Human retinoblastoma, when manipulated to differentiate in vitro, may provide the basis for a useful model for the elucidation of cell surface molecules and their alteration during neoplasia and embryogenesis. The overall objective of this proposal is the detection and characterization of retinoblastoma differentiation antigens for clinical application. The approach will involve production of monoclonal antibodies for the investigation of oncofetal antigens which may play important roles in the surface modulation of cell-cell interactions during in vitro retinoblastoma differentiation. We intend to: 1. Extend our in vitro model of retinoblastoma differentiation, using chemical induction of differentiation in tissue culture; 2. Develop a panel of monoclonal antibodies against retinoblastoma cell surface antigens at varying stages of in vitro differentiation; 3. Perform characterization of the monoclonal antibody lines and their corresponding antigens detectable with the monoclonal antibody probes. The immunologic study of differentiation antigens detectable with monoclonal antibody probes during in vitro chemical induction of retinal differentiation in cultured retinoblastoma cells has important clinical implications for (1) immunohistochemical approaches to diagnosis and predictions of prognosis in retinoblastoma, as well as implications for (2) the understanding of retinal tumorigenesis and morphogenesis, (3) immunotherapeutic approaches to retinoblastoma, (4) chemical modulation of retinoblastoma tumor growth in vivo, and (5) elucidation of retinal developmental anomalies such as retinal dysplasia.