SEARCH

SEARCH BY CITATION

Keywords:

  • FGF23;
  • polymorphism;
  • prostate cancer

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

Objective

  • To determine whether sequence variants within the FGF23 gene are associated with the risk of developing prostate cancer in a Korean population.

Patients and Methods

  • Five common single nucleotide polymorphisms (SNPs) in the FGF23 gene were assessed in 272 patients with prostate cancer and 173 control subjects with benign prostatic hyperplasia.
  • Single-locus analyses were conducted using conditional logistic regression.
  • In addition, we performed a haplotype analysis for the five FGF23 SNPs tested.

Results

  • Three SNPs in the FGF23 gene (rs11063118, rs13312789 and rs7955866) were associated with an increased risk of prostate cancer in our study population.
  • Odds ratios for homozygous variants vs wild-type variants ranged from 1.68 (95% confidence interval [CI]: 1.15–2.46) to 1.79 (95% CI: 1.16–2.75).

Conclusion

  • This is the first study showing that genetic variations in FGF23 increase prostate cancer susceptibility.

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

Prostate cancer remains the second leading cause of cancer deaths in the western world [1]. Similarly, in Korea, the incidence rate of prostate cancer has rapidly increased during the last decade [2]. It is also a major cause of cancer-related morbidity. Although epidemiological investigations over several decades have studied exogenous risk factors for prostate cancer, including diet, occupation and sexually transmitted agents, the only established risk factors are age, ethnicity and family history of prostate cancer [3]. Since age is thought to be one of the crucial factors in prostate cancer development, fibroblast growth factor (FGF)-23, which forms a complex with Klotho, the age-related gene product, could play a key role in this regard. We hypothesized that sequence variation in genes involved in aging were candidates for risk factors for the development of prostate cancer. We chose to investigate genetic variation in FGF-23, which acts as a ligand for KLOTHO, responsible for aging [4, 5].

FGF-23 is a peptide hormone member of the FGF-19 subfamily, which, like other FGFs, signals through FGF receptors (FGFRs) [6, 7]; however, because of its atypical heparin-binding domain, FGF-23 binds FGFRs with a low affinity [7], except in the presence of the coreceptor α-Klotho, which confers high-affinity binding [8]. FGF-23 has a critical role in phosphate homeostasis [9, 10] and its signalling is mediated via the complex formed by FGF-23, FGFR1c and Klotho [9]. FGF-23 is a key component of an endocrine feedback loop, along with the hormonal vitamin D metabolite 1,25(OH)2D. The regulation of 1,25(OH)2D by FGF-23 could also have implications for cancer, given that vitamin D metabolites have been shown to have anticarcinogenic effects.

The hormone 1,25(OH)2D has antiproliferative and prodifferentiation effects in cancer cell lines [10]. It has been reported to reduce the growth of many prostate cancer cell lines by several mechanisms, including altering the secretion and signalling of growth factors [11-13], induction of apoptosis [14], and/or induction of a cell-cycle arrest, most commonly in G1 [15, 16]. Epidemiological investigations have shown that vitamin D metabolite levels are associated strongly with a subsequent risk of prostate cancer [17]. Given the antineoplastic effects of 1,25(OH)2D, the regulation of this hormone by FGF-23 may be of importance in prostate carcinogenesis.

Two epidemiological investigations between FGF-23 and cancer have been reported. Tebben et al. [18] reported significantly greater concentrations of FGF-23 in patients with advanced ovarian cancer compared with those with early-stage cancer or control subjects. Jacobs et al. [19] reported circulating FGF-23 to be associated with an increased risk of metachronous colorectal adenoma; however, to date, no epidemiological investigations of FGF-23 and risk for prostate neoplasia have been reported.

Because of the critical role of FGF-23 in the maintenance of 1,25(OH)2D and its relation to Klotho, the anti-aging molecule, the present study was performed to assess whether single nucleotide polymorphisms (SNPs) in the FGF23 gene were associated with the development of prostate cancer in Korean men. This is the first study to evaluate the association between FGF23 polymorphisms and prostate cancer risk.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

