Precancerous Lesions of the Cervix

  • Thomas C. WrightEmail author
  • Brigitte M. Ronnett
  • Robert J. Kurman
Living reference work entry


This chapter covers the epidemiology, pathogenesis, current nomenclature, and histological appearances of precancerous lesions of the cervix. For over 20 years, it has been evident that specific types of human papillomavirus (HPV) cause almost all squamous cell carcinomas of the cervix as well as the vast majority of adenocarcinomas of the cervix. However, much remains to be learned regarding the precise molecular pathways by which these DNA tumor viruses produce tumors and why only a small minority of HPV- infected individuals develop disease. The chapter discusses the current terminology used to refer to precancerous lesions including both squamous and glandular lesions. The chapter then provides an in- depth description of the histological features of both squamous intraepithelial lesions and glandular neoplasia. This includes both their differential diagnoses and the use of adjunctive histological methods including p16.

Precursors of Squamous Cell Carcinoma

Terminology and Historical Perspective

The histopathological classification of a disease should reflect both current concepts of its pathogenesis and its clinical behavior. Over the last 50 years, our understanding of the pathobiology and behavior of cervical cancer precursors has evolved considerably. As a result, the terminology used to classify preinvasive lesions of the cervix has frequently changed. Although these changes in nomenclature and the resulting lack of a uniform terminology have been an ongoing source of confusion to both gynecologists and pathologists, each change has actually reduced the number of specific pathological categories and has made clinical decision-making more straightforward.

The existence of precursor lesions for invasive cervical cancer has been recognized for over a century. As early as 1886, Sir John Williams commented on the presence of noninvasive epithelial abnormalities adjacent to invasive squamous cell carcinomas of the cervix (Williams 1888). The spatial relationships and histologic appearance of these noninvasive epithelial lesions were better described by Cullen in 1900 who recognized that these intraepithelial lesions histologically resembled the adjacent invasive cancers (Cullen 1900). In the 1930s, Broders reintroduced the term carcinoma in situ that was first used by Schottlander and Kermauner to refer to these intraepithelial cervical lesions (Broders 1932). A temporal relationship between carcinoma in situ and invasive cancer was subsequently reported by Smith and Pemberton, as well as by Galvin, Jones, and Telinde, who diagnosed carcinoma in situ in several patients months to years before the development of invasive cervical cancer (Pemberton and Smith 1929). The recognition that there was both a spatial and temporal relationship between carcinoma in situ and invasive squamous cell carcinoma led to the hypothesis that invasive squamous cell carcinoma develops from a histologically well-defined precursor lesion (Broders 1932). This hypothesis was subsequently substantiated by long-term follow-up studies, which clearly demonstrated that a significant proportion of untreated patients with carcinoma in situ subsequently develop invasive squamous cell carcinoma (Kolstad and Klem 1976; Koss et al. 1963).

Once it was accepted that carcinoma in situ was a precursor to invasive squamous cell carcinoma, population-based cytological screening programs were begun to detect and treat precursor lesions prior to the actual development of cancer. As large numbers of women began to be screened for cervical disease, it became apparent that many women had cervical epithelial abnormalities that were cytologically/histologically less severe than carcinoma in situ. These lesions formed a histologic spectrum that ranged from lesions in which the majority of the cells had the cytological features of carcinoma in situ to those in which the degree of atypicality was much less. A variety of confusing terms were used to refer to the spectrum of cervical abnormalities with features intermediate between those of carcinoma in situ and normal cervical epithelium including anaplasia, basal cell hyperplasia, and atypical hyperplasia, but the term that came to be most widely used was dysplasia. Dysplasia is derived from the Greek word “dys” that translates to “bad” and “plasia” that translates to “molding” and is a term that has been widely used in many areas of pathology to describe nonmalignant processes. Dysplasia was usually graded as mild, moderate, or severe. The key distinguishing feature of dysplasia was that the atypical cells did not extend through the full thickness of the epithelium or invade the basement membrane. In the cytological nomenclature, dysplasia was considered to be a benign to possibly malignant squamous epithelial atypia, whereas carcinoma in situ was designated as positive for malignant cells.

The separation of noninvasive cervical lesions into two groups, dysplasia and carcinoma in situ, implied that there was a biologic distinction between these two entities and that the two could be reproducibly distinguished from each other. In most centers, dysplasia was considered to be a potentially reversible process and therefore was either ignored, followed, or treated depending on a variety of clinical factors, whereas carcinoma in situ was considered to be a highly significant lesion and patients with this diagnosis were usually treated with hysterectomy. This classification of noninvasive precursor lesions into dysplastic and carcinoma in situ lesions was based solely on arbitrary histologic differences which were often quite subtle (Burghardt 1991; Koss 1978). In the 1960s, several studies of inter- and intraobserver variability of histologic diagnosis demonstrated that pathologists could not reproducibly distinguish between severe dysplasia and carcinoma in situ (Crocker et al. 1968; Kirkland 1963).

A number of studies in the late 1960s suggested that the cellular changes of dysplasia and carcinoma in situ were qualitatively similar and remained constant throughout the histologic spectrum. Both dysplasia and carcinoma in situ were found to be monoclonal proliferations of abnormal squamous epithelial cells with an aneuploid nuclear DNA content (Fu et al. 1983). On the basis of these descriptive biologic studies, Richart introduced the concept that all types of precursor lesions to squamous cell carcinoma of cervix represented a single disease process which he termed cervical intraepithelial neoplasia (CIN) (Richart 1973).

The CIN terminology divided cervical cancer precursors into three groups. CIN 1 corresponded to lesions previously diagnosed as mild dysplasia, CIN 2 corresponded to moderate dysplasia, and CIN 3 to both severe dysplasia and carcinoma in situ, since pathologists could not reproducibly distinguish between the two. At the time of its introduction, CIN was thought to define a spectrum of histological changes that shared a common etiology, biology, and natural history. Furthermore, the diagnostic term cervical intraepithelial neoplasia (CIN) implied that such lesions, if untreated, had a significant, albeit individually unknown, risk of developing into invasive carcinoma in the future. As a corollary, it was presumed that when the histologic changes of CIN were diagnosed and the lesion adequately treated, the development of invasive cancer could be prevented. The CIN terminology became the most widely used histologic terminology for cervical cancer precursors in the 1970s and the 1980s. However, over the last three decades, there has been an explosion of information about the etiology of cervical cancer and its precursor lesions. In the 1950s, Koss and Durfee described epithelial cells from the cervix that had ballooned cytoplasm (Koss and Durfee 1956). They called these cells “koilocytes” from the Greek word “empty space.” In describing these cells, they noted the similarities between them and cells seen in mild dysplasia. Twenty years later, Meisels and Fortin linked koilocytes with infection with human papillomavirus (HPV) (Meisels and Fortin 1976). Today it is now widely accepted that a high proportion of both invasive squamous cell carcinomas and adenocarcinomas of the cervix, as well as their respective precursor lesions, are caused by specific types of high-risk human papillomavirus (HPV) that infect the anogenital tract (Hausen 2002; Hausen 2009). Since infection with specific high-risk types of HPV plays a critical role in the development of cervical cancer, a new model of cervical carcinogenesis has been developed, Fig. 1. This model has three discrete steps: (1) initial infection with a high-risk type of HPV, (2) progression to a histologically defined precursor lesion which requires persistence of the HPV infection, and (3) invasion (Wright and Schiffman 2003; Snijders et al. 2006; Doorbar 2018; Schiffman et al. 2007). Based on this biological model, it is highly unlikely that cervical cancer develops according to a step-wise progression that envisions CIN 1, progressing to CIN 2, to CIN 3 and on to cancer. Moreover, it now appears that the basic premise underlying the CIN terminology is incorrect; the spectrum of histologic changes which are referred to as CIN do not represent a single disease process at different stages in its development but instead two distinct biological entities, one a productive viral infection and the other a true neoplastic process confined to the epithelium (Wright 2006; Schiffman et al. 2007).
Fig. 1

Steps in the development of cervical cancer. Three steps in cervical carcinogenesis. (Used with permission from Wright and Schiffman 2003)

Productive HPV infections of the cervical squamous epithelium are self-limited in the majority of patients and commonly result in flat cervical lesions (Fig. 2) and less frequently in exophytic ones (condylomata acuminata). Flat HPV-associated lesions in which productive viral infection is occurring display cytoplasmic cavitation and nuclear abnormalities. Traditionally, flat or exophytic condyloma were not classified as CIN 1 since they do not have the degree of nuclear atypia usually present in CIN. However, in the 1980s it was found that the HPV types found in cervical flat and exophytic condylomas are the same types that can be found in CIN 1 lesions. Moreover, flat condylomas and CIN 1 lesions are similar with respect to their ploidy status (Fu et al. 1988). Therefore, both exophytic and flat condylomatous lesions of the cervix were combined into the CIN 1 category. These lesions have been designated in the past as koilocytotic atypia, koilocytosis, or flat condyloma.
Fig. 2

Flat HPV-associated cervical lesion. This flat white lesion seen after the application of a weak solution of acetic acid represents a production HPV infection of the cervix

The other entity subsumed within the morphologic CIN spectrum is histologically “high grade.” High-grade lesions are frequently aneuploid and represent true intraepithelial neoplasia with a potential to progress to invasive squamous cell carcinoma if left untreated. High-grade lesions are composed of proliferating basal-type atypical cells with a high nuclear: cytoplasmic ratio and have been designated moderate dysplasia, severe dysplasia, carcinoma in situ, or CIN 2 or 3. There is a common misconception that low grade lesions are “viral,” whereas high-grade lesions are not. The prevalence of high-risk HPV in both low- and high-grade lesions is similar, approximately 80–90% (Clifford et al. 2005, 2006; Smith et al. 2007). However, in low-grade lesions, large numbers of viral particles are produced, whereas in high-grade lesions viral DNA is present, but infectious viral particles are produced in comparatively lower amounts, Fig. 3.
Fig. 3

HPV-infected cells. Electron microscopy of cells productively infected with HPV. (a) There are intranuclear aggregates of HPV in a koilocytotic, superficial cell of a LSIL. The marginated nuclear chromatin is agglutinated, and the cytoplasmic substance displays vacuolar degeneration (vd). The latter corresponds to koilocytotic ballooning on light microscopy. (b) Higher magnification of HPV particles in the nucleus

Because of our increased understanding of the pathogenesis of cervical cancer precursors, it has been suggested that the terminology used to refer to these lesions be changed to better reflect the biological processes that underlie the histologic patterns. The most widely accepted modification is the terminology that has been incorporated into The Bethesda System of cytological diagnosis. This terminology uses the term “low-grade squamous intraepithelial lesion” (LSIL) for lesions previously classified as koilocytotic atypia and CIN 1 and “high-grade squamous intraepithelial lesion” (HSIL) for lesions previously called CIN 2 and CIN 3 (The Bethesda 1988; Solomon et al. 2002). For histopathological reporting, it has been suggested that the terms low-grade squamous intraepithelial lesion (LSIL) and high-grade squamous intraepithelial lesion (HSIL) also be used (Wright and Kurman 1994). The rationale for adapting the two-tiered terminology for histopathology is the same as for using a two-tiered terminology for cytology. First, LSIL (previously referred to as CIN 1) represents a biologically distinct group compared to HSIL (previously referred to as CIN 2 and CIN 3). LSIL is heterogeneous with respect to associated HPV types, clonality, ploidy, and loss of heterozygosity at specific chromosomal loci, whereas HSIL is homogeneous with respect to these parameters. Moreover, the natural history of LSIL is characterized by higher rates of spontaneous regression and lower rates of progression than compared to HSIL. In our opinion, using the term “lesion” rather than “neoplasia” better reflects the natural history of these histopathological entities, since the majority of cervical intraepithelial neoplasia (CIN) are histologically low-grade and represent self-limited HPV infections that will spontaneously resolve in the absence of therapy. “Intraepithelial lesion” better describes these low-grade viral infections than does the term “intraepithelial neoplasia.” It must be emphasized that different terminologies are currently in use by different pathology laboratories. Although many laboratories in the USA have switched to the two-tiered LSIL/HSIL terminology, some continue to utilize the original three-tiered CIN terminology or even the terms dysplasia and carcinoma in-situ. In 2010 the Armed Forces Institute of Pathology’s Fasicle on Tumors of the Cervix began using the LSIL/HSIL terminology for both histopathology as well as cytology (Kurman et al. 2010). In 2011 the College of American Pathologists (CAP) and the American Society for Colposcopy and Cervical Pathology (ASCCP) sponsored the Lower Anogenital Squamous Terminology (LAST) project for HPV-associated squamous lesions of the lower anogenital tract in men and women. The LAST terminology recommends the use of LSIL and HSIL when diagnosing HPV-associated noninvasive squamous lesions at all lower anogenital tract sites including the cervix (Darragh et al. 2012). Although pathologists have long recognized that the histologic diagnosis of CIN 2 is poorly reproducible, there remain clinical reasons for pathologists to try and separate CIN 2 from CIN 3 in some women. Clinical follow-up studies of women with CIN 2 have shown that after 24 months half of them spontaneously regress in the absence of treatment (Tainio et al. 2018). Therefore, clinical management guidelines from both the ASCCP as well as the American College of Obstetrics and Gynecology (ACOG) continue to recommend the separation of CIN 2 from CIN 3 in women desiring future childbearing (Massad et al. 2013a; ACOG 2013). Observation is preferred for women desirous of future childbearing with a diagnosis of CIN 2, acceptable if they have a diagnosis of CIN 2,3 and not recommended for those with CIN 3. To address this clinical need, the LAST terminology allows for the separation of histologic HSIL into two entities, HSIL (CIN 2) and HSIL (CIN 3). In 2014, the World Health Organization (WHO) Classification of Tumors of Women of Reproductive Organs adopted the same terminology (Kurman et al. 2014). Correlations between the different terminologies are shown in Table 1. It should be noted that because older studies utilized either the dysplasia/carcinoma in situ or the CIN terminology, in many places in this chapter we have used the older terminology.
Table 1

Terminologies for cervical cancer precursor lesions

Older classification

CIN classification

WHO/Bethesda System classification

Mild dysplasia



Moderate dysplasia



Severe dysplasia


Carcinoma in situ

aHSIL can be subdivided into HSIL (CIN 2) and HSIL (CIN 3) in specific clinical situations

General Features


SIL is predominantly a disease of women in their reproductive years, with a large population impact and risk factors characteristic of a sexually transmitted disease (STD). The prevalence of SIL in different countries and populations within a country varies widely depending on the underlying risk factors in the population and the extent of cytological screening. Although SIL is not a reportable disease, in most states good estimates of the prevalence of cytologic abnormalities in women undergoing screening in the United States are available from a number of sources. One of the most comprehensive surveys is the College of American Pathologists interlaboratory comparison study that compiles rates of cytological abnormalities diagnosis from over 600 United States cytology laboratories. According to this survey, in 2006 the median reporting rate of LSIL was 2.5% and it was 0.5% for HSIL (Eversole et al. 2010). Age has a profound impact on both the rate of detection of cytological abnormalities and histological SIL. In the ATHENA study which enrolled 46,887 women 21 years or older undergoing routine cervical cancer screening in the USA, the highest rate of LSIL was in women 21–24 years (6.5%), Fig. 4 (Wright et al. 2012). The LSIL rate drops with increasing age and was 3.8% in women 25–29 yrs, 2% in women 30–39 yrs, and < 1% in women 40 years and older. In ATHENA, the rate of HSIL showed a similar age distribution. HSIL was found in 0.7% of women 21–24 years, 0.4% of those 25–39 years and dropped to 0.2% and 0.1% in women 40–49 years and ≥50 years, respectively. Almost identical rates and age distributions for LSIL and HSIL among US women undergoing cervical cancer screening have recently been reported in the BD Onclarity study (Stoler 2018). A similar impact of age as is seen with cytological abnormalities is also found for histologic LSIL and HSIL in ATHENA, Fig. 5 (Wright et al. 2012). The peak age for the diagnosis of histologic LSIL and HSIL is 25–29 years of age and rates drop considerably in older women. In contrast, the rate of histologically confirmed invasive cervical cancer reported in the US Surveillance Epidemiology and End Results (SEER) cancer registry increases until 40 years of age and then shows only minimal changes through 65 years of age, Fig. 6 (Wang et al. 2004a). After age 65 years there is a minimal decline. It is also important to note that during the 1970s–1990s the rates of HSIL appear to have increased in both Western Europe and the USA. In Iceland, for example, the rate of detection of HSIL in women 20–24 years of age increased almost four fold from 1979–1983 to 1994–1998 and then leveled out, Fig. 7 (Sigurdsson and Sigvaldason 2007). In women 25–29 years of age it almost doubled during this time period and then appears to have begun to decrease.
Fig. 4

Impact of age on cytological abnormalities. (Data come from a large US cervical cancer screening trial Wright et al. 2012)

Fig. 5

Impact of age on histologic SIL. (Data come from a large US cervical cancer screening trial Wright et al. 2012)

Fig. 6

Incidence of cervical cancer in the USA. The impact of age on the incidence of cervical cancer is derived from surveillance data from the National Cancer Institutes’ Surveillance Epidemiology and End Results (SEER) cancer registry. (Modified from Wang et al. 2004a)

Fig. 7

Changes in detection of HSIL in Iceland between 1979 and 1998. (Modified from Sigurdsson and Sigvaldason 2007)

Vaccination against HPV is already causing dramatic reductions in the rates of cytological abnormalities and histologic HSIL in countries such as Australia and Denmark that have achieved high vaccine coverage. The Australia national vaccination program was carried out between 2007 and 2009 and resulted in over half of women 12–26 years being vaccinated with the quadrivalent HPV vaccine (HPV 6, 11, 16, 18) (Brotherton 2016). By 2013 the Victorian Cervical Cytology Registry showed a significant reduction in both high-grade cytological abnormalities and histologic HSIL among women who had been vaccinated in school-based programs between ages 12 and 17 years. The prevalence rate of HSIL+ was 4.8 per 1000 person years in vaccinated women compared to 6.4 per 1000 in unvaccinated women (Gertig et al. 2013). By 2016 this same registry reported that the prevalence rate of HSIL+ among women 25–29 years had declined by 17% compared to the preceding 2 years (Brotherton et al. 2016). A Danish registry study found that between 2010 and 2013 there was a markedly decreasing trend in the incidence of HSIL+ for young women 12–20 years who had been targeted by the Danish HPV vaccination program (Baldur-Felskov et al. 2014).


Epidemiologic studies have documented that the major risk factors for cervical cancer are infection with specific high-risk types of HPV and a lack of cervical cancer screening, Table 2. Other risk factors that have classically been associated with cervical cancer play a much less important role. Among women infected with high-risk types of HPV, factors such as smoking, immunosuppression, and long-term use of oral contraceptives can result in a doubling or tripling of risk for HSIL and invasive cancer (Schiffman et al. 2007). The mechanisms of action of any of the risk factors other than infection with HPV and lack of screening are not understood.
Table 2

Risk factors associated with SIL in various epidemiological studies

Sexual activity

Number of sexual partners

Early sexual activity (especially less than 16 years of age)

Sexually transmitted diseases

Human papillomavirus

Herpes simplex virus

Chlamydia trachomatis

Early age of first pregnancy


Low socioeconomic class

Cigarette smoking

Human immunodeficiency virus

Immunosuppression from any cause

Interval since last Pap smear

Oral contraceptive use

SIL squamous intraepithelial lesion, CIN cervical intraepithelial neoplasia

Human Papillomaviruses

In the late 1970s based on theoretical considerations, Dr. Harald zur Hausen suggested that there might be an association between HPV and cervical cancer, an achievement for which he received the Nobel Prize in Medicine in 2008 (Hausen 1977, 2009). A large number of epidemiological, clinicopathological, and molecular studies subsequently linked the presence of specific types of HPV to the development of anogenital cancers and their precursors, and it is now accepted that infection with high-risk HPVs plays a critical role in the pathogenesis of most cervical cancers and their precursor lesions (Human Papillomaviruses 2007).

Classification of HPV and Association with Specific Types of Anogenital Lesions

Papillomaviruses are classified as members of the family Papillomaviridae (Human Papillomaviruses 2007). The papillomaviruses are double-standard DNA tumor viruses that have a DNA genome of approximately 8,000 base pairs in length, a nonenveloped virion that measures 45–55 nm in diameter, and an icosahedral capsid composed of 72 capsomers. Papillomaviruses are widely distributed throughout nature. There are bovine, canine, avian, rabbit, deer, and human papillomaviruses. They are all highly species specific viruses which infect only one species. This suggests an evolutionary history spanning over 300 hundred millions years (Herbst et al. 2009). Within a given species, many types and subtypes of papillomaviruses may exist. Unlike many other viruses in which specific viral isolates have capsid proteins with different antigenic structures, the capsid proteins of papillomavirus are highly conserved and antibodies directed against bovine papillomavirus (BPV) capsid proteins cross-react with human papillomaviruses (Jenson et al. 1980). Therefore, DNA sequence is used to classify different viral types (genotypes).

The classification of papillomaviruses is based on phylogenetic algorithms that compare either the whole viral genomic sequence or specific subgenomic segments. As of 2004 the phylogenetic classification of papillomaviruses includes 118 distinct types, Fig. 8 (de Villiers et al. 2004). The human papillomaviruses are included in the genus Alphapapillomavirus and consists of a number of closely related groupings referred to as species or clades. The most important of these are the A9 clade which has HPV 16 as the prototypic virus and the A7 clade which includes HPV 18 and 45. To date, over 200 genotypes have been sequenced and listed in the NIH’s Papillomavirus Episteme Database (Papillomavirus Episteme 2018). In order to be classified as a separate HPV genotype, there needs to be at least 10% difference in the base-pair sequence of the highly conserved L1 region (major capsid protein) compared to other genotypes (Burk et al. 2013). Although the different types of HPV are quite similar structurally, they have significant specificity with regard to the anatomic location of the epithelia that they infect and the type of lesions that they produce at the site of infection. Within a given genotype, there can also be a number of genetic variants. HPV 16 is now recognized to have 4 main variant lineages that differ in their L1 sequence by less than 10% and there are 9 subvariants that can have as little as 0.5% L1 variation (Burk et al. 2013). Similarly, there are three distinct lineages of HPV 18 variants. The three lineages of HPV 18 variants appear to have diverged about the time homosapiens began establishing residence in different continents (Bernard et al. 1994). There is now compelling data that intratypic variants are important in determining the pathogenicity of the viruses. HPV 16 variants are important in influencing whether the virus will become persistent, whether cancer will develop, and even the histological subtype of the cancer that develops (Burk et al. 2013).
Fig. 8

Phylogenetic classification of papillomaviruses. Based on their DNA sequence, papillomaviruses can be group in closely related clusters or clades. (Used with permission from de Villiers et al. 2004)

Papillomaviruses are epitheliotrophic viruses which predominantly infect skin and mucous membranes and produce characteristic epithelial proliferations at the sites of infection. These benign epithelial proliferations or papillomas have the capacity to undergo malignant transformation under certain circumstances. Examples of this in animals include the papillomas induced in domestic rabbits by the cottontailed rabbit papillomavirus (CRPV) which can progress to invasive squamous cell carcinomas when treated with topical applications of methylcholantrene and alimentary tract papillomas induced in cattle by bovine papillomavirus (BPV) which undergo malignant transformation when the animals eat radiomimetic bracken ferns. In humans, HPV infections occur on the skin and mucous membranes, on the conjunctiva, oral cavity, larynx, tracheobronchial tree, esophagus, bladder, anus, and genital tract of both sexes. HPVs appear to be fastidious in their growth requirements and replicate only in the nucleus of infected cells. In addition to being species specific, papillomaviruses are also relatively tissue and site specific.

More than 40 types of HPV can infect the anogenital tract of which about 20 types are commonly encountered, Table 3. Based on their associations with cervical and anogenital cancers, 13 anogenital HPVs have been classified by the International Agency for Research on Cancer (IARC) as oncogenic. These are HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68 (Human Papillomaviruses 2007). Others also consider HPV 82 to be oncogenic (Munoz et al. 2006). HPV types 6 and 11, which are the two types most commonly found in association with condyloma accuminata, are not implicated in the development of cervical cancer, but are associated with squamous cell carcinomas of the larynx and various carcinomas of the vulva, penis, and anus (Human Papillomaviruses 2007).
Table 3

Oncogenic risk of common types of anogenital human papillomavirus

Low oncogenic risk:a

6, 11, 42, 43, 44, 53

High oncogenic risk

16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68

Unclear oncogenic risk:

26, 66, 73, 82

aFor cervical cancer

Essentially all HSIL is found to be associated with HPV when sensitive molecular methods are used to detect HPV DNA (Petry et al. 2016). LSIL can be associated with any of the anogenital types identified in women in the general population. A meta-analysis of HPV types identified in women with either cytologic LSIL or histologic LSIL found that HPV was detected in 80% of LSIL from North America and approximately 70% for LSIL from other regions of the world (Clifford et al. 2005, 2006). This difference most likely is due to regional differences in how lesions are diagnosed. HPV 16 was the most common genotype, being found in 19% of HPV positive cases, Fig. 9. HPV types 31, 51, and 53 were the next most commonly identified types, each being identified in approximately 7–8% of lesions. HPV 16 was less likely to be found in lesions from Africa and HPV 18 was more common in lesions from North America than in lesions from Europe or Latin America. All of the associations are complicated by the fact that multiple types of HPV are frequently found in association with low-grade lesions.
Fig. 9

Prevalence of anogenital HPV types in LSIL. (Modified from Clifford et al. 2006)

In contrast to LSIL, most studies have found that approximately half of HSIL is associated with HPV 16. The prevalence of HPV 16 in cytologic HSIL and histologic HSIL in studies from different areas of the globe ranges from 30% to 70% (Smith et al. 2007; Clifford et al. 2006). A meta-analysis of the distribution of HPV in HSIL has concluded that HPV 16 is identified in 45.3% of the lesions, HPV 18 in 6.9% and HPV 31 in 8.6%, Fig. 10a (Smith et al. 2007). HPV types 33, 58, and 52 are the next most common types found in HSIL. Multiple types of HPV are less commonly found in HSIL than in LSIL. In order to understand subtle differences in the distribution of HPV genotypes in women with HSIL and the effects of age across Europe, two large multicenter studies have been conducted using PCR on paraffin-embedded tissue blocks from over 3,000 women with histologic HSIL (Tjalma et al. 2013). Only 1.5% of the women with HSIL were HPV negative and women with HPV negative HSIL tended to be older, with most cases diagnosed in women >61 years. Multiple types of HPV were identified in 17% of the HSIL. In women with a single HPV type and HSIL, the most common types were HPV 16 (60%), HPV 33 (11%), HPV 31 (9%), HPV 52 (4%), and HPV 18 (4%). The prevalence of HPV 16 increased with lesion grade being found in 43% of CIN 2 lesions and 64% of CIN 3. The average age of women with HPV 16 positive CIN 3 was younger (34 years) than for those with HPV 18 positive CIN 3 (42 years).
Fig. 10

Prevalence of anogenital HPV types. (a) HSIL. (b) Invasive cervical cancer. (Modified from Smith et al. 2007)

The distribution of HPV types in invasive cervical cancers shows an even stronger enrichment for HPV 16 and 18 that is observed in women with HSIL, Fig. 10b (Smith et al. 2007). One study used a sensitive PCR method to identify and type HPV in paraffin-embedded samples from over 10,000 women with invasive cervical cancer (de Sanjose et al. 2010). The most common HPV types were 16, 18, 31, 33, 35, 45, 52, and 58. These genotypes combined accounted for 91% of the HPV positive cases. HPV 16 was the single most commonly identified genotype and was found in 61% of cervical cancers. HPV 16 was more commonly found in squamous cell carcinomas (62%) than in adenocarcinomas (50%). In contrast, HPV 18 and 45 are more commonly found in adenocarcinomas (32% and 12%, respectively) than in squamous cell carcinomas (8% and 5%, respectively). HPV 16, 18, and 45 combined accounted for 94% of cervical adenocarcinomas. HPV 31, 33, and 52 are each found in 3–4% of cervical cancers (de Sanjose et al. 2010). None of the other high-risk types are found in more than 2% of cervical cancers and most are found in 1% or less. As with HSIL, multiple HPV types are rarely found in association with invasive cervical cancers.

