Moderate Cytologic Atypia Research Articles

Primary Sézary Syndrome Commonly Shows Low-Grade Cytologic Atypia and an Absence of Epidermotropism

The dermatopathologic findings in cases of Sezary syndrome (SS) that arise in patients without a previous diagnosis of mycosis fungoides have not been well characterized. We evaluated the histologic findings in skin biopsy specimens from 31 patients with such primary SS and correlated them with clinical and hematologic parameters at the time of biopsy. The most characteristic histologic finding was the presence of a dermal perivascular lymphoid infiltrate, usually with mild to moderate cytologic atypia and variable numbers of eosinophils; epidermotropism was absent or minimal in 19 cases (61%). Reactive epidermal changes such as spongiosis, parakeratosis, and acanthosis also were present frequently (27 [87%], 17 [55%], 19 [61%] cases, respectively). The number of eosinophils present in skin biopsy specimens correlated with the level of peripheral blood lymphocytosis. In erythrodermic patients or patients with persistent xerosis and pruritus, it is important to carefully evaluate the degree of lymphocyte atypia in the dermal perivascular infiltrate and correlate with blood flow cytometric findings to diagnose primary SS. Many cases will lack the epidermotropism usually seen in mycosis fungoides.

Primary Sézary Syndrome Commonly Shows Low-Grade Cytologic Atypia and an Absence of Epidermotropism

The dermatopathologic findings in cases of Sézary syndrome (SS) that arise in patients without a previous diagnosis of mycosis fungoides have not been well characterized. We evaluated the histologic findings in skin biopsy specimens from 31 patients with such primary SS and correlated them with clinical and hematologic parameters at the time of biopsy. The most characteristic histologic finding was the presence of a dermal perivascular lymphoid infiltrate, usually with mild to moderate cytologic atypia and variable numbers of eosinophils; epidermotropism was absent or minimal in 19 cases (61%). Reactive epidermal changes such as spongiosis, parakeratosis, and acanthosis also were present frequently (27 [87%], 17 [55%], 19 [61%] cases, respectively). The number of eosinophils present in skin biopsy specimens correlated with the level of peripheral blood lymphocytosis. In erythrodermic patients or patients with persistent xerosis and pruritus, it is important to carefully evaluate the degree of lymphocyte atypia in the dermal perivascular infiltrate and correlate with blood flow cytometric findings to diagnose primary SS. Many cases will lack the epidermotropism usually seen in mycosis fungoides.

