NEW ORLEANS – A urine assay test can provide a high negative predictive value to rule out the presence of high-grade, clinically significant prostate cancer without the need for a digital rectal exam or prostate massage, according to a study presented by Dr. James McKiernan at the annual meeting of the American Urological Association.
“There is an unmet medical need in prostate cancer biomarker development, and new biomarkers should be designed to detect high-grade prostate cancer, not all prostate cancer,” said Dr. McKiernan, director of urologic oncology at Columbia University Medical Center in New York.
Meg Barbor/Frontline Medical News
Dr. James McKiernan
He and his colleagues designed a novel noninvasive urine exosome gene expression assay (EXO106) and evaluated its performance in testing for high-grade prostate cancers (CaP) among a group of male patients.
Unmet need in prostate cancer
“The role of PSA [prostate-specific antigen] screening in the United States is controversial, with a high false-positive rate,” said Dr. McKiernan. Over 1 million prostate biopsies are performed annually in the United States, and the vast majority of them reveal low-grade or very-low-risk prostate cancer.
Serious complications associated with prostate biopsy, including hospitalization and infection, are increasing, and unnecessary or potentially unnecessary treatments for low-risk disease continue at a relatively high rate.
The role of exosomes in EXO106
Dr. McKiernan and his colleagues sought to utilize exosomes in the development of a urine assay test to predict high-grade CaP. Exosomes are lipid bilayer-protected vesicles, which makes them stable under varying conditions and protects them from degradation: they contain RNA, DNA, and protein.
“Exosomes exist in all cells, both malignant and benign, and are believed to be involved in intracellular communication both locally and at a distance,” Dr. McKiernan said.
The EXO106 assay is a simple voided urine test developed over the past 4 years; it does not require a digital rectal exam (DRE) or prostate massage. Exosomes were separated from the urine tested in this population by a process of ultrafiltration, and then the expression level of genes of interest was determined.
The EXO106 assay relied on an expression level of three genes – ERG qPRC, PCA3 qPCR, and SPDEF qPCR. Multivariate analysis was performed to correlate the expression level in the exosomes of these three genes with the presence or absence of high-grade cancer on prostate biopsy.
Study design
In this study, the intended use of the EXO106 risk score was to predict the presence of high-grade (Gleason Score [GS] ≥ 7) prostate cancer for men aged 50 years or older with a PSA between 2 and 10 ng/mL, with either a normal or abnormal DRE, who were referred for their first-ever prostate biopsy – this was not a rebiopsy population. First catch, non-DRE, random urine was collected at all sites, kept stable without chemical preservatives, and then shipped to a central lab for analysis.
The objective of this national trial was to evaluate whether the use of this assay could improve upon the standard of care (SOC) to detect high-grade prostate cancer. The SOC in this study was defined by typical clinical practice based on PSA, age, race, and family history.
“The primary study endpoint was to calculate the area under the receiver operating characteristic curve (AUC) for the combination of the EXO106 assay and the standard of care, and to determine if this was significantly better than the standard of care alone at detecting high-grade prostate cancer [(AUC (EXO106 + SOC) > AUC (SOC)],” Dr. McKiernan explained.
The secondary endpoint was to establish a binary cut-point to determine a normal and abnormal level for the EXO106 assay and then to interrogate the negative predictive value (NPV) of this binary cut-point.
The study evaluated 519 patients (median age, 63; median PSA, 5.12 ng/mL), of whom 17% were African-American, 82% had a normal DRE, and 23% had a family history of CaP. Nearly half (48%) of the biopsies overall were positive, and 28% of biopsies in the cohort were positive for high-grade CaP (GS ≥ 7).
Results
The PSA alone was found to be a relatively noneffective discriminator for high-grade CaP, with an AUC of .545, with .50 indicating absolute lack of discriminating power.
The EXO106 assay alone had a .711 AUC by itself and .725, compared with the SOC. “This met the primary objective as being significantly better than the SOC alone for determining the presence or absence of high-grade prostate cancer,” Dr. McKiernan said.
The binary cut-point resulted in an NPV of 91%. “So if a patient underwent this assay alone and the assay was normal, there was a 91% chance that they did not have high-grade prostate cancer,” he explained. “If this was applied in routine clinical practice in this cohort, it would reduce the unnecessary biopsy rate by 26%.”