- Email: [email protected]
Cytological screening of endocervical adenocarcinoma
Annales de pathologie (2012) 32, e8—e14
Disponible en ligne sur
www.sciencedirect.com
REVIEW
Cytological screening of endocervical adenocarcinoma Le dépistage cytologique de l’adénocarcinome du col Luigi Di Bonito a, Christine Bergeron b,∗ a
Direttore dell’ UCO, Anatomia e Istologia patologica, Ospedale Universitario di Cattinara, Trieste, Italy b Laboratoire Cerba, boîte postale, 95066 Cergy Pontoise cedex 9, France Accepted for publication on 13 September 2012 Available online 22 November 2012
KEYWORDS Cervical screening; Adenocarcinoma; Cytology; Bethesda terminology
MOTS CLÉS Dépistage du cancer col ; Adénocarcinome ;
∗
Summary Invasive endocervical adenocarcinoma represents on average 15% of cervical carcinomas and it is associated with the human papillomavirus infection high risk types 16 and 18 in most cases. Its detection has some special features compared to squamous cell carcinoma; glandular precancerous lesions are less known and only adenocarcinoma in situ is diagnosed by consensus among pathologists; adenocarcinoma in situ develops in the squamocolumnar junction by reserve cells but it is hard to be located by colposcopy in the endocervical canal or in the deep glandular recess. Sampling of endocervical cells requires brushes rather than an Ayre spatula. Cytological diagnosis of glandular cells abnormalities is based on the Bethesda System 2001 terminology which redefined endocervical cells abnormalities and also introduced the entity of adenocarcinoma in situ. This entity is characterized by specific morphological features, such as the radial arrangement of nuclei in the periphery, like ‘‘at the end of the feathers of a bird’s wing’’ (feathering of cells), images of nuclei palissading or rosette without tumoral diathesis. Glandular cells abnormalities are rare and represent less than 0.1% of all smears and less than 5% of abnormal smears. By improving the collection and the interpretation of abnormal endocervical cells, cytological screening should allow the diagnosis of in situ adenocarcinoma and detection of invasive adenocarcinoma at a very early stage. This will lead to a decrease in mortality from endocervical adenocarcinoma, especially in young women. © 2012 Elsevier Masson SAS. All rights reserved.
Résumé L’adénocarcinome invasif du col utérin correspond en moyenne à 15 % des carcinomes du col utérin. Il est le plus souvent associé à une infection à papillomavirus humain de type 16 ou 18. Son dépistage présente des particularités par rapport au carcinome malpighien ; les lésions glandulaires précancéreuses sont moins bien connues et seul l’adénocarcinome in situ est diagnostiqué de manière consensuelle par les pathologistes ; l’adénocarcinome in situ se
DOI of original article: http://dx.doi.org/10.1016/j.annpat.2012.09.199. Corresponding author. E-mail address: [email protected] (C. Bergeron).
0242-6498/$ — see front matter © 2012 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.annpat.2012.09.230
Cervical adenocarcinoma and cytopathology
Cytologie ; Terminologie Bethesda 2001
e9
développe autour de la jonction squamocylindrique à partir de cellules de réserve et peut être difficile à localiser en colposcopie dans le canal endocervical. Le prélèvement des cellules cylindriques nécessite une brosse plutôt que la spatule d’Ayre. Le diagnostic cytologique des cellules glandulaires se fait selon la terminologie Bethesda 2001 qui a redéfini le cadre des anomalies des cellules glandulaires et a introduit l’entité « adénocarcinome in situ ». Cette entité est caractérisée par des critères spécifiques, comme la disposition radiaire des noyaux en périphérie, donnant aux cellules un aspect ressemblant « à des plumes à l’extrémité d’une aile d’oiseau », des images de noyaux en palissade ou en rosette sans diathèse tumorale. Les anomalies des cellules glandulaires restent rares et correspondent à moins de 0,1 % de la totalité des frottis et moins de 5 % des frottis anormaux. En améliorant le prélèvement et l’interprétation des cellules cylindriques anormales, le dépistage cytologique devrait permettre le diagnostic d’adénocarcinome in situ et d’adénocarcinome invasif à un stade précoce. Une diminution de la mortalité par adénocarcinome du col, en particulier chez les femmes jeunes, devrait en résulter. © 2012 Elsevier Masson SAS. Tous droits réservés.
Introduction Endocervical adenocarcinoma’s incidence remains stable or is slightly increasing in 2011, including those countries where a parallel decrease in squamous cancers’ incidence is observed [1,2]. In some countries, like England or Australia, where there is a well-organized screening, invasive adenocarcinoma makes up almost one fourth of all invasive cervical cancers [3]. This increase has been particularly striking in women under 40 [1,4]. Over the last 20 years, invasive cervical adenocarcinoma’s mortality has also increased in absolute and relative terms compared to squamous cell carcinoma [5]. Should we conclude that cytological screening is not suitable for detecting precancerous glandular lesions or invasive adenocarcinomas at a very early stage?
Precancerous lesions of the endocervical cylindrical epithelium Adenocarcinoma in situ is the only histological recognized entity by consensus as a preinvasive lesion of the cervix [6] (Fig. 1). The adenocarcinoma in situ is sometimes referred to as high-grade intraepithelial glandular lesion or dysplasia. Progression from adenocarcinoma in situ to invasive adenocarcinoma is estimated to last between 5 and 13 years [7,8]. The adenocarcinoma in situ arises at the squamocolumnar junction after persistent infection with high-risk human papillomavirus (HPV) infection, types 16 and 18 being the most common [9]. Persistent HPV16 and 18 infections are usually associated with cervical squamous intraepithelial neoplasia (CIN 2—3) starting at the junction because these viruses have a higher affinity with keratinocytes and they can more easily reach the basal layers of the squamous epithelium where the viral cycle starts. HPV can also infect at the same time or, more rarely, separately reserve cells around the squamocolumnar junction which may differentiate towards glandular columnar cells [10]. The infection usually remains latent and in case of persistence it may allow the expression of viral oncogenes E6 and E7 in the columnar epithelium causing a transforming infection. Immunohistochemical expression of p16 is an indirect way of highlighting the transforming infection in the endocervical columnar epithelium [11]. Morphologically, the transforming infection results in adenocarcinoma in situ,
Figure 1. Adenocarcinoma in situ on a biopsy. A. The abnormal epithelium affects only a part of the endocervical glands. B. The abnormal epithelium reaches the surface of endocervical glandular structures which can be reached by a brush. Adénocarcinome in situ sur une biopsie. A. L’épithélium cylindrique anormal n’intéresse qu’une partie des glandes endocervicales. B. L’épithélium cylindrique anormal atteint la surface des récessus endocervicaux. Une brosse peut donc prélever les cellules glandulaires anormales.