Sertoli Cell Tumors of the Ovary

Ovarian Sertoli cell tumors are rare, and their morphologic spectrum, behavior, and factors influencing the latter are not clearly established. They may be mimicked by many different tumors, some of them more frequent than Sertoli cell tumors; immunohistochemistry may aid in this differential, but its role has not been analyzed in a large series. We studied the clinicopathologic features of 54 Sertoli cell tumors, including the immunohistochemical profile of 23 of them. The patients, 6 of whom had Peutz-Jeghers syndrome, ranged from 2 to 76 years of age (mean, 30 years). Eleven patients had estrogenic and 4 had androgenic manifestations. The tumors ranged from 0.8 to 30 cm, with the majority being in the range of 4 to 12 cm. They were all unilateral, usually solid, and often yellow. The predominant microscopic pattern was tubular, seen, albeit often only focally, in all tumors; other patterns were cords or trabeculae (28), diffuse (21), pseudopapillary (4), retiform (3), islands or alveolar arrangements (3), and spindled (3). The tubules were solid or hollow with the former being somewhat more common. Delicate septa were occasionally seen and were conspicuous in areas of one tumor. The stroma was abundant in 15 tumors with marked sclerosis in 4. The cells usually had pale to occasionally densely eosinophilic cytoplasm, but 6 tumors were composed of cells with prominent foamy cytoplasm, falling in the category of "lipid-rich" Sertoli cell tumor, and one had cells with clear non-foamy cytoplasm. Forty-four tumors were stage I (42 of them were stage Ia and 2 were stage Ic), 1 was stage II, 3 were stage III, and 6 were not adequately staged. Follow-up was available for 27 patients with stage I tumors, and all were alive and well at last follow-up except for 2 patients with stage Ia and 1 with stage Ic disease. Those 3 patients had pelvic-abdominal recurrences 18, 36, and 9 months, respectively, after the initial diagnosis. Two of the three clinically malignant stage I tumors had moderate to severe cytologic atypia and brisk mitotic activity (>5 or more mitoses/10 high power fields [HPFs]), and one of these had tumor cell necrosis. Among the 10 clinically benign stage I tumors with more than 5 years of follow-up, only 3 had >5 mitoses/10 HPFs, but none had more than mild cytologic atypia and none had tumor cell necrosis. Two of the three patients with stage III disease had follow-up information and one was alive at 16 months and the second developed splenic metastases 2 years after the initial diagnosis. Two of the three stage III tumors had at least moderate cytologic atypia and brisk mitotic activity. Immunohistochemical stains showed positivity for AE1/3-Cam5.2 in 15 of 23 tumors; Epithelial membrane antigen (EMA) was negative in all the tumors. Inhibin was positive in 18 of 22 tumors, calretinin in 10 of 20, CD99 in 19 of 22, vimentin in 17 of 18, smooth muscle actin in 4 of 18, neuron specific enolase in 8 of 16, S-100 in 2 of 20, and chromogranin was negative in all 21 cases studied. Although Sertoli cell tumors usually have a distinctive tubular pattern that facilitates the diagnosis, other patterns may occasionally predominate, causing confusion with various other primary and metastatic ovarian tumors. EMA, inhibin, and chromogranin represent the most helpful triad of immunomarkers serving to exclude two common mimics of Sertoli cell tumors (endometrioid carcinoma [inhibin-; EMA+; chromogranin-] and carcinoid tumor [inhibin-; EMA+; chromogranin+]). Although CD99 and calretinin are often expressed in these tumors, they are much less specific and not as helpful in the differential diagnosis. Most Sertoli cell tumors are stage I, unilateral, cytologically bland, and clinically benign, but occasional examples are high stage, and about 11% of stage I tumors have worrisome histologic features that may portend an adverse outcome. The tumors typically occur in young females, sometimes children who typically present with sexual precocity, and occasional patients have Peutz-Jeghers syndrome.

Chordoma Cutis: An Unusual Malignant Tumor Metastatic to Skin

A seventy‐year‐old woman developed back pain associated with a sacral mass seven years prior to presentation. Biopsy of the sacral mass showed chordoma. The patient underwent two surgical resections because of recurrent disease. She presented after developing a firm, non‐tender skin nodule on the left back. Biopsy of the lesion showed a dermal neoplasm composed of nests of cells in a mucinous stroma. There was moderate cytologic atypia of the cells, and occasional mitotic figures were seen. Physaliphorous cells were present. The immunostains S‐100, cytokeratin, and EMA were all positive. A diagnosis of metastatic chordoma to the skin or chordoma cutis was made. The patient died 15 months later from complications of metastatic disease. Chordomas are rare, representing only 1–4% of malignant bone tumors, which most often occur in the sacral area. Metastasis to the skin is unusual. In one study, 207 cases of chordoma were evaluated, and only 19 had skin involvement. Chordoma cutis has been described as nodules which are firm, blue domed, cystic‐appearing, erythematous, translucent, flesh‐colored, and reddish brown. Cutaneous metastasis of chordoma has been reported to the nose, chest, back, thigh, shoulders, left arm, face, neck, and scalp.

Smooth muscle tumors of the ovary: a clinicopathologic study of 54 cases emphasizing prognostic criteria, histologic variants, and differential diagnosis.

