Ellen E. Sheets
Since the advent of widespread Papanicolaou (Pap) smear screening in the United States during the 1950s, the incidence of invasive cervical cancer and mortality from this disease has fallen approximately 70%. The Pap smear collects exfoliated cells from the surface of the cervix. Exfoliation occurs from normal, precancerous, and cancerous cervical epithelium. It is the detection of precancerous cells, which predate invasive disease by years, that leads to treatment of the cervical epithelium and prevention of invasive disease.
The management of the abnormal Pap test has undergone numerous updates, which are reflected in the consensus guidelines from the American Society of Colposcopy and Cervical Pathology (ASCCP). Full guideline disclosure can be reviewed at www.asccp.org. These guidelines were the result of a widely inclusive consensus conference convened in 2001 and reflect the new terminology for Pap test classification, known as the Bethesda System, which also underwent consensus review in 2001. A full classification can be reviewed at http://bethesda2001.cancer.gov.
Treatment should be based on a diagnosis that results from evaluation of the tissue's abnormal cervical cytology. Screening and subsequent treatment are aimed at the detection and elimination of preinvasive cervical disease, thus eliminating subsequent invasive disease. Management guidelines have been developed for preinvasive cervical disease.
EVOLUTION OF SCREENING CERVICAL CYTOLOGY AND THE BETHESDA SYSTEM
The evolution of the Pap smear is instructive in many aspects. First devised as a simple method to determine the reproductive cycle of laboratory animals by George Papanicolaou, it has evolved into a source of cellular material for sophisticated molecular and diagnostic techniques. Although apparently ever changing, the one consistent aspect of its past and future is the success Pap screening has had in the prevention of invasive cervical cancer.
Papanicolaou introduced collection of cervical cells from the posterior vaginal fornix as a method of finding early invasive cervical cancer. Ayers introduced direct sampling of the cervix with the spatula that still bears his name. Such direct sampling significantly increased the cellular yield.
The next major advance was the support that the beginnings of the American Cancer Society provided for increasing the visibility of the Pap smear. Such exposure led to increasing adoption of the technique, and this coincided with the realization that Pap smears could detect preinvasive disease, as well. Richart and colleagues identified the preinvasive component to squamous cell cancer of the cervix. This was achieved, in part, by use of the colposcope in defining the cervical transformation zone.
Terminology was developed for squamous preinvasive disease and reflected its origin in the cervical epithelium. The term cervical intraepithelial neoplasia, or CIN, was developed, and increasing involvement in the epithelial layer was reflected by CIN grades. CIN grade 1 (CIN 1) was used when fewer than one half of cells of the squamous cervical epithelium were abnormal. CIN 2 represented approximately two thirds involvement and CIN 3 a full-thickness epithelial abnormality. Until the early 1990s, some iteration of these categories was used to describe cervical cytology results. These terms still are used to describe the histology of squamous cervical lesions.
Technology also has led to changes in the way that we obtain cervical cytologic specimens. The first advance was the Ayers spatula, as noted above. Next, cotton swabs, often moistened with saline, were used to obtain cells directly from the cervical transformation zone. Although an improvement, a major advance in obtaining endocervical cells occurred when the endocervical brush was introduced in the 1980s. Although earlier thinking was that squamous metaplastic cells or endocervical cells were important in determining the adequacy of cervical screening, this concept has not been supported by the data. It has since become evident that a certain amount of squamous cellularity is more reflective of adequate sampling.
In the late 1980s, increased attention was paid to the potential false-negative result rate of the Pap smear, widely quoted to be as high as 50%. A false-negative Pap smear result does not reflect the current condition of the cervical epithelium and can arise from errors in screening, interpretation, or sampling. Several studies have shown that the inability of a Pap smear to render the true cervical diagnosis is more likely due to the smeared cellular sample not representing the state of the epithelium rather than the technical interpretation being misleading. Given these data, several efforts were made to reduce the false-negative rate and resulted in technology that allows for the cervical cellular sample to be rinsed into a preservative solution. Data have shown that up to 80% of the cells collected are thrown away after a conventional smear is made. Less than 10% are left on the collection device when the device is rinsed rather than smeared onto glass. From the concept of improving the cellular sample came the basis for liquid-based Pap tests, thus attempting to decrease the false-negative rate.
The Food and Drug Administration (FDA) has approved two liquid-based Pap tests. Both allow for the cellular material to be deposited in a liquid preservative that rinses the collection device and fixes the cells. The liquid-based Pap tests are the ThinPrep Pap Test (TPPT, Cytyc Corporation, Boxborough, MA, 1996 approval) and the AutoCyte PREP (TriPath Imaging, Burlington, NC, 1999 approval).
FDA labeling for the TPPT allows the makers of this system to claim that it is more sensitive than the conventional Pap smear for the detection of low-grade squamous intraepithelial lesion (LSIL) or greater disease. In 2001, they were allowed a claim that it was more sensitive for the detection of high-grade squamous intraepithelial lesion (HSIL) or greater disease. Further data had been required by the FDA to substantiate the LSIL or greater claim, and the data from a six-site clinical trial resulted in a greater than 53% increase in detection of HSIL or greater disease.
FDA labeling allows the makers of AutoCyte PREP to state that it is a replacement for the conventional Pap smear and leads to fewer unsatisfactory specimens. There is less published literature on this system than on the TPPT. Although there have been editorials indicating that the two systems are probably equivalent, the data sets as yet do not support equivalency, and there has been no head-to-head trial comparing the two systems.