e10
L. Di Bonito, C. Bergeron
Figure 2. The detection of p16 protein by immunohistochemistry showed a diffuse staining of all abnormal cells. La détection de la protéine p16 par immunohistochimie montre un marquage diffus sur l’ensemble des cellules anormales.
with diffuse atypical endocervical cells immunopositivity for p16 [12] (Fig. 2). This marker may help in the differential diagnosis of glandular hyperplasia and its variants such as glandular intraepithelial low-grade lesions. These have no reproducible morphological criteria and in most of cases they are not associated with an HPV infection. They are negative for p16 immunohistochemical stain. Tubal metaplasia is another differential diagnosis. In these cases, p16 stains only ciliated cells (Fig. 3).
Cytological screening Sampling of the endocervical epithelium Cytological sampling can be performed both for conventional smears using the cytobrush and for liquid-based cytology using the cervexbrush and its variants. Improvement of cytology brushes allows better sampling of the endocervical epithelium. Endocervical cells sampling can be difficult because adenocarcinoma in situ may have a multifocal development and endocervical glands can be partially or totally involved. If abnormal endocervical cells do not reach the superficial part of the glandular structures, they may not shed on the smears. This explains why sometimes adenocarcinoma in situ is diagnosed late on the cone biopsy [10,12].
Terminology for abnormal endocervical cells To ensure better communication among clinicians involved in cervical cancer screening, the Terminology Bethesda System (TBS) 1988 proposed diagnostic categories for reporting of cervical smear. The TBS was revised in 2001, especially for cervical glandular cell abnormalities [13,14]. The TBS provides the following categories.
Atypical glandular cells (AGC) of the endocervix, endometrium, or glandular not otherwise specified (NOS) Site specifying allows clinician to have different managements according to the patient age.
Figure 3. A. Tubal metaplasia presents high columnar and ciliated cells. B. The detection of p16 protein by immunohistochemistry highlights the presence of ciliated cells. A. La métaplasie tubaire présente des cellules cylindriques hautes de type intercalaire et des cellules ciliées. B. La détection de la protéine p16 par immunohistochimie souligne la présence des cellules ciliées.
The diagnosis of atypical glandular cells is made in presence of slight atypias (nuclear enlargement without chromatin abnormality) and when columnar cells shed either isolates or in clusters. The main differential diagnosis included in this group are endocervical cell changes associated with the presence of a polyp, endocervical hyperplasia, reparative changes occurring after conization, and tubal metaplasia, especially when cells lose their upper ciliated part. When the columnar cells are of an endometrial origin the differential diagnosis regards reparative or metaplastic changes of the uterine isthmus or of endometrium. After menopause, it is preferable to report the presence of endometrial cells without specifying their nature (normal/abnormal). In this case an endometrial biopsy has to be performed.
Atypical glandular cells, favor neoplastic This diagnosis is suggested when the morphological abnormalities of cylindrical cells are more severe but quantitatively and qualitatively fall short of definitive adenocarcinoma diagnosis. The differential diagnosis of these anomalies includes cellular changes associated with the
Cervical adenocarcinoma and cytopathology
e11
Figure 4. In presence of an intrauterine device, the smear shows three-dimensional clusters of columnar cells with a regular but slightly enlarged nucleus. Such changes should not be confused with glandular atypia. Le frottis fait en présence d’un stérilet comporte des amas tridimensionnels de cellules cylindriques avec un noyau régulier mais légèrement augmenté de volume. Ce type de modifications ne doit pas être confondu avec des atypies glandulaires.
Figure 6. Adenocarcinoma in situ on liquid-based smear (ThinPrep® ). The abnormal cells arranged in rosettes around a core. The background is clean. Adénocarcinome in situ (AIS) sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales ont une disposition en rosettes autour d’une lumière. Le fond est propre.
presence of an intrauterine device (Fig. 4) or due to radiotherapy and chemotherapy.
distributed chromatin but absence of prominent nucleoli and tumor diathesis.
The adenocarcinoma in situ (AIS) was proposed for the first time in the TBS 2001
The invasive adenocarcinoma
This entity is characterized by specific morphological features, such as the radial arrangement of nuclei in the periphery, like ‘‘at the end of the feathers of a bird’s wing’’ (feathering of cells), images of nuclei palissading or rosette on a clean background without evidence of necrosis (Figs. 5 and 6). Abnormal cells display increased nucleo-cytoplasmic ratio, with slightly altered and coarsely
Figure 5. Adenocarcinoma in situ on liquid-based smear (ThinPrep® ). The abnormal cells have a cylindrical palisad arrangement of the nuclei and tapered cytoplasm like bird’s feathers. The background is clean and lacks of necrosis. Adénocarcinome in situ (AIS) sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales ont une disposition en palissade du noyau et un cytoplasme effilé en plumes d’oiseau. Le fond est propre sans nécrose.
