Established Prostate Cancer Susceptibility Variants are not Associated with Disease Outcome

Introduction

Recently, several single nucleotide polymorphisms (SNP) consistently replicated in multiple study populations (1-10) have been implicated in prostate cancer etiology by genome-wide association studies (11-19). In contrast, explorations of these variants with respect to clinicopathologic features of the malignant phenotype have been inconsistent. Genetic variants showed a moderate trend of increasing frequency across increasing degrees of disease aggressiveness in some (1-5, 11, 13, 15) but not in other studies (6, 7, 12, 14). We recently did a systematic evaluation of 15 confirmed risk variants in a cohort of 1,563 cases who underwent radical prostatectomy for the treatment of prostate cancer revealing no association with pathologic stage or grade (20).

To this end, the importance of these risk variants in prostate cancer prognosis remains unknown. We therefore explored all confirmed prostate cancer risk variants reported to date with respect to cancer-specific mortality in a Swedish population–based case-control study of prostate cancer etiology.

Materials and Methods

Study Cohort

Cancer Prostate in Sweden is a population-based case-control study of prostate cancer etiology. The study design has been described in detail elsewhere (10). Briefly, cases were all men between 35 and 79 years of age with a histologically or cytologically verified adenocarcinoma of the prostate (ICD-10: C61) between 2001 and 2003. Clinical characteristics for all patients were retrieved through linkage to the National Prostate Cancer registry (21). Study participants donated blood on average 4.9 mo (range, 0.7-23.7 mo) after the date of diagnosis and serum was stored at −70°C until analysis. For the present study, DNA samples from blood were available for 2,875 case patients. All participants gave written informed consent and the Research Ethics Committees at Karolinska Institutet and Umeå University approved this investigation.

Follow-up Assessment

With the use of each study participant's unique national registration number, complete follow-up regarding vital status was assessed from date of blood draw up until January 15, 2008 through record linkage to the Swedish Population Registry. Determination of cause of death was obtained through linkage with the Cause of Death Registry up to December 31, 2005. Review of death certificates, done by an oncologist, established cause of death for individuals deceased after December 31, 2005.

SNP Selection and Genotyping

We selected 16 SNPs implicated in four genome-wide association studies and replicated in at least one independent study in the original reports (11-19). They included five SNPs at 8q and 17q (three separate subregions at 8q24, and one region each at 17q12 and 17q24.3) and one SNP each from 2q15, 3p12, 6q25, 7p15, 7q21, 9q33, 10q11, 10q26, 11q13, 19q13, and Xp11. These 16 SNPs were genotyped among all patients using a MassARRAY QGE iPLEX system (Sequenom, Inc.). The average genotype call rate was 98.3% and the average concordance rate among duplicated control samples was 99.8%. Each of the SNPs in the autosomal chromosomes was in Hardy-Weinberg equilibrium (P ≥ 0.05).

Statistical Methods

To test the association between these 16 risk variants and fatal prostate cancer, we did time-to-event analyses using death from prostate cancer as outcome. Survival time was censored at the time of death for patients dying from causes other than prostate cancer. Association between genotypes and prostate cancer death was assessed in Cox regression models adjusted for age, clinical stage, pathologic grade, nodal or distant metastases, and diagnostic prostate-specific antigen level. For each risk variant, an additive genetic model was assumed. All P values reported were based on two-sided hypotheses.

Results

The clinical characteristics of the study population are presented in Table 1 . More than 40% of the patients were diagnosed with an advanced cancer (defined by meeting at least one of the following criteria: T₃/T₄, N+, M+, Gleason score of 8 to 10 or prostate-specific antigen level of ≥50 ng/mL). Overall, 626 (21%) of the 2,875 patients died during follow-up, and of those, 440 (15%) had prostate cancer classified as their underlying cause of death. The average follow-up time among nondeceased patients was 4.9 years (range, 3.7-6.8 years).

No significant association between disease outcome and single risk variants were observed; only one SNP (rs6465657) on chromosome 7q21 showed nominal association with prostate cancer–specific survival with the reported risk allele (C) associated with decreased prostate cancer–specific survival (hazard ratio, 0.87; 95% confidence interval, 0.75-1.00; P = 0.05; Table 2 ). However, this association was not significant after applying a Bonferroni correction to adjust for 16 independent tests (P = 0.003 required to reach a 5% study-wide significance level). Exploring genotype-specific association with disease outcome did not reveal any significant association with prostate cancer outcome (data not shown). Only heterozygous carrier of the rs6465657 SNP showed nominal association with prostate cancer–specific survival (hazard ratio, 0.73; 95% confidence interval, 0.58-0.92, P = 0.007); however, this finding was not study-wide significant.

Finally, we explored the cumulative effect of these 16 variants on prostate cancer mortality. The average number of risk alleles carried in the study cohort was 13 (note that each subject could carry a maximum of 31 risk alleles). Using this as a reference group, we observed no study-wide significant association with prostate cancer–specific mortality either among men carrying fewer than 13 risk alleles or among men carrying more than 13 risk alleles (Table 3 ). A trend test for increasing number of risk alleles carried revealed no significant association with prostate cancer mortality (hazard ratio, 0.94; 95% confidence interval, 0.91-0.98; P = 0.07).

Discussion

In this prospectively followed cohort of patients with incident prostate cancer, we observed no evidence in support of the hypothesis that prostate cancer risk variants discovered to date from genome-wide association studies are associated with the lethal potential of prostate cancer. Consistent with previous attempts to explore the association between reported prostate cancer susceptibility variants and clinicopathologic features, our findings indicate that these SNPs are only informative for the initiation of prostate cancer, as opposed to the prognosis of this disease.

Considering the adverse effects associated with radical therapy (22), and that most men affected with prostate cancer will die with—rather than from—this disease (23), there is an urgent need for improved tools to distinguish lethal from indolent disease at diagnosis. Identification of genetic markers associated with lethal prostate cancer hold promise to advance the understanding of the biological aspects behind the metastatic potential of prostate cancer.

It is important to note that failure to detect the association between established prostate cancer susceptibility variants and disease outcome does not imply a lack of such genetic variants in the genome. Recent epidemiologic findings support a genetic component in cancer prognosis (24, 25), and it is probable that hitherto performed efforts to discover prostate cancer susceptibility variants—using case-control designs—are suboptimal in identifying genetic variants associated with the prognosis of this disease. Genome-wide efforts exploring disease progression and fatal outcome in cohorts of patients affected by prostate cancer, as compared with prostate cancer case-control studies, should be more efficient to identify genetic markers associated with the prognosis of prostate cancer.

In conclusion, we observed no association between established prostate cancer susceptibility variants and cancer-specific mortality. Future efforts prospectively assessing disease outcome in prostate cancer cohorts should be more efficient to identify genetic variants associated with the prognosis and cancer-specific mortality of prostate cancer.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.