We studied 54 ovarian smooth muscle tumors with an emphasis on histologic criteria for malignancy. Twenty-two leiomyomas were identified, including 7 typical, 11 cellular, 2 mitotically active, 1 with bizarre nuclei, and 1 myxoid. Follow-up ranging from 12 to 240 months (mean, 77.6 months) was available for 14 patients; all were alive with no evidence of disease. Of 26 leiomyosarcomas, including 2 myxoid leiomyosarcomas, most were readily diagnosed by the presence of at least two of the following: moderate or severe cytologic atypia, mitotic rate > or =10 mitotic figures per 10 high power fields, and tumor cell necrosis. Some cytologically atypical tumors demonstrated lesser mitotic activity of 5 to 9 mitotic figures per 10 high power fields, in the absence of tumor cell necrosis. Sixty percent of these were clinically malignant, supporting a diagnosis of leiomyosarcoma in such tumors. Follow-up was available for 21 patients. Seventy-one percent developed recurrent disease at a mean of 19 months, and 62% died of their disease at a mean of 24 months. Four tumors were deemed of uncertain malignant potential, and two that were stage II both recurred in the pelvis. One case of ovarian intravenous leiomyomatosis had a benign outcome at 42 months, as did one case of ovarian leiomyoma with leiomyomatosis peritonealis disseminata at 180 months. Overall, ovarian smooth muscle tumors encompass the same varied histologic spectrum as their uterine counterparts. The main tumors in the differential diagnosis are those in the fibroma/thecoma category, spindle cell carcinomas, and metastatic gastrointestinal stromal tumors.

Leiomyosarcoma of the breast: a pathologic and comparative genomic hybridization study of two cases

Leiomyosarcoma is an extremely rare form of primary breast sarcoma. We present the pathologic and genetic findings of two cases of leiomyosarcoma of the breast. The patients were 44 and 52 years of age and they presented with circumscribed masses of 3.0 and 4.5 cm, (greatest dimension) respectively. Microscopically, the two tumors showed diffuse proliferation of spindle cells with oval and blunt-ended nuclei arranged in short fascicles or bundles. There was moderate cytologic atypia in both cases, and 6 and 12 mitotic figures per 10 high power fields, respectively. No epithelial component was identified. The tumor cells were strongly immunoreactive for markers of smooth-muscle differentiation, including desmin, muscle-specific actin, and smooth-muscle actin. Comparative genomic hybridization analysis showed losses of 10q (two of two cases), 13q (two of two cases), 17p (one of two cases), and gains of 1q (one of two cases) and 17p (one of two cases). The patterns of chromosomal imbalances identified in leiomyosarcoma of the breast are similar to those reported in leiomyosarcoma of soft tissue and uterus and are different from those reported for leiomyoma, indicating that these alterations may be important for development of malignant smooth-muscle tumors regardless of site or organ of origin.

A system for the surgical staging of musculoskeletal sarcoma. 1980.

Historically, an adequate surgical procedure has been the most effective means of treating the majority of primary musculoskeletal sarcomas, and amputation has figured prominently in the surgical armamentarium. 4, 7, 9, 19, 21, 29, 41 The recent evidence that certain chemotherapeutic agents may have significant anti-sarcoma activity 2, 15, 17, 38 and coincident technical advances in irradiation therapy, radiographic localization, and reconstructive surgery have fostered enthusiastic interest in extremity-saving treatments. Almost all such treatments emphasize limb salvage as an alternative to amputation and are usually performed under a protective cloak of adjunctive chemotherapy, irradiation or immunoactive agents. 20, 23, 24, 30, 37, 39 Since neither chemotherapy nor irradiation therapy alone has been shown to assure long-term local control of bulk disease, surgical intervention remains an essential step in the overall management of musculoskeletal sarcomas. 3, 9, 17, 18, 29 Questions concerning the magnitude and timing of the surgical procedure are as unanswered as those relating to the most appropriate use of the adjuncts themselves. Increasingly, the surgeon and his patient are confronted with a bewildering array of therapeutic options, the long-term outcomes of which are unknown. These relatively rare sarcomas increasingly are distributed among a variety of treatment protocols in which multiple parameters differ. This trend necessitates interinstitutional cooperation if sufficient numbers of patients are to be available for the timely evaluation of treatments in clinical use. Such cooperation and even effective interinstitutional communication are seriously hampered by the lack of uniform language, so that meaningful comparison of treatments is currently impossible. Prime factors include the lack of a consistent definition of the surgery performed and a serviceable surgical staging system encompassing bone and soft tissue. Standard terminology will assure that like and unlike treatments are appropriately compared. Although an effective staging system should serve all members of the multidisciplinary team, the biologic behavior of musculoskeletal sarcomas suggests that the most useful staging system will articulate with the surgical procedure.