Study Population

The study cohort comprised two groups, a prostate cancer and BPH (control) group, originated from a population of older men treated for urological problems at the Seoul National University Bundang Hospital (Kyonggi, Seoul) and Chung-Ang University Hospital (Seoul, Korea). Peripheral blood leukocyte samples were obtained for genotyping from 445 men (prostate cancer, n = 272; BPH, n = 173) and were stored at −80 °C. Subjects with BPH (n = 173) were included after PSA blood tests, DRE and true biopsy confirming that they were prostate cancer-free. The median ages of the BPH and prostate cancer cohorts were 67.3 and 68.2 years, respectively. BPH samples were used as the control group for two reasons; firstly, most males have evidence of BPH by the time they are aged 70–80 years, so the presence of some degree of BPH would be ‘normal’ at the median age of diagnosis (68.2 years) in our prostate cancer cohort. Truly ‘normal’ samples would only be obtained in a much younger control cohort, which could introduce bias. Secondly, collection of blood samples requires hospital visiting and a prostate cancer screening procedure would only be undertaken in subjects with evidence of symptoms of prostate enlargement. Written informed consent was obtained from all the study participants. The study was approved by the Institutional Review Board of Chung-Ang University Hospital and Seoul National University Bundang Hospital (IRB numbers C2008035, B-0905/075-011, respectively). Blood samples were collected in tubes containing sodium EDTA. A QIAamp blood extraction kit (Qiagen, Seoul, Korea) was used for DNA extraction. The Gleason score was classified as low (2–6), intermediate (4+3, 3+4), or high (8–10) grades. The clinical and pathological regional stages were categorized as localized (T1 or T2N0M0), locally advanced (T3 or T4N0M0), and metastatic (TxN+ or M+) based on pathological and/or radiological reports. Clinical characteristics of the study cohort are shown in Table 1, and were similar to those reported in the previous Korean study [20].

Table 1. Characteristics of patients in the prostate cancer and BPH (control) group
 Prostate cancer groupBPH (control) group
Number of patients272173
Mean (sd) age, years68.2 (6.8)67.3 (8.8)
Mean (sd) body mass index, kg/m224.1 (3.3)24.0 (3.0)
Mean (sd) prostate volume, cm337.2 (18.6)48.4 (26.2)
Mean (sd) PSA, ng/mL48.2 (192.8)5.2 (6.7)
Gleason score, n (%)  
Low grade29 (11) 
3+4 or 4+3202 (75) 
High grade39 (14) 
Stage, n (%)  
Localized252 (92.6) 
Locally advanced10 (3.7) 
Metastatic8 (2.9) 
Unknown2 (0.7) 

Selection and Genotyping of SNPs

We selected five SNPs from two international databases (International HapMap and the National Center for Biotechnology Information [NCBI] dbSNPs). SNP selection from the International HapMap database (Han Chinese and Japanese) was conducted as follows: (i) extraction of all genotypes from the Han Chinese and Japanese population in the FGF23 gene region using HapMart of the International HapMap database (version: release #27; http://www.hapmap.org); (ii) calculation of minor allele frequency (MAF) and linkage disequilibrium (LD) using Haploview software (Cambridge, MA, USA; http://www.broad.mit.edu/mpg/haploview); and (iii) selection of SNPs having MAF >0.05 and tagging SNPs if several SNPs showed a high LD (>0.98). Furthermore, we added the SNPs from the NCBI dbSNPs in the FGF23 gene region. The selection criteria included location (SNPs in exons were preferred) and amino acid changes (non-synonymous SNPs were preferred). Selected SNPs were genotyped using the TaqMan® assay [21]. Genotyping quality control was performed in 10% of the samples by duplicate checking (rate of concordance in duplicates >99.9%). The assay names (Applied Biosystems, Foster City, CA, USA) of each of the SNPs are described in Table S1.

Statistics

The SNP genotype frequencies were examined for Hardy–Weinberg equilibrium (HWE) using the chi-squared statistic, and all were found to be consistent (P > 0.05) with HWE among the Korean control subjects. Data were analysed using unconditional logistic regression to calculate an odds ratio (OR) as an estimate of the relative risk of prostate cancer associated with SNP genotypes [22].

To determine the association between the genotype and haplotype distributions of patients and control subjects, logistic analysis was carried out, controlling for age (continuous value) as a covariate, to eliminate or reduce any confounding that might influence the findings. A P value ≤0.05 was considered to indicate statistical significance. To address the problem of multiple comparisons, Bonferroni correction was used. ‘Lewontin's D’ and the LD coefficient r2 were examined to measure the LD between all pairs of biallelic loci 9 [23]. Using phase algorithm ver. 2.0 [24], the haplotypes were inferred from the successfully genotyped SNPs, and association analysis was performed using sas version 9.1 (SAS Inc., Cary, NC, USA). To achieve the optimum correction for multiple testing of markers, representing SNPs in LD with each other, the effective number of independent marker loci (3.95) was calculated using snpspd software (http://genepi.qimr.edu.au/general/daleN/SNPSpD/), a program that is based on the spectral decomposition of matrices of pairwise LD among markers [25].

