Abstract
BACKGROUND
Prostate cancer (PC) is the most frequently diagnosed non-skin malignancy in men in the Western world, yet few disease-associated mutations have been found. Recently, a low frequency recurring mutation in the HOXB13 gene was reported among both hereditary PC families and men from the general population.
MATERIALS and METHODS
We determined the distribution and frequency of the G84E HOXB13 variant in 1,310 incipient PC cases and 1,259 age-mated controls from a population-based, case control study of PC.
RESULTS
The G84E mutation was more frequent in cases than controls (1.3% versus 0.4%, respectively), and men with the HOXB13 G84E variant had a 3.3-fold higher relative risk of PC compared with noncarriers (95% CI, 1.21–8.96). There was a stronger association between the G84E variant and PC among men with no first-degree relative with PC (OR, 4.04; 95% CI, 1.12–14.51) compared to men with a family history of PC (OR, 1.49; 95% CI, 0.30–7.50; p=0.36 for interaction). We observed some evidence of higher risk estimates associated with the variant for men with higher versus lower Gleason score (OR, 4.13; 95% CI, 1.38–12.38 vs OR, 2.71; 95% CI, 0.88–8.30), and advanced versus local stage (OR, 4.47; 95% CI, 1.28–15.57 vs OR, 2.98; 95% CI, 1.04–8.49), however these differences were not statistically different.
CONCLUSIONS
These results confirm the association of a rare HOXB13 mutation with PC in the general population and suggest that this variant may be associated with features of more aggressive disease.
Keywords: polymorphism, prostate neoplasm, genetic susceptibility
Introduction
Prostate cancer (PC) is the most frequently diagnosed non skin malignancy in men in the Western world, with more than 241,740 new cases and approximately 28,170 PC-specific deaths estimated for 2012 (1). Among the most significant risk factors for PC is a positive family history. Men with a first-degree family history of PC have a two- to three-fold increase in risk relative to men without such a history (2). This is supported by twin studies, which estimate that inherited genetic factors account for approximately 42% of PC incidence (3). Family-based genome-wide linkage studies have identified over 20 loci that segregate with the disease in hereditary PC (HPC) families; however, only a few of these loci have been replicated in multiple independent sets of HPC families (4–6).
One locus, which has been replicated in multiple data sets is at 17q21-22, located at approximately 57–60 cM (7). Originally reported on the basis of a linkage study using 175 HPC pedigrees (8), initial linkage reports in this region showed only marginal evidence for linkage (LOD = 2.36), reflecting the modest size of the families (average of 2.6 affected men per family with DNA), locus heterogeneity associated with the disease, and presence of phenocopies. The strongest evidence was provided by pedigrees with four or more affected individuals (LOD = 3.17) (8). Interest in the region increased with replication of the original finding (9,10), and narrowing of the region to a 1-LOD support interval of about 10cM (11).
In 2012, the homeobox transcription factor HOXB13 was proposed as a gene for PC risk from the 17q21-22 region (12). This was based on analysis of more than 200 genes. Specifically, in an analysis of HPC families, investigators reported that 18 affected men in four families carried a rare but recurring mutation in the second position in codon 84, resulting in a non-conservative replacement of a glutamic acid for a glycine (G84E). These findings were validated in an assessment of unrelated affected men of European descent, in which the mutation rate was 1.4%, compared to 0.1% found in controls. Among men with early onset familial PC, the carrier rate was 3.1%, compared to those with late-onset non-familial PC (0.6%; p=2.0×10−6). Replication studies have focused on clinical features of disease (13), studies of familial disease (14), particular ethnic groups (15,16), or hospital- based series (17).
None of the above analyses address the mutation frequency among men from the general population in the United States. Nor do they allow for stratification based on long-term follow up. Toward that end, we have investigated the frequency and distribution of HOXB13 mutations among 1,457 cases and 1,342 controls from two Western Washington population-based, case- control studies of PC.
Methods
Study Population
We used data and biological specimens from two previous population-based case-control studies. As previously described (18,19), study subjects were Caucasian and African American residents of King County, Washington. Cases from the first study were 40–64 year old men diagnosed with histologically confirmed PC between 1993 and 1996. Cases from the second study were 35–74 year old men diagnosed with histologically confirmed PC between 2002 and 2005. For both studies, cases were identified from the metropolitan Seattle-Puget Sound population based tumor registry that is operated as part of the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program.
