© 2003 by Oxford University Press
Journal of the National Cancer Institute Monographs, No. 31, 72-79,
2003
© 2003 Oxford University Press
ARTICLE |
Chapter 11: Future Directions in Cervical Pathology
Affiliation of author: Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD.
Correspondence to: Mark E. Sherman, M.D., National Institutes of Health, 6120 Executive Blvd., Rm. 7080, Bethesda, MD 20892-7374 (e-mail: shermanm{at}mail.nih.gov).
 |
ABSTRACT |
|---|
 TopNotes Abstract OverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Technologic innovations, including the development of improved sampling devices, liquid-based collection systems, and computer-assisted screening, have revolutionized cervical cytology. Biologic discoveries in upcoming years promise to transform our understanding of the pathogenesis of cervical neoplasia, leading to another quantum leap in our approach to screening and prevention. This review summarizes the opportunities and challenges that recent and anticipated advances in pathology present for epidemiologic research.
|
OVERVIEW |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
This review highlights the importance of recent technologic innovations in planning future research on cervical cancer (1). The first sections are organized by study design and objectives. The latter sections address issues related to specific technologies, pathologic classifications, and the need for multidisciplinary studies to address basic research questions.
|
PRIMARY SCREENING STUDIES |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Extremely sensitive single screens may permit a lengthening of the screening interval for low-risk women, thereby allowing allocation of additional resources for screening patients at the greatest risk. To this end, randomized clinical trials are needed to determine the cost-effectiveness of combined primary screening using cytology and human papillomavirus (HPV) testing. Ideally, these studies should address different ways of combining screening technologies over the span of a lifetime to achieve cancer prevention rather than focusing on the detection of precursors at a single point in time or during short-term follow-up. This paradigm shift would place greater emphasis on the negative predictive value of screening as opposed to sensitivity. At different ages, the optimal approach based on test performance, cancer risk, and HPV prevalence might favor the use of cytology alone, HPV testing alone, or combined testing. The cost-effectiveness of using conventional as opposed to liquid-based cytology (LBC) in combination with HPV testing should be assessed, given that the high sensitivity of HPV testing may obviate the need for a more expensive albeit a more sensitive cytologic technique. Future screening studies should analyze the results by age, because both cancer risk and test performance are age dependent. Specifically designed studies will be required to assess new screening approaches in populations with special concerns (e.g., immunosuppression or pregnancy).
Data suggest that conventional smears mainly detect prevalent disease but have minimal value in identifying women who are destined to develop cervical neoplasia in the future. In contrast, HPV testing demonstrates both high cross-sectional sensitivity and utility in identifying women at the highest risk for developing cervical neoplasia in the future (2). More important, a negative HPV test result provides strong reassurance that a woman is unlikely to develop carcinoma for several years. Studies are needed to compare the performance of LBC to HPV testing in predicting the future risk of cervical neoplasia. Bimodality screening studies should compare the sensitivity of combined testing to the improvement predicted by chance alone (3) and, most important, should assess the clinical relevance of the absolute gain in sensitivity that is achieved.
The value of self-collected specimens for HPV testing has recently received considerable attention (4), but less emphasis has been placed on developing protocols for combined testing, despite historic evidence suggesting that self-collected samples may be useful for cytologic examination (5). Implementation of effective self-collected screening may reduce the number of unscreened women, may improve compliance with follow-up, and may increase cost-effectiveness. Research is needed to develop improved self-collection devices, transport media, and educational tools for patients. The possibility of using multiple pooled samples collected at several different times and the effect of experience on self-collection should also be explored.
|
COLPOSCOPY TRIAGE STUDIES |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
The Atypical Squamous Cells of Undetermined Significance (ASCUS)/Low-Grade Squamous Intraepithelial Lesion (LSIL) Triage Study (ALTS) demonstrated that testing for oncogenic types of HPV DNA is useful for determining which women with ASCUS smears harbor an underlying cervical intraepithelial neoplasia 3 (CIN3) and, therefore, require colposcopy (6). In contrast, HPV testing was not useful for managing women with LSILs because more than 80% of the patients tested positive (7). In assessing the general applicability of the conclusions of ALTS, two issues require consideration: 1) The enrollment was based on the original interpretations of smears rather than LBC and 2) eligibility was based on reports from community laboratories located exclusively within the United States.
Data suggest that HPV testing is equally useful for managing ASCUS, irrespective of whether conventional cytology or LBC has been used. Both the frequency of ASCUS interpretations (8) and the association of ASCUS with HPV detection (9) seem to be comparable with the two cytologic methods. Studies to determine whether preparing a second thin-layer slide from the same liquid-based collection has utility in reducing the frequency of equivocal cytology reports are needed; available data are sparse and conflicting (10,11). Many ASCUS interpretations reflect the reluctance of pathologists to commit to a specific interpretation based on few atypical cells. Examining a second slide may provide the additional evidence required for a definitive classification of squamous intraepithelial lesions (SILs) in some cases.
