An exosomal messenger RNA-based urine diagnostic demonstrated high negative predictive value for high-grade prostate cancer.
James M. McKiernan, MD
An exosomal messenger RNA (mRNA)-based urine diagnostic demonstrated high predictive value for high-grade prostate cancer, according to a report at the 2015 American Urological Association Annual Meeting annual meeting.
The assay, labeled EXO106, demonstrated significantly better diagnostic performance compared with prostate-specific antigen (PSA) testing alone or PSA plus standard clinical characteristics. In the study, the test missed fewer than 5% of higher-grade Gleason score (GS) ≥4+3 cancers, researchers indicated. If utilized, the assay could result in a 27% reduction in prostate needle biopsies.
“Exosomal mRNA can be isolated and analyzed, using a first-catch, random urine,” said principal investigator James M. McKiernan, MD, a professor of Urology and director of Urologic Oncology at NewYork-Presbyterian Hospital/Columbia University Medical Center in New York. “The EXO106 provides a high negative predictive value for high-grade cancer.”
Exosomes are secreted by virtually all cells into biological fluids, as a means of cellular communication. They are lipid bilayer-protected vesicles that remain stable under varying conditions and afford protection for their contents against degradation. Exosomes contain multiple forms of RNA, as well as DNA and protein, said McKiernan.
The structural and molecular characteristics of exosomes lend themselves to potential use as an assay to diagnose high-grade prostate cancer. EXO106 is a urine-based liquid biopsy test that assesses expression of three genes on exosomal RNA:ERG, PCA3, and SPDEF.
Expression patterns undergo evaluation by a multivariate algorithm to predict the presence of high-grade prostate cancer (GS >6), with results reflected in an EXO106 risk score.
Overall, the EXO106 test missed 12 of 148 cases of high-grade (GS ≥7). However, GS 4-predominant pattern (4+3) accounted for three of the false-negatives, and the lower-risk 3+4 pattern associated with three or fewer positive biopsy cores accounted for the remaining false-negatives.
When the definition of high-grade disease was limited to GS ≥4+3, the assay had a false-negative rate of <5%. The EXO106 assay also detected biopsy-confirmed high-grade prostate cancer in more than 90% of cases.
McKiernan reported findings from a multicenter evaluation of EXO106. Investigators at 22 centers in the United States enrolled 1560 patients with GS >7 prostate cancer. After exclusion of 9% of urine samples, the study population comprised a training set of 499 patients and a 519-patient intended-use population.
The study included men 50 or older with no previous prostate biopsy, PSA level of 2 ng/mL to 10 ng/mL, and a scheduled biopsy for suspected prostate cancer. Men receiving treatments known to influence PSA levels were excluded. Assay results were based on a first-catch, randomize urine not associated with digital rectal exam (DRE). Samples were shipped to a central laboratory for analysis.
The primary objective was to determine whether the EXO106 assay added to standard of care (PSA, age, race, and family history for prostate cancer) would result in an area under the curve (AUC) superior to that of standard of care. The secondary objective was the binary performance of the EXO106 at a predefined cutpoint (sensitivity, specificity, negative predictive value [NPV], and positive predictive value [PPV]).
The 519-patient primary study population had a media age of 63 and a median prebiopsy PSA level of 5.12 ng/mL. DRE was suspicious in 18% of cases, and 23% of the men had a positive family history.
Biopsies involved a median of 12 cores. The biopsy result was positive in 48% of cases. Biopsies showed GS 6 prostate cancer in 20% of cases, GS 3+4 in 16%, GS 4+3 in 7%, GS 8 in 2%, and GS 9 in 3%. Overall, 28% of the biopsies showed GS ≥3+4.
Analysis of performance characteristics showed that PSA alone had an AUC of 0.545, increasing to 0.631 for standard of care. The AUC increased to 0.711 for EXO106 alone and to 0.725 for EXO106 plus standard of care. The trial met its primary endpoint by demonstrating significantly better diagnostic performance with the addition of EXO106 to standard of care compared with the standard of care (P<.00004).
Analysis of the EXO106 by performance of the prespecified risk score cutpoint showed a sensitivity of 91.89%, specificity of 33.96%, PPV of 35.70%, and NPV of 91.30%.
Exosome Diagnostics, Inc, the Massachusetts-based company developing the assay, said further outcomes and economic studies are planned for this year with the goal of launching the test commercially in 2016. The company also intends to seek FDA approval for an in vitro diagnostic version of the test.
Full integration of the assay into the current prognostic standard of care before the initial biopsy could save approximately $166 million per year for national health plans and about $29 million per year, Exosome Diagnostics maintains.
McKiernan J, Donovan M, O’Neill V, et al. Validation of a novel non-invasive urine exosome gene expression assay to predict high-grade prostate cancer in patients undergoing initial biopsy with an equivocal PSA. Presented at: AUA Annual Meeting; May 15-19, 2015; New Orleans, LA. Abstract PII-LBA2.