Incorporating germline genetics into personalised risk-assessment in the detection of prostate cancer in men with an elevated genetic risk

Abstract

Hypothesis - Prostate biopsy outcomes for men with a genetic predisposition to prostate cancer (PrCa) can be predicted from germline genetic profiling and clinical factors. Prostate cancer (PrCa) is the second most common male malignancy worldwide and has a large heritable component. the use of prostate specific antigen (PSA) as a screening test has limitations in both diagnostic accuracy and inability to discriminate between clinically significant and insignificant disease. Given that most men have a low lifetime risk of developing lethal PrCa, a proposed improved screening strategy could target certain populations of men at increased risk due to a genetic predisposition such as those with a family history (GH) of prostate cancer. To date, approximately 170 common variants (SNPs) associated with PrCa risk exist and can be detected by analysing germline DNA. A polygenic risk score (PRS) can assign men a risk category. The use of prostate multiparametric MRI (mpMRI) has become the standard of care in men with a clinical suspicion of PrCa, with its clinical utility proving useful in helping target prostate biopsy towards a diagnosis of clinically significant prostate cancer. The clinical utility of mpMRI in men without a clinical suspicion of PrCa, but at increased risk of PrCa is undefined. The PROFILE study offers up front mpMRI of the prostate and biopsy to men with a FH of PrCa, regardless of PSA in addition to SNP analysis. This thesis examined the association of PRS with biopsy outcome in addition to known clinical risk variables and biopsy outcome. For this thesis, I performed an interim analysis of the PROFILE study participants' mpMRI, PRS and biopsy outcomes. The incidence of PrCa was 30%, occurring across a spectrum of PSA values, but with 70% occurring at a PSA of 3.0ng/ml or less and 38% of significant cancers occurring at a PSA of 3.0ng/ml or less. I found that PRS was associated with cancer detection and when men were categorised into percentiles of risk, those in the top 20% were the most affected. mpMRI performed well in detecting clinically significant prostate cancer, and a PIRADS 1-2 MRI showed high sensitivity at ruling out clinically significant disease. In men with an abnormal mpMRI, their predicted probability of cancer detection changed according to PRS. Overall mpMRI did not appear as specific in clinically significant caner detection compared to PROMIS data, which is important to recognise in the use of PiRADS reporting in young men with a FH of PrCa if used as a risk-stratification or diagnostic tool on its own. Both PRS and MRI have the potential to play an important role in risk-stratifying men with a FH of PrCa and incorporation into targeted screening algorithms.