The clinical efficacy, dose-response relationship, and safety of TRH-T (thyrotropin releasing hormone tartrate) were assessed in 290 patients with spinocerebellar degeneration (SCD) in a 2-week, double-blind study using placebo as control. 254 patients satisfied the criteria for inclusion in evaluation of the drug efficacy. The patients were treated with TRH-T in an intramuscular dose of 2 mg, 0.5 mg or 0 mg (placebo) as TRH once a day for 2 weeks. Clinical responses to these treatments were evaluated 3 times: at the end of weeks 1 and 2 of treatment and a week after the end of treatment. The results of "global improvement rating" as well as those of "ataxia improvement rating" showed that both 2 mg and 0.5 mg TRH-T treatments were significantly superior to placebo treatment in patients with predominantly cerebellar form of SCD. The effect was well maintained a week after the end of the 2-week treatment in the patients who were given TRH-T in daily dose of 2 mg and showed improvement at the end of treatment. The results of "improvement rating of each symptom" revealed that 2 mg treatment was significantly more effective than placebo for disorders of standing, gait, speech and writing. In the patients who had no pyramidal involvement or disorder of deep sensation, the drug efficacy and dose-response relationship were evident. Adverse reactions to the drug such as headache, feeling febrile and nausea were observed in 50% of the patients on 2 mg treatment, in 38% of those on 0.5 mg treatment and in 21% of those on placebo patient, however, discontinued treatment because of adverse reactions.