TY - JOUR
AU - Selvaggi, Suzanne M.
AB - Since its introduction by Papanicolaou in 19431 cytologic examination of cervical smears has led to early detection of cervical carcinoma and its precursors, with a reduction in both morbidity and mortality. And even though Koss has suggested that the cervical smear was perhaps the only effective cancer screening test,2 screening for cervical cancer has not eliminated the disease in any population studied. Reported estimates of the false-negative rate have ranged from 6% to 50% of all cervicovaginal samples.2,3 Three major sources of error include variability among clinicians obtaining the sample and the area of the cervix sampled; the cell collection and preparation techniques used; and errors in interpretation, with lack of diagnostic reproducibility for some cervical abnormalities. Detection and interpretation of cervical abnormalities depend on the methods used to obtain and prepare the sample. Sampling factors account for the majority of false-negative results.4 Cellular material must be collected from the squamocolumnar junction and the endocervix. Use of the cotton-tipped swab fell out of favor due to its cell-trapping effect.5 Increased cell sample size and improved cell preservation resulted when the cytobrush was used to obtain cells from the endocervix.6 The presence of endocervical cells/squamous metaplastic cells is a component of smear adequacy. Reported detection rates for cervical epithelial abnormalities are higher in smears containing these cell types.3,4 Today, clinicians obtaining cervical samples commonly use plastic spatulas, cytobrushes, and broom-brushes. In addition, new technologies are enabling better samples to be processed, improving the ability of a cytotechnologist or an automated screening system to detect cervical abnormalities. Two liquid-based thin-layer cell preparation technologies are available as alternatives to the conventional Pap smear.7,8 The goal of each system is to improve the quality of the cervical sample analyzed. Clinical trial7,8 and direct-to-vial9,10 studies using these techniques reported marked improvements in specimen adequacy and clarity, with fewer cervical samples obscured by blood and inflammation, compared with the Pap smear. Detection rates for squamous intraepithelial lesions (SILs) also have increased with these liquid-based technologies,9,10 although their impact on morbidity and mortality rates for cervical cancer remains undefined. Another important and clinically relevant issue involves diagnostic terminology and reproducibility. Nomenclature systems have evolved for cytologic and histologic diagnosis for cervical lesions. In 1953, Reagan and colleagues11 used the terms dysplasia (mild, moderate, severe) and squamous cell carcinoma in situ (CIS). In 1973, Richart12 used the term cervical intraepithelial neoplasia (CIN) to introduce the concept of cervical neoplasia as a disease continuum. In Richart's system, 3 categories existed: CIN1 (mild dysplasia), CIN2 (moderate dysplasia), and CIN3. Because both the cytologic and histologic differences between severe dysplasia and CIS were often subjective, CIN3 encompassed these 2 categories. This approach is the standard classification system for histologic diagnosis of cervical lesions today. However, the simultaneous use of the Pap, Reagan, and Richart classification systems, including multiple modifications of each, created a professional Tower of Babel. To address this issue, the National Cancer Institute (NCI) sponsored a workshop in 1988 to develop a uniform terminology system that was reproducible, would correlate with the histology of the lesion, and would facilitate communication between the laboratory and the clinician. The result was the Bethesda Nomenclature System for Cervicovaginal Cytology.13 By the late 1980s, human papillomavirus (HPV) was becoming established as an etiologic agent in the development of CIN and cervical carcinoma. Most low-risk HPV types were found in condylomas and CIN1 lesions. Most high-risk HPV types were associated with CIN2/CIN3 lesions. Since the differentiation of mild, moderate, and severe dysplasia/CIS had poor diagnostic reproducibility, the Bethesda conference participants defined only 2 clinically relevant categories: low-grade squamous intraepithelial lesion (LSIL) and high-grade squamous intraepithelial lesion (HSIL). The classification LSIL encompassed HPV, mild dysplasia, and CIN1; HSIL included moderate dysplasia, severe dysplasia, CIS, CIN2, and CIN3. The term atypical squamous cells of undetermined significance (ASCUS) was introduced at the first Bethesda workshop in 1988 and was retained by the second Bethesda conference in 1991.14 This category was intended to indicate a squamous epithelial abnormality for cases not diagnostic of a reactive/inflammatory, preneoplastic/neoplastic condition, effectively replacing the class II Pap smear. Furthermore, ASCUS