The invasive adenocarcinoma is characterized by the presence of atypical glandular cells showing nuclear changes comprising enlargement, hyperchromasia, coarse granulation of chromatin and large, sometimes irregular multiple nucleoli. The abnormal cells often form spherical or oval clusters and the smear background often shows blood, necrosis, and cell debris. The TBS 2001 asks, if possible, to specify if the adenocarcinoma is endocervical, endometrial or extra-uterine. The criteria which point to an endocervical origin are abundant shedding of large columnar cells (Fig. 7). Endometrial origin is suggested when there is less shedding of abnormal glandular cells which are smaller and round shaped with a centrally positioned nucleus (Fig. 8) and tend to arrange in morules. The extra-uterine origin is rare but can be observed in rectal cancers infiltrating the vaginal wall, in ovarian cancers shedding through the tubal lumen or in the breast cancer metastasis. All these findings are rare if compared to squamous abnormalities [15,16]. As described by the Centre de regroupement informatique et statistique en anatomie pathologique en Île-de-France (CRISAP-IDF), out of 250,000 smears, glandular abnormalities are rare and represent less than 0.1% of all smears and less than 5% of abnormal smears [15]. These percentages do not distinguish atypical glandular cells and adenocarcinoma in situ or invasive. According to the Trieste data records between 1999 and 2010, 101 diagnosis of cervical invasive adenocarcinoma were reported on cytological smears; histological correlation was available in 95% of cases. Of these, 63% were endometrial adenocarcinomas, 20% were endocervical adenocarcinomas, 6% were CIN, 5% were metastasis and 4% were adenocarcinoma in situ. The remaining 2% were represented by poorly differentiated carcinomas (Di Bonito, not published data).
e12
L. Di Bonito, C. Bergeron
Figure 7. Endocervical invasive adenocarcinoma on liquid-based smear (ThinPrep® ). The abnormal columnar cells shed in clusters. Nucleoli are evident. The background is necrotic with cellular debris. Adénocarcinome invasif de l’endocol sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales desquament en amas. Elles ont des noyaux nucléolés. Le fond comporte des débris cellulaires et de la nécrose.
Natural history and management The TBS distinguishes atypical glandular cells not otherwise specified (AGC-NOS), atypical glandular cells favor neoplastic (AGC-favor neoplastic) and adenocarcinoma in situ because these entities have a different risk of association with a CIN 3 or adenocarcinoma on biopsy. Women with an AGC-NOS result have a 9—41% risk of having a CIN 3 or adenocarcinoma in situ or invasive adenocarcinoma [17—21]. In contrast, women with an AGC favor neoplasia have a 27—96% risk of having the same kind of lesions on biopsy. A cytological result of AIS is associated with a 48—69% risk of biopsy confirmed AIS and a 38% risk of
Figure 8. Endometrial invasive adenocarcinoma on liquid-based smear (ThinPrep® ). The abnormal columnar cells shed in clusters. They have central nuclei with no visible cytoplasm. Adénocarcinome invasif de l’endomètre sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales de plus petite taille desquament en amas. Elles ont un noyau de petite taille en position centrale et un cytoplasme peu visible.
Figure 9. A. High-grade squamous intraepithelial neoplasia (CIN 3) involving endocervical glands on biopsy. Liquid-based smear (ThinPrep® ). B. The abnormal cells are shed in clusters and should not be interpreted as glandular cells. A. Néoplasie cervicale intraépithéliale (CIN) de grade 3 impliquant les récessus endocervicaux sur une biopsie. Frottis en milieu liquide ThinPrep® . B. Les cellules malpighiennes anormales desquament en amas mais ne doivent pas être interprétées comme des cellules cylindriques.
invasive adenocarcinoma [17—21]. Because of this different risk, women with AGC-NOS should be managed less aggressively than women with an AGC favor neoplasia result or a cytological diagnosis of AIS. The recommendations for management are however mostly similar for these different entities [22,23]. Colposcopy with biopsy and endocervical curettage is recommended. CIN 2—3 is diagnosed on biopsy under colposcopy or cone in about 40% of cases with a cytological diagnosis of ACG [17—19]. High-grade squamous intraepithelial lesions often involve endocervical glands and the squamous cells shed on Pap smear can simulate abnormal columnar cells (Fig. 9). The best criterion for diagnosing CIN 2—3 involving endocervical glands is the arrangement of abnormal squamous cells in the centre of the cluster (Figs. 9 and 10). On the other hand, abnormal glandular cells have tendency to escape from the centre of the clusters. Furthermore, endocervical elements display nuclei at the cell periphery, with cytoplasmic vacuolization, while squamous elements tend to have more centrally located nuclei and cytoplasms are dense (Fig. 10). Based on the
Cervical adenocarcinoma and cytopathology
e13
Figure 10. The abnormal cells are shed in clusters. A. Peripherally located nuclei and vacuolated cytoplasm correspond to adenocarcinoma in situ. B. Clusters made up of abnormal cells with centrally located nuclei and scanty not vacuolated cytoplasm correspond to a squamous intraepithelial high-grade lesion. Les cellules cylindriques anormales desquament en amas. A. Elles ont un noyau périphérique et un cytoplasme mucosécrétant correspondant à un adénocarcinome in situ. B. Les cellules malpighiennes anormales qui desquament en amas ont un noyau central et un cytoplasme de petite taille non sécrétant correspondant à une lésion malpighienne intraépithéliale de haut grade.
high prevalence of CIN 2, 3 in women with AGC, Bethesda 2001 Consensus Guidelines recommend that all women in reproductive age with an AGC cytological result have to be referred for a colposcopic evaluation and endocervical sampling. This allows detecting lesions in the endocervical canal not visible in colposcopy. Endocervical curettage is not recommended for pregnant women. There is no consensus on the indication of HPV testing after a diagnosis of ACG. Prospective studies have shown that HPV testing among premenopausal women has an excellent negative predictive value. Authors propose to include HPV testing using the same algorithm as for patients with a diagnosis of atypical squamous cells of undetermined significance (ASC-US) [24,25]. Patients with a diagnosis of ACG favor neoplasia or AIS should be referred directly for colposcopy, or diagnostic conization if the colposcopy is unsatisfactory. Postmenopausal women being more likely to have an endometrial lesion should undergo an endometrial biopsy [21].
Conclusion Cervical cytology is not a perfect tool for diagnosis of glandular abnormalities because its main role is the detection of squamous lesions. Though, the TBS 2001 allows better communication between the pathologist and the clinician and thus better care for patients with abnormal glandular cells. What kind of impact will the diagnosis of glandular abnormalities, particularly of AIS, have in the context of cervical cancer screening? Better detection of AIS will have an impact on the incidence and mortality of invasive adenocarcinoma in the coming years, more or less rapidly according to their developmental potential, if they are prevalent and
stay stable for several years or if they are prevalent and tend to evolve fast towards invasive lesions. Epidemiological data in Australia and England has shown that the risk of developing invasive cervical adenocarcinoma decreases as the level of protection of a normal smear with an adequate endocervical component increases particularly among women over 30 years old [26].
