In any animal model of human disease a positive control therapy that demonstrates efficacy in both the animal model and the human disease can validate the application of that animal model to the discovery of new therapeutics. Such a therapy has recently been reported by Fornai et al. using chronic lithium carbonate treatment and showing therapeutic efficacy in both the high-copy SOD1G93A mouse model of familial amyotrophic lateral sclerosis (ALS), and in human ALS patients.

Seeking to verify this positive control therapy, we tested chronic lithium dosing in a sibling-matched, gender balanced, investigator-blinded trial using the high-copy (average 23 copies) SOD1G93A mouse (n = 27–28/group). Lithium-treated mice received single daily 36.9 mg/kg i.p. injections from 50 days of age through death. This dose delivered 1 mEq/kg (6.94 mg/kg/day lithium ions). Neurological disease severity score and body weight were determined daily during the dosing period. Age at onset of definitive disease and survival duration were recorded. Summary measures from individual body weight changes and neurological score progression, age at disease onset, and age at death were compared using Kaplan-Meier and Cox proportional hazards analysis. Our study did not show lithium efficacy by any measure.

Rigorous survival study design that includes sibling matching, gender balancing, investigator blinding, and transgene copy number verification for each experimental subject minimized the likelihood of attaining a false positive therapeutic effect in this standard animal model of familial ALS. Results from this study do not support taking lithium carbonate into human clinical trials for ALS.