Genotypes of major homozygotes (A/A), heterozygotes (A/B) and minor homozygotes (B/B) were given codes of 0, 1 and 2; 0, 1 and 1; and 0, 0 and 1 in the codominant, dominant and recessive models, respectively. Genetic effects of inferred haplotypes were analysed in the same way as SNPs. For example, 0 copy of Haplotype-1 (-/-, major homozygote), one copy of Ht-1 (-/Ht1, heterozygote) and two copies of Ht1 (Ht1/Ht1, minor homozygote) were coded as 0, 1 and 2 in the codominant model.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

Five sequence variants in the FGF23 gene were examined in this study: three were located in introns; one in exon; one was in the 3'-untranslated region (UTR) (Fig. 1A). The MAFs of the five genotyped SNPs are shown in Fig. 1A and Table S1. No significant deviations from the HWE were observed in the SNPs (P > 0.05; Table S1).

figure

Figure 1. (A) Genetic map of FGF23 on chromosome 12p13.3. Coding exons are marked by black blocks, and 5'- and 3'UTRs by white blocks. (B) Haplotypes of FGF23. ‘Others’ category contains rare haplotypes. (C) LD among FGF23 polymorphisms.

Download figure to PowerPoint

The genotype frequencies for each SNP in both the prostate cancer group and the control group were analysed using a logistic regression model (Table 2). Among the five SNPs examined, three (rs11063118, rs13312789 and rs7955866) were found to be significantly associated with prostate cancer risk in a codominant model (OR = 1.68, Pcorr = 0.03; OR = 1.72, Pcorr = 0.04; OR = 1.79, Pcorr = 0.03, respectively, Table 2).

Table 2. Logistic regression analysis of the association of FGF23 SNPs with the risk of prostate cancer in the Korean study population: MAFs and P values for three models, codominant, dominant and recessive models, controlling for age as a covariate
SNPIDChromosomePositionAA changeAllelesMAFCodominantPcorrDominantPcorrRecessivePcorr
Prostate cancer group, n = 272BPH group, n = 173OR (95% CI)POR (95% CI)POR (95% CI)P
  1. To achieve optimum correction for multiple testing of SNPs in LD with each other, the effective number of independent marker loci (3.95) in FGF23 was calculated using the software snpspd (http://genepi.qimr.edu.au/general/daleN/SNPSpD/) on the basis of the spectral decomposition of matrices of pairwise LD between SNPs. Significant associations (P ≤ 0.05) are shown in boldface. NS, nonsignificant.

rs177824695Intron T>G0.3470.3840.81 (0.61–1.09)0.17NS0.73 (0.49–1.10)0.13NS0.84 (0.47–1.51)0.57NS
rs110631185Intron T>C0.2020.1211.68 (1.15–2.46)0.0080.031.91 (1.22–3.01)0.0050.021.71 (0.59–4.93)0.32NS
rs133127895Intron C>T0.1580.0901.72 (1.11–2.65)0.010.042.03 (1.23–3.34)0.0060.021.24 (0.36–4.20)0.73NS
rs79558665CDST239MG>A0.1640.0921.79 (1.16–2.75)0.0080.032.12 (1.30–3.48)0.0030.011.24 (0.36–4.20)0.73NS
rs1331279453'UTR T>G0.0200.0260.86 (0.34–2.19)0.75NS0.86 (0.34–2.19)0.75NS   
FGF23_ht15   0.4280.4620.91 (0.68–1.21)0.51NS0.87 (0.56–1.34)0.52NS0.90 (0.55–1.49)0.69NS
FGF23_ht25   0.3420.3840.79 (0.59–1.06)0.12NS0.71 (0.47–1.07)0.10NS0.80 (0.45–1.44)0.46NS
FGF23_ht35   0.1560.0871.76 (1.14–2.72)0.010.042.10 (1.26–3.48)0.0040.021.24 (0.36–4.20)0.73NS

In an analysis involving only patients in the prostate cancer group, no associations were detected between the five SNPs examined in the present study and clinical characteristics (PSA, clinical tumour stage or Gleason score [Tables 3-5]); however, significant associations were found between some SNPs and pathological tumour stage (Table 6). One SNP (rs17782469) and haplotype 2 showed a significant association with pathological aggressiveness in a recessive model (OR = 2.82, P = 0.01; OR = 3.12, P = 0.007, respectively), while one haplotype (ht1) showed a significant protective effect from pathological aggressiveness in codominant and dominant models (OR = 0.66, P = 0.04; OR = 0.56, P = 0.04, respectively [Table 6]).

