Melarsoprol has remained the chosen drug for the late-stage treatment of human African trypanosomiasis (HAT) due both to Trypanosoma brucei (T.b.) gambiense and T.b. rhodesiense; however, arsenical encephalopathies, which are often fatal, occur in 5-10% of the treated cases. To date, two major problems have not been solved. The first one is the precise diagnosis of early involvement of the central nervous system (CNS) which determines the therapeutics to be administered. The second one is linked to the lack of data on in vivo efficacy of products which are effective in vitro against trypanosomes. Answers have to be provided by experimental animal models of HAT. Such models would allow for better studies of the pathology and pathogenesis of the disease, as well as therapeutic trials of potentially effective new drugs or combinations. We have developed acute and chronic murine and sheep experimental animal models of HAT infected by T. b. brucei. Meningoencephalitis and neurological signs are relatively difficult to obtain in murine models and require artificial means, such as suramin treatment on day 21 after-infection. The chronic murine model has demonstrated CNS involvement with meningitis, followed by meningoencephalitis with progressive astrocytosis. The sheep model develops a disease with CNS complications and cerebrospinal fluid can be collected. In the sheep model, we have described anti-galactocerebrosides antibodies, which represent major components of myelin, which may indicate an autoimmune process in the CNS. We then described these antibodies in the cerebrospinal fluids and sera from patients at a late-stage of the disease. From a therapeutic point of view, we have cured mice or sheep with low doses of melarsoprol, or with the nitroimidazole derivatives Ro 15-0216 and megazol, alone or combined with suramin. Further studies of these nitroimidazole compounds, which could be proposed for human use, have to be carried out on a-primate model infected by T.b. gambiense. To our knowledge, this primate model is not available. This is why we have recently developed a T. b. gambiense primate model of HAT on Cercopithecus aethiops.