Glandular Intraepithelial Neoplasia Research Articles

Upregulation of heat shock protein 27 in metaplastic and neoplastic lesions of the endocervix

Heat shock proteins (hsps) are molecular chaperones that are known to play a pivotal role in regulating intracellular homeostasis. hsp27 may have diagnostic and prognostic values for different gynecological malignancies. A cross-sectional analytical study was conducted at the Department of Pathology, The University of Liverpool, Liverpool, UK. Included in the study were 80 cervical glandular lesions of various histologic types, representing tuboendometrial metaplasia/endometriosis (n = 19), cervical glandular intraepithelial neoplasia (n = 33), and invasive adenocarcinoma (n = 28). Paraffin-embedded sections were stained using a commercial mouse monoclonal anti-hsp27 antibody with prior pressure-cooking for antigen retrieval. Sections of 11 normal cervices were used as controls. The median percentage of cells expressing hsp27 in each group was calculated. Normal cervical glands showed minimal expression of hsp27 (median: 10%, interquartile ranges [IQ]: 5-15). Expression was significantly more widespread in tuboendometrial metaplasia/endometriosis (median: 35%, IQ: 15-80), cervical glandular intraepithelial neoplasia (median: 60%, IQ: 32-80), and invasive adenocarcinoma (median: 40%, IQ: 25-80) when compared with normal endocervix (P = 0.007, < 0.001, and 0.001, respectively). However, no significant difference in hsp27 protein expression was found between cervical glandular intraepithelial neoplasia and invasive adenocarcinoma. In invasive adenocarcinoma, hsp27 showed no correlation with tumor grade, lymph node involvement, and lymphovascular space invasion. Our data highlight early dysregulation of hsp27 expression in both metaplastic and neoplastic lesions of the cervix.

Upregulation of heat shock protein 27 in metaplastic and neoplastic lesions of the endocervix

Heat shock proteins (hsps) are molecular chaperones that are known to play a pivotal role in regulating intracellular homeostasis. hsp27 may have diagnostic and prognostic values for different gynecological malignancies. A cross-sectional analytical study was conducted at the Department of Pathology, The University of Liverpool, Liverpool, UK. Included in the study were 80 cervical glandular lesions of various histologic types, representing tuboendometrial metaplasia/endometriosis (n= 19), cervical glandular intraepithelial neoplasia (n= 33), and invasive adenocarcinoma (n= 28). Paraffin-embedded sections were stained using a commercial mouse monoclonal anti-hsp27 antibody with prior pressure-cooking for antigen retrieval. Sections of 11 normal cervices were used as controls. The median percentage of cells expressing hsp27 in each group was calculated. Normal cervical glands showed minimal expression of hsp27 (median: 10%, interquartile ranges [IQ]: 5–15). Expression was significantly more widespread in tuboendometrial metaplasia/endometriosis (median: 35%, IQ: 15–80), cervical glandular intraepithelial neoplasia (median: 60%, IQ: 32–80), and invasive adenocarcinoma (median: 40%, IQ: 25–80) when compared with normal endocervix (P= 0.007, &lt;0.001, and 0.001, respectively). However, no significant difference in hsp27 protein expression was found between cervical glandular intraepithelial neoplasia and invasive adenocarcinoma. In invasive adenocarcinoma, hsp27 showed no correlation with tumor grade, lymph node involvement, and lymphovascular space invasion. Our data highlight early dysregulation of hsp27 expression in both metaplastic and neoplastic lesions of the cervix.

A retrospective clinical audit of cervical smears reported as 'glandular neoplasia'.

The aims of this study were to review the diagnostic pathway of women with smears reported as 'glandular neoplasia' and to outline the management, colposcopy findings, treatment and final histological diagnosis in these women. The design was a retrospective review. A total of 114 women were identified over a 5-year period from the cytology database at the Royal Liverpool University Hospital Cytology Department, whose hospital case notes were available for review. Methods included a review of the case notes for the demographic details, indication for smear, colposcopic findings, investigation and/or treatment procedures, histology, final diagnosis and current disease status. Of 114 smears reported as 'glandular neoplasia', 67 were reported as consistent with cervical glandular intra-epithelial neoplasia (CGIN), six with endocervical adenocarcinoma, 36 with endometrial adenocarcinoma and five with other glandular neoplastic abnormalities. The average age was 46.5 years. 79 (69.3%) smears were routine call/recall and 36 (30.7%) women were symptomatic. The positive predictive value (PPV) for a significant histological abnormality in the CGIN smear group was 80.6% (23.9% invasive carcinomas, 43.3% CGIN and 13.4% CIN) and the PPV of an 'endometrial adenocarcinoma' smear was 86.1%. Smears indicating glandular neoplasia are associated with a high probability of clinically significant lesions, the PPV of a CGIN smear being over 80%. Immediate referral for colposcopy and assessment by an experienced colposcopist is recommended.

