ABSTRACT
Pathogenic mutations in BRCA1 (BReast CAncer gene 1) confer high risks of both breast (up to 70%) and ovarian (up to 40%) cancers. Zinc (Zn) and copper (Cu) are essential for various physiological functions, including antioxidant reactions. Their balance, reflected in the Zn/Cu ratio, plays a crucial role in maintaining redox homeostasis, which is vital for cancer prevention. This study examines the antioxidant properties of Zn and Cu, specifically focusing on the blood Zn/Cu ratio as a potential marker for cancer risk among BRCA1 mutation carriers. The study cohort consisted of 989 initially unaffected women, followed up for 7.5 years. Blood samples were analyzed using inductively coupled plasma mass spectrometry. Although individual Zn and Cu levels did not significantly correlate with overall cancer risk, those women with a Zn/Cu ratio above 6.38 experienced a significantly lower cancer risk than women with a ratio below this cut-off point. This suggests that the Zn/Cu ratio may be a valuable biomarker for cancer prevention in this high-risk group. Given the increased cancer risk in BRCA1 mutation carriers, optimizing Zn and Cu levels through dietary and active interventions could provide a preventive strategy.
ABSTRACT
Breast cancer and ovarian cancer pose a significant risk for BRCA1 carriers, with limited risk-reduction strategies. While improved screening helps in the early detection of breast cancer, preventive measures remain elusive. Emerging evidence suggests a potential link between iodine levels and modulation of cancer risk, but comprehensive studies are scarce. We conducted a prospective study among 989 BRCA1 carriers to assess the association between blood iodine levels and breast and ovarian cancer risk. Using inductively coupled plasma mass spectrometry, we measured blood iodine levels and observed a negative association with breast cancer risk, with a significantly lower risk observed in quartile 4 (iodine > 38.0 µg/L) compared with quartile 1 (iodine < 30 µg/L) (HR = 0.49; 95%CI: 0.27-0.87; p = 0.01). Conversely, a suggestive increase in ovarian cancer risk was observed at higher iodine levels (HR = 1.91; 95%CI: 0.64-5.67; p = 0.25). No significant association was found between iodine levels and overall cancer risk. Our results suggest the potential of iodine to reduce breast cancer risk in BRCA1 carriers after prophylactic oophorectomy but require further validation and investigation of its effect on ovarian cancer risk and overall mortality. These findings highlight the need for personalized strategies to manage cancer risk in BRCA1 carriers.
Subject(s)
BRCA1 Protein , Breast Neoplasms , Iodine , Ovarian Neoplasms , Humans , Female , Breast Neoplasms/blood , Breast Neoplasms/genetics , Ovarian Neoplasms/blood , Ovarian Neoplasms/genetics , Iodine/blood , Middle Aged , Adult , Prospective Studies , BRCA1 Protein/genetics , BRCA1 Protein/blood , Risk Factors , Heterozygote , Biomarkers, Tumor/blood , AgedABSTRACT
BRCA1 mutations predispose women to breast and ovarian cancer. The anticancer effect of zinc is typically linked to its antioxidant abilities and protecting cells against oxidative stress. Zinc regulates key processes in cancer development, including DNA repair, gene expression, and apoptosis. We took a blood sample from 989 female BRCA1 mutation carriers who were initially unaffected by cancer and followed them for a mean of 7.5 years thereafter. There were 172 incident cases of cancer, including 121 cases of breast cancer, 29 cases of ovarian cancers, and 22 cancers at other sites. A zinc level in the lowest tertile was associated with a modestly higher risk of ovarian cancer compared to women with zinc levels in the upper two tertiles (HR = 1.65; 95% CI 0.80 to 3.44; p = 0.18), but this was not significant. Among those women with zinc levels in the lowest tertile, the 10-year cumulative risk of ovarian cancer was 6.1%. Among those in the top two tertiles of zinc level, the ten-year cumulative risk of ovarian cancer was 4.7%. There was no significant association between zinc level and breast cancer risk. Our preliminary study does not support an association between serum zinc level and cancer risk in BRCA1 mutation carriers.