Müllerian adenosarcoma of vagina arising in persistent endometriosis: report of a case and review of the literature

Background Primary adenosarcoma arising in vaginal endometriosis poses a diagnostic challenge, especially in superficial vaginal biopsies. Case A 56-year-old woman, with a prior diagnosis of ovarian endometriosis, presented with a rapidly enlarging mass of the vaginal vault. Two prior biopsies were benign and showed endometriosis. The third vaginal biopsy revealed benign endometriotic glands cuffed by a cellular stroma with moderate cytologic atypia, a histological appearance diagnostic of Müllerian adenosarcoma. A 16-cm vaginal mass that had infiltrated the pelvic structures was resected. Conclusions Close clinical follow-up of extrauterine endometriosis and clinical–pathologic correlation is necessary. Histological features such as cellular stromal cuffing around benign endometriotic glands are critical in arriving at a timely diagnosis of adenosarcoma in patients with persistent extrauterine endometriosis, even in superficial vaginal biopsies.

The spectrum of micronodular thymic epithelial tumours with lymphoid B-cell hyperplasia.

A rare type of thymoma, micronodular thymoma with lymphoid B-cell hyperplasia, was recently reported by Suster and Moran. Thymic epithelial tumours with a similar pattern but with varied cytological features of the tumour cells are analysed. A total of 11 cases of thymic epithelial tumours characterized by micronodular proliferation of tumour cells separated by abundant lymphoid stroma with prominent germinal centres were reviewed clinicopathologically and examined immunohistochemically. The presence of Epstein-Barr virus (EBV) genome was also examined by in-situ hybridization. Based on the morphology of tumour epithelial cells, cases were subdivided into four groups: group 1 (two cases) having spindle epithelial cells; group 2 (two cases) showing an admixture of spindle and polygonal epithelial cells; group 3 (five cases) having polygonal epithelial cells, with mild to moderate cytological atypia in four cases, and group 4 (two cases) representing lymphoepithelioma-like carcinoma. The degree of cytological atypia and the number of tumour cells positive for MIB-1 and p53 gradually increased towards group 4. The abundant lymphoid stroma in all cases contained many CD20-positive B-cells and CD3 and CD45RO-positive T-cells. CD99-positive immature T-cells were present in all cases of groups 1 and 2 and in most cases of group 3, but not in both cases of group 4 tumours. IgG, IgM and IgD-positive plasma cells and lymphocytes were also present in all cases, more prominent in those of groups 3 and 4. The EBV genome was detected in only a few lymphocytes in five cases. The tumours in this series belong to a distinct category of thymic epithelial tumours and each of the above groups may constitute a spectrum in the continuum of cytological atypia. The aetiological relationship of EBV with these tumours could not be proved. The lymphoid B-cell hyperplasia may result from a host immune response and may suggest a favourable clinical course of this type of tumour.

Cyclin D1 expression in dysplastic nevi: an immunohistochemical study.

