Introduction

To date, more than 300 susceptibility loci associated with prostate cancer (PCa) have been identified mainly from studies of European ancestry populations with much less contributions from Asian populations. As there was a genetic difference according to ethnic groups, we performed the first Korean population based exome wide association study for PCa in this study.

Methods

Our study was performed in a two-stage design._x000D_ Stage I – 1,001 PCa patients using a custom HumanExome BeadChip v1.0 (Illumina Inc) containing 242,186 SNPs. After quality control, genotype data from 988 PCa patients were compared with 2,641 normal controls from community cohort (KARE). _x000D_ Stage II – 5 exome-wide significant lead SNPs (P < 1.0 × 10-4) from the Stage I were selected to genotype and replicated in independent cohort (514 PCa and 548 control). Genotyping of these SNPs was performed using the Fluidigm 192.24 Dynamic Array TM IFC and Biomark HD systems. To construct a genetic risk score (GRS), the cumulative number of risk alleles was calculated using an additive model. Then, we obtained estimates of the area under the receiver-operating characteristic curve (AUC) to evaluate its discriminative ability._x000D_

Results

Stage I – we detected 5 lead variants (p < 8.30E-07) – rs1512268 (8p21, NKX3-1), rs1016343 (8q24, PRNCR1), rs7837688 (8q24, CASC8), rs7501939 (17q12, HNF1B) and rs2735839 (19q13, KLK3) (figure 1)_x000D_ Stage II – mean GRS score was 4.23 ± 1.44 for the controls and 4.78 ± 1.43 for the cases. The GRS was associated with increased PCa risk (GRS 6 : OR=1.85, GRS 7 : OR=2.11 in reference of GRS 4) (figure 2). The trend of increase in PCa risk according to the increase in GRS is very strong (p<0.0001, test for trend). The predictive accuracy of GRS was 0.605 (95%CI 0.573-0.637). _x000D_

Conclusions

We detect 5 risk variants for PCa in Korean population, confirmed in many previously reported findings and build the GRS to predict PCa more effectively. Our findings suggest that similarities than difference in genetic susceptibilities might be more important aspect across populations

Funding

None