Controls were recruited by random-digit telephone dialing (20). In both studies, household census information was obtained, and men within the study age range and without a self-reported physician’s diagnosis of PC and who agreed to receive information, were contacted. Household census information was obtained for 94% and 81% of residential telephone numbers contacted in the first and second studies respectively. For both studies, controls were recruited evenly throughout the ascertainment period for cases, and were frequency matched to cases by 5-year age groups in both studies. Of eligible controls identified, 75% and 63% were interviewed in the first and second studies respectively.
Of 753 eligible interviewed cases in the first study, 631 (84%) provided blood samples yielding sufficient DNA for genotyping, and of 703 eligible interviewed controls, 565 (80%) provided blood samples. Of 1,001 eligible interviewed cases in the second study, we collected 827 (83%) blood samples for genotyping, and of 942 eligible interviewed controls, 787 (84%) provided blood samples. Out of a total of 230 genotyped African American men, there was only one case and no controls with the HOXB13 variant. As such, we were unable to stratify by race and we therefore confined all further analyses to Caucasian men. We conducted analyses for the present study on a total of 1,310 cases and 1,259 controls with sufficient DNA for this specific study and who were successfully genotyped for the HOXB13 variant, all Caucasian.
As previously described (18,19), we interviewed study participants using a structured questionnaire eliciting demographic characteristics, medical and lifestyle history. The Seattle–Puget Sound SEER cancer registry provided information on stage and Gleason score of PC at diagnosis for cases. Vital status and underlying cause of death were obtained through the cancer registry, which links quarterly to the Washington State Center for Health Statistics database, and death certificates were requested from the state to confirm cause of death. The Fred Hutchinson Cancer Research Center’s Institutional Review Board approved the studies and the Internal Review Board of the National Human Genome Research Institute approved genotyping. Written informed consent was obtained from all study participants.
Single Nucleotide Polymorphism Genotyping
A custom designed TaqMan SNP genotyping assay (Applied Biosystems, Foster City, CA) was used to genotype the HOXB13 G84E (rs138213197) SNP on the ABIPrism 7900HT sequence detection system according to the manufacturer’s instructions. The 40 blind duplicates showed 100% concordance.
Statistical Analysis
Genotype frequencies for HOXB13 (rs138213197) were consistent (P > 0.05) with Hardy-Weinberg equilibrium (HWE) among Caucasian controls. We estimated the odds ratio (OR) and 95% confidence intervals (CI) for the association between the HOXB13 G84E variant and PC by fitting unconditional logistic regression models, adjusted for age.
To investigate whether an association between the HOXB13 G84E mutation was modified by family history of PC, we stratified subjects by first-degree relative with PC. To statistically assess departures from multiplicative effects we included a product term in stratified analyses, and used a log-likelihood ratio test to compare logistic models with and without the product term.
We used polytomous logistic regression to explore differences by stage (local versus regional/distant) and Gleason score (2–7, including 3+4 versus 7–10, including 4+3). We grouped Gleason score 7 (4+3) tumors with Gleason scores 8–10 because outcomes such as PC-specific mortality for 4+3 tumors have been shown to be similar to 8–10 tumors (21). We additionally characterized tumors by a composite measure of aggressiveness and conducted polytomous logistic regression. Tumors were classified as “less aggressive” if they met the following criteria: local stage, Gleason score 2–6 or 7 (3 + 4), and diagnostic serum PSA <20 ng/mL. “More aggressive” cancer was defined by regional/distant stage, Gleason score 7 (4 + 3) or 8–10, or serum PSA ≥20 ng/mL.
To analyze risk of death from PC associated with the HOXB13 G84E variant, we calculated PC-specific hazard ratios (HR) and 95% CIs by Cox proportional hazards regression, adjusting for age. Time to PC-specific mortality was defined as the time from diagnosis to death. The censoring date was the date of last vital status update from the cancer registry (July 1, 2012).
In addition, we calculated the adjusted HR associated with the variant for risk of recurrence/progression. Time to recurrence/progression was defined as time from diagnosis to first reported evidence of recurrence, as previously described (22). The censoring date was the date that the follow-up questionnaire was returned. For those who died of PC prior to administration of the follow-up survey, time to recurrence was imputed (22). We used Stata statistical software (version 11.0, Stata Corp.) for all analyses.
Results
Compared with controls, cases were more likely to report a first-degree family history of PC (Table 1). Cases were more likely to have had a Prostate-Specific Antigen (PSA) test in the last five years before reference date (date of diagnosis for cases and a randomly assigned date for controls that approximated the distribution of diagnosis dates for cases). The majority of cases had localized stage disease (78%), moderate Gleason score (79%), and less aggressive tumors (66%).
Table I.