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
References [1] Bray F, Carstensen B, Moller H, Zappa M, Zakelj MP, Lawrence G, et al. Incidence trends of adenocarcinoma of the cervix in 13 European countries. Cancer Epidemiol Biomarkers Prev 2005;14:2191—9. [2] Ries L, Harkins D, Krapcho M, Mariotto A, Miller BA, Feurer EJ, et al. SEER cancer statistics review, 1975—2003, National Cancer Institute. Bethesda, MD, available at http://seer.cancer.gov/csr/1975 2003/-24k-2010-06-28 [3] Sasieni P, Adams J. Changing rates of adenocarcinoma and adenosquamous carcinoma of the cervix in England. Lancet 2001;357:1490—3. [4] Zappa M, Visioli CB, Ciatto S, Iossa A, Paci E, Sasieni P. Lower protection of cytological screening for adenocarcinoma and shorter protection for younger women: the results of a casecontrol in Florence. Br J Cancer 2004;90:1784—6. [5] Sherman ME, Wang SS, Carreon J, Devesa SS. Mortality trends for cervical squamous and adenocarcinoma in the
e14
[6]
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
L. Di Bonito, C. Bergeron United States: relation to incidence and survival. Cancer 2005;103:1258—64. Wells M, Ostor AG, Crum CP, Franceschi S, Tommasino M, Nesland JM, et al. Epithelial tumours, tumours of the uterine cervix. In: Tavassoli, Fand, Devilee P, editors. Tumours of the Breast and female genital organs. Lyon: WHO IARC; 2003. p. 262—79. Plaxe SC, Saltzstein SL. Estimation of the duration of the preclinical phase of cervical adenocarcinoma suggests that there is ample opportunity for screening. Gynecol Oncol 1999;75:55—61. Lee KR, Flynn CE. Early invasive adenocarcinoma of the cervix. Cancer 2000;89:1048—55. Castellsague X, Diaz M, de Sanjose S, Munoz N, Herrero R, Franceschi S, et al. Worlwide human papillomavirus etiology of cervical adenocarcinoma and its cofactors: implications for screening and prevention. J Natl Cancer Inst 2006;98: 303—15. Ruba S, Schoolland M, Allpress S, Sterrett G. Adenocarcinoma in situ of the uterine cervix: screening and diagnostic errors in Papanicolaou smears. Cancer 2004;102:280—7. Ansari-Lari MA, Staebler A, Zaino RJ, Shah KV, Ronnett BM. Distinction of endocervical and endometrial adenocarcinoma: immunohistochemical p16 expression correlated with human papillomavirus (HPV) DNA detection. Am J Surg Pathol 2004;28:160—7. Kalir T, Simsir A, Demopoulos HB, Demopoulos RI. Obstacles to the early detection of endocervical adenocarcinoma. Int J Gynecol Pathol 2005;24:414—21. Solomon D, Davey D, Kurman R, Moriarty A, O’Connor D, Prey, et al. The 2001 Bethesda System. Terminology for reporting results of cervical cytology. JAMA 2002;287:2114—9. Schneider V. Cytopathology of adenocarcinoma in situ of the endocervix and its differential diagnosis. Monogr Clin Cytol 2011;20:26—33 [Epub 2010 Dec 10]. Bergeron C, Cartier I, Guldner L, Lassalle M, Savignon A, Asselain B. Lésions précancéreuses et cancers du col de l’utérus diagnostiqués par le frottis cervical, Île-de-France, enquête Crisap, 2002. BEH 2005;2:4—5. Jones BA, Novis DA. Follow-up of abnormal gynaecologic cytology: a College of American Pathologists Q-Probes study of
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
16,132 cases from 306 laboratories. Arch Pathol Lab Med 2000;124:665—71. Laverty CR, Farnsworth A, Thurloe J, Bowditch R. The reliability of a cytological prediction of cervical adenocarcinoma in situ. Aust N Z J Obstet Gynaecol 1988;28:307—12. Lee KR, Manna EA, St John T. Atypical endocervical glandular cells: accuracy of cytological diagnosis. Diagn Cytopathol 1995;13:202—8. Westin MC, Derchain SF, Rabelo-Santos SH, Angelo-Andrade LA, Sarian LO, Oliveira E, et al. Atypical glandular cells and adenocarcinoma in situ according to Bethesda 2001 Classification cytohistological correlation and clinical implications. Eur J Obstet Gynecol Reprod Biol 2008;13:79—85. Adhya AK, Mahesha V, Srinivasan R, Nijhawan R, Rajwanshi A, Suri V, et al. Atypical glandular cells in cervical smearshistological correlation and a suggested plan of management based on age of the patient in a low resource setting. Cytopathology 2009;20:375—9. Chieng DC, Elgert P, Cohen JM, Cangiarella JF. Clinical significance of atypical glandular cells of undetermined significance in postmenopausal women. Cancer 2001;93:1—7. Wright Jr TC, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D. 2006 Consensus guidelines for the management of women with abnormal cervical cancer screening tests. Am J Obstet Gynecol 2007;197:346—55. HAS 2002: clinical practice guidelines. Management of a patient with an abnormal cervical smear, update, available at: http://www.has-sante.fr/portail/upload/docs/application/ pdf/frottis anglais.pdf Liao SY, Rodgers WH, Kauderer J, Bonfiglio TA, Walker JL, Darcy KM, et al. Carbonic anhydrase IX (CA-IX) and human papillomavirus (HPV) as diagnostic biomarkers of cervical dysplasia/neoplasia in women with a cytologic diagnosis of atypical glandular cells (AGC): a Gynecologic Oncology Group (GOG) Study. Int J Cancer 2009;125:2434—40. Saqi A, Gupta PK, Erroll M, Babiac A, Blackmun D, Mansukhani M, et al. High-risk human papillomavirus DNA testing: a marker for atypical glandular cells. Diagn Cytopathol 2006;34:235—9. Mitchell H, Hocking J, Saville M. Improvement in protection against adenocarcinoma of the cervix resulting from participation in cervical screening. Cancer 2003;99:336—41.
