), Ann Moriarty (endometrial cells), Dennis O'Connor (educational notes and recommendations), Marianne Prey (computer-assisted interpretation), Steve Raab (ancillary testing), Mark Sherman (atypical squamous cells), Sana Tabbara (other malignant neoplasms), Tom Wright (squamous lesions), Nancy Young (non-neoplastic findings).

ASC consultants: ASC 2002/2003 presidents: Diane Davey and Dave Wilbur.

Dr. Young, a member of the CAP Cytopathology Committee, is in the Department of Pathology, Fox Chase Cancer Center, Philadelphia.