Prostate Cancer, Hereditary, 10

For a general discussion of hereditary prostate cancer, see 176807.

Amundadottir et al. (2006) reported a common variant on 8q24 associated with prostate cancer, discovered in a genomewide linkage scan in 871 Icelandic men with prostate cancer that grouped into 323 extended families. The linkage signal had a maximum lod score of 2.11 (D8S529 at 148.25 cM). To refine the location of the linkage signal, they performed an association study and found the strongest association to prostate cancer for allele -8 of the microsatellite DG8S737, with an odds ratio (OR) of 1.79. Of SNPs identified within the linkage disequilibrium block that contains DG8S737, allele A of rs1447295 showed the strongest association with prostate cancer (OR = 1.72, P = 1.7 x 10(-9)). The result was replicated in 3 case-control series of European ancestry in Sweden and the United States. Combining results from the 3 case-control groups of European ancestry gave an estimated OR of 1.62 (P = 2.7 x 10(-11)) for DG8S737 -8 and an OR of 1.51 (P = 1.0 x 10(-11)) for rs1447295. Amundadottir et al. (2006) found that about 19% of affected men and 13% of the general population carried at least 1 copy of allele -8, yielding a population attributable risk (PAR) of about 8%. The association was also replicated in an African American case-control group with a similar OR, in which 41% of affected individuals and 30% of the population were found to be carriers. This led to a greater estimated PAR (16%) for the African American group, which may contribute to the higher incidence of prostate cancer in African American men than in men of European ancestry.

Freedman et al. (2006) performed a whole-genome admixture scan in 1,597 African Americans and identified a 3.8-Mb interval on chromosome 8q24 as significantly associated with susceptibility to prostate cancer (lod = 7.1). The increased risk associated with African ancestry at 8q24 was greater in men diagnosed before 72 years of age (P less than 0.00032), and Freedman et al. (2006) suggested that this finding supports the epidemiologic observation that the higher risk for prostate cancer in African Americans is greater in younger men and attenuates with older age. The authors noted that the same region had been identified by Amundadottir et al. (2006) through linkage analysis of prostate cancer, followed by fine mapping. Freedman et al. (2006) strongly replicated this association (P less than 4.2 x 10(-9)) but found that the previously described alleles (DG8S737, rs1447295) did not explain more than a fraction of the admixture signal.

Following up on the study of Amundadottir et al. (2006), which identified the correlated variants allele A rs1447295 and allele -8 of microsatellite marker DG8S737 associated with prostate cancer, Gudmundsson et al. (2007) performed a genomewide association scan of 1,453 affected Icelandic individuals and 3,064 controls using microarray analysis followed by 4 replication studies. The rs1447295 gave the most significant results for association (OR = 1.71, P = 1.6 x 10(-14)). They constructed a 14-SNP haplotype that efficiently tagged a susceptibility variant, rs16901979, relatively uncommon (2 to 4%) in individuals of European descent but very common (approximately 42%) in African Americans. The rs16901979 variant showed a stronger association with affected individuals who had an earlier age at diagnosis. Gudmundsson et al. (2007) estimated the PAR of the rs16901979 A allele alone to be 24% in African Americans and approximately 13% in populations of European ancestry; with updated estimates for the effect of DG8S737 -8 in African Americans, the 2 variants had a combined estimated PAR of 31% in that population. The authors suggested that these 2 variants could account for a large fraction of the excess of prostate cancer rates in African Americans relative to European Americans. The genomic region containing rs16901979 did not contain any known genes or microRNAs.

Following up on the admixture scan of Freedman et al. (2006) in African Americans with prostate cancer, Haiman et al. (2007) genotyped 2,973 SNPs in up to 7,518 men with or without prostate cancer from African American, Japanese American, Native Hawaiian, Latino, and European American populations. They identified 7 risk variants, 5 theretofore undescribed, spanning 430 kb and each independently predicting risk for prostate cancer. The strongest association was for rs6983561 (P = 7.9 x 10(-19)). The variants defined common genotypes that span more than a 5-fold range of susceptibility to prostate cancer in some populations. None of the variants aligned to a known gene or altered the protein sequence of an encoded protein.