Disponible en ligne sur
www.sciencedirect.com
REVIEW
Cytological screening of endocervical adenocarcinoma Le dépistage cytologique de l’adénocarcinome du col Luigi Di Bonito a, Christine Bergeron b,∗ a
Direttore dell’ UCO, Anatomia e Istologia patologica, Ospedale Universitario di Cattinara, Trieste, Italy b Laboratoire Cerba, boîte postale, 95066 Cergy Pontoise cedex 9, France Accepted for publication on 13 September 2012 Available online 22 November 2012
KEYWORDS Cervical screening; Adenocarcinoma; Cytology; Bethesda terminology
MOTS CLÉS Dépistage du cancer col ; Adénocarcinome ;
∗
Summary Invasive endocervical adenocarcinoma represents on average 15% of cervical carcinomas and it is associated with the human papillomavirus infection high risk types 16 and 18 in most cases. Its detection has some special features compared to squamous cell carcinoma; glandular precancerous lesions are less known and only adenocarcinoma in situ is diagnosed by consensus among pathologists; adenocarcinoma in situ develops in the squamocolumnar junction by reserve cells but it is hard to be located by colposcopy in the endocervical canal or in the deep glandular recess. Sampling of endocervical cells requires brushes rather than an Ayre spatula. Cytological diagnosis of glandular cells abnormalities is based on the Bethesda System 2001 terminology which redefined endocervical cells abnormalities and also introduced the entity of adenocarcinoma in situ. This entity is characterized by specific morphological features, such as the radial arrangement of nuclei in the periphery, like ‘‘at the end of the feathers of a bird’s wing’’ (feathering of cells), images of nuclei palissading or rosette without tumoral diathesis. Glandular cells abnormalities are rare and represent less than 0.1% of all smears and less than 5% of abnormal smears. By improving the collection and the interpretation of abnormal endocervical cells, cytological screening should allow the diagnosis of in situ adenocarcinoma and detection of invasive adenocarcinoma at a very early stage. This will lead to a decrease in mortality from endocervical adenocarcinoma, especially in young women. © 2012 Elsevier Masson SAS. All rights reserved.
Résumé L’adénocarcinome invasif du col utérin correspond en moyenne à 15 % des carcinomes du col utérin. Il est le plus souvent associé à une infection à papillomavirus humain de type 16 ou 18. Son dépistage présente des particularités par rapport au carcinome malpighien ; les lésions glandulaires précancéreuses sont moins bien connues et seul l’adénocarcinome in situ est diagnostiqué de manière consensuelle par les pathologistes ; l’adénocarcinome in situ se
DOI of original article: http://dx.doi.org/10.1016/j.annpat.2012.09.199. Corresponding author. E-mail address: [email protected] (C. Bergeron).
0242-6498/$ — see front matter © 2012 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.annpat.2012.09.230
Cervical adenocarcinoma and cytopathology
Cytologie ; Terminologie Bethesda 2001
e9
développe autour de la jonction squamocylindrique à partir de cellules de réserve et peut être difficile à localiser en colposcopie dans le canal endocervical. Le prélèvement des cellules cylindriques nécessite une brosse plutôt que la spatule d’Ayre. Le diagnostic cytologique des cellules glandulaires se fait selon la terminologie Bethesda 2001 qui a redéfini le cadre des anomalies des cellules glandulaires et a introduit l’entité « adénocarcinome in situ ». Cette entité est caractérisée par des critères spécifiques, comme la disposition radiaire des noyaux en périphérie, donnant aux cellules un aspect ressemblant « à des plumes à l’extrémité d’une aile d’oiseau », des images de noyaux en palissade ou en rosette sans diathèse tumorale. Les anomalies des cellules glandulaires restent rares et correspondent à moins de 0,1 % de la totalité des frottis et moins de 5 % des frottis anormaux. En améliorant le prélèvement et l’interprétation des cellules cylindriques anormales, le dépistage cytologique devrait permettre le diagnostic d’adénocarcinome in situ et d’adénocarcinome invasif à un stade précoce. Une diminution de la mortalité par adénocarcinome du col, en particulier chez les femmes jeunes, devrait en résulter. © 2012 Elsevier Masson SAS. Tous droits réservés.
Introduction Endocervical adenocarcinoma’s incidence remains stable or is slightly increasing in 2011, including those countries where a parallel decrease in squamous cancers’ incidence is observed [1,2]. In some countries, like England or Australia, where there is a well-organized screening, invasive adenocarcinoma makes up almost one fourth of all invasive cervical cancers [3]. This increase has been particularly striking in women under 40 [1,4]. Over the last 20 years, invasive cervical adenocarcinoma’s mortality has also increased in absolute and relative terms compared to squamous cell carcinoma [5]. Should we conclude that cytological screening is not suitable for detecting precancerous glandular lesions or invasive adenocarcinomas at a very early stage?
Precancerous lesions of the endocervical cylindrical epithelium Adenocarcinoma in situ is the only histological recognized entity by consensus as a preinvasive lesion of the cervix [6] (Fig. 1). The adenocarcinoma in situ is sometimes referred to as high-grade intraepithelial glandular lesion or dysplasia. Progression from adenocarcinoma in situ to invasive adenocarcinoma is estimated to last between 5 and 13 years [7,8]. The adenocarcinoma in situ arises at the squamocolumnar junction after persistent infection with high-risk human papillomavirus (HPV) infection, types 16 and 18 being the most common [9]. Persistent HPV16 and 18 infections are usually associated with cervical squamous intraepithelial neoplasia (CIN 2—3) starting at the junction because these viruses have a higher affinity with keratinocytes and they can more easily reach the basal layers of the squamous epithelium where the viral cycle starts. HPV can also infect at the same time or, more rarely, separately reserve cells around the squamocolumnar junction which may differentiate towards glandular columnar cells [10]. The infection usually remains latent and in case of persistence it may allow the expression of viral oncogenes E6 and E7 in the columnar epithelium causing a transforming infection. Immunohistochemical expression of p16 is an indirect way of highlighting the transforming infection in the endocervical columnar epithelium [11]. Morphologically, the transforming infection results in adenocarcinoma in situ,
Figure 1. Adenocarcinoma in situ on a biopsy. A. The abnormal epithelium affects only a part of the endocervical glands. B. The abnormal epithelium reaches the surface of endocervical glandular structures which can be reached by a brush. Adénocarcinome in situ sur une biopsie. A. L’épithélium cylindrique anormal n’intéresse qu’une partie des glandes endocervicales. B. L’épithélium cylindrique anormal atteint la surface des récessus endocervicaux. Une brosse peut donc prélever les cellules glandulaires anormales.
