A familial history is one of the most important risk factors of breast cancer. To date, several susceptibility genes, including BRCA1, BRCA2, ATM, CHEK2, BRIP2, PALB2 and RAD51, were identified; however, these genes would account for only 25% of familial breast cancer risk.1, 2

ZNF350/ZBRK1 (zinc finger and BRCA1-interacting protein with a KRAB domain 1) serves as a transcription repressor by interacting with BRCA1, and regulates the transcription of DNA damage response genes, such as GADD45a, p21, Bax and GADD153.3 ZNF350 is an interesting candidate gene in association with breast cancer susceptibility because the mutations in BRCA1-interacting proteins or the genes that are involved in the BRCA1 pathway were found in familial breast cancer patients.

In this issue of the journal, Plourde et al.4 reported the association between promoter variants of ZNF350 and breast cancer susceptibility in 96 high-risk non-BRCA1/2 French Canadian breast cancer families. They investigated the single-nucleotide polymorphisms (SNPs) in the ZNF350 gene region in previous report5 and promoter regions in this study.4 However, neither SNPs nor haplotypes in this promoter region were significantly associated with breast cancer risk, although the haplotype analysis combining the SNPs in promoter and gene regions showed associations with P<0.05. In the functional analysis by reporter assay, two of three haplotypes (H11 and H12) showed significantly increased transcriptional activities compared with H8, the most common haplotype.

There are, however, two major problems. First, they should consider multiple testing because they examined 23 SNPs and 33 haplotypes. When Bonferoni’s correction is applied, none of the SNPs had P-value<α=0.00217 (0.05/23) and no haplotype had P-value<α=0.00152 (0.05/33). Second, the risk haplotypes (which were not significant in the association study; H11, P=0.38, odds ratio≈2 and H12, P=1.0, odds ratio≈1) were associated with increased transcriptional activities, although ZNF350 serves as a tumor/metastatic suppressor and its expression was low in tumor tissue. They discussed about this critical discrepancy as the increase of ZNF350 expression could be involved in initiation rather than in development during tumorigenesis; however, there is no evidence supporting their speculation at present. Also, in a previous report investigating the association of ZNF350 SNPs with breast cancer susceptibility, no significant association was observed in three SNPs,6 including c.425 T>C (rs2278420), which showed marginal association (P=0.040) in this study. Although some SNPs in ZNF350 showed associations with P<0.05, consistent results have not been observed.7, 8, 9 These results suggest that ZNF350 SNPs, especially promoter SNPs, are not more likely to be the causative SNPs associated with familial breast cancer risks. To identify positive signal of low minor allele frequency and high odds ratio (moderate-penetrance susceptibility gene) in case–control association study, much more sample size is needed (more than 500 cases and 500 controls to obtain 80% statistical power to detect an effect with odds ratio of 2 or greater for allele (haplotype) with 2% frequency at α=0.05). Further analysis with larger sample size or replication study will be required to evaluate the impact of ZNF350 on breast cancer susceptibility.