cases were to be qualified to favor either a reactive or premalignant/malignant process. Unfortunately, the term has become somewhat of a catchall diagnosis. A recent study15 by the College of American Pathologists (CAP) Interlaboratory Comparison Program in Cervicovaginal Cytology (PAP) reported a 61% increase in the median ASCUS rate from 2.8% of all cervical samples in 1993 to 4.5% in 1996. In contrast, LSIL and HSIL rates have remained relatively constant. Potential reasons for the ASCUS rate increase include medicolegal concerns for false-negative cytology reports. Consistent lack of diagnostic reproducibility within ASCUS is a hot issue in the pathology and gynecology communities. The critical issue is determination of the clinical and management implications for patients with an ASCUS diagnosis on cervical cytology. Cases involving ASCUS and LSIL represent the largest volume of patients who require evaluation and clinical management. Interim guidelines for the management of abnormal cytology, in particular ASCUS and LSIL, were developed in 199416 when the NCI began the formal planning for the ASCUS-LSIL Triage Study (ALTS), which was designed to evaluate 3 alternative methods of managing patients with LSIL and ASCUS diagnoses.17 The ALTS study was a multicenter, randomized, prospective clinical trial that enrolled a large cohort of women aged 18 years and older according to strict eligibility criteria. Patients were randomized to 1 of 3 study groups: (1) immediate referral for colposcopy at enrollment, (2) use of HPV testing to triage to colposcopy, or (3) follow-up cytology only and referral for colposcopy with a diagnosis of HSIL. In this issue of THE JOURNAL, Stoler and Schiffman report findings from the ALTS study18 focusing on the interobserver reproducibility between the clinical centers' original cytologic and histologic diagnoses as compared to those of a Pathology Quality Control Group (QC). The authors examined the variation in diagnostic interpretation for 4948 monolayer cytology specimens, 2237 cervical punch biopsies, and 535 tissues obtained by the loop electrosurgical excision procedure (LEEP). Assuming sound statistical analysis of the large database, diagnostic reproducibility was virtually identical for the 3 specimen types, but showed substantial variability in interpretation. Cytologic diagnoses of ASCUS had the greatest variability. The QC group concurred with the clinical center diagnosis in only 43% of cases and downgraded 38.6% of the cases to negative. The data on the reproducibility of ASCUS are not surprising, as the readers must know and adhere to the cytologic criteria of ASCUS. It is paramount to limit the number of cases placed into this category. The ASCUS diagnosis is not a synonym for "I don't know," but rather must be qualified as to whether a reactive or premalignant process is favored. The role of HPV testing, alluded to by Stoler and Schiffman, will likely have impact on women whose cervical cytology showed ASCUS, favor SIL.19 A positive result on HPV testing would warrant colposcopy with possible biopsy/LEEP follow-up. A negative result would indicate routine cytologic examination. High-grade squamous intraepithelial lesions were another source of variation among reviewers. Concordance was reached in only 47.1% of the cases and the remainder were interpreted as LSIL or ASCUS. In a 2-year CAP PAP study, Woodhouse and colleagues20 examined the issue of interobserver variability in the subclassification of SILs. The discrepant rate between LSILs and HSILs ranged from 9.8% to 15% for cytotechnologists, pathologists, and the laboratory as a whole in comparison to the reference diagnosis. Confusion arose in differentiating mild dysplasia (LSIL) from moderate dysplasia (HSIL), pointing out the importance of adherence to cytologic criteria. In the study by Stoler and Schiffman, diagnostic reproducibility for cervical biopsies and LEEP specimens with CIN1 were equivalent to cytologic reproducibility. An interpretation of CIN1 by the clinical center pathologist was corroborated by the QC group in only 42.6% of the cases. Forty-one percent of the biopsy specimens originally diagnosed as CIN1 were interpreted as negative by the QC group. The findings were similar for LEEP specimens. These findings suggest that the histologic category of CIN1 is just as inflated as the cytologic category of ASCUS. To help ensure reproducibility and reduce interpretative variability for cervical specimens, diagnostic criteria for cytologic and histologic classification systems must be communicated frequently to the practicing community and incorporated into ongoing quality assurance programs. In addition, professional and educational societies such as the educational committee of the American Society of Cytopathology should provide well-supported recommendations for residency program committees in pathology to consider increasing the