Given the significant concern regarding the quality of Pap smears, the U.S. Government developed the Clinical Laboratories Improvement Amendment released in 1988. This guideline addressed the number of Pap smears that could be screened by cytotechnicians and the need for quality assurance review. Additionally, this amendment directed the National Institutes of Health to establish guidelines on Pap smear terminology in order for the cytology results to be more consistent. As a result of this direction, the first Bethesda conference was held in 1990, leading to the development of the Bethesda System for Pap cytology reports.
The Bethesda System used the categories of:
1. No evidence of malignant cells
2. Atypical squamous cells of undetermined significance (ASCUS) for cells felt to represent some of the spectrum of changes found in precancerous cells but not diagnostic of such
3. LSIL for cells with findings consistent with either human papillomavirus (HPV) effects or consistent with CIN 1–type changes. The thinking was that there was little ability to discern cytologically koilocytic effect from CIN 1 changes.
4. HSIL for cells with findings consistent with CIN 2, 3, or carcinoma in situ. The ability to discern among these categories was limited, but these cytologic changes were distinct from CIN 1–HPV effect.
In 2001, the third Bethesda conference was held and resulted in some more subtle, yet very important, changes. At major issue was the category of ASCUS. This category is difficult to interpret clinically, and it was felt that more recent data regarding this category supported a change. The change was to remove from the ASCUS category those Pap tests that contained cells suspicious for HSIL but not diagnostic. Therefore, the category of atypical squamous cells suspicious for HSIL (ASC-H) was developed. Further, the remaining atypical Pap test results would be designated atypical squamous cells of undetermined significance (ASC-US).
The 2001 Bethesda System update did not alter the categories of LSIL and HSIL. It did designate a specific cytology category for adenocarcinoma in situ. In the glandular preinvasive categories, atypical glandular cells (AGC) also have an undetermined significance (AGC-US) and suggest neoplasia. Within the glandular neoplasia section, the cytopathologist is asked to designate further whether the cells are from the endocervix or endometrium, if possible.
Further, those in compliance with the 2001 Bethesda System will now report benign endometrial cells, when seen in the cytology specimen, for women age 40 and older. As a women nears and goes beyond the menopausal years, the presence of benign endometrial cells can be indicative of more significant endometrial cavity pathology, such as polyps, hyperplasia (both simple and atypical), and endometrial cancer.
Educational notes are recommended at the end of the Bethesda System report and now will reflect the recommendations of the ASCCP consensus conference guidelines. A further discussion of these consensus conference guidelines is presented later in this chapter.
Interestingly, the 2001 Bethesda conference also took into account data from the Atypical/Low-grade Triage Study (ALTS), which looked at the overall best way to find histologic high-grade precancer within the cytology categories of ASCUS and LSIL. The expectation of the 2001 Bethesda conference was to triage ASC-US cytology by testing for HPV high-risk (HR)-type DNA and sending to colposcopy those whose ASC-US Pap tests were positive. The full results and recommendation of the ALTS and the ASCCP consensus conference will be discussed later in this chapter.
EPIDEMIOLOGY AND MOLECULAR BIOLOGY OF HUMAN PAPILLOMAVIRUS INFECTION
Since the early 1990s, HPV has been accepted as a necessary but not sufficient cause in the development of invasive cervical cancer, both squamous cell cancer and adenocarcinoma. It is the natural history of an HPV infection that has given rise to the not-sufficient portion. Excellent data support the concept that the majority of HPV infections, particularly the initial exposure in a woman's teens and early 20s, regress spontaneously.
Although HPV-related diseases have been noted in the medical literature since the Roman-Hellenic era, it was not until researchers using electron microscopy identified mature viral particles in condylomata during the 1950s that a viral cause for these diseases was entertained strongly. The next breakthrough came with the isolation of HPV type 6 DNA from condylomata by zur Hausen and colleagues during the early 1980s. Since that time, more than 100 HPV types have been identified. A new type is designated when there are sufficient differences in the DNA sequences but still enough homology such that it is consistent with the overall family of papovaviruses.
Interestingly, HPV types have specific preferences for the type of epithelium that they infect. Our interest is in those HPV types that exclusively infect lower genital tract epithelium. These HPV types have been categorized further according to their ability to cause cervical neoplasia. Those HPV types that rarely, if ever, are found in preinvasive or invasive cervical cancer are put in the category of low-risk viruses. Conversely, those types found at least occasionally in high-grade cervical intraepithelial neoplasia or cancer are categorized as high-risk viruses (Table 53.1). The prototypes of HR HPV are HPV 16 and 18. Combined, these viral types are present in more than 90% of high-grade precancers and over 80% of invasive disease.
|
|
|
TABLE 53.1. Human papillomavirus types |
The genetic areas of interest in HPV are the E6 and E7 (E, early) reading frames. The protein products of these areas bind tumor suppressor genes p53 and pRB, respectively. Simply, these genetic regions can cause transformation but not immortalization in cell culture when derived from the HR HPV types. Low-risk types do not give rise to such changes. Disruption of these tumor suppressor genes causes a cascade of events including alteration of the cell cycle. The viral capsule is quite uniform within the types and is formed from the L1 and L2 (L, late) reading frames. These capsule proteins are utilized in prophylactic vaccine therapy, whereas various manipulations of either the protein or HPV DNA from E6 and E7 are used in therapeutic immunologic approaches. An in-depth discussion of these events can be found in reviews of HPV molecular biology.