Internationally, the systematic differences in diagnostic thresholds among pathologists (12) and the use of classifications other than the Bethesda System complicate the direct translation of ALTS conclusions to clinical practice outside the United States. Therefore, countries either will have to launch their own studies to evaluate the clinical utility of HPV testing or will need to calibrate their cytologic interpretations with readings in ALTS and develop projected results. Economic considerations favor the latter approach, which could potentially be achieved through a web-based resource or slide exchange.
The use of HPV testing in clinical practice combined with a microscopic review of cytologic slides might allow pathologists to hone their interpretations by using the HPV test as a gold standard. Optimistically, this could lead to a reduction in the frequency of ASCUS reporting by training pathologists to classify slides with specific morphologic patterns as normal based on accumulated experience showing that such appearances are associated with negative HPV tests. Longitudinal studies are required to evaluate the trends in reporting frequencies using the 2001 Bethesda classification and the association between the subcategories of atypical squamous cells (ASC) ("undetermined significance [ASC-US]") and "cannot exclude HSIL [high-grade squamous intraepithelial lesion] (ASC-H)," HPV detection, and age (13). Specifically, it is hoped that the elimination of the category ASCUS favor reactive in the revised Bethesda System will result in fewer equivocal cytologic reports. Similarly, routine use of HPV testing for colposcopy referral of women with ASC-US in combination with reduced misclassification of LSILs (now strongly associated with oncogenic HPV detection) may increase the expectations of finding CIN, leading to lowered thresholds among colposcopists for performing biopsies and the use of less stringent criteria for diagnosing CIN1 among histopathologists. As with the implementation of LBC and other new technologies, it is imperative for investigators and practitioners to monitor performance over time.
Data from ALTS demonstrate that the risk of future detection of CIN3 among women with negative colposcopic examinations, negative biopsy specimens, or histologic CIN1 is similar (Schiffman M: personal communication). This result should permit follow-up without the immediate treatment of women with all of these results, which would be reassuring given the poor reproducibility of the histologic diagnosis of CIN1 (in contrast to the cytologic interpretation of LSIL) (14) and the risk of unnecessary treatment attendant to false-positive diagnoses. The use of HPV testing may result in a worrisome increase in the performance of the loop electrosurgical excision procedure (LEEP), even without demonstrable CIN2 or worse, secondary to concerns about colposcopically unrecognized disease. Studies are required to monitor the age-specific rates for the performance of LEEPs and the diagnostic findings in these specimens. These data could be generated by using large surveys or perhaps by developing a population-based registry for tracking CIN, treatment patterns, and the associated short- and long-term complications of treatment, especially with respect to miscarriage rates.
|
ENDOCERVICAL ADENOCARCINOMA |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Although cervical screening has reduced the mortality from cervical cancer overall in developed nations, screening is of unproven benefit in preventing fatal cervical adenocarcinomas. Historically, adenocarcinoma in situ (AIS) was often detected incidentally during a work-up for a concurrent squamous lesion that was identified cytologically. Development and widespread dissemination of cytologic criteria for AIS are relatively recent occurrences, and the category of AIS was not formally included in the Bethesda System until the 2001 revision (13). Histopathologically, the distinction between endocervical and endometrial adenocarcinoma may be difficult, further complicating etiologic studies (15). Although atypical endocervical cells (atypical glandular cells, endocervical type) are more often associated with high-grade lesions than ASC-US, the majority of these display squamous rather than glandular differentiation [reviewed in (16)]. Adenocarcinoma has been identified disproportionately among "rapidly progressing" cancers (17), but it is unclear what proportion of these cases reflected ineffective screening as opposed to aggressive biology. Unlike squamous carcinomas, HPV type 18 is found approximately as often as HPV type 16 (HPV16) in these tumors, and it is the predominant type in some studies.
AIS is the recognized precursor of invasive adenocarcinoma, but unlike squamous carcinoma, milder forms of glandular dysplasia, analogous to CIN1 and 2, have not been widely accepted as a cytologic, histopathologic, or biologic entity. These attributes, in combination with the relatively subtle cytologic and histopathologic features in some cases, difficulties in sampling, and less experience among pathologists in recognizing AIS, have historically compromised the value of cytologic screening. Descriptive studies of endocervical adenocarcinoma should be performed to determine whether improved sampling of the endocervix, dissemination of cytologic criteria, and formal inclusion of AIS in the Bethesda System have increased the ratio of in situ to invasive lesions and led to improved survival. Molecular characterizations of benign-appearing endocervical epithelium and adjacent AIS should be performed to elucidate the pathogenesis of cervical adenocarcinoma. Benign endocervical cells express receptors for estrogen and progesterone, whereas endocervical adenocarcinomas generally do not, suggesting that, if hormonal risk factors are etiologically related to the development of these tumors, they must act early in the pathogenesis of these neoplasms, primarily affect the virus, or produce indirect effects. Molecular epidemiologic studies that include a prominent pathology component are needed to better understand the relationships between hormonal exposures and adenocarcinoma.
