The purpose of this study is to compare the therapeutic combination of I.V. eflornithine + oral nifurtimox to the standard IV eflornithine regimen in terms of therapeutic efficacy and clinical safety, in patients suffering from Trypanosoma brucei gambiense (Tbg) human African trypanosomiasis (HAT) in the meningoencephalitic phase.

Melarsoprol is the most commonly used product for the treatment of patients suffering from human African trypanosomiasis (HAT) in the meningoencephalitic (second, late) phase. This treatment is frequently complicated by fatal reactive encephalopathy, and at the same time resistance is beginning to appear in various countries. Eflornithine is effective and better tolerated, but it is more difficult to use. Nifurtimox, registered in several South American countries for treatment of Chagas' disease but used off label since the 1970's in series of cases of meningo-encephalitic HAT, is at present the only other potential alternative for the treatment of late-stage HAT.

The very limited number of compounds available, the lack of prospects for the development of new products and the emergence of resistance are arguments for the use of therapeutic combinations. Ideally, drug combinations should allow for reductions in the dosages of the drugs used in a way that, in particular in the case of toxic drugs such as those used for second stage HAT, the toxicity of the combination does not exceed that of either monotherapy. Of the three drug combinations nowadays possible: melarsoprol-nifurtimox, melarsoprol-eflornithine and eflornithine-nifurtimox, the last one has (in two different dosing regimens) shown the least treatment-associated toxicity and mortality in the 69 patients treated in one previous and this clinical trial to date. Good tolerability was also observed in a case series of 31 patients. The efficacy data to date suggest that efficacy is comparable to that of eflornithine and that of melarsoprol (in areas without high melarsoprol failure rates).

Inclusion Criteria:

• confirmed second-stage T.b. gambiense infection : presence of the parasite in blood, lymph node fluid or CSF and >20 white blood cells/µL in CSF. In presence of blood in the CSF, lumbar puncture must be done again or the patient cannot be included in the study (see 10.9.1).
• and of age 15 years or older.
• and resident in __________ (each site will define boundaries or maximum distance)
• and written informed consent of the patient or of a legally acceptable representative if the patient is a minor (<18 years for both genders in Uganda and Angola, <18 years for males and <16 years for females in the Democratic Republic of Congo) or unable to communicate.

Exclusion Criteria:

• pregnant woman (systematic testing of women of childbearing potential)
• treated for late-stage HAT during the last 36 months. Patients previously treated for first-stage (pentamidine) can be included.
• unlikely to have access to the treatment centre or be accessible at their place of residence for 18 months after treatment
• unable to take oral medication
• suffering from conditions other than second stage HAT that seriously limit the chances of survival over 18 months time
• Severe anemia (Hb< 5g/dl)
• Severe underlying diseases upon admission (e.g. Active tuberculosis and/or being treated for TB; Bacterial or cryptococcal meningitis; Stages 3 or 4 HIV/AIDS according to the WHO clinical definition) (WHO, 1986).
• Severe renal failure based on clinical examination combined with biochemistry if available: creatinine clearance <20mL/min
• Severe hepatic failure based on clinical examination combined with biochemistry if available: total bilirubin >50 µmol/L, ALAT/GPT >70 UI/L, unless these laboratory values are determined by the investigator as likely due to conditions other than hepatic failure.