Histopathology ofcervical precursor lesions and cancer
S. Lax
KEY WORDS
cervical cancer, precursor lesions, cytological screening, histomorphology
-Abstract
The most frequent types of cervical cancer are squamous-cell carcinoma and adenocarcinoma, which develop from the distinctive precursor lesions cervical intraepithelial neoplasia (CIN) / squamous intraepithelial lesion (SIL), and adenocarcinoma in situ (AIS), respectively. Their tumorigene-sis is HPV-related. High-risk HPV (e.g., types 16 and 18) is integrated into the genome and leads to tumor progression. Cytological screening leads to detection of precursors and their mimics. P16 and Ki-67 immunohistochemistry assists in the histological differential diagnosis of precursors to reactive and metaplastic epithelium. For invasive cervical carcinoma, stage is the strongest prognostic factor. Per definition, microinvasive (pT1a1 / pT1a2) carcinoma is diagnosed histologically on cone biopsies and treated less radically. The distinction between adenocarcinomas of the cervix and endometrial adenocarcinomas is important and can be supported by immunohistochemistry (e.g., ER, p16, CEA, and vimentin) and HPV in-situ hybridization. The rarer adenoid-basal and neuroendocrine carcinomas are less frequently HPV-related.
Introduction
Although invasive cervical carcinoma has become rare in most European countries, from the global perspective it must still be considered a public health burden. In particular, many countries in Africa, southeast Asia, and Latin America reveal an incidence that is more than 10 times as frequent compared to the incidence in central Europe (1). Various factors seem to be responsible for these epidemiological differences, such as socioeconomic standards, immunodeficiency, and HPV infection (2). In particular, because the patho-
genesis has been clearly linked to HPV infection, cervical carcinoma has become apreventable disease.
Histologically, the most frequent type of cervical carcinoma is squamous-cell carcinoma followed by adenocarcinoma (3, 4), of which various subtypes are distinguished (Table 1). Both squamous-cell carcinoma and adenocarcinoma develop through distinctive precursor lesions. For some of the rare types of cervical carcinoma, such as adenoid-cystic, adenoid-basal, and small-cell carcinoma, no precursor lesions are known. The practical value of the precursor lesions
is their presence in cervicovaginal smears and the possibility of early detection by cytological screening. The frequency of precursor lesions has significantly increased in most European countries along with the decrease in cervical carcinoma incidence.
Pathogenesis and histomorphology of squamous-cell carcinoma and its precursor lesions
Our current understanding of the pathogenesis of squamous-cell carcinoma considers it to develop from precursor lesions designated as cervical intraepithelial neoplasia (CIN) (5). CIN is categorized into three grades (CIN1—3) based on the degree of proliferation of atypical basaloid cells (1). The atypical basaloid cell proliferation involves the basal third of the epithelium in CIN1, reaches the middle third in CIN2, and extends to the superficial third in CIN3. A more recent approach based on the Bethesda system for cervical cytology distinguishes between two categories with distinctive biology: low- and high-grade squamous intraepithelial lesions (LSIL and HSIL, respectively). LSIL is characterized by extensive HPV-related cytological changes such as koilocytosis and proliferation of the basal and parabasal cells with mild atypia and mitosis. In contrast, HSIL consists of small to medium-sized atypical basal cells that may involve the entire thickness of the epithelium and it often lacks clearly visible HPV-related cytological changes. If compared to the WHO classification, CIN1 relates to LSIL, whereas CIN2 and 3 are related to HSIL. The various classification schemes are compared in Table 2.
From a biological point of view the dualistic Bethesda approach is reasonable because LSIL and HSIL reveal a different pathogenesis. LSIL is mostly associated with low- or intermediate-risk HPV such as HPV-6 and -11, whereas HSIL harbors clearly oncogenic HPV DNA such as types 16, 18, 31, 33, and 45. There are also fundamental differences between LSIL and HSIL effects at the cellular level (6). LSIL is characterized by infection of terminally differentiated cells that are unable to divide. Therefore the cytological changes involve only the superficial layers of the epithelium. On the other hand, HPV infection in HSIL involves the basal and parabasal cells, which are still capable of dividing. This leads to morphological changes in all or almost all layers of the epithelium. LSIL is typically characterized by low-risk HPV—induced DNA synthesis without accumulation of abnormal DNA, whereas HSIL shows the latter as the consequence of a disrupted cell cycle by high-risk HPV. This leads to aneuploid cells that are
Table 1. WHO classification of malignant tumors of the uterine cervix and their precursors, modified according to (1).