ABSTRACT
BRCA1 mutations substantially elevate the risks of breast and ovarian cancer. Various modifiers, including environmental factors, can influence cancer risk. Lead, a known carcinogen, has been associated with various cancers, but its impact on BRCA1 carriers remains unexplored. A cohort of 989 BRCA1 mutation carriers underwent genetic testing at the Pomeranian Medical University, Poland. Blood lead levels were measured using inductively coupled plasma mass spectrometry. Each subject was assigned to a category based on their tertile of blood lead. Cox regression analysis was used to assess cancer risk associations. Elevated blood lead levels (>13.6 µg/L) were associated with an increased risk of ovarian cancer (univariable: HR = 3.33; 95% CI: 1.23-9.00; p = 0.02; multivariable: HR = 2.10; 95% CI: 0.73-6.01; p = 0.17). No significant correlation was found with breast cancer risk. High blood lead levels are associated with increased risk of ovarian cancer in BRCA1 carriers, suggesting priority for preventive salpingo-oophorectomy. Potential risk reduction strategies include detoxification. Validation in diverse populations and exploration of detoxification methods for lowering lead levels are required.
Subject(s)
BRCA1 Protein , Lead , Ovarian Neoplasms , Humans , Female , Ovarian Neoplasms/blood , Ovarian Neoplasms/genetics , Lead/blood , Adult , Middle Aged , BRCA1 Protein/genetics , Risk Factors , Poland , Heterozygote , Mutation , Biomarkers, Tumor/blood , Biomarkers, Tumor/genetics , Genetic Predisposition to Disease , Aged , Proportional Hazards ModelsABSTRACT
Importance: Magnetic resonance imaging (MRI) surveillance is offered to women with a pathogenic variant in the BRCA1 or BRCA2 gene who face a high lifetime risk of breast cancer. Surveillance with MRI is effective in downstaging breast cancers, but the association of MRI surveillance with mortality risk has not been well defined. Objective: To compare breast cancer mortality rates in women with a BRCA1 or BRCA2 sequence variation who entered an MRI surveillance program with those who did not. Design, Setting, and Participants: Women with a BRCA1 or BRCA2 sequence variation were identified from 59 participating centers in 11 countries. Participants completed a baseline questionnaire between 1995 and 2015 and a follow-up questionnaire every 2 years to document screening histories, incident cancers, and vital status. Women who had breast cancer, a screening MRI examination, or bilateral mastectomy prior to enrollment were excluded. Participants were followed up from age 30 years (or the date of the baseline questionnaire, whichever was later) until age 75 years, the last follow-up, or death from breast cancer. Data were analyzed from January 1 to July 31, 2023. Exposures: Entrance into an MRI surveillance program. Main Outcomes and Measures: Cox proportional hazards modeling was used to estimate the hazard ratios (HRs) and 95% CIs for breast cancer mortality associated with MRI surveillance compared with no MRI surveillance using a time-dependent analysis. Results: A total of 2488 women (mean [range] age at study entry 41.2 [30-69] years), with a sequence variation in the BRCA1 (n = 2004) or BRCA2 (n = 484) genes were included in the analysis. Of these participants, 1756 (70.6%) had at least 1 screening MRI examination and 732 women (29.4%) did not. After a mean follow-up of 9.2 years, 344 women (13.8%) developed breast cancer and 35 women (1.4%) died of breast cancer. The age-adjusted HRs for breast cancer mortality associated with entering an MRI surveillance program were 0.20 (95% CI, 0.10-0.43; P < .001) for women with BRCA1 sequence variations and 0.87 (95% CI, 0.10-17.25; P = .93) for women with BRCA2 sequence variations. Conclusion and Relevance: Results of this cohort study suggest that among women with a BRCA1 sequence variation, MRI surveillance was associated with a significant reduction in breast cancer mortality compared with no MRI surveillance. Further studies of women with BRCA2 sequence variations are needed to ascertain these women obtain the same benefits associated with MRI surveillance.
Subject(s)
Breast Neoplasms , Female , Humans , Adult , Aged , Middle Aged , Breast Neoplasms/pathology , BRCA1 Protein/genetics , Genes, BRCA2 , BRCA2 Protein/genetics , Mastectomy , Cohort Studies , Genes, BRCA1 , Mutation , Risk Management , Magnetic Resonance ImagingABSTRACT
The APOBEC3B gene belongs to a cluster of DNA-editing enzymes on chromosome 22 and encodes an activation-induced cytidine deaminase. A large deletion of APOBEC3B was associated with increased breast cancer risk, but the evidence is inconclusive. To investigate whether or not APOBEC3B is a breast cancer susceptibility gene, we sequenced this gene in 617 Polish patients with hereditary breast cancer. We detected a single recurrent truncating mutation (c.783delG, p.Val262Phefs) in four of the 617 (0.65%) hereditary cases by sequencing. We then genotyped an additional 12,484 women with unselected breast cancer and 3740 cancer-free women for the c.783delG mutation. The APOBEC3B c.783delG allele was detected in 60 (0.48%) unselected cases and 19 (0.51%) controls (OR = 0.95, 95% CI 0.56-1.59, p = 0.94). The allele was present in 8 of 1968 (0.41%) familial breast cancer patients from unselected cases (OR = 0.80, 95% CI 0.35-1.83, p = 0.74). Clinical characteristics of breast tumors in carriers of the APOBEC3B mutation and non-carriers were similar. No cancer type was more frequent in the relatives of mutation carriers than in those of non-carriers. We conclude the APOBEC3B deleterious mutation p.Val262Phefs does not confer breast cancer risk. These data do not support the hypothesis that APOBEC3B is a breast cancer susceptibility gene.
