Endometrial stromal sarcoma (ESS), uterine cellular leiomyoma (UCL), and uterine leiomyosarcoma (ULS) are composed mainly of spindle cells that express similar antigens such as desmin, smooth muscle actin (SMA), and muscle-specific actin (MSA). The differential diagnosis of an ESS versus a uterine smooth muscle tumor or an extrauterine spindle cell sarcoma can be problematic based solely on clinical presentation, histologic assessment, or routine immunohistochemistry. Recently, we reported that normal endometrium, but not myometrium, as well as five cases of ESS, were positive for CD10. We now report the results of CD10 immunohistochemistry in an additional 11 cases of ESS (total 16 cases), 10 cases of UCL, and nine cases of ULS. CD10 immunoreactivity was detected in 16 of 16 cases of ESS (100%) as compared to only 2 of 10 cases of UCL (20%) and none of nine cases of ULS (0%). We compared the utility of CD10 immunoreactivity with that of desmin, SMA, MSA, estrogen receptor (ER), and inhibin in these tumors. Although the majority of cases of UCL and ULS were positive for SMA, MSA, and desmin, a substantial portion of cases of ESS were also positive for SMA, MSA, and desmin. We conclude that in combination with SMA, MSA, and desmin, CD10 is a useful immunohistochemical marker in the differential diagnosis of ESS versus UCL or ULS.

No abstract

A b s t r a c t CD43 expression on B cells is an immunophenotypic feature suggestive of malignancy. In the light of its diagnostic importance, we performed a comprehensive survey ofCD43 expression in various types ofnonHodgkin lymphoma (NHL) The expression of CD43 in non-Hodgkin lymphoma (NHL) has been previously studied (summarized in Arber and Weiss 1 ). However, many of these studies were performed before the widespread use of the heat-induced epitope retrieval technique in immunohistochemistry and were limited by a relatively small number of cases examined. In addition, the use of different anti-CD43 antibodies in these studies has generated a wide range of frequencies of CD43 expression in different types of NHL. Because of its diagnostic importance, a comprehensive reevaluation of CD43 expression in various types of NHL using current immunohistochemical techniques was undertaken to determine the frequency of CD43 expression in NHL, with the intent to maximize the diagnostic usefulness of this marker.Because CD43 + follicular lymphomas have been described only rarely in the literature, 610 another objective of the present study was to further characterize these cases by using CD 10 immunodetection and the polymerase

Immunophenotyping has become common in the diagnosis and classification of acute leukemias and is particularly important in the proper identification of cases of minimally differentiated acute myeloid leukemia (AML-MO). To evaluate the immunophenotype of adult AML, 106 cases were studied by cytochemical analysis and by flow cytometry with a panel of 22 antibodies. The results were compared with the French-American-British (FAB) Cooperative Group classification, as well as with available cytogenetic data on each case. CD45, CD33, and CD13 were the most commonly expressed antigens (97.2%, 95.3%, and 94.3%, respectively). Lymphoid-associated antigens were expressed in 48.1% of cases. CD20 was the most commonly expressed lymphoid antigen (17%), although often expressed in only a subpopulation of leukemic cells, followed by CD7 (16%), CD19 (9.8%), CD2 (7.5%), CD3 (6.7%), CD5 (4.8%), and CD10 (2.9%). Some immunophenotypes correlated with FAB type, including increased frequency of CD2 expression in AML-M3; lack of CD4, CDllc, CD36, CD117, and HLA-DR expression in AML-M3; increased frequency of CD20 and CD36 expression and lack of CD34 expression in AML-M5; increased frequency of CD5 expression in AML-M5a; and increased frequency of CD14 expression in AML-M5b, when compared with all other AMLs (P < .05). When compared with AML-

We studied 48 cases of invasive breast carcinoma for evidence of Epstein-Barr virus (EBV), which is associated with many human malignancies. In situ hybridization studies to detect the presence of EBV-encoded small nonpolyadenylated RNA (EBER)-1 were performed in paraffin sections. Immunohistochemical studies to detect EBV nuclear antigen (EBNA)-1, latent membrane protein (LMP)-1, and the transactivating immediate-early BZLF1 (ZEBRA) protein were also performed in paraffin sections. The presence of EBV genomic DNA was studied by polymerase chain reaction (PCR) amplification using sets of primers flanking the EBNA-4 and the EBV-LMP-1 genes in frozen tissues. Southern blot analysis using a probe flanking the EBV terminal repeat region was then attempted in cases that were PCR-positive. Five of 48 cases (10%) of breast carcinoma showed focal EBER-positive tumor cells. Twelve cases (25%) were positive for EBNA-1 by immunohistochemistry, all but one different from the EBER-positive cases. None of the cases were positive for LMP-1 or ZEBRA protein by immunohistochemistry. PCR studies for EBNA-4 and LMP-1 were each positive in five cases (including three cases in common). However, Southern blot studies successfully performed in all but one of the PCR-positive cases were completely negative. The identification of EBV by any methodology was not correlated with tumor size, grade, or lymph node status. This study demonstrated evidence of EBV infection in tissues involved by invasive breast carcinomas in a significant subset of cases. However, the lack of localization of EBV infection to a significant population of the tumor cells in any case, the negativity by Southern blot hybridization, and the lack of expression of multiple antigens in any case strongly argue against a significant role for EBV in the pathogenesis of breast carcinoma.