A cross-sectional study of 137 consecutive women with smears reported as atypical glandular cells of undetermined significance found that testing for oncogenic HPV types detected 94% of high-grade lesions (AIS and CIN2 or worse), with a positive predictive value of 41% (18). These results are promising, but larger studies with follow-up are required to demonstrate the utility of HPV testing in this setting. In addition, better means of post-treatment surveillance are needed. Currently, hysterectomy is the preferred management for women with AIS, because it is difficult to exclude residual disease after cone biopsy and the sensitivity of follow-up is unknown. A demonstration that LBC, HPV testing, or another screening modality would permit safe post-treatment follow-up for AIS would enable young patients to safely choose conservative management if preservation of fertility was desired. Clinicopathologic studies have generally concluded that LBC performs as well or better than conventional cytology with respect to the identification of glandular lesions (19–23). However, most of these studies were designed to retrospectively analyze the outcomes in women with atypical glandular cell (AGC) cytology reported at a single institution and, therefore, have included few cases of AIS or invasive endocervical adenocarcinoma. Furthermore, these studies have generally lacked sufficient follow-up to ensure complete ascertainment of cases and have not collected epidemiologic data or biologic specimens for HPV testing or other assays. Although the insensitivity of cytologic screening tends to dominate clinical discussions, many women with AGCs have benign reactive changes, suggesting that overtreatment is an important public health concern given that large cones are often required to exclude glandular neoplasia.
|
NATURAL HISTORY STUDIES |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Repeated cytologic examinations provide a method for studying the natural history of cervical neoplasia with relatively little disruption of the lesions under study. Whereas biopsies remove the full thickness of the epithelium and induce substantial inflammation and wound repair with epithelial regeneration, sampling for cytology removes primarily the superficial layers of the epithelium. Although cytologic testing may have less effect on natural history than performing biopsies, cytologic samples have some limitations for research. Alterations that occur at the base of the epithelium cannot be effectively assessed using cytology; therefore, tissue-based studies may be required to conclusively address certain fundamental questions, such as whether latent HPV infections may persist undetected in the basal epithelium for many years. An important consideration in evaluating markers in cytologic samples is that the assays must detect these markers in a mixed-cell population consisting mainly of surface epithelium.
Insensitive baseline testing may bias natural history data because delayed detection of prevalent disease may masquerade as incident disease when finally discovered. Conclusions about the future risk of cervical neoplasia associated with the detection of HPV, high viral load, or other markers at baseline among "normal" women are limited by the methods used to exclude an abnormality at the outset. Insensitive baseline testing may lead to inflated assessments about the value of biomarkers for identifying women at the greatest risk of developing future disease. Optimized screening methods may also improve the recognition and the grading of incident precursors, especially when the severity of the disease has been underestimated colposcopically or histologically. In the future, the development and validation of devices that permit repeated sampling of a specific anatomic location on the cervix may shift the focus of natural history studies from outcomes in subjects to the behavior of individual lesions.
|
NEW TECHNOLOGIES: LIQUID-BASED CYTOLOGY |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Liquid-based cytology offers several theoretic advantages over conventional Pap smears, including a more complete collection of exfoliated cells, random and presumably more representative transfer of exfoliated cells to slides, and improved microscopic visualization attributable to reduced overlapping, obscuring blood, and inflammation. Cell suspensions remaining after thin-layer slide preparation are suitable for ancillary studies (e.g., HPV DNA testing), ensuring that testing has been performed on a representative specimen. A systematic comparison of conventional cytology and LBC found that more LSILs and HSILs were identified with LBC in all eight direct-to-vial studies (the entire sample used for LBC) reviewed, with a summary odds ratio (OR) of 2.15 (95% confidence interval [CI] = 2.05 to 2.26) for LSILs and an OR of 2.26 (95% CI = 2.06 to 2.47) for HSILs (8). Based on limited evidence, two meta-analyses (24,25) found that LBC is probably more sensitive but less specific than smears.
The success of Pap smear screening in preventing cervical cancer in developed countries argues for a conservative approach in implementing new screening techniques. Nonetheless, the gain in sensitivity and the logistical advantage of having a representative specimen for ancillary testing support the use of LBC, provided that implementation of LBC is realistic in the target population under study and that the preparation and interpretation of LBC specimens can be optimized.
Investigators must recognize that introducing LBC into pathology laboratories that have previously relied on smears is associated with a learning curve. Accordingly, LBC results should be monitored over time to detect systematic shifts in performance. In addition, the optimization of LBC may require clinicians to change cervical samplers, which can affect both the appearance and the quality of the specimens. Therefore, appropriate training and quality-assurance monitoring are important when implementing these methods in studies or in clinical practice. Finally, future population-based analyses are needed to determine the impact of LBC on cancer incidence and mortality. The data for adenocarcinoma should be given particular scrutiny, given that the incidence of this tumor is increasing among young white women in several industrialized nations (26) and that the features of endocervical lesions in LBC may be subtle compared with smears.
