Search

The Duke experience with 56 vulvar Paget disease patients was analyzed emphasizing pathologic features and controversial issues. Nearly all patients were Caucasian, and their mean age was 69 years. The average length of follow-up was 5.6 years. For each case, the following histologic features were evaluated and their association with disease course was examined: pseudo-invasion, adnexal involvement, signet-ring cells, cytologic atypia, glands formation, epidermal acantholysis, parakeratosis, hyperkeratosis, and chronic inflammation. The recurrence rate after surgical management was 32%, with epidermal acantholysis being the only statistically significant risk factor. Stromal invasion occurred in 10 patients (18%), and was not a statistically significant adverse prognostic indicator, although the single patient who died of the disease had the deepest stromal invasion. Recurrence was more common after resections with positive surgical margins, but this correlation was not statistically significant. Intraoperative frozen section analysis of the margins did not reduce recurrence rate, nor was it useful in attaining permanent free margins. The Paget cells were consistently reactive with cytokeratin-7 and carcinoembryonic antigen and unreactive with S-100 protein, HMB-45, and Mart-1. In addition, the tumor cells were usually positive for mucin stains. This profile helps distinguish vulvar Paget disease from its mimics, Pagetoid squamous cell carcinoma and malignant melanoma.

Struma ovarii that display extraovarian spread or later recurrence is exceedingly rare. Among 88 patients with "malignant" struma ovarii followed for prolonged periods, several features helped to predict the adverse clinical course. Adhesions (graded 2 to 4+), peritoneal fluid (or =1 L) or ovarian serosal rent were worrisome features, occurring in 74% of 27 biologically malignant tumors but only 10% of 61 clinically benign tumors. The size of the strumal component rather than the overall size of the ovarian teratoma also had some predictive value. Tumors with a strumal componentor =6 cm recurred rarely (7%), whereas 33% of the consult and 88% of the literature casesor =12 cm were clinically malignant. Except for a papillary pattern or poorly differentiated cancer, no microscopic feature reliably predicted the clinical outcome, including those typically associated with malignancy in primary thyroid tumors. Among the consult cases, 7% with histologic follicular adenomas and 29% with papillary carcinomas were clinically malignant. Unequivocal vascular invasion was rare, precluding assessment of its effect. Optically clear nuclei, when extensive, were useful to diagnose papillary carcinoma, but were present nevertheless in smaller numbers in both macrofollicular and microfollicular adenomas. Eight tumors confined initially to the ovary (stage 1) recurred. Papillary carcinomas recurred earlier (average 4 y) than follicular adenomatous neoplasms (average 11 y, range: 1-29 y). Overall, the survival rate for all patients was 89% at 10 years and 84% at 25 years, indicating the need for routine long-term follow-up.

Endometrial "hyperplasia," as currently diagnosed, includes the changes caused by an abnormal hormonal state and those caused by a separate category of monoclonal premalignant disease. The appearance of the disease in these 2 functional categories is discontinuous, permitting more specific diagnosis of the condition using the terms "benign endometrial hyperplasia" and "endometrial intraepithelial neoplasia" (EIN), respectively. Benign endometrial hyperplasia involves the entire endometrial compartment and, with protracted estrogen exposure, shows the progressive development of cysts, remodeled glands, vascular thrombi, and stromal microinfarcts. They are best construed as a sequence of changes whereby the appearance at any single time point is uniquely dependent on the preceding combination and the duration of hormonal exposures. In contrast, the premalignant clone of an EIN lesion is characteristically offset from the background endometrium by its altered cytology and crowded architecture. The use of an internal standard for cytology assessment, combined with the distinctive topography of a clonal process, enables the diagnosis of EIN lesions with a long-term cancer risk 45-fold greater than that of their benign endometrial hyperplasia counterparts. The resolution of hormonal and premalignant subsets of traditional "endometrial hyperplasias" is possible using redefined diagnostic criteria, enabling patient therapy to be appropriately matched with the underlying disease mechanisms.