Objective We previously surveyed cyclin D1 expression in common acquired nevi, Spitz nevi, and malignant melanomas and reported that benign nevi maintain a zonal pattern of cyclin D1 expression, in contrast with malignant melanomas. Our aim was to extend those observations by examining cyclin D1 expression in dysplastic nevi. Methods Cyclin D1 overexpression in 23 dysplastic nevi was detected by an immunohistochemical technique. The extent of atypia of the nevi was graded as mild, moderate, or severe, using previously established criteria. Results Cyclin D1 overexpression in dysplastic nevi maintained a zonal pattern, similar to Spitz nevi. Cyclin D1 overexpression was greatest in the region of the epidermal-dermal junction and was significantly less prominent in the papillary and reticular dermis, suggesting that cyclin D1 expression is under cell control and correlates with maturation of nevus cells. Cyclin D1 overexpression also correlated with cytologic atypia, as dysplastic nevi with moderate or severe cytologic atypia contained a greater percentage of cyclin D1-positive cells than did nevi with mild atypia. Six dysplastic nevi with many cyclin D1--positive cells were assessed by fluorescence in situ hybridization studies using cyclin D1--specific and chromosome 11 centromeric probes. In all cases, there was no evidence of 11q13 translocation, amplification, or trisomy of chromosome 11. Conclusions Cyclin D1 may be involved in the pathogenesis of dysplastic nevi. Cyclin D1 overexpression does not appear to be explained by cyclin D1 locus amplification or translocation in most cases, and it may be a result of other cell abnormalities that up-regulate the protein level of cyclin D1.

Cyclin D1 expression in dysplastic nevi: an immunohistochemical study.

We previously surveyed cyclin D1 expression in common acquired nevi, Spitz nevi, and malignant melanomas and reported that benign nevi maintain a zonal pattern of cyclin D1 expression, in contrast with malignant melanomas. Our aim was to extend those observations by examining cyclin D1 expression in dysplastic nevi. Cyclin D1 overexpression in 23 dysplastic nevi was detected by an immunohistochemical technique. The extent of atypia of the nevi was graded as mild, moderate, or severe, using previously established criteria. Cyclin D1 overexpression in dysplastic nevi maintained a zonal pattern, similar to Spitz nevi. Cyclin D1 overexpression was greatest in the region of the epidermal-dermal junction and was significantly less prominent in the papillary and reticular dermis, suggesting that cyclin D1 expression is under cell control and correlates with maturation of nevus cells. Cyclin D1 overexpression also correlated with cytologic atypia, as dysplastic nevi with moderate or severe cytologic atypia contained a greater percentage of cyclin D1-positive cells than did nevi with mild atypia. Six dysplastic nevi with many cyclin D1--positive cells were assessed by fluorescence in situ hybridization studies using cyclin D1--specific and chromosome 11 centromeric probes. In all cases, there was no evidence of 11q13 translocation, amplification, or trisomy of chromosome 11. Cyclin D1 may be involved in the pathogenesis of dysplastic nevi. Cyclin D1 overexpression does not appear to be explained by cyclin D1 locus amplification or translocation in most cases, and it may be a result of other cell abnormalities that up-regulate the protein level of cyclin D1.

Mixed Endometrial Stromal and Smooth Muscle Tumors of the Uterus

Uterine tumors composed of a prominent component of smooth muscle (SM) and endometrial stroma (ES) (so-called stromomyomas) have received little attention in the literature. The features of 15 of these tumors, defined as those containing more than 30% of each component, were evaluated. Many of the tumors were referred because of problems in the differential diagnosis. Patient age ranged from 29 to 68 years (mean, 46 years). The tumors ranged from 3 to 27 cm (average 9.6 cm) in diameter, and most were grossly well circumscribed. The sectioned surfaces often had soft, tan-yellow areas admixed with firm, whorled areas. Microscopic evaluation disclosed that nine tumors were well circumscribed, and six had infiltrating tongues typical of endometrial stromal sarcoma (ESS). The endometrial stromal component, which predominated in five cases, typically was characterized by a diffuse growth of closely packed, minimally atypical small cells accompanied by numerous arterioles and was desmin-negative in all cases tested, except for rare desmin-positive cells in three tumors. Five tumors showed sex-cord-like differentiation in these areas. The smooth muscle component, which predominated in seven cases, was composed predominantly of spindle cells in disorganized short fascicles, longer fascicles, or nodules with prominent central hyalinization. This component appeared benign, except in one case with moderate cytologic atypia, focal tumor cell necrosis, and 4 mitotic figures/10 high-power fields. The smooth muscle component was strongly desmin-positive in all the tumors tested. Follow-up of more than 1 year was available for seven patients. Six patients were alive and well, but one tumor with infiltrative borders recurred at 48 months as a pure endometrial stromal sarcoma. Mixed endometrial stromal and smooth muscle tumors should be distinguished from highly cellular leiomyomas, pure endometrial stromal tumors, and "uterine tumors resembling ovarian sex cord tumors," at least until knowledge of their clinicopathologic features is more complete. For treatment purposes, these tumors should be reported as endometrial stromal nodules or as endometrial stromal sarcomas with smooth muscle differentiation and any unusual features of either component recorded in a notation.