|
NEW TECHNOLOGIES: LOGISTICAL CONSIDERATIONS |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Epidemiologic studies require large-scale, stable, long-term specimen storage in suitable facilities equipped with sophisticated tracking procedures. Improved techniques for processing and storing exfoliative cellular specimens are needed. For example, techniques to convert cell suspensions to cell pellets suitable for storage in microtubes at -80°C would facilitate retrospective molecular testing of multiple specimens collected from subjects at specific time points with established outcomes. However, such approaches would not be suitable for morphologically based assays (e.g., immunohistochemistry and in situ hybridization), which would most likely require the preparation and storage of material as slides. Unanswered questions about the possible loss of immunoreactivity on cytologic and histologic slides after storage may be a concern (27). As emphasis shifts from HPV testing alone to include other molecular assays, critical issues related to specimen preparation and storage will emerge. Methodologic studies are needed urgently so that future studies with intensive pathology components can move forward. Data suggest (Castle PE, Solomon D, Hildesheim A, Herrero R, Bratti MC, Sherman ME, et al: unpublished data) that current collection methods do not provide optimal morphology and preservation of DNA after several years of storage at room temperature. The development of a liquid collection medium that provided excellent morphology and preservation of macromolecules is needed.
|
NEW TECHNOLOGIES: AUTOMATED SCREENING |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Clinically available instruments for automated screening have been used mainly to reduce the labor required for manual screening rather than to improve interpretation (28). Wilbur et al. (29) compared the results of manual screening and a protocol in which only slides considered to be potentially abnormal on the basis of review of 10 computer images were manually screened. Expert interpretation of 1275 thin-layer slides served as the reference. The goal of the automated approach was to reduce the number of slides requiring manual screening without sacrificing accuracy. Although this slide set was highly enriched for abnormal slides, only 438 (34%) slides were referred to the pathologist using the automated approach, including 98% of the slides classified as HSILs. For comparison, 91% of the slides classified as HSILs were referred to that pathologist using manual screening of every slide. Emerging evidence suggests that automated cytology may be useful in large studies by providing rapid, accurate, and standardized screening while reducing labor and costs.
In the future, automated screening instruments that provide a visual record of cells that have been identified as potentially abnormal may permit expert cytologic reviews without microscopic re-examination of glass slides. New devices are being evaluated that would guide the screeners to microscopic fields containing potentially abnormal cells. To date, interpretations based on computer-selected images have not always agreed well with microscopic classification, even though automated instruments have proved to be effective in selecting slides that require microscopic examination (29). However, the development of automated systems that improve the classification of slides is an area of active research. The invention of an automated screening technique that would provide a same-day result could have important benefits, especially in rural environments with limited access to medical care. Interactive systems may also provide ongoing training for users with limited access to collegial consultation; however, the performance of the automated instruments themselves will also require ongoing quality-assurance monitoring. Finally, automated systems may be sensitive to variations in slide preparation, which are difficult to standardize when slides are prepared in multiple laboratories.
|
NEW TECHNOLOGIES: STANDARDS FOR EVALUATION |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Cytologic screening serves two main functions: 1) identifying women who have a prevalent cancer precursor and require treatment and 2) identifying at-risk women with oncogenic HPV infections that have not progressed to cancer precursors.
The performance of cytology in identifying women with oncogenic HPV infections is best assessed by comparison with a sensitive HPV DNA test. Neither colposcopy nor histology represents a suitable gold standard. The colposcopic distinction between HPV-related changes and squamous metaplasia is unreliable (30) and the histologic diagnosis of CIN1 is also irreproducible (14).
Assessing the performance of cytology in identifying cancer precursors is complex for at least two reasons. First, the exact definition of a true cancer precursor is undefined. Second, we lack completely sensitive methods for detecting putative precursors in cross-sectional studies.
Most histologic CIN3 lesions probably represent cancer precursors, especially if confirmed by an expert panel review. Evidence for regression of these lesions is sparse, although natural history data are limited because the long-term follow-up of these lesions would be unethical. However, many clinically detected CIN3 lesions in developed countries are extremely small and are diagnosed among women less than 30 years old, an age range in which invasive cancer is extremely rare (31). Accordingly, most CIN3 lesions in screened populations do not represent "proximate" or "immediate" precursors because carcinomas typically develop a decade or more later among these women. The use of rigorously confirmed CIN3 as a study end point is challenging because these lesions are relatively rare. The use of CIN2 and above as a study end point increases the statistical power and also reflects the threshold for treatment in most developed nations; however, a major disadvantage of this approach is that many CIN2 lesions do not represent true precursors. A substantial proportion of CIN2 lesions spontaneously regress (32), suggesting that cytologic HSIL (which includes moderate dysplasia) and histologic CIN2 represent hybrid categories that amalgamate reversible HPV infections and cancer precursors. Further studies to determine the natural history and biologic characteristics of CIN2 are needed.
The second challenge in defining the performance of cytology in detecting precursors is that even multimodality evaluations of the cervix are incompletely sensitive. Emerging data demonstrate that colposcopic examination with histologic sampling may miss cases of CIN2 and 3, which can profoundly affect the assessment of screening tests (3). Paradoxically, more accurate ascertainment of prevalent disease requires a follow-up period to identify cases missed at initial evaluation. When cases are detected during follow-up, it may be difficult to determine whether they are incident or missed prevalent cases. It may be possible to provide informed guesses about whether a case is incident or prevalent by assessing the size of the lesion, using data from repeated HPV testing to determine the interval of type-specific persistence before the diagnosis, and review of preceding pathology to identify false-negative interpretations. However, determinations of whether a lesion is incident or prevalent remain probabilistic.
