There are currently no animal models of primary uveal melanoma, a potentially blinding as well as lethal condition. Presently, the most popular experimental model for studying uveal melanoma requires the transplantation of Greene hamster cutaneous melanoma or cultured human melanoma cells into the rabbit eye, but these transplantation-based models cannot be used to study the development of melanoma from the normal choroid. Likewise, although the development of human uveal melanoma cell lines represents a major advance, these in vitro studies cannot duplicate the clinical and pathologic appearances of various stages of tumor development. In addition, experimental models of conjunctival melanoma have yet to be described. During the past 2 years, we successfully induced primary melanocytic lesions in the rabbit choroid and conjunctiva by means of the chronic, local application of the carcinogen, 7,12-dimethylbenz[a]anthracene (DMBA). We now propose to promote these DMBA-induced choroidal and conjunctival melanoma precursor lesions to melanoma with additional applications of DMBA and/or the promoting agent, phorbol 12-myristate 13-acetate (PMA), to photographically document the clinical evolution of DMBA-induced choroidal and conjunctival melanocytic lesions, to determine if these experimentally- induced lesions are locally invasive or give rise to metastases, and to study these experimentally-induced lesions by light and electron microscopy to correlate the pathology with clinical appearance and clinical behavior. Finally, we propose to compare the clinical and pathological features of DMBA-induced choroidal and conjunctival pigmented lesions with human choroidal nevi and melanomas and conjunctival nevi, primary acquired melanosis and melanoma. The experimental model of primary uveal melanoma permits the pathologic documentation of clinically observable stages of tumor development, impossible in patients because of the difficulty in obtaining representative choroidal biopsies without ocular morbidity. The experimental models of both primary uveal and conjunctival melanomas would facilitate the evaluation of existing diagnostic and therapeutic strategies and the development of new techniques that can be tested in a model before coming to clinical trial.
