CD10 is expressed in cutaneous clear cell lesions of different histogenesis.

BACKGROUND: CD10, the Common Acute Lymphoblastic Leukemia Antigen, is a neutral endopeptidase commonly used as a marker of early B-cell differentiation in the classification of lymphomas. Neoplasms of other histogenesis may express CD10, including renal cell carcinoma. Renal cell carcinoma metastatic to the skin (MRCC) can simulate other more common clear cell lesions in which expression of CD10 has not been described. METHODS: Fifty-two cutaneous clear cell lesions including xanthomas (CX), xanthelasmas (XA), xanthogranulomas (XG), balloon cell nevi (BCN), nodular/clear cell hidradenomas (CCH), and MRCC were examined by immunohistochemistry for the expression of CD10, noting frequency and pattern of labeling. RESULTS: CD10 was expressed in 32/35 of the xanthomatous lesions (CX, XA, and XG), 3/3 MRCC, but only 2/8 BCN and 2/6 CCH. BCN and CCH expressed CD10 in fewer than 10% of the clear cells, whereas all MRCC and most xanthomatous lesions had labeling in greater than 10% (p < 0.001). Xanthomatous lesions exhibited a predominantly membranous pattern of labeling compared to the cytoplasmic pattern of MRCC (p < 0.025). CONCLUSIONS: Cutaneous clear cell lesions of different histogenesis express CD10, limiting its use as a specific diagnostic marker for MRCC. Among other clear cell lesions, however, BCN and CCH have a lower frequency of labeling than does MRCC, and xanthomatous lesions show a membranous pattern compared to the cytoplasmic pattern of MRCC, BCN, and CCH. This latter observation may be indicative of altered protein function or trafficking.

Reactive stroma in human prostate cancer: induction of myofibroblast phenotype and extracellular matrix remodeling.

PURPOSE: Generation of a reactive stroma environment occurs in many human cancers and is likely to promote tumorigenesis. However, reactive stroma in human prostate cancer has not been defined. We examined stromal cell phenotype and expression of extracellular matrix components in an effort to define the reactive stroma environment and to determine its ontogeny during prostate cancer progression. EXPERIMENTAL DESIGN: Normal prostate, prostatic intraepithelial neoplasia (PIN), and prostate cancer were examined by immunohistochemistry. Tissue samples included radical prostatectomy specimens, frozen biopsy specimens, and a prostate cancer tissue microarray. A human prostate stromal cell line was used to determine whether transforming growth factor beta1 (TGF-beta1) regulates reactive stroma. RESULTS: Compared with normal prostate tissue, reactive stroma in Gleason 3 prostate cancer showed increased vimentin staining and decreased calponin staining (P < 0.001). Double-label immunohistochemistry revealed that reactive stromal cells were vimentin and smooth muscle alpha-actin positive, indicating the myofibroblast phenotype. In addition, reactive stroma cells exhibited elevated collagen I synthesis and expression of tenascin and fibroblast activation protein. Increased vimentin expression and collagen I synthesis were first observed in activated periacinar fibroblasts adjacent to PIN. Similar to previous observations in prostate cancer, TGF-beta1-staining intensity was elevated in PIN. In vitro, TGF-beta1 stimulated human prostatic fibroblasts to switch to the myofibroblast phenotype and to express tenascin. CONCLUSIONS: The stromal microenvironment in human prostate cancer is altered compared with normal stroma and exhibits features of a wound repair stroma. Reactive stroma is composed of myofibroblasts and fibroblasts stimulated to express extracellular matrix components. Reactive stroma appears to be initiated during PIN and evolve with cancer progression to effectively displace the normal fibromuscular stroma. These studies and others suggest that TGF-beta1 is a candidate regulator of reactive stroma during prostate cancer progression.