Criteria used to assess pathogenicity
PVS1 and PVS1_strong were considered met according to Tayoun decision
tree (Abou Tayoun et al., 2018; ClinGen-TP53_Expert_Panel, 2019), PS3
was weighted when a functional defect was found in at least 2
independent studies in absence of discordant results. PS4 was weighted
for variants with an odds ratio (OR) >5.0 in case-control
studies, PS4_moderate for low-moderate penetrant genes if the OR was
between 1.5 and 5, with a p-value <0.01 as long as the
phenotype was in accordance with the described for the gene. PM1, if the
variant affected a highly conserved amino acid located in the FHA and/or
kinase domain. PM2 was weighted when the variant was absent or in less
than 1 out of 100,000 alleles in gnomAD v2.1.1 from “all” non-cancer
population dataset; if present in ≥ 2 individuals within any
sub-population, it should be present in <1 out of 50,000
alleles in that subpopulation. Since some CHEK2 variants in spite
of being frequent in the population, the associated risk is significant,
PM2_supporting was applied if the variant was present in ≤ 1 out of
20,000 alleles in gnomAD v2.1.1 dataset (Karczewski et al., 2019). PP3
was weighted if the in silico predictors suggested a splicing
alteration (reduction of ≥20% in Alamut score) and/or protein function
alteration according to Varsome genome interpreter (Kopanos et al.,
2018). Variant classification was performed using different combination
of rules according to classical ACMG-AMP guidelines (S. Richards et al.,
2015), ClinGen-TP53 suggested modifications to ACMG
(ClinGen-TP53_Expert_Panel, 2019) and to ACMG-Bayesian modelling
(Tavtigian et al., 2018) (Table 1).
Risk allele categorization was ascertained when possible as previously
described (Senol-Cosar et al., 2019) (Table S2). Accordingly, ERA
classification was given to variants reported in multiple association
studies or to those determined by robust meta-analysis; likely risk
allele (LRA) was assigned if either the variant showed association in at
least 2 independent studies, had been reported in a large study of high
quality or in multiple studies with almost complete concordance .