A Ketogenic Diet Successfully Treats Refractory Epileptic Spasms in a Child with AICA-Ribosiduria
Biochemical/Metabolic and Therapeutics
-
Primary Categories:
-
Secondary Categories:
Introduction:
5-aminoimidazole-4-carboxamide (AICA)-ribosiduria (AICAR, OMIM #608688) is an ultra-rare inborn error of purine biosynthesis caused by biallelic pathogenic variants in the gene encoding AICAR transformylase/IMP cyclohydrolase (ATIC) (MIM #601731, NM_004044.7), a bifunctional enzyme that catalyzes the last two steps of de novo purine synthesis (DNPS), converting AICA-riboside to inosine monophosphate (IMP). Defects in this process lead to the accumulation of various purine intermediates in urine and CSF, including AICA-riboside, succinyl-AICA (SAICA)-riboside, and succinyl adenosine (S-Ado). To date, nine patients with AICAR have been reported. The classic phenotype includes visual impairment due to chorioretinal atrophy, severe, pharmacoresistant epilepsy, structural brain abnormalities, developmental delay and severe intellectual disability (ID), ante-natal growth impairment, hypotonia, dysmorphic features (i.e. coarse facies and an upturned nose), and various hepatic, renal, and cardiac manifestations. Currently, there are no effective therapies to treat the underlying disorder.
Methods:
We performed untargeted metabolomic analyses in a patient with AICA-ribosiduria using profiles generated from plasma derived from venous whole blood, urine, and CSF. Individual patient samples were compared to a set of invariant anchor specimens included in each batch to achieve semiquantitative analysis of metabolites. Raw spectral intensity values were normalized to the anchor samples, log-transformed, and compared to a normal reference population to generate Z-scores. Samples were obtained pre- and post-treatment.
Results:
We report a six-year-old female with epilepsy who first presented at nine months of age with global developmental delay and abnormal movements of truncal flexion and extremity extension. Evaluation by the Neurology service revealed diffuse hypotonia but an absence of dysmorphic features. A continuous EEG revealed flexion-extension movements to be epileptic spasms and demonstrated multifocal epileptiform discharges. A brain MRI was notable for mild thinning of the corpus callosum. Initial metabolic evaluations in plasma (lactate, acylcarnitine profile, and amino acids) and urine (organic acids and purine) were normal. Untargeted metabolomic analyses revealed elevations of N6-S-Ado (Z-score >= +6.0) in CSF and plasma and reduction of hypoxanthine (Z-score<-2) in CSF, indicating a defect in purine metabolism. Trio exome sequencing identified two compound heterozygous pathogenic variants in trans in ATIC (NM_004044.7): c.1277A>G (p.K426R), inherited from the father, and c.1113C>A (p.Y371Ter), inherited from her mother. Genomic evaluation together with metabolomic findings were consistent with a diagnosis of AICAR. She received ACTH for spasm treatment and zonisamide given her epileptiform EEG. However, spasms recurred several times, requiring addition of vigabatrin to her antiseizure medication regimen. A repeat EEG demonstrated multiple subclinical seizures, so she was ultimately started on a ketogenic diet (KD). A KD led to complete seizure and spasm resolution, and she discontinued vigabatrin and zonisamide. Her overall development subsequently improved, enabling her to participate in therapies. Neuropsychological testing revealed mild ID. At six years of age, she can run, transfer objects from hand to hand, smile, and laugh. Furthermore, urine metabolomics analysis demonstrated reduction of N6-S-Ado (Z-score = -2.7) indicating a dietary impact of the KD on the purine pathway.
Conclusion:
Our study is the first to successfully treat refractory epileptic spasms in AICAR using a KD. In contrast to the classic phenotype of severe ID in AICAR, our patient's diagnosis of mild ID suggests that a KD may benefit development as well. Our work further expands the clinical spectrum of AICAR to include a thin corpus callosum on brain MRI and the absence of dysmorphic features. Our report also indicates that untargeted metabolomics analyses can help establish the diagnosis of AICAR when combined with genomic evaluation and may be used to monitor treatment efficacy. Our patient, the tenth AICAR reported, adds to the limited number of cases and should help identify and manage other patients with this condition.
