Cervical Cancer Precursors and Their Management
L. Stewart Massad
DEFINITIONS
The Bethesda System (TBS)—A classification system originally introduced in 1988 for grading cytologic abnormalities.
Cervical cytology—A diagnostic technique that involves microscopic examination of individual cells or cell clusters scraped, brushed, or washed from the surface of the cervix and stained with Papanicolaou stain. The nuclear characteristics and the nuclear-cytoplasmic ratio are used to diagnose the degree of cervical abnormality present. It is a common screening test for cervical cancer.
Cervical dysplasia—Dysplasia is a pathologic term to indicate noninvasive epithelial atypia that involves various degrees of the cervical epithelium. By convention, in mild dysplasia, the nuclear atypia, mitoses, and cellular irregularity involve the lower one third of the squamous epithelium, moderate dysplasia the middle third, and severe dysplasia the upper third. Full-thickness dysplasia is termed carcinoma in situ.
Cervical intraepithelial neoplasia (CIN)—A cytologic and histologic classification of preinvasive cervical atypias or neoplastic changes. In general, CIN 1 = mild dysplasia, CIN 2 = moderate dysplasia, and CIN 3 = severe dysplasia/carcinoma in situ.
Colposcopy—A diagnostic technique involving visual inspection of the cervix and other areas of the lower genital tract with a low-power (3 to 15×) binocular scope. The surface contour, vascular pattern, and staining pattern of changes seen with dilute acetic acid and Lugol iodine reflect the severity and extent of cervical abnormality and help the clinician to select a location for biopsy confirmation.
Conization—Removal of the cervical transformation zone and part of the endocervix in a cone-shaped specimen. Traditionally performed with a scalpel, it also can be done with laser or wire loops.
Human papillomavirus—One of a family of protein-encapsulated DNA viruses, some of which are the necessary though not sufficient cause of cervical cancer. DNA sequence variations allow distinction of more than 150 genotypes.
Lower Anogenital Squamous Terminology (LAST)—A classification system introduced in 2011 for grading histologic abnormalities while incorporating the impact of papillomavirus infection on disease. Its two tiers mirror TBS terminology, including low- and high-grade squamous intraepithelial lesions. It includes provisions for molecular testing for equivocal lesions.
Loop electrosurgical excision procedure (LEEP)—Also called a large loop excision of the transformation zone (LLETZ). Cone biopsy performed with a thin wire electrosurgical loop electrode 1 to 2.5 cm, usually under local anesthesia.
Squamocolumnar junction (SCJ)—The junction between the squamous and columnar epithelium on the cervix. As columnar epithelium undergoes metaplasia to squamous epithelium on the cervical portion, the SCJ has a dynamic location.
Transformation zone—A colposcopic term for the area of metaplasia that is observed between the original prepubertal (“original SCJ”—lateral edge) and present squamocolumnar junction (medical edge).
Cervical cancer prevention has been very successful because of two important factors. First, preinvasive cervical changes that develop and progress slowly over years have been identified; second, relatively simple and inexpensive methods to detect and destroy these preinvasive cervical changes have been developed over the past 50 years. In recent years, the cause of these cervical changes has been identified as due to the human papillomavirus (HPV), and this has led to major changes in our understanding and management of both screening and treatment of cervical cancer precursors. Because of the sensitivity of screening tests, which rely on HPV detection and the fact that most women, even with high-risk HPV infection, will not progress to invasive cervical cancer, balancing the risk and costs of triage and treatment of these early cervical changes requires a thorough understanding of the natural history of these lesions and the risk and benefits of the triage and treatment options available.
