User:Dpaulbick/sandbox

David Bick, MD at HudsonAlpha Institute for Biotechnology A gene of uncertain significance (GUS) is a gene that has not been associated with a phenotype in humans or it is a gene connected with a phenotype in humans that previously had not been connected to that gene.[1]

History With the success of the Human Genome Project and the development of Next-generation sequencing the number of human Genomes and Exomes sequenced each year is increasing dramatically [1]. DNA sequencing of the Genome or Exome (together referred to as genomic sequencing) is available world-wide to aid in the diagnosis of patient's with genetic disorders.

Laboratories that carry out genomic sequencing for patients find DNA differences between the patient's sequence and the sequence of the reference genome. These differences are called Genetic variants. Some variants do not cause disease and these are referred to as benign variants. The variants that result in different human Eye colors are benign variants. Some variants cause disease such as the Delta-F508 variant in CFTR, the gene associated with Cystic fibrosis. Variants that cause disease are referred to a pathogenic variants. Clinical laboratories also find variants in genes that they cannot interpret. These are called variants of uncertain significance.

A consistent variant classification system is central to the use of Genomics in patient care. Laboratories throughout the world need an evidence-based system to decide which Genetic variants are pathogenic and which ones are benign. Without such a system one laboratory using their own internal system will classify a variant as pathogenic while a second laboratory will classify the same variant as benign. This is confusing to the physician and dangerous to patients. Take for example BRCA1, the gene associated with Hereditary breast–ovarian cancer syndrome. A patient found to have a pathogenic variant in BRCA1 may elect prophalactic surgery [2]. If laboratories cannot agree on which BRCA1 variants are pathogenic then patients cannot make accurate decisions concerning bilateral mastectomy and oophorectomy.

The problem of uniform variant classification is further compounded by the many Incidental findings identified through genomic sequencing. When a patient has genomic testing the laboratory may find a pathogenic variant in a gene that is not related to the reason for testing. For example, when a child with a seizure disorder has genomic testing the laboratory may find a pathogenic variant in BRCA1. This finding is unrelated to the primary purpose of the test: a search for pathogenic variants that cause the seizure disorder. In this situation finding a pathogenic BRCA1 variant is an incidental finding (also referred to as a secondary finding). Accurate and consistent variant classification is proving to be a substantial problem in the assessment of incidental findings [2].

Efforts to create evidence-based variant classification systems for use by all laboratories has been underway for some time. The American College of Medical Genetics and Genomics proposed five categories of sequence variations for the purposes of clinical reporting in 2000 [3]. These guidelines were updated in 2008 describing six interpretative categories of sequence variation [3]. The International Agency for Research on Cancer (IARC) [4] proposed guidelines specifically targeted to cancer susceptibility genes [4] in 2008. They descibe a system of five classes of variants based on the degree of likelihood of pathogenicity. Recommendations for clinical management of at-risk individuals is based on class and syndrome. In 2011, the American College of Medical Genetics and Genomics proposed guidelines for interpretation and reporting of postnatal constitutional Copy-number variations. Three main categories of significance were proposed with three subdivisions within one catagory [5]. The Association for Clinical Genetic Science [5] proposed a 5 tiered classification system in 2013 [6]. A variant classification system for the genes BRCA1 and BRCA2 was published in 2014 [6]. In 2015, a five-category classification system was proposed for somatic variants identified in molecular profiling of cancer to simplified reporting of variant interpretations to treating oncologists [7]. Efforts to apply a 5-tiered scheme for standardized classification of mismatch repair gene variants has been published [8].

The American College of Medical Genetics and Genomics, the Association of Molecular Pathology and the College of American Pathologists convened a working group in 2013 to further refine standards and guidelines for the interpretation of sequence variants. Published in 2015, this guideline included standard terminology for variants [9]. The five types of variants described are: pathogenic variant, likely pathogenic variant, variant of uncertain significance, likely benign variant and benign variant. The working group [10] and others [7] argued that the terms Mutation and Polymorphism (biology) be abandoned when describing variants in humans as these lead to confusion.

The human genome has about 20,000 genes. About 5000 genes have been associated with a human disease. For these genes one or more pathogenic variants have been found that reliably link changes in the gene with a specific disease. As noted above The American College of Medical Genetics and Genomics, the Association of Molecular Pathology and the College of American Pathologists convened a working group in 2013 to refine standards and guidelines for the interpretation of sequence variants (ref) It became clear in working group discussions that this 5-tiered classification system could not be applied to genes that had never been connected to a human disease therefore one member proposed that these genes be classified as 'genes of uncertain significance'. The term GUS (pronounce 'goose') was proposed and adopted. No previous classification system provided guidance for this type of gene.

Application Genomic sequencing is now commonly used to try to establish the cause of disease in children and adults with possible genetic disorders where other methods have failed [11], [12]. Yet among the 20,000 genes in the Human genome only a few thousand have been associated with a human disese [8]. Therefore when clinical laboratories test patients they will sometimes find variants in a Gene that inactivate its function where the gene has not been connected to any human disease. Examples of the kind of variants that inactivate a gene's function include Nonsense mutations, Splice site mutations, and Frameshift mutations to name a few. As previously mentioned the word 'variant' is preferred to 'Mutation'. When this situation arises the laboratory may find that there is data from Model organisms connecting the gene in question with a Phenotype in an animal that is similar to the patient's disease. Additionally there may be other data about the gene's normal function in humans. If all of the data taken together is compelling, then this gene is designed a GUS.

Similarly, a clinical laboratory may find a variant in a gene that has been associated with a human disease but not the disease found in the patient that was tested. If there is sufficient data from Model organisms and other data sources the gene in this situation is also designed a GUS. It is important to realize that a single gene can result in several different diseases (patient Phenotypes). The gene LMNA is a good example [9]. We do not at present, appreciate all of the possible patient Phenotypes that result from pathogenic variants in different parts of a gene.

A number of clinical laboratories will report a gene that is a GUS. This provides families and the physician ordering the test an opportunity to engage with the research community to try to conclusively establish an association between a particular patient phenotype and a gene. While this can result in a wild GUS chase, many Phenotype-Genotype relationships have been established this way.

References Jump up ^. . Missing or empty |title= (help) Jump up ^. . Missing or empty |title= (help) Jump up ^. . Missing or empty |title= (help) Jump up ^. PMID 18951446. Missing or empty |title= (help) Jump up ^. PMID 21681106. Missing or empty |title= (help) Jump up ^. . Missing or empty |title= (help) Jump up ^. . Missing or empty |title= (help) Jump up ^. PMID 24362816. Missing or empty |title= (help) Jump up ^. PMID 25741868. Missing or empty |title= (help) Jump up ^. PMID 25741868. Missing or empty |title= (help) Jump up ^. . Missing or empty |title= (help) Jump up ^. . Missing or empty |title= (help)