Exposure to HPV appears to occur primarily by intimate sexual contact but does not require intercourse. Other avenues of exposure have been discussed but not verified, such as common pools, hot tubs, and bathrooms. How HPV infects the skin has not been proven definitively but most likely requires breaks in the epithelial surface. Once exposed, the recipient's options are to kill the virus or become infected. When infection occurs, replication of the viral particle requires mature squamous keratinocytes for capsule assembly. This is the underlying reason why HPV cannot yet be cultured successfully in routine microbiology laboratories as a method of HPV detection.
Prolonged HPV infection allows the virus to produce its proteins and, in the case of high-risk viral types, that means E6 and E7 proteins. It is this process that is tied intimately to the morphologic changes that we perceive as CIN and invasive disease. For low-risk types, expression of the viral proteins leads to a proliferative epithelial response and the formation of condylomata or, in some cases, low-grade CIN.
Epidemiologic case studies, often using HPV DNA detection techniques, indicate that widespread exposure of the female and male sexually active population has occurred, yet the majority of these HPV-related infections appear to regress spontaneously. Koutsky and colleagues followed young women manifesting evidence of their first exposure to HPV and found that all evidence of HPV appears to regress spontaneously within 9 to 18 months from first exposure. The reason for eradication of this viral infection is not known, but immune response to the infection is a leading theory.
Reasons for suspecting that an immune response to HPV infection is the source of regression of disease are well founded but have not yet been proven definitively. Those whose immune systems are compromised, such as organ transplant recipients and human immunodeficiency virus–infected individuals, have virtually no ability to clear their HPV infections, even with repetitive treatment. Even when patients who are seropositive for human immunodeficiency virus or who have acquired immunodeficiency syndrome are treated aggressively with antiretroviral therapy, their HPV infections remain.
Interestingly, young women first exposed to HPV will be most likely to manifest their infections by developing CIN 1, which is detected by Pap test and can be confirmed by biopsy. However, the HPV types involved in these lesions are predominately in the high-risk category. Although this finding has not led to a change in the definition of low-risk versus high-risk types, it indicates that clinically detectable cervical cytologic abnormalities most likely represent HR HPV no matter what the degree of morphologic abnormality.
More significant precancerous changes, such as CIN 2 or 3, tend to occur in the mid to late 30s and into the 40s. These morphologic changes are indicative of risk for progression to invasive disease and, again, are caused by HR HPV types. How these women, often in apparent monogamous relationships, became exposed to HPV remains an area of strong research. Some believe that this represents reexposure, while others feel that perhaps a form of viral latency is at play. Obviously, HPV does not exhibit the classic manifestations of a latent virus, and perhaps the best way to describe this situation is a lack of clinical symptoms of the HPV infection after apparent spontaneous regression during the woman's 20s.
Underlying the information regarding HPV discussed to this point is that studies increasingly are showing that cytologically normal women who harbor HR HPV types in the cervix over a period of years are at increased risk for the development of high-grade preinvasive disease. Although not yet completely accepted clinically, certainly these women represent the at-risk pool among those who have had sequentially normal cytology findings on Pap testing. These data become increasingly significant as the woman's age advances to 30 years or older.
Many clinicians are not aware that women who practice a homosexual life style are also at risk for HPV exposure. They, like their heterosexual counterparts, require Pap test screening and develop evidence of HPV exposure with positive cervical cytology findings. An abnormal Pap test result in a homosexual women is no different in its etiology than in those with other sexual orientations.
HUMAN PAPILLOMAVIRUS TESTING METHODS
Because HPV cannot be cultured, its detection depends on the identification of its DNA. Since the advent of polymerase chain reaction (PCR) and similar DNA amplification methods, very minute amounts of viral DNA can be detected from infected cells. Generally, these methods require the extraction of the cell's DNA, and then the aggregate DNA is probed specifically for the presence of viral DNA. Using such techniques, although very sensitive, does not allow for identification of what cell in the sample carried the viral DNA. Therefore, contamination from sperm, white cells, or mucus of male origin may lead to false-positive HPV DNA test results. The most common methods for HPV DNA detection are PCR based and RNA-DNA hybrid detection, which is the method used in Hybrid Capture (HC; Digene, Gaithersburg, MD).
HC II is a relatively simple and inexpensive method of detecting and amplifying a specific DNA signal and is clinically relevant as the only FDA-approved test for HPV. Cells are disrupted with a base solution that liberates the aggregate sample's DNA. RNA probes specific for HPV types are then added to the solution and allowed to adhere to target DNA to form DNA-RNA hybrids. This sample is then exposed to antibodies to DNA-RNA hybrids. These antibodies are already adherent to the walls of a well and, thus, the DNA-RNA hybrids are fixed to the walls. Multiple antibodies that have alkaline phosphatase attached to them, thus amplifying the signal, then further detect these hybrids. The signal amplification can be up to 3,000-fold. A chemiluminescent dioxetane substrate is then added to the sample and it is cleaved by the alkaline phosphatase. The subsequent substrate produces a type of light that can be measured in a luminometer. Such light is reported as relative light units (RLUs). The RLU of an unknown is compared with that produced from a control sample containing 10 pg/mL of relevant HPV DNA. Final data is reported as a RLU-to-PC ratio, and a positive sample must have a value of 1.0 or greater.