e10
L. Di Bonito, C. Bergeron
Figure 2. The detection of p16 protein by immunohistochemistry showed a diffuse staining of all abnormal cells. La détection de la protéine p16 par immunohistochimie montre un marquage diffus sur l’ensemble des cellules anormales.
with diffuse atypical endocervical cells immunopositivity for p16 [12] (Fig. 2). This marker may help in the differential diagnosis of glandular hyperplasia and its variants such as glandular intraepithelial low-grade lesions. These have no reproducible morphological criteria and in most of cases they are not associated with an HPV infection. They are negative for p16 immunohistochemical stain. Tubal metaplasia is another differential diagnosis. In these cases, p16 stains only ciliated cells (Fig. 3).
Cytological screening Sampling of the endocervical epithelium Cytological sampling can be performed both for conventional smears using the cytobrush and for liquid-based cytology using the cervexbrush and its variants. Improvement of cytology brushes allows better sampling of the endocervical epithelium. Endocervical cells sampling can be difficult because adenocarcinoma in situ may have a multifocal development and endocervical glands can be partially or totally involved. If abnormal endocervical cells do not reach the superficial part of the glandular structures, they may not shed on the smears. This explains why sometimes adenocarcinoma in situ is diagnosed late on the cone biopsy [10,12].
Terminology for abnormal endocervical cells To ensure better communication among clinicians involved in cervical cancer screening, the Terminology Bethesda System (TBS) 1988 proposed diagnostic categories for reporting of cervical smear. The TBS was revised in 2001, especially for cervical glandular cell abnormalities [13,14]. The TBS provides the following categories.
Atypical glandular cells (AGC) of the endocervix, endometrium, or glandular not otherwise specified (NOS) Site specifying allows clinician to have different managements according to the patient age.
Figure 3. A. Tubal metaplasia presents high columnar and ciliated cells. B. The detection of p16 protein by immunohistochemistry highlights the presence of ciliated cells. A. La métaplasie tubaire présente des cellules cylindriques hautes de type intercalaire et des cellules ciliées. B. La détection de la protéine p16 par immunohistochimie souligne la présence des cellules ciliées.
The diagnosis of atypical glandular cells is made in presence of slight atypias (nuclear enlargement without chromatin abnormality) and when columnar cells shed either isolates or in clusters. The main differential diagnosis included in this group are endocervical cell changes associated with the presence of a polyp, endocervical hyperplasia, reparative changes occurring after conization, and tubal metaplasia, especially when cells lose their upper ciliated part. When the columnar cells are of an endometrial origin the differential diagnosis regards reparative or metaplastic changes of the uterine isthmus or of endometrium. After menopause, it is preferable to report the presence of endometrial cells without specifying their nature (normal/abnormal). In this case an endometrial biopsy has to be performed.
Atypical glandular cells, favor neoplastic This diagnosis is suggested when the morphological abnormalities of cylindrical cells are more severe but quantitatively and qualitatively fall short of definitive adenocarcinoma diagnosis. The differential diagnosis of these anomalies includes cellular changes associated with the
Cervical adenocarcinoma and cytopathology
e11
Figure 4. In presence of an intrauterine device, the smear shows three-dimensional clusters of columnar cells with a regular but slightly enlarged nucleus. Such changes should not be confused with glandular atypia. Le frottis fait en présence d’un stérilet comporte des amas tridimensionnels de cellules cylindriques avec un noyau régulier mais légèrement augmenté de volume. Ce type de modifications ne doit pas être confondu avec des atypies glandulaires.
Figure 6. Adenocarcinoma in situ on liquid-based smear (ThinPrep® ). The abnormal cells arranged in rosettes around a core. The background is clean. Adénocarcinome in situ (AIS) sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales ont une disposition en rosettes autour d’une lumière. Le fond est propre.
presence of an intrauterine device (Fig. 4) or due to radiotherapy and chemotherapy.
distributed chromatin but absence of prominent nucleoli and tumor diathesis.
The adenocarcinoma in situ (AIS) was proposed for the first time in the TBS 2001
The invasive adenocarcinoma
This entity is characterized by specific morphological features, such as the radial arrangement of nuclei in the periphery, like ‘‘at the end of the feathers of a bird’s wing’’ (feathering of cells), images of nuclei palissading or rosette on a clean background without evidence of necrosis (Figs. 5 and 6). Abnormal cells display increased nucleo-cytoplasmic ratio, with slightly altered and coarsely
Figure 5. Adenocarcinoma in situ on liquid-based smear (ThinPrep® ). The abnormal cells have a cylindrical palisad arrangement of the nuclei and tapered cytoplasm like bird’s feathers. The background is clean and lacks of necrosis. Adénocarcinome in situ (AIS) sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales ont une disposition en palissade du noyau et un cytoplasme effilé en plumes d’oiseau. Le fond est propre sans nécrose.