EPIDEMIOLOGY AND ONCOGENESIS
In 1842, Rigoni-Stern noted that cervical cancer was found in married women, but was virtually absent in celibate women, such as Catholic nuns. Subsequent epidemiologic studies reported that number of sexual partners, early age at first sexual
intercourse, low socioeconomic status, cigarette smoking, and early age at first pregnancy increased a woman’s risk of cervical neoplasia. More recently, it has become clear that immunosuppression from any cause, including infection with human immunodeficiency virus and therapeutic immunosuppression after organ transplantation, is associated with an increased rate of cervical neoplasia. Oral contraceptive use has been reported to increase a woman’s risk of cervical cancer, especially adenocarcinoma. Cellular changes on Pap smear described as “koilocytosis” have been recognized since the 1940s as correlates of cervical disease, but the human papillomavirus (HPV) was not identified as the etiologic agent in cervical oncogenesis until 1980 by zur Hausen. This discovery and the demonstration that HPV infection is the necessary though far from sufficient cause of cervical cancer won him the Nobel Prize in 2008.
intercourse, low socioeconomic status, cigarette smoking, and early age at first pregnancy increased a woman’s risk of cervical neoplasia. More recently, it has become clear that immunosuppression from any cause, including infection with human immunodeficiency virus and therapeutic immunosuppression after organ transplantation, is associated with an increased rate of cervical neoplasia. Oral contraceptive use has been reported to increase a woman’s risk of cervical cancer, especially adenocarcinoma. Cellular changes on Pap smear described as “koilocytosis” have been recognized since the 1940s as correlates of cervical disease, but the human papillomavirus (HPV) was not identified as the etiologic agent in cervical oncogenesis until 1980 by zur Hausen. This discovery and the demonstration that HPV infection is the necessary though far from sufficient cause of cervical cancer won him the Nobel Prize in 2008.
The current understanding of the role of HPV in cervical oncogenesis is illustrated in Figure 50.1. The more than 150 types of HPV are members of a large family of encapsulated DNA viruses known as the Papovaviridae. Unlike many viruses that infect humans and that are serotyped by their surface antigens, the papillomaviruses are classified according to their DNA base-pair sequence. The degree of relatedness is determined by the degree of their DNA hybridization homology. HPVs are epitheliotropic viruses that infect virtually all surface epithelia, including the skin and mucous membranes. In addition to cervical cancer, they have been shown to cause significant proportions of vulvar, vaginal, anal, and oropharyngeal cancers and may contribute to malignancies at other sites.
The most prevalent anogenital HPVs can be divided into three groups that are predictive of their ability to produce neoplasia (Table 50.1). Low-risk HPV types are commonly associated with condylomas and cervical intraepithelial neoplasia (CIN) 1 and some CIN 2 s, but they almost never are associated with true precancer or cancer. HPV types 6 and 11 are responsible for approximately 90% of all condylomas of the lower genital tract. On the other hand, the anogenital HPV types with a high oncogenic risk are mainly types 16, 18, 31, 33, 35, 39, 45, 51, 52, 58, 59, and 68. HPVs in this risk group are overrepresented in high-grade lesions and cancers. HPV types 16 and 18 are highly oncogenic, accounting for two thirds of cancers and about half of precancers. The carcinogenicity of a few HPV types, such as 56, remains uncertain at this time.
TABLE 50.1 Anogenital HPV Types | ||||||
|---|---|---|---|---|---|---|
|
HPVs infect the cervix by entering basal cells following microtrauma to thin, immature portions of the transformation zone adjacent to the SCJ. Some 80% of sexually active adults appear to have been infected with HPV at some point. When a mitotically active epithelial cell is infected, the virus may remain in the cell in a form termed a latent infection. Latent infections can be detected only by molecular techniques and may be undetectable. In most women with an initial HPV infection, the infection clears spontaneously as a result of innate and cell-mediated immunologic host defense mechanisms. About half the women who test positive for HPV will test negative at 6 months later, about 70% will be negative at 12 months, and by 2 years only 8% to 20% continue to test positive. The highly oncogenic types 16 and 18 are more slowly cleared and more commonly persist compared with less oncogenic HPV types. Although women with transient infections may develop cervical epithelial abnormalities that result in abnormal Pap tests and even colposcopic abnormalities, these are usually low- or mid-grade lesions that often clear spontaneously. Approximately 50% of women with CIN 1 will resolve their lesions over 2 years without therapy. Among adolescents, with a more active immune defense system clearance rates approach 90%.