Distributions for selected characteristics among Caucasian prostate cancer cases and controls, King County, Washington, 1993–1996 and 2002–2005
| Characteristic | Cases (n=1,310) | Controls (n=1,259) | ||
|---|---|---|---|---|
| N | % | n | % | |
| Age (years) | ||||
| < 50 | 102 | 7.8 | 105 | 8.3 |
| 50–59 | 522 | 39.9 | 519 | 41.2 |
| 60–69 | 542 | 41.4 | 491 | 39.0 |
| 70–74 | 144 | 11.0 | 144 | 11.4 |
| Family history of prostate cancer | ||||
| No | 1,027 | 78.4 | 1,118 | 88.8 |
| Yes | 283 | 21.6 | 141 | 11.2 |
| Number of PSA testsa | ||||
| None | 290 | 22.1 | 445 | 35.4 |
| 1–2 | 321 | 24.5 | 239 | 19.0 |
| 3–4 | 244 | 18.6 | 153 | 12.2 |
| ≥ 5 | 392 | 29.9 | 225 | 17.9 |
| Missing | 63 | 4.8 | 197 | 15.7 |
| PSA | ||||
| 0 – 3.9 | 179 | 13.7 | 1,169 | 92.9 |
| 4 – 9.9 | 721 | 55.0 | 74 | 5.9 |
| 10 – 19.9 | 189 | 14.4 | 14 | 1.1 |
| 20+ | 118 | 9.0 | 2 | 0.2 |
| Missing | 103 | 7.9 | ||
| Stage of disease | ||||
| Local | 1,020 | 77.9 | ||
| Regional | 257 | 19.6 | ||
| Distant | 33 | 2.5 | ||
| Gleason score | ||||
| Low (2–4) | 70 | 5.3 | ||
| Moderate [5–7 (3+4)] | 1,034 | 78.9 | ||
| High [7 (4+3), 8–10] | 202 | 15.4 | ||
| Missing | 4 | 0.3 | ||
| Cancer aggressiveness | ||||
| Less aggressive b | 870 | 66.4 | ||
| More aggressive c | 440 | 33.6 | ||
PSA tests done in the 5 years before reference date.
Local stage, and Gleason score 2–6 or 7 (3 + 4), and diagnostic serum PSA <20 ng/mL.
Regional/distant stage, or Gleason score 7 (4 + 3) or 8–10, or serum PSA ≥20 ng/mL.
Overall in our study, the HOXB13 G84E mutation was identified in 18 out of 1,457 cases and 5 out of 1,342 controls. Due to the limited numbers of African Americans with the HOXB13 G84E variant (one case and no controls), we were unable to perform meaningful stratified analyses and therefore conducted all further analyses with the 1,310 Caucasian cases and 1,259 Caucasian controls. The HOXB13 G84E mutation was more frequent in Caucasian cases than controls (1.3% in cases versus 0.4% in controls) and had a 3.3-fold higher risk of PC compared with noncarriers of G84E, adjusted for age (95% CI, 1.21–8.96; Table 2). There appeared to be a greater association between the HOXB13 G84E variant and PC among men with no first-degree relative with PC (OR, 4.04; 95% CI, 1.12–14.51; Table 3) compared with men with a family history of PC (OR, 1.49; 95% CI, 0.30–7.50); however, this interaction was not significant (likelihood ratio p=0.36).
Table II.
Genotype distribution and odds ratios (ORs; 95% CI) for associations between HOXB13 genotypes and prostate cancer risk
| Genotype | Cases (n=1,310) | Controls (n=1,259) | ORa | 95% CI | ||
|---|---|---|---|---|---|---|
| n | % | n | % | |||
| GG | 1,293 | (98.7) | 1,254 | (99.6) | 1.00 | (reference) |
| GA | 17 | (1.3) | 5 | (0.4) | 3.30 | (1.21–8.96) |
Adjusted for age.
Table III.
Genotype distribution and odds ratios (ORs; 95% CI) for associations between HOXB13 genotypes and prostate cancer risk, stratified by family history of prostate cancer
| Genotype | Cases (n=1,310) | Controls (n=1,259) | ORa,b | 95% CI | ||
|---|---|---|---|---|---|---|
| n | % | n | % | |||
| First-degree relative with prostate cancer | ||||||
| GG | 277 | (97.9) | 139 | (98.6) | 1.00 | (reference) |
| GA | 6 | (2.1) | 2 | (1.4) | 1.49 | (0.30–7.50) |
| No first-degree relative with prostate cancer | ||||||
| GG | 1,016 | (98.9) | 1,115 | (99.7) | 1.00 | (reference) |
| GA | 11 | (1.1) | 3 | (0.3) | 4.04 | (1.12–14.51) |
Adjusted for age.