Epithelial tumors
Squamous tumors and precursors Squamous-cell carcinoma, NOS Keratinizing Non-keratinizing Basaloid Verrucous Warty Papillary
Lymphoepithelioma-like Squamotransitional
Early invasive (microinvasive) squamous-cell carcinoma Squamous intraepithelial neoplasia Cervical intraepithelial neoplasia (CIN3) Squamous-cell carcinoma in situ Glandular tumors and precursors Adenocarcinoma Mucinous adenocarcinoma Endocervical Intestinal Signet-ring cell Minimal deviation Villoglandular Endometrioid adenocarcinoma Clear cell adenocarcinoma Serous adenocarcinoma Mesonephric adenocarcinoma Early invasive adenocarcinoma Adenocarcinoma in situ Glandular dysplasia Other epithelial tumors Adenosquamous carcinoma Glassy cell carcinoma variant Adenoid cystic carcinoma Adenoid basal carcinoma Neuroendocrine tumors Carcinoid Atypical carcinoid Small-cell carcinoma Large-cell neuroendocrine carcinoma Undifferentiated carcinoma
able to replicate and survive. The mechanism responsible for this process in HSIL is mainly induced by the viral proteins E6 and E7 and further involves host regulatory proteins such as cyclins, cyclin-dependant kinases, and cyclin-dependant kinase inhibitors. This leads to deregulation of the cell cycle and the apop-totic pathway. Important apoptotic proteins such as p53 and Rb lose their function and others such as pl6 are deregulated. HSIL is further characterized by in-
Table 2: Comparison of different classification systems of precursor lesions of cervical squamous-cell carcinoma.
Traditional classification	WHO classification	Bethesda classification
Mild dysplasia	CIN1	LSIL
Moderate dysplasia	CIN2	HSIL
Severe dysplasia	CIN3	
Carcinoma in situ		
CIN = cervical intraepithelial neoplasia; LSIL = low-grade squamous intraepithelial lesion; HSIL = high-grade squamous intraepithelial lesion.
tegration of the viral DNA into the host genome whereas LSIL shows an episomal location of HPV DNA. LSIL associated with low-risk HPV is usually polyclonal, whereas those associated with high-risk HPV tend to be monoclonal (7). Most HSIL are monoclonal but polyclonality may occur. There is some evidence that polyclonal lesions tend to regress whereas monoclonal lesions show progression (8).
There is evidence that only a subset of CIN1/ LSIL progresses into CIN2 and 3/HSIL because most LSIL have the potential to regress over some years (9). It is unclear whether all CIN3 develop from CIN1 but it has been hypothesized that CIN3 originates "de novo" from metaplastic squamous epithelium under the transition of atypical squamous metaplasia. However, this has not been proven at all. One further problem is that atypical squamous metaplasia cannot be easily distinguished from CIN3 and shows poor interobserver agreement even among experts.
The hallmark of LSIL/CIN1 is moderate-to-marked nuclear atypia on the surface of the epithelium (10). Keeping this in mind helps avoid overdiagnosis. In addition, LSIL is very rare in postmenopausal women. Non-atypical cells with perinuclear halos do not qualify for koilocytosis and are instead designated "pseudokoilocytes." Furthermore, LSIL/CIN1 usually do not harbor a lot of mitosis and abnormal mitotic figures. Therefore, lesions with a high mitotic index must be upgraded to HSIL. With respect to HSIL, its variable histological presentation must be stressed. HSIL may be associated with marked koilocytosis, hyperkeratosis or even metaplastic features. Reflecting the multipotential nature of transformation zone cells, HSIL may even combine squamous and mucinous features.
Differential diagnosis includes metaplastic and reparative processes as well as atrophy (10). Criteria that help on H&E sections are loss of polarity,
distribution of chromatin, mitosis and, in particular, nuclear polymorphism. Furthermore, most non-neoplastic lesions tend to show maturation on the epithelial surface.