In some individuals, however, for reasons that are not well understood, the latent papillomavirus begins to replicate independent of the host cell cycle, and large numbers of complete virions are produced. This is referred to as a productive viral infection. These latent or productive viral infections may be clinically unrecognizable or may be associated with genital warts or CIN 1 or sometimes CIN 2, but they do not result in neoplastic transformation of the infected epithelial cells.
 FIGURE 50.1 Natural history of cervical carcinogenesis. (From Schiffman M, Kjaer SK. Chapter 2: Natural history of anogenital HPV infection and neoplasia. J Natl Cancer Inst Monogr 2003;31:14-19. Reprinted with permission from the Journal of the National Cancer Institute Monographs. Copyright © 2003, Oxford University Press.) |
In contrast to latent infections, some HPV infections are accompanied by alterations in the virus-host interaction critical to the transformation of normal epithelial cells to neoplastic cells. The key alteration seems to be the integration of the viral DNA into host chromosomes. Often, this is associated with loss of regulatory regions of the HPV genome and disruption of a host tumor suppressor gene. The E7 protein binds to the pRB protein, which in turn triggers E2F transcription factor and induces p16ink4 expression, which leads to DNA synthesis and cell cycling. Other gene products employ host cell synthetic mechanisms to produce new viral DNA and viral proteins. These changes in cellular mechanisms would ordinarily lead to activation of epithelial cell p53 protein, an important protector of the normal cell cycle. However, the HPV E6 protein targets p53 for proteolytic degradation, allowing oncogenic HPV types to bypass this key cellular control mechanism. Identification of E6 and E7 mRNA has been proposed as a more specific marker of epithelial cell transformation leading to cancer than simple HPV detection.
The likelihood of this oncogenic transformation after HPV infection varies by HPV type. In an analysis of HPV screening data of more than 20,000 women in the Portland, Oregon, area, Khan and colleagues found that women who had a negative Pap test and tested negative for any HPV had a less than 1% risk of developing CIN 3 within 10 years of follow-up. This contrasted sharply with women who had a negative Pap but who tested positive for HPV 16 on study entry; those women had a 17.3% risk of developing CIN 3 or a more severe lesion during the next 10 years. At least half of these women developed high-grade CIN within 24 months. For those who tested positive for other HPV types, the risk of developing CIN 3 was only 3%.
Although Rubin recognized “incipient carcinoma” in 1909, it was not until the 1950s that the preinvasive potential of carcinoma in situ of the cervix became generally accepted. In their 1952 publication, Galvin, Jones, and Te Linde described their observations of the natural history of carcinoma in situ. Their report confirmed and advanced previous studies by Cullen and by Pemberton and Smith. The concept of a clearly identifiable preinvasive neoplastic change in the cervical epithelium represented a major breakthrough in an understanding of the development and natural history of cervical cancer, but carcinoma in situ is clinically occult. Techniques such as the Schiller test with application of Lugol iodine and random four-quadrant biopsy proved imprecise in identifying women with carcinoma in situ, and many women with this diagnosis were treated with radical hysterectomy because of the uncertainty of the natural history of this diagnosis and of the risk of concomitant cancer. It was the parallel development of cervical cytology by George Papanicolaou and Herbert Trout in the United States after 1941 and of colposcopy by Hans Hinselmann in Germany after 1927 that provided techniques that rendered clinically useful the concept of a precancerous cervical lesion whose treatment would prevent the development of invasive cancer. Cytology allowed identification of women at risk, and colposcopy allowed clinicians to identify preinvasive lesions without conization, determine their extent, select sites for biopsy confirmation, and institute fertility-sparing therapy. These developments have led to a 70% reduction in cervical cancer deaths in the United States over the past 50 years. This chapter discusses how these preinvasive lesions evolve, how they can be diagnosed and distinguished from HPV-related lesions destined to clear, and how they can be treated.