Genomic Organization of HPV

The genomic organization of the different types of HPV appears to be similar, Fig. 11. The viral genome can be divided into three regions: an upstream regulatory region (also referred to as the long control region or LCR), the early region, and the late region. The long control region (LCR) is a noncoding region of the viral genome that is important in regulating viral replication and transcription of downstream sequences in the early region. The early region is transcribed early in the viral life cycle (hence its name) and encodes predominately for proteins that are important in viral replication, whereas the late region encodes for viral structural proteins that are produced late in the viral life cycle.
Fig. 11

Genomic organization of HPV. HPV is a double-stranded, circular, DNA tumor virus whose genome can be divided into three regions; the upstream regulatory region (URR), the early region, and the late region

The early region open reading frames (ORFs) encode for proteins required for viral replication and maintenance of a high viral copy number in infected cells (Doorbar et al. 2012). The early region also includes the transforming regions of the HPV genome: E5, E6, and E7. The E6 and E7 ORFs encode the major transforming genes of HPV (Hausen 2002). The E5 ORF encodes a protein with weak transforming capacity. Both the E6 and E7 proteins are small zinc-binding proteins that lack endogenous enzymatic activity and exert their transforming activity through binding to cell regulatory proteins (Doorbar 2007, 2018). E6 can bind to p53 and results in rapid proteolytic degradation of p53 through an ubiquitin-dependent pathway, thus blocking apoptosis. E7 binds to the retinoblastoma (Rb) gene product, as well as other “Rb-like proteins.” Binding of E7 to Rb blocks the cell proliferation inhibiting function of these endogenous tumor suppressors. E7 can also activate cyclins A and E as well as block the cell proliferation inhibiting functions of WAF-1 and p27, two cyclin-dependent kinase inhibitors. The end result of overexpression of E6 and E7 within cells is unrestricted cell proliferation and a blockage of apoptosis.

The late region of HPV is downstream of the early region and contains two ORFs termed L1 and L2 which encode capsid proteins. The L1-encoded protein is the major capsid protein and is highly conserved among papillomaviruses from all species. The L2-encoded protein is a minor capsid protein which is much more variable among viral types. Transcription from the L1 and L2 ORFs occurs as a late event in the viral life cycle at a time when infectious virus is being produced. L1 capsid proteins produced in in vitro culture systems are capable of associating and forming viral-like particles (VLPs) that are similar to native virions, but lack the viral genome. VLPs composed of L1 capsid proteins are the antigens used in the prophylactic HPV vaccines.

Life Cycle of HPV

Although the HPV life cycle is not completely characterized, the rough outlines of the process are known Fig. 12 (Doorbar 2018; Snijders et al. 2006; Doorbar et al. 2012). In order to initiate an infection, virus needs to gain access to the basal lamina and either basal cells or primitive “basal-like” cells of the immature squamous epithelium. HPV presumably reaches these cells through small defects or microtraumas of the epithelium. The development of transforming infections that can evolve into HSIL lesions and invasive cervical cancers is postulated to require that the virus gain access to specialized squamocolumnar junction cells that have a unique morphology and gene expression profile (Herfs et al. 2012). Localization in the basal layer may be due to the presence of specific receptors for HPV on the basal cells. Infection of basal cells appears to require the cells to be actively dividing. Once HPV enters into the proliferating basal cell, it limits the ability of the infected basal cell to differentiate. This appears to be largely mediated by E6. In the lower epithelial layers, the expression of the E7 protein stimulates cellular proliferation. The combination of a delayed commitment to differentiate and sustained cell proliferation allows for viral persistence and expansion of the infected basal cell population as noninfected basal cells differentiate (Doorbar 2018; Doorbar et al. 2012). Interference by HPV with the cellular proliferation and differentiation pathways occurs in an ordered way. This assures a reservoir of HPV-infected basal cells that can lead to long-term persistence as a nonproductive HPV infection. In the literature, nonproductive HPV infections have frequently been referred to as latent infections. Usually in nonproductive latent infections, a small number of copies of the HPV genome remain in the nucleus in a free circular form called an episome. Replication of the episomal DNA in latent infections is coupled to the replication of the epithelial cells and occurs in concert with replication of the host cell’s chromosomal DNA (Hoffmann et al. 2006). Since complete viral particles are presumably not produced in latent infections, the characteristic cellular effects of a HPV infection are not present and HPV can only be identified using molecular methods. Latently infected epithelium displays no morphologic abnormality. Thus, the term latent infection is used to characterize HPV infections in which there is no gross or microscopic evidence of a HPV-induced epithelial lesion, and the virus is present at such low copy number in the epithelium that it cannot be detected with routine molecular detection methods (Gravitt 2011).
Fig. 12

HPV life cycle. (Source: Kahn 2009)

The other form of HPV infection is a productive viral infection. The development of a productive infection depends on a steady migration of HPV-infected cells from the basal reservoir into the upper layers of the epithelium. In productive viral infections, viral DNA replication occurs independently of host chromosomal DNA synthesis. This independent viral DNA replication produces large amounts of viral DNA and results in infectious virions. Viral DNA replication takes place predominantly in the intermediate and superficial cell layers of the stratified squamous epithelium. As the virally infected epithelial cells mature and move towards the epithelial surface, there is continued expression of E6 and E7 proteins. HPV E7 expression results in cells of the parabasal and mid-epithelial layers re-entering the cell cycle and providing the DNA replication machinery required for viral replication. Cell-derived, differentiation-specific transcriptional factors produced by the epithelium stimulate the production of viral capsid proteins. Viral gene expression patterns change to favor genome amplification over viral persistence. This change appears to involve alterations to viral promotor usage and mRNA splicing (Doorbar 2018). This allows large amounts of intact virions to be assembled in the superficial cell layers and produces the characteristic viral changes that can be detected both cytologically and histologically, Figs. 13a, b. These viral changes include acanthosis, cytoplasmic vacuolization, koilocytosis, multinucleation, and nuclear atypia. Deregulated E6 and E7 expression is believed to lead to the accumulation of genetic errors in productively infected cells that predisposes them to progress to HSIL (CIN 3) and invasive cervical cancer (Isaacson Wechsler et al. 2012).
Fig. 13

Cytopathic effects of HPV. The cytopathic effects of HPV include nuclear enlargement, nuclear pyknosis or hyperchromaticity, anisocytosis, multinucleation, and perinuclear cytoplasmic vacuolization. (a) Histologic features of a lesion classified as LSIL. (b) Cytological features of LSIL

In order for a HPV infection to persist and progress to HPV-associated precancerous and cancerous lesions, it is important that it avoids the host immune response (de Sanjose et al. 2018). One strategy that is used to avoid the host immune response is to maintain a low profile (Stanley 2012). All of the HPV life-cycle takes place in differentiating keratinocytes that are programed to undergo cell death and be released from the epithelial surface remote from immune activity. HPV does not cause virus-induced cell death and hence there is no inflammation or danger signal made to the immune system. Langerhans cells are not recruited to the sites of HPV infection. Although keratinocytes normally have the capacity to respond to cell injury and the presence of pathogens and mediate immune response, HPV infections downregulate innate immune sensors on the keratinocytes and suppress the type 1 interferon response as well as downregulates cytokine responses (Stanley 2012). Through these and other mechanisms, HPV evades the innate immune response and there is a delay in activation of adaptive immunity.

Epidemiology of HPV Infections

Anogenital HPV infections are primarily transmitted by direct skin-to-skin or mucosa-to-mucosa contact. Based on prevalence studies among both men and women in the population, it appears that although sexual intercourse is the most common route of infection, intercourse is not required for transmission (Burchell et al. 2006; Gravitt 2011). The efficiency of transmission during sexual intercourse is unknown, but clearly quite high with some estimates indicating a 40% transmission risk per act of intercourse (Burchell et al. 2006). Infection with multiple types of HPV is seen in 20%-30% of infected young men and women. Condom use reduces but does not eliminate the risk of transmission to women. It also appears that male circumcision reduces transmission and carriage of HPV (Winer et al. 2006).

Most women become infected with HPV within several years of initiating sexual activity (Winer et al. 2003; Brown et al. 2005). Cumulative exposure is difficult to measure because most infections are transient and relatively short-lived and serology is an insensitive indicator of exposure since many infected women do not produce measurable levels of antibody (Ho et al. 2004). However, prospective follow-up studies of female college students report that after 5 years of follow-up 80% have been infected at some point with HPV (Winer et al. 2003). Follow-up studies indicate that a frequent pattern of infection in sexually active young women is multiple serial infections with different types of HPV with each infectious episodes being of relatively short duration (Brown et al. 2005).

The majority of HPV infections are transient and undergo clearance or become latent within 1–2 years of detection (Burchell et al. 2006; Schiffman et al. 2011; Moscicki et al. 2012). Clearance or development of latency is presumed to be mediated by cell-mediated immunity (Gravitt 2011; Doorbar 2018; Stanley 2012). After an infection has been persistent for 36 months, the potential for future clearance diminishes considerably (Schiffman et al. 2007). Moreover, the longer an infection persists, the greater the likelihood that a high-grade precursor lesion will develop (Schiffman et al. 2007; Elfgren et al. 2017; Moscicki et al. 2012). Therefore, infections caused by high oncogenic risk HPV infections that persist for 2 or more years pose the greatest risk to women since these are the infections that may progress to a high-grade cervical cancer precursor or even an invasive cervical cancer. Approximately 10% of HPV infections will persist for 2 or more years. Long-term prospective cohort studies have demonstrated that in some women HPV infections that have appeared to have cleared can reappear. Moreover, follow-up studies of HIV-infected women show a strong relationship between increasing levels of immune suppression and detection of HPV (Wright and Kuhn 2006). This suggests that in many instances women who appear to have cleared HPV infections actually harbor latent HPV infections in which the virus remains in the epithelium at a low copy number (Gravitt 2011; Moscicki et al. 2012). The combination of multiple anogenital types of HPV, all of which appear to be highly transmissible in sexually active populations, combined with the transient nature of most infections explains the prevalence of HPV in the population. Prevalence of high-risk HPV DNA positivity is high in women with normal cytology in their late teens and 20s and drops with increasing age, Fig. 14 (Datta et al. 2008). Reactivity of latent infections could explain the slight increase in HPV prevalence seen in postmenopausal women (de Sanjose et al. 2007, 2018). Developmental status of the country of residence also appears to be important (de Sanjose et al. 2007, 2018). The reasons for the increased HPV prevalence in less developed countries are unknown, but many relate to sexual practices and poor hygienic conditions as well as an increase burden of comorbid disease in the population.
Fig. 14

Prevalence of high-risk HPV DNA in women with normal cytology in the USA. This data comes from a large, CDC surveillance survey. (Modified from Datta et al. 2008)

Development of Cervical Disease After HPV Infection

About one-third of the HPV-DNA-positive women will have a cytological abnormality (Schiffman et al. 2007). The cumulative incidence of minor cytological abnormalities in HPV-infected women with initial normal cytology is 25–50% within 1–2 years. The risk of a cytological abnormality declines to baseline level in the population by 4 years (Castle et al. 2002; Moscicki et al. 2001). Risk factors for HPV persistence and the subsequent development of HSIL are not yet well characterized; however, HPV type is clearly important. HPV 16 infections are especially oncogenic. The 3 year cumulative risk for CIN 3 is approximately 25% for women 25 years and older with a prevalent HPV 16 infection, and the 12 year cumulative risk for CIN 3 is approximately 50% in women with a persistent HPV 16 infection for 2 years (Kjaer et al. 2010; Wright et al. 2015). Risks of persistence and subsequent development of HSIL are also increased for HIV-infected and other immunosuppressed women (Wright and Kuhn 2006).

Most HSIL begins at the squamocolumnar junction of the transformation zone, with one edge of the lesion bordering the endocervical columnar epithelium. Only about 10% of SIL will involve the endocervical canal without involving the squamocolumnar junction (Abdul-Karim et al. 1982). In general, the portion of SIL on the exocervical portio surface is low-grade, whereas the portion of SIL that extends into the endocervical canal is high-grade, Fig. 15. It has been proposed that the cell of origin of most invasive cervical cancers is a distinct population of residual embryonic cells found within the cervical transformation zone. These cells are cuboidal in shape and have been referred to squamocolumnar junction cells (Herfs et al. 2012). They express a 5 protein signature that is also found in almost all junctional LSIL, HSIL, and invasive cervical cancers but is not expressed by the regular basal cells of the portio epithelium or in LSIL arising on the ectocervix. The 5 protein signature of these squamocolumnar junction cells includes cytokeratin 7, anterior gradient 2 (AGR2), cluster differentiation-63 (CD-63), matrix metalloproteinase 7 (MMP7), and guanine deaminase (GDA) (see chapter “Benign Diseases of the Cervix”).
Fig. 15

Distribution of SIL on the cervix. When both HSIL and LSIL are present on the cervix, the HSIL generally develops internally to the LSIL. It is much more likely that HSIL will extend into the endocervical canal

Because of the insensitivity of colposcopy to detect small lesions, there are limited data on how long it takes for a persistent HPV infection to develop into a HSIL or invasive cancer. Some authors hold the belief that the subsequent development of HSIL and invasive cervical cancer in persistently infected women takes an average of 5–14 years if undetected and not treated (Woodman et al. 2007). Others believe that persistence of a high-risk HPV infection for more than 7 years without the development of HSIL (CIN 3) is uncommon (Moscicki et al. 2012; Castle et al. 2011). The peak prevalence of HPV infection in the population is in the late teens/early 20s, whereas cervical cancer incidence in unscreened populations plateaus from around 35 to 55 years of age. HSIL is consistently more common than are invasive squamous cell carcinomas in unscreened populations, suggesting that only a minority of HSIL develops the capacity to become invasive cancers (Holowaty et al. 1999; Gravitt 2011). Long-term follow-up studies of untreated or partially treated HSIL have found that 30–50% progress to invasive cancer over a 30 year follow-up period (McCredie et al. 2008).

Other Risk Factors

It should be stressed that although infection with specific high-risk types of HPV is necessary for the development of invasive cervical cancer, it is not sufficient for the development of cervical cancer. The long latency between the initial exposure to HPV and the development of cervical cancer as well as the fact that only a small fraction of women exposed to HPV develop cervical disease suggest that additional steps and perhaps other co-factors are necessary in the pathogenesis of cervical neoplasia. It should also be recognized that because of the strength of the association between HPV infection and cervical disease, it is difficult to evaluate the role of other risk factors (Gravitt 2011).

Cigarette smoking has been associated with the development of cervical cancer (Bosch and de Sanjose 2007). In a comprehensive review of the literature, Szarewski concluded that a positive association between cigarette smoking and the development of cervical cancer had been reported by the majority of studies designed to address this question (Szarewski and Cuzick 1998). There are a number of possible mechanisms that could account for an association between cervical cancer and cigarette smoking. One is the secretion of cigarette smoke by-products, including nicotine and cotinine, in cervical mucous of tobacco users and women passively exposed to cigarette smoke (McCann et al. 1992). Another possible mechanism that could account for the association is the effect of cervical smoke by-products on the number and distribution of immune effector and regulatory cells such as Langerhans cells in the cervix (Szarewski et al. 2001).

Multiparity and the use of combined oral contraceptives are also recognized as a significant risk factor for the development of both cervical cancer precursors and invasive cervical cancers. A meta-analysis of 28 studies evaluating the risk of cervical cancer in women on oral contraceptives demonstrated that the relative risk of invasive cervical cancer increased with increasing duration of contraceptive use. After 10 years of use, the summary relative risks for were 2.2 (95% CI. 1.9–2.2) for all women and 2.5 (95% CI. 1.6–3.9) for HPV positive women (Smith et al. 2003). A recent analysis of the 308,036 women enrolled and followed up to 9 years in the European Prospective Investigation into Cancer and Nutrition Study (EPIC) found that increasing number of pregnancies was associated with an increased risk of CIN 3 (Roura et al. 2016). In the EPIC study, the use of oral contraceptives for ≥15 years versus never used was associated with a significant increase in risk of both CIN 3 and invasive cervical cancer (Roura et al. 2016).

Cellular immune responses against HPV clearly play an important role in determining whether HPV infections are ultimately cleared or whether they persist and can develop into cervical neoplasia (Stanley 2012). Immunosuppression is a well-established risk factor for the development of both SIL and cervical cancer. Renal transplant recipients have a relative risk of 13.6 for the development of cervical carcinoma in situ compared to women in the general population (Wright and Kuhn 2006). Infection with human immunodeficiency virus (HIV) is also an important risk factor for SIL and invasive cervical cancer (Denslow et al. 2014). HPV infections are more prevalent and tend to be more persistent in HIV-infected women (McDonald et al. 2014). Numerous studies have documented a higher prevalence of cervical neoplasia including invasive cancer among HIV-infected women compared to various control groups of HIV-uninfected women (Denny et al. 2012). A recent meta-analysis found that prolonged use of antiretrovirals (ART) in HIV-infected women reduces their risk of acquiring HPV as well as reduces their risk of SIL (Kelly et al. 2018).

Clinical Features

SIL appears to be somewhat more frequently detected on the posterior lip of the cervix compared to the anterior lip of the cervix and is rarely seen at the lateral cervical angles (Pretorius et al. 2006; Guido et al. 2005). SIL may expand horizontally and involve the entire transformation zone, but it usually does not extend onto the native portio epithelium. The endocervical extension of SIL is not restricted and extension along the entire endocervical canal and into the uterus can rarely occur. The size and endocervical distribution of SIL tend to vary directly with increasing severity of lesion grade. HSIL usually has the largest surface area and more frequently involves the endocervical canal.

Pathologic Findings

SIL is characterized by abnormal cellular proliferation, abnormal maturation, and cytologic atypia. The cytologic abnormalities include hyperchromatic nuclei, abnormal chromatin distribution, nuclear pleomorphism, and increased nuclear: cytoplasmic ratio. Nuclear atypia is the hallmark of SIL. The nuclear borders are irregular, and the chromatin is coarse, granular (salt and pepper), or filamentous throughout the nuclear mass.

The traditional grading of SIL was based on the proportion of the epithelium occupied by basaloid, undifferentiated cells, reflecting a progressive loss of epithelial maturation and decreasing glycogenization with increasing lesion severity. The spectrum of epithelial alterations that comprises SIL was therefore semiquantitatively classified into three categories: CIN grade 1 – neoplastic, basaloid cells occupying the lower third of the epithelium; CIN grade 2 – basaloid cells occupying the lower third to two-thirds of the epithelium; and CIN grade 3 – basaloid cells occupying two-thirds to full thickness of the epithelium, Fig. 16. Adoption of The Bethesda System nomenclature to histologic classification as has now been formally done by both the WHO and the College of American Pathologists results in a two tier rather than a three tier grading system.
Fig. 16

Cervical squamous carcinoma precursors. Schematic representation of cervical cancer precursors and the different terminologies that have been used to refer to them. The risk of developing microinvasion from different states of SIL is arbitrarily represented and is not necessarily proportional to that illustrated in this scheme

Low-Grade Squamous Intraepithelial Lesion

HPV-induced cytologic and histologic changes are considered the most characteristic features of LSIL. The most important of these changes is significant nuclear atypia. Nuclear atypia is characterized by variation in nuclear size with nuclear enlargement, hyperchromasia, and irregularity and wrinkling of the nuclear membrane. Nuclei often vary up to three fold in size and have quite variable staining patterns, Fig. 17. However, in LSIL there is usually minimal nuclear atypia in the epithelial cells residing in the lower third of the epithelium, Fig. 18. Cells in the superficial layers of the epithelium may have nuclei that are somewhat smaller and pyknotic. Squamous cells with productive HPV infections also frequently show perinuclear cytoplasmic cavitation or halos that is accompanied by thickening of the cytoplasmic membrane, Fig. 19a. These halos are best appreciated in cytology specimens, Fig. 19b. The halos develop because productive HPV infections induce cytoskeletal abnormalities that may lead to cytoplasmic cavitation. The combination of significant nuclear atypia and cytoplasmic halos has been termed koilocytosis or koilocytotic atypia and is pathognomonic of a productive HPV infected. Mitotic spindle abnormalities also occur in productive HPV infections and these appear to interfere with mitosis and cytokinesis. This leads to the polyploidy and bi- or multinucleated cells that are usually present in productive HPV infections, Figs. 17 and 19 (Fu et al. 1983). Polyploid cells are cytologically atypical and are readily recognized as being “abnormal” on either cytology or histology. Taken together, the histologic and cytologic features of koilocytosis, nuclear atypia, architectural abnormalities, and multinucleation are pathognomonic of an HPV-infected epithelium at any site in the lower genital tract and are especially prominent in LSIL.
Fig. 17

Nuclear atypia in LSIL. The most significant feature of LSIL is nuclear atypia. This is characterized by nuclear enlargement, hyperchromasia, nuclear irregularity, and variation in nuclear size

Fig. 18

LSIL with minimal cytological atypia in lower third of the epithelium. Typically the cells in lower third of the epithelium show minimal atypia in LSIL

Fig. 19

Koilocytosis in LSIL. The cytological features of a productive HPV infection include multinucleation and perinuclear cytoplasmic cavitation or halos. The combination of nuclear atypia and cytoplasmic halos is referred to as koilocytosis. (a) Koilocytosis on histology, (b) Koilocytosis on cytology

The architectural abnormalities associated with LSIL are due to a proliferation of basal and parabasal cells in the infected epithelium. The architectural abnormalities associated with LSIL causes the lesional tissue to appear quite different from the adjacent epithelium. Therefore, LSIL can usually be identified at lower magnifications based on the increased thickness of the epithelium accompanied by cells with prominent hyperchromatic nuclei in the upper levels of the epithelium, Fig. 20. HPV-induced hyperplasia can be highly variable and takes many forms, but is most commonly characterized by papillomatosis and acanthosis, Fig. 21. One of the more common patterns of the acanthosis is that of moderate epithelial thickening and an undulating, slightly raised surface. In the older literature, cervical lesions with HPV-associated cytopathic effects and only a moderate degree of epithelial thickening were referred to as condyloma planum or flat condyloma. Other types of epithelial hyperplasia that can occur in productive HPV infection are multiple papillary fronds containing fibrovascular cores and pointed epithelial spikes, Fig. 22. Colposcopically these can present as a exophytic condyloma accuminatum similar to those that occur on the vulva or vagina or as a less exophytic lesion with prominent, fine surface spikes. The latter are frequently referred to in the colposcopic literature as spiked condyloma. A third type of papillary LSIL is referred to as immature condyloma (e.g., papillary immature metaplasia). These represent infections of the transformation zone epithelium by HPV 6 or 11. These lesions typically have a filiform papillary architecture, Fig. 23a. Immature condylomas can be viewed as part of a histologic spectrum that ranges from squamous papillomas at the benign end to papillary carcinomas at the malignant end. Because HPV 6 and 11 infected transformation zone epithelial cells do not mature, koilocytotic features are minimized and these lesions retain an immature metaplastic phenotype, Fig. 23b. Cells in immature condylomas demonstrate only a mild degree of nuclear atypia and typically have a relatively low mitotic index.
Fig. 20

LSIL recognized at low magnification. LSIL can usually be recognized at relatively low magnification based on the increased thickness of the epithelium accompanied by prominent hyperchromatic nuclei in upper level of the epithelium

Fig. 21

LSIL. Papillomatosis, acanthosis, parakeratosis, and hyperkeratosis are frequently present

Fig. 22

LSIL of type referred to as spiked-condyloma. This lesion shows exaggerated epithelial thickening with multiple papillary fronds, each of which contains a fibrovascular core. The individual cells in these lesions are similar to the cells seen in flat lesions. Such lesions are referred to as “spike-condyloma” on colposcopy

Fig. 23

LSIL of type referred to as immature condyloma. These lesions occur in the transformation zone and are associated with HPV 16 or 11. (a) They typically form thin, papillary projects. (b) Typically the cells have a metaplastic phenotype with minimal koilocytosis

The surface of LSIL frequently has a layer of parakeratosis and somewhat less commonly hyperkeratosis with an associated granular layer, Fig. 24. When gland involvement by the HPV-infected epithelium and acanthosis predominates, the histological pattern appears endophytic and superficially resembles that of an inverted nasal papilloma.
Fig. 24

LSIL with both hyperkeratosis and parakeratosis. Both hyperkeratosis and parakeratosis are frequently associated with LSIL

Differential Diagnosis

Studies measuring interobserver variability of histologic diagnosis of cervical lesions demonstrate that although agreement between pathologists is excellent for invasive lesions, and moderately good for HSIL, it is poor for LSIL (Ismail et al. 1989; Robertson et al. 1989; Shin et al. 2000; Stoler and Schiffman 2001; Malpica et al. 2005). In the NCI-sponsored ALTS multicenter study, 2,237 colposcopically directed biopsies that had been initially diagnosed at the clinical performance sites were reviewed by a quality control panel of pathologists. Only 43% of the biopsies initially diagnosed as LSIL were classified as LSIL after review, Table 4. Many of the discrepancies were due to the inability of the pathologists to distinguish LSIL from reactive squamous proliferations. Other studies have shown that the reproducibility of a diagnosis of HSIL (CIN2) is also poorly reproducible (Gage et al. 2013; Carreon et al. 2007; Reuschenbach et al. 2014). In one study 2 independent reviewers evaluated a series of 357 previously diagnosed cervical biopsies. When the initial diagnosis had been HSIL (CIN 3) the independent reviewers agreed 84% and 81% of the time, respectively, with the initial diagnosis. However, when the initial diagnosis had been HSIL (CIN 2), they agreed with it only 13% and 31% of the time (Carreon et al. 2007). Given that a HSIL (CIN 2) diagnosis is the diagnostic threshold for treatment of cervical lesions in many countries, this suggests that the morphologic criteria routinely used to distinguish between different types of cervical lesions have serious shortcomings.
Table 4

Variability in histopathological diagnoses in ALTS


Quality control panel review diagnosis

Initial Dx

WNL (%)


LSIL (%)

HSIL (%)


























Modified from reference (Stoler 2001 #4675)

