and David C. Wilbur1
(1)
Department of Pathology, Massachusetts General Hospital, Boston, MA, USA
Keywords
EndocervicalAdenocarcinomaAdenocarcinoma in situUsual typeGastric-typeVilloglandularClear cellSerousMetanephricMetaplasiaEctopiaEndocervicosisEndometriosisArias-StellaHyperplasiaCorrect cytological interpretation of glandular lesions of the cervix has challenged cytologists for years. Cervical cytology or the Pap test was initially intended to screen for squamous carcinoma and precursor squamous lesions but with time it became evident that endocervical glandular lesions were recognizable as well. The development of cytologic criteria for the recognition of endocervical lesions lagged behind that of squamous lesions, probably in part because endocervical adenocarcinoma is less common than squamous carcinoma but also because the histologic criteria were defined at a later date. For example, in North America cytologic criteria for endocervical adenocarcinoma in situ (AIS) were only sufficiently delineated, found to be reproducible, and therefore accepted by the cytology community as late as 2004, when they were first included in the second edition of the Bethesda System manual [1, 2]. Lesser degrees of glandular dysplasia or precursor glandular lesions on cervical cytology have yet to be precisely defined either cytologically or histologically. Certainly, the early cytologic emphasis was on the more common squamous lesions; however, additional factors delayed progress with glandular neoplasia. Using early sampling devices, such as fiber swabs, far less cellular material was available for analysis from the upper portions of the endocervical canal. Therefore less cellular material from these areas led to lower sensitivities for these analyses, and less recognition and study of even benign and reactive processes were exhibited by the epithelium of the endocervical canal. This undoubtedly led to lower levels of interest, and hence less progress. The literature on glandular lesions blossomed with the advent of better sampling devices such as brooms and brushes in the late 1980s and 1990s. To some degree this increased attention became absolutely necessary as the cytologist needed to deal with the large volumes of endocervical canal-derived material and its many presentations as will be detailed in this monograph.
Prevalence
Over the last half century or more, adenocarcinoma has made up an increasing proportion of invasive cervical cancers. A study of cervical carcinoma from 1936 to 1967 at the hospital of the University of Pennsylvania found 6.1 % of cases were adenocarcinoma [3]. In several more recent studies of cervical cancer, the rate of adenocarcinoma ranged from 20 to 25 % [4, 5].
Studies of the data collected by the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program in the United States on cervical cancer incidence from 1973 to as recently as 2008 have shown that the incidence of cervical adenocarcinoma is on the rise but the rise is dwarfed by the overall decline in incidence of invasive cervical cancer due to the decreased incidence of squamous cell carcinoma [6–9]. One study analyzing the data from 1973 to 1996 found that the age-adjusted incidence rates of squamous cell carcinoma declined by 41.9 % while the age-adjusted incidence rate of adenocarcinoma increased by 29.1 % and that the proportion of adenocarcinoma relative to squamous cell carcinoma increased by 95.2 % [6]. Analysis of SEER data for years 1973–1989 showed a 14.0 % incidence of adenocarcinoma as compared to 24.2 % for the years 1990–2008 resulting in a 1.95 times higher odds ratio of a newly diagnosed cervical cancer being adenocarcinoma during the later period as compared to the earlier period (Table 1.1) [7]. Likewise an increased incidence of cervical adenocarcinoma has been noted in Europe [4, 10]. The rise in incidence of cervical adenocarcinoma including adenosquamous carcinoma has been particularly noted in younger women in countries with well-established screening programs [4, 5, 8, 11]. Interestingly, the most recent data suggest that levels are universally plateauing or even declining in the later years of data, suggesting improved detection (or possibly changes in epidemiologic drivers such as hormone use) may be having an effect on disease prevalence [12].
Table 1.1
Incidence of cervical cancer over 35 years in the United States using the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program [7]
1973–1989 | 1990–2008 | Change per 100,000 women | ||||
|---|---|---|---|---|---|---|
N (%) | 95 % CI | N (%) | 95 % CI | EAPC | 95 % CI | |
Total | 19,907 (100.0) | – | 20, 456 (100.0) | – | −2.3 | −2.5 to −2.1 |
SCC | 17,113 (86.0) | 85.5–86.4 | 15,504 (75.8) | 75.2–76.4 | −3.0 | −3.2 to −2.1 |
ACA | 2,797 (14.0) | 13.6–14.5 | 4,952 (24.2) | 23.6–24.8 | 0.6 | 0.2–1.0 |
Racial and ethnic differences in the incidence of cervical adenocarcinoma have been identified. Analysis of SEER data from 1976 to 2000 showed that the incidence of cervical adenocarcinoma rose linearly with age in African-American women while the incidence plateaued in white women [8]. Recent analysis of SEER data (1998–2002) has demonstrated higher rates of adenocarcinoma in white women than in African-American women and in Hispanic women as compared to non-Hispanic women [13].
