EAU PCa 2018: Genomics in Prostate Cancer, When and How to Use It

Milan, Italy (UroToday.com) Dr. Waltz and Dr. Gerritsen presented their view on the usage of genomics in prostate cancer in 2018. First, it is important to understand when we should apply genomic testing in prostate cancer. This should be applied in screening, initial diagnosis, initial biopsy, repeat biopsy, classification after biopsy (before initiating active surveillance), and classification after treatment. There are multiple tests available for this purpose, and these were summarized in this discussion.

The first test discussed is the single nucleotide polymorphisms (SNP), measured in the blood nd predicting the diagnosis of prostate cancer This test targets genes associated with prostate cancer. It can either test 16 genes (with an odds ratio of 9.5), or 25 genes (with an odds ratio of up to 31). The area under the curve (AUC) of this test is unfortunately quite low at 0.57, compared to PSA with an AUC of 0.79. Another test discussed is the STHLM3, predicting the diagnosis of prostate cancer. It is a combination of 232 SNPs, PSA, free PSA, intact PSA, HK2, SMB, MIC1, age, family history, digital rectal examination results, and prostate biopsy results. In patients with a PSA >1 ng/ml the AUC is 0.74. This test has been demonstrated to reduce the number of biopsies performed by 32% and reduce the number of biopsies with benign results by 44%.

Next, rare germline mutations were discussed. The BRCA 1+2 mutations were initially mentioned. Men with these mutations have been shown to harbor a 3.8-8.6 higher risk of having prostate cancer, and of higher grade, with increased risk for lymphovascular invasion and higher risk for metastases. The main limitation is the relatively low prevalence in the population with BRCA 1 mutation occurring in 0.44% of prostate cancer patients, and BRCA 2 mutation occurring in 1.2% of prostate cancer patients. PCA3 is a post-digital rectal examination (DRE) urine test, predicting if there need for a repeat biopsy in patients with a previous negative biopsy. It assesses non-coding RNA, and the TMPRSS2-ERG fusion gene transcripts. It has an AUC OF 0.71-0.75. The next test discussed was the SelectMDx test, which is an mRNA expression profile, from post-DRE urine, predicting a result of Gleason >7 in the first biopsy. The gene panel it analyzes includes HOXC6, TDRD1, and DLX1. It as an AUC of 0.77, and when combined with PCA3 test and PSA, it reaches an AUC of 0.81. The Exo Dx Intelliscore test was next discussed. This is a non-DRE urine test, predicting a Gleason >7 in initial biopsy, which analyzes an exosome gene expression assay. Its genomic target is the PCA3 and TMPRSS:ERG. It has a negative predictive value of 0.98, appositive predictive value of 0.35, and an AUC of 0.76. Confirm MDx test uses tissue from the prostate itself (from a biopsy) and predicts the result of cancer at repeat biopsy (in other words, whether the negative biopsy results was actually a false negative). This is a methylation assay quantified in prostate tissue. It assesses epigenetic alteration surrounding tumor lesions (the so-called “Halo effect”). Its genomic targets include GSTP1, APC, and RASSF, and it has a negative predictive value of 0.87-0.9, reducing unnecessary biopsies by approximately 64%. Next, EpiScore test was mentioned, which predicts high grade cancer at repeat biopsy. It is a post-DRE urine test, and includes modified algorithms for the analysis of GSTP1, APC, and RASSF. It has an AUC of 0.74, and a negative predictive value of 0.96. It reduces unnecessary repeat biopsies in 30% of patients. Oncotype Dx test was next discussed. This test predicts high grade cancer (Gleason 4-5) or high stage after biopsy. For this test, prostate tissue which is paraffin embedded, and it analyzes gene expression of 17 genes, including androgen pathway, cellular organization, proliferation, stromal response, and reference genes. Per 20 patients, the test increases the odds ratio increases by 2.3 for high grade disease and it increases the odd ratio by 1.9 for high stage disease. Lastly, Prolaris test was mentioned. This test predicts decision for treatment or active surveillance (AS) in low risk prostate cancer. It requires paraffin-embedded prostate tissue. It analyzes 46 genes, including cell cycle progression and reference genes.

Dr. Walz continued and mentioned the attributes of MRI, as shown in the PROMIS study1, which was a multicenter, paired-cohort, confirmatory study to test the diagnostic accuracy of mpMRI and TRUS-biopsy against a reference test (template prostate mapping biopsy). This study demonstrated that the sensitivity, specificity, positive predictive value and negative predictive value of mpMRI in the diagnosis of clinically significant prostate cancer (Gleason >=3+4=7) disease were 88%, 45%, 65%, and 76%, respectively.

Concluding and summarizing the discussion, the costs of these markers were demonstrated, showing a considerable discrepancy between the various tests. PCA3 costs about $760 US, Oncotype  ~$4000 US dollars, Prolaris ~$3400 , and Exo Dx ~$595. Additionally, Select MDx -250 Euro, Confirm MDx 500 Euro, and mpMRI costs 350 Euro. 

Presented by: J. Walz, Marseille, France and  W.R. Gerritsen, Nijmegen, Netherlands

Written by: Hanan Goldberg, MD, Urologic Oncology Fellow (SUO), University of Toronto, Princess Margaret Cancer Centre, Twitter:@GoldbergHanan at the 2nd EAU Update on Prostate Cancer (PCa18)– September 14-15, 2018 – Milan, Italy

Reference:

1. Ahmed et al. Lancet 2017