Temporal variation in the distribution of hyaluronic acid, CD44s, and CD44v6 in the human endometrium across the menstrual cycle.

Tissues undergoing rapid growth and regeneration contain hyaluronic acid (HA) as a prominent component of the extracellular matrix. The physiologic role of HA is partly mediated by its relationship with CD44, its major cell surface receptor. Given the extensive remodeling of the endometrium during the menstrual cycle, the authors sought to determine whether these changes are related to the levels of HA, CD44s, and CD44v6 in the endometrium. Archival paraffin embedded cell blocks from 10 cases of proliferative endometrium and 20 cases of secretory endometrium were retrieved from the surgical pathology files. Specimens from the secretory phase were subdivided into three categories: early secretory (day 15-18), mid-secretory (day 19-23), and late secretory (day 24-28). All cases were stained for hyaluronic acid, CD44s, and CD44v6. Sections from umbilical cord, tonsil, and squamous cell carcinoma served as positive controls for HA, CD44s, and CD44v6, respectively. Positive staining was defined as droplet to diffuse intracytoplasmic or extracellular staining for HA and uniform membranous staining for CD44. During the proliferative phase, the endometrial glands and the stroma were both negative for CD44s and CD44v6 in all cases. In the secretory phase, the endometrial glands were negative for CD44s in all cases, but CD44v6 was expressed in 12 (60%) of cases. In contrast, the stromal cells expressed CD44s in 18 (90%) cases and were negative for CD44v6 in all cases. HA staining was present in the endometrial stroma throughout the menstrual cycle but was most intense (3+) and diffuse during the midsecretory phase. There was perivascular staining for HA throughout the cycle; it was most intense adjacent to the spiral arterioles in the secretory phase. These data indicate temporal and geographic differences in HA and CD44 staining in the endometrium in concert with the menstrual cycle. The timing of peak staining of HA and CD44s in the stroma and the upregulation of CD44v6 in secretory glands are coincident with the period in which the endometrium is most receptive to embryo implantation. Whether these changes are mere hormonal consequences or actually help modulate the cyclical changes in the endometrium warrants further study.

GLUT-1 is preferentially expressed in atypical endometrial hyperplasia and endometrial adenocarcinoma.

The facilitative transport of monosaccharides in human cells is accomplished by a family of transmembrane proteins, GLUT-1 to GLUT-7, that differ in their tissue distribution, affinities for specific monosaccharides, and physiologic regulation. GLUT-1, a high-affinity glucose transporter, is normally expressed in erythrocytes, the perineurium of peripheral nerves, and capillary endothelial cells of the blood-brain barrier. Although the aberrant expression of GLUT-1 has been reported in a wide spectrum of epithelial malignancies, its possible correlation with the malignant transformation of endometrial epithelium has not been clearly established. The purpose of this study was to evaluate the extent to which benign, hyperplastic, atypical, and malignant endometrial epithelia express GLUT-1. The authors examined the IHC expression of GLUT-1 in cases of proliferative endometrium (n=12), secretory endometrium (n=10), endometrial polyps (n=10), adenomyosis (n=18), simple hyperplasia (n=14), complex hyperplasia without atypia (n=17), complex hyperplasia with atypia (n=17), and adenocarcinoma (n=31). Positive staining was defined as distinct, linear membrane staining, particularly at cell-cell borders. Cells that showed only cytoplasmic staining were considered negative. The percentages of positive cells and staining intensity were assessed in a semiquantitative fashion and scored (1+ to 3+). All cases from proliferative endometrium, secretory endometrium, adenomyosis, and simple hyperplasia and 90% (9/10 cases) of the endometrial polyps were negative for GLUT-1. GLUT-1 was expressed in 24% (4/17 cases) of complex hyperplasia without atypia, 71% (12/17 cases) of complex hyperplasia with atypia, and 90% (28/31 cases) of adenocarcinomas. The extent of staining ranged from occasional positive foci to extensive multifocal staining. GLUT-1 positivity increased in intensity as the distance of tumor cells to stroma increased. The authors conclude that GLUT-1 is preferentially expressed in complex hyperplasia with atypia and in adenocarcinoma and that GLUT-1 immunostaining is useful in distinguishing benign hyperplasia from hyperplasia strongly associated with malignancy. GLUT-1-mediated glucose transport may allow hypoxic tumor cells distant from stromal blood vessels to survive through glycolysis. These data suggest that the expression of GLUT-1 transporter may be closely related to the malignant transformation of epithelial endometrial tumors by supporting their increased need for glucose metabolism.

