Use of an animal model of disease for toxicology enables identification of a juvenile no observed adverse effect level for cyclocreatine in creatine transporter deficiency

In standard general toxicology studies in two species to support clinical development, cyclocreatine, a creatine analog for the treatment of creatine transporter deficiency, caused deaths, convulsions, and/or multi-organ pathology. The potential translatability of these findings to patients was evaluated by comparing toxicity of cyclocreatine in wild-type mice to creatine transporter-deficient mice, a model of the human disease. A biodistribution study indicated greater accumulation of cyclocreatine in the brains of wild-type mice, consistent with its ability to be transported by the creatine transporter. Subsequent toxicology studies confirmed greater sensitivity of wild-type mice to cyclocreatine-induced toxicity. Exposure at the no observed adverse effect level in creatine transporter-deficient (554 μg*hr/ml) mice exceeded exposure at the maximum tolerated dose in wild-type (248 μg*hr/ml) mice. When dosed at 300 mg/kg/day for 3 months, cyclocreatine-related mortality, convulsions, and multi-organ pathology were observed in wild-type mice whereas there were no adverse findings in creatine transporter-deficient mice. Brain vacuolation was common to both strains. Although transporter-deficient mice appeared to be more sensitive, the finding had no functional correlates in this strain. The results highlight the importance of considering models of disease for toxicology in cases where they may be relevant to assessing safety in the intended patient population. •Toxicology studies of cyclocreatine in wild-type animals revealed severe toxicity.•Wild-type mice are more sensitive to toxicity than creatine transporter-deficient.•Mouse model enabled identification of a juvenile No Observed Adverse Effect Level.•Consider relevance of disease models for toxicology, especially in rare diseases.