Likelihood ratio test for interaction by family history: p=0.36.
We did not observe meaningful heterogeneity by stage (p=0.45), Gleason score (p=0.39), or cancer aggressiveness (p=0.50). However, the extremely small number of men with the HOXB13 G84E variant hindered interpretation of stratified analyses. We observed some indication of greater risk of PC for men with higher Gleason score (OR, 4.13; 95% CI, 1.38–12.38; Table 4) compared with lower Gleason score (OR, 2.71; 95% CI, 0.88–8.30). Similarly, associations appeared to be somewhat stronger for regional/distant disease (OR, 4.47; 95% CI, 1.28–15.57; Table 4) versus local disease (OR, 2.98; 95% CI, 1.04–8.49), and more aggressive PC (OR, 4.06; 95% CI, 1.28–12.86; Table 5) versus less aggressive cancer (OR, 2.91; 95% CI, 0.99–8.55).
Table IV.
Genotype distribution and ORs (95% CI) for associations between HOXB13 genotypes and prostate cancer risk stratified by Gleason score and stage
| Genotype | Cases (n=1,310) | Controls (n=1,259) | ORa,b,c | 95% CI | ||
|---|---|---|---|---|---|---|
| n | % | n | % | |||
| Gleason score ≤ 7d (n=750) | ||||||
| GG | 742 | (98.9) | 1,254 | (99.6) | 1.00 | (reference) |
| GA | 8 | (1.1) | 5 | (0.4) | 2.71 | (0.88–8.30) |
| Gleason score ≥ 7e (n=556) | ||||||
| GG | 547 | (98.4) | 1,254 | (99.6) | 1.00 | (reference) |
| GA | 9 | (1.6) | 5 | (0.4) | 4.13 | (1.38–12.38) |
| Local stage (n=1,020) | ||||||
| GG | 1,008 | (98.8) | 1,254 | (99.6) | 1.00 | (reference) |
| GA | 12 | (1.2) | 5 | (0.4) | 2.98 | (1.04–8.49) |
| Regional/distant stage (n=290) | ||||||
| GG | 285 | (98.3) | 1,254 | (99.6) | 1.00 | (reference) |
| GA | 5 | (1.7) | 5 | (0.4) | 4.47 | (1.28–15.57) |
Adjusted for age.
Heterogeneity by Gleason score: p=0.39.
Heterogeneity by stage: p=0.45.
Includes Gleason 3+4.
Includes Gleason 4+3.
Table V.
Genotype distribution and ORs (95% CI) for associations between HOXB13 genotypes and prostate cancer risk stratified by aggressiveness
| Genotype | Cases (n=1,310) | Controls (n=1,259) | ORa,b | 95% CI | ||
|---|---|---|---|---|---|---|
| n | % | n | % | |||
| Less aggressive (n=870) | ||||||
| GG | 860 | (98.8) | 1,254 | (99.6) | 1.00 | (reference) |
| GA | 10 | (1.2) | 5 | (0.4) | 2.91 | (0.99–8.55) |
| More aggressive (n=440) | ||||||
| GG | 433 | (98.4) | 1,254 | (99.6) | 1.00 | (reference) |
| GA | 7 | (1.6) | 5 | (0.4) | 4.06 | (1.28–12.86) |
Adjusted for age.
Heterogeneity by aggressiveness: p=0.50.
There were 82 cases who died of PC in an average 11.6 years of follow-up. We did not observe increased risk of prostate-specific mortality associated with the HOXB13 mutation (HR, 0.94; 95% CI, 0.13–7.01; results not shown). There were 290 events of recurrence/progression in an average 8.5 years of follow-up. We did not find statistically significant evidence of increased risk for recurrence/progression (HR, 1.26; 95% CI, 0.46–3.43; results not shown). However, these analyses were limited as only one man with the mutation died of PC and only four men experienced recurrence/progression.
Discussion
Our data revealed that 1.3% of cases and 0.4% of controls from this population-based, case control study of PC carried the HOXB13 G84E mutation, conferring a risk to carriers of 3.30 (95% CI 1.21–8.96). These numbers are in the range of what has been previously reported. Recall that other studies are focused on either family-based cohorts, hospital-based series and isolated populations. The original report cited a population of 5,083 unrelated subjects of self-reported European descent with PC in which 1.4% were carriers, compared to one in 1,401 (0.1%) controls (12). The finding of unaffected controls with the recurring mutation in our data set is not unexpected. Ewing et al. reported the existence of one HPC family in which a 70-year-old G84E carrier had not been diagnosed with PC. This may reflect the fact that the mutation is not completely penetrant, or that there is age-dependent penetrance.
