Clinical Classification ofBRCA1andBRCA2DNA Sequence Variants: The Value of Cytokeratin Profiles and Evolutionary Analysis—A Report From the kConFab Investigators

Abstract

A B S T R A C T PurposeRare missense substitutions and in-frame deletions of BRCA1 and BRCA2 genes present a challenge for genetic counseling of individuals carrying such unclassified variants. We assessed the value of tumor immunohistochemical markers in conjunction with genetic and evolutionary approaches for investigating the clinical significance of unclassified variants. Patients and MethodsWe studied 10 BRCA1 and 12 BRCA2 variants identified in Australian families with breast cancer. Analyses assumed a prior probability based on revised cross-species sequence alignment methods assessing amino acid evolutionary conservation and position, combined with likelihoods from data on co-occurrence with pathogenic mutations in the same gene, segregation analysis, and immunohistochemistry. We specifically explored the value of estrogen receptor, cytokeratin 5/6, and cytokeratin 14 as tumor markers of BRCA1 mutation status. ResultsPosterior probabilities classified 72% of variants. BRCA1 variants IVS18ϩ1 GϾT (del exon 18) and 5632 T ϾA (V1838E) were classified as pathogenic, with Ͼ99% posterior probability of being deleterious, and tumor histopathology was particularly important for their classification. BRCA2 variant classification was improved over previous studies, largely by incorporating the prior probability of pathogenicity based on amino acid cross-species sequence alignments. ConclusionVariant classification was considerably improved by analysis of estrogen receptor, cytokeratin 5/6, and cytokeratin 14 tumor expression, and use of updated methods estimating the clinical relevance of amino acid evolutionary conservation and position. These methodologies may assist genetic counseling of individuals with unclassified sequence variants.

LOH analysis should not be used as a tool to assess whether UVs of BRCA1/2 are pathogenic or not

“…Thus, since a different result was not obtained for the neutral variants as compared to the pathological variants, we think that the potential pathogenicity or neutrality of the UV or NEW variants, which was the objective of this study, cannot be determined. In any case, and as reported by other authors (Spurdle et al [9]), an increased loss of the allele containing the variant as compared to the wild type allele was seen in tumor tissue, mainly in the variants classified as pathological and neutral or of low clinical significance.…”

“…Therefore, our results agree with what has been proposed by Spurdle et al [9] and disagree with the observations made by Dworkin et al [1]. In other words, loss of heterozygosity analysis in a tumor is not useful to distinguish between a neutral and a pathological variant because each group of variants (neutral, UV, and pathological) does not follow a stable pattern of loss of heterozygosity.…”

Clinical relevance of rare germline sequence variants in cancer genes: evolution and application of classification models

“…A major breakthrough in the field has been the development of a multifactorial likelihood classification model for BRCA1 and BRCA2, which essentially assesses a number of different independent features portrayed by a variant carrier, to establish the likelihood that the variant has the characteristics of known pathogenic mutation in BRCA1 or BRCA2 [3]. The model has been relatively widely used since that point in time, and has also undergone some development since its initial iteration -to refine and/or redefine likelihood ratios (LRs) for the features originally included, and to add additional features to the model [4][5][6][7][8][9][10][11][12]. Table 1 includes a brief comment on which features have been or will be included in multifactorial modelling for BRCA1/2 and MMR genes, with more detailed information included in Supplementary Table 1, together with a description of the caveats currently known to be associated with application of each feature.…”

“…Although variant frequency in unaffected controls has been used as a prescreen to identify likely neutral/low clinical significant variants [6,8], and recently to confirm the low clinical significance of a BRCA2 variant [13], this has yet to be included in the model itself. As shown in Figure 1, the BRCA1/2 multifactorial model currently includes an assessment of the prior probability of pathogenicity of a variant based on amino acid evolutionary conservation and physicochemical properties [14], and can include additional estimates of pathogenicity from LRs based on variant segregation in families [3,15], cooccurrence with a pathogenic mutation [3], reported family history of the proband [7,8,16] and tumour features [6,8,11,17]. There has been much recent development in the field to simplify the calculation of posterior probabilities, with a web-based site available for the assignment of A-GVGD prior probabilities [14] (http://agvgd.iarc.fr/ alignments.php), a simplified web-based method for the calculation of LRs for segregation analysis [15] (http:// www.msbi.nl/cosegregation), and detailed explanations of the derivations of pathology LRs [11,17].…”

“…As shown in Figure 1, the BRCA1/2 multifactorial model currently includes an assessment of the prior probability of pathogenicity of a variant based on amino acid evolutionary conservation and physicochemical properties [14], and can include additional estimates of pathogenicity from LRs based on variant segregation in families [3,15], cooccurrence with a pathogenic mutation [3], reported family history of the proband [7,8,16] and tumour features [6,8,11,17]. There has been much recent development in the field to simplify the calculation of posterior probabilities, with a web-based site available for the assignment of A-GVGD prior probabilities [14] (http://agvgd.iarc.fr/ alignments.php), a simplified web-based method for the calculation of LRs for segregation analysis [15] (http:// www.msbi.nl/cosegregation), and detailed explanations of the derivations of pathology LRs [11,17]. The ongoing availability of large datasets to assess co-occurrence and family history scores would be most advantageous, and this in an intended aim of the recently formed international collaborative research network for the analysis of BRCA1/2 variants entitled ENIGMA (Evidence-based Network for the Interpretation of Germline Mutant Alleles; http:// www.enigmaconsortium.org).…”