Early Prostate Cancer Antigen-2: A Controversial Prostate Cancer Biomarker?

Abstract

It is well known that prostate-specific antigen (PSA)1 has both advantages and disadvantages as a marker of prostate cancer. Advantages include its ability to effectively detect early-stage prostate cancer and to monitor disease progression. A disadvantage of PSA is that prostate cancer cells and normal prostate cells both produce PSA; thus, it is frequently increased in nonmalignant conditions such as prostatitis and benign prostatic hyperplasia. The low diagnostic specificity of PSA leads to many false positives and a large number of biopsies in patients who are suspected to have prostate cancer. These well-recognized limitations of PSA suggest that new, and improved, prostate cancer biomarkers could play a useful role in reducing the number of unnecessary biopsies. Over the last 10 years, many candidate prostate cancer biomarkers have been proposed (1). None of them has as yet reached the clinic, owing to their inferiority when compared with PSA. Getzenberg et al. published two provocative reports, one in the journal Urology (2) and a more recent one in the journal The Prostate (3), claiming that a newly discovered prostate cancer biomarker, early prostate cancer antigen-2 (EPCA-2), may be more effective than PSA in detecting prostate cancer, and more accurate in differentiating between localized and extracapsular disease. There are two important differences between the two papers. In the second paper, the authors claim that by using a different antibody epitope on the same molecule, the discrimination between organ-confined and non-organ–confined prostate cancer, which was a major finding in the first paper, is now nonexistent. Also, despite measuring the same protein (but targeting two different epitopes), the claimed optimal cutpoint in the first paper was 30 ng/mL (μg/L), while in the second the cutpoint was 0.5 ng/mL, nearly 2 orders of magnitude lower. Nevertheless, the authors concluded that the new data confirmed their …