Cerebral malaria is the most severe complication of human infection with Plasmodium falciparum. It was shown that Plasmodium berghei ANKA-induced cerebral malaria was prevented in 100% of mice depleted of CD8+ T cells 1 day prior to the development of neurological signs. However, the importance of parasites in the brains of these mice was never clearly investigated. Moreover, the relevance of this model to human cerebral malaria has been questioned many times, especially concerning the relative importance of leukocytes versus parasitized erythrocytes sequestered in the brain. Here, we show that mice protected from cerebral malaria by CD8+ T-cell depletion have significantly fewer parasites in the brain. Treatment of infected mice with an antimalarial drug 15 to 20 h prior to the estimated time of death also protected mice from cerebral malaria without altering the number of CD8+ T cells in the brain. These mice subsequently developed cerebral malaria with parasitized red blood cells in the brain. Our results clearly demonstrated that sequestration of CD8+ T cells in the brain is not sufficient for the development of cerebral malaria in C57BL/6 mice but that the concomitant presence of parasitized red blood cells is crucial for the onset of pathology. Importantly, these results also demonstrated that the experimental cerebral malaria model shares many features with human pathology and might be a relevant model to study its pathogenesis.