The beauty of HC is that little additional equipment is necessary for the assay. Also, multiple HPV DNA probe types can be placed into the sample at one time, allowing for detection of multiple different HPV DNA sequences. The assay is separated into high-risk and low-risk groups. In the high-risk group, the assay contains RNA probes for types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68.
Problems noted in the literature with HC II surround its reproducibility and potential for cross-reactivity in the high-risk assay with low-risk types, which may be present in overwhelming amounts. Other issues surround whether or not the RLUs are directly proportional to the amount of HPV DNA present in the sample and, therefore, can be used as a method of viral load determination. No final consensus has been reached on that issue. Additionally, it is hard to discern from recent reports whether the relative changes noted in viral load are a byproduct of the size and grade of the cervical lesion or predictive of a larger lesion as Sun and others propose.
The general consensus is that HC II is as sensitive and specific as PCR-based HPV detection methods. PCR techniques have not yet been perfected that will multiplex for the presence of HPV DNA and discern high-risk HPV DNA from low-risk. However, in contrast to HC II, PCR does not appear to suffer from cross-contamination of its results from other HPVs that are not clinically relevant for that given patient or sample. Additionally, some researchers feel that sensitivity of PCR is slightly better than that of HC II and advocate that HPV-negative cytology be retested with PCR using the MY09-MY11 primers. Another potential advantage of PCR is the ability to determine specifically which HPV is present, whereas HC II will be able to determine only if the DNA lies within the high-grade or the low-grade category.
ASCUS/LSIL TRIAGE STUDY
Although only a small percentage, typically 10% to 14%, of ASCUS or of LSIL neoplasia actually represents a histologic high-grade lesion, such Pap test results are common enough to give rise to the majority, greater than 60%, of high-grade histology or CIN 2 or 3. Many clinicians would prefer not to perform colposcopy for the first ASCUS Pap test due to the frequency of this cytologic diagnosis.
More than 5% of all Pap tests per year in the United States reveal ASCUS. This would mean approximately 2 million colposcopies per year solely to evaluate ASCUS if all of these results were immediately triaged to colposcopy. LSIL represents approximately 2% of annual Pap test diagnoses. Given that the vast majority of histologic low-grade lesions or CIN 1 will regress spontaneously, one would prefer to follow patients with ASCUS/LSIL Pap test results, if possible.
The ALTS trial was developed to determine if CIN 3 lesions could be identified within these cytologic categories using either repetitive Pap tests, immediate colposcopy, or ancillary testing for the presence of HR HPV types. The colposcopy arm was considered the gold standard. Patients with a Pap test showing LSIL or ASCUS were randomized into either immediate colposcopy, interval Pap test follow-up, or HPV DNA testing using the HC II test to detect high-risk oncogenic viral types, with referral for colposcopy for positive test results. The main end point of the trial was a histologic diagnosis of CIN 3.
Approximately 640 patients were randomized into the LSIL arm when it was closed. Accrual was terminated due to the high percentage of patients found to have HR HPV by HC II results. Given that more than 80% of HC II results were HR positive, HPV DNA testing could not be utilized in this cytologically defined group for colposcopy triage.
In the ASCUS group, however, the results indicated that HC II HR-positive result was an effective parameter for triage for colposcopy and resulted in the greatest detection rate of CIN 3 from the three ASCUS study arms. HC II detection of HR types resulted in approximately 56% of this arm of the ASCUS study being referred for colposcopy. The repetitive Pap test arm reached equivalency over a year of follow-up when colposcopy was done for a repeat ASCUS result or greater. It is from this large pivotal trial and several smaller antecedent trials that the ASCCP based its recommendations for ASCUS and LSIL evaluation.
Since publication of the ALTS trial data, there has been some dissent regarding HPV testing of ASCUS Pap tests. Herbst and others have cited the difficulties in reproducing the ASCUS Pap test results as a potential pitfall of the trial results. Other obvious concerns for clinicians with the ASCUS triage to HPV testing are as follows: (a) discussing the HPV test results with patients and their families, (b) potential cross-over reactivity when high viral loads of low-risk HPV 6 and 11 are present, (c) potential increase in the ASCUS category because reflexive testing for this category will occur, (d) how to manage the ASCUS-positive, HC II HR-positive patient with a colposcopy that reveals no evidence of disease. The challenge will be to educate the consumer and medical personnel on how to handle these situations.
AMERICAN SOCIETY OF COLPOSCOPY AND CERVICAL PATHOLOGY CONSENSUS GUIDELINES FOR THE MANAGEMENT OF THE ABNORMAL PAP SMEAR
Following the 2001 Bethesda conference, the ASCCP held a consensus conference to discuss the changes in the Bethesda System and to utilize data from the ALTS and other clinical trial data to develop guidelines for the clinician in the management of abnormal Pap test results. These guidelines can be viewed in their entirety at www.ASCCP.org.
For the Pap test that does not reveal any evidence of cytologic abnormalities, the ASCCP has indicated that routine screening should allow for physician-patient decision making. There was considerable discussion regarding how to determine if a woman is in a high-risk category and, therefore, should be screened more frequently compared with those women felt to be at low risk for the development of CIN. No true consensus was reached. The concept of HPV DNA testing for high-risk viral types by HC II in women with normal cytology as a means of determining risk was discussed, but it was felt that the data were insufficient to allow them to reach a conclusion.
