Prognostic Significance of Prostate Cancer Susceptibility Variants on Prostate-Specific Antigen Recurrence after Radical Prostatectomy

Introduction

Prostate cancer is the most common cancer and the second leading cause of cancer deaths among American men (1). However, its etiology remains poorly understood. Incidence and mortality rates of prostate cancer vary substantially worldwide, suggesting the importance of environmental/lifestyle risk factors and perhaps their combination with genetic variants across racial/ethnic populations. Recently, several genomewide association studies have identified several genetic variants as being associated with risk for prostate cancer in men of European ancestry (2-9). Although these novel risk variants have been replicated in several study populations, little is known about the associations of these risk variants with non-European populations and with clinicopathologic features of prostate cancer.

Patients diagnosed with localized prostate cancer are commonly treated with radical prostatectomy. Of these patients, 15% to 46% experience recurrence of disease as detected by a relapse in prostate-specific antigen (10). Although several clinicopathologic indicators, such as prostate-specific antigen level, Gleason score, pathologic stage, and surgical margin status, are currently used to predict outcome following curative intended radical prostatectomy for localized prostate cancer (11), there is a need to find new biomarkers to improve the prediction of disease recurrence and selection of appropriate adjuvant therapy for high-risk patients.

DNA-based genetic biomarkers have certain advantages over clinicopathologic indicators in that they can be done preoperatively, can be conducted easily, and can be interpreted more objectively without individual bias. Despite some recent conducted studies showed a trend of prostate cancer susceptibility variants observed more frequently in prostate cancer patients with early age at diagnosis, higher Gleason score, or more aggressive disease (12-14), the prognostic roles of these risk variants on disease progression remain undetermined. Thus, the aim of this study was to investigate the prognostic significance of the prostate cancer susceptibility variants and recurrence of prostate-specific antigen in clinically localized prostate cancer patients after radical prostatectomy.

Results

As can be seen in Table 1, which summarizes the demographic characteristics of our study population, prostate-specific antigen recurred in 113 (35.3%) of 320 prostate cancer patients who had received radical prostatectomy during the mean and median follow-up of 38.5 and 30.8 months (range, 1-96 months), respectively. All clinicopathologic features were significantly associated with the recurrence of prostate-specific antigen (P < 0.001), except for age (P = 0.821) and body mass index (P = 0.568).

Allele and genotype analyses for the associations of 20 prostate cancer susceptibility single-nucleotide polymorphisms with prostate-specific antigen recurrence after radical prostatectomy are shown in Table 2. We found the allelic frequency of rs1447295 at 8q24, rs7920517, and rs10993994 at 10q11 in patients with recurring prostate-specific antigen to be significantly different from those without recurrence. The rare alleles of these three single-nucleotide polymorphisms were associated with increased risk for prostate-specific antigen recurrence [age-adjusted odds ratios of 1.59 (95% CI, 1.08-2.35) for rs1447295, 1.59 (95% CI, 1.14-2.21) for rs7920517, and 1.66 (95% CI, 1.19-2.31) for rs10993994], and all had a q value < 0.127. In addition, the rare homozygous genotype carriers of these three single-nucleotide polymorphisms were also found to be at greater relative risk for prostate-specific antigen recurrence than in common homozygous carriers [age-adjusted odds ratios of 3.50 (95% CI, 1.10-11.2) for rs1447295, 2.50 (95% CI, 1.27-4.89) for rs7920517, and 2.71 (95% CI, 1.39-5.29) for rs10993994], and all had a q value < 0.170.

The associations between genotypes of these three single-nucleotide polymorphisms and time to disease recurrence were presented in Fig. 1. Kaplan-Meier survival analysis and log-rank test revealed a significant association between rs7920517 and rs10993994 genotypes and prostate-specific antigen–free survival. The median estimated cumulative survivals were significantly lower in rare homozygous carriers than in those common homozygous carriers (32 versus >96 months, P = 0.005, and q = 0.097 for rs7920517; 29 versus >96 months, P = 0.002, and q = 0.076 for rs10993994), showing an earlier recurrence of prostate cancer after radical prostatectomy.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Kaplan-Meier analysis of time to prostate-specific antigen recurrence after radical prostatectomy, stratified by genotypes at rs1447295 (A); rs7920517 (B); and rs10993994 (C).

