The application of glial fibrillary acidic (GFA) protein immunohistochemistry in neurooncology. A progress report.

Glial fibrillary acidic (GFA) protein, a protein originally isolated from fibrous neuroglia (Eng et aI., 1971) and subsequently localized by specific immunofluorescence to fibrillary astrocytes (Bignami et ai, 1972; Bignami and Dahl, 1974; Uyeda et aI., 1972), is currently being used as an antigenic marker for the investigation of normal, developing and pathologically altered supporting cells of the central nervous system (CNS). The protein is, presumptively, the chemical subunit of glial filaments, but the exact relationship of its synthesis to the structural evolution of these filaments remains to be elucidated. In the meantime, the development of the peroxidase-antiperoxidase (PAP) immunocytochemical technique of Sternberger et al. (1970), the demonstration of its applicability to formalin-fixed paraffin-embedded tissue (Taylor and Burns, 1974), and the production of monospecific antiserum to purified GFA protein (Eng and Rubinstein, 1978) have resulted in a new and valuable tool in human and experimental neurooncology, in particular for the more precise identification of diagnostically difficult brain tumors. The contributions of this technique, which permits the GFA protein antigen to be directly visualized at the cellular level in routinely processed tissue, have been described in several recent reports originating from this The purpose of the present review, which is largely based on our personal experience, is to summarize the principal conclusions to date, to examine the implications inherent in some of the findings and, more specifically, to address ourselves to a number of pertinent questions which the application of this technique is currently raising