B-cell lymphoma of MALT type: a review with special emphasis on diagnostic and management problems of low-grade gastric tumours.

B-cell lymphoma of MALT type: a review with special emphasis on diagnostic and management problems of low-grade gastric tumours.

Histopathologic recognition and grading of dysplastic melanocytic nevi: an interobserver agreement study.

Before the controversies surrounding dysplastic melanocytic nevi are resolved, dermatopathologists must be able to reliably distinguish dysplastic nevi from common acquired nevi and malignant melanoma. To establish whether grading of melanocytic dysplasia has any biologic relevance, dermatopathologists must be able to consistently recognize two or more grades of atypia. We studied the concordance among five dermatopathologists for recognition and grading of 60 nevomelanocytic lesions. Ten cases from each of the following categories of melanocytic proliferation were retrieved from the Massachusetts General Hospital files: 1) common melanocytic nevi, 2) melanocytic nevi with features of dysplastic nevi, 3) dysplastic nevi with slight cytologic atypia, 4) dysplastic nevi with moderate cytologic atypia, 5) dysplastic nevi with severe cytologic atypia, and 6) primary malignant melanoma. The slides were reviewed independently; no discussion of diagnostic criteria preceded the review. Overall concordance for diagnosing dysplastic nevi was 77%, with a kappa statistic of 0.55-0.84. Furthermore, in grading dysplastic nevi, experienced dermatopathologists had a concordance ranging from 35% to 58% (kappa value 0.38-0.47). Those with less experience in grading dysplastic nevi had a concordance of 16-65% (kappa value 0.05-0.24). The five observers in this study reliably distinguished dysplastic nevi from common acquired nevi and malignant melanoma. Further refinement of the criteria for grading of nevo-melanocytic dysplasia and experience in grading are critical for accuracy in subcategorization of dysplastic nevi. Consistent, reproducible subcategorization of dysplastic nevi is a requisite before the issue of biologic or prognostic relevance of grading (dysplastic nevi) can be addressed. This study supports the validity of existing criteria for the diagnosis of dysplastic nevi because the problems in diagnosis were at the limits of the spectrum, namely, discrimination of slightly atypical dysplastic nevi from common nevi and severely atypical dysplastic nevi from radial growth phase melanoma.

Histopathologic Recognition and Grading of Dysplastic Melanocytic Nevi: An Interobserver Agreement Study

Before the controversies surrounding dysplastic melanocytic nevi are resolved, dermatopathologists must be able to reliably distinguish dysplastic nevi from common acquired nevi and malignant melanoma. To establish whether grading of melanocytic dysplasia has any biologic relevance, dermatopathologists must be able to consistently recognize two or more grades of atypia. We studied the concordance among five dermatopathologists for recognition and grading of 60 nevomelanocytic lesions. Ten cases from each of the following categories of melanocytic proliferation were retrieved from the Massachusetts General Hospital files: 1) common melanocytic nevi, 2) melanocytic nevi with features of dysplastic nevi, 3) dysplastic nevi with slight cytologic atypia, 4) dysplastic nevi with moderate cytologic atypia, 5) dysplastic nevi with severe cytologic atypia, and 6) primary malignant melanoma. The slides were reviewed independently; no discussion of diagnostic criteria preceded the review. Overall concordance for diagnosing dysplastic nevi was 77%, with a kappa statistic of 0.55-0.84. Furthermore, in grading dysplastic nevi, experienced dermatopathologists had a concordance ranging from 35% to 58% (kappa value 0.38-0.47). Those with less experience in grading dysplastic nevi had a concordance of 16-65% (kappa value 0.05-0.24). The five observers in this study reliably distinguished dysplastic nevi from common acquired nevi and malignant melanoma. Further refinement of the criteria for grading of nevo-melanocytic dysplasia and experience in grading are critical for accuracy in subcategorization of dysplastic nevi. Consistent, reproducible subcategorization of dysplastic nevi is a requisite before the issue of biologic or prognostic relevance of grading (dysplastic nevi) can be addressed. This study supports the validity of existing criteria for the diagnosis of dysplastic nevi because the problems in diagnosis were at the limits of the spectrum, namely, discrimination of slightly atypical dysplastic nevi from common nevi and severely atypical dysplastic nevi from radial growth phase melanoma.