Among various biomarkers, pl6 and Ki-67 seem to be useful for differential diagnosis of intraepithelial neoplasia of the cervix (11). P16 overexpression has been linked to continued expression of the viral oncogene E7 due to HPV infection of the epithelium (12). Therefore, a diffuse strong pl6 staining of squamous epithelium points to infection by high-risk HPV and may occur in HSIL and up to 35 to 50% of LSIL (13). The diagnosis of a lesion is further supported by a high Ki-67 labeling index with many Ki-67 positive nuclei within the superficial half of the epithelium. However, it has to be kept in mind that both LSIL/CIN1 and metaplastic epithelium may show focal, patchy staining for pl6.
Invasive squamous-cell carcinoma and the significance of microinvasion
Invasive squamous-cell carcinoma consists of nests and irregular clusters of tumor cells, which may show either a basal-like appearance or maturation, often with keratinization. Keratin formation is considered a sign of good differentiation. Today a sub-classification into keratinizing and non-keratinizing squamous-cell carcinoma is recommended, in particular to avoid confusion of small-cell squamous carcinoma and small-cell carcinoma of neuroendocrine type. Neither histo-pathological grading nor keratinization seems to influence prognosis. The strongest prognostic factor is tumor stage, which is particularly reflected by the issue of microinvasive carcinoma.
Microinvasive carcinoma is defined by size in
Figure 1. Cervical intraepithelial neoplasia 1 (CIN 1) / low-grade intraepithelial lesion (LSIL). The epithelial changes are characterized by significant nuclear atypia in the superficial half due to extensive koilocytosis and proliferation of basal and parabasal cells. The mitotic index is low. HE, 100x.
Figure 2. Cervical intraepithelial neoplasia 3 (CIN 3) / high-grade intraepithelial lesion (HSIL). The epithelium lacks maturation and consists of small highly atypical cells with hyperchromatic nuclei. HE, 100x.
Figure 3. Microinvasive squamous-cell carcinoma of the cervix (pT1a1), diagnosed on a cone biopsy. Small irregular nests of well-differentiated squamous carcinoma invade the cervical stroma from glands (crypts) (arrows). The surface is covered by CIN3 (asterisks). HE, 20x.
Figure 4. Adenocarcinoma in situ (AIS). Endo-cervical glandular epithelium is replaced by pseudostratified atypical epithelium with goblet cells. HE, 200x.
the absence of a clinically visible tumor (14). By definition, microinvasive carcinoma is diagnosed histologically and thus detected through the histopathological analysis of cone biopsies from patients with CIN3. The current FIGO and UICC classification for cervical carcinoma staging (Table 3) defines microinvasive carcinoma by a maximum horizontal dimension of 7 mm and subdivides two categories with a maximum vertical diameter of 3 mm (Ia1) and 5 mm (Ia2), respectively (14,15). The measurement is taken from the
base of the epithelium, either on the surface or within a gland (crypt) from which the tumors originate (16). The subcategorization of microinvasive carcinoma has important therapeutic repercussions because cone biopsy or simple hysterectomy is usually sufficient for pTlal/Ial tumors.
For the histological diagnosis of microinvasive carcinoma of the cervix, penetration of tumor cells through the basement membrane is required. Invasive foci of tumor cells are usually arranged in a haphazard
Table 3. TNM and FIGO classification of cervical carcinoma (14).