Summary: This is the first report of a silent corticotroph cell pituitary adenoma arising in a struma ovarii. The patient, a 79-year-old woman, was found to have an asymptomatic left-sided adnexal mass confirmed by vaginal sonography to be a complex cystic and solid tumor. Pathological anlysis demonstrated an 8-cm partially cystic struma ovarii in which there was a focus of predominantly basophilic adenohypophyseal cells arranged in a diffuse pattern, adjacent to mature neural tissue containing numerous Herring bodies. The epithelial lining cells of the colloid-filled small and cystic follicles were immunoreactive for thyroglobulin. The pituitary cells were predominantly immunoreactive for adrenocorticotropic hormone, in addition to synaptophysin.

Extragenital metastases to the endometrium are unusual, but several histologic features have been suggested as highly suggestive or even pathognomonic for this diagnosis. We report an endometrial carcinoma with a prominent signet-ring cell morphology and a diffusely permeative pattern of infiltration, features that have been reported as indicating an extragenital metastasis. To the best of our knowledge, this is the first reported case of a signet-ring cell carcinoma of the endometrium. Gynecological pathologists should be aware of this entity because of its potential primary of metastatic signet-ring carcinoma to be endometrium.

This study evaluates the morphologic features of squamous epithelial repair of the uterine cervix, a condition describing a state of regeneration, and compares them with the features of its two histologic mimics, squamous metaplasia and mild dysplasia. The materials examined were from 20 patients with a histologic diagnosis of repair, 42 with cervical biopsy specimens of acceptable quality obtained within 3 weeks of a cervical smear showing repair, and 20 each with squamous metaplasia or mild dysplasia. Specimens with repair disclosed distinctive morphologic characteristics. On low-power magnification, the stroma was chronically inflamed (100%), often floridly (55%). The nuclei were uniform with absent or minimal pleomorphism (90%). The chromatin was bland and evenly distributed (70%). Nucleoli of a bull's eye or macronucleolar appearance (45%) were easily found. Mildly dysplastic epithelium, unlike reparative epithelium, was infrequently associated with an intensely inflamed stroma (20%); its nuclei were pleomorphic (100%) and commonly displayed coarse chromatin (75%) and mitoses (60%). Metaplastic epithelium ws also infrequently associated with an intensely inflamed stroma (10%). Nuclear pleomorphism (10%) and mitotic figures were infrequent (10%), never atypical (0%), and always basally located. Most nuclei had nucleoli, but the majority were small (80%). This study indicates that most cases of repair, mild dysplasia, and metaplasia can be readily distinguished, although due to overlapping features, some cases are difficult to classify as shown by interobserver variability.

This report describes two cases of malignant serous cancers with areas of borderline malignancy, which appear to have arisen within benign glandular inclusions of coelomic origin in pelvic or para-aortic lymph nodes. The patients were 44 and 62 years of age. In both cases the nodes contained benign glandular inclusions lined by a single layer of epithelium which resembled that of tubal epithelium. The location of the glandular epithelium varied from within the fat near the node to intracapsular, subcapsular, or interfollicular positions. The number of glands ranged from few to extensive. In both cases the glandular inclusions disclosed epithelial proliferations, ranging from minor degrees of stratification with formation of small papillae of atypical cells (borderline serous tumor) to almost solid tumor typical of serous cancer. In both cases, the borderline and cancerous tumors exhibited areas of transition which appeared to arise from benign glands. Although benign glandular inclusions of coelomic origin are well documented to occur in pelvic or para-aortic lymph nodes of 5-20% of women and have been considered to be of significance only because of the possibility of the misdiagnosis of cancer, it should now be recognized that the glandular inclusion cysts themselves can become neoplastic.

Introduction Cancer of the ovary and fallopian tube Cancer of the cervix Tumours of the uterus Cancer of the vulva and vagina The management of gestational trophoblastic tumours Molecular biology of gynaecological cancer Surgery Radiotherapy Chemotherapy Palliation of gynaecological malignancy