Akbari et al. studying a set of cases and controls, aged 40–94 from two tertiary care centers in Canada, reported that 10 of 1,525 cases (0.7%) and two of 1,757 controls (0.1%) carried the G84E mutation (17). In a case-control study of familial PC at Vanderbilt University, Breyer and colleagues found that 16 of 858 familial PC probands (1.9%) and two of 825 (0.2%) controls without a personal or family history of PC carried the G84E mutation (14). In a separate series of 268 probands of European descent, the carrier rate was 1.5% (14).
These studies demonstrate that the mutation is rare in both the general and familial PC populations. But it does confer a significant elevation in risk. In the Akbari study, carriers had an increased risk of 5.8 (95% CI 1.3–26.5, P = .01) (17), and in the Breyer study it was 7.9 (95%CI, 1.8–34.5, P = 0.0062) (14). In the original study of Ewing et al., analysis of 5,083 unrelated case subjects and 1,401 controls revealed an OR of 20.1 (95% CI, = 3.5–803.3, P = 8.5×10−7)(12). All of the studies found only heterozygous carriers, which is not unexpected given the rarity of the mutation.
It will be interesting to assess the frequency of the mutation in various populations. The original study included seven African American families and two Asian families, but these were not reported separately. In two population-based case control studies in Sweden, Karlsson and colleagues analyzed 15 HOXB13 polymorphisms in 5,003 cases and 4,693 controls (16). They found that the HOXB13 G84E mutation is more frequent in the Swedish population (1.3% in population controls), and is associated with a 3.5-fold increased risk of PC. The strongest association was reported for younger-onset cases (OR, 8.6; 95% CI, 5.1–14.0) and for familial PC cases (OR, 6.6; 95% CI, 3.3–12.0) (16).
Recently, Lin et al. assayed the gene in a set of study subjects from the Chinese Consortium for Prostate Cancer Genetics (Chinese PCa) (15). They screened an initial set of 96 individuals and found putative disease associated mutations. These, including the G84E mutation, were then assessed in a stage two study that included 671 PC patients and 1,536 controls. Data were collected on an additional 751 cases by imputing mutations from haplotypes. Interestingly, the G84E mutation was not found at all in that study, but a novel mutation, G135E was found once among the 96 patients, and in two of the 575 additional patients in stage two, yielding a P value of 0.027. It was also found in two of the patients from stage three for whom the data were imputed, but not validated. It was not observed within any of the 1,536 controls. These data suggest that other recurring mutations may be found by examination of distinct populations.
One question of intense interest is how HOXB13 protein functions in the prostate and how the mutations might perturb the protein to affect disease risk. The HOX genes are a subfamily of the larger homeobox superfamily that is responsible for pattern formation across many species. In humans, there are four HOX gene clusters that are expressed at precise times during development with a set regulating the anterior and proximal regions and another set, expressed later in development, that regulates the posterior and distal regions of the animal (for review: (23,24)). The HOX genes in the 13 group are expressed in the posterior region of developing vertebrates, including the urogenital system. Interestingly, HOXB13 is one of few genes in the family to maintain a high level of expression during the life of the individual, with high expression levels observed in the normal prostate well into adulthood. HOXB13 is known to play a role in prostate cancer. Specifically, in androgen receptor negative prostate cancer cells, HOXB13 is suppressed and overexpression of HOXB13 results in significant inhibition of cell growth. In addition, HOXB13 has been shown to suppress androgen-stimulated androgen receptor activity by interacting with the receptor (25). This means that HOXB13 could serve as an androgen receptor interacting repressor to modulate hormone-activated androgen receptor signals (25).
The gene, which is itself 70Kb upstream of the major HOXB gene cluster, features two MEIS (myeloid ectropic viral integration site) protein-binding domains. The two MEIS are highly conserved among vertebrates (26,27) and their role is to fine tune HOX functions by regulating the interaction of HOX proteins with specific DNA molecules and proteins (12,15,28). Interesting the two recurring mutations found to date are both within MEIS binding domains. G84E is within the first domain, while G135E is within the second (15).
While there is no evidence as to how the mutations affect protein function, HOXB13 has been considered as a marker of recurrent PC by at least one group of investigators (13). In their study, immunohistochemistry was used to examine 57 organ-confined PC tumors collected after radical prostatectomy. While there was no significant correlation between HOXB13 expression and most clinical or pathological measures, gene expression did correlate with Gleason score and there was a positive correlation with pre-operative prostate specific antigen (PSA) levels, suggesting avenues for further study and validation. Although we examined the association of the HOXB13 variant with risk of recurrence in the present study (HR 1.26; 95% CI, 0.46–3.43), there were only four instances of recurrence among men with the variant, limiting interpretation of the results.