As part of the 2011 the College of American Pathologists (CAP) and the American Society for Colposcopy and Cervical Pathology (ASCCP) sponsored Lower Anogenital Squamous Terminology (LAST) project, a formal literature review was performed of various biomarkers that could assist in the diagnosis of cervical lesions (Darragh et al. 2012). Out of 72 articles that were selected for a full-text review, 53 dealt with the use of p16 immunohistochemistry (IHC). p16ink is a cyclin-dependent protein kinase inhibitor that is overexpressed in almost all HSIL and invasive cervical cancers (Keating et al. 2001; Klaes et al. 2001; Tsoumpou et al. 2009). p16 is particularly attractive as a biomarker for HPV-associated SIL since overexpression within cervical neoplasia has been directly linked to the continued expression of the HPV oncogene E7. Moreover, p16 overexpression appears to be independent of the particular type of high-risk HPV-associated with the lesion. Numerous studies have shown p16 staining to be absent in normal squamous epithelium and benign inflammatory/reparative conditions (Queiroz et al. 2006a; Wang et al. 2004b; Zhang et al. 2007; Tsoumpou et al. 2009). The LAST literature review found that less than 5% of negative biopsies will stain positively for p16 using established criteria for what constitutes “positive” staining (Darragh et al. 2012). In contrast, the LAST literature review concluded that 99% of HSIL (CIN 3) will stain positively. In the recent US FDA trial of a commercially available p16 IHC kit, 7.5% of biopsies with a consensus H&E diagnosis of non-SIL were p16 positive (Stoler et al. 2018). For cases with a H&E consensus diagnosis of LSIL, HSIL (CIN2), and HSIL (CIN 3), the respective rates of p16 positivity were 58%, 100%, and 100%. This suggests that the routine use of p16 IHC would improve the ability of pathologists to differentiate between immature squamous metaplasia or reaction/reparative conditions and HSIL. Unfortunately, p16 IHC is not useful in distinguishing between LSIL and HSIL. Since approximately two-thirds of LSIL is associated with high-risk HPV, it is not surprising that approximately half of LSIL will stain positively for p16, Figs. 25a, b and 27a, b, and another one-third show focal p16 staining (Wang et al. 2004b; Zhang et al. 2007; Hariri and Oster 2007; Stoler et al. 2018; Wright et al. 2012; Stoler 2012 #8258). Therefore, the absence of p16 positivity does not mean that a lesion is not LSIL nor does the presence of p16 positivity indicate that a lesion is HSIL as opposed to LSIL (Fig. 26).
Fig. 25

LSIL of the type referred to by some pathologists as borderline condyloma. (a) Lesions of this type have less nuclear atypia and acanthosis then usually seen in LSIL. This lesion contains considerable numbers of multinucleated cells. (b) It stains positively with p16 and we would classify it as LSIL

Fig. 26

Junction between HSIL and normal squamous epithelium. (a) This section shows the junction between a HSIL and normal squamous epithelium. Both (b) p16 and (c) Ki-67 staining demonstrates how sharp the demarcation usually is between lesional and nonlesional tissue

In order to be classified as showing positive staining for p16, a biopsy needs to show continuous strong nuclear or nuclear plus cytoplasmic staining of the basal cell layer with extension upward to involve at least one third of the epithelial thickness, Fig. 27b. It is important to note that focal or patchy staining for p16 is nonspecific and can be seen in reactive metaplasia as well as in LSIL, Fig. 28a, b. LAST also stressed that the performance of p16 as a biomarker appears to be dependent on a particular monoclonal antibody and that the use of alternative antibodies may not yield the same results. Based on the literature review, LAST recommended the routine use of p16 immunostaining in 4 situations, Table 5. Two large studies have now evaluated the clinical utility of using p16 IHC when interpreting cervical biopsies. One was a European regulatory trial and the other the recent US regulatory trial. In the European study, a number of surgical pathologists were asked to diagnose cervical biopsies that had previously been diagnosed by a panel of 3 expert gynecological pathologists using H&E stained slides (Bergeron et al. 2010). When the surgical pathologists used H&E stained slides they correctly diagnosed 77% of the cases called HSIL by the expert panel. However, when the surgical pathologists evaluated H&E stained slides together with p16 stained slides for all cases, they correctly diagnosed 87% of the HSIL cases and had only a 1% loss in specificity. The US regulatory trial had a similar study design to the European trial. When the surgical pathologists used H&E stained slides alone, they correctly diagnosed 84% of the cases called HSIL by the expert panel. When they used both H&E and p16 stained slides for all cases, they correctly diagnosed 90% of the experts’ HSIL cases with only a small loss of specificity (Stoler et al. 2018).
Fig. 27

p16 and Ki-67 staining of LSIL. Approximately 50–60% of LSIL will show strong diffuse staining of the lower portion of the epithelium with p16. Most LSIL will have an increased Ki-67 labeling index. (a) H&E, (b) p16, (c) Ki-67 staining

Fig. 28

Focal or patchy p16 staining. (a) There is focal staining of the basal/parabasal cell layers, but the staining is not contiguous. (b) There is focal staining of the intermediate and superficial layers of the epithelium, but the basal/parabasal cell layers are not stained. Biopsies with these staining patterns are not classified as p16 positive

Table 5

LAST recommendations for the use of p16 immunostaining (Darragh et al. 2012)



1. p16 IHC is recommended when the H&E differential diagnosis is between HSIL and a mimic of HSIL (such as immature squamous metaplasia, atrophy, reactive epithelial changes, tangential cutting)

Positive p16 staining supports a diagnosis of HSIL

2. Whenever a pathologist is entertaining a H&E diagnosis of HSIL (CIN 2), p16 IHC is recommended to help clarify the diagnosis

Positive p16 staining supports a diagnosis of HSIL. Negative staining supports a diagnosis of LSIL or a non-HPV associated lesion

3. p16 IHC is recommended as an adjudication tool for cases in which there is professional disagreement in histologic specimen interpretation

With the caveat that the disagreement includes HSIL

4. It is recommended against the use of p16 IHC as a routine adjunct to histological assessment of biopsy specimens with morphologic interpretations of normal, LSIL, and HSIL (CIN 3)


LAST also reviewed the clinical utility of other potential biomarkers including IHC with Ki-67 and a less widely used biomarker ProEx C. They concluded that there was insufficient evidence to recommend the routine use of these markers when diagnosing cervical tissue samples (Darragh et al. 2012). When used in combination with p16 IHC, the overall improvement in performance obtained by combining additional biomarkers was minimal compared to p16 alone. Nevertheless, many pathologists use Ki-67 IHC when interpreting difficult cervical biopsies. Ki-67 is a marker of cellular proliferation which shows staining in the upper two-thirds of the epithelium in SIL, whereas normal squamous epithelium typically shows only limited staining in the parabasal cell layers, Figs. 26c and 27c (Queiroz et al. 2006b; Isacson et al. 1996). However, unlike p16, Ki-67 staining may be positive in HPV-negative squamous metaplasia and reactive/reparative conditions which reduces its usefulness in differentiating these conditions from SIL.

Overdiagnosis of LSIL can be reduced if it is remembered that significant nuclear atypia is the hallmark of SIL. Correlation of HPV DNA with specific cytologic/histologic findings has uniformly found that perinuclear halos in the absence of significant nuclear atypia are nonspecific features (Franquemont et al. 1989; Mittal et al. 1990; Ward et al. 1990). Therefore, the diagnosis of LSIL should be made only when significant nuclear atypia accompanies perinuclear halos and the indiscriminate use of the term “koilocytosis” whenever the cervical squamous epithelium demonstrates the slightest hint of “cytoplasmic vacuolization” should be discouraged. Cytoplasmic vacuolization in the absence of nuclear atypia is a nonspecific change that may occur as a reflection of atrophy-related vacuolar degeneration or with prominent glycogen vacuolization of the normal squamous epithelium, Fig. 29. It can also be seen in non-HPV-related infections, such as trichomoniasis, G. vaginalis, and candidiasis, Fig. 30. In contrast to the focal distribution of koilocytes in LSIL, cells of normal squamous epithelium that have perinuclear clearing are not sharply demarcated, the nuclei are not enlarged or atypical, and multinucleated cells are infrequent. In addition to the absence of nuclear atypia, normal stratification and maturation are maintained in such conditions, whereas in HPV-associated lesions, there is some degree of cellular disorganization, particularly near the surface, and there is disturbance in the normal pattern of maturation.
Fig. 29

Atrophy-related perinuclear halos. (a) Biopsy from a postmenopausal patient. The squamous epithelial cells show prominent perinuclear halos, but do not have significant nuclear atypia. (b) This lesion does not stain for p16

Fig. 30

Pseudokoilocytosis secondary to infection. (a) Infection-related perinuclear halos are frequently observed in women with non-HPV-related infections. Generally in infection considerable numbers of inflammatory cells are present. (b) At higher magnification the nuclei are not as enlarged or atypical as usually seen in LSIL and multinucleated cells are infrequent

Overdiagnosis of LSIL as HSIL can be avoided by recognizing that LSIL generally does not have a high mitotic index, loss of cell polarity, and abnormal mitotic figures. Although a diagnosis of LSIL requires significant atypia, the atypia does not usually involve the basal or parabasal cell layers. The classification of lesions that have the characteristic histologic features of LSIL, but also have abnormal mitotic figures (AMFs), is controversial. Abnormal mitotic figures generally indicate that a lesion is aneuploid, Fig. 31. Therefore, an argument could be made that lesions with abnormal mitotic figures should be classified as HSIL, although in the absence of significant basal or parabasal atypia many pathologists would classify low-grade appearing lesions with only one or two abnormal mitotic figures as LSIL. It is also important to note that many HSIL have significant koilocytotic change in the upper layers of the epithelium, so simply having koilocytosis and differentiation in the upper-half of the epithelium should not result in a lesion being classified as LSIL if it has significant basal-parabasal nuclear atypia, Fig. 32. Figure 33 is a lesion that might be classified as LSIL by some pathologists. However, the lesion has significant basal and parabasal nuclear atypia as well as a mitosis in the mid portion of the epithelium and an AMF. Therefore, we would classify this lesion as a HSIL.
Fig. 31

Abnormal mitotic figures in SIL. Abnormal mitotic figures indicate that a lesion is aneuploid and are typically found in HSIL

Fig. 32

HSIL mimicking a LSIL. This lesion has marked koilocytosis in the upper half of the epithelium, but has significant atypia of the basal, parabasal layers and abnormal mitotic figures. Because of the atypia in the basal and parabasal layers, this should be classified as HSIL

Fig. 33

LSIL versus HSIL. Sometimes it is difficult to differentiate between LSIL and HSIL. This lesion has many features of LSIL, but also has a mitosis in the middle of the epithelium and an abnormal mitotic figure and based on this it should be classified as HSIL

High-Grade Squamous Intraepithelial Lesion

In HSIL, atypia should be present in all layers of the squamous epithelium, but to an extent and degree that exceeds what is seen in LSIL. There is significant basal-parabasal nuclear atypia and AMFs. Immature basal-type cells typically occupy more than the lower third of the epithelium, Fig. 34. In addition there is nuclear crowding, pleomorphism and loss of the normal cell polarity. The nuclei of the immature basal-type cells are enlarged when compared to the nuclei of cells at comparable levels of the normal epithelium. This nuclear enlargement is frequently most pronounced in the lower half of the epithelium although in all cases the superficial cells demonstrate some degree of nuclear enlargement. As in LSIL, the nuclei are hyperchromatic and the chromatin is finely to coarsely granular, Fig. 35. Prominent nuclei or chromocenters are uncommon. Normal and abnormal mitotic figures are present and mitoses are usually present in the upper half of the epithelium. Cytoplasm is usually scant resulting in an increase in the nuclear: cytoplasmic ratio. Cell borders between the primitive cells are usually indistinct. The cells overlying the basal-type cells also have atypical nuclei but have more cytoplasm and therefore lower nuclear: cytoplasmic ratios, more distinct cell boundaries and can have prominent HPV cytopathic effects including perinuclear halos and bi- or multinucleation. In the superficial layers of the epithelium, individual dyskeratotic cells may be seen, Fig. 36. These cells are small with pyknotic hyperchromatic nuclei and dense acidophilic cytoplasm. Another characteristic feature of HSIL is the variability in nuclear size (anisonucleosis), Fig. 37. It should be stressed, however, that this is a variable histologic feature. In some HSIL lesions, particularly those that were previously termed carcinoma in situ, the nuclei at first glance appear relatively uniform in size, although careful scrutiny will reveal some variation in nuclear size and shape, Fig. 38. Many HSIL lesions also have koilocytotic features in the upper layers of the epithelium similar to what is seen in LSIL, Fig. 39. However, these koilocytotic features are often associated with bizarre nuclei and abnormal keratinization. These lesions were previously classified as CIN 2.
Fig. 34

HSIL. Undifferentiated neoplastic cells replace 50–70% of the epithelium. The nuclear: cytoplasmic ratio is high, and the cytoplasmic membranes and the basal layer are indistinct

Fig. 35

HSIL. This is a typical HSIL with numerous mitoses in the middle portion of the epithelium and immature basaloid cells extending almost to the surface

Fig. 36

HSIL with dyskeratotic cells. This HSIL shows both parakeratosis and dyskeratotic cells (small dark red cells with pyknotic nuclei) in the upper half of the epithelium

Fig. 37

HSIL with marked variability in nuclear size (anisonucleolis). Anisonucleolis is a variable feature of HSIL

Fig. 38

HSIL. The full thickness of the epithelium is composed of small, undifferentiated neoplastic cells. This is the classic small cell carcinoma in situ. Note: numerous mitotic figures, loss of cellular maturation and organization, and lack of koilocytes

Fig. 39

HSIL with marked koilocytosis. In addition to having immature basaloid cells extending to the middle of the epithelium, this lesion has considerable koilocytotic features. Lesions with this histology were previously classified as CIN 2

The original CIN terminology subdivided high-grade CIN into two categories CIN 2 and CIN 3, and the newer histology terminology based on The Bethesda System allows a distinction to be made between HSIL (CIN 2) and HSIL (CIN 3). This distinction is made based on the proportion of the epithelial thickness occupied by undifferentiated neoplastic cells. In HSIL (CIN2), the immature basaloid-type cells occupy up to two-thirds of the epithelial thickness, but do not extend into the upper third of the epithelium, Fig. 39. Similarly, mitoses are found in the lower two-thirds of the epithelium, but not in the upper third. In HSIL (CIN 3) lesions, immature basaloid-type cells occupy the upper third of the epithelium and mitoses can be present at any level, Fig. 40. Studies of the reproducibility of the histopathological diagnosis of different grades of cervical cancer precursors have shown that diagnosis of HSIL (CIN 2) is not reproducible (Ismail et al. 1989; Robertson et al. 1989; Castle et al. 2007). The lack of reproducibility of the diagnosis of HSIL (CIN 2) is due to the subjective criteria used to separate the different grades of SIL and the fact that the thickness of the epithelium occupied by immature basaloid-type cells varies considerably, even in cervical biopsy specimens.
Fig. 40

HSIL. This form of HSIL has immature basaloid cells and mitoses extending almost to the surface. Lesions with this histology were previously classified as CIN 3

In most tissues, aneuploidy is a marker of malignant potential and multiple studies have shown that the majority of HSIL lesions are aneuploid (Bollmann et al. 2003, 2005; Bocking and Nguyen 2004). A number of studies have compared histologic features with ploidy levels and found that cervical lesions with diploid or polyploid DNA contents generally retain polarity of the basal cell layer and lack abnormal mitotic figures, whereas aneuploid lesions have more marked nuclear atypia and more cellular disorganization. The best histologic correlate of aneuploid is abnormal mitotic figures (AMF) (Bergeron et al. 1987; Fu et al. 1988). Therefore, some authors have suggested that AMFs other than multipolar and dispersed metaphases are an accurate histologic surrogate of aneuploidy and can be used as a histologic determinate for discriminating between LSIL and HSIL (Richart 1990). Although AMFs are commonly seen in HSIL, Fig. 40, AMFs should not be used as the sole criterion for discriminating between LSIL and HSIL for several reasons: (1) AMFs can be difficult to distinguish from karyorrhexis, (2) detection of AMFs is influenced by variables that are independent of ploidy level, including size of biopsy, quality of fixation, quality of the microscopic section, and number of levels examined, and (3) some HSIL lesions, and even some invasive cancers, lack AMFs and are not aneuploid (Hanselaar et al. 1988; Mourits et al. 1992). Therefore, although a lesion showing the above features with unequivocal AMFs should be classified as HSIL, the converse is not true. Lesions with the other histological features of a HSIL should be classified as such even in the absence of AMFs. It is also important to point out that using these criteria, some HSIL will have cells with prominent HPV cytopathic effects similar to those seen in LSIL, Figs. 32 and 33. The presence of such cells should not be taken as evidence that the lesion is low grade if other features of a HSIL are present. Therefore, the criteria that are used for distinguishing LSIL from HSIL include other features such as the distribution of immature, basal-type cells, the level at which mitotic figures in the epithelium are identified, the extent of abnormalities of differentiation and polarity, and the degree of nuclear atypia, Table 6. It should also be stressed that 50–60% of LSIL is p16 IHC positive and therefore p16 positivity does not differentiate between LSIL and HSIL. Therefore, the LAST recommendation that whenever a pathologist is entertaining a H&E diagnosis of HSIL (CIN 2), they should perform p16 IHC to help clarify the diagnosis and that positive p16 staining supports a diagnosis of HSIL is problematic (Darragh et al. 2012). It is likely that different pathologists will have different thresholds for determining which lesions are borderline for HSIL (CIN 2) and will therefore order p16 IHC differently. Therefore, it is reassuring that in the US p16 regulatory trial that the use of p16 adjunctive staining did not increase the overall number of HSIL (CIN 2) diagnoses made (Stoler et al. 2018).
Table 6

Distinguishing features of LSIL and HSIL


Lower third

Upper two-thirds

HPV types

Any anogenital HPV

High-risk typesa


Frequently present

Occasionally present


Mostly diploid or polyploid

Usually aneuploid

Abnormal mitotic figures



Location of undifferentiated cells and mitotic figures

Lower third

Upper two-thirds

aHigh risk types of HPV include 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68

It is important to recognize that HSIL has quite a variable histological presentation, especially with respect to size of the cells, extent of koilocytosis, degree of atypia, and degree of keratinization. The most common different presentations include a mature or koilocytotic form which is characterized by prominent HPV features (i.e., koilocytosis) accompanied by atypia of the basal and parabasal cell layers together with mitotic figures, including abnormal forms, in the upper two-thirds of the epithelium. Another commonly encountered presentation is keratinizing HSIL that has quite prominent superficial keratinization accompanied by cells with the typical histologic and cytologic features of HSIL, Fig. 41. Other HSIL show less keratinization but have a prominent layer of parakeratotic cells, Fig. 42. Perhaps the most difficult forms of HSIL to recognize are those that have an immature metaplastic phenotype. These lesions resemble immature squamous metaplasia and are sometimes encountered underneath an intact columnar epithelium, Fig. 43. Compared to immature squamous metaplasia, HSIL with an immature metaplastic phenotype shows a greater degree of hyperchromasia and anisonucleosis. Nuclear density does not decrease in the superficial cell layers and a syncytium of nuclei can sometimes be seen in the superficial layers. p16 is very useful in diagnosing these lesions, Fig. 44a, b. In the past, some investigators have proposed formally subtyping high-grade precursors into small cell anaplastic, large cell keratinizing, and large cell nonkeratinizing dysplasia (Patten 1978; Ratnam et al. 2000). Because accurate studies concerning the invasive potential of each of these subtypes are lacking, prediction of the likelihood of progression to invasive carcinoma should not be based on the above subclassification.
Fig. 41

HSIL with hyperkeratosis and parakeratosis. Prominent keratinization is frequently seen in HSIL. This HSIL has a thick plaque of keratin on the surface which can often be seen at the time of a gynecological examination. Clinicians call plaques of keratin on the surface of the cervix leukoplakia. Parakeratosis is also present

Fig. 42

HSIL with prominent abnormal parakeratosis. This HSIL has a superficial layer of compacted parakeratotic cells that appear abnormally keratinized

Fig. 43

HSIL with metaplastic features. This HSIL has a metaplastic appearance and is covered by columnar epithelium. The degree of nuclear atypia is too great for this to be metaplasia

Fig. 44

HSIL with metaplastic features. (a) This lesion is relatively bland appearing, but has considerable numbers of mitotic figures. (b) p16 staining demonstrates strong diffuse positivity

Differential Diagnosis

Immature metaplasia, reactive/reparative processes, and atrophy are the most common lesions mistaken for HSIL. This is because cells of these lesions can show immaturity of the squamous epithelium, nuclear atypia, and inflammatory cellular changes. In immature metaplasia, the full thickness of the epithelium is composed of immature parabasal cells with a high nuclear: cytoplasmic ratio, Fig. 45. The cells usually are vertical and the nuclei are only slightly hyperchromatic. The most helpful feature in distinguishing HSIL with an immature metaplastic phenotype from immature metaplasia is the absence of nuclear pleomorphism in the latter. The chromatin in metaplastic squamous epithelium is finer and more evenly distributed than in HSIL. In addition, cellular polarity is retained, cell membranes are clearly defined, and cellular crowding is not marked. Immature metaplasia typically shows a regularity of nuclear spacing and the lesions lack significant variation in nuclear size and staining pattern. Atypical cells are rarely seen in the superficial layers of the epithelium and the superficial layers of the epithelium often appear more normal than the lower half. Mucinous epithelium is often present on the surface of immature metaplastic squamous epithelium but only occasionally overlies HSIL, Fig. 46. Immature metaplasia may have mitotic activity, but abnormal mitotic figures are not present. When significant numbers of mitoses are present it is more likely that the lesion represents a HSIL, Fig. 47. Sometimes there may be more pronounced nuclear atypia within immature metaplasia, Fig. 48a, b. Such lesions are designated by some as atypical immature metaplasia (AIM) (Crum et al. 1983). Histologically, AIM lesions have a monomorphic population of squamous cells that retain their cellular polarity. They are more cellular than the usual immature squamous metaplasia but show a low mitotic activity, usually less than one mitoses per 10 high-powered fields and very few cells are Ki-67 positive (Fu and Reagan 2002). The use of the term AIM is not generally recommended unless clarified with a note because it is not widely accepted and may be confusing to clinicians. Moreover, an AIM diagnosis is poorly reproducible and many of the lesions diagnosed as AIM will be accompanied by a concurrent or subsequent diagnosis of HSIL (Park et al. 1999). In one study 80% of high-risk HPV positive patients with AIM lesions had a follow-up or concurrent diagnosis of HSIL (Geng et al. 1999), suggesting that the lesions classified as AIM were very likely subtle forms of HSIL.
Fig. 45

Immature squamous metaplasia. Metaplastic squamous cells are regularly oriented with uniformly disposed nuclear chromatin and cellular borders. Intracellular bridges are present

Fig. 46

Immature squamous metaplasia. This lesion has a relatively uniform metaplastic appearance, but the nuclei are hyperchromatic and irregular. Some mitoses are present. However, the lesion is found underlying columnar epithelium and the degree of nuclear atypia is insufficient to warrant a diagnosis of HSIL

Fig. 47

HSIL This lesion has a relatively uniform metaplastic appearance, but has considerable numbers of mitoses. Except for the mitoses this lesion is similar to the lesion shown in Fig. 46. However, this lesion showed strong diffuse p16 staining throughout its full thickness (not shown) and was associated with classic HSIL in other sections. Immunohistochemical staining for p16 is essential to classify lesions of this type

Fig. 48

Atypical Immature Metaplasia (AIM). This immature squamous metaplastic lesion has more pronounced nuclear atypia than usually seen in metaplasia. Lesions with this histology are referred to by some as AIM lesions. (b) This particular lesion has negative/patchy p16 staining

Reparative processes are also sometimes very difficult to differentiate from HSIL, Fig. 49b. In reparative processes, atypical basal cells occupy the lower half of the epithelium, but the cells have a regular nuclear outline and prominent nucleoli and usually have distinct cell membranes. In addition, dense acute and chronic inflammatory infiltrates are usually present. In reactive/reparative processes, there is usually intracellular edema which leads to spongiosis of the epithelium. Multinucleated cells are generally not observed and the superficial epithelial cells lack the marked variability in nuclear size, shape, and density that characterize HSIL.
Fig. 49

Reparative process mimicking HSIL. This reparative process shows more nuclear variability than usually seen in repair. However, there is considerable inflammation present

Atrophy can have a variety of histological appearances and is occasionally difficult to distinguish from HSIL because it is composed of basal and parabasal cells showing no differentiation. In most instances the atrophic cells are immature but they are quite bland appearing. Although there is a high nuclear: cytoplasmic ratio, atrophic epithelium is thin and shows no nuclear pleomorphism, mitotic activity, atypia, or lack of polarity. These cases rarely are mistaken for HSIL, provided the patient’s age is known. However, in other instances atrophy results in disturbances in cellular maturation with pseudokoilocytotic cells and cells with nuclear enlargement and nuclear hyperchromasia, Fig. 50. Finally, HSIL, particularly with extensive gland involvement, may be confused with microinvasive carcinoma (see chapter “Carcinoma and Other Tumors of the Cervix”), Fig. 51.
Fig. 50

Atrophy mimicking HSIL. (a) In severe atrophy, there can be considerable variation in nuclear size and irregularity oriented cells and mimic HSIL. (b) Atrophy has a very low Ki-67 staining index and (c) does not stain with p16

Fig. 51

HSIL extending into endocervical glands. When HSIL extends into endocervical crypts or glands, it can sometimes be misinterpreted as an invasive lesion

In our experience, it is sometimes simply impossible to distinguish immature metaplasia, atypical immature metaplasia (AIM), reactive/reparative changes, or atrophy from HSIL based on histopathology alone. In these cases, immunohistochemical staining is essential in differentiating HSIL from the other conditions. P16 is overexpressed in almost all HSIL and invasive cervical cancers and almost all HSIL show strong, diffuse staining of the epithelium, Fig. 52 (Keating et al. 2001; Klaes et al. 2001; Stoler et al. 2018). Both reparative/metaplastic processes and atrophic epithelium are rarely p16 positive, Figs. 48b and 50c. Atrophic epithelium shows either absent or minimal staining with Ki-67, whereas HSIL shows considerable numbers of labeled cells in all layers of the epithelium, Fig 50b.
Fig. 52

p16 immunostaining of HSIL. HSIL typically shows strong, diffuse staining with p16

Behavior of SIL

Before discussing the behavior of SIL, it is important to emphasize that most natural history studies were conducted in the 1970s and 1980s prior to our enhanced understanding of the role of HPV in the pathogenesis of cervical cancer and the realization that LSIL is simply a cytological and histological marker for a productive HPV infection. Moreover, over the last decade we have come to realize that not only is the interpretation of cervical biopsies and cytology subjective and prone to considerable error, but also that colposcopy is much less accurate than previously thought (Massad 2006; Waxman et al. 2017; Pretorius et al. 2011). Nevertheless, the older natural history studies remain important because they provide insight into both the likelihood that a woman with SIL will spontaneously clear her lesion and the likelihood that she will develop an invasive cervical cancer. Since these studies used the dysplasia/CIS or CIN terminology, in this section those terms will be retained in order to accurately describe the findings from these studies. Two approaches have been used to determine the natural history of SIL. These are prospective, clinical follow-up studies of individual women with cervical lesions and epidemiological studies linking cytology records with cancer registries. Prospective clinical follow-up studies of the “natural history” or behavior of SIL have provided widely varying estimates of the rate of progression and regression the different lesions. This is not surprising since various studies have used different entry criteria, different diagnostic criteria for categorizing lesions as SIL, and different study designs. For example, some studies have used punch biopsy and endocervical curettage to establish the diagnosis. These diagnostic methods may remove (treat) lesions and therefore may interfere with long-term analysis by increasing the frequency of spontaneous regression and decreasing the frequency of progression (Nasiell et al. 1983).

