Web-based virtual microscopy has enabled new applications within pathology. Here, we introduce and evaluate a network of academic servers, designed to maximize image accessibility to users from all regions of Europe. Whole-slide imaging was utilized to digitize the entire slide set (n = 154) for the slide seminars of the 21st European Congress of Pathology. The virtual slides were mirrored to five academic servers across Europe using a novel propagation method. Functionality was implemented that automatically selects the fastest server connection in order to optimize the slide-viewing speed ( http://www.webmicroscope.net/ECP2007). Results show that during 6 months of monitoring the uptime of the network was 100%. The average viewing speed with the network was 3.1 Mbit/s, as compared to 1.9 Mbit/s using single servers. A good viewing speed (>2Mbit/s) was observed in 32 of 37 countries (86%), compared to 25 of 37 (68%) using single servers. Our study shows that implementing a virtual microscopy network spanning a large geographical area is technically feasible. By utilizing existing academic networks and cost-minimizing image compression, it is also economically feasible.

Thirty-eight transitional-cell carcinomas (TCC) were analyzed by flow cytometry (FCM) using propidium iodide for DNA analysis and antibodies to cytokeratin by indirect immunofluorescence. By means of two-dimensional FCM analysis, cytokeratin-positive tumor cells could be analysed separately from cytokeratin-negative stromal and inflammatory cells. This resulted in an 18% increase in sensitivity of FCM detection of aneuploidy (10/38 samples with one-parameter DNA analysis versus 15/38 samples with two-parameter DNA and cytokeratin analysis). In addition, S-phase could be determined in the 15 aneuploid samples by means of two-parameter analysis where this was not possible using only DNA content because of the overlap of diploid and aneuploid populations. FCM analysis allowed quantification of the percentage of tumor cells expressing cytokeratin 18 which has previously been shown to correlate quantitatively with higher grade, higher stage TCC. The quantitative measurement of tumor-cell expression of cytokeratin 18 by FCM analysis appears to provide additional information of potential prognostic value, independent of tumor-cell ploidy and proliferative fractions.

Endometrial cells in suspension were stained with propidium iodide and a monoclonal antibody against a cytokeratin intermediate filament protein specific for glandular and columnar cells (RGE 53). In this way columnar epithelial cells of the normal endometrium and of adenocarcinomas can be distinguished and separated by flow cytometry from non‐epithelial cells (fibroblasts and inflammatory cells) and squamous epithelial cells, all of which are negative for RGE 53. This makes it possible to analyse and also sort pure fractions of this particular tissue type for further studies. The use of propidium iodide allows simultaneous DNA flow cytometry of these columnar epithelial cells. Therefore, the use of antibodies to cytokeratin in combination with propidium iodide can be of help in analyzing and sorting pure fractions of both normal and malignant cells. This allows a more refined examination of complex cell mixtures using flow cytometry.

No abstract

A method for the extraction of nuclei from selected regions in paraffin‐embedded tissue is described. Fifty‐micrometer sections are cut, dewaxed, and rehydrated. For the final handling, the sections are manually transferred from one tray to another. The sections are put on a slide under a dissection microscope and the region of interest is isolated by scraping off the irrelevant region with a scalpel. An optimal number of single nuclei is obtained by incubation in a protease solution with intermediate syringing. The nuclei are washed and can be used for flow cytometry. Resuspension of the nuclei in foetal calf serum and cytocentrifugation results in preparations suited for image analysis. DNA cytometric measurements of nuclei in a carcinoma in situ and an invasive carcinoma region in breast tissue present in the same tissue block and in a severe dysplasia/carcinoma in situ (CIN III) region of the cervix are presented.