Protocol for the Design of Virtual Uveal Melanoma Tissue Microarrays From Different Institutions

Abstract

Abstract: : Purpose: Tissue microarrays (TMAs) have been shown to be helpful in simultaneously screening hundreds of tissue samples for the expression of tumor markers by immunohistochemistry or in situ hybridization. The authors designed a protocol for generating TMAs from primary and metastatic uveal melanoma tissue that allows for the correlation of multiple histologic attributes of tumor progression with marker expression and exchange of data between arrays developed at different institutions. Methods: Donor cores measuring 0.6 mm in diameter were placed into the recipient block in patterns unique to the institutional preference of array design. Demographical data, tumor size and location, application of pre–enucleation radiation, and outcomes were recorded for each tumor. For each primary tumor, multiple pairs of cores were obtained from different areas to sample for variations in location (ciliary body vs choroid), cell type (epithelioid vs spindle), and the presence or absence of looping vasculogenic mimicry (VM) patterns. Cores obtained from metatases were sampled on the basis of cell type and VM patterning, and allow for the comparison between matching primary and metastatic lesions. Each TMA has an open access Data Exchange Specification in a self–describing XML document with well defined common data elements. Results: It is possible to correlate multiple histological markers of tumor progression with immunohistochemical or in situ hybridization demonstration of tumor markers independent of the institutional preference for design of the recipient block. Conclusions: In the current era of non–surgical treatment of uveal melanomas and the limited availability of new tumor specimens, the development of TMAs from multiple institutions based upon a standardized protocol facilitates the investigation of new molecular markers while conserving archival tissue. Data extracted from standardized TMAs can then be analyzed as a large virtual array to enhance the statistical power of the observations while TMA blocks stay at the institutions of origin, an innovative aspect of this project. Supported by NIH Grant EY10457

Keywords: melanoma • pathology techniques • microscopy: light/fluorescence/immunohistochemistry