Diagnostic utility of GLUT-1 expression in the cytologic evaluation of serous fluids.

OBJECTIVE: To evaluate the extent to which adenocarcinomas in body cavity fluids express GLUT-1 in comparison to currently available markers for adenocarcinomas. STUDY DESIGN: Archival paraffin-embedded cell blocks of serous fluids from 25 cases of benign effusions containing reactive mesothelial cells and 39 cases of malignant effusions with metastatic adenocarcinoma (11 ovarian, 11 pulmonary, 9 gastrointestinal and 8 breast) were retrieved from the surgical pathology files. All cases were stained with antibodies for GLUT-1, Ber-Ep4, B72.3 and CEA. Positive staining was defined as distinct linear membrane staining for GLUT-1 and Ber-EP4, cytoplasmic staining for CEA, and cytoplasmic or membrane staining for B72.3. Strong staining in at least 10% of the tumor cells was required in order to consider the case positive for the particular marker. RESULTS: GLUT-1 was expressed in 72% (28 of 39) of cases of malignant effusions: 100% (11 of 11) from the ovary, 91% (10 of 11) from the lung, 67% (6 of 9) from the gastrointestinal tract and 12% (1 of 8) from the breast. None (0 of 25) of the benign effusions expressed GLUT-1. Malignant effusions expressed CEA in 74% (29 of 39), Ber-Ep4 in 85% (33 of 39), and B72.3 in 62% (24 of 39). Benign effusions expressed CEA in 3 cases and B72.3 in 2 cases. CONCLUSION: GLUT-1 is a useful marker that can be applied to cytologic specimens. It can be used as a reliable component of an antibody panel to distinguish reactive mesothelial cells from metastatic adenocarcinoma in particular adenocarcinomas of body cavity effusions, in particular adenocarcinomas of ovarian and pulmonary origin.

Perineural invasion in adenoid cystic carcinoma: Its causation/promotion by brain-derived neurotrophic factor.

Adenoid cystic carcinoma (ACC) is a common salivary gland neoplasm with a lengthy clinical course and often late local recurrences after surgical resection. This is accounted for histologically by its infiltrative capacity and distinct propensity for perineural invasion. The expression of biological markers may help explain the clinicopathologic course in ACCs. Brain-derived neurotrophic factor (BDNF) is a growth factor known to be involved in neurogenesis. The aim of this study is to elucidate the expression of BDNF in ACCs, which is currently unknown. Twenty-nine cases of primary ACCs of the head and neck were immunostained to analyze BDNF protein expression. Staining intensity was described as focal or diffuse and graded on a 3-tiered scale. The study group comprised 20 adult females (age 30 to 78) and 9 adult males (age 22 to 70). Sites of involvement included the parotid gland (6 cases), nasopharynx (5), maxilla (4), palate (3), trachea (3), submandibular gland (2), buccal mucosa (2), mandible (1), tongue (1), lacrimal gland (1), and temporal region (1). All tumors exhibited diffuse cytoplasmic staining; 11 cases were classified as 1+ intensity, 12 cases as 2+, and six cases as 3+. Based on the results presented here, BDNF is unformly expressed by ACC and may play a causative role in its predilection for perineural invasion. ACCs may display neurogenesis with high levels of BDNF expression in some tumors. Further studies are warranted to gain better understanding of this possible relationship.

Diagnostic use of muscle markers in the cytologic evaluation of serous fluids.