Table 7 provides a summary of a meta-analysis of clinical follow-up studies of biopsy-confirmed CIN published through the mid-1990s (Mitchell et al. 1996). The higher the grade of a lesion, the more likely it is to persist and the less likely it is to regress. Overall it appears that approximately 57% of CIN 1 lesions spontaneously regress in the absence of therapy and 11% progress to carcinoma in-situ. The rates of persistence and progression are greater for high-grade CIN. Forty-three percent of CIN 2 lesions regress and 22% progress to carcinoma in situ. The equivalent rates for CIN 3 lesions were 32% regression and 12% progression to carcinoma in situ. Overall, the progression of all grades of CIN to invasive cancer in the published observational studies is 1.7%. More recently, Tainio et al. published a meta-analysis of the rate of spontaneous regression of biopsy-confirmed CIN 2 and found that after 24 months of follow-up 50% of the lesions had regressed, which is almost identical to what was found by the meta-analysis of older studies (Tainio et al. 2018).
Table 7

Natural history of CIN is dependent of lesional grade


% Regression

% Persist

Progress to CIS













Source: Mitchell et al. (1996)

In 1998 Melnikow et al. performed a meta-analysis of studies in which women with a cytologic result of SIL were followed (Melnikow et al. 1998). The analysis included 13,226 women with a cytologic result of LSIL who were followed for at least 6 months and had a median weighted follow-up of 29 months. There were also 10,026 women with HSIL who had a median weighted follow-up of 25 months. The pooled estimates for regression to normal were 47% for LSIL and 35% for HSIL, Fig. 53. No evidence of a relationship between the proportion of subjects regressing to normal and the length of follow-up was observed. Rates of progression of LSIL were 7% at 6 months and 21% at 24 months. For HSIL the 6 and 24 month pooled progression rates were 7% and 24%. The pooled progression rates for invasive cancer at 6 and 24 months for LSIL were 0.04% and 0.15%, respectively. For HSIL they were 0.15% at 6 months and 1.44% at 24 months. A much higher rate of regression of cytologically diagnosed LSIL has been reported by Moscicki et al. who prospectively followed a cohort of young women with LSIL. After 12 months of follow-up, 61% had regressed and by 36 months of follow-up 91% had regressed (Moscicki et al. 2004).
Fig. 53

Pooled estimates of the rates of spontaneous regression (left panel) of a cytologic diagnosis of LSIL and HSIL. Pooled estimates of the rates of spontaneous progression of LSIL and HSIL after 6 months (circles) and 12 months (squares). Progression of HSIL is from CIN 2 to CIN 3 or carcinoma in situ. Bars represent 95% confidence intervals. (Modified from reference Melnikow et al. 1998)

Another study evaluated the records of the largest cytology laboratory serving the Toronto (Canada) region and linked these records with the Ontario Tumor Registry for the years of 1962–1980 (Holowaty et al. 1999). During this period of time, most women with dysplasia who were evaluated by this laboratory were managed conservatively and did not undergo treatment. This study provides an unique insight into the long-term natural history of untreated SIL, Table 8. The key findings of this study were that after 10 years of follow-up only 12% of untreated mild dysplasia and 17% of untreated moderate dysplasia were diagnosed with carcinoma in situ. At 10 years 88% of the mild dysplasia and 83% of cases of moderate dysplasia had regressed (Holowaty et al. 1999).
Table 8

Toronto long-term follow-up of abnormal cervical cytology

Grade of lesion

2 years (%)

10 years (%)

Regression ratesa

Mild dysplasia



Moderate dysplasia



Progression ratesb

Mild dysplasia



Moderate dysplasia



Severe dysplasia



aRegression to within normal limits

bProgression to carcinoma in situ or worse. (Modified from reference Holowaty et al. 1999 #4486)

What these studies show is that the vast majority of LSIL spontaneously regress in the absence of treatment and the risk that a woman with a LSIL will be subsequently diagnosed with either carcinoma in situ or invasive cervical cancer is relatively low. They also demonstrate that the likelihood that HSIL will regress in the absence of treatment is higher than many clinicians realize and that it generally takes many years for a HSIL to progress to an invasive carcinoma.


Colposcopy combined with colposcopically directed cervical biopsies are the primary modality by which women with abnormal cervical cytology are evaluated. Colposcopic examination consists of viewing the cervix with a long-focal-length, dissecting-type microscope at a magnification of about 16x after a solution of dilute (4%) acetic acid has been applied to the cervix. The acetic acid solution acts to remove and dissolve the cervical mucus and causes SIL to become whiter than the surrounding epithelium (acetowhite), Fig. 54. This coloration allows the colposcopist to identify and biopsy epithelial lesions. In addition to allowing the detection of acetowhite areas, colposcopy also allows for the detection of blood vessel patterns that can indicate HSIL and the detection of invasive cancers. Colposcopy and appropriately directed biopsy have greatly facilitated the management of patients with preinvasive lesions of the cervix because it allows the clinician to rule out invasive cancer and determine the limits of preinvasive disease. Conservative ablative treatment modalities such as cryosurgery, laser ablation, and loop electrosurgical excision procedure (LEEP) can then be used to treat preinvasive disease, with success rates similar to those obtained with cone biopsies.
Fig. 54

Colposcopic appearance of SIL. An acetowhite, well-circumscribed lesion is present at the external os

Precursors of Cervical Adenocarcinoma

Terminology and Historical Perspective

The first indication that precursor lesions for invasive endocervical adenocarcinomas exist came in 1952 when Helper described highly atypical neoplastic cells lining architecturally normal endocervical glands adjacent to frankly invasive endocervical adenocarcinomas (Helper et al. 1952). Shortly thereafter, Friedell and McKay described two patients with atypical glandular lesions of the cervix and designated these lesions adenocarcinoma in situ (AIS) because of their histologic resemblance to invasive endocervical adenocarcinoma (Friedell and McKay 1953). One of these patients had a coexistent invasive adenocarcinoma of the cervix and one, squamous “carcinoma in situ.”

By analogy to squamous cell cervical cancer precursors, some authors have proposed parallel classification schemas for endocervical adenocarcinoma precursors which include lesions with a lesser degree of abnormality than AIS (Bousfield et al. 1980; Brown and Wells 1986; Genest et al. 1993; Luesley et al. 1987). Such low-grade putative glandular precursor lesions were originally termed endocervical dysplasia by Bousfield et al., but other terms such as atypical hyperplasia are also used to refer to lesions which resemble AIS, but have a somewhat lesser degree of nuclear atypia and mitotic activity (Hopkins et al. 1988; Bousfield et al. 1980). Gloor and associates suggested that the term cervical intraepithelial glandular neoplasia (CIGN) be used to refer to both endocervical glandular dysplasia and AIS and that endocervical glandular dysplasia be classified as either CIGN grade 1 or 2 and AIS be classified as CIGN grade 3 (Genest et al. 1993). The term endocervical glandular dysplasia has been used to describe a “glandular lesion characterized by significant nuclear abnormalities that are more striking than those in glandular atypia, but fall short of the criteria for adenocarcinoma in situ” (Tavassoli and Devilee 2003). However, several different diagnostic criteria have been proposed for this entity and there is not widespread acceptance of any of them (Ioffe et al. 2003; Gloor and Hurlimann 1986a; Jaworski 1990). Because of the relative rarity of endocervical glandular dysplasia, the subjective nature of the morphologic criteria used to distinguish it from AIS, and the infrequent coexistence of endocervical glandular dysplasia with AIS, the significance of endocervical glandular dysplasia is not known and a number of authors question that it even exists as a discrete entity (Goldstein et al. 1998; Lee 2003; Loureiro and Oliva 2014). A number of studies have utilized objective biomarkers including HPV DNA positivity, proliferation markers such as Ki-67/MIB-1, p16, and selective mucin staining to determine whether endocervical glandular dysplasia acts as a precursor to AIS or invasive endocervical adenocarcinoma (Anciaux et al. 1997; Lee et al. 2000; Gloor and Hurlimann 1986b; Riethdorf et al. 2002; Tase et al. 1989a; Baker et al. 2006; Murphy et al. 2004; Leary et al. 1991; Higgins et al. 1992). Although some of these studies have shown a similar pattern of biomarker expression in both endocervical glandular dysplasia and AIS, most studies have not (Lee 2003; Ioffe et al. 2003; Goldstein et al. 1998; Riethdorf et al. 2002). Investigators who advocate the use of endocervical glandular dysplasia do so under the misguided notion that there is a similar relationship of glandular precursors to HPV infection as there is for squamous precursor lesions. As previously discussed the histologic manifestation of productive viral infection is LSIL. Productive HPV infection is tightly linked to squamous differentiation. Glandular epithelium does not support productive infection. Accordingly, there is no comparable low-grade lesion in glandular epithelium. Since the term glandular dysplasia implies a relationship to AIS and invasive carcinoma which does not exist, the use of this term should be discontinued. Instead atypical glandular proliferations that fall short of AIS should be evaluated using biomarkers such as p16 and Ki-67 and classified as reparative changes if they are p16 negative and show a low Ki-67 labeling index. In contrast, if they express p16 strongly and diffusely and show a high Ki-67 proliferation index, they are classified as AIS. The 2014 World Health Organization classification discusses the term endocervical glandular dysplasia which it states is synonymous with low-grade CIGNA but further comments that it is a poorly reproducible diagnosis for which criteria are not well defined (Kurman et al. 2014). It also recommends that IHC be performed and if the profile matches that AIS that the lesion be classified as AIS for management purposes.

Epidemiology and Etiology

Over the last three decades, endocervical glandular lesions have received increasing attention. This is attributable to a variety of factors. One is a perception that the prevalence of adenocarcinomas of the cervix and its precursor lesions is increasing. There has been a documented absolute increase in the prevalence of invasive adenocarcinomas in specific groups of women in both the United States and Europe. This may be due, in part, to the routine use of cytobrushes and HPV testing in screening and the widespread adoption of excisional methods for treating SIL such as the loop electrosurgical excision procedure (LEEP) which permit pathological examination of the entire transformation zone. In addition, there is an increased awareness of these lesions by pathologists and an awareness by colposcopists that certain types of glandular lesions are difficult to recognize colposcopically.

The prevalence of AIS is not well understood, but it is considerably less common than SIL. In most registry series, the ratio between AIS and HSIL has ranged from 1:26 to 1:237. Estimates for the United States as a whole are available in the Surveillance Epidemiology and End Results (SEER) public database that contains data from patients entered into the database between 1976 and 1995, which is the last year that in-situ carcinomas were reported to the database (SEER 2001; Wang et al. 2004a). In the SEER registry, there were a total of 149,178 women with either in situ or invasive cervical cancer. Of these, 96% had squamous lesions and 4% had glandular lesions. 121,793 (82%) of all cervical lesions were classified as in situ and of these, 120,317 (99%) were squamous cell carcinoma in situ, and only 1,476 (1%) were adenocarcinoma in situ. For comparison, of the 27,385 women with invasive cervical cancer, 4,369 (16%) had invasive adenocarcinoma. In 1991–1995 the age-adjusted incidence rate in the SEER database for squamous carcinoma in situ in White women was 27.93 cases per 100,000, whereas the age-adjusted incidence rate for AIS was only 1.25 per 100,000 (Plaxe and Saltzstein 1999; Wang et al. 2004a). Although the overall incidence of AIS remains quite low, the incidence increased approximately six fold in the USA from the 1970s to the 1990s (Wang et al. 2004a). Because AIS is no longer reported, we do not know if the incidence has continued to increase over the last two decades. However, a recent registry study from the Netherlands has reported a significant increase in the incidence of AIS between 2004 and 2013 (van der Horst et al. 2017). The increase in incidence occurred predominantly in women 25–39 years of age and 59% of the cases had a concomitant SIL. Recently, baseline data from two large US cervical cancer screening studies have been published (Wright et al. 2012; Stoler et al. 2018). Both studies referred all women 25 years and older who were HPV or cytology positive to colposcopy and required that a cervical biopsy and/or endocervical curettage be obtained. In one study 16 cases of adjudicated AIS were detected among the 42,695 subjects (37.5 cases per 100,000 women); in the other study 9 cases were detected among the 28,110 subjects (32.0 per 100,000 women). For comparison, the rate of adjudicated HSIL (CIN 3) in the two studies was 594.9 per 100,000 and 569.2 per 100,000, respectively, which gives an AIS to HSIL (CIN 3) ratio in the range of 1:15. The rates of AIS in these two contemporary studies are an order of magnitude higher than rates reported in the 1990s from SEER which is not unexpected given the perception of many pathologists that they have been diagnosing increasing numbers of AIS over the last two decades. Moreover, these two studies are unique in that they referred all HPV-positive women to colposcopy and included histologic sampling in all women.

The age relationship between AIS and invasive adenocarcinoma is similar to that of HSIL and invasive squamous cell carcinoma, suggesting that AIS is a precursor lesion (Plaxe and Saltzstein 1999). AIS rates peak at age 35–44 years in both White and Black women (Wang et al. 2004a). However, unlike squamous lesions of the cervix in which high-grade precursors occur more frequently than invasive cancer, exactly the opposite relationship exists for AIS and invasive adenocarcinoma of the cervix. The incidence of invasive glandular lesions is higher than that of noninvasive glandular lesions in all age groups (Plaxe and Saltzstein 1999). A number of reasons have been proposed for this apparent discrepancy including the fact that AIS is more difficult to detect both cytologically and colposcopically than is SIL and, therefore, might not be detected prior to the development of invasive adenocarcinoma. Additional support implicating AIS as a precursor of invasive adenocarcinoma comes from several anecdotal case reports and two small series of patients who had cytologic or histologic evidence of AIS several years prior to the detection of invasive adenocarcinoma (Boddington et al. 1976; Boon et al. 1981; Kashimura et al. 1990). Although these studies have been interpreted as indicating that AIS is a precursor lesion, it is conceivable that an unrecognized invasive cancer was present at the time of the original Pap test or cervical biopsy.

The reported proportion of AIS that occurs in association with SIL ranges from 24% to 90% (Andersen and Arffmann 1989; Colgan and Lickrish 1990; Denehy et al. 1997; Ault et al. 2011). In the Gardasil HPV vaccine trials, 20 of 22 women diagnosed with AIS had concurrent squamous lesions (Ault et al. 2011). In the two large European multicenter studies of tissue samples from women with cervical cancer precursors, 46 cases of AIS were diagnosed, of which 28 (57%) had coexistent SIL (Holl et al. 2015). This suggests that both AIS and SIL share a similar etiology. Moreover, many of the risk factors are similar between glandular and squamous lesions. These include multiple sex partners, use of oral contraceptives, early onset of sexual activity, and low socioeconomic class (Ursin et al. 1996). Both squamous and glandular precursor lesions also appear to be associated with high-risk types of HPV. Using in situ hybridization, Tase and coworkers examined 8 cases of AIS for the presence of HPV DNA and found that 5 of the cases contained HPV but that, unlike SIL lesions analyzed with the same method, the majority of AIS was associated with HPV 18 as opposed to HPV 16 in SIL (Tase et al. 1989b). Since this initial report, other groups have analyzed AIS for the presence of HPV DNA and almost all AIS is associated with HPV DNA with HPV 16 and 18 being the most commonly encountered types. Duggan analyzed a series of 37 cases of AIS using PCR and identified HPV 18 in 43% of cases and HPV 16 in 23% (Duggan et al. 1994). HPV DNA positivity was not correlated with any clinical variable in that series. More recently Quint et al. analyzed 33 cases of AIS using a highly sensitive PCR method and identified HPV DNA in all of the cases, Table 9. HPV 18 was found in 27%, HPV 16 in 70%, and HPV 45 in 3% (Quint et al. 2010). Not only did this study find a high prevalence of HPV 16, they also found that the Asian American variant of HPV 16 was preferentially associated with glandular lesions of the cervix including AIS. This particular HPV 16 variant was 4 times more common in AIS than in HSIL (CIN 3). In the US ATHENA trial, 16 cases of histologically confirmed AIS were detected. HPV was identified in the screening cytology in 88% of the cases with 38% being HPV 16 positive and 50% positive for HPV 18 (Wright et al. 2012). In the two European multicenter studies of cervical cancer precursors, 94% of the AIS cases were HPV positive with all of the cases having either HPV 16 (57%) or HPV 18 (38%) (Holl et al. 2015).
Table 9

HPV genotypes in AIS


No cases

HPV 16 (%)

HPV 18 (%)

HPV other (%)

HPV negative (%)

Quint et al. (2010)






Ault et al. (2011)






Wright et al. (2012)






Holl et al. (2015)






Clinical Features

The average of women with AIS is 37 years based on a meta-analysis of multiple studies that included 1278 women undergoing treatment (Salani et al. 2009). Most women with AIS are asymptomatic and the lesions are detected either during screening or fortuitously on an endocervical curettage, cervical punch biopsy, cone or loop excisional biopsy performed during the workup for SIL. In women who are symptomatic, the most common complaint is abnormal vaginal bleeding, either postcoital, postmenopausal, or out of phase. Rarely, symptomatic patients present with an abnormal discharge. AIS is difficult to detect both cytologically and colposcopically (Duska 2009; Renshaw et al. 2004). In one early study of 42 women with histologically confirmed AIS, only 45% of the women had atypical glandular cells detected on the preceding cervical cytology (Denehy et al. 1997). The other cases were detected fortuitously on biopsies taken to evaluate SIL. A similar sensitivity of about 50% was found in a retrospective review of previous cytology specimens in histologically confirmed cases of AIS reported to the Australian cancer registry (Ruba et al. 2004). In the Gardasil HPV vaccine trials, a total of 22 subjects with AIS were identified and only 2 of the 22 (10%) had a preceding cervical cytology suggestive of glandular neoplasia (Ault et al. 2011). In the largest study to date, Umezawa et al. reviewed 140 prior cervical cytology specimens from 74 women with AIS (Umezawa et al. 2015). The sensitivity of the original cytology was only 45%. After review by 6 independent cytopathologists diagnostic accuracy increased to 79%. Almost identical results were found in an European multicenter treatment trial of AIS in which only 40% of the referral cytology results were consistent with a glandular lesion (Costa et al. 2012). In that trial 24% of the colposcopies in women with AIS were considered negative.

The distribution of AIS in the cervix is important for determining the clinical management. AIS typically involves both the epithelial surface and endocervical glands or crypts, but in some cases it can be limited to the glands and rarely it can be limited to the surface epithelium (Witkiewicz et al. 2005). Cases confined to the surface epithelium present at a younger average age than cases that involve endocervical glands or crypts, and it has been postulated that they may represent the earliest form of AIS (Witkiewicz et al. 2005). AIS confined to the surface can be quite easily overlooked since they are often small and have less pronounced cytologic atypia and mitoses than the AIS that involves endocervical glands or crypts. In 65% of cases, AIS involves the transformation zone (Andersen and Arffmann 1989; Bertrand et al. 1987) and in the majority of cases it is unifocal. However, AIS can extend for a distance of up to 3 cm. into the endocervical canal and occasionally it can be multicentric with skip lesions in the endocervical canal where foci of AIS are separated from each other by normal endocervical tissue (Bertrand et al. 1987; Cullimore et al. 1992).

The identification of AIS in biopsies and on endocervical curettage can be challenging. One of the first large case series of AIS found that despite colposcopy, cervical biopsy, and endocervical, curettage almost a third of the cases were only identified after conization (Andersen and Arffmann 1989). In the recent European multicenter study that evaluated specimens from 3979 women originally diagnosed with high-grade precancer, 65 cases were diagnosed as AIS at the original institution (Holl et al. 2015). However, only 26 (40%) of these cases were confirmed as AIS by expert pathology review. Moreover, an additional 23 cases of AIS that had not been diagnosed at the original institution were identified during expert review.

Pathologic Findings

This section describes the appearance of “usual” type AIS which is the type that is associated with HPV. A more recently recognized variant, gastric type AIS, is not associated with HPV and is discussed later. AIS is characterized by the presence of endocervical glands lined by atypical columnar epithelial cells that cytologically resemble the cells of invasive adenocarcinoma but which occur in the absence of invasion, Fig. 55. These cells have elongated, cigar-shaped, hyperchromatic nuclei with coarse granular chromatin, Fig. 56. The amount of cytoplasm is greatly reduced and there is only minimal intracellular mucin. This produces an increased nuclear: cytoplasmic ratio. The cells are crowded and pseudostratified, forming two or more rows. AIS may involve glands either focally, multifocally, or diffusely. Typically, some glands show an abrupt transition between normal epithelium and AIS, Fig. 57. Mitotic figures including AMFs are common and apoptotic bodies are also commonly seen in the epithelium, usually subadjacent to the nucleus, Fig. 58. Architecturally, the glands of AIS can have numerous outpouchings and complex papillary infoldings and may display a cribriform pattern focally, Fig. 57. However, since the lesion is confined to the preexisting normal endocervical structures, it retains a lobular architecture.
Fig. 55

Adenocarcinoma In situ (AIS). AIS is characterized by endocervical glands lined by atypical columnar cells resembling the cells of adenocarcinoma

Fig. 56

Adenocarcinoma in situ (AIS). The cells of AIS are pseudostratified and have elongated, hyperchromatic nuclei

Fig. 57

Adenocarcinoma in situ (AIS). Frequently sharp transitions between the normal columnar epithelium and the neoplastic glandular epithelium

Fig. 58

Adenocarcinoma in situ (AIS). Mitotic figures and apototic bodies are usually present

Several histological variants of AIS have been described. Ostor et al. identified two histological variants of which one is the typical or “usual” endocervical-type variant of AIS described above (Ostor et al. 1984). The usual endocervical variant can be pure (58% of cases) or admixed with other variants. The other variant has features of an intestinal, as opposed to endocervical, mucosa with goblet cells and sometimes Paneth cells, Fig. 59. The colonic variant is uncommon and usually occurs in association with the usual endocervical AIS variant. The goblet cells in colonic AIS contain O-acetylated sialomucin which is a marker of intestinal differentiation. Some intestinal types of AIS also contain argentaffin and Paneth cells (Jakobsen et al. 1983; Trowell 1985). Endometrioid variants of AIS and more rarely, adenosquamous, and clear cell AIS also occur (Gloor and Hurlimann 1986b; Jaworski et al. 1988).
Fig. 59

Adenocarcinoma in situ (AIS) of the intestinal type. The intestinal type of AIS shows goblet cells and sometimes even Paneth cells

Some precursor lesions have both squamous and mucinous differentiation, Fig. 60a. This is thought to reflect the multipotential nature of the cells in the transformation zone. These uncommon lesions with both squamous and mucinous differentiation have been referred to as stratified mucin-producing intraepithelial lesion (SMILE) and contain cells with prominent mucin droplets (Park et al. 2000). In the largest case series to date, 69 specimens from 60 patients were identified over a 6 year period from 11,520 cervical specimens (biopsies and resections) (Boyle and McCluggage 2015). SMILE usually show a high Ki-67 labeling index and strong diffuse staining with p16, Fig. 60b, c. These lesions are often found in association with both HSIL and adenocarcinoma in situ. In the original description, 9 of 18 cases of SMILE were associated with invasive lesions (Park et al. 2000). In the larger more recent series, SMILE coexisted with HSIL in 93% of cases, AIS in 42%, and invasive carcinoma in 10% (Boyle and McCluggage 2015). Because of their mixed histologic appearance and association with both HSIL and AIS, it is difficult to classified the lesions as either squamous or glandular in origin. These lesions were initially felt to be a variant of AIS arising in reserve cells of the transformation zone and the squamous differentiation was considered a marker of phenotypic instability (Park et al. 2000). The 2014 WHO terminology classifies them as a variant of AIS (Kurman et al. 2014).
Fig. 60

Stratified mucin-producing intraepithelial lesion (SMILE). (a) This variant of AIS shows both mucinous and squamous differentiation. (b) It has a high Ki-67 labeling index and (c) stains positively for p16

Recently, several case reports and one small case series have been published identifying a rare variant of AIS referred to as gastric AIS (gAIS) (Mikami and McCluggage 2013; Talia et al. 2017). This lesion may be a potential precursor to gastric type cervical adenocarcinomas (Mikami and McCluggage 2013). Like cervical gastric type mucinous carcinoma, these lesions are HPV negative and have a gastric phenotype. They tend to occur in older women than the usual AIS with an average age of 53 years (Talia and McCluggage 2018). Gastric AIS typically arises in the transformation zone and can extend into the endocervical canal and even into the endometrium (Talia et al. 2017). The normal endocervical cells are replaced by columnar cells with abundant eosinophilic or pale-pink cytoplasm and basally located nuclei, Fig. 61. Small numbers of goblet cells can be present. Occasionally the cytoplasm can have a foamy appearance. Compared to the usual variant of AIS, gAIS frequently has less nuclear atypia and nuclear stratification as well as fewer mitoses and apoptotic bodies. However, this not always the case and some gAIS can have considerable nuclear atypia and nuclear stratification, Fig. 62. Since gAIS is not caused by HPV, it does not stain positively for p16, but it may show patchy p16 staining similar to what can be seen with normal endocervical glands. Like the usual variant of AIS, gAIS stains positively for carbonic anhydrase-IX (Liao et al. 2013). The eosinophilic cytoplasm contains neutral mucin that stains predominantly red with Alcian blue/periodic acid Schiff (PAS) stain. The mucus also stains with immunohistochemical markers such as MUC6 and HIK1083 that react with pyloric gland mucin. Immunohistochemical analysis for estrogen and progesterone receptors is usually negative and occasional cases show a mutant p53 expression pattern (diffuse staining or complete absence of staining) as opposed to the usual high-risk HPV-related variant of variant AIS which shows a wild-type p53 expression pattern (Talia et al. 2017).
Fig. 61

Gastric variant of AIS. This variant is characterized by abundant eosinophilic cytoplasm and occasional goblet cells. Typically it has fewer mitotic figures and is less nuclear atypia and stratification than the usual endocervical variant of AIS

Fig. 62

Gastric variant of AIS. This example displays considerable nuclear atypia and stratification, which might suggest the usual type of AIS, but it has the characteristic eosinophilic cytoplasm and a p16 immunostain confirmed that this lesion was p16 negative. An associated invasive gastric-type endocervical adenocarcinoma was present in this case

Differential Diagnosis

The differential diagnosis of AIS includes reparative/reactive glandular atypia secondary to inflammation, radiation or viral infections, Arias-Stella reaction, microglandular hyperplasia, endometriosis, tubal metaplasia, mesonephric remnants, and invasive adenocarcinoma. Endocervical glands may display a wide range of cytological and architectural changes in response to inflammation and radiation. In reactive/reparative atypia, the nuclei become enlarged and have prominent nucleoli, but have nuclear clearing and lack hyperchromasia, Fig. 63. Nuclei may be pleomorphic, but the chromatin is usually smudged and degenerative in appearance. Mitotic activity is usually absent or minimal, as is pseudostratification. Care must be taken to distinguish between true pseudostratification and tangential sectioning through glands which can appear as pseudostratification. Inflamed endocervical glands with reserve cell hyperplasia can also appear as pseudostratification, Fig. 64. Although intraglandular papillary projections should not occur in reactive/reparative processes, exaggerated endocervical papillary projections that project into the endocervical canal can occur. These stromal projections contain infiltrates of chronic inflammatory cells and are lined by a single layer of endocervical cells. Endocervical atypia secondary to repair characteristically has a dense acute and chronic inflammatory infiltrate surrounding the glands, and polymorphonuclear leukocytes may infiltrate into the epithelium.
Fig. 63

Reparative changes of endocervical epithelium. When reactive processes involve the endocervical epithelium, the nuclei become enlarged and have prominent nucleoli, but they lack hyperchromasia and nuclear clearing and have few mitotic figures

Fig. 64

Inflamed endocervical epithelium with reserve cell hyperplasia. When endocervical glands are inflamed and also have reserve cell hyperplasia, they can appear to be pseudostratified and occasionally be mistaken for AIS

