Prostate stem cell antigen (PSCA) is a recently de®ned homologue of the Thy-1/Ly-6 family of glycosylphosphatidylinositol (GPI)-anchored cell surface antigens. PSCA mRNA is expressed in the basal cells of normal prostate and in more than 80% of prostate cancers. The purpose of the present study was to examine PSCA protein expression in clinical specimens of human prostate cancer. Five monoclonal antibodies were raised against a PSCA-GST fusion protein and screened for their ability to recognize PSCA on the cell surface of human prostate cancer cells. Immunohistochemical analysis of PSCA expression was performed on paran-embedded sections from 25 normal tissues, 112 primary prostate cancers and nine prostate cancers metastatic to bone. The level of PSCA expression in prostate tumors was quanti®ed and compared with expression in adjacent normal glands. The antibodies detect PSCA expression on the cell surface of normal and malignant prostate cells and distinguish three extracellular epitopes on PSCA. Prostate and transitional epithelium reacted strongly with PSCA. PSCA staining was also seen in placental trophoblasts, renal collecting ducts and neuroendocrine cells in the stomach and colon. All other normal tissues tested were negative. PSCA protein expression was identi®ed in 105/112 (94%) primary prostate tumors and 9/9 (100%) bone metastases. The level of PSCA expression increased with higher Gleason score (P=0.016), higher tumor stage (P=0.010) and progression to androgen-independence (P=0.021). Intense, homogeneous staining was seen in all nine bone metastases. PSCA is a cell surface protein with limited expression in extraprostatic normal tissues. PSCA expression correlates with tumor stage, grade and androgen independence and may have prognostic utility. Because expression on the surface of prostate cancer cells increases with tumor progression, PSCA may be a useful molecular target in advanced prostate cancer.

Patients infected with human immunodeficiency virus are prone to a wide variety of lymphoproliferative disorders. In these patients the clinical presentation of malignant lymphoma often overlaps with that of benign lymphoid proliferations. Both may include lymphadenopathy, splenomegaly, blood and bone marrow dyscrasias, and lymphocyte-rich effusions. Because benign and malignant lymphocyte-rich effusions, as well as effusions from other malignancies, may contain large cells that resemble immunoblasts or Burkitt's cells, cytomorphologic characteristics alone are unreliable for definitive diagnosis of malignant lymphoma. Usual immunotyping panels using antibodies to B- and T-cell markers frequently fail to demonstrate cell lineage in lymphoma cells of patients with acquired immune deficiency syndrome (AIDS). The authors used gene rearrangement to confirm the diagnosis of malignant lymphoma in effusions from three patients with AIDS when routine cell marker studies failed to demonstrate cell lineage or clonality. Use of biotinylated probes eliminated the need for handling radioactive material and enabled performance of studies in a routine immunohistochemistry laboratory.

Cytologic material, including fine-needle aspirations, bronchial brushings, body cavity fluids, and tissue imprints from 39 neuroendocrine and 26 nonneuroendocrine tumors, was stained for chromogranin by the immunoperoxidase technic. Our results suggest that chromogranin is useful in identifying primary as well as metastatic neuroendocrine lesions from a variety of body sites. All lymphoid proliferations and carcinomas, including small-cell anaplastic carcinomas of the lung (oat-cell carcinomas), were negative. Chromogranin appears to be a useful marker in diagnostic cytology. This technic can be applied to routinely prepared cytologic material.

The specificity and sensitivity of HMB-45, an antimelanoma monoclonal antibody, was evaluated in cytologic specimens from extracutaneous melanomas. Melanoma cells in 23 of 25 (92%) cases stained with this antibody. Staining was intense and diffuse in 22 of these melanomas and focal in 1. Amelanotic tumors and tumors with scanty pigment were included. Neither degree of pigmentation, site of primary, cytologic characteristics of tumor, nor source of specimen predicted the immunostaining pattern. Cytopreparations containing benign (6 cases) and malignant (16 cases) cells with which melanoma may be confused constituted the nonmelanoma group. None of the 22 cases in this group stained. Undifferentiated carcinomas, adenocarcinomas, large cell lymphomas, sarcomas, mesothelial hyperplasias, and a benign nerve sheath tumor were included. Effusions, fine-needle aspirates, bronchial material, and imprints were studied. All smears had been fixed and stained by Papanicolaou's method before immunostaining. Excellent cytomorphologic characteristics were preserved, and immunostaining was not affected. HMB-45 antibody is a highly specific and sensitive marker for malignant melanoma in cytologic material.