pTNM categories (pT = primary tumor)	FIGO stages	Description
PTis	0	Carcinoma in situ (preinvasive)
pT1	I	Cervical carcinoma confined to the uterus
pT1a	IA	Diagnosed only by microscopy
PT1a1	IA1	Depth < 3mm, horizontal spread < 7mm
PT1a2	IA2	Depth < 5mm, horizontal spread < 7mm
pT1b	IB	Clinically visible or microscopic lesion > pT1a2
PT1b1	IB1	Tumor diameter < 4cm
PT1b2	IB2	Tumor diameter > 4cm
pT2	II	Tumor infiltrates beyond the uterus but not to the pelvic wall or to the lower third of the vagina
pT2a	IIA	No parametrial involvement
PT2a1	IIA1	Tumor diameter < 4cm
PT2a2	IIA2	Tumor diameter > 4cm
pT2b	IIB	Infiltration of the parametrium
pT3 and/or N1	III	Tumor infiltrates to the pelvic wall, to the lower third of the vagina or is associated with hydronephrosis
pT3a	IIIA	Lower third of the vagina
pT3b	HIB	Infiltration to the pelvic wall or hydronephrosis
pN1	IIIB	Metastases in pelvic and/or para-aortic lymph nodes
pT4	IVA	Tumor infiltrates mucosa of rectum or urinary bladder or beyond true pelvis
pN - Regional lymph nodes		
pNx		Regional lymph nodes cannot be assessed
pN0		No metastases in regional lymph nodes
pN1		Metastases in regional lymph nodes
pM - Distant metastases		
pMx		Distant metastases cannot be assessed
pMO		No distant metastases
pMl	IVB	Distant metastasis (includes inguinal lymph nodes and intraperitoneal disease; excludes involvement ofvagina, adnexae, and pelvic serosa)
pattern and show irregular margins. They usually display better differentiation than the associated CIN by showing maturation. The use of immunohistochemis-try to demonstrate stromal invasion is of limited value. In particular, disruption of the basement membrane as demonstrated by loss of laminin and collagen IV, respectively, may also occur in normal crypts and CIN.
If microinvasive carcinoma occurs multifocally, the extent of the largest focus is used for classification.
This has been challenged by studies that are based on volumetric measurement of tumor size (17). Because small pTlbl/IBl carcinomas may occur without a clinical visible tumor and be associated with excellent prognosis, expanding the microinvasive carcinoma category has been suggested (18).
Various special types of squamous-cell carcinoma have been described, which are rare and thus of limited clinical value.
Pathogenesis and histomorphology of adenocarcinoma and its precursor lesions
In contrast to squamous-cell carcinoma, the precursor lesion of adenocarcinoma, adenocarcinoma in situ (AIS), is not further subdivided (1). AIS is characterized by cellular atypia similar to colorectal adenoma and may show a variety of cellular differentiation including goblet cells. The nuclei are usually cigar-shaped and pseudostratified, showing coarse chromatin and numerous mitoses. Glandular dysplasia, a lesion with less pronounced changes compared to AIS, has been suggested as a precursor to AIS but has been challenged due to its poor reproducibility and, in particular, its negligible clinical value (19, 20). Recently, a scoring system was proposed to distinguish between glandular dysplasia and AIS but due to its limited clinical value it has been suggested that the term "glandular dysplasia" no longer be used in the clinical setting (21). The term CGIN (cervical glandular intraepithelial neoplasia), which is subdivided into three grades, is used in the UK but not in the U.S. and continental Europe. The differential diagnosis of AIS includes reactive changes of the glandular endo-cervical epithelium and tubal metaplasia. P16 and Ki-67 immunohistochemistry is useful but it needs to be emphasized that tubal metaplasia shows focally strong pl6 immunoreactivity (22).
Cervical adenocarcinoma shows a variety of histological patterns (1). If various histological components are present in one tumor, the classification should be based on the predominant pattern, and the other pattern, if present in at least 10% of the tumor, should just be mentioned in the report (16). The most frequent histological types are the endocervical type, mucinous adenocarcinoma, and endometrioid adenocarcinoma (23). There are divergent reports on the distribution of these two histological types, ranging from a twice-as-frequent incidence of the endocervical type compared to endometrioid adenocarcinoma to a slight predominance of endometrioid adenocarcinoma. More striking is the change in the proportion between adenocarcinoma and squamous-cell carcinoma of the cervix. Cancer registries of several countries have reported a relative increase in the ratios of adenocarcinomas compared to squamous-cell carcinomas. In some countries the incidence of invasive cervical carcinoma decreased from the 1970s to the 1990s by up to one-third, whereas the incidence of ad-enocarcinomaincreasedbyup to 30% (24).
Association with HPV has been found for virtually all types of adenocarcinoma of the cervix, although some data are controversial. Mucinous and endome-
trioid adenocarcinomas frequently harbor HPV DNA 16 (50%) and 18 (40%) and, less frequently, 45 (25%). In contrast to previous findings, HPV DNA was recently also found in adenocarcinoma by using new technologies (26).
