Prostate cancer is the most common malignancy in men in the U.S. and it is the third leading cause of cancer death among men. According to the American Cancer Society, more than 180,000 new cases of prostate cancer are diagnosed each year in the US with 26,000 men succumbing to the disease 1-5. For unknown reason, it occurs more often in African American men then in any other group in the U.S. African American men have the highest risk and are more likely to be diagnosed at an advanced stage of prostate cancer than European American men. This disparity persists after adjusting for clinical factors 1,4-5. While prostate screening is recommended for older men, the screening current tool (PSA) is controversial due to its reliability in prostate cancer diagnosis and its inability to discriminate between clinically relevant (aggressive) or clinically indolent cancer.
The Prostate-specific antigen, or PSA test measures the level of PSA protein in a man’s blood 1-7. Because of these issues, there is need a for improved screening that can detect and discriminate between clinically relevant and clinically indolent cancer. A potential way of improving screening is by analyzing of germline DNA which could identify men who are susceptible to clinically relevant prostate cancer. Research has shown that many aggressive cancers including prostate cancer has a strong genetic heritability with as much as 58% of prostate cancer risk being attributable to genetic factors 1,7-12. Thus, the germline DNA analysis may screenings for clinically relevant prostate cancer by analyzing rare variants.
While many variants have been associated with prostate cancer risk, few are related to aggressive disease. Genome wide association studies (GWAS) have identified multiple loci and variants that influence prostate cancer risk, many of these prostate cancer cases were diagnosed with less aggressive (clinically indolent) tumors. These loci associations, however do not factor in aggressive disease heritable like rare variants. Rare variants (variants with minor allele frequency (MAF <0.05) account for as much as 42% of prostate cancer risk 1,13-19. However, GWAS doesn't typically include rare variants, and focuses on more common variants (MAF >0.05). Hence, identifying rare variants associated with aggressive prostate cancer risk could help reduce mortality rates . The goal of this study was to identify rare variations associated with clinically relevant prostate cancer in two ethnic populations: African Americans (AAs) and European (EAs).
Samples were collected from two institutions: Washington University in St. Louis, Missouri (WU) and Johns Hopkins Medical Institutions in Baltimore, Maryland (JHMI) table 1. WU samples comprised 272 cases (150 EAs & 122 AAs) and 300 controls (150 EAs &150 AAs), and JHMI samples comprised 384 cases (305 EAs & 79 AAs) and 463 controls (305 EAs & 158 AAs). The cases were men diagnosed with aggressive prostate cancer defined as evidence of metastatic disease (either pathologic or radiologic), a PSA > 50ng/ml, or Gleason score of 8-10. No one met the criteria for familial prostate cancer or had a history of a familial tumor syndrome.
Ancestry-matched controls were recruited from both sites who had a minimal risk of developing clinically relevant prostate cancer as they had no personal or family medical history. Both JHMI and WU controls were over the age of 75 years or 65 years for the EAs and AAs, respectively, however JHMI age matched their controls to cases. All study subjects provided informed consent under a protocol approved by Human Research Protection Offices, and genomic DNA was prepared from peripheral blood samples. Genes were selected for this study using two approaches: (1) utilizing genes that were already implicated in prostate cancer from numerous sources (listed in PowerPoint), and (2) utilizing novel prostate genes that were detected during the discovery whole-exome sequencing phase of a subset of the study cohort. All these genes were analyzed to ensure that met the rare variants criteria (MAF<0.05), and 800 were selected for this study. The samples genomic DNA underwent target sequencing with the first stage examining the whole exome using the WU cohort and then the second stage examining 800 promising candidates using custom targeted capture using the JHMI cohort Figure 1. Libraries were prepped according to manufacturer's specifics, and were sequenced on an Illumina HiSeq 2000. 14 Samples that did not meet the minimum coverage threshold of >=20X target base coverage and >=10X exome data prior to analysis. All samples that underwent exome sequencing were genotyped by Affymetrix 600K SNP to help AA and EA samples. All other samples were mapped to GRCH37 reference using BWA-MEM. Data was merged and duplicates were marked using Picard tools, and Samtools V1.16 was used to call variants. Rare variants (MAF <0.05) were annotated using Variant Effect Predictor (VEP), and two aggregation tests (Fisher's exact and SKAT-O) with a threshold of 3.125 x 10-5 was used for statistical analysis. Results found a total of 56 genes that showed suggestive association (P<0.01) and enrichment of rare variants in one of the statistical test. The combined analysis of both exome and targeted sequencing data for the 800 selected genes among the 497 African-Americans (200 cases, 297 controls) and 907 European-Americans (452 cases, 455 controls) found that only one gene meet the 3.125 × 10?5 (0.05 / 1,600 tests) study threshold: the TET2 gene Figure 3. In AA, 24.35% of cases harbored a rare coding variant in TET2 gene compared to 9.61% of controls (FET p = 1.84×10?5, OR=3.0; SKAT-O p= 2.74×10?5). Several Studies have implicated the TET2 gene in different cancer types including prostate cancer 1, 11-17. The PARP2 gene did show a strong association but did not meet the threshold as did BRCA2 in EAs Figure 4. The study also found evidence that the BRCA1, BRCA2, MSH6, & PARP2 in EA are possible cancer predisposition genes which adds to the growing body of evidence that implicates this genes that are in the DNA repair pathway 1,18-21.