Family history of prostate cancer and the incidence of ERG‐ and phosphatase and tensin homolog‐defined prostate cancer

Abstract

Family history is among the strongest known risk factors for prostate cancer (PCa). Emerging data suggest molecular subtypes of PCa, including two somatic genetic aberrations: fusions of androgen‐regulated promoters with ERG and, separately, phosphatase and tensin homolog (PTEN) loss. We examined associations between family history and incidence of these subtypes in 44,126 men from the prospective Health Professionals Follow‐up Study. ERG and PTEN status were assessed by immunohistochemistry. Multivariable competing risks models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for associations between self‐reported family history of PCa and molecular subtypes of disease. Thirteen percent of men had a positive family history of PCa at baseline. During a median follow‐up of 18.5 years, 5,511 PCa cases were diagnosed. Among them, 888 were assayed for ERG status (47% ERG‐positive) and 715 were assayed for PTEN loss (14% PTEN null). Family history was more strongly associated with risk of ERG‐negative (HR: 2.15; 95% CI: 1.71–2.70) than ERG‐positive (HR: 1.49; 95% CI: 1.13–1.95) disease (pheterogeneity: 0.04). The strongest difference was among men with an affected father (HRERG‐negative: 2.09; 95% CI: 1.64–2.66; HRERG‐positive: 1.30; 95% CI: 0.96–1.76; pheterogeneity: 0.01). Family history of PCa was positively associated with both PTEN null (HR: 2.10; 95% CI: 1.26–3.49) and PTEN intact (HR: 1.72; 95% CI: 1.39–2.13) PCa (pheterogeneity: 0.47). Our results indicate that PCa family history may be positively associated with PCa in all ERG and PTEN subtypes, suggesting a role of genetic susceptibility in their development. It is possible that ERG‐negative disease could be especially associated with positive family history.

Analysis of a large prostate cancer family identifies novel and recurrent gene fusion events providing evidence for inherited predisposition

“…A transcriptome‐wide association study by Emami et al 29 observed a positive correlation between imputed TMPRSS2 expression levels in normal prostate tissue (as predicted by germline cis ‐expression quantitative trait loci) and ERG expression in TMPRSS2:ERG fusion‐positive tumour samples 29 . Furthermore, in a prospective study of 888 PrCa cases from the Health Professionals Follow‐up Study, Hashim et al 30 found that family history was strongly associated with ERG‐positive staining in tumour tissue—a surrogate for TMPRSS2:ERG fusion status.…”

“…It thus appears that the frequency of fusion events in PcTas9 is considerably higher than the frequency normally observed in Caucasian tumour cohorts (~40%–50%) 3–6 and suggests an underlying inherited predisposition to the formation of these events in this family. In fact, there is now a strong body of evidence supporting this theory, 3,12–15 including two recent and very diverse studies 29,30 . A transcriptome‐wide association study by Emami et al 29 observed a positive correlation between imputed TMPRSS2 expression levels in normal prostate tissue (as predicted by germline cis ‐expression quantitative trait loci) and ERG expression in TMPRSS2:ERG fusion‐positive tumour samples 29 .…”