In a genomewide association analysis using 550,000 SNPs in a nested case-control study (1,172 cases and 1,157 controls of European origin), Yeager et al. (2007) identified a new association at 8q24 with an independent effect on prostate cancer susceptibility. The most significant signal, rs6983267, was 70 kb centromeric to the previously reported rs1447295, but showed little evidence of linkage disequilibrium with it. A combined analysis with 4 additional studies (total: 4,296 cases and 4,299 controls) confirmed association with prostate cancer for rs6983267 in the centromeric locus (P = 9.42 x 10(-13); heterozygote OR: 1.26, 95% confidence interval: 1.13-1.41; homozygote OR: 1.58, 95% confidence interval: 1.40-1.78). Each SNP remained significant in a joint analysis after adjusting for the other (rs1447295, P = 1.41 x 10(-11); rs6983267, P = 6.62 x 10(-10)). These observations, combined with compelling evidence for a recombination hotspot between the 2 markers, indicated the presence of at least 2 independent loci within 8q24 that contribute to prostate cancer in men of European ancestry. Yeager et al. (2007) estimated that the PAR of the new locus, marked by rs6983267, is higher than the locus marked by rs1447295 (21% versus 9%). They noted that rs6983267 has an overall population frequency in northern Europeans of 50% for the at-risk G allele.

Witte (2007) reviewed the studies of Amundadottir et al. (2006), Freedman et al. (2006), Gudmundsson et al. (2007), Haiman et al. (2007), and Yeager et al. (2007). Combining results for rs1447295, which resides in a region designated 'region 1,' across the studies of Gudmundsson et al. (2007), Haiman et al. (2007), and Yeager et al. (2007) indicated an extraordinarily strong association, with an adjusted P value of 4 x 10(-29). However, Witte (2007) noted that, as in the studies of Amundadottir et al. (2006) and Freedman et al. (2006), relatively weak associations were observed in this region among African Americans. Merging the findings for rs16901979, within region 2 located approximately 350 kb upstream of region 1, gave an adjusted P value of 1 x 10(-19). Combining results for rs6983267, within region 3 between regions 1 and 2, from the 3 later studies gave an adjusted P value of 1 x 10(-11). Witte (2007) concluded that SNPs across all 3 neighboring regions seem to contribute independently to the 8q24 signal, and the combined effects of SNPs across regions closely follow a multiplicative model.

In a large genomewide association study to search for common variants with moderate risk, Thomas et al. (2008) confirmed the association of 2 independent SNPs at 8q24 with prostate cancer, 4242382 (P = 1.12 x 10(-4)) and rs6983267 (P = 3.92 x 10(-4)). Eeles et al. (2008) likewise confirmed the association, identifying 20 SNPs on 8q24 with P less than 10(-6).

Yeager et al. (2008) used next-generation sequencing technology to conduct a resequence analysis of a 136-kb region on chromosome 8q24 in 39 cases of advanced prostate cancer and 40 controls of European origin. The study yielded a comprehensive catalog of common SNPs within this region, including 442 novel SNPs, as well as the pattern of linkage disequilibrium across the region. Yeager et al. (2008) suggested that their results would be useful in choosing SNPs for fine mapping of association signals in 8q24 in prostate and colorectal cancer and for investigations of the functional consequences of select common variants.

Several groups found association of the prostate cancer risk-associated SNP rs6983267 with susceptibility to colorectal cancer; see CRCS2, 611469.

Yeager et al. (2009) reported a genomewide association study in 10,286 cases and 9,135 controls of European ancestry in the Cancer Genetic Markers of Susceptibility (CGEMS) initiative. They identified a new association with prostate cancer risk on chromosome 8q24 (rs620861, P = 1.3 x 10(-10), heterozygote OR, 1.17, 95% CI 1.10-1.24; homozygote OR, 1.33, 95% CI 1.21-1.45). The risk is associated with the C allele of this SNP.

Because previous studies had identified multiple loci on 8q24 associated with prostate cancer risk, Al Olama et al. (2009) performed a comprehensive analysis of SNP associations across 8q24 by genotyping tag SNPs in 5,504 prostate cancer cases and 5,834 controls. They confirmed associations at 3 previously reported loci and identified additional loci in 2 other linkage disequilibrium blocks (rs1006908: per-allele OR = 0.87, P = 7.9 x 10(-8); rs620861: OR = 0.90, P = 4.8 x 10(-8)). Eight SNPs in 5 linkage disequilibrium blocks were independently associated with prostate cancer susceptibility. The 2 highest were rs6983267 with a per allele OR of 1.26, 95% CI of 1.18-1.35, and P value of 6.2 x 10(-23); and rs10090154, with a per-allele OR of 1.47 and P value of 6.8 x 10(-24).