In the category of ASCs, the ALTS data were utilized for ASC-US. Interestingly, little mention is made of ALTS use of the ASCUS category without further subcategories, and the ASCCP guidelines will use the 2001 Bethesda System category of ASC-US. Essentially, ASCCP guidelines indicate that the preferred method of evaluation of the ASC-US Pap test is two-fold. If it comes from a liquid-based sample, reflex testing for HR HPV DNA is done using HC II, and if it comes from a conventional Pap smear, secondary HPV testing is performed using a second co-collected sample. If no co-collection was performed, then for the ASC-US category, Pap smears are repeated at 6-month intervals and the patient is referred for colposcopy if results are repetitively ASC-US or greater. When HC II is done, those patients with ASC-US samples positive for HR HPV should have immediate colposcopy, whereas those negative for HR HPV can revert to yearly screening.
For those patients with ASC-US with HR HPV on HC II who do not have documented CIN at the time of colposcopy, repeat cytology testing at 6-month intervals is recommended, with repeat colposcopy for any positive results. An alternative is to repeat the HPV DNA testing after 12 months and, if positive for high-risk types, repeat the colposcopy. Little is known as to the rates of CIN for the initial ASC-US and less for those who require follow-up of negative colposcopy findings. Postapproval studies will be extremely useful in guiding clinicians and patients.
Patients with ASC-H Pap test results should be referred for immediate colposcopy. Although there are no specific data for this category regarding the possibility of CIN 2 or 3 being found, it is inferred from previous reports that the possibility will be high enough to warrant colposcopic intervention. Postimplementation studies will be of interest. Additionally, the guidelines lead one to assume that HPV DNA testing would not be a useful triage tool for this category.
When negative colposcopy results are noted for ASC-H, the recommendation is to review the cytology and histology for correlation and confirmation. If again supportive of the respective diagnoses, then repeat cytology at 6-month intervals or retesting for HR HPV DNA after 12 months is acceptable, with repeat colposcopy for positive findings.
For LSIL, the ASCCP guidelines note that HPV DNA triage is not useful. This is based primarily on the ALTS data that revealed 83% of LSILs positive for HR types. Given these data, colposcopy is the preferred method of evaluation for LSIL. If colposcopically directed biopsies are consistent with CIN 1 or less, then continued screening with cervical cytology at 6-month intervals is acceptable. Another follow-up method postcolposcopy would be for patients with documented CIN 1 to undergo an HC II assay after 1 year and, if positive, move on to treatment. In any scenario, the guidelines recommend a follow-up period after histologically documented CIN 1 to allow for spontaneous regression. Those patients with cytologically demonstrated LSIL who have CIN 2 or 3 shown by biopsy should follow the guidelines for treatment.
The general guidelines for LSIL are for complete visualization of the transformation zone. When the colposcopy is unsatisfactory, an endocervical curettage is preferred, because it may reveal a diagnosis confirmed by this histology. If all histologic findings are negative, repeat cytologic studies at 6-month intervals or testing for HPV DNA after 1 year is appropriate.
When postmenopausal women have a Pap test consistent with LSIL, one should evaluate the Pap test for evidence of atrophy. If present, treatment with topical estrogen cream should be performed and the Pap test repeated after several weeks. Should the Pap test result then revert to normal, routine screening can be instituted. If still abnormal, colposcopy should be performed.
Pregnant women with LSIL should have colposcopy and biopsy performed when deemed appropriate by examination. Continual follow-up with either cytology or HPV DNA testing is warranted. Adolescents can be followed prospectively without colposcopy for the LSIL category. Either cytologic examination at 6-month intervals or follow-up HPV DNA testing after 1 year is acceptable.
There is no question that an HSIL detected with cytology most likely will reveal CIN 2 or 3 on colposcopically directed biopsies and, therefore, colposcopy is the preferred method. With HSIL cytology results, one would still recommend colposcopy, even if the HC II screen were negative for HR HPV DNA.
Pregnant women should receive colposcopy and directed biopsy of any identified lesion. Vascular changes are more marked during pregnancy, thus making simple interpretation of the lesion colposcopically without biopsy confirmation more difficult. Therefore, even though bleeding may occur, it is recommended that biopsy be performed to rule out invasive disease.
Many clinicians are concerned when HSIL cytology is followed by negative colposcopy and biopsy findings. The guidelines indicate that for a satisfactory colposcopy, follow-up at 4- to 6-month intervals is acceptable for up to 1 year. If cytology results are still positive without lesions noted at that time, then a loop electrical excisional procedure (LEEP) should be performed.
In glandular lesions, the ASCCP guidelines are explicit that the Pap test that shows AGC, whether further qualified or not, must be evaluated with colposcopy and endocervical sampling. For those Pap tests showing AGC that are further qualified as suggestive of neoplasia, a negative colposcopy with negative histologic assessment does not end the evaluation. In that setting, further tissue sampling is required, and the preferred method is by cold knife conization. The same intervention applies for repetitive Pap tests showing AGC that are not explained by colposcopic assessment.
When the glandular cytology is felt to be endometrial in origin, then endometrial sampling is indicated. Also, when the glandular cytology is not further qualified and the woman is perimenopausal or postmenopausal, again endometrial sampling is recommended.
