Breast cancer is one of the most common cancers among women, and susceptibility is explained by genetic, lifestyle and environmental components. Copy Number Variants (CNVs) are structural DNA variations that contribute to diverse phenotypes via gene-dosage effects or cis-regulation. In this study, we aimed to identify germline CNVs associated with breast cancer susceptibility and their relevance to prognosis. We performed whole genome CNV genotyping in 422 cases and 348 controls using Human Affymetrix SNP 6 array. Principal component analysis for population stratification revealed 84 outliers leaving 366 cases and 320 controls of Caucasian ancestry for association analysis; CNVs with frequency > 10% and overlapping with protein coding genes were considered for breast cancer risk and prognostic relevance. Coding genes within the CNVs identified were interrogated for gene-dosage effects by correlating copy number status with gene expression profiles in breast tumor tissue. We identified 200 CNVs associated with breast cancer (q-value < 0.05). Of these, 21 CNV regions (overlapping with 22 genes) also showed association with prognosis. We validated representative CNVs overlapping with APOBEC3B and GSTM1 genes using the TaqMan assay. Germline CNVs conferred dosage effects on gene expression in breast tissue. The candidate CNVs identified in this study warrant independent replication. Breast Cancer is one of the commonly diagnosed cancers among women worldwide 1 , in Canada, breast cancer accounts for about 25% of all diagnosed cancers, and 15% of all cancer deaths 2 . Based on twin studies, estimated heritable genetic factors contribute to about 30% for breast cancer risk, the remaining risk being due to environmental and lifestyle factors 3 . Family based linkage and genome sequencing studies have identified high and moderate penetrant mutations in genes such as BRCA 1 or BRCA 2 4,5 PTEN 6 , PALB2 7 , ATM 8 , TP53 9 , and CHECK2 10 that contribute to the genetic risk of breast cancers. Subsequently, large scale population based Genome Wide Association Studies (GWAS) were successful in identifying several low penetrant common genetic variants (Single Nucleotide Polymorphisms, SNPs) associated with breast cancer risk. Among these, a limited number of GWAS SNPs (7 SNPs) showed effect sizes (odds ratio or ORs) between 1.25-1.5 and the remaining SNPs showed effect sizes < 1.25 11,12 . SNP based GWAS served as a valuable tool in uncovering novel genes or loci associated with breast cancer aetiology. Low, moderate and high penetrant SNPs and mutations together explain up to 50% of the genetic risk associated with breast cancer 11,12 , and the remaining variants to explain the "missing heritability" are yet to be discovered. Copy Number Variations (CNVs) in the germline DNA are currently being investigated to explain missing heritable risk for breast cancer 13 . Germline CNVs are a class of structural variations, and are defined as loss or gain of genomic DNA in size range of 50 bp to 1 Mb 14 . Germline CNVs are studied as ...

Copy Number Variants (CNVs) are a class of structural variations of DNA. Germline CNVs are known to confer disease susceptibility, but their role in breast cancer warrants further investigations. We hypothesized that breast cancer associated germline CNVs contribute to disease risk through gene dosage or other post-transcriptional regulatory mechanisms, possibly through tissue specific expression of CNV-embedded small-noncoding RNAs (CNV-sncRNAs). Our objectives are to identify breast cancer associated CNVs using a genome wide association study (GWAS), identify sncRNA genes embedded within CNVs, confirm breast tissue (tumor and normal) expression of the sncRNAs, correlate their expression with germline copy status and identify pathways influenced by the genes regulated by sncRNAs. We used an association study design and accessed germline CNV data generated on Affymetrix Human SNP 6.0 array in 686 (in-house data) and 495 (TCGA data) subjects served as discovery and validation cohorts. We identified 1812 breast cancer associated CNVs harboring miRNAs (n = 38), piRNAs (n = 9865), snoRNAs (n = 71) and tRNAs (n = 12) genes. A subset of CNV-sncRNAs expressed in breast tissue, also showed correlation with germline copy status. We identified targets potentially regulated by miRNAs and snoRNAs. In summary, we demonstrate the potential impact of embedded CNV-sncRNAs on expression and regulation of down-stream targets.

In this cohort of patients with advanced cancer pain, TAOK3 SNPs were associated with opioid doses. This result supports the original findings from a GWAS in postoperative patients. The proportions of variant homozygotes (8.2% of patients) and their requirement for higher doses of opioids would appear potentially clinically important and should be validated in further studies.

Breast cancer (BC) predisposition in populations arises from both genetic and nongenetic risk factors. Structural variations such as copy number variations (CNVs) are heritable determinants for disease susceptibility. The primary objectives of this study are (1) to identify CNVs associated with sporadic BC using a genome-wide association study (GWAS) design; (2) to utilize 2 distinct CNV calling algorithms to identify concordant CNVs as a strategy to reduce false positive associations in the hypothesis-generating GWAS discovery phase, and (3) to identify potential candidate CNVs for follow-up replication studies. We used Affymetrix SNP Array 6.0 data profiled on Caucasian subjects (422 cases/348 controls) to call CNVs using algorithms implemented in Nexus Copy Number and Partek Genomics Suite software. Nexus algorithm identified CNVs associated with BC (731 autosomal CNVs with >5% frequency in the total sample and Q < 0.05). Thirteen CNVs were identified when Partek algorithm-called CNVs were overlapped with Nexus-identified CNVs; these CNVs showed concordances for frequency, effect size, and direction. Coding genes present within BC-associated CNVs were known to play a role in disease etiology and prognosis. Long noncoding RNAs identified within CNVs showed tissue-specific expression, indicating potential functional relevance of the findings. The identified candidate CNVs warrant independent replication.

We previously identified a novel breast cancer susceptibility variant on chromosome 4q31.22 locus (rs1429142) conferring risk among women of European ancestry. Here, we report replication of findings, validation of the variant in diverse populations and fine‐mapping of the associated locus in Caucasian population. The SNP rs1429142 (C/T, minor allele frequency 18%) showed association for the overall breast cancer risk in Stages 1–4 (n = 4,331 cases/4271 controls; p = 4.35 × 10−8; odds ratio, ORC‐allele,1.25), and an elevated risk among premenopausal women (n = 1,503 cases/4271 controls; p = 5.81 × 10−10; ORC‐allele 1.40) in European populations. SNP rs1429142 was associated with premenopausal breast cancer risk in women of African (T/C; p‐value 1.45 × 10−02; ORC‐allele 1.2) but not from Chinese ancestry. Fine‐mapping of the locus revealed several potential causal variants which are present within a single association signal, revealed from the conditional regression analysis. Functional annotation of the potential causal variants revealed three putative SNPs rs1366691, rs1429139 and rs7667633 with active enhancer functions inferred based on histone marks, DNase hypersensitive sites in breast cell line data. These putative variants were bound by transcription factors (C‐FOS, STAT1/3 and POL2/3) with known roles in inflammatory pathways. Furthermore, Hi‐C data revealed several short‐range interactions in the fine‐mapped locus harboring the putative variants. The fine mapped locus was predicted to be within a single topologically associated domain, potentially facilitating enhancer–promoter interactions possibly leading to the regulation of nearby genes.