Finally, clinical experience suggests that cytologic interpretations of HSILs are often accurate, even when histologic confirmation is lacking, suggesting that women with discordant results require additional follow-up. The conclusion that a cytologic interpretation of HSIL is erroneous should be supported by evidence from pathologic review, follow-up, HPV testing, or other data.
|
CYTOPATHOLOGIC AND HISTOPATHOLOGIC CLASSIFICATIONS: DISSONANT STRAINS |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Cytopathology and histopathology are imperfect measures of the same biologic processes and, ideally, these disciplines should share a single classification reflecting a single biologic truth. Unfortunately, classifications for cytopathology and histopathology have become increasingly standardized but not unified. The use of the term HSIL has been resisted by histopathologists on the basis of a belief that CIN2 and 3 are more reliably distinguished on histology than on cytology and that CIN2 frequently regresses, whereas CIN3 does not. This distinction is sometimes sought in determining whether to treat young women, although proof of its reliability may be lacking. Furthermore, colposcopists, cytopathologists, and histopathologists use similar or identical terms to refer to different entities, further muddling the debate over terminology (Fig. 1
) (9,33).
|
Even if pathologists agreed on classification systems and diagnostic criteria, subjective assessments of morphologic appearances would continue to limit interobserver agreement. The data from ALTS demonstrated that the reproducibility of cytology and that of histology are similar, although the problem areas differ (14). In cytology, the categories of ASCUS and HSIL were especially problematic, whereas in histopathology, CIN1 was a major source of disagreement. Although data have demonstrated that cytologic reproducibility is poor, even when an atlas of classical images is provided as a guide (34), new approaches for improving reproducibility are needed. In particular, validated training sets of slides that emphasize lesions at the border zone between categories rather than straightforward examples should be tested. The development of these slide sets and analytic methods for defining which borderline lesions are most appropriately upgraded or downgraded would be useful. The ongoing development of a web-based atlas illustrating criteria for entities in the Bethesda 2001 System may promote better standardization of criteria, both within the United States and in other countries. Sequential improvements in our biologic understanding of cervical neoplasia have been reflected in successive generations of cytologic classifications, but the challenge of classifying cells that lie on a morphologic continuum remains daunting (Fig. 2
). Perhaps interactive tools may be developed in the future that would allow pathologists to manipulate the appearance of virtual cells and then receive immediate feedback on the association of such changes with regard to risk of disease, HPV infection, or significant molecular alterations.
|
Few studies have addressed whether pathologists can distinguish lesions on the basis of HPV type. In a study in which 200 "atypical" smears were reviewed by a five-member panel, the detection of oncogenic HPV infections increased progressively with the increasing severity of the reclassifications from negative to ASCUS to SIL (37). However, detection of low-risk HPV types was similar across all categories for each reviewer, suggesting that cytologic interpretations are dictated almost exclusively by the presence or the absence of oncogenic HPV types. In a histopathologic study, hyperchromatic markedly enlarged nuclei were particularly characteristic of HPV types 16, 33, and 35, whereas these features were less common in HPV types 6, 11, 18, 31, and 45, even in CIN1 lesions. In addition, HPV capsid protein was detected less often in CIN1 and 2 lesions associated with HPV types 16 and 31. These observations suggested that the HPV types with high oncogenic potential are associated with more pronounced morphologic abnormalities and less prominent productive infections (38). Further studies are needed to clarify the relationship between HPV type and morphology, including HPV type variants.
|
TOWARD A UNIFIED CLASSIFICATION OF PATHOLOGY: THE HOLY GRAIL OF MOLECULAR BIOLOGY |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
Long-term natural history studies of persistent CIN2 and 3 would be considered unethical in most settings, limiting the availability of natural history data for defining the characteristics of "inevitable" or "immediate" cervical cancer precursors. An alternative approach would be to identify a molecular signature of invasive carcinoma and then perform cross-sectional studies on precursor lesions to determine which seem to have progressed from HPV infection to neoplasia on the basis of molecular similarities to cancer. The mechanistic importance of candidate markers identified in cross-sectional studies could be evaluated in prospective studies that tested whether such markers predicted the persistence of HPV infection or progression to more severe pathologic lesions.
The list of available techniques for molecular profiling of DNA, RNA, protein, and methylation status is growing, but currently most of these techniques work best on flash-frozen tissue. This situation presents technical challenges, especially for studying lesions that can be recognized only after histopathologic processing and microscopic examination. Improved methods for microdissection of tissues and cells combined with fixation techniques that preserve nucleic acids may increase the opportunities to characterize molecular changes in precursor lesions (39). Although many studies have related biomarkers to the cytologic and histopathologic interpretation of specimens, more studies are needed in which the morphology of individual cells is meticulously compared with the molecular markers that these cells express. However, many technical challenges remain unresolved. Finally, support for these approaches requires a paradigm shift; specifically, it requires researchers to move beyond studies heavily focused on the morphologic continuum of precursors and HPV biology and to develop classic genetic models of cervical carcinogenesis similar to those constructed to understand other tumor types.
