Consultation informs strategies for improving the use of functional evidence in variant classification

Abstract

When investigating whether a variant identified by diagnostic genetic testing is causal for disease, applied genetics professionals evaluate all available evidence to assign a clinical classification. Functional assays of higher and higher throughput are increasingly being generated and, when appropriate, can provide strong functional evidence for or against pathogenicity in variant classification. Despite functional assay data representing unprecedented value for genomic diagnostics, challenges remain around the application of functional evidence in variant curation. To investigate a growing gap articulated in recent international studies, we surveyed genetic diagnostic professionals in Australasia to assess their application of functional evidence in clinical practice. The survey results echo the universal difficulty in evaluating functional evidence but expand on this by indicating that even self-proclaimed expert respondents are not confident to apply functional evidence, mainly due to uncertainty around practice recommendations. Respondents also identified the need for support resources and educational opportunities, and in particular requested expert recommendations and updated practice guidelines to improve translation of experimental data to curation evidence. We then collated a list of 226 functional assays and the evidence strength recommended by 19 ClinGen Variant Curation Expert Panels. Specific assays for more than 45,000 variants were evaluated, but evidence recommendations were generally limited to lower throughput and strength. As an initial step, we provide our collated list of assay evidence as a source of international expert opinion on the evaluation of functional- evidence and conclude that these results highlight an opportunity to develop additional support resources to fully utilize functional evidence in clinical practice.

Keywords: ACMG/AMP guidelines; assay; clinical genomics; diagnostic genetics; education; functional evidence; pathogenicity assessment; variant classification.

Graphical abstract

Figure 1

Comfort and familiarity of survey participants with the ACMG/AMP framework and associated recommendations and tools for the application of functional evidence (A) Confidence level of participants according to the rating scale “1 = not at all comfortable” to “5 = very comfortable” for each of the indicated types of functional evidence: biochemical assays (e.g., enzyme assays) (n = 32), transcript assays (e.g., splicing assays or transcriptome data) (n = 32), cell models (for, e.g., in vitro cell assays) (n = 31), animal models (e.g., mouse models) (n = 32), and high-throughput functional assays (including multiplexed assays of variant effect) (n = 32). (B–E) Number of participants according to awareness level category (“use,” “are aware of,” or “do not know of”) for the indicated resources for evaluating functional evidence. Note that awareness level was assessed independently of the standard operating procedure used by their practice and/or institution (details for standard operating procedures are shown separately in Figure S5). (B) Functional-evidence recommendations. Awareness of ClinGen Sequence Variant Interpretation (SVI) Working Group functional-evidence recommendations (n = 33). (C) Splicing recommendations. Awareness of ClinGen SVI Splicing Subgroup recommendations (n = 32, one “not applicable” result excluded). (D) SVI functional assay worksheet. Awareness of ClinGen SVI Functional Assay Documentation Worksheet (www.clinicalgenome.org/docs/svi-functional-assay-documentation-worksheet/) (n = 33). (E) VCEP CSpec guidance. Awareness of VCEP specific guidance on the use of functional evidence (https://cspec.genome.network/cspec/ui/svi/) (n = 33).

Figure 2

Participant views related to the application of functional evidence Viewpoints were framed around three questions (A–C) on barriers, enablers, and resources for future use, with the option to select all that apply, and to provide free text answers to an “other” option. (A) Which of the following might prevent your use of functional evidence in variant classification? (B) What would improve your interaction with and use of functional evidence (existing or not)? (C) Which of the following training and/or additional resources would you use to increase/improve your use/understanding of applying functional evidence?

Figure 3

Summary of functional assays specifically mentioned within VCEP CSpecs (A) Number of assays identified for each of the 33 genes covered by the VCEP CSpecs (total n = 226). (B) Number of variants per assay across all VCEP CSpecs. (C) Distribution of pathogenic evidence strength (applicable under ACMG/AMP code PS3) recommended across all the specific assays listed in CSpecs. (D) Distribution of benign evidence strength (applicable under ACMG/AMP code BS3) recommended across all the specific assays listed in CSpecs. A full list of specific assays collated can be found in Table S1. Darker bar regions indicate where VCEPs denote additional considerations and/or modifications for application of the given evidence level.