Familial prostate cancer and HOXB13 founder mutations: geographic and racial/ethnic variations
- First Online:
- Cite this article as:
- Lynch, H.T. & Shaw, T.G. Hum Genet (2013) 132: 1. doi:10.1007/s00439-012-1226-7
- 308 Views
The use of sequencing in cancer genome studies
Genome-wide association studies (GWAS) have recently made major contributions to the identification of mutations inclusive of G84E in HOXB13 for prostate cancer (Ewing et al. 2012) as featured in this editorial regarding the paper by Xu et al. (2012) published in this issue of the journal. However, along with the contributions of GWAS, there has been the rapid emergence of new technologies dealing with next-generation DNA sequencing (NGS) technology, which are providing unique opportunities to advance this knowledge. Specifically by sequencing the entire genome, entire exome or entire transcriptome, NGS allows an unbiased view to detect genetic defects in familial cancer aggregations as demonstrated by recent cancer genetic/genomic studies (Thompson et al. 2012; Hellebrand et al. 2011).
The exome sequencing method, first developed in 2009, uses the next-generation sequencers to sequence only the exome, that is, the coding part, of the targeted genome (Gnirke et al. 2009; Ng et al. 2009; Maher 2009; Choi et al. 2009). When compared to whole genome sequencing, the advantages of exome sequencing include its lower cost and simpler analysis of the sequences, while the mutations identified will provide functional information by focusing on the mutated gene. More importantly, as mutations in the coding region make up 85 % of the genetic disease-causing mutations (Cooper et al. 1995), identifying coding mutations should provide an opportunity with a high chance of determining the mutations contributing to a particular genetic disease.
The major technical challenge of exome sequencing was the standardization of the exome DNA extraction process to ensure inclusion of exon DNA templates for NGS sequencing. This problem has been solved with the development of commercial exome extraction kits by NimbleGen, Agilent, and Illumina companies in 2010–2011. Exome sequencing has become a matured method for genetic study, and provides a powerful means to detect genetic alterations affecting known genes in cancers (Gnirke et al. 2009; Ng et al. 2009; Maher 2009; Choi et al. 2009). This background is being discussed in this editorial with respect to the Xu et al. (2012) paper, given the potential that exome sequencing advances have for scientifically enhancing GWAS.
A recurrent, albeit rare, mutation (G84E) in HOXB13 was recently identified by Ewing et al. (2012) in a previously recognized region of linkage at 17q21-22 as harboring an increased risk for familial prostate cancer. Once confirmed and further clarified, this observation will have the potential to be clinically translatable.
Xu et al. (2012) have utilized a large international sample of prostate cancer-prone families who were recruited by the International Consortium for Prostate Cancer Genetics (ICPCG) in order to confirm the findings of Ewing et al. (2012) that the G84E mutation is rare, but that it is significantly associated with predisposition to prostate cancer. In the Xu et al. cohort, at least one mutation carrier was identified in each of 112 prostate cancer families (4.6 % of all 2,443 prostate cancer families studied), all of whom were of European descent. The G84E mutation was more frequently encountered in males diagnosed with prostate cancer (194 of 382, 51 %) than in unaffected male family members (42 of 137, 30 %) (P = 9.9 × 10−8) (odds ratio 4.42 [95 % confidence interval = 2.56–7.64]). Frequency of the mutation was higher in prostate cancer patients with early-onset disease (age at diagnosis ≤55 years old, 2.2 %) or with a positive family history (2.2 %) and most common in patients with both of these features (3.1 %). G84E was identified as being significantly over-transmitted from parents to affected offspring (P = 6.5 × 10−6).
The authors suggest a conservative interpretation of these findings with the consideration that if major prostate cancer susceptibility genes do exist, they most likely will be identified in regions generating suggestive or significant linkage signals in this sizable study.
History of HOXB13 G84E mutation
Gao et al. (1995) and Williams et al. (1996) found evidence for a tumor suppressor gene associated with prostate cancer in an area near the BRCA1 gene on chromosome 17q. In a study by Lange et al. (2003), the results from 175 prostate cancer-prone pedigrees provided suggestive evidence for linkage on chromosome 17q (LOD = 2.36). Not unexpectedly, the strongest evidence came from a subset of pedigrees with four or more affected individuals (LOD = 3.27). These authors concluded that their genome-wide analysis indicated new areas of the genome that may possibly harbor prostate cancer susceptibility genes and, more specifically, that there exists a prostate cancer susceptibility gene on chromosome 17 that is independent of ELAC2.