TERMINOLOGY
Although carcinoma in situ became well recognized as a full-thickness epithelial change without stromal invasion, the terminology and clinical significance of adjacent, less-than-full-thickness atypia was uncertain. In 1956, Reagan and Hamonic introduced the term dysplasia to designate these cervical epithelial abnormalities that were characterized by cytologic atypia, increased mitotic activity, and loss of polarity, graded as mild, moderate, and severe dysplasia. The poor reproducibility of the distinction between CIS and severe dysplasia and the clear premalignant nature of the latter led Ralph Richart in 1969 to propound the concept of CIN. CIN 1, CIN 2, and CIN 3 were terms used to describe a continuum of disease, representing basaloid changes extending into the lower, middle, and upper thirds of the cervical epithelium with CIN 3 incorporating both CIS and severe dysplasia. Modification of the Bethesda cytologic and histologic grading was proposed by Darragh and colleagues in 2012. CIN 1 and other manifestations of HPV infection are labeled as lowgrade squamous intraepithelial lesions (LSIL) and CIN 3 s a high-grade squamous intraepithelial lesions (HSIL). Since the histologic changes of CIN 2 are heterogenous, including both premalignant and spontaneously regressing lesions, this system encourages use of molecular tests, such as the HPV-related transformation marker p16ink4, to assign mid-grade abnormalities to either the LSIL or HSIL categories. Although the Bethesda System terminology for reporting the results of cervical cytology has been accepted throughout the United States and is used in many other countries, other terminology and definitions are used in Great Britain, Germany, Australia, and other countries, which with different thresholds for colposcopy and treatment creates some difficulty in comparing study results across countries.
The original four- or five-step Papanicolaou cytologic classification, in which “class 1” was normal or benign and “class 5” was suspicious of invasive cancer; the World Health Organization classification of mild, moderate, or severe dysplasia and carcinoma in situ; and the Richart CIN terminal have been replaced in the United States by the Bethesda system (Table 50.2). Development of this consensus system was driven by several factors, including intraobserver variation and a lack of reproducibility. Multiple classification systems without common diagnostic criteria left clinicians unsure about appropriate management.
The Bethesda system was introduced in 1988 and subsequently updated and revised in 1991 and 2001. While cytology allows the identification of various infections and other nonneoplastic findings, the core of any report is the identification of epithelial cell abnormalities, either squamous or glandular. The Bethesda system combined HPV cytopathologic effects, often referred to as koilocytosis, with mild dysplasia or CIN 1 into a category called low-grade squamous intraepithelial lesion (LSIL). More significant lesions— including moderate and severe dysplasia and carcinoma in situ, or CIN 2 and 3—were combined into high-grade squamous intraepithelial lesion (HSIL). Pap tests with cellular abnormalities insufficient for definitive diagnosis of LSIL were categorized as atypical squamous cells (ASC). In 2001, these results were further subcategorized as of undetermined significance (ASC-US) or cannot exclude high grade (ASC-H). Glandular abnormalities less than cancer have been classified as atypical glandular cells (AGC), either not otherwise specified (AGC-NOS) or favoring neoplasia, and as adenocarcinoma in situ (AIS).
In the United States, a typical laboratory that processes cervical cytology from an average, generally low-risk population reports epithelial cell abnormalities in about 5% to 6% of patients. Usually, one half to two thirds of these abnormalities are ASC (2% to 4%), whereas 1% to 2% are LSIL and 0.5% to 1.0% are HSIL diagnoses, with about 0.5% glandular cell abnormalities. The ratio of ASC to SIL diagnoses should
be between 2:1 and 3:1 for most cytology laboratories, and roughly half of ASC-US and more than 80% of LSIL should test positive for high-risk HPV. Most cytology specimens are now processed using automated suspensions of exfoliated cells in liquid-based media (liquid-based cytology, LBC) rather than conventional Pap smears. LBC results in fewer unsatisfactory tests and has a greater yield but does not appear to be more sensitive in detecting true premalignant lesions compared to conventional Pap smears. An important advantage of LBC is the possibility of testing for HPV and other biomarkers from the same liquid-based specimen; women with ASC-US cytology who test negative for high-risk HPV types are at low risk for CIN.
