Tissue microarrays: an overview.

Traditionally, screening for new markers involves using a slide from each of several different patients. A more efficient way is to have one slide that contains several minute specimens, one from each patient. These slides are prepared by transferring paraffin tissue cores from many "donor" blocks to one "recipient" block. Each slide cut from this recipient block is called a tissue microarray (TMA) slide. It can have various histological types of disease that need to be compared or can have the same histological type but different behavior (e.g., responders versus non-responders, etc.). TMAs are ideal for efficient screening of prospective biomarkers by a variety of different mechanisms including immunohistochemistry, fluorescence in situ hybridization of nucleic acids (FISH) and RNA in situ hybridization. Selection and number of cases from patient subsets in a given microarray slide is amenable to statistical modeling to enhance analysis of results. In addition, different microarrays can be constructed to answer different scientific questions. The microarrays can also be produced from retrospective paraffin blocks of well-characterized cases, with clinical follow-up. The TMA slides can be "whole-slide" imaged. This provides a mechanism to share results of experiments with other investigators. There are also ongoing efforts to generate software tools for automated analysis of TMA localization data. There has also been a significant body of work done to standardize data capture, thus facilitating subsequent exchange of information. The preferred current mechanism is to use an "XLM"-based data capture and transfer. There have also been efforts to create "frozen" TMAs. This has been attempted using "donor" frozen tissues embedded in OCT compound. These samples are then arrayed into a recipient OCT block. The presence of OCT can sporadically interfere with certain assays. However, it does provide a novel mechanism for high-throughput evaluation of frozen tissue, with corresponding visualization of tissue morphology.