Glioblastoma multiforme (GBM) is the most common primary human brain tumor. About 7000 new cases are diagnosed yearly in the USA. Despite current neurosurgical and postoperative radiotherapeutic tumor cytoreduction methods, in most cases occult foci of tumor cells infiltrate surrounding edematous brain tissues and cause recurrent disease within one year. GBM is almost invariably fatal within a few years after it is diagnosed. Our goal is to achieve long-term control of GBM by combining immunoprophylaxis with a radiation-based technique, such as boron neutron-capture therapy (BNCT), potentially capable of specifically targeting the infiltrating tumor cells while sparing the surrounding normal brain tissue. It has long been known that the subcutaneous (sc) injection of irradiated cells or untreated cultured cells (and the removal of the resulting tumors) derived from the well characterized, highly immunogenic 9L gliosarcoma (9LGS) rat model into young isogenic rats can prevent tumor growth after subsequent sc or intracranial (ic) injection of untreated, otherwise lethal 9LGS cells. In this study we have confirmed, quantified and extended those findings to study the efficacy of such immunological memory in normal aging rats and in aging rats previously treated for ic 9LGS tumors by BNCT. (1) The sc injection of 5,000,000 untreated 9LGS cells and the surgical removal of the resulting tumors (method A) protected 80% of normal young rats from an ic challenge with 10,000 untreated 9LGS cells, and a single sc injection of 5,000,000 lethally X-irradiated 9LGS cells (method B) protected 66% of them, but multiple sc injections with a crude particulate fraction prepared from 9LGS cells were not protective. Protection is long-lasting since contralateral ic rechallenge of six-month survivors with an injection of 10,000 viable 9LGS cells resulted in 100% survival. (2) Normal one-year-old rats were only slightly less protected than were normal young rats, approximately 70% rather than approximately 80% (method A) and approximately 60% rather than approximately 66% (method B). (3) BNCT treatment alone resulted in partial immunological protection, as 30% of one-year post-BNCT survivors of ic 9LGS tumors prevailed after contralateral ic rechallenge with 10,000 viable 9LGS cells. Moreover a single sc immunization with 5,000,000 untreated 9LGS cells prior to ic rechallenge boosted survival from 30% to 100%. The relevance of these observations to strategies of preclinical experimentation for immunoprophylaxis of malignant gliomas is discussed.