Overexpression of cyclins A and B as markers of neoplastic glandular lesions of the cervix

Introduction. Cyclins are a family of regulatory proteins that play a pivotal role in controlling the cell cycle. While there is evidence of their altered expression in cervical squamous lesions, their precise role in glandular neoplasia is yet to be elucidated. Objectives. To investigate the role of cyclins as markers of early cervical glandular neoplasia by comparing their expression in lesions of different histological type. Methods. Through a cross-sectional analytical study, paraffin wax sections of normal cervix ( n = 11), endometriosis/tubo-endometrioid metaplasia (TEM) ( n = 19), cervical glandular intraepithelial neoplasia (CGIN) ( n = 33), and invasive adenocarcinoma ( n = 28) were studied using monoclonal antibodies for cyclins A, B, D, and E with heat pretreatment for antigen unmasking. A quantitative assessment was employed for the analysis of percentage expression of each marker. Statistical analysis of data was performed using SPSS. Results. A progressive significant increase in cyclin A expression occurred from normal cervix (median: 0, IQ: 0–0), through endometriosis/TEM (median: 1, IQ: 0–15) and CGIN (median: 15, IQ: 0–30) to invasive adenocarcinoma (median: 40, IQ: 21.25–60). Cyclin B exhibited a similar pattern (median: 0, IQ: 0–0, median: 0, IQ: 0–0.5, median: 8, IQ: 0.75–15, and median: 30, IQ: 15–45, respectively). Statistically higher expression of cyclin B was found in CGIN than in TEM/endometriosis ( P < 0.001). Invasive adenocarcinomas expressed higher levels of cyclins A and B than CGIN ( P < 0.001). There was significantly greater cyclin E expression in TEM/endometriosis than in normal cervix ( P = 0.03) with a nonsignificant further increase in CGIN and invasive adenocarcinoma. The expression of cyclin D was not significantly different among all groups. Conclusions. Our data indicate that up-regulation of cyclin A and B expression occurs in neoplastic lesions of the cervix. Cyclin B expression was significantly more widespread in CGIN lesions than in TEM/endometriosis indicating that further assessment of the value of this marker in the diagnosis of cervical glandular neoplasia is warranted.

Guest Lecture 8.45–9.30 Monday 15 September 2003 Endocervical Adenocarcinoma and its Precursors

The incidence of malignant and pre‐malignant endocervical glandular lesions is increasing. Part of this is an apparent increase due to a reduction in the number of invasive cervical squamous carcinomas but there is evidence that there is a real increase in malignant and pre‐malignant endocervical glandular lesions. Different terminologies are in use in the UK where the term cervical glandular intraepithelial neoplasia (CGIN) is commonly used and the rest of the world where pre‐malignant lesions are classified as glandular dysplasia and adenocarcinoma in situ (AIS) (WHO classification). It is well established that high‐grade CGIN (AIS in WHO terminology) is a precursor lesion of cervical adenocarcinoma but it is controversial whether a recognizable precursor to high grade CGIN (namely low‐grade CGIN) exists and criteria for diagnosing this are poorly established and poorly reproducible. Most cases of CGIN are of usual or endocervical type but other morphological subtypes described include endometrioid, intestinal, tubal and stratified mucinous intraepithelial lesion (SMILE). The presence of skip lesions and lesions high up the endocervical canal has been overemphasised in CGIN with most cases occurring close to the transformation zone. Treatment is on an individualized basis but local excision with negative margins and close cytological follow‐up may be employed. There is evidence in the literature that early invasive adenocarcinomas behave in a similar fashion to early invasive squamous carcinomas and that, on selected occasions, conservative therapy can be safely undertaken. However, further studies are needed to ascertain the behaviour and natural history of early invasive cervical adenocarcinoma. In 10%–15% of cases it may be impossible to ascertain whether a malignant endocervical glandular lesion is invasive or in situ. There are many benign mimics of CGIN and adenocarcinoma, including tuboendometrial metaplasia (TEM), endometriosis and microglandular hyperplasia (MGH). Although careful morphological examination usually allows confident distinction of these lesions, a panel of immunohistochemical stains including MIB1, bcl2 and p16 may assist.