Multidisciplinary approaches will be required to integrate the molecular findings into useful clinical paradigms. In particular, cross-sectional studies that compare results for a particular marker with the consensus pathologic classification of cytologic or pathologic material often represent the first approach to assessing the potential value of a new assay. Unfortunately, concordance between the marker and the pathologic classification does not imply utility, only consistency with existing methods. In contrast, the discordant cases are more informative about the marker, but without systematic follow-up or another means for objectively classifying these lesions, the results are not interpretable. Even when a marker seems to be promising in trials, robust, high-throughput assays that can be applied to clinical material of heterogeneous quality must be developed, an effort that may not be a priority for basic sciences laboratories.
Apart from HPV DNA testing, assays for numerous markers have been proposed for use in either screening or predicting progression of HPV infections to cancer precursors, including HPV integration into the host genome, expression of p16, telomerase, 3q amplification, measurement of HPV transcripts and their ratios, proliferation (Ki67), ploidy, and others. None of these markers has attained widespread routine implementation in clinical practice. A critical summary of candidate biomarkers is beyond the scope of this review. Readers are referred to the literature for specific references.
|
TRANSFORMATION ZONE |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
A basic question that remains unresolved about cervical cancer is why precursor lesions seem to develop at the transformation zone. Recognition that cancer frequently develops at sites of metaplasia has led to advances in our understanding of the etiology of several different tumor types. Metaplasia seems to represent a response to chronic irritation in many organs, but in the cervix, metaplasia is a ubiquitous finding among sexually active women. Nonetheless, the low incidence of dysplasia in the vagina (which normally lacks glandular epithelium) compared with the cervix provides strong circumstantial evidence that understanding the properties of the transformation zone and its interactions with HPV may provide clues to understanding the pathogenesis of cervical cancer. A prospective study of young women (40) found that the rate of active metaplasia rather than the extent of the transformation zone was a risk factor for incident LSIL, suggesting that dynamic changes in the transformation zone over time may affect the cellular manifestations of HPV infection. Molecular pathology will be needed to study the microenvironment of the transformation zone and its relationship to risk of cervical disease.
The size and anatomic location of the squamocolumnar junction (SCJ) are not static; the process of squamous metaplasia leads to continuous remodeling (41). The SCJ is generally located more proximally within the endocervical canal among older women, whereas a more distal position is associated with increasing parity. It has been speculated that the association between higher parity and cancer risk reflects increasing exposure of the SCJ to carcinogenic influences and that decreasing parity may account for part of the observed decline in cervical cancer incidence in the early 20th century. Trauma at delivery, hormonal or immunologic changes during pregnancy, or other factors may mediate the effects of parity on the SCJ. Studies using HPV16-transgenic mice suggest that low-dose estrogen exposure may promote multifocal squamous metaplasia, which progresses to CIN and then to carcinoma (42). HPV promoter regions also contain a steroid hormone-binding motif, raising the possibility that hormones may affect transcription of viral genes. The role of hormones as an HPV cofactor deserves further study, given that oral contraceptive use may increase the risk of cervical cancer and that some studies suggest that hormonally mediated exposures (e.g., obesity) may increase the risk of adenocarcinoma in particular (See chapter 1, F. X. Bosch and S. de Sanjosé). Topical estrogen is frequently administered to postmenopausal women with cervical atrophy and equivocal cytologic reports. Prospective systematic studies of these women may elucidate the effect of hormones on the transcription of HPV and human genes, local and systemic immunity, and other factors.
Studies that might increase our understanding of the biology of the SCJ include 1) a comparison of the morphology and gene expression of metaplastic cells between HPV-infected and noninfected women, 2) the localization of specific HPV-binding sites, 3) the assessment of cell turnover in the SCJ as a risk factor for HPV infection or CIN, and 4) the assessment of local immunologic factors that might favor the transformation zone as a site for HPV infection, replication, or transformation. Vaccine trials may provide important clues by allowing comparisons between the transformation zone of the vaccinated and placebo groups. These studies may also provide opportunities to increase our knowledge about immune effector cells, local and systemic antibody production, cytokines, and other factors associated with protection from HPV-induced cervical neoplasia. Identification of these factors may enhance our ability to either prevent HPV infection or minimize the risk of neoplastic events.
|
NOTES |
|---|
The term atypical squamous cells of undetermined significance (ASCUS) has been replaced with the term atypical squamous cells (ASC) in the Bethesda 2001 System and has been subdivided into ASC of undetermined significance (ASC-US), and ASC cannot exclude HSIL (ASC-H). The term atypical glandular cells of undetermined significance (AGUS) has been replaced with the term atypical glandular cells (AGC) in the Bethesda 2001 System.
|
REFERENCES |
|---|
|
TopNotesAbstractOverviewPrimary Screening StudiesColposcopy Triage StudiesEndocervical AdenocarcinomaNatural History StudiesNew Technologies: Liquid-Based...New Technologies: Logistical...New Technologies: Automated...New Technologies: Standards for...Cytopathologic and...Toward a Unified Classification...Transformation ZoneReferences |
|---|
1 Stoler MH. Advances in cervical screening technology. Mod Pathol 2000;13:275–84.[CrossRef][Web of Science][Medline]
2 Sherman ME, Lorincz AT, Scott DR, Wacholder S, Castle PE, Glass AG, et al. Baseline cytology, human papillomavirus testing, and risk for cervical neoplasia: a 10-year cohort analysis. J Natl Cancer Inst 2003;95:46–52.