Subject(s)
Breast Neoplasms , Female , Humans , Breast Neoplasms/genetics , Breast Neoplasms/pathology , Cytidine Deaminase/genetics , Mutation , PolandABSTRACT
Several breast cancer susceptibility genes have been discovered, but more are likely to exist. To identify additional breast cancer susceptibility genes, we used the founder population of Poland and performed whole-exome sequencing on 510 women with familial breast cancer and 308 control subjects. We identified a rare mutation in ATRIP (GenBank: NM_130384.3: c.1152_1155del [p.Gly385Ter]) in two women with breast cancer. At the validation phase, we found this variant in 42/16,085 unselected Polish breast cancer-affected individuals and in 11/9,285 control subjects (OR = 2.14, 95% CI = 1.13-4.28, p = 0.02). By analyzing the sequence data of the UK Biobank study participants (450,000 individuals), we identified ATRIP loss-of-function variants among 13/15,643 breast cancer-affected individuals versus 40/157,943 control subjects (OR = 3.28, 95% CI = 1.76-6.14, p < 0.001). Immunohistochemistry and functional studies showed the ATRIP c.1152_1155del variant allele is weakly expressed compared to the wild-type allele, and truncated ATRIP fails to perform its normal function to prevent replicative stress. We showed that tumors of women with breast cancer who have a germline ATRIP mutation have loss of heterozygosity at the site of ATRIP mutation and genomic homologous recombination deficiency. ATRIP is a critical partner of ATR that binds to RPA coating single-stranded DNA at sites of stalled DNA replication forks. Proper activation of ATR-ATRIP elicits a DNA damage checkpoint crucial in regulating cellular responses to DNA replication stress. Based on our observations, we conclude ATRIP is a breast cancer susceptibility gene candidate linking DNA replication stress to breast cancer.
Subject(s)
Adaptor Proteins, Signal Transducing , Breast Neoplasms , DNA-Binding Proteins , Female , Humans , Adaptor Proteins, Signal Transducing/genetics , Biological Specimen Banks , Breast Neoplasms/genetics , Cell Cycle Proteins/genetics , DNA Damage , DNA-Binding Proteins/genetics , Poland/epidemiology , Replication Protein A/genetics , Replication Protein A/metabolism , United Kingdom/epidemiologyABSTRACT
PURPOSE: The BRCA2 p.K3326* variant is considered a low-penetrance variant for breast cancer. Aldehydes that accumulate in cells under insufficient aldehyde oxidation were most recently shown to trigger carcinogenesis by promoting depletion of BRCA2 protein. Allele T of the common variant rs10744777 in the ALDH2 gene was associated with reduced expression of aldehyde dehydrogenase, the main enzyme in aldehyde oxidation. We hypothesized that this allele could modify breast cancer risk in women with the BRCA2 p.K3326* low-penetrance variant through reduced function of ALDH2, increased accumulation of cellular aldehydes, and depletion of BRCA2 protein. MATERIALS AND METHODS: We genotyped 11,873 Polish women diagnosed with breast cancer and 7,615 ethnically matched controls for these two variants. Next, we extended our analysis of rs10744777 to 231 carriers of pathogenic BRCA2 mutations. RESULTS: BRCA2 p.K3326* variant was associated with significant increase in breast cancer risk only in those who were homozygous for the T allele of the ALDH2 rs10744777 variant (odds ratio = 1.72; 95% CI, 1.19 to 2.48; P = .003). The BRCA2 p.K3326* variant did not increase the risk of breast cancer among those who were heterozygous or homozygous for the C allele of the ALDH2 rs10744777 variant (odds ratio = 1.05; 95% CI, 0.73 to 1.51; P = .81). In the carriers of high-risk BRCA2 mutations, the TT genotype of rs10744777 conferred a modest (18%) and not significant increase in breast cancer risk. CONCLUSION: Our results suggest that BRCA2 p.K3326* variant, which is low-penetrance by itself, confers increased breast cancer risk on the background of the TT genotype of the ALDH2 rs10744777 variant in the Polish population.
