Strong founder effect for BRCA1 c.3629_3630delAG pathogenic variant in Chechen patients with breast or ovarian cancer

Abstract Coding sequences of BRCA1, BRCA2, ATM, TP53, and PALB2 genes were analyzed in 68 consecutive Chechen patients with high‐grade serous ovarian cancer (HGSOC). Pathogenic BRCA1/2 variants were identified in 15 (22%) out of 68 HGSOC cases. Nine out of ten patients with BRCA1 pathogenic alleles carried the same deletion (c.3629_3630delAG), and three out of five BRCA2 heterozygotes had Q3299X allele. The analysis of 49 consecutive patients with triple‐negative breast cancer (TNBC) revealed 3 (6%) additional BRCA1 heterozygotes. All women with BRCA1 c.3629_3630delAG allele also carried linked c.1067G>A (Q356R) single nucleotide polymorphism, indicating that this is a genuine founder variant but not a mutational hotspot. An ATM truncating allele was detected in one HGSOC patient. There were no women with TP53 or PALB2 germline alterations. Genetic analysis of non‐selected HGSOC patients is an efficient tool for the identification of ethnicity‐specific BRCA1/2 pathogenic variants.

over the past time, are usually characterized by distinct patterns of hereditary diseases and pathogenic variants. 3,4 Chechens are a large ethnic group residing mainly in the North Caucasus. The total number of Chechens is approximately 2 million, with the majority of them living in the Chechen Republic (Russia). Interethnic marriage is uncommon in the Chechen community. 5,6 Hence, there is a high probability that hereditary diseases observed in the Chechen population are attributed mainly to recurrent alleles. The primary aim of this study was the analysis of the spectrum of BRCA1/2 germline mutations in patients of Chechen origin. The annual number of new BC and OC cases across the Chechen population is approximately 370 and 40, respectively. 7 Given that only a subset of these patients belongs to clinical subgroups with hereditary features of cancer predisposition (e.g., early-onset, family history, or disease bilaterality for BC patients, or high-grade serous histology for OC patients), we estimated that it may take years to collect a substantial number of these patients prospectively. Several reports demonstrated that consecutive (i.e., unselected for family history or age at onset) series of high-grade serous OC (HGSOC) were characterized by a high frequency of BRCA1/2 mutations. [8][9][10] We anticipated that the retrospective analysis of relatively small numbers of HGSOC patients would reveal whether BRCA1/2 germline mutations persist in the Chechen population and whether the detection of BRCA1/2 pathogenic variants can be simplified because of the existence of a founder effect.
The study included 68 consecutive HGSOC patients who underwent treatment in the Chechen Republican Cancer Center (Grozny, Russia) within the years 2011-2020. Normal archival tissues, which were obtained during medical procedures and were stored in the morphological archive of this hospital, served as a source for DNA (n = 58). In addition, we enrolled prospectively 52 consecutive patients (10 with HGSOC and 42 with BC) who were referred to this clinic between August 2020 and June 2021 and provided blood for genetic analysis. When a high frequency of BRCA1 mutations in HGSOC was revealed, we added 37 archival triple-negative BC cases to the study. The investigation was approved by the local Ethics Committee. Clinical characteristics are described in Table 1.
DNA isolation from formalin-fixed paraffin-embedded (FFPE) samples was performed using Cobas DNA Sample Preparation Kit (Roche). DNA from blood lymphocytes was extracted by conventional protocol. 11 Although the primary goal of the study was the analysis of BRCA1/2 mutations, we added to the panel PALB2, ATM, and TP53 genes considering their well-established role in cancer predisposition. Genomic DNA (400 ng) was subjected to library preparation with KAPA HyperPlus Kit (Roche) according to manufacturer instructions. For FFPE samples, the adapter ligation time was extended to 20 h. Dual-index libraries were used to pool up to 96 samples into one enrichment reaction. Pooled DNA libraries (1000 ng) were enriched with a custom panel of biotinylated probes covering coding sequences as well as exon-intron boundaries and 5′-and 3′-untranslated regions of the above-mentioned genes. Due to the small size of the panel, two-round overnight hybridization was performed. Enriched libraries were sequenced on the Illumina NextSeq 500 platform with the Mid Output Kit v2.5 reagents in a paired-end mode for 150 cycles in both orientations. Mean depth-of-coverage reached ×1500 with 99.9% of bases read at least 100 times. The bioinformatic pipeline included standard steps, i.e., FASTQ files generation, quality assessment, and mapping of the obtained sequences to the hg19 genome using the BWA tool. Aligned reads were subjected to SNVs and indels calling with the HaplotypeCaller [GATK4]. Annotation was made with the SnpEff software instrument.