Well-Differentiated Thymic Carcinoma

Based on a study of 26 cases, the well-differentiated thymic carcinoma is described as a distinct organotypical carcinoma of the thymus with low-grade malignancy. It is characterized by a predominance of epithelial cells with usually low mitotic rate, an epidermoid differentiation with slight to moderate cytological atypia, the constant presence of interepithelial immature cortical thymocytes, lobular growth, and formation of epithelial palisades around perivascular spaces. The tumor occurs at age 14 to 76 years in both sexes. An association with myasthenia gravis is found in 77% of the patients, and 83% of the tumors show invasion of adjacent organs or endothoracic metastasis at primary operation. This rate is higher than in cortical thymomas (47%) but lower than in other thymic carcinomas (92%). Two of 18 patients with follow-up died of tumor recurrence and pleural metastasis. Well-differentiated thymic carcinoma can be related to cortical thymoma by common morphological features and a similar immunophenotype of epithelial cells. It must be differentiated from the lymphocyte-depleted cortical thymomas after corticosteroid treatment and from the benign epithelial-rich medullary thymomas.

Stromal neoplasms of the rectum and anal canal

Nineteen mural-based stromal tumors of the rectum and anal canal were reviewed, with the objective of delineating pathologic features discriminative of malignancy in these uncommon neoplasms. Ten locally excised tumors failed to recur during long-term follow-up and were considered benign. All occurred in the submucosa and ranged in size from 1.0 to 7.0 cm (mean, 2.1 cm). Three were sparsely cellular; seven had the appearance of gastric-type cellular leiomyomas. All lacked nuclear atypia and displayed mitotic activity not exceeding 1 mitosis/50 high-power microscopic fields. In contrast, of nine tumors exhibiting malignant behavior, eight (89%) were located in the muscularis propria. Their mean size was 4.5 cm (range, 1.6 to 11 cm). Necrosis was present in six tumors (67%). Seven sarcomas retained a cellular leiomyomatous appearance but exhibited moderate cytologic atypia. Mitotic counts ranged from 5 to 58 mitoses/50 high-power microscopic fields. Three locally excised sarcomas recurred in the rectum at 2, 2, and 7 years. In five patients tumor recurred in the pelvis. Five patients died of disease 0.67, 1.2, 3, 5, and 11 years postdiagnosis. One patient died with sarcoma at 31 years. Three patients are without evidence of recurrent neoplasm 5, 5, and 7 years postresection. Our data indicate that not all anorectal, mural-based stromal neoplasms are a priori malignant. While location within the muscularis propria, size, nuclear atypia, and tumoral necrosis correlate with malignancy, mitotic activity is the cardinal indicator of sarcomatous behavior in stromal neoplasms of the rectum and anal canal.

Well-differentiated thymic carcinoma: a clinico-pathological study.