Subject(s)
BRCA2 Protein , Breast Neoplasms , Aldehyde Dehydrogenase, Mitochondrial/genetics , Aldehydes , BRCA2 Protein/genetics , Breast Neoplasms/epidemiology , Female , Genetic Predisposition to Disease/genetics , Humans , Polymorphism, Single Nucleotide/geneticsABSTRACT
The current cancer testing gene panels tend to be comprehensive rather than site-specific. BARD1 is one of the genes commonly included in the multi-cancer testing panels. Mutations in BARD1 confer an increase in the risk for breast cancer, but it is not studied whether or not they predispose to prostate cancer. To establish if BARD1 mutations also predispose to prostate cancer, we screened BARD1 in 390 Polish patients with hereditary prostate cancer. No truncating mutations were identified by sequencing. We also genotyped 5715 men with unselected prostate cancer, and 10,252 controls for three recurrent BARD1 variants, including p.Q564X, p.R658C and p.R659=. Neither variant conferred elevated risk of prostate cancer (ORs between 0.84 and 1.15, p-values between 0.57 and 0.93) nor did they influence prostate cancer characteristics or survival. We conclude that men with a BARD1 mutation are not at elevated prostate cancer risk. It is not justified to inform men about increased prostate cancer risk in case of identification of a BARD1 mutation. However, a female relative of a man with a BARD1 mutation may benefit from this information and be tested for the mutation, because BARD1 is a breast cancer susceptibility gene.
ABSTRACT
BACKGROUND: The objective of this study was to establish the contribution of PALB2 mutations to prostate cancer risk and to estimate survival among PALB2 carriers. METHODS: We genotyped 5472 unselected men with prostate cancer and 8016 controls for two Polish founder variants of PALB2 (c.509_510delGA and c.172_175delTTGT). In patients with prostate cancer, the survival of carriers of a PALB2 mutation was compared to that of non-carriers. RESULTS: A PALB2 mutation was found in 0.29% of cases and 0.21% of controls (odds ratio (OR) = 1.38; 95% confidence interval (CI) 0.70-2.73; p = 0.45). PALB2 mutation carriers were more commonly diagnosed with aggressive cancers of high (8-10) Gleason score than non-carriers (64.3 vs 18.1%, p < 0.0001). The OR for high-grade prostate cancer was 8.05 (95% CI 3.57-18.15, p < 0.0001). After a median follow-up of 102 months, the age-adjusted hazard ratio for all-cause mortality associated with a PALB2 mutation was 2.52 (95% CI 1.40-4.54; p = 0.0023). The actuarial 5-year survival was 42% for PALB2 carriers and was 72% for non-carriers (p = 0.006). CONCLUSION: In Poland, PALB2 mutations predispose to an aggressive and lethal form of prostate cancer.
Subject(s)
Fanconi Anemia Complementation Group N Protein/genetics , Mutation , Prostatic Neoplasms/pathology , Sequence Analysis, DNA/methods , Adult , Aged , Aged, 80 and over , Case-Control Studies , Genetic Predisposition to Disease , Humans , Male , Middle Aged , Neoplasm Grading , Poland , Prostatic Neoplasms/genetics , Survival AnalysisABSTRACT
In designing national strategies for genetic testing, it is important to define the full spectrum of pathogenic mutations in prostate cancer (PCa) susceptibility genes. To investigate the frequency of mutations in PCa susceptibility genes in Polish familial PCa cases and to estimate gene-related PCa risks and probability of aggressive disease, we analyzed the coding regions of 14 genes by exome sequencing in 390 men with familial prostate cancer and 308 cancer-free controls. We compared the mutation frequencies between PCa cases and controls. We also compared clinical characteristics of prostate cancers between mutation carriers and noncarriers. Of the 390 PCa cases, 76 men (19.5%) carried a mutation in BRCA1, BRCA2, NBN, ATM, CHEK2, HOXB13, MSH2 or MSH6 genes. No mutations were found in BRIP1, PTEN, TP53, MLH1, PMS2 and SPOP. Significant associations with familial PCa risk were observed for CHEK2, NBN, ATM, and HOXB13. High-grade (Gleason 8-10) tumors were seen in 56% of BRCA2, NBN or ATM carriers, compared to 21% of patients who tested negative for mutations in these genes (OR = 4.7, 95% CI 2.0-10.7, P = .0003). In summary, approximately 20% of familial prostate cancer cases in Poland can be attributed to mutations in eight susceptibility genes. Carriers of mutations in BRCA2, NBN and ATM develop aggressive disease and may benefit from intensified screening and/or chemotherapy.
