Abstract
Our concepts and understanding of the etiology, evolution, and propagation of Graves' ophthalmopathy have become much more sophisticated that they were 10 years ago. Given our current state of knowledge, the following scheme for the pathogenesis of Graves' ophthalmopathy can be proposed. Circulating T cells in patients with Graves' disease, directed against an antigen on thyroid follicular cells, recognize antigenic epitopes that are shared by tissues contained in the retroorbital space. Of the cell types residing in these tissues, fibroblasts are most likely to act as both target and effector cells of the retroorbital immune process. This includes those fibroblasts present in the perimysium of extraocular muscles, which do not appear to be immunologically different from fibroblasts located in the retroorbital connective tissue. By contrast, convincing evidence implicating the human extraocular myocyte itself (rather than the tissue conglomerate of extraocular muscle) as a primary target in GO remains to be demonstrated. Together with adipocytes, fibroblasts may also serve as target and effector cells in pretibial myxedema. How autoreactive T cells escape deletion by the immune system and come to be directed against a self-antigen presented by cells residing in the thyroid gland and extrathyroidal locations is unknown. T cells are recruited to and infiltrate the orbit via certain adhesion receptors, which may also play a costimulatory role in T cell activation and facilitate antigen recognition. Analysis of variable region gene usage of the T cell antigen receptors in retroorbital T cells of patients with active GO reveals limited variability, suggesting that antigen-driven selection and/or expansion of specific T cells may occur early in the evolution of GO.(ABSTRACT TRUNCATED AT 250 WORDS)