The finding of benign endometrial cells on a Pap test in a woman over age 40 should trigger the consideration of endometrial sampling. A careful history of the woman's menstrual pattern or evidence of any recent changes should be sought in addition to questioning, when appropriate, about postmenopausal bleeding.
Cytologic findings consistent with squamous cell cancer always should trigger a further examination, either visual if the lesion is obvious or by colposcopy if not clinically apparent. There is no consensus as to how tissue sampling of possible occult squamous cell cancer should be performed, and many advocate a LEEP be done rather than cold knife conization.
For those whose cytology is consistent with adenocarcinoma in situ, colposcopy should be performed to identify any obvious areas of malignancy, but unless a diagnosis of adenocarcinoma in situ is found on biopsy, a diagnostic procedure is indicated, preferably a cold knife conization. Such a tissue excision should be evaluated for the presence of adenocarcinoma in situ and the possibility of early invasive adenocarcinoma.
TREATMENT
The criteria for treating preinvasive cervical disease depend on the histologic diagnosis, the location and size of the lesion, and the distance from the endocervical canal to the lesion. For CIN 2 or 3, treatment without delay is advised. In the case of CIN 1, persistence of the lesion for longer than 12 months is a reason for treatment. When the histology indicates a glandular lesion, cold knife conization (CKC) is indicated.
Treatment can be either ablative or excisional. Ablative techniques do not generate any further tissue specimens and include laser vaporization and cryosurgery. Such treatment requires a satisfactory colposcopic examination that identifies the entire transformation zone and the full extent of the lesion area. Excisional techniques will generate further tissue for histologic review and include conization by cold knife, LEEP, or laser excisional technique. Generally, when LEEP is used in the office, the same criteria used for ablation apply in order to minimize the potential for bleeding.
Mitchell and colleagues randomly compared cryosurgery, laser vaporization, and LEEP for patients with cervical squamous intraepithelial lesions. No essential difference was noted in success rates or complications among these techniques. As noted above, glandular lesions, either suspected by cytology or documented by histology, require surgical intervention. Because controversy still exists regarding the status of the type of conization for glandular disease, the ASCCP guidelines do recommend CKC for depth of excision and clarity of margins.
TECHNIQUES FOR TREATMENT OF SQUAMOUS INTRAEPITHELIAL LESIONS
Treatment techniques are listed in Table 53.2. It is important to remember that the final pathology after these excisional techniques occasionally will reveal apparently normal epithelium. The most likely reason for this is that the area of morphologic change was removed by a prior colposcopically directed biopsy.
|
|
|
TABLE 53.2. Techniques for treatment of squamous intraepithelial lesions |
Cryosurgery
The technique of cryosurgery initially was reported for treatment of CIN in the early 1970s. Using either carbon dioxide or nitrous oxide under pressure as the coolant, the cervical epithelium is frozen to a depth of 6 to 10 mm. The length of time for the freeze and the absolute temperature of the probe will determine the depth of penetration and subsequent tissue loss. Tissue is sloughed off slowly, over a period of 10 to 14 days, as a watery discharge. The most widely accepted technique for cryosurgery is to freeze the cervix for 3 minutes after formation of an ice ball on the cervix, followed by a 5-minute thaw and a repeated 3-minute freeze.
Although not as widely used as it was prior to LEEP, cryosurgery is less expensive and therapeutically will have results similar to LEEP when adequate caution is used with its application. In general, cryosurgical lesions need to be close to but not extend far into the cervical canal. They ideally need to be within a 2-cm radius from the canal, not occupy more than 50% of the cervical surface, and not involve the endocervical glands. Generally, cryosurgery is performed after childbearing has been completed.
Laser Vaporization
Initially discussed in the gynecology literature in the late 1970s, laser (light amplification by stimulated emission of radiation) became the mainstay of therapy throughout the 1980s and into the 1990s. Its use does require a learning curve, and the length of time for vaporization of cervical lesions should not exceed that taken by cryosurgery. Laser is light generated in the infrared spectrum by running electricity through a gas. For lower genital tract use, the gas used is carbon dioxide. The wavelength of such a laser is absorbed primarily by water. Thus, individual cells are heated instantaneously when the light contacts them, leading to their immediate vaporization by boiling the cellular water content. One can control the depth of laser penetration into the cervical epithelium quite closely by altering the spot size, the power, and the dwell time. Settings used lead to a lower power density that creates a larger area of vaporization and is achieved by a beam spot size of 2.0 to 2.5 mm. Generally, laser vaporization is done to a depth of 7 to 10 mm.
The procedure can be done safely in the office under local anesthesia, using nonsteroidal antiinflammatory drugs to reduce postprocedure discomfort. It is not uncommon to perform laser procedures in the operating room when the area to be treated is extensive.
Laser was widely replaced by LEEP. Tissue generation and ease of use have made LEEP attractive. However, treatment of extensive, preinvasive disease involving multiple lower genital tract structures, such as the cervix and vagina, often are performed in one sitting using laser.
Laser Excisional Conization
Rather than using laser for vaporization leading to ablation, one can use it to excise a conization specimen. The settings for the latter create a significantly greater power density, allowing the laser beam to act in an excisional rather than ablative manner. This is achieved by decreasing the spot size of the beam. Difficulty interpreting cone margins due to “char” artifact caused by laser excision and the ease of LEEP conization have reduced significantly the indications of laser conization.
