Human malignant astrocytomas represent the most common primary tumor of the brain and are a significant cause of mortality and morbidity. The rate and extent of tumor cell migration makes neurosurgical intervention difficult, since the true size and extent of involvement of an astrocytic malignancy are uncertain. For these reasons these tumors have defied attempts at cure. We will utilize an established neural xenografting model of fresh malignant grade -3-4 human astrocytoma cells into cortical pockets in normal and athymic rats. Using a plant lectin, Phaseolus vulgaris leucoagglutinin [PHAL] and rhodamine labeled microspheres, as a graft cell marker we have shown that the xenografted human malignant astrocytoma cells migrate from the implantation pocket into the host rat brain. To improve the prognosis for patients with malignant astrocytoma we will study the differentiation, preferential pathways, spread, rate and pattern of migration of the xenografted human astrocytoma cells to study methods of improving surgical excision and therapy. We will examine the proliferation, tumorgenicity and sites of predilection for cession of migration of these cells to study concepts such as satellitosis. We will further study the relationship of these cells to the blood-brain-barrier of the adult rate hosts to study resistance of human astrocytoma cells to chemo-and immunotherapy. The long term goal of these studies is to develop predictors for the growth, development and sites of termination of malignant astrocytomas in the human brain, resulting in a scientific basis for more effective treatment by surgical methods or removal, adjunctive radiation therapy, chemotherapy and immunotherapy.