3 Franco EL. Statistical issues in human papillomavirus testing and screening. Clin Lab Med 2000;20:345–67.[Web of Science][Medline]
4 Wright TC Jr, Denny L, Kuhn L, Pollack A, Lorincz A. HPV DNA testing of self-collected vaginal samples compared with cytologic screening to detect cervical cancer. JAMA 2000;283:81–6.
5 Bernstein A, Vitner S, Webber JM. Evaluation of a new tampon device for cytologic autocollection and mass screening of cervical cancer and its precursors. Am J Obstet Gynecol 1985;151:351–5.[Web of Science][Medline]
6 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.
7 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.
8 Bernstein SJ, Sanchez-Ramos L, Ndubisi B. Liquid-based cervical cytologic smear study and conventional Papanicolaou smears: a metaanalysis of prospective studies comparing cytologic diagnosis and sample adequacy. Am J Obstet Gynecol 2001;185:308–17.[CrossRef][Web of Science][Medline]
9 Sherman ME, Solomon D, Schiffman M. Qualification of ASCUS. A comparison of equivocal LSIL and equivocal HSIL cervical cytology in the ASCUS LSIL Triage Study. Am J Clin Pathol 2001;116:386–94.
10 Hoerl HD, Wagner JL, De Las Casas LE, Shalkham JE, Hafez GR, Kurtycz DF. Utility of additional slides from residual PreservCytTM material in difficult ThinPrepR gynecologic specimens: a prospective study of 58 cases. Diagn Cytopathol 2001;25:141–7.[CrossRef][Web of Science][Medline]
11 Massarani-Wafai R, Bakhos R, Wojcik EM, Selvaggi SM. Evaluation of cellular residue in the ThinPrepR PreservCytTM vial. Diagn Cytopathol 2000;23:208–12.[CrossRef][Web of Science][Medline]
12 Scott DR, Hagmar B, Maddox P, Hjerpe A, Dillner J, Cuzick J, et al. Use of HPV DNA testing to compare equivocal cervical cytologic interpretations in the United States, Scandinavia, and the United Kingdom. Cancer Cytopathol 2002;96:14–20.
13 Solomon D, Davey D, Kurman R, Moriarty A, O’Connor D, Prey M, et al. The Bethesda System 2001: terminology for reporting the results of cervical cytology. JAMA 2002;287:2114–9.
14 Stoler MH, Schiffman M. Toward optimal laboratory use. Interobserver reproducibility of cervical cytology and histologic interpretations. Realistic estimates from the ASCUS–LSIL Triage Study. JAMA 2001;285:1500–5.
15 Staebler A, Sherman ME, Zaino RJ, Ronnett BM. Hormone receptor immunohistochemistry (ISH) and HPV in situ hybridization are useful for distinguishing endocervical and endometrial adenocarcinomas. Am J Surg Pathol 2002;26:998–1006.[CrossRef][Web of Science][Medline]
16 Veljovich DS, Stoler MH, Andersen WA, Covell JL, Rice LW. Atypical glandular cells of undetermined significance: a five-year retrospective histopathologic study. Am J Obstet Gynecol 1998;179:382–90.[CrossRef][Web of Science][Medline]
17 Hildesheim A, Hadjimichael O, Schwartz PE, Wheeler CM, Barnes W, Lowell DM, et al. Risk factors for rapid-onset cervical cancer. Am J Obstet Gynecol 1999;180:571–7.[CrossRef][Web of Science][Medline]
18 Ronnett BM, Manos MM, Ransley J, Fetterman B, Kinney W, Hurley L, et al. Atypical glandular cells of undetermined significance (AGUS): cytopathologic features, histopathologic results and HPV DNA detection. Hum Pathol 1999;30:816–25.[CrossRef][Web of Science][Medline]
19 Ashfaq R, Gibbons D, Vela C, Saboorian MH, Iliya F. ThinPrep Pap test. Accuracy for glandular disease. Acta Cytol 1999;43:81–5.[Web of Science][Medline]
20 Bai H, Sung CJ, Steinhoff MM. ThinPrep Pap test promotes detection of glandular lesions of the endocervix. Diagn Cytopathol 2000;23:19–23.[CrossRef][Web of Science][Medline]
21 Hecht JL, Sheets EE, Lee KR. Atypical glandular cells of undetermined significance in conventional cervical/vaginal smears and thin-layer preparations. Cancer 2002;96:1–4.[Web of Science][Medline]
22 Roberts JM, Thurloe JK, Bowditch RC, Humcevic J, Laverty CR. Comparison of ThinPrep and Pap smear in relation to prediction of adenocarcinoma in situ. Acta Cytol 1999;43:74–80.[Web of Science][Medline]
23 Wang N, Emancipator SN, Rose P, Rodriguez M, Abdul-Karim FW. Histologic follow-up of atypical endocervical cells. Liquid-based, thin-layer preparation vs. conventional Pap smear. Acta Cytol 2002;46:453–7.[Web of Science][Medline]
24 Nanda K, McCrory DC, Myers ER, Bastian LA, Hasselblad V, Hickey JD, et al. Accuracy of the Papanicolaou test in screening for and follow-up of cervical cytologic abnormalities: a systematic review. Ann Intern Med 2000;132:810–9.