Gillanders et al. (2004) noted that the family history of prostate cancer is one of the most significant risk factors for this disease, but that genetic linkage analysis in attempts to localize prostate cancer susceptibility alleles had been limited by this cancer’s genetic heterogeneity, the lack of DNA samples from parents of individuals with late-onset prostate cancer, disease phenocopies, and its incomplete penetrance. Gillanders et al. (2004) pursued a combined genome-wide linkage analysis of 426 families who were members of 4 hereditary prostate cancer study populations in order to systematically investigate prostate cancer susceptibility genes. In the interest of decreasing the degree of locus heterogeneity, they analyzed subsets of families who harbored similar clinical and demographic characteristics and therein nonparametric multipoint linkage evolved as the primary method of analysis. Results showed strong evidence for prostate cancer linkage at chromosome region 17q22. Stratified analysis revealed several additional chromosomal regions that are likely to segregate prostate cancer susceptibility genes among specific subsets of hereditary prostate cancer families.
Lange et al. (2007) performed an extended linkage analysis including the study of 95 new multiplex prostate cancer-prone families, which showed further evidence in support for a chromosome 17q21-22 prostate cancer susceptibility gene.
The previously mentioned Ewing et al. (2012) findings disclosed a rare but recurrent mutation (G84E) in HOXB13 (rs1382131197) “…a homeobox transcription factor gene that is important in prostate development…” The mutation was significantly more common in men with early-onset, familial prostate cancer (3.1 %) than in those with late-onset, nonfamilial prostate cancer (0.6 %) (P = 2.0 × 10−6). The conclusion was that the HOXB13 G84E variant poses a statistically significant risk of hereditary prostate cancer, while accounting for only a small fraction of all prostate cancers.
Ewing et al. (2012) noted that fine mapping of the 17q21-22 region provided powerful evidence for linkage with an LOD close to marker D17S1820 of 5.49, coupled with a narrow candidate interval for a putative susceptibility gene (1-LOD support interval, approximately 10 cM) (Lange et al. 2007, cited by Ewing et al.).. Ewing et al. noted that “Next-generation sequencing technologies have provided new opportunities to interrogate large genomic intervals that are implicated in human disease in a rapid and comprehensive manner.”
The study by Xu et al. (2012) published in this issue capitalizes upon the results of the 2012 study by Ewing et al. (2012). In summary, germline mutations of the HOXB13 gene in 2,443 prostate cancer families from the ICPCG provided confirmation for the observation that G84E mutation is significantly associated with family histories of prostate cancer among individuals of European descent with the results remaining significant when families used in the original report were not included in the analysis.
In retrospect, it would have been of great value for Ewing et al. (2012) and Xu et al. (2012) to have pursued extra-prostate cancer cases, some of which may have been associated with G84E in HOXB13 and found to be part of hereditary prostate cancer syndromies. If identified, the cancer control benefit might have been immense in terms of screening for not only prostate cancer but, in addition, for other cancers that may be integral to that particular form of hereditary prostate cancer syndromy.
Prostate cancer appears to show extensive geographic, racial, and ethnic differences, both phenotypically and genotypically. For example, Xu et al. (2012) provide a high level of geographic and racial information. Their data show marked variation in prostate cancer frequency among the European, North American, and Australian families, including a geographic frequency variation of the G84E mutation throughout the European continent. Noteworthy was that the G84E mutation appears to be more common in Nordic countries, notably Finland, and less common in individuals from North America and Australia. Xu et al. note that the mutation was not found in families of any other race or ethnicity, including but not limited to African and Ashkenazi Jewish descent, while stating appropriately that larger numbers of families are needed to confirm the mutation’s population distribution. Markers flanking the G84E mutation show that it resides in the same haplotype in 95 % of carriers, consistent with a founder effect.
Another interesting follow-up to the Ewing et al. (2012) paper has been a study by Lin et al. (2012) that did not find the G84E mutation among 671 Chinese prostate cancer patients and 1,536 controls. However, a novel mutation in the HOXB13 gene, G135E, was identified in three of the cases but none of the controls, making it significantly more frequent among the cases (P = 0.027). Furthermore, it was found that the G135E mutation carriers shared a unique haplotype, pointing to a likely founder mutation. Contrast this with two other studies based on that of Ewing et al. that identified the G84E mutation in Canadian men of European descent (Akbari et al. 2012) and found it to be prevalent in more than 1 % of the Swedish population (Karlsson et al. 2012).
Further exploitation of these findings, as well as others that may give evidence of geographic and racial variation in HOXB13 mutations in the future, particularly from the clinical and pathology standpoint, could give new insights into the etiology of prostate cancer.