All medical testing has associated benefits, risks, and limitations. Some unique issues arise when assessing the genome for variants associated with health conditions. Each type of genetic test (e.g., condition-specific tests, broad multigene tests, exome testing, and whole genome sequencing) has its own specific benefits, risks, and limitations.

Genetic tests vary widely in the scope of analysis. The number of genes included in a genetic test ranges from a single gene to the whole genome. Tests with a larger scope may or may not be the best choice for the clinical situation. It is important to consider the sensitivity of the test for the suspected condition, the likelihood of getting back uncertain or unexpected results, and the cost in the context of the patient and family.

Different types of genetic variants cause disease. Genetic conditions are caused by different types of variants, such as single nucleotide changes, extra or missing bases, methylation errors, and trinucleotide repeats, among others. Tests vary in their ability to detect these variants: not all tests detect all types of variants, and there may be extensive variability among tests.

Genes vary in terms of the strength of association with disease. While research continues to identify new genomic associations with different neurologic diseases, the clinical impact of all associated genes is not necessarily equal. For genes with limited clinical data, it can be difficult to predict the impact of finding a variant. In addition, for these genes, there is a greater likelihood of identifying a variant of uncertain clinical significance. Labs and specific tests differ in the genes included on panels.

Uncertain and unexpected information is more likely to be identified on larger genomic tests. The lab may identify variants that are of unknown clinical significance. These are called variants of uncertain clinical significance (VUS). Additionally, patients may learn they have an unexpected genetic condition causing their symptoms. Results may also identify a secondary finding, which is a medically significant result for the patient and/or family members but unrelated to the reason for testing. Genomic testing may identify undisclosed family relationships, such as nonpaternity.

The exome is the portion of the genome containing the protein-coding regions of the genes known as exons. While representing only about 1% of the genome, exome testing has a high diagnostic yield because most variants associated with hereditary conditions are located within the exons or in the regions immediately flanking them. Exome testing is also referred to as exome sequencing (ES).

Whole genome sequencing (WGS), on the other hand, examines all coding and non-coding DNA, or about 3 billion base pairs of DNA, and results in a much larger amount of data to be interpreted. It includes testing of the non-coding and regulatory regions of the genome, where additional variants associated with disease could possibly be located.

Exome testing is generally less expensive and faster than WGS. Whole genome sequencing may have a higher diagnostic yield for some patients than exome testing, but these data are still emerging. Currently, due to the large amount of data that comes from WGS and downstream interpretation challenges, exome testing is often (but not always) the preferred diagnostic test when the goal is to assess the breadth of the genome.

More information is available in the Comparing Genetic Tests and Variant Classification and Reanalysis in Exome Testing resources.

Part of the utility of genomic testing is to obtain information that can inform management. For situations in which the clinical diagnosis is uncertain, genomic testing may be able to provide a definitive diagnosis. Having a diagnosis allows clinicians to identify treatment options and management approaches that are matched to the underlying biological process causing disease. Even when the clinical diagnosis is certain, knowing the specific variant causing the condition in the patient can help determine whether a treatment targeted to address a specific variant is applicable. There are only a few targeted treatments currently available, either clinically or through clinical trials, for pediatric neurologic conditions, but it is an area of very active research and is rapidly developing.

Genomic information can be used to target treatment and management strategies in many ways.

• Avoid unnecessary, ineffective, or harmful treatments. For example, identifying a known pathogenic variant in SCN1A in a child with medically refractory generalized epilepsy allows avoidance of phenytoin.
• Initiate preventive measures. Some neurogenetic conditions are associated with an increased risk of cancer (e.g., neurofibromatosis) or disease triggers (e.g., channelopathies such as paramyotonia congenita), which would be appropriate for screening and pre-emptive knowledge.
• Provide more precise symptomatic treatment. For example, supplementation (e.g., pyridoxine supplementation in pyridoxine-dependent epilepsy).
• Use treatments targeting the identified variant at a cellular or molecular level. For example, exon skipping for Duchenne muscular dystrophy.
• Avoid adverse drug reactions. There are some neurogenetic conditions that are known to be caused by variants in pathways that are involved with drug metabolism (e.g., Alpers-Huttenlocher syndrome is caused by variants in POLG variants which helps metabolize valproic acid; use of this drug can precipitate liver failure).

The use of genomic information to develop treatments that directly target the underlying variant has proved challenging in part because of the multiple sources of variation, which include the following:

• Clinical variability among patients who carry variants within the same gene (allelic heterogeneity). This variability has many causes, including the location of the variant within the gene and the type of variant (e.g., single nucleotide change vs. deletion), both of which affect the impact of the variant on gene function. For example, different variants in KCNA2, cause infantile-onset epilepsy similar to Dravet syndrome and more severe phenotypes including epilepsy, ataxia, and intellectual disability.
• Some genomic variants can be present in all cells of the body or found only in a subset of cells or tissues (somatic mosaicism). For example, in a subset of individuals with neurofibromatosis type 1 the underlying cause is somatic mosaicism, which results in symptoms being localized to one or more segments of the body.
• Clinical presentation can be affected by the presence of other genes and/or genomic variants (epistatis).

The availability of emerging technologies that allow for different approaches to targeting underlying variants, such as CRISPR-Cas9, holds promise for future efforts.

To aid in variant interpretation, the laboratory may use "trio testing." This involves limited analysis of samples from both biological parents of the patient (proband). Data from the parents' samples serve as controls in variant analysis for the patient. Trio analysis is particularly useful when classifying de novo pathogenic variants (a variant that is discovered for the first time in a proband). If an unaffected parent and the proband carry the same rare variant, it is less likely that the variant is the underlying cause of the proband's condition. Parental information, therefore, helps reduce the number of variants the lab must consider in their data analysis. Although trio testing is often more informative, patients without available biological parents still benefit from exome testing.

Setting parental expectations about trio testing

Parents may have the misconception that they are also receiving full exome analysis. Generally, this is not the case; the tests are run primarily for the proband's benefit. Unless consented separately, parental exome information is typically not used for the parent's own health information. In certain circumstances, genetic information important to the parent's health care may be reported. For instance, upon closer study, a parent who carries the same variant as their child may be discovered to be mildly affected.

Additionally, parents may have the option of learning if they carry pathogenic variants for unrelated but medically actionable conditions (secondary findings). The ordering provider should check with the lab about what kind of parental results may be disclosed if any; such policies often vary between labs.

Genomic tests can identify different types of variants within a gene, including those that are benign, pathogenic, or those with unknown or unclear effects.
The clinical significance of a variant may be uncertain for several different reasons:

• A variant may be rare or novel, with insufficient data about its effects on the function of the gene and its relationship to disease.
• Data about the effect of the variant on the gene may be conflicting.

While it is tempting to believe that variants identified in genomic testing contribute to disease risk, a VUS is not a clinically useful result. Changes in management based on the finding of a VUS are not recommended. If you have questions about a VUS result, you can call the laboratory to ask for any information they might have about it and why it is classified as a VUS. In some cases, a VUS may be reclassified as actionable or benign as new data is obtained.

Important points to discuss with patients before testing
Ideally, the likelihood of VUS results should be discussed with the patient before genomic testing. This is especially true for large panels and exome testing, where the likelihood of detecting multiple uncertain variants is high.

Patients should be aware that:

• Genetic variants are commonly reported
• Genetic variants can be either harmful or benign
• Genomic testing is a rapidly evolving field and some variants have not yet been classified
• Reclassification of a VUS may change over time, and most will be reclassified as benign

It is important to note that not all test reports will include VUS. Whether or not a test will detect and report out VUS can be determined by contacting the lab.