Gudmundsson et al. (2009) reported a prostate cancer genomewide association follow-up study. They identified association with 2 SNPs on chromosome 8q24.21, 16902094G (OR 1.21, P = 6.2 x 10(-15)) and rs445114T (OR = 1.14, P = 4.7 x 10(-10)). In a multivariate analysis using 22 prostate cancer risk variants typed in the Icelandic population, Gudmundsson et al. (2009) estimated that carriers in the top 1.3% of the risk distribution are at 2.5 times greater risk of developing the disease than members of the general population.

Tuupanen et al. (2009) and Sotelo et al. (2010) independently determined that rs6983267 lies within a functional TCF4 (TCF7L2; 602228)-binding enhancer element of the MYC gene (190080). Tuupanen et al. (2009) were unable to show a definite correlation between rs6983267 genotype and MYC expression. However, Sotelo et al. (2010) found that the T allele of rs6983267 consistently stimulated activity of a MYC reporter to a greater extent than the G allele in both the presence and absence of beta-catenin (CTNNB1; 116806)/TCF4. The effect of rs6983267 was not large, but it was highly reproducible, with p less than 0.0022.

Gudmundsson et al. (2012) analyzed 32.5 million variants discovered by whole-genome sequencing 1,795 Icelanders, and identified a new low-frequency variant at 8q24 associated with prostate cancer in European populations, rs188140481-A (odds ratio = 2.90; p(combined) = 6.2 x 10(-34)), with an average risk allele frequency in controls of 0.54%. This variant is only very weakly correlated (r(2) less than or equal to 0.06) with previously reported risk variants at 8q24, and its association remained significant after adjustment for all risk-associated variants known to that time. Carriers of rs188140481-A were diagnosed with prostate cancer 1.26 years younger than noncarriers (p = 0.0059).

Chung et al. (2011) performed resequencing and fine mapping of a region in chromosome 8q24 (region 2) that had been shown to have the strongest association with prostate cancer susceptibility in Japanese Americans and African Americans (e.g., Haiman et al., 2007). They mapped the association peak within a region of approximately 13 kb (chr8:128.16-128.175 Mb, NCBI36) and demonstrated that this region was transcribed, yielding a long noncoding RNA (PRNCR1; 615452) whose expression was upregulated in prostate cancer and its precursor lesions. Chung et al. (2011) identified a 4-SNP haplotype (ATTT) involving rs1456315, SNP34 (chr8:128,173,151, NCBI36), rs5013678, and rs7463708 that showed significant association with prostate cancer in the Japanese population (p = 2.00 x 10(-24), OR = 1.74, 95% confidence interval = 1.56-1.93). They hypothesized that the 8q24 region 2 prostate cancer susceptibility locus is likely to be located within a linkage disequilibrium block encompassing these SNPs (chr8:128,173,119-128,173,237, NCBI36).

Breyer et al. (2014) found that 8q24 genomewide association study (GWAS) SNPs are cis-acting expression quantitative trait loci (eQTLs) that modify POU5F1B (615739) expression in normal prostate. In normal prostate tissues, risk alleles for prostate cancer susceptibility by GWAS were consistently accompanied by reduced POU5F1B expression, whereas the protective alleles are accompanied by increased POU5F1B expression. Breyer et al. (2014) found that POU5F1B and POU5F1 (164177) are coexpressed in normal prostate. The risk allele rs6983267 was significantly associated with reduced POU5F1B (p = 0.001) and POU5F1 (p = 0.005) expression in normal prostate of white subjects and was trending for all subjects combined. POU5F1B and POU5F1 were coexpressed in prostate cancer, but POU5F1 expression in prostate cancer accompanied indices of poor prognoses. Breyer et al. (2014) found that the risk allele of rs6983267 marked several common haplotypes encompassing the adjacent gene, POU5F1B. These haplotypes did not have consistent risk effects either within or across the tested populations. In contrast, the protective allele of rs6983267 marked a haplotype that was consistently inversely associated with prostate cancer risk among United States study subjects of European descent, United States study subjects of African descent, and Finnish study subjects. Although this protective haplotype was the most common one in the subjects of European descent (42% frequency in control subjects), it was notably less common among those of African descent (17% frequency in control subjects). Predicted deleterious alleles of 2 POU5F1B missense SNPs, rs6998061 (G176E) and rs7002225 (E238Q), mark that haplotype. In the POU5F1B region, the effect of the GWAS SNP rs6983267 is correlated with the effects of these 2 missense SNPs, the latter more directly detecting risk effects among United States subjects of European descent.