A Negative Trial of Creatine in ALS

Abstract & Commentary

Source: Jan Groeneveld G, et al. A randomized sequential trial of creatine in amyotrophic lateral sclerosis. Ann Neurol. 2003;53:437-445.

This is the first reported clinical trial of the effects of creatine in sporadic ALS patients. The trial was carried out in The Netherlands. The rationale for the trial was based on work done by my laboratory. We demonstrated that creatine had significant effects in improving survival and preventing loss of motor neurons in a transgenic mouse model of ALS, in which human mutated Cu,Zn-superoxide dismutase (mSOD1) is overexpressed. In that animal study, administration of creatine starting at 50 days of age prolonged survival by 26 days, which was markedly better than the effects of riluzole. The present clinical trial was a double-blind, placebo-controlled, sequential clinical trial. The primary end points were death, persistent assisted ventilation, or tracheotomy. Secondary outcome measurements were the rate of decline in isometric arm muscle strength, forced vital capacity, functional status, and quality of life. The trial was terminated when 34 patients had died and the sequential monitoring of the statistics suggested that there was no difference in the null hypothesis between the 2 treatment arms. Jan Groeneveld and associates did observe a slightly slower decline in arm strength and in vital capacity at 1 month. Otherwise, no significant benefits occurred. There were no significant adverse effects. The dose used was 10 g daily. This is consistent with the doses that were effective when used in mice. These doses were 1 or 2% creatine in the diet, which correspond to 6-12 g daily in humans.


These results are extremely disappointing. The results with creatine in the transgenic mouse model have been replicated by numerous groups. This suggests that studying this transgenic mouse model, which represents mutations found in a small percentage of familial ALS patients, may not be predictive of clinical efficacy in sporadic ALS patients. Whether this conclusion is generally valid remains to be proven. At present, trials of minocycline, as well as Celebrexâ, in sporadic ALS are ongoing. Both of these agents have been shown to have significant therapeutic benefits in the transgenic mouse model of ALS. There are also 2 further trials of creatine under way in the United States, and it will be interesting to see the outcome of these efforts. The present results, however, suggest that caution is needed in extrapolating results in certain transgenic mouse models caused by mutations that occur in a very small proportion of ALS patients. One expects that this will not be the case with other transgenic mouse models, such as Huntington’s disease, where a common genetic defect occurs in all patients. — M. Flint Beal

Dr. Beal is Professor and Chairman; Department of Neurology; Cornell University Medical College New York, NY