Loop Electrical Excisional Procedure
LEEP, variably known as simply loop or LLETZ (large loop excision of the transformation zone), was made possible after modifications were developed in the wire loop and the electrosurgical current running through it. Now the current, a blend of coagulative and cutting electrosurgical currents, allows for excision with minimal bleeding and minimal coagulative artifact of the excisional margin.
Most practitioners who perform LEEP will do so in the office on patients with satisfactory colposcopy findings, with the extent of the lesion visible. Vasoconstrictive agents, such as epinephrine or pitressin, are combined with local anesthetics given in an intrastromal technique, circumferentially over the cervical face and avoiding the lesion area. Such treatment maximizes vasoconstriction and provides local anesthesia. Many practitioners will identify the transformation zone and lesion tissue by staining the cervix with modified Schiller iodine solution prior to excision. The procedure is best performed as a single pass encompassing the transformation zone and lesion. Coagulation of the surgical bed is then done with a roller ball cautery using a spark technique. Monsel solution commonly is placed in the surgical bed for hemostasis, thus avoiding sutures.
Cold Knife Cervical Conization
The role of CKC has become increasingly limited due to the widespread use of LEEP. Most clinicians feel that CKC should be performed when the margins of resection are important, because LEEP can leave a cautery effect difficult to interpret at the margin. The uses of CKC have been focused on evaluation of microscopic invasive disease, both squamous cell cancer and adenocarcinoma. Certainly there is support for CKC in adenocarcinoma in situ for evaluation of margins and depth of the conization sample. The incidence of adenocarcinoma in situ is increasing in women of childbearing age, and sparing of fertility is an important concern. Given this rise in incidence, much has been made regarding conservative management. Shin and others point out that deep conization of greater than 1½ to 2 cm with negative margins can assess the extent of adenocarcinoma in situ involvement and margin status. With negative margins, the patient can be highly reassured that no residual disease exists in the residual cervix.
CKC generally is performed under general anesthetic with the patient in the dorsal lithotomy position. The cervical transformation zone is identified by use of iodine solution, and the areas that are iodine negative are not part of the transformation zone.
Hysterectomy
Although not commonly used at this time, there still is a role for hysterectomy in the treatment of the abnormal Pap test. Such a need does arise in the scenario of repetitive high-grade CIN that recurs despite less invasive treatments. It is not uncommon in this situation to end up with cervical scarring that makes Pap test and colposcopy virtually impossible to perform. The patient, particularly postchildbearing, may opt for more definitive management with hysterectomy. However, she should be counseled that her abnormal cytology result rate posthysterectomy will not be zero. Preinvasive disease can occur in the upper vaginal vault at the hysterectomy cuff area.
Immunotherapy
Because HPV is necessary for the development of CIN and invasive cancer, one would assume that a prolonged infection is required for transformation. HPV-related cell surface proteins have been identified and certainly should induce an immune response. This immune response potentially could be supplemented by inducing a further immune response by the administration of similar HPV-related proteins. The concept is to enhance a T-lymphocyte response that will lead to eradication of the cervical lesion.
Several methodologies have been evaluated in phase 1 and 2 clinical trials looking to treat CIN 2 and 3 or invasive squamous cell carcinoma. The results have shown some responses but are not yet definitive.
Vaccines have been developed that employ the use of the HPV viral capsule. The capsule is produced by the L1 and L2 regions of the HPV genome and have very little variation from one type of HPV to another. Therefore, capsules generated in vivo from the proteins of these regions are being given as viruslike particles, or VLP, as vaccines against initial infection. The concept is to develop an antibody response to the VLP and thus create an antibody barrier to HPV infection transvaginally. It will be several more years before the preliminary clinical results are known, but initial antibody data from the cervix appear to support the induction of an immune response.
POSTTREATMENT SURVEILLANCE
Although the excisional and ablative methods are extremely effective for the treatment of CIN, there are instances of persistence and failure posttreatment. In general, these are found at the time of follow-up Pap testing. However, there is a growing interest in using HPV DNA after LEEP to predict which woman with positive margins after LEEP are at greatest risk of recurrence. At this point, there is no specific recommendation for HPV DNA testing post-LEEP, but it is certain to be evaluated further.
Most clinicians follow patients posttreatment with Pap tests every 6 months for a year and, if normal, yearly. Some have advocated the additional use of an endocervical curettage in the follow-up of women with glandular lesions, particularly adenocarcinoma in situ. If the Pap test result is abnormal, colposcopy should be performed to define the extent of disease and allow for directed biopsies.
SUMMARY POINTS
· The Pap test has a proven track record for success. Since its widespread inception in the 1950s, the incidence of invasive cervical cancer and related death has dropped significantly. Since then, the technique surrounding how cervical cytology is obtained and processed has changed, with increases in transformation zone sampling and decreases in apparent false-negative test result rates.
· In the 1960s, the categories of cervical intraepithelial neoplasia were used to describe preinvasive cervical cytology. CIN ranged from minor or grade 1 changes to the most severe cytologic atypia, designated CIN 3. In the 1990s, the Bethesda System created the categories of ASC-US, HSIL, AGC, and adenocarcinoma in situ to complement the invasive squamous cell cancer and adenocarcinoma categories.