Next-generation sequencing (NGS) analysis was performed for 68 consecutive Chechen patients with HGSOC. BRCA1/2 pathogenic variants were observed in 15 (22%) out of 68 HGSOC cases (  25 we expected to increase the number of BRCA1 heterozygotes. Two additional cases of BRCA1 pathogenic variants (c.1338_1339delAG: n = 1; c.5296delA: n = 1) were revealed. All 9 OC and 1 BC Chechen patients with the BRCA1 c.3629_3630delAG allele also carried linked c.1067G>A (Q356R; rs1799950) single nucleotide polymorphism (SNP). The population frequency of rs1799950 according to the gnomAD database is 0.05, while in the current study, it was observed in 11.5% of women of Chechen ethnicity. This SNP is not associated with pathogenic BRCA1 variants in patients of Slavic origin (data not shown). Taken together, BRCA1 Q356R is likely to be a benign substitution, being a passenger for BRCA1 c.3629_3630delAG pathogenic variant. We investigated further whether this allele is located in cis or trans relative to the BRCA1 c.3629_3630delAG. Five carcinomas were informative for the loss-of-heterozygosity analysis; all these tumors demonstrated concurrent loss of c.3629_3630delAG and c.1067G>A alleles indicating that they are located on the same chromosome. Therefore, BRCA1 c.3629_3630delAG is a genuine founder allele but not a mutational hot spot. BRCA2 Q3299X is described in the ClinVar database as a pathogenic variant. Because it is located relatively close to the end of the coding portion of this gene, the evidence for its disease-predisposing role is considered to be weak. To clarify the contribution of this variant to cancer predisposition, we examined the loss-of-heterozygosity (LOH) status at this locus in the tumor tissue and revealed somatic loss of the normal BRCA2 allele in two out of three analyzed BCs. One HGSOC patient, who was negative for BRCA1/2 germline mutations, was heterozygous for ATM pathogenic variant (c.3511C>T [Q1171X; rs876659067]) ( Table 2). There were no women with TP53 or PALB2 germline alterations.
This study revealed a strong founder effect for BRCA1 c.3629_3630delAG allele in Chechen BC and OC patients. BRCA1 c.3629_3630delAG was firstly described in a French study. 12 The analysis of large series of Russian BC and OC patients revealed one carrier of this variant in a woman with apparent Russian ethnicity. 26 It is safe to conclude that currently available data do not demonstrate the elevated frequency of BRCA1 c.3629_3630delAG allele in any of the studied populations, i.e., this variant appears to be characteristic specifically for Chechens. The detection of the founder mutations in Chechen patients is anticipated from historical and genetic studies. 5,6,27,28 Haplotyping analysis revealed a well-preserved genetic identity of Chechens residing in various parts of the Caucasus mountains. 27 Strong founder effect was demonstrated for patients with cystic fibrosis registered in the Chechen Republic. 28 The identification of recurrent Chechen mutation BRCA1 c.3629_3630delAG may have immediate clinical implication, as it allows non-expensive screening of non-selected BC and OC patients as well as healthy people with instances of BC or OC in their families. The analysis of founder pathogenic variants is significantly cheaper as compared to full-length BRCA1/2 testing; therefore, this approach may therefore be more financially viable.
The analysis of consecutive or randomly selected HGSOC patients is apparently the most efficient approach for the analysis of the spectrum of BRCA1/2 mutations in previously unstudied populations. However, it may cause some biases given that some mutations located within BRCA1 and BRCA2 genes demonstrate a preferential association with OC, while other pathogenic alleles tend to be linked to BC predisposition. Consequently, if the study is limited to HGSOC patients, some recurrent BC predisposing alleles may be missed. Notably, the BRCA1 c.3629_3630delAG allele identified in the present study is located within the OC cluster region. 26,29 In conclusion, this small-scale study identified recurrent BRCA1 pathogenic variant characteristic for Chechen BC and OC patients. It also demonstrated the utility of HGSOC testing for rapid ethnicity-specific mapping of the spectrum of BRCA1/2 germline mutations.