Genomic testing, variant interpretation, and counselling — branching viva

Opening framework

My framework has four layers. First, the test and its yield — the hierarchy from karyotype through chromosomal microarray (first-tier for unexplained developmental disability, 10 to 20 percent yield) to exome sequencing (25 to 40 percent) and genome sequencing (40 to 50 percent). Second, the variant and its classification — the ACMG/AMP five-tier framework that converts raw sequence data into a probabilistic call, distinguishing pathogenic from uncertain from benign. Third, the counselling — the pre-test conversation that covers yield, VUS possibility, secondary findings, and consent, and the post-test disclosure that contextualises the result. Fourth, the family — cascade testing of parents, siblings, and at-risk relatives, and periodic re-analysis as knowledge grows. [1] [9]

Why the normal microarray does not exclude a genetic diagnosis

Chromosomal microarray detects copy-number variants — deletions and duplications of chromosomal material — at sub-microscopic resolution, and it is the first-tier test for unexplained developmental disability, developmental delay, autism with dysmorphism, and multiple congenital anomalies. But it does not detect single-nucleotide variants (the point mutations that cause most single-gene disorders), repeat expansions such as the fragile X CGG expansion, balanced rearrangements such as reciprocal translocations, or epigenetic changes such as methylation defects. A normal microarray has excluded pathogenic CNVs, not the broader universe of genetic disease. The next test is exome or genome sequencing, ideally as a trio (proband plus both parents), because trio sequencing identifies de novo variants and filters rare inherited variants, increasing yield and reducing the VUS burden. [2] [9]

Classifying variants: the ACMG/AMP framework and VUS management

The ACMG/AMP framework classifies each variant into one of five tiers: pathogenic (greater than 99 percent probability), likely pathogenic (90 to 99 percent), variant of uncertain significance (5 to 90 percent), likely benign (1 to 5 percent), and benign (less than 1 percent). The classification is built on evidence codes: PVS for very strong pathogenic evidence (typically a predicted null variant in a haploinsufficient gene), PS for strong, PM for moderate, PP for supporting, and BA1, BS, BP for benign evidence. The combining rules specify which combinations produce each tier — for example, one very strong plus one strong yields pathogenic. A VUS is a variant for which the evidence is insufficient to classify as either pathogenic or benign. The critical management principle is that a VUS should never be the sole basis for a clinical decision. I manage a VUS by offering segregation testing in the family, functional studies where available, and scheduling periodic re-analysis every 2 to 3 years, while anchoring the child's management on the clinical picture. [1]

The counselling framework

Pre-test counselling covers the estimated diagnostic yield (I would cite the phenotype-specific yield), the possibility of a non-diagnostic result, the possibility of a VUS and what that means, the option of secondary findings under the ACMG SF v3.0 list (73 medically actionable genes, opt-in), the implications for insurance and employment with reference to the Disability Discrimination Act, and consent for data sharing and sample storage. Post-test counselling is structured around the disclosure of the result in its full clinical context: a pathogenic result prompts a discussion of the diagnosis, natural history, management, reproductive options, and cascade testing; a VUS prompts a discussion of uncertainty and the plan for re-analysis; a non-diagnostic result prompts a discussion of what was excluded and what remains possible. The genetic counsellor is central to this process. [5] [9]

Branch: the critically ill infant and rapid genome sequencing

For a critically ill infant in the NICU with a suspected but undiagnosed genetic disorder, I would request rapid trio genome sequencing with a turnaround time of typically 7 to 14 days. The Willig and Petrikin study demonstrated that rapid whole-genome sequencing in critically ill newborns can identify Mendelian disorders, change acute management, and reduce the iatrogenic harm of undirected investigation. A rapid result can guide decisions about surgery, transplantation, escalation or withdrawal of intensive care, and disease-specific therapy. The consent is obtained under time pressure, and the genetic counsellor supports the family before, during, and after the disclosure. A rapid result that excludes a genetic diagnosis is also valuable, because it redirects the investigation pathway. [7]

Closing: secondary findings and the cascade-testing obligation

The closing point is the management of secondary findings and the cascade-testing obligation. Under the ACMG SF v3.0 list, a clinically actionable variant in a gene unrelated to the indication for testing is reported if the family opted in at the pre-test visit — and a secondary finding triggers disease-specific surveillance and management. The cascade-testing obligation follows any pathogenic or likely pathogenic result: both parents are tested to confirm de novo versus inherited status, siblings are tested, and at-risk relatives on the relevant side of the family are offered testing through the genetics service. The plan is shared: the clinical geneticist owns the variant interpretation and re-analysis, the genetic counsellor owns the family testing and reproductive counselling, and the general paediatrician owns the coordination. And every non-diagnostic result is re-analysed every 2 to 3 years, because the answer may emerge as knowledge grows. [5] [9]