An automated segmentation approach for highlighting the histological complexity of human lung cancer.

Lung cancer nodules, particularly adenocarcinoma, contain a complex intermixing of cellular tissue types: incorporating cancer cells, fibroblastic stromal tissue, and inactive fibrosis. Quantitative proportions and distributions of the various tissue types may be insightful for understanding lung cancer growth, classification, and prognostic factors. However, current methods of histological assessment are qualitative and provide limited opportunity to systematically evaluate the relevance of lung nodule cellular heterogeneity. In this study we present both a manual and an automatic method for segmentation of tissue types in histological sections of resected human lung cancer nodules. A specialized staining approach incorporating immunohistochemistry with a modified Masson's Trichrome counterstain was employed to maximize color contrast in the tissue samples for automated segmentation. The developed, clustering-based, fully automated segmentation approach segments complete lung nodule cross-sectional histology slides in less than 1 min, compared to manual segmentation which requires multiple hours to complete. We found the accuracy of the automated approach to be comparable to that of the manual segmentation with the added advantages of improved time efficiency, removal of susceptibility to human error, and 100% repeatability.

BRCA2 alteration is important in clear cell carcinoma of the ovary.

BRCA2 has been shown to play a significant role in hereditary ovarian carcinoma. Several cases of clear cell carcinoma (CCC) of the ovary containing BRCA2 mutations have been identified. We hypothesize that sequence variants of the BRCA2 gene are common in CCC of the ovary. Multiple methods were utilized to detect BRCA2 genetic alterations in a cohort of 13 ovarian CCC. These included an LOH analysis for copy number, real-time and methylation-specific polymerase chain reaction (PCR) to probe for BRCA2 promoter methylation, in addition to protein truncation testing (PTT) gel screening for nonsense BRCA2 mutations, and finally direct gene sequencing to either confirm the nonsense mutations or to detect candidate missense mutations in the remaining tumor samples. Whenever a sequence variation was detected in a tumor sample, the corresponding region was sequenced from a blood sample to determine germline status. Seven BRCA2 sequence variations were identified in 6 of the 13 CCC (46%); three tumors contained an alteration in BRCA2 copy number. Only one subject carried a germline sequence variation that might alter BRCA2 function despite the fact that a family history of breast, ovarian or colon cancer was common in this population. The 5-year disease-specific survival probability for patients with a BRCA2 alteration is 87.5%, compared to only 40% for those patients without a BRCA2 alteration (p = 0.39). Alterations in BRCA2 gene sequence, copy number, or expression are extremely common in CCC and may contribute to a paradoxical better clinical outcome.

Structural and functional connectivity of marine fishes within a semi-enclosed Newfoundland fjord.

The interplay between structural connectivity (i.e. habitat continuity) and functional connectivity (i.e. dispersal probability) in marine fishes was examined in a coastal fjord (Holyrood Pond, Newfoundland, Canada) that is completely isolated from the North Atlantic Ocean for most of the year. Genetic differentiation was described in three species (rainbow smelt Osmerus mordax, white hake Urophycis tenuis and Atlantic cod Gadus morhua) with contrasting life histories using seven to 10 microsatellite loci and a protein-coding locus, PanI (G. morhua). Analysis of microsatellite differentiation indicated clear genetic differences between the fjord and coastal regions; however, the magnitude of difference was no more elevated than adjacent bays and was not enhanced by the fjord's isolation. Osmerus mordax was characterized by the highest structure overall with moderate differentiation between the fjord and St Mary's Bay (F(ST)c.0.047). In contrast, U. tenuis and G. morhua displayed weak differentiation (F(ST) < 0.01). Nonetheless, these populations did demonstrate high rates (< 75%) of Bayesian self-assignment. Furthermore, elevated differentiation was observed at the PanI locus in G. morhua between the fjord and other coastal locations. Interestingly, locus-specific genetic differentiation and expected heterozygosity were negatively associated in O. mordax, in contrast to the positive associations observed in U. tenuis and G. morhua. Gene flow in these species is apparently unencumbered by limited structural connectivity, yet the observed differentiation suggests that population structuring exists over small scales despite high dispersal potential.