5-aminoimidazole-4-carboxamide (AICA)-ribosiduria (AICAR, OMIM #608688) is an ultra-rare inborn error of purine biosynthesis caused by biallelic pathogenic variants in the gene encoding AICAR transformylase/IMP cyclohydrolase (ATIC) (MIM #601731, NM_004044.7), a bifunctional enzyme that catalyzes the last two steps of de novo purine synthesis (DNPS), converting AICA-riboside to inosine monophosphate (IMP). Defects in this process lead to the accumulation of various purine intermediates in urine and CSF, including AICA-riboside, succinyl-AICA (SAICA)-riboside, and succinyl adenosine (S-Ado). To date, nine patients with AICAR have been reported. The classic phenotype includes visual impairment due to chorioretinal atrophy, severe, pharmacoresistant epilepsy, structural brain abnormalities, developmental delay and severe intellectual disability (ID), ante-natal growth impairment, hypotonia, dysmorphic features (i.e. coarse facies and an upturned nose), and various hepatic, renal, and cardiac manifestations. Currently, there are no effective therapies to treat the underlying disorder.
Methods:
We performed untargeted metabolomic analyses in a patient with AICA-ribosiduria using profiles generated from plasma derived from venous whole blood, urine, and CSF. Individual patient samples were compared to a set of invariant anchor specimens included in each batch to achieve semiquantitative analysis of metabolites. Raw spectral intensity values were normalized to the anchor samples, log-transformed, and compared to a normal reference population to generate Z-scores. Samples were obtained pre- and post-treatment.
Results:
We report a six-year-old female with epilepsy who first presented at nine months of age with global developmental delay and abnormal movements of truncal flexion and extremity extension. Evaluation by the Neurology service revealed diffuse hypotonia but an absence of dysmorphic features. A continuous EEG revealed flexion-extension movements to be epileptic spasms and demonstrated multifocal epileptiform discharges. A brain MRI was notable for mild thinning of the corpus callosum. Initial metabolic evaluations in plasma (lactate, acylcarnitine profile, and amino acids) and urine (organic acids and purine) were normal. Untargeted metabolomic analyses revealed elevations of N6-S-Ado (Z-score >= +6.0) in CSF and plasma and reduction of hypoxanthine (Z-score<-2) in CSF, indicating a defect in purine metabolism. Trio exome sequencing identified two compound heterozygous pathogenic variants in trans in ATIC (NM_004044.7): c.1277A>G (p.K426R), inherited from the father, and c.1113C>A (p.Y371Ter), inherited from her mother. Genomic evaluation together with metabolomic findings were consistent with a diagnosis of AICAR. She received ACTH for spasm treatment and zonisamide given her epileptiform EEG. However, spasms recurred several times, requiring addition of vigabatrin to her antiseizure medication regimen. A repeat EEG demonstrated multiple subclinical seizures, so she was ultimately started on a ketogenic diet (KD). A KD led to complete seizure and spasm resolution, and she discontinued vigabatrin and zonisamide. Her overall development subsequently improved, enabling her to participate in therapies. Neuropsychological testing revealed mild ID. At six years of age, she can run, transfer objects from hand to hand, smile, and laugh. Furthermore, urine metabolomics analysis demonstrated reduction of N6-S-Ado (Z-score = -2.7) indicating a dietary impact of the KD on the purine pathway.
Conclusion:
Our study is the first to successfully treat refractory epileptic spasms in AICAR using a KD. In contrast to the classic phenotype of severe ID in AICAR, our patient's diagnosis of mild ID suggests that a KD may benefit development as well. Our work further expands the clinical spectrum of AICAR to include a thin corpus callosum on brain MRI and the absence of dysmorphic features. Our report also indicates that untargeted metabolomics analyses can help establish the diagnosis of AICAR when combined with genomic evaluation and may be used to monitor treatment efficacy. Our patient, the tenth AICAR reported, adds to the limited number of cases and should help identify and manage other patients with this condition.