Differentiating reactive mesothelial cells from malignant mesotheliomas and from adenocarcinomas can be diagnostically challenging when based solely on the morphologic examination of serous fluids. The diagnosis even after the use of standard immunohistochemical stains may at times be inconclusive because of the variable reactivity of mesothelial cells for these markers. Pathologists and cytologists underutilize reactivity for desmin, a feature of mesothelial cells apparently not shared by adenocarcinomas. The purpose of this study was to evaluate the extent to which mesothelial cells express muscle differentiation and to assess the diagnostic utility of muscle markers in distinguishing reactive mesothelial cells from malignant mesotheliomas and adenocarcinomas. Archival paraffin-embedded cell blocks of serous fluids from 24 cases of reactive mesothelial cells, 14 cases of malignant mesothelioma, and 56 cases (14 cases from each) of metastatic adenocarcinoma from the lung, breast, ovary, and gastrointestinal tract were retrieved. Five cases of omentum with unremarkable mesothelial cells were also included in the study. All cases were stained for desmin, actin, myoglobin, and myogenin and evaluated independently by two observers. Strong cytoplasmic reactivity for desmin was noted in 22 of 24 cases (92%) of reactive mesothelial cells. The reactive mesothelial cells did not express actin, myoglobin, or myogenin. All cases of malignant mesothelioma and metastatic adenocarcinoma were negative for the four muscle markers. The mesothelial lining and scattered subserosal cells in the omental sections were positive for desmin. Because desmin was expressed only in benign mesothelial cells, it may serve as a reliable marker in distinguishing reactive mesothelial cells from mesothelioma or from adenocarcinoma. Awareness of this staining pattern is also important to avoid pitfalls when evaluating body fluid specimens from patients with a history of tumors expressing muscle differentiation.

Lipomatous variant of nasopharyngeal angiofibroma: a case report.

Nasopharyngeal angiofibroma is an uncommon benign vascular tumor that occurs in adolescent boys and young men. We describe an unusual histological variant of this entity characterized by a prominent adipose tissue component. The patient was a 35-year-old man with a 20-year history of nasal obstruction. Results of physical examination and angiography demonstrated a nasal mass consistent with a nasopharyngeal angiofibroma. We discuss the differential diagnoses pathologically for this lesion.

Expression of hyaluronic acid and its receptors, CD44s and CD44v6, in normal, hyperplastic, and neoplastic endometrium.

The interaction between epithelial tumor cells and their surrounding stroma is important in tumor progression and metastasis. This is accomplished through a number of transmembrane receptors that interact with stromal extracellular matrix molecules. One of these receptors, CD44, binds to extracellular matrix component hyaluronic acid (HA). The purpose of this study was to evaluate the significance of HA, CD44s, and CD44v6 in benign, hyperplastic, atypical, and malignant endometrial epithelia. Archival paraffin-embedded cell blocks from proliferative endometrium (n = 11), secretory endometrium (n = 12), simple hyperplasia (n = 13), complex hyperplasia without atypia (n = 9), complex hyperplasia with atypia (n = 17), and adenocarcinoma (n = 21) were stained for HA, CD44s, and CD44v6. HA was detected throughout the normal menstrual cycle but was more intense during the secretory phase. Only during the secretory phase was CD44s expressed in the stromal cells in 11 cases (92%), whereas CD44v6 was detected in glandular epithelium in 9 (75%). CD44s was expressed in the glandular epithelium in 2 (15%) cases of simple hyperplasia, 4 (44%) of complex hyperplasia without atypia, 14 (82%) of complex hyperplasia with atypia, and in 16 (76%) of adenocarcinoma. CD44v6 was expressed in the glandular epithelium in 1 (11%) case of complex hyperplasia without atypia, 17 (100%) cases of complex hyperplasia with atypia, and in 18 (86%) cases of adenocarcinoma, but in none of the cases of simple hyperplasia. The endometrial stromal cells expressed CD44v6 in 1 (8%) case of simple hyperplasia, 6 (67%) of complex hyperplasia without atypia, 8 (47%) of complex hyperplasia with atypia, and in 3 (14%) of adenocarcinoma. We concluded that in the normal menstrual cycle, the timing of peak staining of HA and CD44s in the stroma and the up-regulation of CD44v6 in secretory glands are coincident with the period in which the endometrium is most receptive to embryo implantation. HA is more abundant in the stroma adjacent to the tumor, suggesting that interactions between tumor cells and stromal HA promote tumorigenesis. With progression from hyperplasia and with increasing atypia to adenocarcinoma, levels of stromal HA, glandular CD44v6, and glandular and stromal CD44s all increase. Thus, HA and CD44 are both involved in the development and progression of endometrial cancer.