Malignant mixed mullerian tumor of the ovary and bilateral breast cancer: an argument for BRCA3, or a coincidental cluster of unconnected cancers?

OBJECTIVES: Malignant mixed mullerian tumors (MMMTs) of the ovary are a rare, aggressive subtype of ovarian cancer without a clear relationship to familial breast-ovarian cancer syndromes. CASE: We present the case of a woman with bilateral breast cancers who subsequently developed a stage IIIc MMMT of the ovary. The patient had a first-degree female relative with breast and ovarian cancer (not MMMT), as well as second- and third-degree female relatives each with bilateral breast cancers. BRCA1 and BRCA2 sequencing of germline DNA revealed no evidence of a heritable mutation. CONCLUSIONS: Ovarian MMMTs may be a hallmark of breast/ovarian cancer secondary to genetic risk independent of classic BRCA1/2 pathways.

Immunohistochemical analysis of hepatocellular and adenocarcinoma in the liver: MOC31 compares favorably with other putative markers.

Distinguishing hepatocellular carcinoma (HCC) from metastatic adenocarcinoma (MA) and cholangiocarcinoma (CC) can, at times, be difficult and sometimes requires immunohistochemical analysis. Recently, MOC31, an antibody directed against a cell surface glycoprotein, has been shown to be useful in separating HCC from both MA and CC; however, no study has compared MOC31 and other frequently used immunostains. We compare MOC31 with other commonly used immunostains for HCC, MA, and CC. Formalin-fixed, paraffin-embedded tissue sections from 57 previously characterized hepatic neoplasms (13 HCC, 14 CC, 3 combined HCC-CC, and 27 MA) were immunostained with antibodies directed against MOC31, cytokeratin (CK) 7, CK20, alpha-fetoprotein (AFP), polyclonal carcinoembryonic antigen, Ber-EP4, and Factor XIII-A. Two pathologists reviewed slides, and positivity was defined as more than 1% of cells staining with the appropriate pattern. Positive MOC31 immunostaining was seen in 0 of 13 HCC, 13 of 14 CC, 3 of 3 HCC-CC, and 27 of 27 MA; the staining was strong and diffuse. CK20 reactivity was observed in 0 of 13 HCC, 2 of 14 CC, 0 of 3 HCC-CC, and 12 of 27 MA; CK7 immunostained 4 of 13 HCC, 13 of 14 CC, 3 of 3 HCC-CC, and 15 of 27 MA; AFP was detected in 4 of 13 HCC and 2 of 3 HCC-CC, whereas all CC and MA were negative; polyclonal carcinoembryonic antigen showed immunoreactivity in 12 of 13 HCC and 3 of 3 HCC-CC in a canalicular pattern, whereas diffuse positivity was identified in 13 of 14 CC and 26 of 27 MA; Ber-EP4 immunostained 1 of 13 HCC, 14 of 14 CC, 2 of 3 HCC-CC, and 26 of 27 MA; and Factor XIII-A was negative in all HCC, CC, and MA. MOC31 expression distinguished HCC from adenocarcinoma in 56 of 57 cases. AFP was specific for HCC but was not sensitive. CK7 and CK20 have limited utility in distinguishing HCC from CC or MA, and Factor XIII-A is not useful. Ber-EP4 staining was similar to MOC31, but one HCC did stain with Ber-EP4. Polyclonal CEA yields similar numerical results as MOC31, but the focal nature of the staining and occasional difficulty in evaluating the pattern can make interpretation problematic. We conclude that MOC31 should be a component of the immunohistochemical panel to distinguish HCC from CC and MA.

MOC-31 aids in the differentiation between adenocarcinoma and reactive mesothelial cells.

