Neurogenetic conditions and precision diagnosis: Viva

Q1. The microarray is normal but you still think this is genetic. Why?

Because a chromosomal microarray detects copy-number variants, such as deletions and duplications, and not the single-nucleotide variants and small insertions or deletions that an exome reads. A normal microarray therefore rules out a pathogenic copy-number change but does not rule out a single-gene disorder, which is the more likely mechanism in a child with regression, drug-resistant epilepsy, and a movement disorder after normal development. The microarray is the right first-tier test for a copy-number burden, and a normal result simply moves the investigation to the next tier. [9][3]

Probe: What is the yield of the microarray? Around fifteen to twenty per cent for unexplained developmental disability, which is why it is first-tier for copy-number variants but not the whole answer for a single-gene picture like this. [3]

Q2. What test do you order next, and why does a trio matter?

A trio exome, sequencing the proband and both parents together. The regression after normal development in a child of unaffected parents points to a de novo dominant variant, and trio design is what allows the laboratory to distinguish a de novo change from an inherited one at the variant level. The pooled diagnostic yield of exome in neurodevelopmental disorders is around thirty-six per cent, and it is higher with a trio because the parental samples resolve segregation and support a confident pathogenic call. [3]

Probe: What if only the child can be sampled? A proband-only exome still works but the yield is lower and the interpretation is harder, because a variant of unknown inheritance is harder to classify. Wherever possible I would obtain parental samples, even after the event, to convert a proband-only result into an effective trio. [1]

Q3. How do you interpret a variant of uncertain significance, and what do you tell the family?

Every variant is classified by the American College of Medical Genetics five-tier framework into pathogenic, likely pathogenic, uncertain significance, likely benign, or benign, using population frequency, computational prediction, segregation, and functional evidence. A variant of uncertain significance is an honest statement that the evidence is insufficient to call, and I manage the child by their phenotype, segregate the variant through the family where possible, and reanalyse it as evidence accrues. I never report it to the family as the cause of the disease, because over-calling it produces a wrong label, wrong recurrence counselling, and wrong therapy. [1]

Probe: Can a variant of uncertain significance ever become pathogenic? Yes, and that is the point of reanalysis. As new gene-disease associations and new cases are published, an uninterpretable variant may be reclassified within one to two years, so the family keeps a scheduled review rather than a closed file. [3]

Q4. Which treatable conditions could a diagnosis unlock here?

Glucose transporter one deficiency is the leading candidate in this child, presenting with epilepsy and a paroxysmal movement disorder and responding to a ketogenic diet that supplies the brain with ketone bodies as an alternative fuel, and I would support the diagnosis with a low cerebrospinal fluid glucose. Beyond it, the treatable neurogenetic conditions include creatine transporter deficiency responding to creatine, dopa-responsive dystonia responding to levodopa, pyridoxine-dependent epilepsy responding to pyridoxine, and biotinidase deficiency responding to biotin, and the mutation-specific therapies for spinal muscular atrophy and Duchenne muscular dystrophy all require a confirmed molecular diagnosis first. [11][9]

Probe: Why does speed matter for these? Because the treatable conditions do not give back the developmental time lost to delay, and a child whose glucose transporter one deficiency is diagnosed and treated at three has a better developmental trajectory than one diagnosed at six. [11]

Q5. What do you do if the exome is negative?

I do not discharge the child as undiagnosed. I move to genome sequencing, which captures the non-coding and structural change and the repeat expansions that the exome misses, and I arrange periodic reanalysis of the existing data, because new gene-disease associations are published continuously and an uninterpretable variant may be reclassified within one to two years. I give the family a clear plan with scheduled review and keep the door to a diagnosis open, and in parallel I would reconsider the acquired mimics and repeat the phenotyping if new features emerge. [9][3][7]

Probe: When would rapid whole-genome sequencing be the right first move instead? In the acutely ill infant with an unexplained encephalopathy or refractory seizures, where a diagnosis in days rather than weeks changes acute management, from treatable-condition therapy to the redirection of care, and reduces the cost and length of stay. [7]