While not significantly different, we did observe a stronger association with Gleason score ≥7 (OR 4.13, 95% CI 1.38–12.38) compared with Gleason score ≤ 7 (OR, 2.71; 95% CI, 0.88–8.302), with regional/distant stage disease (OR, 4.47; 95% CI, 1.28–15.57) versus local disease (OR, 2.98; 95% CI, 1.04–8.49), and with more aggressive (OR, 4.06; 95% CI, 1.28–12.86) versus less aggressive PC (OR, 2.91; 95% CI, 0.99–8.55). These risk estimates, however, are based on small numbers and further investigation of these clinical features of PC in relation to HOXB13 mutation status is needed. While Breyer et al also noted this trend (14), Ewing and colleagues found no evidence of significant differences in Gleason grade between carriers and non-carriers, even accounting for age at diagnosis (12), although they did not present data that would allow for assessment of trend. Similarly, this information was not presented by Karlsson et al., (16).
The small number of mutation carriers observed limits generalization of our findings. Although there were 230 African American men in the present study, only one case had the HOXB13 mutation. It is notable that none of the mutations reported by Ewing were in men who self identified as African American (12). We also only tested for the G84E mutation, as have other studies. At least four additional mutations have been reported (12). One useful exercise would be to take our most at risk cases and scan the entire gene for those additional mutations. Finally, the size of our carrier pool, and the fact that the data were from a population-based study, and not HPC families, also prohibited us from doing any type of penetrance analysis.
In this population-based, case-control study of PC, we found that while rare, as reported in other studies, the HOXB13 G85E mutation is strongly associated with PC risk. There was also some suggestion that the association may vary with Gleason score, disease aggressiveness and stage, but additional studies and meta-analyses will be needed to confirm this observation. In addition, other mutations recurring in the general population remain to be determined.
Acknowledgments
This work was supported by NIH grants R01 CA0566787, R01 CA092579, R01 CA082664, and P50 CA097186.
J.L.S., S.K., and M.S.M. acknowledge additional support from the Fred Hutchinson Cancer Research Center. E.A.O., D.M.K., E.M.K. and T.K. gratefully acknowledge the Intramural Program of the National Human Genome Research Institute.
Footnotes
The authors have no conflict of interest to disclose.
References
- 1.Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA: A Cancer Journal for Clinicians. 2012;62(1):10–29. doi: 10.3322/caac.20138. [DOI] [PubMed] [Google Scholar]
- 2.Whittemore AS, Wu AH, Kolonel LN, John EM, Gallagher RP, Howe GR, West DW, Teh CZ, Stamey T. Family history and prostate cancer risk in black, white, and Asian men in the United States and Canada. Am J Epidemiol. 1995;141(8):732–740. doi: 10.1093/oxfordjournals.aje.a117495. [DOI] [PubMed] [Google Scholar]
- 3.Lichtenstein P, Holm NV, Verkasalo PK, Iliadou A, Kaprio J, Koskenvuo M, Pukkala E, Skytthe A, Hemminki K. Environmental and heritable factors in the causation of cancer--analyses of cohorts of twins from Sweden, Denmark, and Finland [see comments] N Engl J Med. 2000;343(2):78–85. doi: 10.1056/NEJM200007133430201. [DOI] [PubMed] [Google Scholar]
- 4.Schaid DJ. The complex genetic epidemiology of prostate cancer. Hum Mol Genet. 2004;13(Spec No 1):R103–121. doi: 10.1093/hmg/ddh072. [DOI] [PubMed] [Google Scholar]
- 5.Easton DF, Schaid DJ, Whittemore AS, Isaacs WJ. Where are the prostate cancer genes?--A summary of eight genome wide searches. Prostate. 2003;57(4):261–269. doi: 10.1002/pros.10300. [DOI] [PubMed] [Google Scholar]