Reactive glandular atypia secondary to irradiation is characterized by nuclear enlargement and pleomorphism, but the cytoplasm is frequently vacuolated or granular. Pseudostratification and mitotic figures are absent. Atypia due to irradiation has greater cell to cell variation in size and shape than AIS or endocervical glandular atypia. Glands with the Arias-Stella reaction have a single layer of hyperchromatic, enlarged nuclei that frequently protrude into the gland lumen (i.e., hobnail cells). Typically Arias-Stella reaction involves only a portion of a gland and mitotic activity is absent. Although microglandular hyperplasia, especially when inflamed, can occasionally be confused with AIS, microglandular hyperplasia lacks significant nuclear atypia, lacks pseudostratification, and has few mitotic figures, Fig. 65. Moreover, microglandular hyperplasia has a characteristic pattern of closely packed, small, uniform glands. Atypical forms of microglandular hyperplasia have been described which form solid masses of epithelium and have significant degrees of cytological atypia (Young and Scully 1989). These lesions almost always contain areas of typical microglandular hyperplasia which allow them to be identified as atypical forms of microglandular hyperplasia (see chapter “Benign Diseases of the Cervix”). Similarly, endometriosis of the cervix is usually readily recognizable and easily distinguished from AIS. Typical endometriosis consists of glands and endometrial-type stroma. The cells lining the glands are basally located endometrial type cells that can be pseudostratified and mitotically inactive. Both tubal metaplasia and mesonephric remnants should not be mistaken for AIS since they have bland, nonmitotically active nuclei and typical histological features that allow them to recognized (see chapter “Benign Diseases of the Cervix”). However, tubal metaplasia can occasionally have more enlarged nuclei with coarse chromatin that can make it difficult to distinguish from AIS, Fig. 66.
Fig. 65

Microglandular hyperplasia with reparative changes. When significant inflammation is present, microglandular hyperplasia can become somewhat atypical and mistaken for AIS. Even when inflamed, mitotic figures are uncommon in microglandular hyperplasia and there is a lack of pseudostratification and minimal nuclear atypia

Fig. 66

Atypical tubal metaplasia. Occasionally tubal metaplasia becomes more hyperchromatic and has enlarged nuclei with coarse chromatin. When this occurs, it can be misinterpreted as AIS. However, such lesions typically lack mitotic activity and pseudostratification

Immunohistochemical staining can be quite useful in distinguishing AIS from other glandular lesions. Most, but not all, AIS lesions demonstrate diffuse nuclear and/or cytoplasmic immunoreactivity for p16, Fig. 67 (Negri et al. 2003; Riethdorf et al. 2002; Cameron et al. 2002; Murphy et al. 2003; Tringler et al. 2004; Volgareva et al. 2004). A recent systematic review and meta-analysis of tissue-based biomarkers in glandular neoplasia found that p16 is the most widely studied biomarker and estimated that ≥94% of AIS will stain positively for p16 (Lee et al. 2016). Reactive/reparative lesions and microglandular hyperplasia are usually p16 negative. Tubal metaplasia is not infrequently p16 positive, but the staining pattern is typically different than that seen with AIS. In tubal metaplasia, the staining with p16 is usually focal and weak, as opposed to diffuse and strong as seen in AIS, Fig. 68 (McCluggage 2007). However, some cases of tubal metaplasia can be diffusely positive and in these instances a Ki-67 IHC is useful to help distinguish these lesions from AIS. The Ki-67 proliferation index in tubal metaplasia and endometriosis is usually quite low, less than 10% of the cells will stain positively, whereas the Ki-67 proliferation index in AIS is much higher (Cameron et al. 2002; Pirog et al. 2002; McCluggage et al. 1995). Over 30% of the cells in AIS lesions usually show nuclear staining with Ki-67 and in most instances the majority of AIS cells will stain positively, Fig. 67b. AIS also stains positive for an antigen extracted from HeLa cells (mn antigen which is now known to represent carbonic anhydrase-IX) (Liao et al. 1994). More recent studies have shown that >92% of AIS stain positively with carbonic anhydrase-IX, whereas normal endocervical glands do not (Liao et al. 2013; Choschzick et al. 2014). Although less well studied than p16, one potential advantage of carbonic anhydrase-IX is that it is reported to stain gastric-type AIS, whereas p16 does not (Liao et al. 2013). ProEx-C which is a combination of several biomarkers had a 93% sensitivity for distinguishing AIS from reactive glandular lesions in one study (Sanati et al. 2010). Bcl-2 has been used to help differentiate AIS from endometriosis and tubal metaplasia. Bcl-2 appears to have antiapoptotic properties and both tubal metaplasia and endometriosis stain positively for bcl-2, whereas AIS lesions which show significant degrees of apoptosis stain either negatively or only focally positive with bcl-2 (Cameron et al. 2002). Carcinoembryonic antigen (CEA) is expressed in the cytoplasm of 67% of AIS, whereas the normal columnar epithelium of the endocervix is either negative or demonstrates only luminal, as opposed to cytoplasmic staining (Hurlimann and Gloor 1984; Marques et al. 1996). Immunohistochemical staining for vimentin typically is negative in AIS lesions but shows cytoplasmic positivity in tubal metaplasia and endometriosis (Marques et al. 1996). Similarly both tubal metaplasia and endometriosis usually stain positively for estrogen receptors, whereas AIS does not. Some authors have recommended that a panel of antibodies be utilized when trying to identify difficult cases of AIS. This panel includes Ki-67, p16, bcl-2, monoclonal CEA, vimentin, and estrogen receptor antibodies (McCluggage 2007). In our experience, a panel of p16, Ki-67, ER, and PR suffices.
Fig. 67

p16 staining of adenocarcinoma in situ (AIS). (a) Classic AIS showing sharp demarcation between normal and neoplastic tissue. (b) The neoplastic tissue stains strongly with p16. (c) It has a high Ki-67 labeling index

Fig. 68

Typical tubal metaplasia. Tubal metaplasia typically shows focal weak staining with p16 as opposed to the strong diffuse staining seen in AIS. (a) H&E, (b) p16 immunostaining

Invasion should be suspected if the involved glands extend beyond the glandular field or beyond the deepest uninvolved endocervical crypt. In addition, in AIS there should be no desmoplasia or stromal reaction around the involved glands. Other worrisome features that can be associated with invasion are exuberant glandular budding, an extensive cribriform pattern, foci in which the glands become confluent or back-to-back and the formation of papillary projections from the endocervical surface (Kudo et al. 1991; Ostor et al. 2000).

Clinical Behavior and Treatment

Due to the relative rarity of AIS, no natural history studies have been published and therefore the evidence that these lesions are precursors for invasive endocervical adenocarcinoma remains circumstantial. Despite this, until recently it was believed that women with AIS should undergo hysterectomy because there were numerous reports in the literature of persistent AIS or invasive disease occurring after conservative management of AIS with conization (Kennedy and Biscotti 2002; Muntz 1996). Because AIS most often occurs in young women who may be desirous of preserving their fertility, there has been a movement towards managing these women conservatively with conization, either cold-knife, large loop electrosurgical excision (LEEP), or laser (Massad et al. 2013a; Costa et al. 2012; Munro et al. 2017). There have been a number of series of patients with AIS who have been followed after cone biopsy. These studies have generally shown low rates of recurrence among women treated by means of a cone biopsy, provided the endocervical margin is negative. In the European multicenter trial of treatment of AIS with conization, 12.6% of patients were diagnosed during follow-up with AIS/SIL or invasive carcinoma (Costa et al. 2012). The most important risk factor for recurrence after conization is margin status (Munro et al. 2017; Costa et al. 2012; Goldstein and Mani 1998; Wolf et al. 1996; Salani et al. 2009). Women with positive endocervical margins are at significant risk for having an undiagnosed invasive cervical adenocarcinoma or for developing recurrent AIS. Although residual disease has been reported in up to 40% of patients with uninvolved margins, this can increase to up to 80% if the margin is involved (Costa et al. 2012). A 2009 meta-analysis found that of 671 patients followed after conization with surveillance only, 26% of women with negative margins developed recurrent disease, but only 0.1% developed invasive adenocarcinoma (Salani et al. 2009). The meta-analysis found that invasive adenocarcinoma was detected in 5.2% of patients with positive margins. Other factors for residual or recurrent disease are age >30 years, pure AIS (e.g., no coexistent HSIL), and larger lesions (>8 mm) (Munro et al. 2017; Costa et al. 2012). Based on these studies, conservative management by cone biopsy alone is now considered to be an option in women with AIS desirous of maintaining their fertility, if the cone biopsy margins are negative.

Cervical Cytology

Strengths and Limitations of Cervical Cytology

Although it was introduced over a half century ago, cervical cytologic screening continues to be one of the most effective cancer prevention test available. Over the half century since it was introduced, so much epidemiological and modeling data have accumulated demonstrating the effectiveness of cytology that it has become the index by which all other cancer screening tests are compared. Cytologic screening performed only twice in a woman’s lifetime can reduce her risk for invasive cervical cancer by up to 43% and yearly screening is estimated to reduce a woman’s risk by over 90% (Parkin 1991; Goldie et al. 2004). However, despite the effectiveness of cytologic screening, it is important to remember that no screening, diagnostic, or therapeutic technique used in medicine is perfect, and cervical cytology is no exception. Some women will develop invasive cervical cancer, despite routine cytologic screening.

Over the last decades numerous advances have been made in cervical cytology collection techniques, how cytological preparations are evaluated, and the classification systems used for reporting cytologic diagnosis. One of the most important advances has been the introduction of liquid-based cytology. With liquid-based cytology, the cells collected from the cervix are transferred directly to a liquid fixation solution that is shipped to the cytology laboratory where the slide is prepared. One of the primary advantages of liquid-based cytology is that molecular testing for sexually transmitted infections such as HPV DNA, chlamydia, N. gonorrhea can be performed directly from liquid-based specimens. HPV testing is particularly useful when a diagnosis of ASC-US is made (i.e., “reflex” HPV DNA testing).

The Bethesda System (TBS) Terminology

In 1988, The Bethesda System (TBS) for reporting cervical/vaginal cytological diagnoses was developed to provide uniform guidelines for reviewing and reporting gynecological Papanicolaou tests (The Bethesda 1988). The Bethesda System classification was subsequently modified in 1991 and 2001 (Workshop 1991; Solomon et al. 2002). In 2014 in response to additional experience with liquid-based cytology, additional insights into the biology of HPV, the widespread introduction of HPV vaccination, and adoption of HPV testing either alone or in combination with cytology the 2001 Bethesda System was reviewed and updated (Nayar and Wilbur 2015). The Bethesda System is now the standard classification for cervical cytology used in the United States. There are multiple distinct parts to the report, but a statement of specimen adequacy and the Interpretation/Results are the key ones, Table 10. Other parts of the report can provide additional information. This is designed to assist clinicians by answering three basic questions: (1) was the sample adequate? (2) was the cytology test normal? and (3) if the test was not completely normal, what specifically was wrong?
Table 10

The Bethesda System 2014 Classification (Nayar and Wilbur 2015)

Specimen type:

Indicate conventional smear (Pap smear) versus liquid-based preparation versus other

Specimen adequacy:

 Satisfactory for evaluation

 Unsatisfactory for evaluation (specify reason)

General categorization (optional)

 Negative for intraepithelial lesion or malignancy

 Other: See interpretation/result

 Epithelial cell abnormality: See interpretation/result (specify “squamous” or “glandular” as appropriate)


Negative for intraepithelial lesion or malignancy

  Nonneoplastic findings (optional to report)

   Nonneoplastic cellular variations

   Reactive cellular changes

   Glandular cells status posthysterectomy



  Endometrial cells (in a women > 45 yrs of age)

Epithelial cell abnormalities

  Squamous cell

   Atypical squamous cells

    – Of undetermined significance (ASC-US)

    – Cannot exclude HSIL (ASC-H)

   Low-grade squamous intraepithelial lesion (LSIL) (encompassing: HPV/milddysplasia/CIN1)

   High-grade squamous intraepithelial lesion (HSIL) (encompassing: moderate and severe dysplasia, CIS: CIN 2 and CIN 3)

   Squamous cell carcinoma

  Glandular cell


    – Endocervical cells (NOS or specify in comments)

    – Endometrial cells (NOS or specify in comments)

    – Glandular cells (NOS or specify in comments)


    – Endocervical cells, favor neoplastic

    – Glandular cells, favor neoplastic

   Endocervical adenocarcinoma in situ


    – Endocervical

    – Endometrial

    – Extrauterine

    – Not otherwise specified (NOS)

Other malignant neoplasms: (specify)

Adjunctive testing:

Provide a brief description of the test method(s) and report the result so that it is easily understood by the clinician

Computer-assisted interpretation of cervical cytology

If case examined by an automated device, specify device, and result

Educational notes and comments appended to cytology reports (optional)

Cytological Appearance of Cervical Cancer Precursors

The Bethesda System classification categorizes precursors to cervical cancer as “epithelial cell abnormalities.” The category “epithelial cell abnormalities” is subdivided into abnormalities of squamous cells and abnormalities involving glandular cells, either endocervical or endometrial. Cytological changes previously classified as mild squamous cytological atypia and atypical endocervical cells are also included in this category.

Squamous Cell Abnormalities
Atypical Squamous Cells (ASC)

The ASC category is used to designate cytological changes suggestive of a squamous intraepithelial lesion (SIL) that are quantitatively or qualitatively insufficient for a definitive diagnosis of SIL (Workshop 1991; Solomon et al. 2002; Nayar and Wilbur 2015). There are several points that need to be made with respect to ASC. First, a diagnosis of ASC is one of exclusion; the cells are abnormal, but they do not warrant a diagnosis of SIL. Second, a diagnosis of ASC should not be used when the underlying process is inflammatory or reactive, such slides should be carefully reviewed and classified as “negative for intraepithelial lesion or malignancy” whenever possible rather than ASC. Third, although the ASC category is sometimes disparagingly referred to as a “cytological wastebasket,” there are specific criteria that should be used for making this diagnosis. If these criteria are adhered to, the median rate of ASC in US laboratories is approximately 5% of all cytology specimens and the ASC rate should be no more than twice the SIL rate (Eversole et al. 2010).

The 2014 Bethesda System subdivides the ASC category into two subdivisions: Atypical Squamous Cells – Undetermined Significance (ASC-US) refers to samples in which the cytological changes are suggestive of LSIL but lack sufficient cytological abnormalities to allow a definitive diagnosis, and Atypical Squamous Cells – Cannot Exclude an HSIL (ASC-H) refers to samples in which the cytological changes are suggestive of HSIL, but the cytological abnormalities are insufficient to allow a definitive interpretation (Sherman et al. 1999).

The specific criteria used to diagnose ASC are given in Table 11. One of the major criteria used to distinguish ASC-US from benign cellular changes is nuclear size. In ASC-US, the nuclei are typically 2.5–3 times the size of a normal intermediate cell or twice the size of a squamous metaplastic cell nucleus, Fig. 69. In addition there can be a slightly increased nuclear to cytoplasmic ration (N/C ratio). Other features that are found in ASC-US are minimal nuclear hyperchromasia and irregularity in chromatin distribution or nuclear shape (Nayar and Wilbur 2015). ASC-US can include “atypical parakeratosis” which has cells with nuclear atypia associated with dense orangeophilic cytoplasm. A diagnosis of ASC-US is sometimes made when there are cells with some, but not all of the criteria necessary for a diagnosis of LSIL, Fig. 70. This typically occurs when there are cytoplasmic changes that suggest HPV effect (nuclear halos), but the cells have minimal nuclear changes.
Table 11

Criteria used to diagnosis atypical squamous cells (ASC)

Atypical squamous cells – undetermined significance (ASC-US)

 Cells resemble superficial or intermediate squamous cells in size and configuration

 Nuclei are 2.5–3 times the size of a normal intermediate cell nuclei or 2 times the size of a metaplastic cell nucleus

 Nuclei are round to oval with minimal irregularities

 Nuclei are normochromatic to slightly hyperchromatic

Atypical squamous cells – cannot exclude HSIL (ASC-H)

Small cells with high N/C ratios

Cells resemble parabasal or basal cells in size and configuration, but the nucleus is 1.5–2.5 times larger resulting in a high N/C ratio

 Cells occur singularly or in small groups

 Nuclei often have uneven chromatin and are hyperchromatic

 Nuclear contour is often irregular

Crowded sheet pattern

Crowded clusters of cells resembling parabasal or basal cells with atypical nuclear features including hyperchromasia and high N/C ratios

 Cells have a loss of polarity and can be difficult to visualize

Fig. 69

Atypical squamous cells – undetermined significance (ASC-US). Intermediate squamous epithelial cells demonstrate nuclear enlargement and hyperchromasia. No organisms or inflammatory changes were identified on the smear

Fig. 70

Atypical squamous cells – undetermined significance (ASC-US). These cells are suggestive but not diagnostic of a LSIL since they have a considerable degree of nuclear enlargement and perinuclear halos. However, the findings are not sufficient to allow a diagnosis LSIL and there only a limited number of such cells were present so a diagnosis of ASC-US was rendered

It should be emphasized that the degree of nuclear changes considered sufficient to warrant a diagnosis of ASC-US is highly subjective and varies between cytologists. This introduces a degree of uncertainty with respect to a diagnosis of ASC-US, and studies have shown that a diagnosis of ASC-US is the least reproducible of all cytological diagnoses (Confortini et al. 2003, 2007; Gatscha et al. 2001). Approximately 3–5% of women with a diagnosis of ASC-US will have histologic HSIL (CIN 3) when colposcopy is performed (Stoler et al. 2013; Stoler et al. 2011; Tewari et al. 2017).

The second category of ASC is ASC-H in which the cells resemble parabasal or basal cells in size and configuration. ASC-H appears as two different types of patterns. One is small cells with high N/C ratios. These cells occur singly or in small groups. The size of the cells is similar to that of metaplastic cells, but the nuclei are 1.5–2.5 times larger than normal which results in a high N/C ratio. The cells frequently are hyperchromatic with irregular nuclear contours and uneven chromatin, Fig. 71. The differential diagnosis in such cases is between atypical immature squamous metaplasia and HSIL. The other pattern of ASC-H is the “crowded sheet pattern” (Nayar and Wilbur 2015). This pattern shows a crowded cluster of squamous cells with atypical nuclear features including hyperchromasia and high N/C ratios, Fig. 72. The cells have a loss of polarity and can be difficult to visualize. These crowded clusters of cells can represent HSIL that has grown into endocervical glands or crypts, reactive or neoplastic endocervical cells, or atrophy.
Fig. 71

Atypical squamous cells – cannot exclude HSIL (ASC-H). A cluster of atypical immature metaplastic type cells is present. The cells have an increased N:C ratio, hyperchromasia, and slightly irregular nuclei. However, the cells have more cytoplasm than usually associated with HSIL and only a few clusters were present

Fig. 72

Atypical squamous cells – cannot exclude HSIL (ASC-H). This crowded sheet of cells is hyperchromatic and appears to have high N/C ratios. However, it is difficult to see the individual cells so one does not want to diagnose it as HSIL. Instead a diagnosis of ASC-H should be given

ASC-H is an uncommon finding and typically accounts for less than 10% of all ASC diagnoses. The median reporting rate of ASC-H in the US laboratories in 2006 was 0.3% according to the College of American Pathologists (CAP) survey (Eversole et al. 2010). The majority of women with ASC-H are high-risk HPV DNA positive, and histologic HSIL is identified at the time of colposcopy in 12–68% of women with ASC-H (Sherman et al. 1999; Bandyopadhyay et al. 2008; Liman et al. 2005). Because of the high prevalence of histologic HSIL in women with ASC-H, it has been suggested that ASC-H would be more appropriately referred to as “equivocal HSIL” (Wright et al. 2007).

Low-Grade Squamous Intraepithelial Lesions (LSIL)
The LSIL category in the Bethesda System includes both HPV effects and mild dysplasia (CIN 1). The cells of LSIL are of the superficial or intermediate cell type and are found either as individual cells or as sheets of cells with well-defined cell borders. The cells are typically enlarged with abundant cytoplasm. Nuclei are usually enlarged to >3 times the size of a normal intermediate cell nucleus and in some instances can be quite large, Fig. 73. There are usually prominent perinuclear halos, and the nuclei are usually hyperchromatic and multinucleation is common, Fig. 74. The chromatin is finely granular and uniformly distributed.
Fig. 73

LSIL. The cells are of the intermediate type with nuclear enlargement and prominent koilocytosis. One of the nuclei is more than 10 times the size of a normal intermediate cell

Fig. 74

LSIL. Cells from this lesion demonstrate considerable koilocytosis with multinucleation and prominent halos

With cytology methods such as liquid-based cytology and computerized imaging systems, the rate of LSIL appears to be increasing in the USA. In surveys taken in the 1990s the median reporting rate of LSIL in US laboratories was 1.6%, but by 2006 this has increased to 3% (Jones and Davey 2000; Eversole et al. 2010). A recent report of the impact of implementing a computerized cytology imaging system in a tertiary military center reported that after implementation the rate of LSIL increased from 2.6% to 3.9% (Duby and DiFurio 2009). Age is an important factor in determining the prevalence of LSIL. In a large US screening trial, the prevalence of LSIL decreased from 6.5% in women 21–24 years of age to 3.8% in women 25–29 years and then to 1.4% in women 40–49 years (Wright et al. 2012).

High-Grade Squamous Intraepithelial Lesions (HSIL)
Because The Bethesda System combines moderate and severe dysplasia together with carcinoma in situ in the HSIL category, there is a wide variation in the cytological appearance of HSIL. As the severity of the lesion increases, the degree of differentiation and the amount of cytoplasm decreases, the nuclear: cytoplasmic ratio increases, and the degree of nuclear atypia increases, Table 12. Although cellular size varies considerably, the cells of HSIL are typically smaller and have less cytoplasm than do the cells of LSIL. Some HSIL are quite small and are the size of basal cells, The degree of nuclear enlargement also varies considerably, Fig. 75. In some cases, the nucleus is as large as that of the typical LSIL, but since there is less cytoplasm the N/C ratio is higher than in LSIL. In cases in which the overall size of the HSIL cells is small, the nuclei may not much larger than that of a typical intermediate cell, but the N/C ratio is quite high, Fig. 76. The nucleus is typically quite hyperchromatic and the contour of the nucleus is usually irregular and indentations are often present. The chromatin can be fine or coarsely granular and is evenly distributed. Nucleoli are generally not present. The cytoplasm of HSIL also varies considerably. In some cases, it can be thin or lacey, in other cases it is densely metaplastic, and in others it can be densely keratinized, Fig. 77. The cells can occur singly, in sheets and clusters, or as syncytial aggregates.
Table 12

Criteria used to diagnosed squamous epithelial cell abnormalities

Bethesda System



CIN terminology




Older terminology

Mild dysplasia

Mod. dysplasia

Severe dysplasia


Cell type

Superficial or intermediate



Basal, spindle, pleomorphic

Cell arrangement

Singly or sheets

Singly or sheets

Singly or sheets

Singly or sheets or syncitia

Number abnormal cells










Nuclear size










Nuclear:cytoplasmic ratio





Fig. 75

HSIL. There is considerable variability in nuclear size. Many of the cells have a considerable amount of cytoplasm, but the N:C ratio is higher than usually seen in LSIL

Fig. 76

HSIL. These cells are the size of parabasal cells, but the nucleus is the same size as an intermediate cell. Therefore, the N/C ratio is greatly increased

Fig. 77

HSIL. Many of the HSIL cells are keratinized and have quite hyperchromatic nuclei. It is often difficult to distinguish between HSIL of this type and keratinizing invasive squamous cell carcinoma

The number of abnormal cells that are present can vary dramatically case to case. When only a few small HSIL cells of the basal cell type are present, it can be quite challenging to correctly classify the case as a HSIL, and these cases account for a disproportionate percentage of false negative cervical cytology. According to the College of American Pathologists (CAP) 2006 survey, the median reporting rate of HSIL in US laboratories was 0.6% (Eversole et al. 2010). The rate of HSIL varies with age. In a large US screening trial, the prevalence of HSIL decreased from 0.7% in women 21–24 years of age to 0.4% in women 25–29 years and then to 0.2% in women 40–49 years (Wright et al. 2012). A diagnosis of HSIL connotes a high risk for significant cervical disease. Histologic HSIL is found in approximately 60% of women with HSIL (Massad et al. 2013b).

Invasive Squamous Cell Carcinoma
Squamous cell carcinomas of the cervix are subdivided into keratinizing and non-keratinizing types. Nonkeratinizing carcinomas typically have large numbers of malignant cells that form loose cell sheets and syncytial arrangements. The cells are usually somewhat smaller than HSIL but have most of the features of HSIL. The nuclei have coarsely clumped chromatin and focal chromatin clearing and prominent macronucleoli may be present, Fig. 78. Squamous cell carcinomas often have a “dirty” background containing blood, cellular debris, fibrin, and necrotic material. This is often referred to as a tumor diathesis. This characteristic background is less prominent in liquid-based cytology specimens. However, in liquid-based cytology, there is often a distinctive necrotic background that is easy to recognize since it surrounds the cellular material in a “clumped” appearance and large, necrotic tissue fragments are sometimes present, Fig. 79.
Fig. 78

Invasive squamous cell carcinoma. The cells of this nonkeratinizing squamous cell carcinoma are polygonal and arranged in syncytial sheets. They are highly atypical, but smaller than the cells of many intraepithelial lesions

Fig. 79

Tumor diathesis. This liquid-based cytology preparation from a woman with squamous cell carcinoma shows necrotic material surrounding malignant cells giving it a “clumped” appearance

Cytology specimens from women with keratinizing carcinomas contain malignant cells demonstrating a variety of cell shapes and sizes, Fig. 80. Some of the cells are pleomorphic or tadpole-shaped. These cells have abundant orangophilic cytoplasm. There is frequently abundant hyperkeratosis and parakeratosis. The nuclei are irregular in shape and quite hyperchromatic. Sometimes the nuclei are degenerated, appearing as opaque masses or “ink blots.” Unlike nonkeratinizing squamous cell carcinoma, keratinizing squamous cell carcinomas often do not have “dirty” background or evidence of tumor diathesis.
Fig. 80

Invasive squamous cell carcinoma. The cells of this keratinizing squamous cell carcinoma are quite pleomorphic and include spindle-shaped, elongate, and caudate forms. The nuclei of some cells are extremely hyperchromatic

Glandular Cell Abnormalities

In the 2014 Bethesda System, all types of glandular cell abnormalities, including both atypical endocervical and endometrial cells, are combined together in a single entity referred to as glandular cell abnormalities. Benign appearing endometrial cells occurring in postmenopausal women are classified as “other.” Glandular cell abnormalities are divided into three categories: atypical glandular cells, either unqualified or favor neoplastic; adenocarcinoma in situ; and invasive adenocarcinoma.

Atypical Glandular Cells (AGC)

All atypical glandular cells lacking the diagnostic features of adenocarcinoma, irrespective of whether they are of endometrial or endocervical origin, are classified by the 2014 Bethesda System as atypical glandular cells with a specification as to whether they are endocervical, endometrial, or of uncertain origin. There are two categories of AGC. The first is atypical glandular cells (either endocervical, endometrial, or unclassified) that are not qualified and the second is atypical glandular cells; favor neoplastic.

Glandular cytological abnormalities are considerably less common than squamous abnormalities, and most cytologists tend to be less comfortable recognizing and diagnosing them. In addition, the criteria used to differentiate reactive endocervical changes, endocervical dysplasia, endocervical adenocarcinoma in situ and invasive endocervical adenocarcinoma are less well established than those used for squamous lesions. Cytologists even have difficulty in differentiating atypical endocervical cells from cases of HSIL that have extended into endocervical crypts. This accounts for the high prevalence of squamous abnormalities (approximately 30%) detected in women referred for AGC to colposcopy (Kim et al. 1999; Ronnett et al. 1999).

