Chiam et al. [1] stated that prostate cancer (PCa) is a major global health problem that imposes a significant economic burden in nations with an aging population. The annual percentage change (APC) of the incidence of PCa in Korean men was 13.7% from 1999 to 2009, and APC of mortality rates was 17.5% from 1999 to 2002 [2]. The widespread use of prostate-specific antigen (PSA)-based screening testing (PSA-ST) leads to an increased incidence of PCa because it enables the earlier detection of occult or asymptomatic disease [3-5].

As PSA is not a specific marker of PCa [1], recommendations on PSA-ST for PCa vary in terms of the screening age and interval [6,7]. Of note, the 2012 U.S. Preventive Services Task Force guideline [8] recommended against routine screening for PCa, because the benefits of PSA-ST for PCa do not outweigh the harms.

The harms of PSA-ST can be summarized as overdiagnosis, unnecessary biopsies with potential associated adverse effects, anxiety, and excessive treatment [7,9,10]. As such, the most serious limitation of PSA-ST as a screening modality is the fact that PSA levels can be elevated in patients with benign prostatic hyperplasia or prostatitis, as well as in PCa patients [7,11]. This phenomenon may give rise to overdiagnosis, resulting in overtreatment [1,6,12,13]. In addition to this, PSA-ST has very poor sensitivity, specificity, and predictive values because there are no absolute cutoff PSA levels defining PCa [1,13]. Thus, Lee et al. [14] concluded that PSA-ST alone did not increase earlystage PCa detection or reduce mortality.

To overcome these limitations of PSA-ST, PSA velocity [15], testing for 4 prostate-specific kallikreins [3], the prostate health index test [16], the percentage of free PSA [17], and tests for noncoding prostate-tissue-specific RNA [18] have been introduced. However, these PSA derivatives may be impractical or only helpful in specific situations [1,7]. Thus, novel biomarkers capable of replacing serum PSA for PCa screening must be investigated [19-22]. In addition, reliable and accurate biomarkers for discriminating between indolent and aggressive tumors at the early stage of PCa are needed [23].

As age, race, and environment are known to be the main risk factors for PCa, epigenetic studies investigating the carcinogenesis of PCa through gene-environment interactions have been conducted [1,24]. Current evidence suggests that epigenetic alterations of aberrant DNA methylation, histone modifications, and noncoding microRNA are associated with the carcinogenesis of PCa [25-28]. Thus, potential biomarkers related to a high frequency of epigenetic changes may improve the sensitivity and specificity of the diagnosis (including early detection) and prognosis of PCa [1,13,25,27,29].

Chiam et al. [1] tabulated the epigenetic biomarkers associated with the diagnosis, prognosis, and treatment response of PCa. Furthermore, Yegnasubramanian [13] suggested that methylation in the regulatory regions of GSTP1, APC, PTGS2, RARB, and RASSF1A may be epigenetic biomarkers for PCa screening. In particular, measurements of GSTP1 promoter methylation in plasma, serum, whole blood, urine, ejaculate, or prostatic secretions may complement PSA-ST for PCa based on a meta-analysis of 22 studies [30]. However, all those studies were case-control studies with a small sample size. Thus, a population-based cohort study in asymptomatic men with a large sample size is needed to evaluate the effectiveness of GSTP1 for the early detection of PCa and/or the identification of aggressive tumors.

In conclusion, the controversies regarding PSA-ST have led to the need for a more accurate biomarker suitable for the early detection of PCa [31]. This unmet need could be satisfied by epigenetic biomarkers related to the pathogenesis of PCa [13,29].

However, potential epigenetic markers require further research to be validated for screening in diverse populations [25,32]. Further studies may lead to the development of epigenetic markers that could replace, rather than complement, PSA-ST due to advantages in sensitivity.