- 6.Ostrander EA, Markianos K, Stanford JL. Finding prostate cancer susceptibility genes. Annu Rev Genomics Hum Genet. 2004;5:151–175. doi: 10.1146/annurev.genom.5.061903.180044. [DOI] [PubMed] [Google Scholar]
- 7.Levin AM, Machiela MJ, Zuhlke KA, Ray AM, Cooney KA, Douglas JA. Chromosome 17q12 variants contribute to risk of early-onset prostate cancer. Cancer Res. 2008;68(16):6492–6495. doi: 10.1158/0008-5472.CAN-08-0348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lange EM, Gillanders EM, Davis CC, Brown WM, Campbell JK, Jones M, Gildea D, Riedesel E, Albertus J, Freas-Lutz D, Markey C, Giri V, Dimmer JB, Montie JE, Trent JM, Cooney KA. Genome-wide scan for prostate cancer susceptibility genes using families from the University of Michigan prostate cancer genetics project finds evidence for linkage on chromosome 17 near BRCA1. Prostate. 2003;57(4):326–334. doi: 10.1002/pros.10307. [DOI] [PubMed] [Google Scholar]
- 9.Gillanders EM, Xu J, Chang BL, Lange EM, Wiklund F, Bailey-Wilson JE, Baffoe-Bonnie A, Jones M, Gildea D, Riedesel E, Albertus J, Isaacs SD, Wiley KE, Mohai CE, Matikainen MP, Tammela TL, Zheng SL, Brown WM, Rokman A, Carpten JD, Meyers DA, Walsh PC, Schleutker J, Gronberg H, Cooney KA, Isaacs WB, Trent JM. Combined genome-wide scan for prostate cancer susceptibility genes. J Natl Cancer Inst. 2004;96(16):1240–1247. doi: 10.1093/jnci/djh228. [DOI] [PubMed] [Google Scholar]
- 10.Xu J, Dimitrov L, Chang BL, Adams TS, Turner AR, Meyers DA, Eeles RA, Easton DF, Foulkes WD, Simard J, Giles GG, Hopper JL, Mahle L, Moller P, Bishop T, Evans C, Edwards S, Meitz J, Bullock S, Hope Q, Hsieh CL, Halpern J, Balise RN, Oakley-Girvan I, Whittemore AS, Ewing CM, Gielzak M, Isaacs SD, Walsh PC, Wiley KE, Isaacs WB, Thibodeau SN, McDonnell SK, Cunningham JM, Zarfas KE, Hebbring S, Schaid DJ, Friedrichsen DM, Deutsch K, Kolb S, Badzioch M, Jarvik GP, Janer M, Hood L, Ostrander EA, Stanford JL, Lange EM, Beebe-Dimmer JL, Mohai CE, Cooney KA, Ikonen T, Baffoe-Bonnie A, Fredriksson H, Matikainen MP, Tammela T, Bailey-Wilson J, Schleutker J, Maier C, Herkommer K, Hoegel JJ, Vogel W, Paiss T, Wiklund F, Emanuelsson M, Stenman E, Jonsson BA, Gronberg H, Camp NJ, Farnham J, Cannon-Albright LA, Seminara D. A combined genomewide linkage scan of 1,233 families for prostate cancer-susceptibility genes conducted by the international consortium for prostate cancer genetics. Am J Hum Genet. 2005;77(2):219–229. doi: 10.1086/432377. Epub 2005 Jun 2029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lange EM, Robbins CM, Gillanders EM, Zheng SL, Xu J, Wang Y, White KA, Chang BL, Ho LA, Trent JM, Carpten JD, Isaacs WB, Cooney KA. Fine-mapping the putative chromosome 17q21-22 prostate cancer susceptibility gene to a 10 cM region based on linkage analysis. Hum Genet. 2007;121(1):49–55. doi: 10.1007/s00439-006-0274-2. [DOI] [PubMed] [Google Scholar]
- 12.Ewing CM, Ray AM, Lange EM, Zuhlke KA, Robbins CM, Tembe WD, Wiley KE, Isaacs SD, Johng D, Wang Y, Bizon C, Yan G, Gielzak M, Partin AW, Shanmugam V, Izatt T, Sinari S, Craig DW, Zheng SL, Walsh PC, Montie JE, Xu J, Carpten JD, Isaacs WB, Cooney KA. Germline mutations in HOXB13 and prostate-cancer risk. N Eng J Med. 2012;12(366):141–149. doi: 10.1056/NEJMoa1110000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jeong TO, Oh KJ, Xuan Nguyen NT, Kim YR, Kim MS, Lee SD, Ryu SB, Jung C. Evaluation of HOXB13 as a molecular marker of recurrent prostate cancer. Mol Med Report. 