Table 3. Logistic analysis of FGF23 polymorphisms according to PSA criteria: MAFs and P values for codominant, dominant and recessive models, controlling for age as a covariate
SNP IDMAFCodominantDominantRecessive
PSA ≥10 ng/mL, n = 113PSA 4–10 ng/mL, n = 98PSA <4 ng/mL, n = 62OR (95% CI)POR (95% CI)POR (95% CI)P
  1. Significant associations (P ≤ 0.05) are shown in boldface.

rs177824690.3190.3720.3550.87 (0.62–1.21)0.400.70 (0.45–1.10)0.131.22 (0.61–2.44)0.57
rs110631180.2080.1840.2340.94 (0.65–1.36)0.740.94 (0.59–1.50)0.810.85 (0.33–2.17)0.74
rs133127890.1590.1480.1770.94 (0.62–1.43)0.780.94 (0.58–1.53)0.800.89 (0.26–3.04)0.85
rs79558660.1640.1530.1850.93 (0.62–1.40)0.720.92 (0.57–1.49)0.740.89 (0.26–3.04)0.85
rs133127940.0310.0050.0241.95 (0.60–6.36)0.271.95 (0.60–6.36)0.27  
FGF23_ht10.4420.4340.3871.15 (0.84–1.60)0.391.04 (0.65–1.67)0.871.51 (0.83–2.75)0.17
FGF23_ht20.3140.3670.3470.87 (0.63–1.22)0.430.69 (0.44–1.08)0.101.34 (0.66–2.72)0.41
FGF23_ht30.1590.1430.1770.95 (0.62–1.43)0.790.94 (0.58–1.54)0.810.89 (0.26–3.04)0.85
Table 4. Logistic analysis of FGF23 SNPs according to clinical stage criteria: MAFs and P values for codominant, dominant and recessive models, controlling for age as a covariate
SNP IDMAFCodominantDominantRecessive
Metastatic (n = 8)Locally advanced (n = 10)Localized (n = 252)OR (95% CI)POR (95%CI)POR (95% CI)P
  1. Significant associations (P ≤ 0.05) are shown in boldface.

rs177824690.4380.4000.3451.35 (0.67–2.70)0.401.50 (0.54–4.13)0.441.45 (0.39–5.41)0.58
rs110631180.1880.2000.2040.95 (0.42–2.17)0.911.20 (0.45–3.21)0.72  
rs133127890.1250.2000.1591.07 (0.44–2.61)0.881.28 (0.46–3.57)0.63  
rs79558660.1250.2000.1651.02 (0.42–2.51)0.961.21 (0.44–3.35)0.72  
rs133127940.0000.0500.0201.40 (0.16–11.95)0.761.40 (0.16–11.95)0.76  
FGF23_ht10.3750.3500.4290.74 (0.36–1.53)0.421.23 (0.42–3.58)0.71  
FGF23_ht20.4380.4000.3391.38 (0.69–2.76)0.361.55 (0.56–4.28)0.401.50(0.40–5.59)0.55
FGF23_ht30.1250.2000.1571.08 (0.44–2.63)0.871.30 (0.47–3.62)0.61  
Table 5. Logistic analysis of FGF23 SNPs according to Gleason score criteria: MAFs and P values for codominant, dominant and recessive models, controlling for age as a covariate
SNP IDMAFCodominantDominantRecessive
High grade, n = 39Intermediate, n = 202Low grade, n = 29OR (95% CI)POR (95% CI)POR (95% CI)P
  1. Significant associations (P ≤ 0.05) are shown in boldface.