To confirm the predictive effects of rs1447295, rs7920517, and rs10993994 on prostate-specific antigen recurrence after radical prostatectomy, various clinicopathologic parameters, including age at diagnosis, preoperative prostate-specific antigen, Gleason score, pathologic stage, and surgical margin, were evaluated together with each of these three single-nucleotide polymorphisms, using Cox proportional hazards analysis (Table 3). Our univariate analyses found that high preoperative prostate-specific antigen level, Gleason score 8 to 10, advanced pathologic stage, positive surgical margin, and rare homozygous genotypes of rs7920517 and rs10993994 significantly influenced post–radical prostatectomy prostate-specific antigen–free survival time. After adjusting for all clinicopathologic risk factors in the multivariate analyses, the rare homozygous genotypes of rs1447295, rs7920517, and rs10993994 were further identified as independent prognostic factors for the recurrence of prostate-specific antigen in patients receiving radical prostatectomy (P < 0.01).

Discussion

Biochemical failure (prostate-specific antigen recurrence) is an important indicator of disease recurrence after radical prostatectomy, and many prostate-specific antigen recurrent patients are prone to develop metastatic lesions accompanied by increased risk for mortality (21). Therefore, finding new biomarkers that could predict the recurrence of prostate-specific antigen may make it possible to detect disease recurrence earlier, select treatment strategies, and possibly uncover the underlying mechanism behind the recurrence of the disease. In our investigation on the relationship between 20 single-nucleotide polymorphisms recently highlighted by several genomewide association studies and prostate-specific antigen recurrence, we found that rs7920517 and rs10993994, which were in strong linkage disequilibrium (r² = 0.91), at 10q11 showed significant association with poor prostate-specific antigen–free survival in Kaplan-Meier survival analysis and in multivariate Cox proportional hazards analysis after considering other clinicopathologic risk covariates (Fig. 1 and Table 3). To the best of our knowledge, this study may be the first report of an association between the 10q11 single-nucleotide polymorphisms and prostate-specific antigen recurrence after radical prostatectomy.

Recently, rs10993994, located at 57 bp upstream of the transcription start site of the β-microseminoprotein (MSMB) gene, has been associated with the risk for prostate cancer in two independent genomewide association studies (7, 9). The MSMB gene encodes prostatic secretory protein 94, which is synthesized by prostate epithelial cells and then secreted into the seminal fluid and blood (23). Prostatic secretory protein 94 and prostate-specific antigen, two of the most abundant proteins secreted from the prostate, are thought to be possible markers for prostate cancer (24). More recently, prostatic secretory protein 94 has been found to circulate in blood with low–(free prostatic secretory protein 94) and high (bound to prostatic secretory protein 94– binding protein)–molecular weight forms (25). In the prostate cancer patients who have undergone radical prostatectomy, the bound/free prostatic secretory protein 94 ratio was positively associated with the risk for recurrence after adjusting for prostate-specific antigen, Gleason score, and margin status (26). Furthermore, MSMB is thought to be a tumor suppressor (27) possibly capable of impeding prostate cancer growth, promoting apoptosis, inhibiting the secretion of matrix metalloproteinase implicated in tumor metastasis, and decreasing vascular endothelial growth factor-mediated vascularization.