Microinvasive adenocarcinoma
A category of microinvasive adenocarcinoma has also been established but, in contrast to its squamous-cell counterpart, the diagnosis is more difficult and has been controversial. Basically all histological types of adenocarcinoma may be found in this category but small pT1a/IA tumors are much rarer compared to the squamous-cell carcinoma group. The most important diagnostic criterion, stromal invasion, is not always unequivocally visible in small glandular lesions of the cervix. In particular, well-differentiated and superficially located glandular lesions may be difficult to diagnose. A marked glandular irregularity with haphazardly arranged glands is considered an indication for an infiltrative growth. Recently, the close relationship of glands to blood vessels was assessed as a diagnostic tool for invasive carcinomas (27). A desmoplastic inflammatory stromal response may be of further help and the presence of lymph vascular invasion confirms the carcinoma diagnosis (1). Less difficult to diagnose is a confluent glandular pattern with loss of stroma and formation of cribriform tumor areas or a complex papillary pattern. The vertical diameter of the tumor is usually measured from the surface of the lesion, reflecting tumor thickness rather than the true depth of invasion (28). The prognosis of microinvasive adeno-carcinomais excellent (29).
Adenosquamous carcinoma
Adenosquamous carcinoma, defined as a tumor with both squamous and glandular differentiation, may account for 5 to 20% of all cervical carcinomas, but its diagnosis varies and is controversial. It was categorized among mixed carcinomas but the recent WHO classification considered it a distinctive type (1). The squamous component may even show kera-tinization but for the diagnosis a sufficient formation of glands must be present. Endometrioid adenocarcinomas with benign-looking squamous elements must not be called adenosquamous carcinomas. It is likely that by using strict diagnostic criteria the number of adenosquamous carcinomas may decrease significantly. Due to similar epidemiologic risk factors and prognosis, adenosquamous carcinoma has recently been related to squamous-cell carcinoma. A strong association with HPV was found (25).
Distinction between cervical and endometrial adenocarcinoma
Determining the site of origin of an adenocarcinoma of the cervix may be difficult, particularly in curettage material. In particular, for endometrioid and endocervical types of adenocarcinomas immunohis-tochemistry may be useful in determining the site of origin. Endometrial adenocarcinomas usually express estrogen receptors (ER) and vimentin but lack CEA expression and usually do not contain HPV DNA. In contrast, endocervical adenocarcinomas are negative for ER and vimentin, show positivity for CEA, and contain HPV DNA. Therefore, a combination of ER, vimentin, and CEA can be applied to determine the site of origin (30). HPV in-situ hybridization can be used as an additional tool (31). The value of pl6 im-munohistochemistry has been challenged, in particular, because non-endometrioid adenocarcinomas of the endometrium, such as mucinous and serous carcinomas, frequently express pl6, as do most endocervical adenocarcinomas. For serous and clear cell carcinomas, determining the site of origin can be very difficult.
references
Rare types of cervical carcinoma
Adenoid-basal carcinoma and neuroendocrine carcinoma are encountered within this group. These tumors are rare and thus of limited clinical significance. An association with HPV has been found for most types, although less frequently compared to squa-mous-cell carcinoma. No distinctive precursor lesions have been found for these tumors. Adenoid-basal carcinoma is slow-growing but lacks a clinically visible tumor (6). Metastases are rare.
Two types of neuroendocrine carcinomas of the uterine cervix are distinguished, small-cell carcinoma and large-cell neuroendocrine carcinoma, which are both rare and associated with poor prognosis (32, 33). Both express neuroendocrine markers, in particular NCAM (CD56) and synaptophysin, less frequently chromogranin A (34, 35). Other neuroendocrine pep-tides such as serotonin may be produced but do not cause endocrine symptoms. The Ki-67 labeling index is usually very high.
1.	Tavassoli FA, Devilee P, editors. Tumours of the breast and female genital organs. Lyon: IARCPress; 2003.
2.	Schiffman MH, Bauer HM, Hoover RN, Glass AG, Cadell DM, Rush BB, et al. Epidemiologic evidence showing that human papillomavirus infection causes most cervical intraepithelial neoplasia. J Natl Cancer Inst. 1993;85(12):958-64.
