Abstract

Abstract Background: African-American (AA) men are more often diagnosed with early-onset aggressive prostate cancer than European American (EA) men. Prostate cancer health disparities between AA and EA men are attributed to socioeconomic as well as biologic differences existing between the two groups. The purpose of this study is to investigate how genetic and epigenetic alterations, specifically DNA methylation, contribute to early-onset aggressive prostate cancer in AA men. Methods: We performed 450K methylation array and RNA-sequencing on radical prostatectomy specimens from AA (n=32) and EA (n=5) men treated at Roswell Park Cancer Institute (RPCI). To increase analytical power, we included AA (n=22) and EA (n=172) patients from The Cancer Genome Atlas (TCGA) database. We found no significant difference in the average age between EA and AA men in the RPCI cohort, but the TCGA cohort consisted of AA men who had significantly lower average and median age compared to EA men (57 and 56 vs 61 and 62; p=0.019). We assessed baseline androgen receptor (AR) protein levels in matched benign and malignant AA (n=95) and EA (n=93) radical prostatectomy tissues from RPCI using immunohistochemistry. Gleason score <3+4 and >4+3 were classified as low and high aggressive tumors, respectively. Results: Unsupervised hierarchical clustering of DNA methylation levels in prostate cancers revealed 9 clusters. We focused on the 2 largest clusters, Cluster 1 (n=133) and Cluster 3 (n=73). Cluster 1 consisted predominantly of low aggressive disease (p=0.00002) with lower serum prostate specific antigen (PSA) values (8.97 vs 13.35 ng/ml) compared to Cluster 3. Following this trend, AA patients, but not EA patients, in Cluster 1 had better overall (57 vs 50 months, p=0.48) and disease-free time (47 vs 22 months, p=0.01) as compared to Cluster 3. ERG (ETS-related gene) fusion-positive prostate cancer is believed to be more aggressive than fusion-negative prostate cancer due to increased ERG gene transcription. We did not observe a difference in fusion status between the two DNA methylation clusters. However, overall DNA methylation was higher in fusion-negative samples as compared to fusion-positive samples in the TCGA dataset. ERG is known to bind to and reduce PSA, a direct downstream AR target. In the TCGA dataset, AA fusion-negative patients had higher average and median preoperative PSA values (18.38, 20.15 ng/ml) as compared with AA fusion-positive (7.51, 6.70 ng/ml), EA fusion-positive (9.43, 7.00 ng/ml), and EA fusion-negative patients (11.47, 7.50 ng/ml). There was no significant difference in the AR transcript and AR protein levels between fusion-positive and -negative AA and EA men. AR protein expression in RPCI cohort tissue microarrays showed that adjacent nontumor tissues from AA men had higher percent AR positive nuclei (p<0.05) compared to adjacent nontumor tissues from EA men. Conclusions: Analysis of the RPCI and TCGA cohorts indicated that DNA methylation separated low and high aggressive prostate cancer in AA men. Further analysis will include more patients as well as identify differences in specific CpG loci between Cluster 1 and Cluster 3. Our results suggest that DNA methylation alterations can be important in patients with ERG fusion-negative disease since these prostate cancers exhibit higher levels of overall DNA methylation. Lastly, our observations of higher AR expression in adjacent nontumor tissues suggest that there are underlying genetic variations that contribute to early-onset high aggressive prostate cancer in AA men compared to EA men. Citation Format: Swathi Ramakrishnan, Xuan Peng, Qianya Qi, Qiang Hu, Gissou Azabdaftari, Elena Pop, James Mohler, Kristopher Attwood, Li Yan, Jianmin Wang, Anna Woloszynska-Read. DNA methylation and genetic alterations contribute to aggressive prostate cancer in African American men [abstract]. In: Proceedings of the Tenth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2017 Sep 25-28; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2018;27(7 Suppl):Abstract nr B72.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call