Well-differentiated thymic carcinoma (WDTC) is a recently described epithelial tumour of the thymus previously classified as cortical or predominantly epithelial thymoma. The authors have reviewed a series of 15 cases of WDTC with the aim of further defining the clinicopathological features of this neoplasm. Histologically, the number of lymphocytes was always low; perivascular spaces and epithelial palisading around blood vessels and/or along fibrous septa were prominent features; 6 cases (40%) were associated with areas of typical cortical thymoma. All cases showed slight to moderate cytological atypia and nuclear grooving was frequently detected. Mitotic activity was variable but usually low. Clinically, all but 3 cases (80%) were invasive at surgery; myasthenia gravis was present in 9 cases (60%); 5 patients (33.3%) died due to disease and 2 additional patients (13.3%) had tumor recurrence. Our study indicates that WDTC has fairly distinctive clinicopathological features and that it is histologically and histogenetically related to cortical thymoma. The definition "well-differentiated carcinoma" is justified because of low-grade cytological atypia and retention of some organotypical histological features, in a tumour otherwise often displaying aggressive and sometimes clear-cut malignant clinical behaviour.

Sclerosing Adenosis of the Prostate Gland

Eleven cases of sclerosing adenosis of the prostate gland, a recently reported uncommon pseudoneoplastic lesion with characteristic histological, histochemical, and immunohistochemical features, are described. The well-circumscribed cellular lesions were composed of variably sized and shaped, often compressed, glands and small clusters of epithelial cells embedded in a cellular, often myxoid stroma. Mild cytologic atypia was occasionally present, and one case had moderate cytologic atypia. A distinct basement membrane often surrounded the glands and clusters. Luminal acid mucin was typically present. Keratin-positive basal cells were present in the glands and as spindle cells in the stroma. The basal cells were also immunoreactive for S-100 and muscle-specific actin, suggesting myoepithelial differentiation. Clinical follow-up has shown no evidence of prostatic carcinoma. The available evidence suggests that sclerosing adenosis of the prostate gland is a benign lesion with distinctive features that should enable it to be distinguished from prostatic adenocarcinoma.

Lymphoid Infiltrates of the Stomach

Low-grade gastric lymphomas may be difficult to distinguish from benign inflammatory infiltrates on endoscopic biopsy specimens. Recent reports have suggested that so-called lymphoepithelial lesions (infiltration of lymphocytes into glandular epithelium) are characteristic of primary gastric lymphomas. We evaluated the presence and prominence of lymphoepithelial lesions and other histologic criteria in 25 low-grade gastric lymphomas (21 primary) and 58 benign inflammatory infiltrates to evaluate their utility in distinguishing benign from malignant gastric lymphoid infiltrates. The following features were associated only with lymphomas: (a) prominent (2-3 +) lymphoepithelial lesions (eight of 24 versus none of 58; p less than .0001); (b) Dutcher bodies (three of 25 versus none of 58, p = .05); and (c) moderate cytologic atypia (nine of 25 versus none of 58, p less than .0001). One or more of these features was seen in 18 of 25 gastric lymphomas (72%). Features more often associated with, but not limited to, lymphomas were dense (2-3 +) lymphoid infiltrates (25 of 25 versus five of 58, p less than .0001), rare or questionable lymphoepithelial lesions (11 of 24 versus 17 of 58, p = .01), muscularis mucosae invasion (20 of 20 versus 20 of 47, p less than .0001), ulceration (12 of 24 versus five of 58, p less than .0001), and mild cytologic atypia (eight of 25 versus six of 58, p less than .005). Germinal centers, crypt abscesses, and reactive epithelial atypia were seen with equal frequency in both types of infiltrate. Acute inflammation (2-3 +) was associated more often with inflammatory infiltrates (two of 25 versus 27 of 58, p less than .001). Our results suggest that dense lymphoid infiltrates with either prominent lymphoepithelial lesions, moderate cytologic atypia, or Dutcher bodies are highly suggestive and may be diagnostic of lymphoma. This constellation of findings is present in about 70% of endoscopic biopsy specimens of low-grade gastric lymphoma. In addition, the majority of cases of primary low-grade gastric lymphoma have morphologic, immunophenotypic, and clinical features that justify their inclusion in the category of low-grade lymphomas of mucosa-associated lymphoid tissue, whereas a minority are examples of lymphomatous polyposis of the gastrointestinal tract (centrocytic lymphoma).