be between 2:1 and 3:1 for most cytology laboratories, and roughly half of ASC-US and more than 80% of LSIL should test positive for high-risk HPV. Most cytology specimens are now processed using automated suspensions of exfoliated cells in liquid-based media (liquid-based cytology, LBC) rather than conventional Pap smears. LBC results in fewer unsatisfactory tests and has a greater yield but does not appear to be more sensitive in detecting true premalignant lesions compared to conventional Pap smears. An important advantage of LBC is the possibility of testing for HPV and other biomarkers from the same liquid-based specimen; women with ASC-US cytology who test negative for high-risk HPV types are at low risk for CIN.
TABLE 50.2 The Bethesda System of Cytologic Classification | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The effectiveness of cytology screening has been demonstrated in organized national screening programs. In the United States, demonstration projects in the 1950s showed cytologic screening identified asymptomatic cancers. In Scandinavian countries, the introduction of national screening programs led to a dramatic decline in cervical cancer. However, screening may not be effective in women before ages 25 to 30. Most women who develop cervical cancer in countries with widespread screening availability have not had regular screening.
Despite its effectiveness in cervical cancer prevention programs, limitations to Papanicolaou test screening exist. A single Pap test has a sensitivity of only 50% to 60%. This means that a single test will not detect cervical lesions in many women. However, the slow progression of CIN before the development of an invasive cancer provides the opportunity for multiple screening cytologies over a period of years. So even with limited sensitivity, if three consecutive tests are negative, there is less than a 1% chance that the patient will have a highgrade cervical abnormality. Other screening problems include lesions that do not shed cells, do not transfer from the collecting device, or are not sampled by the clinician. Rarely, the slide preparation or staining is unsatisfactory. Another reason for screening failure is that cytotechnicians and cytopathologists may fail to identify abnormal cells or may misinterpret dysplastic cells as reactive or metaplastic. Glandular lesions are more commonly missed than squamous lesions. Women who are diagnosed with invasive cervical cancer after a reportedly “negative” Papanicolaou test often have abnormal cells on review of their slides that are few in number or obscured by blood or inflammatory changes. The threat of a lawsuit when cancer arises after cytology interpreted as negative may cause cytologists and cytopathologists to “overcall” an equivocal diagnosis and the gynecologist to recommend colposcopy and possibly cervical biopsy or even treatment in a patient with any hint of abnormality. This increases not only the anxiety and morbidity associated with cervical cancer screening but also the costs.
First popularized by Hinselmann in Germany, the colposcope is a magnifying instrument used to examine the cervix. With the colposcope, vascular changes and other epithelial patterns were recognized and correlated with cytologic and histologic changes. These early neoplastic changes were most prominent adjacent to the squamocolumnar junction (SCJ); this area was called the transformation zone, the area of the cervix that had undergone metaplasia from the prepubertal columnar epithelium to mature squamous epithelium. With time, colposcopic patterns associated with the developing stages of dysplasia and early invasive cancer were described. Adoption of colposcopy in North America was limited by cumbersome German terminology and the association of colposcopy with Nazi experimenters, including Hinselmann’s associate Eduard Wirths. However, by the 1970s, colposcopy with directed biopsy had generally replaced cone biopsy as the triage test for women with cytologic abnormalities.
