Gene Panels by Specialty

A gene panel is a next generation sequencing test that reads a chosen set of genes linked to one clinical problem or group of related disorders. The panel is selected by specialty and phenotype, for example cardiomyopathy, epilepsy, inherited retinal disease, hearing loss, hereditary cancer, kidney disease, skeletal dysplasia, immune deficiency or familial hypercholesterolaemia.

Panels sit between a single gene test, which suits a very specific condition, and exome or genome sequencing, which suits a broad or unclear picture. Because only the relevant genes are analysed, coverage is deeper and more even, the data are easier to interpret, and the number of incidental findings is small.

DNA is extracted from blood (3 to 5 mL in EDTA) or saliva. Target regions are captured and sequenced, and variants are compared with a reference genome. Most laboratories also detect exon level deletions and duplications by read depth analysis, and some confirm variants by Sanger sequencing. Results take 2 to 8 weeks, and rapid panels take 3 to 10 days in acutely ill patients.

The diagnostic yield depends on how specific the phenotype is. Typical figures are about 30 to 50% in hypertrophic and dilated cardiomyopathy, 10 to 30% in epilepsy (higher in early infantile onset), 40 to 60% in hearing loss, 60 to 70% in inherited retinal dystrophy, and about 5 to 10% in unselected hereditary cancer testing. A panel result can change surveillance, treatment, reproductive choices and the testing of relatives.

Indications.

• A clinical picture that fits a genetically heterogeneous disorder with many possible genes (cardiomyopathy, channelopathy, epilepsy, neuropathy, ataxia, retinal dystrophy, hearing loss, renal cystic and tubular disease).
• Hereditary cancer syndromes suggested by early onset, multiple primaries or family history (breast, ovarian, colorectal, endometrial, pancreatic, prostate, renal).
• Familial hypercholesterolaemia, aortopathy, skeletal dysplasia, primary immunodeficiency and bleeding or clotting disorders.
• A newborn or child with a metabolic or neuromuscular presentation, where a quick answer allows treatment.
• Sudden unexplained death in a young person, with stored blood or tissue for testing, to protect relatives.
• Confirmation of a clinical diagnosis before gene directed treatment.

Not indicated when a single gene test is clearly the right one, when the phenotype is not suggestive of a genetic cause, or when the result will not change care and the patient has not been counselled.

Specimen requirements.

• EDTA blood (3 to 5 mL), or saliva or buccal swab when a blood sample is difficult. A parent's sample is useful for interpreting variants in a child.
• A request form with the phenotype in detail (age at onset, features, results of other tests, family tree), since the laboratory uses it to choose genes and interpret variants.
• Written consent after counselling, with a decision about reporting of variants of uncertain significance and of incidental findings.
• Stored tissue or blood from a deceased person, with consent from the next of kin.

Normal appearances.

A negative result means no reportable variant was found in the genes tested, with the percentage of the target covered adequately stated. It does not exclude a genetic cause outside the panel.

Interpretation.

• Variants are classified in five tiers: pathogenic, likely pathogenic, variant of uncertain significance, likely benign and benign. Pathogenic and likely pathogenic variants are reported as diagnostic, and uncertain variants are not used for clinical decisions.
• Inheritance and zygosity: a heterozygous variant in a dominant gene, a homozygous or compound heterozygous change in a recessive gene, and a hemizygous change in an X-linked gene. A single variant in a recessive gene is a carrier finding and does not explain disease unless a second variant is found.
• Segregation: testing of parents and affected relatives helps to decide whether a variant is de novo, inherited, or follows the disease in the family.
• Penetrance and expressivity: many dominant conditions show incomplete penetrance and wide variation, so the result is a risk estimate and not a prediction.
• Variants of uncertain significance are common, particularly in people of non-European ancestry whose genetic data are under represented in reference sets. They are re-reviewed when evidence grows.
• Secondary or incidental findings in genes unrelated to the question are reported only when the patient has agreed.
• Results are given with an interpretation in the clinical context and a plan for surveillance, treatment and family testing.

Critical values and findings needing immediate action.

• A pathogenic variant in a gene for a life threatening but preventable condition (for example, long QT syndrome, Lynch syndrome, BRCA1 or BRCA2, familial hypercholesterolaemia, Marfan syndrome): start surveillance and risk reducing treatment and offer testing to relatives.
• A diagnosis in an acutely ill child that points to a specific treatment (for example, a treatable metabolic or epilepsy gene): communication to the clinical team on the day of the result.
• A variant that raises risk to a pregnancy: urgent genetic counselling.
• Results that conflict with the clinical picture: review with the laboratory and repeat testing.

Limitations and artefacts.

• A panel covers only selected genes, so it can miss a gene outside the list, and genes added later require re-testing.
• Some variant types are poorly detected: repeat expansions, deep intronic and regulatory variants, balanced structural changes, low level mosaicism, and variants in genes with pseudogenes or high sequence similarity.
• Coverage is uneven in some regions, and uncovered exons need to be reported.
• Interpretation depends on the knowledge at the time. A variant of uncertain significance may be reclassified, and a pathogenic call may be downgraded.
• Variants of uncertain significance can cause anxiety and wrong clinical action, so they are not used to change management.
• Mitochondrial DNA is not covered by nuclear gene panels.
• Privacy, insurance and employment concerns, and effects on relatives, need counselling before testing.
• Costs and access vary, and results from different laboratories may be classified differently.

Special groups.

• Children: tests for adult onset conditions without a childhood intervention are not done in a child, who can choose later.
• Pregnancy: prenatal panels need a defined question, and results can lead to urgent decisions.
• Families with a known variant: single variant testing is used.
• People of non-European ancestry: a higher rate of uncertain results, and local variant data help interpretation.
• Deceased patients: DNA from stored blood or tissue, with consent, can end a diagnostic search and protect relatives.
• Regions: panel testing is available in few laboratories, and samples are often sent abroad at high cost, with slow turnaround. Founder variants (for example, haemoglobinopathies, hearing loss genes, and some cancer variants in specific populations) allow low cost targeted tests that cover most local cases. Few genetic counsellors are available, so clinicians carry much of the counselling task.

Follow up tests.

• Parental and family testing for segregation of the variant.
• Deletion and duplication analysis (MLPA or microarray) when sequencing is negative.
• Exome or genome sequencing when a panel is negative and suspicion remains.
• Functional tests (enzyme assays, biochemical tests, muscle or skin biopsy) to support uncertain variants.
• Cascade testing of at risk relatives.
• Specialist surveillance (cardiac, oncological, ophthalmological, audiological).
• Re-analysis of the data after 1 to 2 years, or when new genes are discovered.