rs177824690.3840.2410.3480.92 (0.61–1.38)0.670.87 (0.50–1.52)0.630.93 (0.40–2.17)0.88
rs110631180.1610.3100.2021.26 (0.80–1.98)0.331.45 (0.82–2.58)0.200.84 (0.26–2.78)0.78
rs133127890.1310.1900.1571.57 (0.94–2.60)0.081.77 (0.96–3.24)0.071.51 (0.34–6.67)0.58
rs79558660.1360.2070.1631.45 (0.88–2.41)0.151.58 (0.87–2.88)0.131.51 (0.34–6.67)0.58
rs133127940.0200.0340.0200.56 (0.14–2.19)0.400.56 (0.14–2.19)0.40  
FGF23_ht10.4100.4330.4140.98 (0.65–1.45)0.901.01 (0.56–1.81)0.990.92 (0.44–1.89)0.81
FGF23_ht20.2440.3790.2240.95 (0.63–1.43)0.820.88 (0.51–1.52)0.651.09 (0.47–2.56)0.84
FGF23_ht30.2690.1290.1901.57 (0.94–2.61)0.081.77 (0.97–3.26)0.071.51 (0.34–6.67)0.58
Table 6. Logistic analysis of FGF23 SNPs according to pathological stage criteria: MAFs and P values for codominant, dominant and recessive models, controlling for age as a covariate
SNP IDMAFCodominantDominantRecessive
Locally advanced, n = 97Localized, n = 150OR (95% CI)POR (95% CI)POR (95% CI)P
  1. Significant associations (P ≤ 0.05) are shown in boldface.

rs177824690.3870.3331.27 (0.86–1.86)0.230.97 (0.58–1.64)0.922.82 (1.27–6.29)0.01
rs110631180.2060.1871.14 (0.73–1.77)0.571.26 (0.74–2.17)0.390.81 (0.24–2.77)0.73
rs133127890.1750.1471.23 (0.76–1.99)0.401.23 (0.70–2.16)0.481.69 (0.41–6.99)0.47
rs79558660.1750.1571.14 (0.71–1.84)0.581.11 (0.63–1.94)0.721.69 (0.41–6.99)0.47
rs133127940.0310.0103.46 (0.83–14.38)0.093.46 (0.83–14.38)0.09  
FGF23_ht10.3760.4670.66 (0.45–0.98)0.040.56 (0.32–0.97)0.040.65 (0.32–1.32)0.23
FGF23_ht20.3870.3231.33 (0.90–1.95)0.151.03 (0.61–1.73)0.913.12 (1.37–7.10)0.007
FGF23_ht30.1750.1431.26 (0.78–2.03)0.351.26 (0.72–2.23)0.421.69 (0.41–6.99)0.47

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

The present case–control study was undertaken to investigate the potential association between FGF23 gene polymorphisms and the risk of prostate cancer in a Korean population. In this study, logistic regression analyses suggested that some FGF23 gene polymorphisms were associated with a significantly elevated risk of prostate cancer when compared with healthy control subjects. These results suggest that FGF23 gene polymorphisms may alter susceptibility to prostate cancer and might be used as biomarkers for the disease.

Only few studies have evaluated the association between FGF-23 and cancer. Tebben et al. [18] reported that serum or plasma FGF-23 concentrations are elevated in patients with advanced-stage epithelial ovarian cancer. They used two assays for FGF-23. The iFGF23 assay does not detect carboxy-terminal fragments, and is thus a measure of intact, and presumably bioactive, protein. The cFGF23 assay measured intact protein and carboxy-terminal fragments. They observed a significant positive correlation between serum iFGF23 and cFGF23 concentrations and stage of disease. Jacobs et al. [19] provided evidence of a relationship between FGF-23 and the risk for colorectal neoplasia.

In the present study, we found that three SNPs were associated with an increased risk of prostate cancer. Of these, two were intronic and one caused amino acid substitution (T239M). One study has explained T239M variation of FGF23 in stone formers with renal phosphate leak [26]. In vitro studies have shown that the T239M change increases FGF-23 secretion and that FGF239M variant induces a higher activation of the the FGF receptor/Extracellular signal-regulated kinase (ERK) pathway compared with FGF23239T. FGF239M may influence the growth of prostate cells by activating the ERK pathway.

Both antiproliferative and prodifferentiating effects of 1,25(OH)2D on cancer cell lines have been reported [10], along with growth inhibition in prostate cancer cell lines [11-16]. Several mechanisms of action have been proposed to explain the inhibitory action of vitamin D on carcinogenesis. Induction of G0/G1 cell-cycle arrest by 1,25(OH)2D in HL-60 human leukaemia and breast cancer cell lines, possibly via increased expression of the cyclin-dependent kinase inhibitor proteins p27kip1, has been reported [10]. Colorectal carcinoma cell differentiation may be induced by 1,25(OH)2D-mediated transcription of E-cadherin, a tumour suppressor gene. Also, 1,25(OH)2D inhibits the expression of bcl-2, a suppressor of apoptosis, in HL-60 cells [27], and 1,25(OH)2D has been shown to reduce the transcriptional activity of β-catenin, a key proto-oncogene in colorectal carcinogenesis. Given the potential for antineoplastic activity by 1,25(OH)2D, the influence of FGF-23 on the production of this hormone is a critical consideration.

