Machado–Joseph disease (MJD) is an autosomal dominant spinocerebellar degeneration (SCD).1 Reports indicate that 5-hydroxytryptamine 1A (5-HT1A) agonists are efficacious for the treatment of cortical cerebellar atrophy2,3 but its effectiveness in MJD has not been fully investigated.4,5 We report the case of a patient with MJD whose ataxia, anorexia, and leg pain were remarkably alleviated after treatment with tandospirone, a 5-HT1A agonist.5 In the current report we present the 18 month follow-up findings in this patient, with special attention paid to the ‘drug holiday’ effects of this drug.

A 52-year-old man with MJD was admitted for his fourth time to Hokuyukai Neurology Hospital on 9 March 2002 because of worsening ataxic gait, leg pain, and insomnia. He first developed balance problems on walking in 1985 at the age of 35. He was diagnosed with MJD in 1996, at which time one of his brothers had already been diagnosed with MJD, which had been confirmed by gene analysis. On 26 March 2001 his ataxic gait worsened and he developed leg pain, anorexia, depression, and insomnia, for which he was admitted for the third time Hokuyukai Neurology Hospital. All of these symptoms showed significant improvement following tandospirone treatment (30 mg per day), which was not associated with any accompanying side-effects including dizziness, daytime sleepiness, and headache. The details of this patient's third hospitalization have been reported.6 The patient's total Ataxia Rating Scale (ARS) score6 upon discharge was 13 (6 for posture and gait, 4 for kinetic, 0 for dysarthria, and 3 for oculomotor movement). The patient continued to take tandospirone after being discharged, and was tested using the ARS and the Self-Rating Depression Scale (SDS)7 bimonthly. His ataxic gait, leg pain, and insomnia began to worsen in February 2002 and took a significant downturn on 9 March, necessitating a fourth hospitalization.

Neurological examination showed staring eyes, facial twitching, and slight dysarthria. The patient's eye movements were saccadic and were accompanied by a persistent gaze-evoked nystagmus. He demonstrated hyperreflexive knee jerks and displayed decomposition and hypermetria on finger–nose and heel–shin testing. He had great difficulty standing with his feet together, and his walking was wide-based and irregular. No objective sensory disturbance was detected. His total ARS was 45–19 for posture and gait, 22 for kinetic, 1 for dysarthria, and 3 for oculomotor movement, while his SDS score was 28. Brain MRI revealed mild atrophy of the cerebellum and pons. The diagnosis of MJD was confirmed by identification of 68 repeats of the CAG trinucleotide on chromosome 14.

After written informed consent was obtained, tandospirone treatment was stopped for 4 days starting on 10 March, after which his ataxia and ability to walk deteriorated quickly over the ensuing 4 days. His ARS and SDS scores were higher than they were prior to the discontinuance of tandospirone (55 for the total ARS, 22 for posture and gait, 28 for kinetic, 2 for dysarthria, and 3 for oculomotor movement, 38 for SDS), and his leg pain and insomnia worsened significantly. He did not develop any other complications during his 4-day drug withdrawal period. Tandospirone treatment was re-started on 14 March. Six days later his heel–shin tests showed improvement, and he was able to stand with his feet together and walk steadily. His ARS and SDS scores fell to levels below those seen before his drug holiday (22 for the total ARS, 6 for posture and gait, 13 for kinetic, 0 for dysarthria, and 3 for oculomotor movement, 26 for SDS). His insomnia and leg pain also improved and he was discharged on 10 April, after which he continued to take 30 mg/day of tandospirone. His ARS and SDS scores did not increase for 9 months after his last discharge.

In the present case, significant alleviation of cerebellar ataxia was observed after initiation of tandospirone therapy, as we already reported.5 Machado–Joseph disease has been reported to be resistant to 5-hydroxytryptophan2 and buspirone treatment,3 although Friedman reported success in one patient.4 Our patient responded in a manner similar to that of Dr Friedman's patient, possibly because his condition was mild. Interestingly, the CAG repeat length in the present patient was 68, which was nearly identical to that reported for Dr Friedman's patient (69). Some reports have suggested that 5-HT1A agonists directly affected cerebellar excitatory pathways by inhibiting glutamate release from cerebellar parallel fibers.8 Although its exact mechanism of action remains to be elucidated, it is conceivable that tandospirone might have similar direct effects on the cerebellum.