The invasive adenocarcinoma is characterized by the presence of atypical glandular cells showing nuclear changes comprising enlargement, hyperchromasia, coarse granulation of chromatin and large, sometimes irregular multiple nucleoli. The abnormal cells often form spherical or oval clusters and the smear background often shows blood, necrosis, and cell debris. The TBS 2001 asks, if possible, to specify if the adenocarcinoma is endocervical, endometrial or extra-uterine. The criteria which point to an endocervical origin are abundant shedding of large columnar cells (Fig. 7). Endometrial origin is suggested when there is less shedding of abnormal glandular cells which are smaller and round shaped with a centrally positioned nucleus (Fig. 8) and tend to arrange in morules. The extra-uterine origin is rare but can be observed in rectal cancers infiltrating the vaginal wall, in ovarian cancers shedding through the tubal lumen or in the breast cancer metastasis. All these findings are rare if compared to squamous abnormalities [15,16]. As described by the Centre de regroupement informatique et statistique en anatomie pathologique en Île-de-France (CRISAP-IDF), out of 250,000 smears, glandular abnormalities are rare and represent less than 0.1% of all smears and less than 5% of abnormal smears [15]. These percentages do not distinguish atypical glandular cells and adenocarcinoma in situ or invasive. According to the Trieste data records between 1999 and 2010, 101 diagnosis of cervical invasive adenocarcinoma were reported on cytological smears; histological correlation was available in 95% of cases. Of these, 63% were endometrial adenocarcinomas, 20% were endocervical adenocarcinomas, 6% were CIN, 5% were metastasis and 4% were adenocarcinoma in situ. The remaining 2% were represented by poorly differentiated carcinomas (Di Bonito, not published data).
e12
L. Di Bonito, C. Bergeron
Figure 7. Endocervical invasive adenocarcinoma on liquid-based smear (ThinPrep® ). The abnormal columnar cells shed in clusters. Nucleoli are evident. The background is necrotic with cellular debris. Adénocarcinome invasif de l’endocol sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales desquament en amas. Elles ont des noyaux nucléolés. Le fond comporte des débris cellulaires et de la nécrose.
Natural history and management The TBS distinguishes atypical glandular cells not otherwise specified (AGC-NOS), atypical glandular cells favor neoplastic (AGC-favor neoplastic) and adenocarcinoma in situ because these entities have a different risk of association with a CIN 3 or adenocarcinoma on biopsy. Women with an AGC-NOS result have a 9—41% risk of having a CIN 3 or adenocarcinoma in situ or invasive adenocarcinoma [17—21]. In contrast, women with an AGC favor neoplasia have a 27—96% risk of having the same kind of lesions on biopsy. A cytological result of AIS is associated with a 48—69% risk of biopsy confirmed AIS and a 38% risk of
Figure 8. Endometrial invasive adenocarcinoma on liquid-based smear (ThinPrep® ). The abnormal columnar cells shed in clusters. They have central nuclei with no visible cytoplasm. Adénocarcinome invasif de l’endomètre sur un frottis en milieu liquide ThinPrep® . Les cellules cylindriques anormales de plus petite taille desquament en amas. Elles ont un noyau de petite taille en position centrale et un cytoplasme peu visible.
Figure 9. A. High-grade squamous intraepithelial neoplasia (CIN 3) involving endocervical glands on biopsy. Liquid-based smear (ThinPrep® ). B. The abnormal cells are shed in clusters and should not be interpreted as glandular cells. A. Néoplasie cervicale intraépithéliale (CIN) de grade 3 impliquant les récessus endocervicaux sur une biopsie. Frottis en milieu liquide ThinPrep® . B. Les cellules malpighiennes anormales desquament en amas mais ne doivent pas être interprétées comme des cellules cylindriques.
invasive adenocarcinoma [17—21]. Because of this different risk, women with AGC-NOS should be managed less aggressively than women with an AGC favor neoplasia result or a cytological diagnosis of AIS. The recommendations for management are however mostly similar for these different entities [22,23]. Colposcopy with biopsy and endocervical curettage is recommended. CIN 2—3 is diagnosed on biopsy under colposcopy or cone in about 40% of cases with a cytological diagnosis of ACG [17—19]. High-grade squamous intraepithelial lesions often involve endocervical glands and the squamous cells shed on Pap smear can simulate abnormal columnar cells (Fig. 9). The best criterion for diagnosing CIN 2—3 involving endocervical glands is the arrangement of abnormal squamous cells in the centre of the cluster (Figs. 9 and 10). On the other hand, abnormal glandular cells have tendency to escape from the centre of the clusters. Furthermore, endocervical elements display nuclei at the cell periphery, with cytoplasmic vacuolization, while squamous elements tend to have more centrally located nuclei and cytoplasms are dense (Fig. 10). Based on the
Cervical adenocarcinoma and cytopathology
e13
Figure 10. The abnormal cells are shed in clusters. A. Peripherally located nuclei and vacuolated cytoplasm correspond to adenocarcinoma in situ. B. Clusters made up of abnormal cells with centrally located nuclei and scanty not vacuolated cytoplasm correspond to a squamous intraepithelial high-grade lesion. Les cellules cylindriques anormales desquament en amas. A. Elles ont un noyau périphérique et un cytoplasme mucosécrétant correspondant à un adénocarcinome in situ. B. Les cellules malpighiennes anormales qui desquament en amas ont un noyau central et un cytoplasme de petite taille non sécrétant correspondant à une lésion malpighienne intraépithéliale de haut grade.
high prevalence of CIN 2, 3 in women with AGC, Bethesda 2001 Consensus Guidelines recommend that all women in reproductive age with an AGC cytological result have to be referred for a colposcopic evaluation and endocervical sampling. This allows detecting lesions in the endocervical canal not visible in colposcopy. Endocervical curettage is not recommended for pregnant women. There is no consensus on the indication of HPV testing after a diagnosis of ACG. Prospective studies have shown that HPV testing among premenopausal women has an excellent negative predictive value. Authors propose to include HPV testing using the same algorithm as for patients with a diagnosis of atypical squamous cells of undetermined significance (ASC-US) [24,25]. Patients with a diagnosis of ACG favor neoplasia or AIS should be referred directly for colposcopy, or diagnostic conization if the colposcopy is unsatisfactory. Postmenopausal women being more likely to have an endometrial lesion should undergo an endometrial biopsy [21].
Conclusion Cervical cytology is not a perfect tool for diagnosis of glandular abnormalities because its main role is the detection of squamous lesions. Though, the TBS 2001 allows better communication between the pathologist and the clinician and thus better care for patients with abnormal glandular cells. What kind of impact will the diagnosis of glandular abnormalities, particularly of AIS, have in the context of cervical cancer screening? Better detection of AIS will have an impact on the incidence and mortality of invasive adenocarcinoma in the coming years, more or less rapidly according to their developmental potential, if they are prevalent and
stay stable for several years or if they are prevalent and tend to evolve fast towards invasive lesions. Epidemiological data in Australia and England has shown that the risk of developing invasive cervical adenocarcinoma decreases as the level of protection of a normal smear with an adequate endocervical component increases particularly among women over 30 years old [26].