P16INK4a is a useful marker for the diagnosis of adenocarcinoma of the cervix uteri and its precursors: an immunohistochemical study with immunocytochemical correlations.

p16 is a tumor suppressor gene that plays a central role in the regulation of the cell cycle. In squamous cervical cancers, overexpression of p16 is induced by HPV and associated with the carcinogenesis of cervical epithelia. The aim of this study was to determine whether immunostaining of p16 is useful in detecting adenocarcinomas of the cervix uteri in histologic and cytologic routine specimens. A total of 45 surgical specimens, including 18 cases of invasive carcinoma, 8 cases of adenocarcinoma in situ, 4 cases of endocervical glandular atypia (cervical glandular intraepithelial neoplasia), and 15 reactive lesions of the endocervical glands were immunostained using a specific anti-human p16 monoclonal antibody (clone E6H4, mtm laboratories AG, Heidelberg, Germany). Furthermore, immunocytochemical analysis was performed on 10 preoperative ThinPrep cytologic samples with abnormal glandular cells and compared with the human papillomavirus status as assessed with the Hybrid Capture II test. p16 was detected immunohistochemically in all 26 cases of adenocarcinoma of the cervix uteri, including 18 invasive and 8 in situ carcinomas. Only a focal expression was evidenced in the four specimens with endocervical glandular atypia, and no reaction was found in reactive lesions. Also, the immunocytochemical analysis on the 10 ThinPrep specimens evidenced a strong expression of p16 in neoplastic endocervical cells. In all cases this was associated with a high-risk HPV-positive typing. p16 is a useful marker for the detection of the adenocarcinoma of the cervix uteri and its precursors. The immunocytochemical detection on ThinPrep specimens may contribute to an early detection of endocervical lesions.

Cytology suggestive of glandular neoplasia: outcomes and suggested management.

Eighty-three cases having a cervical smear result showing abnormal glandular cells were identified and matched up with the diagnostic histology result. Thirty-four (41.0%) were associated with malignancy and 26 (31.3%) with a cervical intraepithelial lesion without invasion. Thirty-eight (45.8%) had conditions of the cervix as follows: 12 cases had invasive disease of the cervix; nine (10.8%) adenocarcinoma of cervix and three (3.6%) squamous carcinoma of cervix. Nineteen (22.9%) had cervical intraepithelial neoplasia (CIN/SIL) alone and seven (8.4%) had cervical glandular intraepithelial neoplasia (CGIN) +/- CIN. There were 16 (19.3%) cases with malignancies of the uterine corpus and six (7.2%) had a malignancy arising from another primary site. Twenty-three (27.7%) had no malignant or pre-malignant condition. The risk of malignancy was related to age and ranged from 18.2% in those under 35 years to 67.9% in those 55 years and over. A protocol for the management of these cases is described.

How significant is a cervical smear showing glandular dyskaryosis?

Objective: To evaluate the incidence, outcome and predictive value of cytology showing glandular dyskaryosis. Participants: Fifty-seven women with a smear diagnosis of glandular dyskaryosis registered between January 1997 and December 2001. Setting: Colposcopy and cytopathology units in a large district general hospital. Results: Sixty smears in 57 women showing glandular dyskaryosis were identified from a cohort of 135,120 smears, giving an incidence of 0.05%. Hospital records were available for 50 women. Final diagnosis included 13 cases of cervical glandular intraepithelial neoplasia (CGIN), 4 microinvasive cervical adenocarcinomas, 2 undifferentiated tumours, 1 microinvasive squamous carcinoma, 21 cases of CIN and 13 cases of endometrial pathology (8 endometrial cancers). Twelve women had coexistent squamous and glandular disease. Forty-five out of 50 women had significant pathology (positive predictive value 90%). Colposcopy was seen to be of limited value in assessment of smears showing glandular dyskaryosis. Only 1 out of 13 glandular lesions was diagnosed by colposcopy. Conclusion: Smears showing glandular dyskaryosis are associated with significant pathology in 90% of cases and malignancy in 32% of cases. Hence, women with a smear showing glandular dyskaryosis should be referred urgently to a colposcopy clinic and flagged up as suspected cancer. Glandular dyskaryosis should be included in the national referral criteria for suspected gynaecological cancer.