Subject(s)
Ataxia Telangiectasia Mutated Proteins/genetics , BRCA2 Protein/genetics , Cell Cycle Proteins/genetics , Mutation , Nuclear Proteins/genetics , Prostatic Neoplasms/pathology , Adult , Aged , Aged, 80 and over , Case-Control Studies , Genetic Association Studies , Genetic Predisposition to Disease , Humans , Male , Middle Aged , Neoplasm Grading , Pedigree , Poland , Prostatic Neoplasms/genetics , Exome SequencingABSTRACT
Bloom Syndrome is a rare recessive disease which includes a susceptibility to various cancers. It is caused by homozygous mutations of the BLM gene. To investigate whether heterozygous carriers of a BLM mutation are predisposed to breast cancer, we sequenced BLM in 617 patients from Polish families with a strong family history of breast cancer. We detected a founder mutation (c.1642C>T, p.Gln548Ter) in 3 of the 617 breast cancer patients (0.49%) who were sequenced. Then, we genotyped 14,804 unselected breast cancer cases and 4698 cancer-free women for the founder mutation. It was identified in 82 of 14,804 (0.55%) unselected cases and in 26 of 4698 (0.55%) controls (OR = 1.0; 95%CI 0.6-1.6). Clinical characteristics of breast cancers in the BLM mutation carriers and non-carriers were similar. Loss of the wild-type BLM allele was not detected in cancers from the BLM mutation carriers. No cancer type was more common in the relatives of mutation carriers compared to relatives of non-carriers. The BLM founder mutation p.Gln548Ter, which in a homozygous state is a cause of Bloom syndrome, does not appear to predispose to breast cancer in a heterozygous state. The finding casts doubt on the designation of BLM as an autosomal dominant breast cancer susceptibility gene.
ABSTRACT
BACKGROUND: NBN 657del5 founder mutation predisposes to breast and prostate cancer. Recently, it has been reported that the pathogenicity of this mutation with regard to prostate cancer risk is modified by a missense variant of the same gene (E185Q). METHODS: To evaluate the interaction of the 657del5 and E185Q founder alleles of NBN on breast cancer risk in Poland, 4964 women with breast cancer and 6152 controls were genotyped for these two recurrent variants of NBN (657del5 truncating variant and E185Q missense variant). RESULTS: The NBN 657del5 mutation was detected in 57 of 4964 unselected cases and in 35 of 6152 controls (OR = 2.0, p = 0.001). The E185Q GG genotype was detected in 2167 of 4964 unselected cases and in 2617 of 6152 controls (OR = 1.04, p = 0.3). In carriers of the 657del5 deletion, the elevated cancer risk was restricted to women with the GG genotype of the E185Q variant (OR = 3.6, 95% CI 1.9-6.6; p < 0.0001). Among women with other E185Q genotypes, the OR associated with 657del5 was 1.0 (95% CI 0.5-1.8; p = 0.9). The interaction between the two alleles was statistically significant (homogeneity p = 0.003). CONCLUSION: In Poland, the pathogenicity of the NBN 657del5 mutation is restricted to women with a homozygous GG genotype of missense variant of the same gene (E185Q). This is the first clear example whereby a moderate penetrance breast cancer gene is impacted by a genetic modifier.