3.	Brinton LA, Tashima KT, Lehman HF, Levine RS, Mallin K, Savitz DA, et al. Epidemiology of cervical cancer by cell type. Cancer Res. 1987;47(6):1706-11.
4.	Vizcaino AP, Moreno V, Bosch FX, Munoz N, Barros-Dios XM, Borras J, et al. International trends in incidence of cervical cancer: II. Squamous-cell carcinoma. Int J Cancer. 2000;86(3):429-35.
5.	Richart RM. Cervical intraepithelial neoplasia. Pathol Annu. 1973;8:301-28.
6.	Cviko A, Briem B, Granter SR, Pinto AP, Wang TY, Yang YC, et al. Adenoid basal carcinomas of the cervix: a unique morphological evolution with cell cycle correlates. Hum Pathol. 2000;31(6):740-4.
7.	Park TW, Riethdorf S, Schulz G, Riethdorf L, Wright T, Loning T. Clonal expansion and HPV-induced immortalization are early molecular alterations in cervical carcinogenesis. Anticancer Res. 2003;23(lA):155-60.
8.	Ueda Y, Enomoto T, Miyatake T, Ozaki K, Yoshizaki T, Kanao H, et al. Monoclonal expansion with integration of high-risk type human papillomaviruses is an initial step for cervical carcinogenesis: association of clonal status and human papillomavirus infection with clinical outcome in cervical intraepithelial neoplasia. Lablnvest. 2003;83(10):1517-27.
9.	Melnikow J, Nuovo J, Willan AR, Chan BK, Howell LP. Natural history of cervical squamous intraepithelial lesions: a meta-analysis. Obstet Gynecol. 1998;92(4Pt2):727-35.
10.	Wright TC, Ronnett BM, Kurman RJ, Ferenczy A. Precancerous lesions of the cervix. In: Kurman RJ, Ellenson LH, Ronnett BM, editors. Blaustein's Pathology of the Female Genital Tract. 6th ed. New York: Springer Verlag; 2011. p. 194-252.
11.	Tsoumpou I, Arbyn M, Kyrgiou M, Wentzensen N, Koliopoulos G, Martin-Hirsch P, et al. pl6(INK4a) immunostaining in cytological and histological specimens from the uterine cervix: a systematic review and meta-analysis. CancerTreatRev. 2009;35(3):210-20.
12.	Keating JT, Cviko A, Riethdorf S, Riethdorf L, Quade BJ, Sun D, et al. Ki-67, cyclin E, and pl6INK4 are complimentary surrogate biomarkers for human papilloma virus-related cervical neoplasia. Am J Surg Pathol. 2001;25(7):884-91.
13.	Klaes R, Friedrich T, Spitkovsky D, Ridder R, Rudy W, Petry U, et al. Overexpression of pl6(INK4A) as a specific marker for dysplastic and neoplastic epithelial cells of the cervix uteri. Int J Cancer. 2001;92(2):276-84.
14.	Sobin LH, Gospodarowicz M, Wittekind C. TNM Classification of Malignant Tumours. 7th ed. Oxford: Wiley-Blackwell, 2009.
15.	Pecorelli S, Zigliani L, Odicino F. Revised FIGO staging for carcinoma of the cervix. Int J Gynaecol Obstet. 2009;105(2):107-8.
16.	Wright TC, Ferenczy A, Kurman RJ. Carcinoma and other tumors of the cervix. In: Kurman RJ, editor. Blaustein's Pathology of the Female Genital Tract. 5th ed. New York: Springer Verlag, 2002. p. 325-81.
17.	Reich 0, Pickel H. Multifocal stromal invasion in microinvasive squamous cell carcinoma of the cervix: how to measure and stage these lesions. Int J Gynecol Pathol. 2002;21(4):416-7.
18.	Girardi F, Burghardt E, Pickel H. Small FIGO stage IB cervical cancer. Gynecol Oncol. 1994;55(3 Pt l):427-32.
19.	Wright TC, Kurman RJ, Ferenczy A. Precancerous lesions of the cervix. In: Kurman RJ, editor. Blaustein's Pathology ofthe Female Genital Tract. 5th ed. New York: Springer Verlag; 2002. p. 253-324.
