Is genetic testing sophisticated enough to make PSA screening viable for mainstream use?

In the end, we should be advocates and participants in comparative trials to achieve data to provide cost-effective testing combinations and guide recommendations for smarter prostate cancer screening. Population-based scientists should continue to study utilization of these tests in clinical practice and estimate benefits to prostate cancer–specific outcomes in the absence of another large randomized clinical trial in prostate cancer screening.


EA Heijnsdijk, et al. N Engl J Med. 2012;doi:10.1056/NEJMoa1201637.

Konety B, et al. Rev Urol. 2015;17:231–240.

Stattin P, et al. Eur Urol. 2015;doi:10.1016/j.eururo.2015.01.009.

Tomlins SA, et al. Eur Urol. 2015;doi:10.1016/j.eururo.2015.04.039.

Vickers AJ. Urology. 2015;doi:10.1016/j.urology.2014.11.003.

Wei JT, et al. J Clin Oncol. 2014;doi:10.1200/JCO.2013.52.8505.

Behfar Ehdaie, MD, MPH, is a urologic surgeon at Memorial Sloan Kettering Cancer Center. He can be reached at ehdaieb@mskcc.org. Disclosure: Ehdaie reports no relevant financial disclosures.



Although some genetic tests show promise, none has been proven to reduce harms or improve health outcomes. PSA screening for prostate cancer has been an uncontrolled experiment performed annually on millions of older men in the United States since the early 1990s. Only in the past few years have the U.S. Preventive Services Task Force and other medical groups recognized that routine PSA screening does more harm than good, and that targeted testing should occur only after a shared decision-making discussion in which the patient expresses a clear preference to be screened.

Kenneth W. Lin, MD, MPH
Kenneth W. Lin

The search is on to identify more specific biomarkers that can either replace PSA as a screening test, or augment PSA by predicting which men with elevated levels are at the greatest risk for harboring clinically important — and potentially curable — cancers. However, utilizing genetic tests for this purpose outside of clinical trials is premature. The only genetic test for prostate cancer approved by the FDA is the PCA3 urine assay. In 2014, the Evaluation of Genomic Applications in Practice and Prevention Working Group concluded that PCA3 has insufficient supporting evidence to inform decisions to conduct initial or repeat biopsies for prostate cancer in at-risk men.

Other commercial genetic tests or test combinations include urine assays — TMPRSS2-ERG fusion gene, Mi-Prostate score and ExoDx Prostate(Intelliscore) (Exosome Diagnostics) — and multigene prostate tissue assays, such as Oncotype DX (Genomic Health), Prolaris (Myriad Genetics), Decipher Prostate Cancer Classifier (GenomeDx), ProMark (Metamark) and ConfirmMDx (MDxHealth). Obviously, tissue assays are not suitable for determining which men to biopsy, although they may have a role in reducing unnecessary repeat biopsies or treatments.

The Mi-Prostate score — which incorporates serum PSA, urine PCA3 and TMPRSS2-ERG — is limited by the low prognostic value of the fusion gene test and uncertainty regarding cutoff values across populations. ExoDx may predict high-grade prostate tumors, but its prognostic value has not been studied. A 2016 systematic review commissioned by Agency for Healthcare Research and Quality found insufficient evidence to assess analytic validity of 18 commercially — or close to — available multigene panels for prostate cancer risk assessment, evidence of modest clinical validity beyond patient age and family history, and no studies of clinical utility (eg, effects on process of care, health outcomes, harms and economic outcomes).