Approximately 9 months after his last discharge, the patient's condition deteriorated again. Possible reasons for  his  deterioration  are  threefold.  First,  side-effects of tandospirone may have played a role. This seems unlikely because his symptoms deteriorated further after cessation of drug therapy. Second, the disease itself may have progressed. This also seems unlikely because his condition dramatically improved following re-treatment with tandospirone. This improvement lasted nearly 9 months, suggesting that there was an absence of irreversible disease progression and the presence of significant numbers of functioning neuronal 5-HT1A receptors. Third, serotonin receptors may have become desensitized as a result of the chronic administration of tandospirone; this explanation seems most plausible. Moreover, it is likely that post-synaptic 5-HT1A receptors on Purkinje cells were re-sensitized by tandospirone withdrawal during the patient's short drug holiday. The long-term effectiveness of tandospirone argues against a placebo effect.

Drug holiday effects have been reported for levodopa (l-dopa)9–11 and bromocriptine,12 the latter of which is a dopamine receptor agonist, in Parkinson's patients. Wieland et al. reported the presence of residual desensitization of post-synaptic 5-HT1A receptors in response to chronic tandospirone infusion.13 However, to our knowledge there have been no previously published reports of a drug holiday effect with the use of serotonin receptor agonists for the treatment of cerebellar ataxia. A 4 day drug holiday was decided upon for the present patient based on a report of the effectiveness of this approach as regards to l-dopa therapy.11


  1. Top of page
  • 1
    Takiyama Y, Oyanagi S, Kawashima S et al. A clinical and pathological study of a large Japanese family with Machado–Joseph disease tightly linked to the DNA markers on chromosome 14Q. Neurology 1994; 44: 13021308.
  • 2
    Trouillias P. L-5-Hydroxytryptophan treatment in hereditary ataxia: clinical and etiological correlations. Mov. Disord. 1990; 5 (Suppl. 1): 6.
  • 3
    Trouillas P, Xie J, Adeleine P et al. Buspirone: a 5-hydroxytryptamine1A agonist, is active in cerebellar ataxia. Results of a double-blind drug placebo study in patients with cerebellar cortical atrophy. Arch. Neurol. 1997; 54: 749752.
  • 4
    Friedman JH. Machado-Joseph disease/spinocerebellar ataxia 3. Responsive to buspirone. Mov. Disord. 1997; 12: 613614.
  • 5
    Takei A, Honma S, Kawashima A et al. Beneficial effects of tandospirone on ataxia of a patient with Machado–Joseph disease. Psychiatry Clin. Neurosci. 2002; 56: 181185.
  • 6
    Trouillas P, Takayanagi T, Hallet M et al. International cooperative ataxia rating scale for pharmacological assessment of the cerebellar syndrome. J. Neurol. Sci. 1997; 145: 205211.
  • 7
    Zung WWK, Durham NC. A self-rating depression scale. Arch. Gen. Psychiatry 1963; 12: 6370.
  • 8
    Maura G, Reiteri M. Serotonin 5-HT1D and 5-HT1A receptors respectively mediate inhibition of glutamate release and inhibition of cyclic GMP production in rat cerebellum in vitro. J. Neurochem. 1996; 66: 203209.
  • 9
    Corona T, Rivera C, Otero E, Stopp L. A longitudinal study of the effects of an L-dopa drug holiday on the course of Parkinson's disease. Clin. Neuropharmacol. 1995; 18: 325332.
  • 10
    Feldman RG, Kaye JA, Lannon MC. Parkinson's disease: follow-up after ‘drug holiday’. J. Clin. Pharmacol. 1986; 26: 662667.
  • 11
    Martinez F, Castillo J, Castro A, Lema M, Noya M. The results of drug holiday in Parkinson's disease. Neurologia 1992; 7: 254259.
  • 12
    Teychenne PF, Bergsrud D, Elton RL, Racy A. Bromocriptine: long-term low-dose therapy in Parkinson's disease. Clin. Neuropharmacol. 1986; 9: 138145.
  • 13
    Wieland S, Fischette C, Lucki I. et al. Effect of chronic treatments with tandospirone and imprimine on serotonin-mediated behavioral responses and monoamine receptors. Neuropharmacology 1993; 32: 561573.