Cervical squamous and glandular intraepithelial neoplasia: identification and current management approaches.

Certain types of human papillomaviruses (HPV) are associated with squamous intraepithelial lesions and cancer and these are termed high-risk. HPV type 16 is detected in approximately half of the high-grade squamous intraepithelial lesions and cancer. Because of the high rate of spontaneous regression of low-grade squamous lesions, follow-up by cytology, colposcopy and possible biopsy appears preferable. Due to the higher rate of progression to malignancy of the high-grade lesions conservative treatment is recommended. One of the most common reasons for persistence relates to the human immunodeficiency virus. Adenocarcinoma in situ is an uncommon disorder and not well identified by cytologic sampling or colposcopic inspection. The diagnosis is made by cone biopsy, the specimen having negative margins for disease. Hysterectomy is the treatment procedure of choice unless fertility is an issue. Excisional methods (particularly electrosurgical loop) can interfere with accurate histological interpretation in some cases of both squamous disease and adenocarcinoma in situ. This paper is available too at: http://www.insp.mx/salud/index.html.

P16INK4A as a marker for cervical dyskaryosis: CIN and cGIN in cervical biopsies and ThinPrep™ smears

Aim: To examine the potential of p16INK4A as a biomarker for dysplastic squamous and glandular cells of the cervix in tissue sections and ThinPrep™ smears. Methods: Immunocytochemical analysis of p16INK4A...

Cytological and histological outcome following a borderline cervical smear

A prospective cytological and histological study was conducted in 627 women with borderline nuclear abnormality to assess its clinical significance in an attempt to reach a sound policy of management. The incidence of dyskaryosis was 10.9% at a 6-month repeat smear after the first borderline smear. The detection of cervical intraepithelial neoplasia (CIN) after two borderline smears increased to 21.15% on cytology and 40.75% on histology. Additionally, there were two cases of glandular intraepithelial neoplasia and one case of adenocarcinoma detected on histology after having borderline nuclear changes reported on cytology; 39.7% of these women with initial borderline cervical smear were under the age of 30 years. The high prevalence of borderline nuclear abnormality in younger women and a poor correlation between these abnormalities and severity of CIN suggest that repeat cytology is a less than optimal management for these women.

Immunohistochemical staining with MIB1, bcl2 and p16 assists in the distinction of cervical glandular intraepithelial neoplasia from tubo-endometrial metaplasia, endometriosis and microglandular hyperplasia.

Preinvasive endocervical glandular lesions, termed cervical glandular intraepithelial neoplasia, are increasing in incidence. The distinction of cervical glandular intraepithelial neoplasia from benign mimics, especially tubo-endometrial metaplasia, endometriosis and microglandular hyperplasia, can be difficult. This study investigates the value of immunohistochemical staining with MIB1, bcl2 and p16 in the distinction of cervical glandular intraepithelial neoplasia from these benign mimics. Immunohistochemical staining using the monoclonal antibodies MIB1, bcl2 and p16 was performed on cases of cervical glandular intraepithelial neoplasia (n = 21), tubo-endometrial metaplasia (n = 13), endometriosis (n = 7) and microglandular hyperplasia (n = 14). With tubo-endometrial metaplasia and microglandular hyperplasia staining with MIB1 was either negative or involved <10% of cells, while with cervical glandular intraepithelial neoplasia the majority of cases (86%) exhibited >10% positive cells. Two cases of endometriosis exhibited a MIB1 index of 10-30% while in the other cases <10% cells stained. With bcl2 the cells of microglandular hyperplasia were negative although there was staining of associated reserve cells in 43% of cases. All cases of tubo-endometrial metaplasia except one and all cases of endometriosis stained diffusely positive with bcl2. Cases of cervical glandular intraepithelial neoplasia were negative or exhibited focal staining. With p16 all cases of cervical glandular intraepithelial neoplasia exhibited diffuse strong positivity, generally involving 100% of cells, while all cases of microglandular hyperplasia were negative. Sixty-two percent of cases of tubo-endometrial metaplasia showed focal positivity, the remainder being negative. Cases of tubo-endometrial metaplasia were never diffusely positive with p16. In three cases of endometriosis there was staining of >50% of cells while the other cases were either focally positive or negative. A panel of antibodies, comprising MIB1, bcl2 and p16, is a useful adjunct to histology in distinguishing cervical glandular intraepithelial neoplasia from tubo-endometrial metaplasia, endometriosis and microglandular hyperplasia. Cases of cervical glandular intraepithelial neoplasia are diffusely positive for p16 and generally exhibit a high proliferation index with MIB1, while bcl2 is negative or, at most, focally positive. Tubo-endometrial metaplasia and endometriosis are characterized by strong diffuse positivity with bcl2 and a low proliferation index with MIB1 (although occasional cases of endometriosis show moderate proliferative activity). p16 is negative or exhibits focal positivity in tubo-endometrial metaplasia but in endometriosis there may be quite widespread positivity. Microglandular hyperplasia shows a low proliferation index with MIB1 and is negative for bcl2 and p16.