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
References [1] Bray F, Carstensen B, Moller H, Zappa M, Zakelj MP, Lawrence G, et al. Incidence trends of adenocarcinoma of the cervix in 13 European countries. Cancer Epidemiol Biomarkers Prev 2005;14:2191—9. [2] Ries L, Harkins D, Krapcho M, Mariotto A, Miller BA, Feurer EJ, et al. SEER cancer statistics review, 1975—2003, National Cancer Institute. Bethesda, MD, available at http://seer.cancer.gov/csr/1975 2003/-24k-2010-06-28 [3] Sasieni P, Adams J. Changing rates of adenocarcinoma and adenosquamous carcinoma of the cervix in England. Lancet 2001;357:1490—3. [4] Zappa M, Visioli CB, Ciatto S, Iossa A, Paci E, Sasieni P. Lower protection of cytological screening for adenocarcinoma and shorter protection for younger women: the results of a casecontrol in Florence. Br J Cancer 2004;90:1784—6. [5] Sherman ME, Wang SS, Carreon J, Devesa SS. Mortality trends for cervical squamous and adenocarcinoma in the
e14
[6]
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
L. Di Bonito, C. Bergeron United States: relation to incidence and survival. Cancer 2005;103:1258—64. Wells M, Ostor AG, Crum CP, Franceschi S, Tommasino M, Nesland JM, et al. Epithelial tumours, tumours of the uterine cervix. In: Tavassoli, Fand, Devilee P, editors. Tumours of the Breast and female genital organs. Lyon: WHO IARC; 2003. p. 262—79. Plaxe SC, Saltzstein SL. Estimation of the duration of the preclinical phase of cervical adenocarcinoma suggests that there is ample opportunity for screening. Gynecol Oncol 1999;75:55—61. Lee KR, Flynn CE. Early invasive adenocarcinoma of the cervix. Cancer 2000;89:1048—55. Castellsague X, Diaz M, de Sanjose S, Munoz N, Herrero R, Franceschi S, et al. Worlwide human papillomavirus etiology of cervical adenocarcinoma and its cofactors: implications for screening and prevention. J Natl Cancer Inst 2006;98: 303—15. Ruba S, Schoolland M, Allpress S, Sterrett G. Adenocarcinoma in situ of the uterine cervix: screening and diagnostic errors in Papanicolaou smears. Cancer 2004;102:280—7. Ansari-Lari MA, Staebler A, Zaino RJ, Shah KV, Ronnett BM. Distinction of endocervical and endometrial adenocarcinoma: immunohistochemical p16 expression correlated with human papillomavirus (HPV) DNA detection. Am J Surg Pathol 2004;28:160—7. Kalir T, Simsir A, Demopoulos HB, Demopoulos RI. Obstacles to the early detection of endocervical adenocarcinoma. Int J Gynecol Pathol 2005;24:414—21. Solomon D, Davey D, Kurman R, Moriarty A, O’Connor D, Prey, et al. The 2001 Bethesda System. Terminology for reporting results of cervical cytology. JAMA 2002;287:2114—9. Schneider V. Cytopathology of adenocarcinoma in situ of the endocervix and its differential diagnosis. Monogr Clin Cytol 2011;20:26—33 [Epub 2010 Dec 10]. Bergeron C, Cartier I, Guldner L, Lassalle M, Savignon A, Asselain B. Lésions précancéreuses et cancers du col de l’utérus diagnostiqués par le frottis cervical, Île-de-France, enquête Crisap, 2002. BEH 2005;2:4—5. Jones BA, Novis DA. Follow-up of abnormal gynaecologic cytology: a College of American Pathologists Q-Probes study of
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
16,132 cases from 306 laboratories. Arch Pathol Lab Med 2000;124:665—71. Laverty CR, Farnsworth A, Thurloe J, Bowditch R. The reliability of a cytological prediction of cervical adenocarcinoma in situ. Aust N Z J Obstet Gynaecol 1988;28:307—12. Lee KR, Manna EA, St John T. Atypical endocervical glandular cells: accuracy of cytological diagnosis. Diagn Cytopathol 1995;13:202—8. Westin MC, Derchain SF, Rabelo-Santos SH, Angelo-Andrade LA, Sarian LO, Oliveira E, et al. Atypical glandular cells and adenocarcinoma in situ according to Bethesda 2001 Classification cytohistological correlation and clinical implications. Eur J Obstet Gynecol Reprod Biol 2008;13:79—85. Adhya AK, Mahesha V, Srinivasan R, Nijhawan R, Rajwanshi A, Suri V, et al. Atypical glandular cells in cervical smearshistological correlation and a suggested plan of management based on age of the patient in a low resource setting. Cytopathology 2009;20:375—9. Chieng DC, Elgert P, Cohen JM, Cangiarella JF. Clinical significance of atypical glandular cells of undetermined significance in postmenopausal women. Cancer 2001;93:1—7. Wright Jr TC, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D. 2006 Consensus guidelines for the management of women with abnormal cervical cancer screening tests. Am J Obstet Gynecol 2007;197:346—55. HAS 2002: clinical practice guidelines. Management of a patient with an abnormal cervical smear, update, available at: http://www.has-sante.fr/portail/upload/docs/application/ pdf/frottis anglais.pdf Liao SY, Rodgers WH, Kauderer J, Bonfiglio TA, Walker JL, Darcy KM, et al. Carbonic anhydrase IX (CA-IX) and human papillomavirus (HPV) as diagnostic biomarkers of cervical dysplasia/neoplasia in women with a cytologic diagnosis of atypical glandular cells (AGC): a Gynecologic Oncology Group (GOG) Study. Int J Cancer 2009;125:2434—40. Saqi A, Gupta PK, Erroll M, Babiac A, Blackmun D, Mansukhani M, et al. High-risk human papillomavirus DNA testing: a marker for atypical glandular cells. Diagn Cytopathol 2006;34:235—9. Mitchell H, Hocking J, Saville M. Improvement in protection against adenocarcinoma of the cervix resulting from participation in cervical screening. Cancer 2003;99:336—41.