The rate of precursor squamous lesions far outweighs that of glandular lesions of the cervix in most series [14–17]. The SEER registry kept track of in situ carcinomas separately from invasive carcinomas from 1976 to 1995; thereafter, in situ and invasive carcinomas were grouped as one category. During that time period, of the 149,178 women with cervical carcinoma, 96 % were squamous lesions and 4 % were glandular lesions; of the total, 121,793 (82 %) were in situ lesions. Of the in situ lesions, 99 % were squamous cell carcinoma in situ and only 1 % were AIS [18]. The data also showed that AIS increased at a steady rate in both white and African-American women during this period with the pace being greatest for white women [8]. Unlike squamous cell carcinoma, the incidence of invasive adenocarcinoma is greater than that of AIS [18]. Studies have suggested that screening is less effective in detection of endocervical adenocarcinoma [19, 20]. Wang et al., in an analysis of SEER data from 1976 to 1995, noted in white women that the in situ-to-invasive ratio for squamous cell carcinoma rose from 2.25 to 6.69 (threefold) while that for adenocarcinoma rose from 0.2 to 0.8 (fourfold) [8]. In both tumor types, the reason for these increases of in situ versus invasive tumors is almost certainly better detection as well as potentially an increased prevalence. In a retrospective study of cases from large Australian cytology laboratories, the sensitivity for detection of AIS in cytology samples was found to be up to 54.3 % [21].
Etiology of Endocervical Adenocarcinoma and AIS
Almost all cervical squamous cell carcinomas are associated with high risk human papillomavirus (hrHPV) [22–24]. Endocervical adenocarcinoma is also associated with hrHPV but the prevalence is somewhat lower, approximately 90 % [25–28]. Endocervical adenocarcinomas positive for hrHPV are predominantly of the usual type which make up about 80 % of all endocervical adenocarcinomas [29]. The presumed precursor lesion of the usual type, AIS is positive for hrHPV in nearly 100 % of cases [28]. Although less commonly found in squamous lesions of the cervix, HPV18 and its associated type 45 are identified in a substantial number of in situ and invasive adenocarcinomas. HPV16 is also very commonly identified, but is not in the majority as is noted in squamous lesions [17, 25, 28, 30, 31]. Recently the Asian American variant of HPV16 has been preferentially associated with glandular neoplasia. It has been suggested that this variant is more prone to cause effects in the estrogen-associated areas of the host genome in line with the data that shows that hormone use may be more commonly associated with glandular carcinogenesis [28, 30–35]. The type of HPV present has not been found to impact survival in women with endocervical adenocarcinoma [31].
The epidemiologic risk factors for the development of the most common forms of endocervical adenocarcinoma are similar to squamous cell carcinoma and are those associated with sexually transmitted disease: use of oral contraceptives, multiple sexual partners, and young age at first intercourse [23]. Unlike squamous cell carcinoma, cigarette smoking is not associated with increased risk of endocervical adenocarcinoma. Unlike squamous cell carcinoma, endocervical adenocarcinoma appears to be less commonly associated with high parity but has a greater association with obesity (which is a known cause of increased endogenous estrogen production) [32]. Exogenous hormone use also has been shown to be a factor in the development of glandular cancers [11, 20, 36].
Clinical Features of Endocervical Adenocarcinoma and AIS
In women with early stage tumors, endocervical adenocarcinoma may be asymptomatic or may present with abnormal vaginal bleeding (irregular, heavy, or postcoital) [33, 37, 38]. More advanced tumors may present with pelvic or low back pain. Depending on the type of adenocarcinoma present, up to 50 % of patients will have a visible tumor mass on physical exam [33]. Deeply infiltrative tumors can cause a diffusely enlarged, barrel-shaped cervix. The absence of a grossly identifiable tumor may correspond to an early stage tumor or if higher stage, to a deeply invasive tumor high in the endocervical canal [38].
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