BACKGROUND: Evaluation of effusion specimens for the presence of adenocarcinoma often is complicated by the presence of reactive mesothelial cells that can mimic adenocarcinoma. Ancillary studies, in particular immunohistochemistry, can be helpful in making this distinction. MOC-31 is an antibody that recently was reported to be useful in distinguishing adenocarcinoma from mesothelioma in tissue specimens. In this study we examined the utility of this antibody in pleural effusions. METHODS: Eighty-nine archival, formalin fixed, paraffin embedded cell blocks representing 59 adenocarcinomas, 12 other neoplasms (including 6 mesotheliomas), and 18 reactive effusions were retrieved. After protease digestion, recut slides were immunostained with the MOC-31 antibody utilizing a modified avidin-biotin complex technique. Only membrane-based reactivity was considered as positive. RESULTS: In two adenocarcinomas there was insufficient material remaining in the cell block. Among the 57 remaining cases, reactivity was observed in 54 cases. Reactivity also was observed in one of six mesotheliomas and one small cell carcinoma. The remaining cases, including all 18 reactive effusions, were nonreactive. In distinguishing adenocarcinoma from reactive mesothelial cells, the presence of MOC-31 reactivity was found to be 95% sensitive and 100% specific with a positive predictive value of 100% and a negative predictive value of 95%. CONCLUSIONS: MOC-31 is useful in differentiating between adenocarcinoma and reactive mesothelial cells in pleural effusion specimens. Cancer (Cancer Cytopathol)

MOC-31 aids in the differentiation of metastatic adenocarcinoma from hepatocellular carcinoma.

BACKGROUND: Fine-needle aspiration biopsy (FNAB) is frequently used to diagnose mass lesions in the liver. Differentiating metastatic adenocarcinoma from primary hepatocellular carcinoma can be difficult. Despite a number of morphologic criteria, there remain occasional cases in which the cytologic features fail to resolve this differential reliably; in these cases ancillary studies may be useful. Recently, it has been reported that the antibody MOC-31 reliably separates metastatic adenocarcinoma from hepatocellular carcinoma. In this study we examine the utility of MOC-31 in liver FNAB material. METHODS: Thirty-three archival, alcohol-fixed, paraffin-embedded cell blocks representing 17 cases of hepatocellular carcinomas and 16 cases of metastatic adenocarcinoma were retreived. After protease digestion, the sections were immunostained with the antibody MOC-31 (Dako, Carpinteria, CA) utilizing a modified avidin-biotin complex technique. Only membrane-based reactivity was considered positive. RESULTS: In five cases there was insufficient diagnostic material remaining in the cell block for immunohistochemical staining. Among the remaining cases, MOC-31 reactivity was observed in 10 of 12 metastatic adenocarcinomas and 2 of 16 hepatocellular carcinomas. For metastatic adenocarcinoma the presence of MOC-31 reactivity yields a sensitivity of 83%, a specificity of 87%, a positive predictive value of 83%, and a negative predicitive value of 87%. CONCLUSIONS: MOC-31 is useful in separating metastatic adenocarcinoma from hepatocellular carcinoma in FNAB cell block material. Cancer (Cancer Cytopathol)

Altered expression of gamma-glutamyl transpeptidase in human tumors.

Elevated levels of gamma-glutamyl transpeptidase (GGT) activity and intracellular glutathione in tumor cells have been correlated with resistance to several classes of chemotherapeutic drugs. In this study, the first comprehensive analysis of GGT expression in human malignant neoplasms, 451 tumors were immunostained with an antibody directed against a c-terminus peptide of the human GGT protein. Analysis of the immunostaining revealed that GGT was expressed in 22 of 44 lung carcinomas and 16 of 22 ovarian surface epithelial carcinomas, although normal pulmonary and ovarian epithelium are GGT-negative. The tumor samples were obtained from patients before the start of therapy; therefore, GGT was not induced by radiation or chemotherapy. There was no GGT expression in mesotheliomas, Hodgkin's disease, non-Hodgkin's lymphomas, melanomas, basal cell carcinomas, and most soft tissue sarcomas, all of which are derived from GGT-negative cells. Carcinomas arising from some GGT-positive epithelium retained their GGT-positive phenotype. These included renal cell carcinomas, hepatocellular and cholangiocarcinomas, and carcinomas of the prostate and thyroid whereas both pancreatic adenocarcinomas and infiltrating carcinomas of the breast showed a wide range of GGT expression. Further studies are underway to determine whether expression of GGT plays a role in the inherent resistance of some tumors to alkylating agents and other classes of chemotherapeutic drugs.

Altered localization of E-cadherin and alpha-catenin in rat esophageal tumors.