The cytological features of atypical endocervical cells vary depending on the degree of the underlying histopathologic abnormality. Cases of the type designated by cytopathologists as atypical endocervical cells – not otherwise specified (NOS) – have variability in nuclear size and shape, Fig. 81. These cells occur in sheets or strips with some cell crowding and nuclear overlap. Nuclei are typically enlarged compared to normal endocervical cells with up to 3–5 times greater nuclear area. There is mild hyperchromasia and mild chromatin irregularity. Mitoses are rare and there are only occasional nucleoli. Atypical endocervical cells favor neoplasia includes those cases where the cytological features are suggestive of adenocarcinoma in situ, but are insufficient to allow a definitive diagnosis. These cases typically have more nuclear hyperchromasia, variability in nuclear size, and granularity of the chromatin than is observed in cases of atypical endocervical cells, NOS, Fig. 82. When the cells occur in strips, they often are pseudostratified.
Fig. 81

Atypical glandular cells – endocervical type (AGC-EC). These endocervical cells have enlarged nuclei and prominent nucleoli and vary somewhat in size and shape. The smear was obtained 6 weeks postpartum and follow-up examination was completely negative

Fig. 82

Atypical glandular cells – favor neoplasia. These endocervical cells are somewhat suggestive of adenocarcinoma in situ. The nuclei are hyperchromatic and the chromatin is coarsely clumped. There is variation in nuclear size and the cells form a three-dimensional aggregate

Adenocarcinoma In Situ
In cases of adenocarcinoma in situ, there are usually a larger number of atypical glandular cells that form crowded cellular clusters. The sheets are usually three-dimensional and sometimes retain the architecture of the underlying glands, Fig. 83. The cells within these sheets occasionally form rosettes and have extensive feathering of the cells at the periphery. Individual endocervical cells are highly atypical with enlarged round, oval, or elongated nuclei that vary in size from cell to cell. In most cases, the chromatin is coarsely clumped and multiple mitoses are seen, Fig. 84. Sometimes it is difficult for the cytologist to determine whether the atypical cells represent atypical glandular cells or HSIL cells that have extended into an endocervical crypt or “gland.” In these cases, highly atypical nuclei are identified in the center of a cell aggregate and some of the cells at the periphery of the aggregate appear to be endocervical cells.
Fig. 83

Adenocarcinoma in situ. These endocervical cells form a tight three-dimensional structure that is similar to the outline of an endocervical gland. A large number of these formations are often present in specimens from AIS

Fig. 84

Adenocarcinoma in situ. The individual atypical endocervical cells are hyperchromatic with coarsely clumped chromatin. They show the characteristic feathering of the nuclei at the edge of the cluster which is typical of AIS

The Bethesda System subclassifies invasive adenocarcinomas into “adenocarcinoma-endocervical type,” “adenocarcinoma-endometrial type,” and “adenocarcinoma-not otherwise specified”. The cytological diagnosis of invasive adenocarcinoma is relatively straightforward. Adenocarcinoma cells from either an endocervical or an endometrial primary have enlarged nuclei, high nuclear: cytoplasmic ratios, coarsely clumped chromatin, and prominent nucleoli, Fig. 85. They can occur singly or in clusters. Cytologists should try to distinguish between endometrial and endocervical primary adenocarcinomas whenever possible. Key features that allow discrimination between endometrial and endocervical origin in cytology include number of abnormal cells, size of the cells, retention of columnar configuration, appearance of cytoplasm, and nuclear structure (Ng 1993). Typically, adenocarcinoma of the cervix shows considerably larger numbers of cells than does endometrial adenocarcinoma. The cells of endometrial adenocarcinoma typically occur singly or as small clusters, whereas the cells of endocervical adenocarcinoma occur as larger 2 dimensional sheets, 3 dimensional clusters, or syncytial sheets. Cells derived from endocervical adenocarcinoma typically retain a columnar configuration which is lost in most endometrial carcinomas. The cytoplasm of cells exfoliated from endocervical adenocarcinoma is typically finely vacuolated, whereas the cytoplasm from cells of endometrial adenocarcinoma is typically scant and cyanophilic. The nuclei of cells of endometrial adenocarcinoma vary in size becoming larger with higher grade tumors. The chromatic is less granular and the nuclei are less hyperchromatic than the nuclei of cells of endocervical adenocarcinoma. They also less frequently have multiple nucleoli.
Fig. 85

Endocervical adenocarcinoma. These endocervical cells have the features of frank adenocarcinoma. The nuclei are quite enlarged, the chromatin is coarsely clumped and marginated, and there are prominent nucleoli. The background shows inflammation and necrosis indicating tumor diathesis is present

Management of Cytologic Abnormalities and Cervical Cancer Precursors

In 2012 the American Society for Colposcopy and Cervical Pathology sponsored a consensus workshop to update prior Consensus Guidelines for the Management of Women with Cytological Abnormalities and Cervical Cancer Precursors (Wright et al. 2007; Massad et al. 2013a). These guidelines are widely used in the USA and are evidence-based with each recommendation accompanied by a grading of both the strength of the recommendation and the strength of the data supporting the recommendation. What follows is a brief synopsis of the guidelines. The complete recommendations and management algorithms are available at

Atypical Squamous Cells (ASC)

The prevalence of biopsy-confirmed HSIL among women undergoing colposcopy for an ASC cytology varies from 5% to 17% (Wright et al. 2007; Stoler et al. 2011, 2013). The prevalence of HSIL in women with ASC depends on a number of factors including the patient’s age, history, and the subclassification of the ASC result. Overall, it appears that approximately half of women with histologic HSIL have ASC as their initial abnormal cervical cytology result (Lonky et al. 1999; Kinney et al. 1998). However, it should be noted that the risk that a woman with ASC has invasive cervical cancer is quite low (about one per thousand).

Atypical Squamous Cells: Undetermined Significance (ASC-US)
In Kaiser Northern California, the 5 year cumulative risk for HSIL in women 30–64 years of age with ASC-US is 6.9% (Katki et al. 2013a). Two methods are considered acceptable for managing women in the general population with ASC-US: high-risk HPV DNA testing and repeating the cervical cytology at 1 year, Fig. 86 (Massad et al. 2013a). HPV DNA testing identifies more cases of HSIL than does a single repeat cervical cytology, but refers approximately equivalent numbers of women for colposcopy (Arbyn et al. 2006). Moreover, cost-effectiveness modeling has demonstrated that HPV DNA testing for women with ASC-US is highly attractive when the initial ASC-US cytology was obtained from a liquid-based sample (Kulasingam et al. 2006; Pedersen et al. 2016). Thus, high-risk HPV DNA testing is the preferred approach to managing women with ASC-US whenever liquid-based cytology is used for screening (Massad et al. 2013a). Women found to be high-risk HPV DNA positive should be referred to colposcopy, whereas HPV DNA negative women should undergo repeat cotesting in 3 years.
Fig. 86

ASCCP Consensus Conference algorithm for managing women with ASC-US

Since the prevalence of HPV DNA positivity is much higher in young women with ASC-US than in older women, HPV DNA testing is not recommended for young women with ASC-US (Sherman et al. 2002; Boardman et al. 2005; Stoler et al. 2011, 2013). Instead, women 21–24 years of age with ASC-US are managed using annual repeat cytological examinations and only referred to colposcopy if the repeat Pap tests are diagnosed as ASC-H, HSIL, or AGC or are persistently abnormal for a period of 2 years, Fig. 87 (Massad et al. 2013a). Management options for pregnant patients with ASC-US are identical to those for nonpregnant patients with the exception that it is acceptable to defer the colposcopic examination until the patient is 6 weeks post-partum.
Fig. 87

ASCCP Consensus Conference algorithm for managing young women with either ASC-US or LSIL

Atypical Squamous Cells: Cannot Exclude HSIL (ASC-H)

ASC-H is a much more concerning cytology result than ASC-US since biopsy-confirmed HSIL is identified in 13% to 66% of women with ASC-H (Xu et al. 2016). Thus for the purposes of management, ASC-H should be considered to be an equivocal HSIL result and all women with ASC-H should be referred for a colposcopic evaluation (Massad et al. 2013a). If after colposcopy the patient has histologic LSIL or less, follow-up utilizing either repeat cotesting at 12 and 24 months, a loop electrosurgical excision (LEEP), or review of the cytological, histological, and colposcopic findings is acceptable.

Low-Grade Squamous Intraepithelial Lesions (LSIL)

In Kaiser Northern California, the 5 year cumulative risk for histologic HSIL in women 30–64 years of age with LSIL is 16% (Katki et al. 2013b). The risk of HSIL varies considerably depending on the patient’s age and HPV status. The 5 year cumulative risk of HSIL is 19% for HPV-positive LSIL and only 5.1% for HPV-negative LSIL. In women 30–34 years with LSIL the risk of histologic HSIL is 17%, whereas it drops to 7.3% in women 60–64 years. Therefore, the management of women with a cytological result of LSIL varies depending on HPV status and age. Women with LSIL who either are of unknown HPV status or HPV positive should be referred to colposcopy, Fig. 88 (Massad et al. 2013a). The preferred approach to managing HPV negative women with LSIL is to do a repeat cotest in 12 months. Since invasive cervical cancer is very uncommon in young women and prospective studies have shown that over 90% of LSIL will spontaneously clear, young women with LSIL should not be referred to colposcopy, but should be followed using yearly cytology tests for a period of 2 years, Fig. 87 (Moscicki et al. 2001, 2006; Massad et al. 2013a). Another “special population” is postmenopausal women with LSIL. Both the prevalence of HPV DNA and the prevalence of histologic HSIL are lower in postmenopausal women with LSIL than in women in the general population with LSIL. Therefore, postmenopausal women with LSIL and no HPV test can be managed using either HPV testing, repeat cytologic testing at 6 and 12 months, and colposcopy.
Fig. 88

ASCCP Consensus Conference algorithm for managing women with LSIL

High-Grade Squamous Intraepithelial Lesions (HSIL)

Histologic HSIL is identified in 53–97% of women with a cytological result of HSIL and invasive cervical cancer is found in approximately 2% (Wright et al. 2007). In Kaiser, the 5 year cumulative detection rate of histologic HSIL in HPV negative women with cytologic HSIL is 49%. This increases to 71% in HPV-positive women with cytologic HSIL. Therefore, women with a cytological result of HSIL irrespective of HPV status should be referred for either a colposcopic evaluation or an immediate loop electrosurgical excisional procedure, Fig. 89 (Massad et al. 2013a). If after colposcopy the patient has histologic LSIL or less, followed-up utilizing either repeat cotesting at 12 and 24 months, a loop electrosurgical excision (LEEP), or review of the cytological, histological, and colposcopic findings is acceptable. For women 21–24 years with HSIL, initial colposcopy is recommended. If histologic HSIL is not identified, observation for up to 24 months using 6 monthly cytology and colposcopy is recommended. Nonpregnant women with HSIL who have an unsatisfactory colposcopic examination require a diagnostic excisional procedure.
Fig. 89

ASCCP Consensus Conference algorithm for managing women with HSIL

Histologic LSIL

LSIL is the histological manifestation of a HPV infection. The majority of histologic LSIL will spontaneously regress in the absence of therapy and few cases progress to histologic HSIL (Moscicki et al. 2004; Cox et al. 2003; Trimble et al. 2005). Therefore, it is recommended that histologic LSIL preceded by ASC-US, LSIL cytology, HPV 16 or 18 positivity, or HPV persistence undergo conservative follow-up consisting of cotesting at 1 year, Fig. 90 (Massad et al. 2013a). If histologic LSIL persists for at least 2 years, either continued follow-up or treatment is acceptable. Since the risk of an undetected histologic HSIL or glandular lesion is expected to be higher in women referred for the evaluation of an ASC-H, HSIL, or atypical glandular cells (AGC) on cytology, acceptable approaches to women with histologic LSIL preceded by HSIL or AGC cervical cytology includes followed-up utilizing either repeat cotesting at 12 and 24 months, loop electrosurgical excision (LEEP), or review of the cytological, histological, and colposcopic findings provided the colposcopic examination is satisfactory and endocervical sampling is negative.
Fig. 90

ASCCP Consensus Conference algorithm for managing women with histologic LSIL

Histologic HSIL

Women with histologic HSIL (CIN 3) are at significantly high risk of progressing to invasive cervical cancer and therefore treatment is recommended. Provided the colposcopic examination is satisfactory and there is no suggestion of invasive disease (e.g., by either colposcopy, cytology, or histology), both ablative or excisional treatment modalities are considered acceptable forms of treatment (Massad et al. 2013a). A diagnostic excisional procedure is recommended for all women with histological HSIL and unsatisfactory colposcopic examination or with recurrent disease. The management of women with histologic HSIL (CIN 2) is controversial and recommendations vary between countries. It is clear that regression rates and progression rates to invasive cervical cancer are lower in women with histologic HSIL (CIN 2) than in women with HSIL (CIN 3) (Tainio et al. 2018; Ostor 1993; Moscicki et al. 2010; Silver et al. 2018). In the USA follow-up is recommended for women desirous of maintaining childbearing capacity with histologic HSIL or HSIL (CIN 2), but treatment is considered acceptable (Massad et al. 2013a). For all other women, treatment is recommended.

Endocervical Curettage

Endocervical curettage (ECC) is performed to evaluate lesion distribution and morphology within the endocervical canal and to exclude the presence of invasive carcinoma, and unsuspected cervical adenocarcinoma in situ and invasive adenocarcinoma. Over the last decade the utility of endocervical curettage (ECC) has become the subject of considerable debate (Driggers and Zahn 2008). In ALTS, the ECC provided only a minimal 2.2% increase in the detection of HSIL when performed in women under the age of 40 years, but provided a 13% increased detection of histologic HSIL when performed in women 40 years and older (Solomon et al. 2007). A recent National Cancer Institute (NCI) study found that ECC detected histologic HSIL in 14% of women undergoing colposcopy (Liu et al. 2017). It was more likely to find HSIL in women with a high-grade cytologic abnormality, those who were HPV 16 positive and those with a high-grade colposcopic impression. In women with ASC-US or LSIL cytology with an unsatisfactory colposcopy, the ECC detected histologic HSIL in 13% of women, whereas when the colposcopic examination was normal or the examination was satisfactory it detected less than 5% HSIL. Many clinicians routinely perform an endocervical curettage during colposcopy. The endocervical curettage specimen consists of endocervical tissue fragments, blood, mucus, and, when positive, strips of atypical epithelium, Fig. 91. To avoid the loss of tiny tissue fragments during processing, the clinician should collect and concentrate the sample, including mucus and blood, on a small square of lens paper or using a cytobrush and immediately place it in the fixative (Hoffman et al. 1993). By this method, even the smallest tissue fragments can be recovered easily in the laboratory, embedded, and sectioned entirely.
Fig. 91

HSIL present in an endocervical curetting

In most instances, when atypical epithelium is detected in the endocervical curettage, it lacks underlying stroma and orientation is not possible. As a result, the pathologist can neither rule out underlying invasion nor grade an intraepithelial lesion. In other cases, where the atypical epithelium is well oriented, the pathologist is able to grade the lesion and can, if desired. It is also helpful if the pathologist conveys an estimate of the amount of atypical epithelium that is present in the endocervical curettage. If only a few small fragments of atypical epithelium are present in the endocervical curettage, these may represent “pick-ups” from a lesion that is actually confined to the portio and does not extend into the endocervical canal. In such cases it may be preferable to reexamine the patient with the colposcope rather than proceeding directly with a diagnostic excisional procedure. Frequently the second, carefully performed curettage yields no atypical epithelium and the patient may be managed on a conservative, outpatient basis. Conversely, the pathologist should be careful not to discount or overlook a few or even a single fragment of atypical epithelium in an endocervical curettage. In a review of 21 women who developed invasive cancer after cryotherapy, 7 out of 18 endocervical curettages taken before cryotherapy were found on review to contain SIL that had been missed at the time of original diagnosis (Schmidt et al. 1992).