· HPV is a necessary but not sufficient cause for the development of cervical precancerous and cancerous lesions. Testing for the presence of HPV DNA can improve the triage pattern of women whose Pap tests fall within the category of ASC-US. Data from the ALTS have shown that screening such Pap tests for the presence of HR HPV will better identify those women within the ASC-US cytology screening category who harbor high-grade CIN.
· Treatment advances stem from understanding that preinvasive and invasive cervical disease arises at the transformation zone. Treating this zone of change will completely eliminate preinvasive disease approximately 90% of the time or more. Therapies range from simple excisions of the abnormal cervical epithelium to laser or cryosurgical ablations, to surgical excisional biopsies of the cervix, to hysterectomy.
RECOMMENDED READINGS
Screening Cervical Cytology and the Bethesda System
Chatelain P, Mottot C. Liquid-based cytology for primary cervical cancer screening: a multi-centre study. Br J Cancer 2001;84:360–366.
Gay JD, Donaldson LD, Goellner JR. False-negative results in cervical cytologic studies. Acta Cytol 1985;29:1043–1046.
Germain M, Heaton R, Erickson D, et al. A comparison of the three most common Papanicolaou smear collection techniques. Obstet Gynecol 1994;84:168–173.
Jones HW. The Bethesda System. Cancer 1995;76:1914–1918.
Monsonego J, Autillo-Touati A, Bergeron C, et al. The 2001 Bethesda system: terminology for reporting results of cervical cytology. JAMA 2002;287:2114–2119.
Solomon D, Schiffman M, Tarrone R. Comparison of three management strategies for patient with atypical squamous cells of undetermined significance. J Natl Cancer Inst 2001;93:387–392.
Epidemiology and Molecular Biology of Human Papillomavirus Infection
Howley PM, Lowy DR. Papillomaviruses and their replication. In: Knipe DM, Howley PM, Griffin DE, et al., eds. Fields virology, fourth ed. Philadelphia: Lippincott Williams & Wilkins, 2001.
Lowy DR, Howley PM. Papillomaviruses. In: Knipe DM, Howley PM, Griffin DE, et al., eds. Fields virology, fourth ed. Philadelphia: Lippincott Williams & Wilkins, 2001.
Xi LF, Carter JJ, Galloway DA, et al. Acquisition and natural history of human papillomavirus type 16 variant infection among a cohort of female university students. Cancer Epidemiol Biomarkers Prev 2002;11:343–351.
Ylitalo N, Josefsson A, Melbye M, et al. A prospective study showing long-term infection with human papillomavirus 16 before the development of cervical carcinoma in situ. Cancer Res 2000;60:6027–6032.
Human Papillomavirus Testing Methods
Brennan MM, Lambkin HA, Sheehan CD, et al. Detection of high-risk subtypes of human papillomavirus in cervical swabs: routine use of the Digene Hybrid Capture Assay and polymerase chain reaction analysis. Br J Biomed Sci 2001;58:24–29.
Schiffman MH, Kiviat NB, Burk RD, et al. Accuracy and interlaboratory reliability of human papillomavirus DNA testing by hybrid capture. J Clin Microbiol1995;33:545–550.
Sun CA, Liu JF, Wu DM, et al. Viral load of high-risk human papillomavirus in cervical squamous intraepithelial lesions. Int J Gynaecol Obstet 2002;76:41–47.
Vernon SD, Unger ER, Williams D. Comparison of human papillomavirus detection and typing by cycle sequencing, line blotting, and hybrid capture. J Clin Microbiol 2000;38:651–655.
ASCUS/LSIL Triage Study (ALTS)
ALTS Group. Human papillomavirus testing for triage of women with cytologic evidence of low-grade squamous intraepithelial lesions: baseline data from a randomized trial. JNCI 2000;92:397–402.
Herbst AL, Pickett KE, Follen M, et al. The management of ASCUS cervical cytologic abnormalities and HPV testing: a cautionary note. Obstet Gynecol2001;98:849–851.
Kinney WK, Manos MM, Hurley LB, et al. Where is the high-grade cervical neoplasia? The importance of minimally abnormal Papanicolaou diagnoses. Obstet Gynecol 1998;6:973–976.
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. JNCI 2001;93:293–299.
Stoler MH, Schiffman M. Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL triage study. JAMA 2001;285:1500–1555.
Guidelines for the Management of the Abnormal Pap Smear
Crum CP, Generst DR, Krane JF, et al. Subclassifying atypical squamous cells in thin-prep cervical cytology correlates with the detection of high-risk human papillomavirus DNA. Am J Clin Pathol 1999;112:384–390.
Wright TC Jr, Cox JT, Massad LS, et al. 2001 consensus guidelines for the management of women with cervical cytological abnormalities. JAMA2002;287:2120–2129.
Treatment
Ferenczy A. Management of patients with high grade squamous intraepithelial lesions. Cancer 1995;76:1928–1933.
Mitchell MF, Tortolero-Luna G, Cook E, et al. A randomized clinical trial of cryotherapy, laser vaporization, and loop electrosurgical excision for treatment of squamous intraepithelial lesions of the cervix. Obstet Gynecol 1998;92:737–744.
Schiller JT, Lowy DR. Papillomavirus-like particle based vaccines: cervical cancer and beyond. Expert Opin Biol Ther 2001;1:571–581.
Shin CH, Schorge JO, Lee KR, et al. Conservative management of adenocarcinoma in situ of the cervix. Gynecol Oncol 2000;79:6–10.