2012;5(4):901–904. doi: 10.3892/mmr.2012.769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Breyer JP, Avritt TG, McReynolds KM, Dupont WD, Smith JR. Confirmation of the HOXB13 G84E Germline Mutation in Familial Prostate Cancer. Cancer Epidemiol Biomarkers Prev. 2012 Jul 18; doi: 10.1158/1055-9965.EPI-12-0495. [Epub ahead of pring] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lin X, Qu L, Chen Z, Xu C, Ye D, Shao Q, Wang X, Qi J, Chen Z, Zhou F, Wang M, Wang Z, He D, Wu D, Gao X, Yuan J, Wang G, Xu Y, Wang G, Dong P, Jiao Y, Yang J, Ou-Yang J, Jiang H, Zhu Y, Ren S, Zhang Z, Yin C, Wu Q, Zheng Y, Turner AR, Tao S, Na R, Ding Q, Lu D, Shi R, Sun J, Liu F, Zheng SL, Mo Z, Sun Y, Xu J. A novel germline mutation in HOXB13 is associated with proste cancer risk in Chinese men. Prostate. 2012 doi: 10. 1002/pros.22552. [Epub ahead of print] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Karlsson R, Aly M, Clements M, Zheng L, Adolfsson J, Xu J, Gronberg H, Wiklund F. A population-based assessment of germline HOXB13 G84E mutation and prostate cancer risk. Eur Urol. 2012 doi: 10.1016/j.eururo.2012.07.027. (In Press) [DOI] [PubMed] [Google Scholar]
- 17.Akbari MR, Trachtenberg J, Lee J, Tam S, Bristow R, Loblaw A, Narod SA, Nam RK. Association between germline HOXB13 G84E mutation and risk of prostate cancer. J Natl Cancer Inst. 2012 doi: 10.1093/jnci/djs288. [EPub ahead of print] [DOI] [PubMed] [Google Scholar]
- 18.Agalliu I, Salinas CA, Hansten PD, Ostrander EA, Stanford JL. Statin use and risk of prostate cancer: results from a population-based epidemiologic study. Am J Epid. 2008;168(3):250–260. doi: 10.1093/aje/kwn141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Stanford JL, Wicklund KG, McKnight B, Daling JR, Brawer MK. Vasectomy and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev. 1999;8(10):881–886. [PubMed] [Google Scholar]
- 20.Waksberg J. Sample methods for random digit dialing. JAm Stat Soc. 1978;73:40–46. [Google Scholar]
- 21.Wright JL, Salinas CA, Lin DW, Kolb S, Koopmeiners J, Feng Z, Stanford JL. Prostate cancer specific mortality and Gleason 7 disease differences in prostate cancer outcomes between cases with Gleason 4 + 3 and Gleason 3 + 4 tumors in a population based cohort. J Urol. 2009;182(6):2702–2707. doi: 10.1016/j.juro.2009.08.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Holt SK, Kwon EM, Lin DW, Ostrander EA, Stanford JL. Association of hepsin gene variants with prostate cancer risk and prognosis. Prostate. 2010;70(9):1012–1019. doi: 10.1002/pros.21135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Graham A. Developmental patterning: The Hox code out on a limb. Curr Biol. 1994;4(12):1135–1137. doi: 10.1016/s0960-9822(00)00256-6. [DOI] [PubMed] [Google Scholar]
- 24.Goodman FR, Scambler PJ. Human HOX gene mutations. Clin Genet. 2001;59:1–11. doi: 10.1034/j.1399-0004.2001.590101.x. [DOI] [PubMed] [Google Scholar]
- 25.Kim SD, Park RY, Kim YR, Kim IJ, Kang TW, Nam KI, Ahn KY, Bae CS, Kim BY, Park SS, Jung C. HOXB13 is co-localized with androgen receptor to suppress androgen-stimulated prostate-specific antigen expression. Anat Cell Biol. 2010;43(4):284–293. doi: 10.5115/acb.2010.43.4.284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zeltser L, Desplan C, Heintz N. Hoxb-13: A new Hox gene in a distant region of the HOXB cluster maintains colinearity. Development. 1966;122(8):2475–2484. doi: 10.1242/dev.122.8.2475. [DOI] [PubMed] [Google Scholar]
- 27.Mortlock DP, Sateesh P, Innis JW. Evolution of N-terminal sequences of the vertebrate HOXA13 protein. Mamm Genome. 2000;11(2):151–158. doi: 10.1007/s003350010029. [DOI] [PubMed] [Google Scholar]
- 28.Shen WF, Montgomery JC, Rozenfeld S, Moskow JJ, Lawrence HJ, Buchberg AM, Largman C. AbdB-like Hox proteins stabilize DNA binding by the Meis1 homeodomain proteins. Mol Cell Biol. 1997;17(11):6448–6458. doi: 10.1128/mcb.17.11.6448. [DOI] [PMC free article] [PubMed] [Google Scholar]