Prostate cancer meta-analysis from more than 143,000 men identifies 57 novel prostate cancer susceptibility loci


Session type:

Rosalind Eeles1,Fredrick R Schumacher2,Ali Amin Al Olama3,Sonja I Berndt4,Fredrik Wilklund5,David V Conti6,Mahbubl Ahmed1,Sara Benlloch7,Douglas F Easton8,Peter Kraft9,Stephen J Chanock4,Brian E Henderson (deceased)6,Zsofia Kote-Jarai1,Christopher A Haiman6
1The Institute of Cancer Research,2Case Western Reserve University,3Cambridge Medical School,4National Cancer Institute,5Karolinkska Institute,6University of Southern California,7PRACTICAL Study, Cambridge Medical School,8Centre for Cancer Genetics Epidemiology,9Harvard School of Public Health



Genome-wide association studies (GWAS) have identified >100 prostate cancer (PrCa) susceptibility loci, capturing 35% of the PrCa familial relative risk (FRR) within European-ancestry populations. To further elucidate the missing FRR we conducted the largest known PrCa GWAS of ~47,000 cases and ~28,000 controls among individuals of European descent using the OncoArray, a platform consisting of a 260K GWAS backbone and 310K custom content selected from previous GWAS of multiple cancers.


OncoArray genotype data was imputed to the October 2014 release of the 1000 genomes project in conjunction with several previous PrCa GWAS of European ancestry: UK stage 1 (1,906 cases/1,934 controls) and stage 2 (3,888/3,956); CaPS 1 (498/502) and CaPS 2 (1,483/519); BPC3 (2,137/3,101); NCI PEGASUS (4,622/2,954); and iCOGS (21,209/ 20,440). Risk analyses for overall PrCa risk, aggressive PrCa, and Gleason score were performed. Logistic and linear regression summary statistics were meta-analysed using an inverse variance fixed effect approach. We excluded regions surrounding previously associated PrCa variants (+/- 500kb of the reported index variant) to identify novel associations


The PrCa meta-analysis identified 57 novel loci significantly associated (P<5.0x10-8) with overall PrCa, 3 loci for advanced PrCa (Gleason ≥8, death from PrCa, PSA>100, or disease stage “Distant”), and 3 loci for early-onset PrCa (≤55 years of age). When combined multiplicatively, the 57 novel PrCa loci identified here capture 5.6% of the FRR in the OncoArray samples. In total nearly 41% of the FRR is explained by the novel and previously identified PrCa loci. In risk stratification, men in the top 10% of the genetic risk score group have a relative risk of 2.91 for developing PrCa compared with the average and for the top 1%, nearly 5x that of the average of the population.


These results will improve the utility of genetic risk scores for targeted screening and prostate cancer prevention.