Subject(s)
Alleles , Breast Neoplasms/genetics , Cell Cycle Proteins/genetics , Genetic Predisposition to Disease , Mutation , Nuclear Proteins/genetics , Adult , Aged , Aged, 80 and over , Amino Acid Substitution , Female , Genotype , Humans , Middle Aged , Odds Ratio , Risk Assessment , Risk Factors , Young AdultABSTRACT
BACKGROUND: XRCC2 participates in homologous recombination and in DNA repair. XRCC2 has been reported to be a breast cancer susceptibility gene and is now included in several breast cancer susceptibility gene panels. METHODS: We sequenced XRCC2 in 617 Polish women with familial breast cancer and found a founder mutation. We then genotyped 12,617 women with breast cancer and 4599 controls for the XRCC2 founder mutation. RESULTS: We identified a recurrent truncating mutation of XRCC2 (c.96delT, p.Phe32fs) in 3 of 617 patients with familial breast cancer who were sequenced. The c.96delT mutation was then detected in 29 of 12,617 unselected breast cancer cases (0.23%) compared to 11 of 4599 cancer-free women (0.24%) (OR = 0.96; 95% CI 0.48-1.93). The mutation frequency in 1988 women with familial breast cancer was 0.2% (OR = 0.84, 95% CI 0.27-2.65). Breast cancers in XRCC2 mutation carriers and non-carriers were similar with respect to age of diagnosis and clinical characteristics. Loss of the wild-type XRCC2 allele was observed only in one of the eight breast cancers from patients who carried the XRCC2 mutation. No cancer type was more common in first- or second-degree relatives of XRCC2 mutation carriers than in relatives of the non-carriers. CONCLUSION: XRCC2 c.96delT is a protein-truncating founder variant in Poland. There is no evidence that this mutation predisposes to breast cancer (and other cancers). It is premature to consider XRCC2 as a breast cancer-predisposing gene.
Subject(s)
Breast Neoplasms/genetics , DNA-Binding Proteins/genetics , Adult , Aged , Alleles , Breast Neoplasms/epidemiology , Breast Neoplasms/pathology , Female , Genetic Association Studies , Genetic Testing , Genotype , Humans , Middle Aged , Mutation , Mutation Rate , Poland/epidemiologyABSTRACT
To optimize genetic testing, it is necessary to establish the spectrum of breast cancer-predisposing mutations in particular ethnic groups. We studied 1,018 women with a strong family history for breast cancer (families with hereditary breast cancer; HBC) from genetically homogenous population of Poland, which is populated by ethnic Slavs, for mutations in 14 cancer susceptibility genes. Additionally, we compared the frequency of candidate pathogenic variants in breast cancer cases and controls. Germline mutations were detected in 512 of 1,018 probands with breast cancer (50.3%), including BRCA1/2 mutations detected in 420 families and non-BRCA mutations seen in 92 families. Thirteen BRCA1/2 founder mutations represented 84% of all BRCA1/2-positive cases. Seven founder mutations of CHEK2, PALB2, NBN and RECQL represented 73% of all non-BRCA-positive cases. Odds ratios for hereditary breast cancer were 87.6 for BRCA1, 15.4 for PALB2, 7.2 for CHEK2, 2.8 for NBN and 15.8 for RECQL. Odds ratios for XRCC2, BLM and BARD1 were below 1.3. In summary, we found that 20 founder mutations in six genes (BRCA1/2, CHEK2, PALB2, NBN and RECQL) are responsible for 82% of Polish hereditary breast cancer families. A simple test for these 20 mutations will facilitate genetic testing for breast cancer susceptibility in Poland. It may also facilitate genetic testing for breast cancer susceptibility in other Slavic populations and women of Slavic descent worldwide.
Subject(s)
Breast Neoplasms/genetics , Germ-Line Mutation/genetics , Adult , Aged , Female , Genetic Predisposition to Disease/genetics , Genetic Testing/methods , Humans , Middle Aged , PolandABSTRACT
PURPOSE: To establish the contribution of four founder alleles of NBN to prostate cancer risk and cancer survival. MATERIALS AND METHODS: Five thousand one hundred eighty-nine men with prostate cancer and 6,152 controls were genotyped for four recurrent variants of NBN (657del5, R215W, I171V, and E185Q). RESULTS: The NBN 657del5 mutation was detected in 74 of 5,189 unselected cases and in 35 of 6,152 controls (odds ratio [OR], 2.5; p < 0.001). In carriers of 657del5 deletion, the cancer risk was restricted to men with the GG genotype of the E185Q variant of the same gene. Among men with the GG genotype, the OR associated with 657del5 was 4.4 (95% confidence interval [CI], 2.4 to 8.0). Among men with other E185Q genotypes, the OR associated with 657del5 was 1.4 (95% CI, 0.8 to 2.4) and the interaction was significant (homogeneity p=0.006). After a median follow-up of 109 months, mortality was worse for 657del5 mutation carriers than for non-carriers (hazard ratio [HR], 1.6; p=0.001). The adverse effect of 657del5 on survival was only seen on the background of the GG genotype of E185Q (HR, 1.9; p=0.0004). CONCLUSION: The NBN 657del5 mutation predisposes to poor prognosis prostate cancer. The pathogenicity of this mutation, with regards to both prostate cancer risk and survival, is modified by a missense variant of the same gene (E185Q).