An alteration in the localization of E-cadherin and its associated proteins has been observed in many epithelial neoplasms. No data exist, however, for the expression of these proteins in an animal model system for esophageal cancer or in cultured rat esophageal epithelial cell lines. The present study investigated the localization of E-cadherin and its associated protein, alpha-catenin, in rat esophageal epithelial cell lines of differing tumorigenic potential; in tumors induced after transplantation of these cell lines into syngeneic hosts; and, in esophageal tumors induced in rats by the carcinogen, N-nitrosomethylbenzylamine (NMBA). Immunofluorescent staining of the cultured cell lines revealed staining for both E-cadherin and alpha-catenin at cell-cell boundaries. Western blot analysis confirmed the membrane-bound localization of E-cadherin and alpha-catenin in the cells. However, tumors induced by these cell lines in syngeneic rats showed reduction in the expression of both E-cadherin and á-catenin in the plasma membrane of invasive epithelial cells. Immunohistochemical analysis of NMBA-induced esophageal neoplasms in rats revealed E-cadherin and alpha-catenin to be abnormally expressed in poorly differentiated tumors when compared to well differentiated tumors. These results suggest that the microenvironment may have an important role in regulating the expression of these adhesion molecules in rat esophageal epithelial cells, and that alteration in the cellular localization of E-cadherin and alpha-catenin may be indicative of tumor progression in NMBA-induced rat esophageal cancer.

CD31 immunoreactivity in carcinomas and mesotheliomas.

CD31 is a specific and sensitive marker of endothelial differentiation. Previous reports have described its immunoreactivity in large series of soft tissue neoplasms, as well as its comparison with other available and commonly used endothelial markers. CD31 reactivity in carcinomas or mesotheliomas has been incompletely addressed, however. Hence, we applied anti-CD31 (JC70/A, DAKO, Carpinteria, Calif) to 290 previously characterized neoplasms by using a modified avidin-biotin-peroxidase complex technique following microwave epitope retrieval. Seven carcinomas showed plasmalemmal-based immunoreactivity (2 papillary thyroid carcinomas, 2 mucoepidermoid salivary gland carcinomas, 1 cutaneous adnexal tumor, 1 cutaneous squamous cell carcinoma, and 1 esophageal squamous cell carcinoma); the remaining 283 lesions were negative for this marker. We conclude that anti-CD31 immunostaining in carcinomas and mesotheliomas is rare. These findings support the concept that CD31 is a reliable marker of endothelial differentiation and should be included in diagnostic immunohistochemical panels when vascular tumors enter the differential diagnosis.

Biopsy bag artifact.

We describe a distinctive tissue artifact that results from the use of biopsy bags for processing small impressionable pieces of tissue. In its fully developed form, the artifact produces a tic-tac-toe pattern, while in less pronounced cases it may produce elongated oval spaces in the tissue or a serrated contour at the periphery of the tissue. The artifact was observed in 60% of endometrial specimens, 55% of endocervical curettings, and sporadically in other small specimens. In the endometrial and endocervical specimens, the artifact was focal and did not interfere with the diagnosis. In occasional small lung and prostate specimens, there was focally significant distortion that potentially could compromise the diagnosis.

Anti-MOC-31: a potential addition to the pulmonary adenocarcinoma versus mesothelioma immunohistochemistry panel.

MOC-31 expression has recently been advocated as an immunohistochemical marker for distinguishing mesothelioma from adenocarcinoma in tissue sections. We studied formalin-fixed, paraffin-embedded tissue from 23 pleural mesotheliomas and 23 primary pulmonary adenocarcinomas for immunoreactivity with anti-MOC-31, a human epithelial-related antigen. All of the 23 adenocarcinomas strongly expressed the marker, whereas only one of the mesotheliomas showed weak reactivity. These results demonstrate the usefulness of anti-MOC-31 in differentiating pulmonary adenocarcinoma from mesothelioma.

Primary osteosarcoma of the uterus: report of a case with immunohistochemical study.

A 63-yr-old female presented with uncontrollable uterine bleeding. Pathological evaluation revealed a uterine mesenchymal neoplasm which histologically represented a primary osteosarcoma. Immunohistochemistry was utilized to help support the diagnosis. The English literature regarding this rare uterine neoplasm is briefly reviewed.