As reviewed by Marsell and Jonsson [9] and Liu and Quarles [28], modulation of phosphate homeostasis and vitamin D metabolism in the kidney are the principal functions of FGF-23. The biological interaction between 1,25(OH)2D and FGF-23 is intricate and appears to manifest itself via phosphate regulation [28]. In the intestine, 1,25(OH)2D stimulates the absorption of calcium and phosphate, whereas 1,25(OH) 2D and phosphate induce the expression and release of FGF-23, primarily from bone, to prevent phosphate levels from rising too rapidly [10, 28]. In turn, FGF-23 inhibits further synthesis of 1,25(OH)2D by inducing the kidney vitamin D 24-hydroxylase (CYP24A1), which catalyses 1,25(OH)2D catabolism [10], as well as by suppressing the kidney CYP27B1. Jacobs et al. [19] proposed that the observed suppression of 1,25(OH)2D by FGF-23 results in increased odds for metachronous adenoma in individuals with higher FGF-23 concentrations than in those with lower levels.

Despite the relationship between FGF-23 and 1,25(OH) 2D, the possibility remains that FGF-23 has independent effects on prostate neoplasia or may act via another pathway, perhaps related to body size. Previous work has shown that higher circulating FGF-23 is significantly associated with an increased risk of cardiovascular disease, which shares many common risk factors with colorectal cancer, although the mechanism of action is currently unknown [9]. We didn't measure FGF-23 concentrations in the present study, but the current work presents novel preliminary hypothesis-generating evidence for a potent relationship between FGF-23 and prostate cancer. It merits further investigation with a larger sample size to better understand the epidemiology and biology of FGF-23.

Some limitations of this study should be acknowledged. Our analysis was based on comparison of samples from patients with prostate cancer and samples from patients with non-malignant BPH. This comparison might be somewhat different from the comparison with the general population of men without any prostatic disease. Furthermore, all the men in the BPH group were potentially at risk for development of prostate cancer and may have had latent prostate cancer at the time of designation as control subjects, leading to disease misclassification. One Korean study provides a useful statistic; its authors evaluated the cancer detection rate between patients who underwent TRUS biopsy before TURP (group A) and those who did not (group B) [29]. The cancer detection rates of the two groups were 8.9 and 7.5%, respectively. This study demonstrated the percentage of Korean men incidentally detected with prostate cancer after initial diagnosis with BPH.

Despite these limitations, the present study provides the first evidence of an important and novel association between FGF23 SNPs and the risk of prostate cancer. It supports a potential mechanism of action through the vitamin D pathway, though it is not clear whether FGF-23 acts via 1,25(OH)2D or has independent effects. A larger study population and well-designed and complementary biochemical experiments are necessary to clarify whether FGF-23 and vitamin D metabolites are acting independently or in concert to effect prostate carcinogenesis.

We have shown that three SNPs (rs11063118, rs13312789 and rs7955866) in the FGF23 gene are good candidates for genetic markers that can help predict the risk of prostate cancer. Our finding, we believe, is the first study to show that SNPs in FGF23 might influence the susceptibility of prostate cancer. Although further studies are needed, this study provides a potential use for FGF23 SNPs as diagnostic markers for prostate cancer.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

This work was supported by a grant from the Korea Healthcare Technology R&D Project, Ministry of Health, Welfare and Family Affairs, Republic of Korea (A085138).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information
Abbreviations
SNP

single nucleotide polymorphism

FGF

fibroblast growth factor

FGFR

fibroblast growth factor receptor

NCBI

National Center for Biotechnology Information

MAF

minor allele frequency

LD

linkage disequilibrium

HWE

Hardy–Weinberg equilibrium

OR

odds ratio

UTR

untranslated region

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
bju12396-sup-0001-ts1.docx15K

Table S1 Assay names (IDs) of FGF23 SNPs.

bju12396-sup-0002-ts2.doc45K

Table S2 Frequencies of FGF23 SNPs in patients with prostate cancer and normal control subjects (N = 445).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.