Pathogenicity of AIFM1 Variants and Beneficial Effect of Riboflavin Need to Be Appropriately Confirmed

We read with interest the article by Heimer et al. about two pediatric patients with X-linked cerebellar ataxia due to two different mutations in the AIFM1 gene. We have the following comments and concerns.

We do not agree that the mutation c.422C>T is truly pathogenic. Pathognicity was claimed after in-silico prediction and after protein modelling. No other family member carried the mutation, thus there was no segregation, and no biochemical or functional investigations were carried out. Location of the variant in a region where pathogenic mutation cluster not necessarily implies that the variant c.422T>C is also pathogenic.

Patient 2 was initially diagnosed with epilepsia partialis continua and three antiepileptic drugs (AEDs), benzodiazepines, valproic acid, and levetirazetam, were tried, but without effect. Why were only 3 AEDs given and no further AEDs tried? Were they given in monotherapy or in combination? Why was valproic acid chosen from which it is well known that it is mitochondrion-toxic and can even cause fatalities in some mitochondrial disorders (MIDs). Did the authors try pirazetam or was the patient ever put on a ketogenic diet?