How predictive is a cervical smear suggesting glandular neoplasia?

The prevalence of endocervical adenocarcinoma and its precursors has increased, in part due to increased diagnostic awareness of these lesions. To date, limited information has been published regarding the predictive value of glandular abnormalities in cervical smears. This study details the histological follow up of 418 cervical smears showing glandular abnormality, reported in our department over a six year period from 1993 to 1998. Histological follow up was available for 395 of the 418 smears (94.50%). The overall positive predictive value (PPV) for this group of smears was 72.66% for either significant glandular or squamous pathology (at least low grade cervical glandular intraepithelial neoplasia or CIN2 on follow up biopsy), and 55.70% for significant glandular pathology alone. Examination of subcategories of abnormal glandular smear showed that the PPV increased with the degree of abnormality reported within the smears.

Utility of Trichrome and Reticulin Stains in the Diagnosis of Superficial Endometriosis of the Uterine Cervix

Superficial endometriosis of the uterine cervix is a not uncommon lesion and the cells on the cervicovaginal smear shed from it can be easily mistaken for cervical glandular intraepithelial neoplasia (CGIN). The correct diagnosis can not always be easily made on H&E stained tissue sections unless it is suspected. The endometriotic stroma is often misinterpreted as stromal hypercellularity or postinflammatory fibrosis following erosion or ulceration of the cervical mucosa. Moreover, the endometriotic glands may resemble tubo-endometrioid metaplasia of the endocervical glands. This article describes the utility of trichrome and reticulin stains in the diagnosis of superficial cervical endometriosis. The absence of abundant thick collagen bundles and the investment of individual stromal cells by a fine reticulin network within the endometriotic foci are characteristic histologic features. These findings are not observed in the surrounding normal cervical stroma nor in the usual conditions in the differential diagnosis.

Interval cervical cancer following treatment for cervical intraepithelial neoplasia.

The objective of this study was to determine whether those women who developed cervical cancer following treatment for preinvasive disease had common features in their history which could identify those at increased risk of progression and therefore be used to modify management protocols. A retrospective case note review from clinical and histopathologic records was undertaken at a teaching hospital in Wessex, Southern England. The review included 33 women diagnosed with cervical carcinoma between 1985 and 1996 who had previously undergone treatment for cervical intraepithelial neoplasia (CIN) or cervical glandular intraepithelial neoplasia (CGIN). The diagnosis prior to treatment was CIN 3 in 19 cases, CGIN 3 in 2 cases, CIN 2 in 9 cases (97% high grade CIN/CGIN) and CIN 1 in 1 case. At primary treatment, among those treated by knife cone biopsy or Large Loop Excision of the Transformation Zone (LLETZ), and for whom the margins of the treatment specimen were reported, 14 out of 15 had incomplete margins. Local ablation (in which completeness of excision could not be histologically assessed) was performed in 12 cases. In 58% (19/33) of cases, the patient was 40 years or older at the time of initial treatment. Fifteen women had one or more negative smears after treatment, of which only 6 had transformation zone sampling. The interval between treatment of CIN/CGIN and diagnosis of invasion ranged from 8 to 216 months. (mean 40.4 months), with 67% of cases of invasive cancer occurring within 5 years of treatment for CIN/CGIN and 94% within 10 years. Screen detection was achieved in 91% (30/33) of cases with 53% diagnosed while stage 1A. In conclusion, most treatment screen detection of invasive disease at an early (and often microinvasive) stage was achieved for most patients, although a third of patients were diagnosed more than 5 years after initial treatment. The data suggest the need to follow up longer than 5 years when there are risk factors such as incomplete excision of high grade CIN/CGIN and in women over 40 years of age at the time of initial diagnosis.

A clinical review of borderline glandular cells on cervical cytology.

