Variant classification and reanalysis in exome testing

Variant interpretation is a complex, evolving science, and interpretations can change over time as new information is learned. When interpreting exome results, it is important to be aware that some variants initially detected through testing may not be included on the final report. Understanding the technical process of exome analysis and variant interpretation can help ordering and referring providers maximize the return of clinically relevant variants for their patients. 

Variant filtration and classification

DNA is first isolated from the patient’s sample, then specially designed probes isolate the exons, or coding regions, from the other 98-99% of the patient’s genome. After amplification and sequencing the exons, the patient’s exome data is analyzed using bioinformatic pipelines developed by the lab.

  • Variants that differ from the reference sequence are researched using multiple databases.
  • Typically, complex computer algorithms are used to predict the biological effect of the variants. Many common variants not believed to be pathogenic are filtered out.
  • Exome data is also filtered to prioritize the most likely causes related to the patient’s symptoms.

Importantly, the lab’s algorithms for variant filtration include an input of the patient’s clinical features and symptoms. Clinicians can improve the utility of exome testing in identifying a patient diagnosis by providing as much clinical information as possible, which prevents the elimination of possible causal variants from the test report. 

Variant reclassification

The classification of a variant – benign, uncertain, or pathogenic – can change over time. Genetic sequence variants are interpreted and classified by the laboratory which does the testing, based on current knowledge at the time of the testing. When more information is learned about these variants, the interpretation can change. Patients and families, as well as clinicians, should be aware of the possibility for reclassification of variants. 

Family studies

One method that is sometimes used to gather more information about a genetic variant is a family variant study. In such studies, the ordering provider – typically a genetic specialist - works with the family and the laboratory to purposefully select specific relatives within the family whose variant status could inform the likelihood that the variant is, or is not, the cause of the clinical findings in the family. Sometimes, these family studies can result in a reclassification of the variant.

Variant reanalysis

As technology and bioinformatic pipelines improve, and information regarding new gene-disease associations is published, there may be value to reanalyzing previously generated exome data. Clinical knowledge bases of gene-disease and variant-disease associations rapidly add new information each year. Recent studies showed an increase in diagnostic yield of 15-41% when reanalysis was performed one year later. 

Some labs automatically conduct the reanalysis, while others accept requests for a certain period of time after the original analysis. There may or may not be an additional charge. As part of ongoing patient care, clinicians may refer, or re-refer, to a genetic specialist if they have questions about reanalysis. 

The following are some situations when reanalysis can be helpful.

  • A variant of uncertain significance (VUS) was reported. Additional information about the impact of VUS on gene function may become available based on new research.
  • A variant in a gene suspected, but not known, to be associated with the patient’s reported phenotype was reported. There could be new information about the gene’s role in the condition of interest.
  • The lab reports that their bioinformatic pipelines have improved. These changes in lab processes may aid in the identification of variants previously undetected or not included in previous variant filtration processes.
  • The patient or family develops new clinical findings. Expansion of the clinical presentation may affect the variant analysis process and may lead to identification actionable variants.