20.	Goldstein NS, Ahmad E, Hussain M, Hankin RC, Perez-Reyes N. Endocervical glandular atypia: does a preneoplastic lesion of adenocarcinoma in situ exist? Am J Clin Pathol. 1998;110(2):200-9.
21.	Ioffe OB, Sagae S, Moritani S, Dahmoush L, Chen TT, Silverberg SG. Proposal of a new scoring scheme for the diagnosis of noninvasive endocervical glandular lesions. Am J Surg Pathol. 2003;27(4):452-60.
22.	McCluggage WG. Immunohistochemistry as a diagnostic aid in cervical pathology. Pathology. 2007;39(1):97-111.
23.	Young RH, Clement PB. Endocervical adenocarcinoma and its variants: their morphology and differential diagnosis. Histopathology. 2002;41(3):185-207.
24.	Vizcaino AP, Moreno V, Bosch FX, Munoz N, Barros-Dios XM, Parkin DM. International trends in the incidence of cervical cancer: I. Adenocarcinoma and adenosquamous cell carcinomas. Int J Cancer. 1998;75(4):536-45.
25.	Pirog EC, KleterB, Olgac S, Bobkiewicz P, Lindeman J, Quint WG, et al. Prevalence of human papillomavirus DNA in different histological subtypes of cervical adenocarcinoma. Am J Pathol. 2000;157(4):1055-62.
26.	Nofech-Mozes S, Khalifa MM, Ismiil N, Dube V, Saad RS, Sun P, et al. Detection ofHPV-DNA by a PCR-based method in formalin-fixed, paraffin-embedded tissue from rare endocervical carcinoma types. Appl Immunohistochem Mol Morphol. 2010;18(l):80-5.
27.	Wheeler DT, Kurman RJ. The relationship of glands to thick-wall blood vessels as a marker of invasion in endocervical adenocarcinoma. Int J Gynecol Pathol. 2005;24(2):125-30.
28.	Ostor AG. Early invasive adenocarcinoma of the uterine cervix. Int J Gynecol Pathol. 2000;19(l):29-38.
29.	Ostor A, Rome R, Quinn M. Microinvasive adenocarcinoma of the cervix: a clinicopathologic study of 77 women. Obstet Gynecol. 1997;89(l):88-93.
30.	Alkushi A, Irving J, Hsu F, Dupuis B, Liu CL, Rijn M, et al. Immunoprofile of cervical and endometrial adenocarcinomas using a tissue microarray. Virchows Arch. 2003;442(3):271-7.
31.	Staebler A, Sherman ME, Zaino RJ, Ronnett BM. Hormone receptor immunohistochemistry and human papillomavirus in situ hybridization are useful for distinguishing endocervical and endometrial adenocarcinomas. Am J Surg Pathol. 2002;26(8):998-1006.
32.	Abeler VM, Holm R, Nesland JM, Kjorstad KE. Small cell carcinoma of the cervix. A clinicopathologic study of 26 patients. Cancer. 1994;73(3):672-7.
33.	Gersell DJ, Mazoujian G, Mutch DG, Rudloff MA. Small-cell undifferentiated carcinoma of the cervix. A clinicopathologic, ultrastructural, and immunocytochemical study of 15 cases. Am J Surg Pathol. 1988;12(9):684-98.
34.	Conner MG, Richter H, Moran CA, Hameed A, Albores-Saavedra J. Small cell carcinoma of the cervix: a clinicopathologic and immunohistochemical study of 23 cases. Ann Diagn Pathol. 2002;6(6):345-8.
35.	Ishida GM, Kato N, Hayasaka T, Saito M, Kobayashi H, Katayama Y, et al. Small cell neuroendocrine carcinomas of the uterine cervix: a histological, immunohistochemical, and molecular genetic study. Int J Gynecol Pathol. 2004;23(4):366-72.
AUTHOR'S Sigurd Lax, PhD, Department ofPathology, Graz West General Hospital, ADDRESS Göstingerstrasse 22, A-8020 Graz, E-mail: sigurd.lax@medunigraz.at,
sigurd.lax@lkh-grazwest.at