  1. Abdul-Karim FW, Fu YS, Reagan JW, Wentz WB (1982) Morphometric study of intraepithelial neoplasia of the uterine cervix. Obstet Gynecol 60:210–214PubMedPubMedCentralGoogle Scholar
  2. ACOG (2013) Practice bulletin no. 140: management of abnormal cervical cancer screening test results and cervical cancer precursors. Obstet Gynecol 122(6):1338–1367CrossRefGoogle Scholar
  3. Anciaux D, Lawrence WD, Gregoire L (1997) Glandular lesions of the uterine cervix: prognostic implications of human papillomavirus status. Int J Gynecol Pathol 16:103–110PubMedCrossRefPubMedCentralGoogle Scholar
  4. Andersen ES, Arffmann E (1989) Adenocarcinoma in situ of the uterine cervix: a clinico-pathologic study of 36 cases. Gynecol Oncol 35:1–7PubMedCrossRefPubMedCentralGoogle Scholar
  5. Arbyn M, Sasieni P, Meijer CJ, Clavel C, Koliopoulos G, Dillner J (2006) Chapter 9: Clinical applications of HPV testing: a summary of meta-analyses. Vaccine 24(Suppl 3):S78–S89CrossRefGoogle Scholar
  6. Ault KA, Joura EA, Kjaer SK, Iversen OE, Wheeler CM, Perez G et al (2011) Adenocarcinoma in situ and associated human papillomavirus type distribution observed in two clinical trials of a quadrivalent human papillomavirus vaccine. Int J Cancer 128(6):1344–1353. Scholar
  7. Baker AC, Eltoum I, Curry RO, Stockard CR, Manne U, Grizzle WE et al (2006) Mucinous expression in benign and neoplastic glandular lesions of the uterine cervix. Arch Pathol Lab Med 130(10):1510–1515PubMedPubMedCentralGoogle Scholar
  8. Baldur-Felskov B, Dehlendorff C, Junge J, Munk C, Kjaer SK (2014) Incidence of cervical lesions in Danish women before and after implementation of a national HPV vaccination program. Cancer Causes Control 25(7):915–922. Scholar
  9. Bandyopadhyay S, Austin RM, Dabbs D, Zhao C (2008) Adjunctive human papillomavirus DNA testing is a useful option in some clinical settings for disease risk assessment and triage of females with ASC-H Papanicolaou test results. Arch Pathol Lab Med 132(12):1874–1881PubMedPubMedCentralGoogle Scholar
  10. Bergeron C, Ferenczy A, Shah K et al (1987) Multicentric human papillomavirus infections of the female genital tract. Correlation of viral types with abnormal mitotic figures, colposcopic presentation, and location. Obstet Gynecol 69:736–742PubMedGoogle Scholar
  11. Bergeron C, Ordi J, Schmidt D, Trunk MJ, Keller T, Ridder R (2010) Conjunctive p16INK4a testing significantly increases accuracy in diagnosing high-grade cervical intraepithelial neoplasia. Am J Clin Pathol 133(3):395–406.PubMedCrossRefPubMedCentralGoogle Scholar
  12. Bernard H-U, Chan S-Y, Delius H (1994) Evolution of papillomaviruses. Curr Top Microbiol Immunol 186:33–54PubMedGoogle Scholar
  13. Bertrand M, Lickrish GM, Colgan TJ (1987) The anatomic distribution of cervical adenocarcinoma in situ: implications for treatment. Am J Obstet Gynecol 157:21–25PubMedCrossRefGoogle Scholar
  14. Boardman LA, Stanko C, Weitzen S, Sung CJ (2005) Atypical squamous cells of undetermined significance: human papillomavirus testing in adolescents. Obstet Gynecol 105(4):741–746PubMedCrossRefGoogle Scholar
  15. Bocking A, Nguyen VQ (2004) Diagnostic and prognostic use of DNA image cytometry in cervical squamous intraepithelial lesions and invasive carcinoma. Cancer 102(1):41–54PubMedCrossRefGoogle Scholar
  16. Boddington MM, Spriggs AI, Cowdell RH (1976) Adenocarcinoma of the uterine cervix: cytological evidence of a long preclinical evolution. Br J Obstet Gynecol 83:900–903CrossRefGoogle Scholar
  17. Bollmann R, Mehes G, Torka R, Speich N, Schmitt C, Bollmann M (2003) Human papillomavirus typing and DNA ploidy determination of squamous intraepithelial lesions in liquid-based cytologic samples. Cancer 99(1):57–62PubMedCrossRefGoogle Scholar
  18. Bollmann R, Mehes G, Speich N, Schmitt C, Bollmann M (2005) Aberrant, highly hyperdiploid cells in human papillomavirus-positive, abnormal cytologic samples are associated with progressive lesions of the uterine cervix. Cancer 105(2):96–100PubMedCrossRefGoogle Scholar
  19. Boon ME, Baak JPA, Kurver PJH, Overdiep SH, Verdonk GW (1981) Adenocarcinoma in situ of the cervix: an underdiagnosed lesion. Cancer 48:768–773PubMedCrossRefGoogle Scholar
  20. Bosch FX, de Sanjose S (2007) The epidemiology of human papillomavirus infection and cervical cancer. Dis Markers 23(4):213–227PubMedPubMedCentralCrossRefGoogle Scholar
  21. Bousfield L, Pacey F, Young Q, Krumins I, Osborn R (1980) Expanded cytologic criteria for the diagnosis of adenocarcinoma in situ of the cervix and related lesions. Acta Cytol 24(4):283–296PubMedGoogle Scholar
  22. Boyle DP, McCluggage WG (2015) Stratified mucin-producing intraepithelial lesion (SMILE): report of a case series with associated pathological findings. Histopathology 66(5):658–663. Scholar
  23. Broders AC (1932) Carcinoma in situ contrasted with benign penetrating epithelium. J Am Med Assoc 99:1670–1674CrossRefGoogle Scholar
  24. Brotherton JM (2016) HPV vaccines: so much learnt, so many more lessons to come. Lancet Oncol 17(1):8–9. Scholar
  25. Brotherton JM, Gertig DM, May C, Chappell G, Saville M (2016) HPV vaccine impact in Australian women: ready for an HPV-based screening program. Med J Aust 204(5):184–1e1PubMedCrossRefGoogle Scholar
  26. Brown LJR, Wells M (1986) Cervical glandular atypia associated with squamous intraepithelial neoplasia: a premalignant lesion? J Clin Pathol 39:22–28PubMedPubMedCentralCrossRefGoogle Scholar
  27. Brown DR, Shew ML, Qadadri B, Neptune N, Vargas M, Tu W et al (2005) A longitudinal study of genital human papillomavirus infection in a cohort of closely followed adolescent women. J Infect Dis 191(2):182–192PubMedCrossRefGoogle Scholar
  28. Burchell AN, Winer RL, de Sanjose S, Franco EL (2006) Chapter 6: Epidemiology and transmission dynamics of genital HPV infection. Vaccine 24(Suppl 3):S52–S61CrossRefGoogle Scholar
  29. Burghardt E (1991) Colposcopy-cervical pathology, 2nd edn. Thieme Medical Publishers, New YorkGoogle Scholar
  30. Burk RD, Harari A, Chen Z (2013) Human papillomavirus genome variants. Virology 445(1–2):232–243. Scholar
  31. Cameron RI, Maxwell P, Jenkins D, McCluggage WG (2002) Immunohistochemical staining with MIB1, bcl2 and p16 assists in the distinction of cervical glandular intraepithelial neoplasia from tubo-endometrial metaplasia, endometriosis and microglandular hyperplasia. Histopathology 41(4):313–321PubMedCrossRefPubMedCentralGoogle Scholar
  32. Carreon JD, Sherman ME, Guillen D, Solomon D, Herrero R, Jeronimo J et al (2007) CIN2 is a much less reproducible and less valid diagnosis than CIN3: results from a histological review of population-based cervical samples. Int J Gynecol Pathol 26(4):441–446. 00004347-200710000-00013 [pii]CrossRefPubMedPubMedCentralGoogle Scholar
  33. Castle PE, Schiffman M, Gravitt PE, Kendall H, Fishman S, Dong H et al (2002) Comparisons of HPV DNA detection by MY09/11 PCR methods. J Med Virol 68(3):417–423PubMedCrossRefPubMedCentralGoogle Scholar
  34. Castle PE, Stoler MH, Solomon D, Schiffman M (2007) The relationship of community biopsy-diagnosed cervical intraepithelial neoplasia grade 2 to the quality control pathology-reviewed diagnoses: an ALTS report. Am J Clin Pathol 127(5):805–815PubMedCrossRefPubMedCentralGoogle Scholar
  35. Castle PE, Rodriguez AC, Burk RD, Herrero R, Wacholder S, Hildesheim A et al (2011) Long-term persistence of prevalently detected human papillomavirus infections in the absence of detectable cervical precancer and cancer. J Infect Dis 203(6):814–822. Scholar
  36. Choschzick M, Woelber L, Gieseking F, Oosterwijk E, Tennstedt P (2014) Carbonic anhydrase IX is strongly overexpressed in adenocarcinoma in situ of the cervix uteri. Histopathology 64(4):600–602. Scholar
  37. Clifford GM, Rana RK, Franceschi S, Smith JS, Gough G, Pimenta JM (2005) Human papillomavirus genotype distribution in low-grade cervical lesions: comparison by geographic region and with cervical cancer. Cancer Epidemiol Biomarkers Prev 14(5):1157–1164PubMedCrossRefPubMedCentralGoogle Scholar
  38. Clifford G, Franceschi S, Diaz M, Munoz N, Villa LL (2006) Chapter 3: HPV type-distribution in women with and without cervical neoplastic diseases. Vaccine 24(Suppl 3):S26–S34CrossRefGoogle Scholar
  39. Colgan TJ, Lickrish GM (1990) The topography and invasive potential of cervical adenocarcinoma in situ, with and without associated squamous dysplasia. Gynecol Oncol 36:246–249PubMedCrossRefGoogle Scholar
  40. Confortini M, Carozzi F, Dalla Palma P, Ghiringhello B, Parisio F, Prandi S et al (2003) Interlaboratory reproducibility of atypical squamous cells of undetermined significance report: a national survey. Cytopathology 14(5):263–268PubMedCrossRefGoogle Scholar
  41. Confortini M, Bondi A, Cariaggi MP, Carozzi F, Dalla Palma P, Ghiringhello B et al (2007) Interlaboratory reproducibility of liquid-based equivocal cervical cytology within a randomized controlled trial framework. Diagn Cytopathol 35(9):541–544PubMedCrossRefPubMedCentralGoogle Scholar
  42. Costa S, Venturoli S, Negri G, Sideri M, Preti M, Pesaresi M et al (2012) Factors predicting the outcome of conservatively treated adenocarcinoma in situ of the uterine cervix: an analysis of 166 cases. Gynecol Oncol 124(3):490–495. Scholar
  43. Cox JT, Schiffman M, Solomon D (2003) Prospective follow-up suggests similar risk of subsequent cervical intraepithelial neoplasia grade 2 or 3 among women with cervical intraepithelial neoplasia grade 1 or negative colposcopy and directed biopsy. Am J Obstet Gynecol 188(6):1406–1412PubMedCrossRefPubMedCentralGoogle Scholar
  44. Crocker J, Fox H, Langley FA (1968) Consistency in the histological diagnosis of epithelial abnormalities of the cervix uteri. J Clin Pathol 21:67–70CrossRefGoogle Scholar
  45. Crum CP, Egawa K, Fu YS et al (1983) Atypical immature metaplasia (AIM): a subset of human papillomavirus infection of the cervix. Cancer 51:2214–2219PubMedCrossRefGoogle Scholar
  46. Cullen TS (1900) Cancer of the uterus. Appleton and Company, New YorkGoogle Scholar
  47. Cullimore JE, Luesley DM, Rollason TP, Byrne P, Buckley CH, Anderson M et al (1992) A prospective study of conization of the cervix in the management of cervical intraepithelial glandular neoplasia (CIGN) – a preliminary report. Br J Obstet Gynecol 99:314–318CrossRefGoogle Scholar
  48. Darragh TM, Colgan TJ, Cox JT, Heller DS, Henry MR, Luff RD et al (2012) The lower Anogenital squamous terminology standardization project for HPV-associated lesions: background and consensus recommendations from the College of American Pathologists and the American Society for Colposcopy and Cervical Pathology. J Low Genit Tract Dis 16(3):205–242. Scholar
  49. Datta SD, Koutsky LA, Ratelle S, Unger ER, Shlay J, McClain T et al (2008) Human papillomavirus infection and cervical cytology in women screened for cervical cancer in the United States, 2003–2005. Ann Intern Med 148(7):493–500PubMedCrossRefGoogle Scholar
  50. de Sanjose S, Diaz M, Castellsague X, Clifford G, Bruni L, Munoz N et al (2007) Worldwide prevalence and genotype distribution of cervical human papillomavirus DNA in women with normal cytology: a meta-analysis. Lancet Infect Dis 7(7):453–459PubMedCrossRefGoogle Scholar
  51. de Sanjose S, Quint WG, Alemany L, Geraets DT, Klaustermeier JE, Lloveras B et al (2010) Human papillomavirus genotype attribution in invasive cervical cancer: a retrospective cross-sectional worldwide study. Lancet Oncol 11(11):1048–1056. Scholar
  52. de Sanjose S, Brotons M, Pavon MA (2018) The natural history of human papillomavirus infection. Best Pract Res Clin Obstet Gynaecol 47:2–13. Scholar
  53. de Villiers EM, Fauquet C, Broker TR, Bernard HU, zur Hausen H (2004) Classification of papillomaviruses. Virology 324(1):17–27PubMedCrossRefPubMedCentralGoogle Scholar
  54. Denehy TR, Gregori CA, Breen JL (1997) Endocervical curettage, cone margins, and residual adenocarcinoma in situ of the cervix. Obstet Gynecol 90(1):1–6PubMedCrossRefPubMedCentralGoogle Scholar
  55. Denny LA, Franceschi S, de Sanjose S, Heard I, Moscicki AB, Palefsky J (2012) Human papillomavirus, human immunodeficiency virus and immunosuppression. Vaccine 30(Suppl 5):F168–F174. Scholar
  56. Denslow SA, Rositch AF, Firnhaber C, Ting J, Smith JS (2014) Incidence and progression of cervical lesions in women with HIV: a systematic global review. Int J STD AIDS 25(3):163–177. Scholar
  57. Doorbar J (2007) Papillomavirus life cycle organization and biomarker selection. Dis Markers 23(4):297–313PubMedPubMedCentralCrossRefGoogle Scholar
  58. Doorbar J (2018) Host control of human papillomavirus infection and disease. Best Pract Res Clin Obstet Gynaecol 47:27–41. Scholar
  59. Doorbar J, Quint W, Banks L, Bravo IG, Stoler M, Broker TR et al (2012) The biology and life-cycle of human papillomaviruses. Vaccine 30(Suppl 5):F55–F70. Scholar
  60. Driggers RW, Zahn CM (2008) To ECC or not to ECC: the question remains. Obstet Gynecol Clin North Am 35(4):583–597; viiiPubMedCrossRefGoogle Scholar
  61. Duby JM, DiFurio MJ (2009) Implementation of the ThinPrep imaging system in a tertiary military medical center. Cancer Cytopathol 117(4):264–270CrossRefGoogle Scholar
  62. Duggan MA, Benoit JL, McGregor SE, Inoue M, Nation JG, Stuart GCE (1994) Adenocarcinoma in situ of the endocervix: human papillomavirus determination by dot blot hybridization and polymerase chain reaction amplification. Int J Gynecol Pathol 13(2):143–149PubMedCrossRefPubMedCentralGoogle Scholar
  63. Duska LR (2009) Can we improve the detection of glandular cervical lesions: the role and limitations of the Pap smear diagnosis atypical glandular cells (AGC). Gynecol Oncol 114(3):381–382PubMedCrossRefPubMedCentralGoogle Scholar
  64. Elfgren K, Elfstrom KM, Naucler P, Arnheim-Dahlstrom L, Dillner J (2017) Management of women with human papillomavirus persistence: long-term follow-up of a randomized clinical trial. Am J Obstet Gynecol 216(3):264 e1–264 e7. Scholar
  65. Eversole GM, Moriarty AT, Schwartz MR, Clayton AC, Souers R, Fatheree LA et al (2010) Practices of participants in the college of american pathologists interlaboratory comparison program in cervicovaginal cytology, 2006. Arch Pathol Lab Med 134(3):331–335. Scholar
  66. Franquemont DW, Ward BE, Anderson WA, Crum CP (1989) Prediction of “high-risk” cervical papillomavirus infection by biopsy morphology. Am J Clin Pathol 92(5):577–582PubMedCrossRefPubMedCentralGoogle Scholar
  67. Friedell GH, McKay DG (1953) Adenocarcinoma in situ of endocervix. Cancer 6:887–897PubMedCrossRefPubMedCentralGoogle Scholar
  68. Fu YS, Reagan JW (2002) Pathology of the uterine cervix, vagina, and vulva, 2nd edn. W.B. Saunders Company, PhiladelphiaGoogle Scholar
  69. Fu YS, Reagan JW, Richart RM (1983) Precursors of cervical cancer. Cancer Surv 2:359–382Google Scholar
  70. Fu YS, Huang I, Beaudenon S, Ionesco M, Barrasso R, de Brux J et al (1988) Correlative study of human papillomavirus DNA, histopathology and morphometry in cervical condyloma and intraepithelial neoplasia. Int J Gynecol Pathol 7:297–307PubMedCrossRefPubMedCentralGoogle Scholar
  71. Gage JC, Schiffman M, Hunt WC, Joste N, Ghosh A, Wentzensen N et al (2013) Cervical histopathology variability among laboratories: a population-based statewide investigation. Am J Clin Pathol 139(3):330–335. Scholar
  72. Gatscha RM, Abadi M, Babore S, Chhieng D, Miller MJ, Saigo PE (2001) Smears diagnosed as ASCUS: interobserver variation and follow-up. Diagn Cytopathol 25(2):138–140PubMedCrossRefPubMedCentralGoogle Scholar
  73. Genest DR, Stein L, Cibas E, Sheets E, Zitz JC, Crum CP (1993) A binary (Bethesda) system for classifying cervical cancer precursors: criteria, reproducibility, and viral correlates. Hum Pathol 24(7):730–736PubMedCrossRefPubMedCentralGoogle Scholar
  74. Geng L, Connolly DC, Isacson C, Ronnett BM, Cho KR (1999) Atypical immature metaplasia (AIM) of the cervix: is it related to high-grade squamous intraepithelial lesion (HSIL)? Hum Pathol 30(3):345–351PubMedCrossRefPubMedCentralGoogle Scholar
  75. Gertig DM, Brotherton JM, Budd AC, Drennan K, Chappell G, Saville AM (2013) Impact of a population-based HPV vaccination program on cervical abnormalities: a data linkage study. BMC Med 11:227. Scholar
  76. Gloor E, Hurlimann J (1986a) Cervical intraepithelial glandular neoplasia (adenocarcinoma in situ and glandular dysplasia). A correlative study of 23 cases with histologic grading, histochemical analysis of mucins and immunohistochemical determination of the affinity for four lectins. Cancer 58:1272–1280PubMedCrossRefPubMedCentralGoogle Scholar
  77. Gloor E, Hurlimann J (1986b) Cervical intraepithelial glandular neoplasia (adenocarcinoma in situ and glandular dysplasia). A correlative study of 23 cases with histologic grading, histochemical analysis of mucins, and immunohistochemical determination of the affinity for four lectins. Cancer 58(6):1272–1280PubMedCrossRefPubMedCentralGoogle Scholar
  78. Goldie SJ, Kim JJ, Wright TC (2004) Cost-effectiveness of human papillomavirus DNA testing for cervical cancer screening in women aged 30 years or more. Obstet Gynecol 103(4):619–631PubMedCrossRefPubMedCentralGoogle Scholar
  79. Goldstein NS, Mani A (1998) The status and distance of cone biopsy margins as a predictor of excision adequacy for endocervical adenocarcinoma in situ. Am J Clin Pathol 109(6):727–732PubMedCrossRefPubMedCentralGoogle Scholar
  80. Goldstein NS, Ahmad E, Hussain M, Hankin RC, Perez-Reyes N (1998) Endocervical glandular atypia: does a preneoplastic lesion of adenocarcinoma in situ exist? Am J Clin Pathol 110(2):200–209PubMedCrossRefPubMedCentralGoogle Scholar
  81. Gravitt PE (2011) The known unknowns of HPV natural history. J Clin Invest 121(12):4593–4599. Scholar
  82. Guido RS, Jeronimo J, Schiffman M, Solomon D (2005) The distribution of neoplasia arising on the cervix: results from the ALTS trial. Am J Obstet Gynecol 193(4):1331–1337PubMedCrossRefPubMedCentralGoogle Scholar
  83. Hanselaar AG, Vooijs GP, Oud PS, Pahlplatz MM, Beck JLM (1988) DNA ploidy patterns in cervical intraepithelial neoplasia grade III, with and without synchronous invasive squamous cell carcinoma: measurements in nuclei isolated from paraffin-embedded tissue. Cancer 62:2537–2545PubMedCrossRefPubMedCentralGoogle Scholar
  84. Hariri J, Oster A (2007) The negative predictive value of p16INK4a to assess the outcome of cervical intraepithelial neoplasia 1 in the uterine cervix. Int J Gynecol Pathol 26(3):223–228PubMedCrossRefPubMedCentralGoogle Scholar
  85. Helper TK, Dockerty MB, Randall LM (1952) Primary adenocarcinoma of the cervix. Am J Obstet Gynecol 63:800–808CrossRefGoogle Scholar
  86. Herbst LH, Lenz J, Van Doorslaer K, Chen Z, Stacy BA, Wellehan JF Jr et al (2009) Genomic characterization of two novel reptilian papillomaviruses, Chelonia mydas papillomavirus 1 and Caretta caretta papillomavirus 1. Virology 383(1):131–135. Scholar
  87. Herfs M, Yamamoto Y, Laury A, Wang X, Nucci MR, McLaughlin-Drubin ME et al (2012) A discrete population of squamocolumnar junction cells implicated in the pathogenesis of cervical cancer. Proc Natl Acad Sci U S A 109(26):10516–10521. Scholar
  88. Higgins GD, Phillips GE, Smith LA, Uzelin DM, Burrell CJ (1992) High prevalence of human papillomavirus transcripts in all grades of cervical intraepithelial glandular neoplasia. Cancer 70:136–146PubMedCrossRefPubMedCentralGoogle Scholar
  89. Ho GY, Studentsov YY, Bierman R, Burk RD (2004) Natural history of human papillomavirus type 16 virus-like particle antibodies in young women. Cancer Epidemiol Biomarkers Prev 13(1):110–116PubMedCrossRefPubMedCentralGoogle Scholar
  90. Hoffman MS, Sterghos S Jr, Gordy LW, Gunasekaran S, Cavanagh D (1993) Evaluation of the cervical canal with the endocervical brush. Obstet Gynecol 82(4 Pt 1):573–577PubMedPubMedCentralGoogle Scholar
  91. Hoffmann R, Hirt B, Bechtold V, Beard P, Raj K (2006) Different modes of human papillomavirus DNA replication during maintenance. J Virol 80(9):4431–4439. Scholar
  92. Holl K, Nowakowski AM, Powell N, McCluggage WG, Pirog EC, Collas De Souza S et al (2015) Human papillomavirus prevalence and type-distribution in cervical glandular neoplasias: results from a European multinational epidemiological study. Int J Cancer 137(12):2858–2868. Scholar
  93. Holowaty P, Miller AB, Rohan T, To T (1999) Natural history of dysplasia of the uterine cervix. J Natl Cancer Inst 91(3):252–258PubMedCrossRefPubMedCentralGoogle Scholar
  94. Hopkins MP, Roberts JA, Schmidt RW (1988) Cervical adenocarcinoma in situ. Obstet Gynecol 71:842–844PubMedPubMedCentralGoogle Scholar
  95. Human Papillomaviruses (2007) IARC monographs on the evaluation of carcinogenic risks to humans. IARC, LyonGoogle Scholar
  96. Hurlimann J, Gloor E (1984) Adenocarcinoma in situ and invasive adenocarcinoma of the uterine cervix. An immunohistologic study with antibodies specific for several epithelial markers. Cancer 54(1):103–109PubMedCrossRefPubMedCentralGoogle Scholar
  97. Ioffe OB, Sagae S, Moritani S, Dahmoush L, Chen TT, Silverberg SG (2003) Symposium part 3: should pathologists diagnose endocervical preneoplastic lesions “less than” adenocarcinoma in situ?: point. Int J Gynecol Pathol 22(1):18–21PubMedCrossRefPubMedCentralGoogle Scholar
  98. Isaacson Wechsler E, Wang Q, Roberts I, Pagliarulo E, Jackson D, Untersperger C et al (2012) Reconstruction of human papillomavirus type 16-mediated early-stage neoplasia implicates E6/E7 deregulation and the loss of contact inhibition in neoplastic progression. J Virol 86(11):6358–6364. Scholar
  99. Isacson C, Kessis TD, Hedrick L, Cho KR (1996) Both cell proliferation and apoptosis increase with lesion grade in cervical neoplasia but do not correlate with human papillomavirus type. Cancer Res 56(4):669–674PubMedPubMedCentralGoogle Scholar
  100. Ismail SM, Colelough AB, Dinnen JS, Eakins D, Evans DM, Gradwell E et al (1989) Observer variation in histopathological diagnosis and grading of cervical intraepithelial neoplasia. BMJ 298:707–710PubMedPubMedCentralCrossRefGoogle Scholar
  101. Jakobsen A, Kristensen PB, Poulsen HK (1983) Flow cytometric classification of biopsy specimens from cervical intraepithelial neoplasia. Cytometry 4:166–169PubMedCrossRefGoogle Scholar
  102. Jaworski RC (1990) Endocervical glandular dysplasia, adenocarcinoma in situ, and early invasive (microinvasive) adenocarcinoma of the uterine cervix. Semin Diagn Pathol 7(3):190–204PubMedGoogle Scholar
  103. Jaworski RC, Pacey NR, Greenberg ML, Osborn RA (1988) The histologic diagnosis of adenocarcinoma in situ and related lesions of the cervix uteri. Adenocarcinoma in situ. Cancer 61:1171–1181PubMedCrossRefGoogle Scholar
  104. Jenson AB, Rosenthal JD, Olson C, Pass F, Lancaster WD (1980) Immunologic relatedness of papillomavirus from different species. J Natl Cancer Inst 64:495–500PubMedGoogle Scholar
  105. Jones BA, Davey DD (2000) Quality management in gynecologic cytology using interlaboratory comparison. Arch Pathol Lab Med 124(5):672–681PubMedPubMedCentralGoogle Scholar
  106. Kahn JA (2009) HPV vaccination for the prevention of cervical intraepithelial neoplasia. N Engl J Med 361:273Google Scholar
  107. Kashimura M, Shinohara M, Oikawa K, Hamasaki K, Sato H (1990) An adenocarcinoma in situ of the uterine cervix that developed into invasive adenocarcinoma after 5 years. Gynecol Oncol 36:128–133PubMedCrossRefGoogle Scholar
  108. Katki HA, Schiffman M, Castle PE, Fetterman B, Poitras NE, Lorey T et al (2013a) Five-year risks of CIN 3+ and cervical cancer among women with HPV testing of ASC-US Pap results. J Low Genit Tract Dis 17(5 Suppl 1):S36–S42. Scholar
  109. Katki HA, Schiffman M, Castle PE, Fetterman B, Poitras NE, Lorey T et al (2013b) Five-year risks of CIN 2+ and CIN 3+ among women with HPV-positive and HPV-negative LSIL Pap results. J Low Genit Tract Dis 17(5 Suppl 1):S43–S49. Scholar
  110. Keating JT, Cviko A, Riethdorf S, Riethdorf L, Quade BJ, Sun D et al (2001) Ki-67, cyclin E, and p16INK4 are complimentary surrogate biomarkers for human papilloma virus-related cervical neoplasia. Am J Surg Pathol 25(7):884–891PubMedCrossRefGoogle Scholar
  111. Kelly H, Weiss HA, Benavente Y, de Sanjose S, Mayaud P, ART and HPV Review Group et al (2018) Association of antiretroviral therapy with high-risk human papillomavirus, cervical intraepithelial neoplasia, and invasive cervical cancer in women living with HIV: a systematic review and meta-analysis. Lancet HIV 5(1):e45–e58. Scholar
  112. Kennedy AW, Biscotti CV (2002) Further study of the management of cervical adenocarcinoma in situ. Gynecol Oncol 86(3):361–364PubMedCrossRefPubMedCentralGoogle Scholar
  113. Kim TJ, Kim HS, Park CT, Park IS, Hong SR, Park JS et al (1999) Clinical evaluation of follow-up methods and results of atypical glandular cells of undetermined significance (AGUS) detected on cervicovaginal Pap smears. Gynecol Oncol 73(2):292–298PubMedCrossRefPubMedCentralGoogle Scholar
  114. Kinney WK, Manos MM, Hurley LB, Ransley JE (1998) Where’s the high-grade cervical neoplasia? The importance of minimally abnormal Papanicolaou diagnoses. Obstet Gynecol 91(6):973–976PubMedPubMedCentralGoogle Scholar
  115. Kirkland JA (1963) Atypical epithelial changes in the uterine cervix. J Clin Pathol 16:150–154PubMedPubMedCentralCrossRefGoogle Scholar
  116. Kjaer SK, Frederiksen K, Munk C, Iftner T (2010) Long-term absolute risk of cervical intraepithelial neoplasia grade 3 or worse following human papillomavirus infection: role of persistence. J Natl Cancer Inst 102(19):1478–1488. Scholar
  117. Klaes R, Friedrich T, Spitkovsky D, Ridder R, Rudy W, Petry U et al (2001) Overexpression of p16(INK4A) as a specific marker for dysplastic and neoplastic epithelial cells of the cervix uteri. Int J Cancer 92(2):276–284PubMedCrossRefPubMedCentralGoogle Scholar
  118. Kolstad P, Klem V (1976) Long-term followup of 1121 cases of carcinoma in situ. Obstet Gynecol 48(2):125–129PubMedPubMedCentralGoogle Scholar
  119. Koss LG (1978) Dysplasia. A real concept or a misnomer? Obstet Gynecol 51:374PubMedCrossRefGoogle Scholar
  120. Koss L, Durfee GR (1956) Unusual patterns of squamous epithelium of uterine cervix: cytologic and pathologic study of koilocytotic atypia. Ann N Y Acad Sci 63:1245–1261PubMedCrossRefPubMedCentralGoogle Scholar
  121. Koss LG, Stewart FW, Foote FW, Jordan MJ, Bader GM, Day E (1963) Some histological aspects of behavior of epidermoid carcinoma in situ and related lesions of the uterine cervix. Cancer 16(9):1160–1211PubMedCrossRefGoogle Scholar
  122. Kudo R, Sagai S, Hayakawa O, Ito E, Horimoto E, Hashimoto M (1991) Morphology of adenocarcinoma in situ and microinvasive adenocarcinoma of the uterine cervix. Acta Cytol 35:109–116PubMedPubMedCentralGoogle Scholar
  123. Kulasingam SL, Kim JJ, Lawrence WF, Mandelblatt JS, Myers ER, Schiffman M et al (2006) Cost-effectiveness analysis based on the atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesion Triage Study (ALTS). J Natl Cancer Inst 98(2):92–100PubMedCrossRefGoogle Scholar
  124. Kurman RJ, Ronnett BM, Sherman ME, Wilkinson EJ (2010) Tumors of the cervix, vagina and vulva. AFIP atlas of tumor pathology, vol fourth series. American Registry of Pathology in conjunction with Armed Forces Institute of Pathology, Washington, DCGoogle Scholar
  125. Kurman R, Carcangiu ML, Herrington CS, Young RH (2014) WHO classification of tumors of female reproductive organs. World Health Organization Classification of Tumors. International Agency for Research on Cancer, LyonGoogle Scholar
  126. Leary J, Jaworski R, Houghton R (1991) In-situ hybridization using biotinylated DNA probes to human papillomavirus in adenocarcinoma in-situ and endocervical glandular dysplasia of the uterine cervix. Pathology 23:85–89PubMedCrossRefGoogle Scholar
  127. Lee KR (2003) Symposium part 4: should pathologists diagnose endocervical preneoplastic lesions “less than” adenocarcinoma in situ?: counterpoint. Int J Gynecol Pathol 22(1):22–24PubMedCrossRefGoogle Scholar
  128. Lee KR, Sun D, Crum CP (2000) Endocervical intraepithelial glandular atypia (dysplasia): a histopathologic, human papillomavirus, and MIB-1 analysis of 25 cases. Hum Pathol 31(6):656–664PubMedCrossRefGoogle Scholar
  129. Lee S, Rose MS, Sahasrabuddhe V, Zhao R, Duggan M, Afiero A (2016) Tissue-based immunohistochemical biomarker accuracy in the diagnosis of malignant glandular lesions of the uterine cervix: a systematic review of the literature and meta-analysis. Int J Gynecol Pathol 36:310–322CrossRefGoogle Scholar
  130. Liao SY, Brewer C, Zavada J, Pastorek J, Pastorekova S, Manetta A et al (1994) Identification of the MN antigen as a diagnostic biomarker of cervical intraepithelial squamous and glandular neoplasia and cervical carcinomas. Am J Pathol 145(3):598–609PubMedPubMedCentralGoogle Scholar
