CanVIG-UK Consensus Specification

Changelog (v3.30)​

• Explanatory Notes: Amended threshold for Likely Benign/Benign boundary to match Tavtigian et al 2020 and ACGS 2026 guidance

• PS4: Updated wording to reflect access of UK Biobank variant counts via gnomAD v4.1; case-counting updated to confirm use of all of gnomAD v4.1, and added caution where the ancestry of probands/families used for case-counting are poorly represented in gnomAD.

• PVS1: Guidance for nonsense variants at the 5’ end of the gene updated to align with ACGS 2026

• PS1: Formatting update to separate protein impact guidance for missense and initiation codon variants

• PM5: Clarification added to confirm Pathogenic reference variant requirement for Moderate, and updated the requirements to meet Supporting.

• PP1/BS4: Added allowance for use of COOL tool for co-segregation analysis

• PP3/BP4: Reworded to confirm SpliceAI as the preferred splicing predictor, and updated REVEL threshold to align with ACGS 2026

• PP4: Updated wording to point to gene-specific guidance for MMR/SDHB/SDHD genes

• BA1/BS1: Removed reference to female-only populations per current gene-specific guidelines

• BS2/BP2: Added reference to use of VCEP guidance

• BP7: Clarified that listed conservation tools are examples

• Combinations table: Removed restriction for use of PM5 and PP3 in combination (previously combination restriction no.12)

General Guidance Notes:

• Evidence items for which CanVIG-UK has offered additional specification are shaded in grey. Evidence items are shaded in white where there is no additional specification beyond ACGS Best Practice Guidelines version 1.2 (20/02/2024).

• Gene specific guidance for specific CSGs can be viewed at https://www.cangene-canvaruk.org/gene-specific-recommendations and should be followed for genes where these exist. These include CanVIG-UK gene specific guidance and gene specific guidance from ClinGen Sequence Variant Interpretation (SVI) Working Groups (+/- notes from CanVIG-UK).

• Evidence items can be combined using evidence (exponent) points for evidence towards pathogenicity (Very Strong= 8, Strong= 4, Moderate= 2, Supporting= 1) or towards benignity (Very Strong= -8, Strong= -4, Moderate= -2, Supporting= -1). Thresholds: ≥10 (Pathogenic), 6-9 (Likely Pathogenic), (-1) – (-6) (Likely Benign), ≤-7 (Benign). It is recommended that evidence criteria and evidence (exponent) scores are included on clinical reports.

• ≥2 concordant evidence items are required for a classification of likely pathogenic/pathogenic/likely benign/benign, with the exception of BA1, which provides standalone evidence towards benignity

• Variants should be reported using HGVS nomenclature, including the clinically appropriate transcript and version number (e.g. MANE select and/or MANE clinical plus) and human reference genome build.

• This specification can be used for small sequence variants and intragenic copy number variants. For copy number variants including the whole gene, or overlapping either end of the gene (including UTRs), refer to the ACMG CNV guidance and ACGS Best Practice Guidelines 2024. Professional judgement should always be used when evaluating the evidence surrounding a particular genomic variant and assigning a classification.

Changelog: 

• v1.1: Added clarification that BS2 may also be downgraded by one benignity evidence strength level 

General Guidance Notes: None