Following the announcement of the 2001 Bethesda System terminology for reporting the results of cervical cytology, the American Society of Colposcopy and Cervical Pathology has sponsored consensus conferences to develop and update management guidelines, expanding recommendations to cover
results of cotesting. The guidelines are based on 5-year risk of CIN 3+ and prescribe similar management for problems with similar risks. They also incorporate cotesting into follow-up to improve sensitivity of surveillance for persistent or recurrent lesions, with longer intervals between follow-up tests to minimize the risk of identifying transient lesions. When cytology, colposcopy, and histology findings are discordant, expert review may be in order. Multiple biopsies minimize the likelihood of missing CIN 3+.
results of cotesting. The guidelines are based on 5-year risk of CIN 3+ and prescribe similar management for problems with similar risks. They also incorporate cotesting into follow-up to improve sensitivity of surveillance for persistent or recurrent lesions, with longer intervals between follow-up tests to minimize the risk of identifying transient lesions. When cytology, colposcopy, and histology findings are discordant, expert review may be in order. Multiple biopsies minimize the likelihood of missing CIN 3+.
Once the importance of HPV in cervical oncogenesis had been demonstrated, other investigators suggested that a vaccine might prevent infection and perhaps even eliminate the HPVs in already infected individuals. Koutsky and colleagues tested immunization against HPV type 16 in a clinical trial, prospectively randomizing almost 2,400 young women who tested negative for HPV type 16 infection to three doses of a vaccine made from synthetic capsule protein or placebo. This and other trials showed that vaccinated HPV-naïve women develop essentially 100% immunity to targeted HPV types, with substantial reduction in subsequent CIN. Currently approved prophylactic HPV vaccines are targeted against L1 capsid proteins and thus are HPV-type specific. The quadrivalent vaccine targets the high-risk HPV types 16 and 18, as well as low-risk types 6 and 11, while the bivalent vaccine targets only HPV 16 and 18. Minor side effects are common and include injection site pain and fever. Serious side effects are rare, but anaphylaxis and vagal reactions have been reported. Vaccines appear only prophylactic, with little or no effect on established infections. Multivalent and therapeutic vaccines are in development.
Screening for Cervical Neoplasia
The goal of cervical cancer prevention is the identification and destruction of precancer before the development of invasive disease that threatens life, health, and reproductive capacity. This must be balanced against the harms of screening, including stigmatization from the identification of HPV as a sexually transmitted infection, fear of cancer, and the potential for bleeding, local injury, and preterm delivery after treatment when lesions identified through screening are not destined to lead to cancer. For this reason, screening should not begin until age 21, since the risk of cancer in teens is negligible, and although precancers do occur earlier, they are unlikely to progress to invasion prior to detection in subsequent screening rounds. Similarly, cancer late in life is unlikely among women with long negative screening histories, so screening should cease in such women at age 65. The sensitivity of screening appears sufficient to allow 3-year intervals between Pap tests and 5-year intervals between Pap and HPV cotests. These screening recommendations were revised in 2013 (www.ASCCP.org).
A variety of devices have been devised for cervical sampling, including spatulas, brooms, and brushes. The spatula is rotated with gentle pressure twice around the cervix with the longer lobe in the os to enhance sampling. A broom is also rotated against the ectocervix. An endocervical brush is placed in the cervical canal with some bristles visible and rotated 90 to 180 degrees; deeper penetration and more vigorous rotation increase bleeding without improving cell collection. Cells should be transferred to slide or vial according to manufacturer’s instructions; inadequate cell transfer may result in uninterpretable results.
Visual Inspection
Although colposcopy has been used instead of or in addition to cytology for cervical cancer screening in some areas of Europe and South America, it is time-consuming, requires well-trained clinicians, and is expensive. Simple visual inspection of the cervix after application of acetic acid (VIA) or Lugol iodine (VILI) has been studied in resource-poor settings for cervical cancer screening, alone or in combination with HPV testing. Areas of cervical dysplasia turn white after application of acetic acid and turn yellow after application of iodine. Inspection of the cervix using handheld low-power magnifying lenses can identify many lesions. Although these techniques have proved useful and cost-effective in resource poor settings, they are not appropriate for developed countries because of limited sensitivity and specificity. In most settings, VIA and VILI remain investigational.
Human Papillomavirus Testing
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