  131. Liao SY, Rodgers WH, Kauderer J, Darcy KM, Carter R, Susumu N et al (2013) Endocervical glandular neoplasia associated with lobular endocervical glandular hyperplasia is HPV-independent and correlates with carbonic anhydrase-IX expression: a Gynaecological Oncology Group Study. Br J Cancer 108(3):613–620. Scholar
  132. Liman AK, Giampoli EJ, Bonfiglio TA (2005) Should women with atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion, receive reflex human papillomavirus-DNA testing? Cancer 105(6):457–460PubMedCrossRefPubMedCentralGoogle Scholar
  133. Liu AH, Walker J, Gage JC, Gold MA, Zuna R, Dunn ST et al (2017) Diagnosis of cervical Precancers by Endocervical curettage at colposcopy of women with abnormal cervical cytology. Obstet Gynecol 130(6):1218–1225. Scholar
  134. Lonky NM, Sadeghi M, Tsadik GW, Petitti D (1999) The clinical significance of the poor correlation of cervical dysplasia and cervical malignancy with referral cytologic results. Am J Obstet Gynecol 181(3):560–566PubMedCrossRefPubMedCentralGoogle Scholar
  135. Loureiro J, Oliva E (2014) The spectrum of cervical glandular neoplasia and issues in differential diagnosis. Arch Pathol Lab Med 138(4):453–483. Scholar
  136. Luesley DM, Jordan JA, Woodman CBJ, Watson N, Williams DR, Waddell C (1987) A retrospective review of adenocarcinoma-in-situ and glandular atypia of the uterine cervix. Br J Obstet Gynaecol 94:699–703PubMedCrossRefPubMedCentralGoogle Scholar
  137. Malpica A, Matisic JP, Niekirk DV, Crum CP, Staerkel GA, Yamal JM et al (2005) Kappa statistics to measure interrater and intrarater agreement for 1790 cervical biopsy specimens among twelve pathologists: qualitative histopathologic analysis and methodologic issues. Gynecol Oncol 99(3 Suppl 1):S38–S52. Scholar
  138. Marques T, Andrade LA, Vassallo J (1996) Endocervical tubal metaplasia and adenocarcinoma in situ: role of immunohistochemistry for carcinoembryonic antigen and vimentin in differential diagnosis. Histopathology 28(6):549–550PubMedCrossRefPubMedCentralGoogle Scholar
  139. Massad LS (2006) More is more: improving the sensitivity of colposcopy. Obstet Gynecol 108(2):246–247PubMedCrossRefPubMedCentralGoogle Scholar
  140. Massad LS, Einstein MH, Huh WK, Katki HA, Kinney WK, Schiffman M et al (2013a) 2012 updated consensus guidelines for the management of abnormal cervical cancer screening tests and cancer precursors. Obstet Gynecol 121(4):829–846. Scholar
  141. Massad LS, Einstein MH, Huh WK, Katki HA, Kinney WK, Schiffman M et al (2013b) 2012 updated consensus guidelines for the management of abnormal cervical cancer screening tests and cancer precursors. J Low Genit Tract Dis 17(5 Suppl 1):S1–S27. Scholar
  142. McCann MF, Irwin DE, Walton LA, Hulka BS, Morton JL, Axelrad CM (1992) Nicotine and cotinine in the cervical mucus of smokers, passive smokers, and nonsmokers. Cancer Epidemiol Biomarkers Prev 1:125–129PubMedPubMedCentralGoogle Scholar
  143. McCluggage WG (2007) Immunohistochemistry as a diagnostic aid in cervical pathology. Pathology 39(1):97–111PubMedCrossRefPubMedCentralGoogle Scholar
  144. McCluggage WG, Maxwell P, McBride HA, Hamilton PW, Bharucha H (1995) Monoclonal antibodies Ki-67 and MIB1 in the distinction of tuboendometrial metaplasia from endocervical adenocarcinoma and adenocarcinoma in situ in formalin-fixed material. Int J Gynecol Pathol 14(3):209–216PubMedCrossRefPubMedCentralGoogle Scholar
  145. McCredie MR, Sharples KJ, Paul C, Baranyai J, Medley G, Jones RW et al (2008) Natural history of cervical neoplasia and risk of invasive cancer in women with cervical intraepithelial neoplasia 3: a retrospective cohort study. Lancet Oncol 9(5):425–434PubMedCrossRefPubMedCentralGoogle Scholar
  146. McDonald AC, Tergas AI, Kuhn L, Denny L, Wright TC Jr (2014) Distribution of human papillomavirus genotypes among HIV-positive and HIV-negative women in cape town. S Afr Front Oncol 4:48. Scholar
  147. Meisels A, Fortin R (1976) Condylomatous lesions of the cervix and vagina. I. Cytologic patterns. Acta Cytologica 20:505–509PubMedPubMedCentralGoogle Scholar
  148. Melnikow J, Nuovo J, Willan AR, Chan BK, Howell LP (1998) Natural history of cervical squamous intraepithelial lesions: a meta-analysis. Obstet Gynecol 92(4 Pt 2):727–735PubMedPubMedCentralGoogle Scholar
  149. Mikami Y, McCluggage WG (2013) Endocervical glandular lesions exhibiting gastric differentiation: an emerging spectrum of benign, premalignant, and malignant lesions. Adv Anat Pathol 20(4):227–237. Scholar
  150. Mitchell MF, Tortolero-Luna G, Wright T, Sarkar A, Richards-Kortum R, Hong WK et al (1996) Cervical human papillomavirus infection and intraepithelial neoplasia: a review. J Natl Cancer Inst Monogr 21:17–25Google Scholar
  151. Mittal KR, Chan W, Demopoulos RL (1990) Sensitivity and specificity of various morphological features of cervical condylomas. Arch Pathol Lab Med 114:1038–1041PubMedPubMedCentralGoogle Scholar
  152. Moscicki AB, Hills N, Shiboski S, Powell K, Jay N, Hanson E et al (2001) Risks for incident human papillomavirus infection and low-grade squamous intraepithelial lesion development in young females. J Am Med Assoc 285:2995–3002CrossRefGoogle Scholar
  153. Moscicki AB, Shiboski S, Hills NK, Powell KJ, Jay N, Hanson EN et al (2004) Regression of low-grade squamous intra-epithelial lesions in young women. Lancet 364(9446):1678–1683PubMedCrossRefPubMedCentralGoogle Scholar
  154. Moscicki AB, Schiffman M, Kjaer S, Villa LL (2006) Chapter 5: Updating the natural history of HPV and anogenital cancer. Vaccine 24(Suppl 3):S42–S51CrossRefGoogle Scholar
  155. Moscicki AB, Ma Y, Wibbelsman C, Darragh TM, Powers A, Farhat S et al (2010) Rate of and risks for regression of cervical intraepithelial neoplasia 2 in adolescents and young women. Obstet Gynecol 116(6):1373–1380. 00006250-201012000-00021 [pii]CrossRefPubMedPubMedCentralGoogle Scholar
  156. Moscicki AB, Schiffman M, Burchell A, Albero G, Giuliano AR, Goodman MT et al (2012) Updating the natural history of human papillomavirus and anogenital cancers. Vaccine 30(Suppl 5):F24–F33. Scholar
  157. Mourits MJE, Pieters WJ, Hollema H, Burger M (1992) Three-group metaphase as a morphologic criterion of progressive cervical intraepithelial neoplasia. Am J Obstet Gynecol 167(3):591–595PubMedCrossRefPubMedCentralGoogle Scholar
  158. Munoz N, Castellsague X, De Gonzalez AB, Gissman L (2006) Chapter 1: HPV in the etiology of human cancer. Vaccine 24S3:S1–S10CrossRefGoogle Scholar
  159. Munro A, Codde J, Spilsbury K, Stewart CJ, Steel N, Leung Y et al (2017) Risk of persistent or recurrent neoplasia in conservatively treated women with cervical adenocarcinoma in situ with negative histological margins. Acta Obstet Gynecol Scand 96(4):432–437. Scholar
  160. Muntz HG (1996) Can cervical adenocarcinoma in situ be safely managed by conization alone? Gynecol Oncol 61(3):301–303. Scholar
  161. Murphy N, Ring M, Killalea AG, Uhlmann V, O’Donovan M, Mulcahy F et al (2003) p16INK4A as a marker for cervical dyskaryosis: CIN and cGIN in cervical biopsies and ThinPrep smears. J Clin Pathol 56(1):56–63PubMedPubMedCentralCrossRefGoogle Scholar
  162. Murphy N, Heffron CC, King B, Ganuguapati UG, Ring M, McGuinness E et al (2004) p16INK4A positivity in benign, premalignant and malignant cervical glandular lesions: a potential diagnostic problem. Virchows Arch 445(6):610–615PubMedCrossRefGoogle Scholar
  163. Nasiell K, Nasiell M, Vaclavinkova V (1983) Behavior of moderate cervical dysplasia during long-term follow-up. Obstet Gynecol 61:609–614PubMedGoogle Scholar
  164. Nayar R, Wilbur DC (2015) The Bethesda system for reporting cervical cytology. Definitions, criteria and explanatory notes, 3rd edn. Springer International Publishing, Heidelberg/New York/Dordrecht/LondonGoogle Scholar
  165. Negri G, Egarter-Vigl E, Kasal A, Romano F, Haitel A, Mian C (2003) p16INK4a is a useful marker for the diagnosis of adenocarcinoma of the cervix uteri and its precursors: an immunohistochemical study with immunocytochemical correlations. Am J Surg Pathol 27(2):187–193PubMedCrossRefPubMedCentralGoogle Scholar
  166. Ng A (1993) Glandular diseases of the uterus. In: Keebler CM, Somrak TM (eds) The manual of Cytotechnology. American Society of Clinical Pathologists, ChicagoGoogle Scholar
  167. Ostor AG (1993) Natural history of cervical intraepithelial neoplasia: a critical review. Int J Gynecol Pathol 12:186–192PubMedCrossRefPubMedCentralGoogle Scholar
  168. Ostor AG, Pagano R, Davoren RAM, Fortune DW, Chanen W, Rome R (1984) Adenocarcinoma in situ of the cervix. Int J Gynecol Pathol 3:179–190PubMedCrossRefPubMedCentralGoogle Scholar
  169. Ostor AG, Duncan A, Quinn M, Rome R (2000) Adenocarcinoma in situ of the uterine cervix: an experience with 100 cases. Gynecol Oncol 79(2):207–210PubMedCrossRefPubMedCentralGoogle Scholar
  170. Papillomavirus Episteme [database on the Internet] (2018). Available from: Accessed
  171. Park JJ, Genest DR, Sun D, Crum CP (1999) Atypical immature metaplastic-like proliferations of the cervix: diagnostic reproducibility and viral (HPV) correlates. Hum Pathol 30(10):1161–1165PubMedCrossRefPubMedCentralGoogle Scholar
  172. Park JJ, Sun D, Quade BJ, Flynn C, Sheets EE, Yang A et al (2000) Stratified mucin-producing intraepithelial lesions of the cervix: adenosquamous or columnar cell neoplasia? Am J Surg Pathol 24(10):1414–1419PubMedCrossRefPubMedCentralGoogle Scholar
  173. Parkin DM (1991) Screening for cervix cancer in developing countries. In: Miller AB, Chamberlain J, Day NE, Hakama M, Prorok PC (eds) Cancer screening. Cambridge University Press, Cambridge, pp 184–198Google Scholar
  174. Patten SF (1978) Diagnostic cytopathology of the uterine cervix, 2nd edn. S. Karger, BasalGoogle Scholar
  175. Pedersen K, Burger EA, Sy S, Kristiansen IS, Kim JJ (2016) Cost-effective management of women with minor cervical lesions: revisiting the application of HPV DNA testing. Gynecol Oncol 143(2):326–333. Scholar
  176. Pemberton FA, Smith GV (1929) The early diagnosis and prevention of carcinoma of the cervix: a clinical pathologic study of borderline cases treated at the free hospital for women. Am J Obstet Gynecol 17:165CrossRefGoogle Scholar
  177. Petry KU, Cox JT, Johnson K, Quint W, Ridder R, Sideri M et al (2016) Evaluating HPV-negative CIN2+ in the ATHENA trial. Int J Cancer 138(12):2932–2939. Scholar
  178. Pirog EC, Isacson C, Szabolcs MJ, Kleter B, Quint W, Richart RM (2002) Proliferative activity of benign and neoplastic endocervical epithelium and correlation with HPV DNA detection. Int J Gynecol Pathol 21(1):22–26PubMedCrossRefPubMedCentralGoogle Scholar
  179. Plaxe SC, Saltzstein SL (1999) Estimation of the duration of the preclinical phase of cervical adenocarcinoma suggests that there is ample opportunity for screening. Gynecol Oncol 75(1):55–61PubMedCrossRefPubMedCentralGoogle Scholar
  180. Pretorius RG, Zhang X, Belinson JL, Zhang WH, Ren SD, Bao YP et al (2006) Distribution of cervical intraepithelial neoplasia 2, 3 and cancer on the uterine cervix. J Low Genit Tract Dis 10(1):45–50PubMedCrossRefPubMedCentralGoogle Scholar
  181. Pretorius RG, Belinson JL, Burchette RJ, Hu S, Zhang X, Qiao YL (2011) Regardless of skill, performing more biopsies increases the sensitivity of colposcopy. J Low Genit Tract Dis 15(3):180–188. Scholar
  182. Queiroz C, Silva TC, Alves VA, Villa LL, Costa MC, Travassos AG et al (2006a) P16(INK4a) expression as a potential prognostic marker in cervical pre-neoplastic and neoplastic lesions. Pathol Res Pract 202(2):77–83PubMedCrossRefPubMedCentralGoogle Scholar
  183. Queiroz C, Silva TC, Alves VA, Villa LL, Costa MC, Travassos AG et al (2006b) Comparative study of the expression of cellular cycle proteins in cervical intraepithelial lesions. Pathol Res Pract 202(10):731–737PubMedCrossRefPubMedCentralGoogle Scholar
  184. Quint KD, de Koning MN, van Doorn LJ, Quint WG, Pirog EC (2010) HPV genotyping and HPV16 variant analysis in glandular and squamous neoplastic lesions of the uterine cervix. Gynecol Oncol 117(2):297–301. S0090-8258(10)00131-9 [pii]CrossRefPubMedPubMedCentralGoogle Scholar
  185. Ratnam S, Franco EL, Ferenczy A (2000) Human papillomavirus testing for primary screening of cervical cancer precursors. Cancer Epidemiol Biomarkers Prev 9(9):945–951PubMedPubMedCentralGoogle Scholar
  186. Renshaw AA, Mody DR, Lozano RL, Volk EE, Walsh MK, Davey DD et al (2004) Detection of adenocarcinoma in situ of the cervix in Papanicolaou tests: comparison of diagnostic accuracy with other high-grade lesions. Arch Pathol Lab Med 128(2):153–157PubMedPubMedCentralGoogle Scholar
  187. Reuschenbach M, Wentzensen N, Dijkstra MG, von Knebel Doeberitz M, Arbyn M (2014) p16INK4a immunohistochemistry in cervical biopsy specimens: a systematic review and meta-analysis of the interobserver agreement. Am J Clin Pathol 142(6):767–772. Scholar
  188. Richart RM (1973) Cervical intraepithelial neoplasia: a review. In: Sommers SC (ed) Pathology annual. Appleton-Century-Crofts, East Norwalk, pp 301–328Google Scholar
  189. Richart RM (1990) A modified terminology for cervical intraepithelial neoplasia. Obstet Gynecol 75(1):131–133PubMedPubMedCentralGoogle Scholar
  190. Riethdorf L, Riethdorf S, Lee KR, Cviko A, Loning T, Crum CP (2002) Human papillomaviruses, expression of p16, and early endocervical glandular neoplasia. Hum Pathol 33(9):899–904PubMedCrossRefPubMedCentralGoogle Scholar
  191. Robertson AJ, Anderson JM, Beck JS, Burnett RA, Howatson SR, Lee FD et al (1989) Observer variability in histopathological reporting of cervical biopsy specimens. J Clin Pathol 42(3):231–238PubMedPubMedCentralCrossRefGoogle Scholar
  192. Ronnett BM, Manos MM, Ransley JE, Fetterman BJ, Kinney WK, Hurley LB et al (1999) Atypical glandular cells of undetermined significance (AGUS): cytopathologic features, histopathologic results, and human papillomavirus DNA detection. Hum Pathol 30(7):816–825PubMedCrossRefPubMedCentralGoogle Scholar
  193. Roura E, Travier N, Waterboer T, de Sanjose S, Bosch FX, Pawlita M et al (2016) The influence of hormonal factors on the risk of Developing Cervical Cancer and Pre-Cancer: results from the EPIC Cohort. PLoS One 11(1):e0147029. Scholar
  194. Ruba S, Schoolland M, Allpress S, Sterrett G (2004) Adenocarcinoma in situ of the uterine cervix: screening and diagnostic errors in Papanicolaou smears. Cancer 102(5):280–287. Scholar
  195. Salani R, Puri I, Bristow RE (2009) Adenocarcinoma in situ of the uterine cervix: a metaanalysis of 1278 patients evaluating the predictive value of conization margin status. Am J Obstet Gynecol 200(2):182.e1–182.e5CrossRefGoogle Scholar
  196. Sanati S, Huettner P, Ylagan LR (2010) Role of ProExC: a novel immunoperoxidase marker in the evaluation of dysplastic squamous and glandular lesions in cervical specimens. Int J Gynecol Pathol 29(1):79–87. Scholar
  197. Schiffman M, Castle PE, Jeronimo J, Rodriguez AC, Wacholder S (2007) Human papillomavirus and cervical cancer. Lancet 370(9590):890–907. Scholar
  198. Schiffman M, Wentzensen N, Wacholder S, Kinney W, Gage JC, Castle PE (2011) Human papillomavirus testing in the prevention of cervical cancer. J Natl Cancer Inst 103(5):368–383. Scholar
  199. Schmidt C, Pretorius RG, Bonin M, Hanson L, Semrad N, Watring W (1992) Invasive cervical cancer following cryotherapy for cervical intraepithelial neoplasia or human papillomavirus infection. Obstet Gynecol 80(5):797–800PubMedPubMedCentralGoogle Scholar
  200. SEER Program – National Cancer Institute, USA (2001). Accessed 20 Mar 2003, 2009
  201. Sherman ME, Tabbara SO, Scott DR, Kurman RJ, Glass AG, Manos MM et al (1999) “ASCUS, rule out HSIL”: cytologic features, histologic correlates, and human papillomavirus detection. Mod Pathol 12(4):335–342PubMedPubMedCentralGoogle Scholar
  202. Sherman ME, Schiffman M, Cox JT, ASCUS LSIL Triage Study Group (2002) Effects of age and HPV load on colposcopic triage: data from the ASCUS LSIL Triage Study (ALTS). J Natl Cancer Inst 94:102–107PubMedCrossRefPubMedCentralGoogle Scholar
  203. Shin CH, Schorge JO, Lee KR, Sheets EE (2000) Conservative management of adenocarcinoma in situ of the cervix. Gynecol Oncol 79(1):6–10PubMedCrossRefPubMedCentralGoogle Scholar
  204. Sigurdsson K, Sigvaldason H (2007) Is it rational to start population-based cervical cancer screening at or soon after age 20? Analysis of time trends in preinvasive and invasive diseases. Eur J Cancer 43(4):769–774PubMedCrossRefGoogle Scholar
  205. Silver MI, Gage JC, Schiffman M, Fetterman B, Poitras NE, Lorey T et al (2018) Clinical outcomes after conservative Management of Cervical Intraepithelial Neoplasia Grade 2 (CIN2) in women ages 21–39 years. Cancer Prev Res (Phila) 11(3):165–170. Scholar
  206. Smith JS, Green J, Berrington de Gonzalez A, Appleby P, Peto J, Plummer M et al (2003) Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 361(9364):1159–1167PubMedCrossRefGoogle Scholar
  207. Smith JS, Lindsay L, Hoots B, Keys J, Franceschi S, Winer R et al (2007) Human papillomavirus type distribution in invasive cervical cancer and high-grade cervical lesions: a meta-analysis update. Int J Cancer 121(3):621–632PubMedCrossRefGoogle Scholar
  208. Snijders PJ, Steenbergen RD, Heideman DA, Meijer CJ (2006) HPV-mediated cervical carcinogenesis: concepts and clinical implications. J Pathol 208(2):152–164PubMedCrossRefPubMedCentralGoogle Scholar
  209. Solomon D, Davey D, Kurman R, Moriarty A, O’Connor D, Prey M et al (2002) The 2001 Bethesda system: terminology for reporting results of cervical cytology. JAMA 287(16):2114–2119PubMedCrossRefPubMedCentralGoogle Scholar
  210. Solomon D, Stoler M, Jeronimo J, Khan M, Castle P, Schiffman M (2007) Diagnostic utility of endocervical curettage in women undergoing colposcopy for equivocal or low-grade cytologic abnormalities. Obstet Gynecol 110(2):288–295PubMedCrossRefPubMedCentralGoogle Scholar
  211. Stanley MA (2012) Epithelial cell responses to infection with human papillomavirus. Clin Microbiol Rev 25(2):215–222. Scholar
  212. Stoler MH, Schiffman M (2001) Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL Triage Study. JAMA 285(11):1500–1505PubMedCrossRefGoogle Scholar
  213. Stoler MH, Wright TC Jr, Sharma A, Apple R, Gutekunst K, Wright TL (2011) High-risk human papillomavirus testing in women with ASC-US cytology: results from the ATHENA HPV study. Am J Clin Pathol 135(3):468–475. Scholar
  214. Stoler MH, Wright TC Jr, Cuzick J, Dockter J, Reid JL, Getman D et al (2013) APTIMA HPV assay performance in women with atypical squamous cells of undetermined significance cytology results. Am J Obstet Gynecol 208(2):144.e1–144.e8. Scholar
  215. Stoler MH, Wright TC Jr, Ferenczy A, Ranger-Moore J, Fang Q, Kapadia M et al (2018) Routine use of adjunctive p16 immunohistochemistry improves diagnostic agreement of cervical biopsy interpretation: results from the CERTAIN Study. Am J Surg Pathol. Scholar
  216. Szarewski A, Cuzick J (1998) Smoking and cervical neoplasia; a review of the evidence. J Epidemiol Biostat 3:229Google Scholar
  217. Szarewski A, Maddox P, Royston P, Jarvis M, Anderson M, Guillebaud J et al (2001) The effect of stopping smoking on cervical Langerhans’ cells and lymphocytes. BJOG 108(3):295–303PubMedGoogle Scholar
  218. Tainio K, Athanasiou A, Tikkinen KAO, Aaltonen R, Cardenas J, Hernandes J et al (2018) Clinical course of untreated cervical intraepithelial neoplasia grade 2 under active surveillance: systematic review and meta-analysis. BMJ 360:k499. Scholar
  219. Talia KL, McCluggage WG (2018) The developing spectrum of gastric-type cervical glandular lesions. Pathology 50(2):122–133. Scholar
  220. Talia KL, Stewart CJR, Howitt BE, Nucci MR, McCluggage WG (2017) HPV-negative gastric type adenocarcinoma in situ of the cervix: a Spectrum of rare lesions exhibiting gastric and intestinal differentiation. Am J Surg Pathol 41(8):1023–1033. Scholar
  221. Tase T, Okagaki T, Clark BA, Twiggs LB, Ostrow RS, Faras AJ (1989a) Human papillomavirus DNA in glandular dysplasia and microglandular hyperplasia: presumed precursors of adenocarcinoma of the uterine cervix. Obstet Gynecol 73(6):1005–1008PubMedGoogle Scholar
  222. Tase T, Okagaki T, Clark BA, Twiggs LB, Ostrow RS, Faras AJ (1989b) Human papillomavirus DNA in adenocarcinoma in situ, microinvasive adenocarcinoma of the uterine cervix and coexisting cervical squamous intraepithelial neoplasia. Int J Gynecol Pathol 8(1):8–17PubMedCrossRefGoogle Scholar
  223. Tavassoli FA, Devilee P (eds) (2003) Pathology and genetics of Tumours of the breast and female genital organs. World Health Organization classification of tumors. IARC Press, LyonGoogle Scholar
  224. Tewari D, Novak-Weekley S, Hong C, Aslam S, Behrens CM (2017) Performance of the cobas HPV test for the triage of atypical squamous cells of undetermined significance cytology in cervical specimens collected in SurePath. Am J Clin Pathol 148(5):450–457. Scholar
  225. The (1988) Bethesda system for reporting cervical/vaginal cytologic diagnoses. Developed and approved at a National Cancer Institute Workshop, Bethesda, 12–13, 1988 Dec. 1988;11(5):291–297Google Scholar
  226. Tjalma WA, Fiander A, Reich O, Powell N, Nowakowski AM, Kirschner B et al (2013) Differences in human papillomavirus type distribution in high-grade cervical intraepithelial neoplasia and invasive cervical cancer in Europe. Int J Cancer 132(4):854–867. Scholar
  227. Trimble CL, Piantadosi S, Gravitt P, Ronnett B, Pizer E, Elko A et al (2005) Spontaneous regression of high-grade cervical dysplasia: effects of human papillomavirus type and HLA phenotype. Clin Cancer Res 11(13):4717–4723. Scholar
  228. Tringler B, Gup CJ, Singh M, Groshong S, Shroyer AL, Heinz DE et al (2004) Evaluation of p16INK4a and pRb expression in cervical squamous and glandular neoplasia. Hum Pathol 35(6):689–696PubMedCrossRefGoogle Scholar
  229. Trowell JE (1985) Intestinal metaplasia with argentaffin cells in the uterine cervix. Histopathology 9:561–569CrossRefGoogle Scholar
  230. Tsoumpou I, Arbyn M, Kyrgiou M, Wentzensen N, Koliopoulos G, Martin-Hirsch P et al (2009) p16(INK4a) immunostaining in cytological and histological specimens from the uterine cervix: a systematic review and meta-analysis. Cancer Treat Rev 35(3):210–220PubMedPubMedCentralCrossRefGoogle Scholar
  231. Umezawa T, Umemori M, Horiguchi A, Nomura K, Takahashi H, Yamada K et al (2015) Cytological variations and typical diagnostic features of endocervical adenocarcinoma in situ: a retrospective study of 74 cases. Cytojournal 12:8. Scholar
  232. Ursin G, Pike MC, Preston-Martin S, d’Ablaing G 3rd, Peters RK (1996) Sexual, reproductive, and other risk factors for adenocarcinoma of the cervix: results from a population-based case-control study (California, United States) [see comments]. Cancer Causes Control 7(3):391–401PubMedCrossRefPubMedCentralGoogle Scholar
  233. van der Horst J, Siebers AG, Bulten J, Massuger LF, de Kok IM (2017) Increasing incidence of invasive and in situ cervical adenocarcinoma in the Netherlands during 2004–2013. Cancer Med 6(2):416–423. Scholar
  234. Volgareva G, Zavalishina L, Andreeva Y, Frank G, Krutikova E, Golovina D et al (2004) Protein p16 as a marker of dysplastic and neoplastic alterations in cervical epithelial cells. BMC Cancer 4:58. Scholar
  235. Wang SS, Sherman ME, Hildesheim A, Lacey JV Jr, Devesa S (2004a) Cervical adenocarcinoma and squamous cell carcinoma incidence trends among white women and black women in the United States for 1976–2000. Cancer 100(5):1035–1044PubMedCrossRefPubMedCentralGoogle Scholar
  236. Wang SS, Trunk M, Schiffman M, Herrero R, Sherman ME, Burk RD et al (2004b) Validation of p16INK4a as a marker of oncogenic human papillomavirus infection in cervical biopsies from a population-based cohort in Costa Rica. Cancer Epidemiol Biomarkers Prev 13(8):1355–1360PubMedPubMedCentralGoogle Scholar
  237. Ward BE, Burkett BA, Peterson C, Nichols M, Brennan C, Birch LM et al (1990) Cytological correlates of cervical papillomavirus infection. Int J Gynecol Pathol 9:297–305PubMedCrossRefPubMedCentralGoogle Scholar
  238. Waxman AG, Conageski C, Silver MI, Tedeschi C, Stier EA, Apgar B et al (2017) ASCCP colposcopy standards: how do we perform colposcopy? Implications for establishing standards. J Low Genit Tract Dis 21(4):235–241. Scholar
  239. Williams J (1888) Cancer of the uterus: Harveian lectures for 1886. H.K. Lewis, LondonGoogle Scholar
  240. Winer RL, Lee SK, Hughes JP, Adam DE, Kiviat NB, Koutsky LA (2003) Genital human papillomavirus infection: incidence and risk factors in a cohort of female university students. Am J Epidemiol 157(3):218–226PubMedCrossRefGoogle Scholar
  241. Winer RL, Hughes JP, Feng Q, O’Reilly S, Kiviat NB, Holmes KK et al (2006) Condom use and the risk of genital human papillomavirus infection in young women. N Engl J Med 354(25):2645–2654PubMedCrossRefPubMedCentralGoogle Scholar
  242. Witkiewicz A, Lee KR, Brodsky G, Cviko A, Brodsky J, Crum CP (2005) Superficial (early) endocervical adenocarcinoma in situ: a study of 12 cases and comparison to conventional AIS. Am J Surg Pathol 29(12):1609–1614PubMedCrossRefGoogle Scholar
  243. Wolf JK, Levenback C, Malpica A, Morris M, Burke T, Mitchell MF (1996) Adenocarcinoma in situ of the cervix: significance of cone biopsy margins. Obstet Gynecol 88(1):82–86. Scholar
  244. Woodman CB, Collins SI, Young LS (2007) The natural history of cervical HPV infection: unresolved issues. Nat Rev Cancer 7(1):11–22. Scholar
  245. Workshop NCI (1991) The revised Bethesda System for reporting cervical/vaginal cytologic diagnoses. Report of the 1991 Bethesda workshop. J Am Med Assoc 267:1892Google Scholar
  246. Wright TC Jr (2006) Chapter 3: Pathology of HPV infection at the cytologic and histologic levels: basis for a 2-tiered morphologic classification system. Int J Gynaecol Obstet 94(Suppl 1):S22–S31PubMedCrossRefGoogle Scholar
  247. Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D (2007) 2006 consensus guidelines for the management of women with abnormal cervical screening tests. J Low Genit Tract Dis 11(4):201–222PubMedCrossRefGoogle Scholar
  248. Wright TC Jr, Stoler MH, Behrens CM, Apple R, Derion T, Wright TL (2012) The ATHENA human papillomavirus study: design, methods, and baseline results. Am J Obstet Gynecol 206(1):46 e1–46e11. Scholar
  249. Wright TC, Kuhn L (2006) Immunosuppression and the cervix; human immunodeficiency virus (HIV). In: Jordan JA, Singer A (eds) The cervix. Blackwell, Malden, pp 450–517Google Scholar
  250. Wright TC, Kurman RJ (1994) A critical review of the morphologic classification systems of preinvasive lesions of the cervix: the scientific basis of the paradigm. Papillomavirus Rep 5:175–181Google Scholar
  251. Wright TC, Schiffman M (2003) Adding a test for human papillomavirus DNA to cervical-cancer screening. N Engl J Med 348(6):489–490PubMedCrossRefGoogle Scholar
  252. Wright TC, Stoler MH, Behrens CM, Sharma A, Zhang G, Wright TL (2015) Primary cervical cancer screening with human papillomavirus: end of study results from the ATHENA study using HPV as the first-line screening test. Gynecol Oncol 136(2):189–197. Scholar
  253. Xu L, Verdoodt F, Wentzensen N, Bergeron C, Arbyn M (2016) Triage of ASC-H: a meta-analysis of the accuracy of high-risk HPV testing and other markers to detect cervical precancer. Cancer Cytopathol 124(4):261–272. Scholar
  254. Young RH, Scully RE (1989) Atypical forms of microglandular hyperplasia of the cervix simulating carcinoma. Am J Surg Pathol 13:50–56PubMedCrossRefPubMedCentralGoogle Scholar
  255. Zhang Q, Kuhn L, Denny LA, De Souza M, Taylor S, Wright TC Jr (2007) Impact of utilizing p16INK4A immunohistochemistry on estimated performance of three cervical cancer screening tests. Int J Cancer 120(2):351–356PubMedCrossRefPubMedCentralGoogle Scholar
  256. zur Hausen H (1977) Human papillomaviruses and their possible role in squamous cell carcinomas. Curr Top Microbiol Immunol 78:1–30PubMedPubMedCentralGoogle Scholar
  257. zur Hausen H (2002) Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer 2(5):342–350PubMedCrossRefPubMedCentralGoogle Scholar
  258. zur Hausen H (2009) Papillomaviruses in the causation of human cancers – a brief historical account. Virology 384(2):260–265. S0042-6822(08)00772-1 [pii]CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  • Thomas C. Wright
    • 1
    Email author
  • Brigitte M. Ronnett
    • 2
  • Robert J. Kurman
    • 3
  1. 1.Department of Pathology and Cell BiologyColumbia UniversityNew YorkUSA
  2. 2.Department of Pathology, Division of Gynecologic PathologyJohns Hopkins University School of MedicineBaltimoreUSA
  3. 3.Departments of Gynecology, Obstetrics, Pathology and Oncology, Division of Gynecologic PathologyJohns Hopkins University School of MedicineBaltimoreUSA

Personalised recommendations