time residents spend on cytopathology rotations. The current 2 to 3 months in most programs is at best marginal, especially considering that new technologies such as HPV testing and other molecular studies use the same minimally invasive cytologic samples. Moreover, the Bethesda nomenclature system for cervicovaginal cytology has been in use for 10 years. It is time to critically evaluate and address terminology, diagnostic reproducibility, and the role of HPV testing in the management of ASCUS/LSIL cases. Although nominal cost HPV testing would allow triaging of all ASCUS cases, HPV analysis currently is too expensive for reflex testing and reimbursement for liquid-based technologies has just begun. The third NCI-sponsored Bethesda workshop is scheduled for April 2001, and will be followed by a clinical management conference. These issues of detection, interpretation, and management should be addressed with an evidence-based framework and a consensus reached on appropriate care for women with cervical cytologic abnormalities. References 1. Papanicolaou GN, Traut HF. Diagnosis of Uterine Cancer by Vaginal Smears. New York, NY: The Commonwealth Fund; 1943. 2. Koss LG. The Papanicolaou test for cervical cancer detection: a triumph and a tragedy. JAMA.1989;261:737-743.Google Scholar 3. Van der Graaf Y, Vooijs GP. False negative rate in cervical cytology. J Clin Pathol.1987;40:438-442.Google Scholar 4. Vooijs PG, Elias A, Van der Graaf Y. et al. Relationship between the diagnosis of epithelial abnormalities and the composition of cervical smears. Acta Cytol.1985;29:323-328.Google Scholar 5. Rubio CA. The false negative smear, II: the trapping effect of collecting instruments. Obstet Gynecol.1977;49:576-580.Google Scholar 6. Ros E, Jimenez Ayala M, Vilaplana E. et al. New technique for endocervical cytologic sampling with Stormby's brush. Cytologia (Tokyo).1983;3:9-20.Google Scholar 7. Lee KR, Ashfaq R, Birdsong GG. et al. Comparison of conventional Papanicolaou smears and a fluid-based thin-layer system for cervical cancer screening. Obstet Gynecol.1997;90:278-284.Google Scholar 8. Bishop JW, Bigner SH, Colgan JJ. Multicenter masked evaluation of AutoCyte PREP thin layers with matched conventional smears including initial biopsy results. Acta Cytol.1998;42:189-197.Google Scholar 9. Vassilakos P, Saurel J, Rondez R. Direct-to-vial use of AutoCyte PREP liquid-based preparation for cervical-vaginal specimens in three European laboratories. Acta Cytol.1999;43:65-68.Google Scholar 10. Corkill M, Knapp D, Martin J, Hutchinson ML. Specimen adequacy of ThinPrep sample preparations in a direct-to-vial study. Acta Cytol.1997;41:39-44.Google Scholar 11. Reagan JW, Seidemand IL, Saracusa Y. The cellular morphology of carcinoma in situ and dysplasia or typical hyperplasia of the uterine cervix. Cancer.1953;6:224-235.Google Scholar 12. Richart RM. Cervical intraepithelial neoplasia. Pathol Annu.1973;8:301-328.Google Scholar 13. National Cancer Institute Workshop. The 1988 Bethesda System for reporting cervical/vaginal cytological diagnoses. JAMA.1989;262:931-934.Google Scholar 14. Kurman RJ, Solomon D. The Bethesda System for Reporting Cervical/Vaginal Cytologic Diagnoses: Definitions, Criteria, and Explanatory Notes for Terminology and Specimen Adequacy. New York, NY: Springer-Verlag; 1994. 15. Davey D, Woodhouse S, Styer P. et al. Atypical epithelial cells and specimen adequacy: current laboratory practices of participants in the College of American Pathologists Interlaboratory Comparison Program in Cervicovaginal Cytology. Arch Pathol Lab Med.2000;124:203-211.Google Scholar 16. Kurman RJ, Henson DE, Herbst AL. et al. Interim guidelines for management of abnormal cervical cytology. JAMA.1994;271:1866-1869.Google Scholar 17. Schiffman M, Adrianza ME. ASCUS-LSIL Triage Study: design, methods and characteristics of trial participants. Acta Cytol.2000;44:726-742.Google Scholar 18. Stoler MH, Schiffman M.for the ALTS Group. Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL Triage Study. JAMA.2001;285:1500-1505.Google Scholar 19. Crum CP, Genest DR, Krane JF. et al. Subclassifying atypical cells in Thin-Prep cervical cytology correlates with detection of high-risk human papillomavirus DNA. Am J Clin Pathol.1999;112:384-390.Google Scholar 20. Woodhouse SL, Stastny JF, Styer PE. et al. Interobserver variability in subclassification of squamous intraepithelial lesions: results of the College of American Pathologists Interlaboratory Comparison Program in Cervicovaginal Cytology. Arch Pathol Lab Med.1999;123:1079-1084.Google Scholar
TI - Implications of Low Diagnostic Reproducibility of Cervical Cytologic and Histologic Diagnoses
JF - JAMA
DO - 10.1001/jama.285.11.1506
DA - 2001-03-21
UR - https://www.deepdyve.com/lp/american-medical-association/implications-of-low-diagnostic-reproducibility-of-cervical-cytologic-sQjq1cGa1i
SP - 1506
EP - 1508
VL - 285
IS - 11
DP - DeepDyve
ER -