In addition to the previous genomewide association studies association results, the prostate cancer risk allele T of rs10993994 has been reported to greatly reduce the tumor suppressor MSMB promoter activity in the follow-up functional analyses (28-30). Replacing the T allele with the C allele destroys the binding site of cAMP response element binding protein, and the mean mRNA expression level of MSMB with the T allele is significantly lower than that with the C allele (30). These findings suggest that rs10993994 is functionally important and might partially account for the observed association with prostate cancer susceptibility. In contrast, we estimate that men with two rare allele C at rs10993994 are 2.5 times more likely to experience disease recurrence than those with no rare alleles in our study population (Tables 2 and 3). Our findings are not consistent with the notion that C alleles protect against prostate cancer. There are several possible explanations for this inconsistency. First, prostate cancer susceptibility loci have not been confirmed in multiple populations, including people of Han Chinese origin. If rs10993994 is a prostate cancer risk single-nucleotide polymorphism in men of Han Chinese ancestry, it is possible that this risk single-nucleotide polymorphism might also influence other genes besides MSMB and account for the disparate effects. The rs10993994 is also located at 23 kb upstream of an interesting prostate cancer candidate gene, nuclear receptor coactivator 4, providing another biological relevance for this genetic association. Nuclear receptor coactivator 4, also known as the androgen receptor coactivator ARA70, is an androgen receptor–associated protein and enhances the transcriptional activity of androgen receptor in human prostate cancer cells in a ligand-dependent manner (31). ARA70-induced androgen receptor transactivation could result in the decreased apoptosis, increased cell proliferation, promoted cell invasion, and facilitated tumor progression in prostate cancer cells (32, 33). Thus, genetic variation at rs10993994 might influence oncogene and tumor suppressor in a context-dependent manner. If rs10993994 is not a prostate cancer risk single-nucleotide polymorphism in men of Han Chinese ancestry, it is possible that Han Chinese and European men do in fact share a true prostate cancer risk variant at 10q11. However, the true prostate cancer risk variant is in the strong linkage disequilibrium with rs10993994 in European men but not in Han Chinese. Fine mapping of 10q11 has suggested a recombination hotspot immediately telomeric of the rs10993994 (29), indicating the possibility of the separation of disease risk and recurrence risk loci. Second, results on associations of rs10993994 with clinicopathologic features of prostate cancer were inconsistent. Although some studies have found a higher frequency of the prostate cancer risk allele T in patients with more aggressive prostate cancer (9, 34), others have not (14, 35). Third, rs10993994 has not been significantly associated with other clinicopathologic characteristics of prostate cancer, such as age at diagnosis, Gleason score, pathologic stage, or prostate cancer–specific survival (13, 14, 34-36). Finally, although it has been suggested that MSMB/prostatic secretory protein 94 is a tumor suppressor, increased intratumoral expression of prostatic secretory protein 94 also seems to be associated with worse survival outcomes (37, 38). Therefore, further investigation is needed to determine the role of rs10993994, MSMB/prostatic secretory protein 94, and ARA70 in the etiology of prostate cancer.

The rare allele A of rs1447295 at 8q24, also the reported risk allele, is associated with a significantly increased risk for prostate-specific antigen recurrence in logistic regression and multivariate Cox proportional hazards models (Tables 2 and 3). However, this result should be consider with caution because our study only had a limited number of rs1447295 rare homozygous carriers (n = 14). Chromosome 8q24 has been found to contain several single-nucleotide polymorphisms associated with risk for prostate (2), colorectal (39), breast (40), ovarian (41), and bladder (42) cancers, but no genes have been identified in the region of interest to date. Therefore, further fine mapping of 8q24 in multiethnic samples might help in the identification of the strongest markers in each population and possibly add to our understanding of the contribution of 8q24 single-nucleotide polymorphisms to prostate cancer progression.

In conclusion, systematically evaluating 20 recently highlighted prostate cancer susceptibility single-nucleotide polymorphisms, we provided the first evidence for the association of these variants and recurrence of the disease. However, this study is limited by sample size in analyses of outcomes and in subset analyses. In addition, our homogeneous Chinese Han population may make our findings less generalizable to other ethnic groups. Thus, further functional analyses and large independent studies in other ethnic populations are required to validate the relevance of the observed associations to the progression of prostate cancer after radical prostatectomy.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.