25 Sulik SM, Kroeger K, Schultz JK, Brown JL, Becker LA, Grant WD. Are fluid-based cytologies superior to the conventional Papanicolaou test? A systematic review. J Fam Pract 2001;50:1040–6.[Web of Science][Medline]
26 Vizcaino AP, Moreno V, Bosch FX, Munoz N, Barros-Dios XM, Parkin DM. International trends in the incidence of cervical cancer: I. Adenocarcinoma and adenosquamous cell carcinomas. Int J Cancer 1998;75:536–45.[CrossRef][Web of Science][Medline]
27 van den Broek LJ, van de Vijver MJ. Assessment of problems in diagnostic and research immunohistochemistry associated with epitope instability in stored paraffin sections. Appl Immunohistochem Mol Morphol 2000;8:316–21.[Web of Science][Medline]
28 Rosenthal DL. Automation and the endangered future of the Pap test. J Natl Cancer Inst 1998;90:738–49.
29 Wilbur DC, Parker EM, Foti JA. Location-guided screening of liquid-based cervical cytology specimens: a potential improvement in accuracy and productivity is demonstrated in a preclinical feasibility trial. Am J Clin Pathol 2002;118:399–407.
30 Schneider DL, Herrero R, Bratti C, Greenberg M, Hildesheim A, Balmaceda I, et al. Cervicography screening for cervical cancer among 9,062 women in a high-risk population. Am J Obstet Gynecol 1999;180:290–8.[CrossRef][Web of Science][Medline]
31 Sherman ME, Wang SS, Tarone R, Rich L, Schiffman M. Histopathologic extent of CIN3 lesions in the ASC-US LSIL Triage Study (ALTS): implications for subject safety and lead-time bias. Cancer Epidemiol Biomarkers Prev. In press 2003.
32 Nasiell K, Nasiell M, Vaclavinkova V. Behavior of moderate cervical dysplasia during long-term follow-up. Obstet Gynecol 1983;61:609–14.[Web of Science][Medline]
33 Crum CP, Egawa K, Fu YS, Lancaster WD, Barron B, Levine RU, et al. Atypical immature metaplasia (AIM). A subset of human papilloma virus infection of the cervix. Cancer 1983;15:2214–9.
34 Smith AE, Sherman ME, Scott DR, Tabbara SO, Dworkin L, Olson J, et al. Review of the Bethesda System atlas does not improve reproducibility or accuracy in the classification of atypical squamous cells of undetermined significance. Cancer Cytopathol 2000;90:201–6.
35 Richart RM. Cervical intraepithelial neoplasia. Pathol Ann 1973;8:301–28.[Medline]
36 Reagan JW, Hamonic MJ. Part IV. The cytology of early cancer. Dysplasia of the uterine cervix. Ann N Y Acad Sci 1956;63:1236–44.
37 Sherman ME, Schiffman MH, Lorincz AT, Manos MM, Scott DR, Kurman RJ, et al. Toward objective quality assurance in cervical cytopathology. Correlation of cytopathologic diagnoses with detection of high-risk human papillomavirus types. Am J Clin Pathol 1994;102:182–7.[Web of Science][Medline]
38 Kadish AS, Hagan RJ, Ritter DB, Goldberg GL, Romney SL, Kanetsky PA, et al. Biologic characteristics of specific human papillomavirus types predicted from morphology of cervical lesions. Hum Pathol 1992;23:1262–9.[CrossRef][Web of Science][Medline]
39 Gillespie JW, Best CJ, Bichsel VE, Cole KA, Greenhut SF, Hewitt SM, et al. Evaluation of non-formalin tissue fixation for molecular profiling studies. Am J Pathol 2002;160:449–57.
40 Moscicki A, Burt VG, Kanowitz S, Darragh T, Shiboski S. The significance of squamous metaplasia in the development of low grade squamous intraepithelial lesions in young women. Cancer 1999;85:1139–44.[CrossRef][Web of Science][Medline]
41 Autier P, Coibion M, Huet F, Grivegnee AR. Transformation zone location and intraepithelial neoplasia of the cervix uteri. Br J Cancer 1996;74:488–90.[Web of Science][Medline]
42 Elson DA, Riley RR, Lacey A, Thordarson G, Talamantes FJ, Arbeit JM. Sensitivity of the cervical transformation zone to estrogen-induced squamous carcinogenesis. Cancer Res 2000;60:1267–75.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. D. Goldhaber-Fiebert, N. K. Stout, J. A. Salomon, K. M. Kuntz, and S. J. Goldie Cost-Effectiveness of Cervical Cancer Screening With Human Papillomavirus DNA Testing and HPV-16,18 Vaccination J Natl Cancer Inst, March 5, 2008; 100(5): 308 - 320. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||