Prognostic factors in malignant mixed tumors of the salivary gland: correlation of immunohistochemical markers with histologic classification.

Malignant mixed tumor of salivary glands is a rare tumor whose variable behavior and prognosis are related for the most part to the clinical stage and histologic grade of the carcinomatous component. The purpose of this study is to predict prognosis by comparing the histologic grading and subclassification of the carcinomatous component with the immunohistochemical reactivity for E-cadherin, P53 mutation protein, and cellular proliferation (Ki67). Stains were performed on formalin-fixed paraffin-embedded tissue sections from 18 cases of malignant mixed tumor. Clinical follow-up was obtained for each patient. Regional lymph node and distant organ metastases were the criteria for poor prognosis. Of seven cases with lymph nodes metastasis, five were high-grade tumors (with one subsequent death from brain metastasis) and two were low-grade. Of the eight high-grade tumors, positivity for Ki67, p53, and E-cadherin were noted in six, four, and two cases, respectively. In contrast, of the 10 low-grade tumors, two stained with Ki67, five with p53, and none with E-cadherin. Most notably, all seven metastatic cases (as opposed to only one of 11 nonmetastatic tumors) had Ki67 reactivity of more than 10%. We conclude that malignant mixed tumor represents a spectrum of malignancies in which the clinical behavior is closely related to the carcinomatous element. In addition to histologic grading, Ki67 is a useful prognostic marker in the evaluation of malignant mixed tumor while p53 and E-cadherin appear to be of limited value.

Are epidermal growth factor and transforming growth factor responsible for pseudoepitheliomatous hyperplasia associated with granular cell tumors?

Granular cell tumors (GCT) are uncommon benign neoplasms that have a predilection for the head and neck region. These tumors can frequently be associated with pseudoepitheliomatous hyperplasia (PEH), which in turn may be mistaken for squamous cell carcinoma. Although epidermal growth factors are overexpressed in squamous cell carcinomas of the head and neck, their presence in PEH, especially its relation to GCT, is unknown. We hypothesize that the expression of epidermal growth factor receptor (EGFR), epidermal growth factor (EGF), and transforming growth factor alpha (TGFalpha) in GCT have a role in the development of PEH overlying some GCT. Sections from 13 cases of GCT (five with overlying PEH) were examined histologically and evaluated immunohistochemically using monoclonal antibodies for EGFR, EGF, and TGFalpha. These were compared with nine cases of PEH independent of GCT. Two of five GCT with overlying PEH and two of six GCT without overlying PEH stained positively for TGFalpha. None of the GCT stained with EGFR or EGF. All cases of PEH, whether or not associated with GCT, were reactive for EGFR and EGF. Four of the five cases of PEH overlying GCT stained with TGFalpha. The staining pattern and intensity of all three antibodies were comparable to that of the adjacent normal squamous mucosa. Among the three antibodies, only TGFalpha in GCT appears to be related to the development of PEH. Epidermal growth factor receptor and EGF do not seem to be directly involved. The reason of PEH formation associated with GCT in the absence of growth factors is unknown.

Melan-A as a useful diagnostic immunohistochemical stain for the diagnosis of primary sinonasal melanomas.

BACKGROUND: Sinonasal melanomas are rare neoplasms whose diagnosis may require confirmatory immunohistochemical stains. S-100 protein and HMB-45, the stains most commonly used, have varying sensitivities and specificities. Melan-A, a more recent melanoma-specific marker, may prove helpful when S-100 protein and HMB-45 stains are negative or equivocal. METHODS: Seven cases of sinonasal melanoma were assessed for reactivity with Melan-A, S-100 protein and HMB-45. RESULTS: The study group consisted of two women and five men ages 40 to 83. Six of the neoplasms were strongly positive for S-100 protein. One case was negative for S-100 protein and HMB-45 but positive for Melan-A. HMB-45 staining varied between diffusely positive (three cases), focally positive (two cases), and negative (two cases). All cases were positive for Melan-A either diffusely (four cases) or focally (three cases). CONCLUSION: Because Melan-A can be positive in cases that are S-100 protein or HMB-45 negative, it is a useful component in the immunohistochemical panel for the diagnosis of sinonasal melanomas.