Vol. 25 • Issue 6 • Page 41
Molecular
Next-generation sequencing (NGS) technology has progressed significantly in recent years, with new chemistries and instrumentation allowing these sequencers to be employed in the clinical laboratory. Advanced bioinformatics have allowed the sensitivities and specificities of NGS sequencing to approach or exceed that of the previous gold standard, Sanger Sequencing.1 The end result is that it is now possible to simultaneously sequence several genes rapidly and in a cost-effective manner.
Prior to the broad use of NGS in the clinical laboratory, the approach to the analysis of patients with a clinical or family history suggestive of hereditary breast and ovarian cancer (HBOC) was limited to Sanger Sequencing of the BRCA1 and BRCA2 genes. The ability to analyze several genes simultaneously at a reasonable cost has led laboratories to offer panels of several genes for these patients. This has led to many surprising observations.
Genes long thought to be associated with a single hereditary cancer syndrome, such as CDH1 in gastric cancer, could cause isolated breast cancer in some families. This observation led some laboratories to offer sequencing panels of known hereditary cancer genes for the evaluation of patients for HBOC. Loss of function variants in these genes-BRCA1 and BRCA2 (HBOC), TP53 (Li-Fraumeni), STK11 (Peutz-Jeghers Syndrome), PTEN (Cowden Syndrome), CDH1 (Hereditary Diffuse Gastric Cancer) and PALB2 (Fanconi Anemia)-is well established as a cause of autosomal dominant hereditary cancer syndromes.
In an individual with breast cancer, pathogenic variants in BRCA1 or BRCA2 are far more likely to be implicated than other genetic variants. Because of this, our lab offers a reflex test where BRCA1 and BRCA2 are analyzed first. If a pathogenic or likely pathogenic variant is found, the analysis stops. If no deleterious variant is found in BRCA1 or BRCA2, the analysis proceeds to the genes in the larger panel. This linear approach prioritizes variants that are more prevalent, well-characterized and clinically actionable.
These “high-risk” genes have incomplete penetrance, meaning not all individuals harboring a pathogenic variant will develop cancer in their lifetime. However, the risks are several-fold higher than the general population and many clinicians will choose to act on the information in terms of recommendations for surveillance and surgical or chemical prophylaxis. It should be noted that large rearrangements in these genes can also cause loss of function, so a complete analysis must include deletion/duplication analysis as well.
As medical knowledge grows, new links between unknown genetic variants and cancer risk will emerge. Physicians, patients and laboratorians must carefully balance the potential benefits and harms of including information about these novel variants in their clinical decisions. So-called “moderate risk” genes, offered in some testing panels, have much less clinical information regarding the cancer risk for individuals harboring pathogenic variants for these genes. Whether to use these expanded panels is a personal choice of the physician and patient, and it involves their tolerance for ambiguity and uncertainty.
Another challenge with the larger expanded gene panels is the greater likelihood of detecting a variant of uncertain significance (VUS). Since experience with these genes and their penetrance is limited, it is extremely difficult to assess rare variants in these genes.
Advanced NGS testing creates a paradox. As an increasing amount of sequencing information is generated, the number of variants discovered in any analysis for any given patient increases. Some variants will be well described and can be reported immediately, but many will be less common or even unique and require a process known as assessment of pathogenicity. This is sometimes referred to as variant annotation.
This assessment requires research using literature, databases, structural predictive tools, family studies, functional studies and other tools. Eventually, the variant will be classified as benign, likely benign, VUS, likely pathogenic or pathogenic. Ironically, as the ðlaboratory costs of NGS sequencing fall, the cost of variant annotation has increased.
The growing number of sequencing results returned to providers and patients will also increase pressure to reclassify VUS into either pathogenic or benign categories. Initially, databases were an important source for reclassification of VUS. For variants with reasonable frequencies, there would be a “critical mass” where conclusions could be based on frequencies alone. However, for rare and unique variants, databasing alone has limited value to reclassify VUSs as even the most extensive databases will have only a handful of patients with these variants.
In order to improve the ability of laboratories to classify variants, a multinational program called BRCA Share has been founded. Participating clinical laboratories and academics can upload BRCA variant data to a common, well-curated database. A committee periodically reviews annotations and has the ability to commission functional and family studies in order to investigate rare VUSs. This effort will allow the clinical laboratory community to have a state-of-the-art resource to aid in the annotation of BRCA variants.
As we move forward in the genomics era of clinical testing, laboratories must establish new areas of expertise in order to help physicians provide efficacious care. We must leverage scientific discovery and diagnostic innovation as well as collaboration. Bioinformatics and variant assessment must be added to the standard repertoire of competences required to operate a clinical laboratory. Physicians and patients need more than a variant list, which is too often the case with today’s gene panels. In order for information on variants to be useful, laboratories must provide insight-not simply data.
Charles (Buck) Strom, MD, PhD, FAAP, FACMG, HCLD, is vice president of Genetics and Genomics at Quest Diagnostics.
References
1. Benowitz, S. New study challenges gold standard for validating DNA sequencing results.National Human Genome Research Institute. https://www.genome.gov/27564480/