To review the diagnoses and diagnostic pathway of women presenting with borderline glandular cells on cervical cytology. To outline the basis of clinical approach of these women. Retrospective review. Forty-three women referred to the hospital department over a 32-month period. Review of the casenotes for the demographic data, previous cervical cytology and/or histology report, indication for the smear resulting in borderline glandular cells, colposcopic findings, diagnostic and/or treatment procedures, final diagnosis and current status. The average age was 36.7 years. Twenty-four women (56%) had clinically significant lesions: seven women (16%) presented with cancers, of which one was endometrial in origin, and 17 (40%) with intraepithelial neoplasia (CIN and cervical glandular intraepithelial neoplasia (CGIN)). Sixty-seven percent of all clinically significant lesions were of squamous origin. Thirty-seven had histological diagnosis, while six went on to cytological surveillance. Colposcopy was the most significant predictor for clinically significant lesions (P < 0.05). Punch biopsies and loop excisions were diagnostic when based on abnormal colposcopic findings. Brush cytology was appropriate follow up for asymptomatic, premenopausal women with no colposcopic abnormality. In addition, endometrial sampling was recommended in the peri- and postmenopausal women. Borderline glandular cells have a high incidence of clinically significant lesions. Immediate referral for colposcopy and assessment is strongly recommended in women with two borderline glandular smears to avoid delays in potential cancer diagnosis.

The abnormal glandular smear: cytologic prediction, colposcopic correlation and clinical management

We reviewed recent cytological reporting of abnormal glandular cells on cervical smears in order to assess the predictive value of these reports and the contribution of colposcopy in the assessment of these abnormalities. The study consisted of a 5-year retrospective review of the clinical management of 80 women with abnormal glandular cells on a cervical smear, with clinical and histopathological data available for review in the interval 1992-1996. There were two groups of women: (i) those referred with gynaecological symptoms and (ii) those with screen detected abnormalities who were asymptomatic and significantly younger than the first group. The predictive value of a glandular smear for malignancy was 42.5% and for premalignancy 28.8%. The most common lesions detected were cervical intraepithelial neoplasia (CIN) (13), endometrial cancer (13), cervical adenocarcinoma (10) and cervical intraepithelial glandular neoplasia (CIGN) (8). Four cases of endometrial carcinoma presented through screening. In the remainder a variety of benign conditions were identified as responsible for the abnormal smear. Failure to find an explanation for the abnormal smear only occurred in 8.8%. In developing a protocol for abnormal glandular smears, our observations indicate that: (a) those with abnormal bleeding require endometrial sampling; (b) for those with screen detected abnormality, colposcopy is valuable as it is a sensitive predictor of early invasion and can predict glandular abnormality; (c)diseases of the entire genital tract, non-gynaecological viscera and metastatic cancer can generate cytological abnormality; (d) screen detected borderline abnormality in endocervical cells is associated with CIN III.

Application of a marker of ciliated epithelial cells to gynaecological pathology.

BACKGROUND: The assessment of neoplastic disease in gynaecological histopathology can be complicated by the high incidence of metaplasia seen in tissues of the female genital tract. There is a need...

CD44 is a marker of endocervical neoplasia.

The expression of CD44s, CD44v4, CD44v5, CD44v7-8, and CD44v10 was investigated immunohistochemically in a variety of neoplastic cervical lesions. Normal endocervical columnar cells exhibited no reactivity for any of the antibodies, whereas the subcolumnar reserve cells were strongly positive for CD44s, CD44v5, and CD44v7-8. In some cases, positive cells were identified in the stroma surrounding the endocervical glands and adjacent to reserve cells. Cervical glandular intraepithelial neoplasia and adenocarcinoma showed consistent immunoreactivity for CD44v5. There was no significant change in CD44 immunoreactivity in squamous cell carcinoma compared with normal epithelia and cervical intraepithelial neoplasia. These findings lend support to the origin of carcinoma of the cervix from a common progenitor reserve cell and suggest the origin of reserve cells from the stroma. CD44v5 may be useful as a diagnostic marker of endocervical neoplasia and could provide a target for therapeutic approaches directed against specific epitopes.

Relation of cervical glandular intraepithelial neoplasia to microinvasive and invasive adenocarcinoma of the uterine cervix: a study of 121 cases.

AIMS: To examine the relation between invasive adenocarcinoma and its alleged precursor, cervical glandular intraepithelial neoplasia (CGIN), and to assess the management and outcome of CGIN and the validity of...