Recent progress and perspectives on prostate cancer biomarkers

In 2015, prostate cancer (PC) was the most commonly diagnosed male malignancy, not only in western countries [1], but also in Japan [2]. PSA is the most commonly used biomarker for the early detection of PC. After the introduction of PSA testing, the rate of PC diagnosis increased and PC-associated mortality decreased. Elevated PSA levels are associated with an increased risk of PC, a higher pathological grade, and an increased risk of metastatic disease [3]. However, the use of PSA as a biomarker has a number of limitations. First, PSA is not a PC-specific biomarker. In addition, PSA levels are influenced by several factors, including age, acute prostatitis, ejaculation, catheterization, and certain medications. Furthermore, there is no precise value indicative of a lack of PC risk, and PSA levels cannot distinguish between indolent and aggressive disease, particularly at PSA levels below 20 ng/mL. In one study, the conventional cutoff value of 4 ng/mL PSA predicted PC in 10- or 12-core needle biopsies in only 30–40 % of patients [4]. In addition, ~15 % of men with serum PSA levels below 4 ng/mL are reported to be at risk for PC [5]. A precise PSA cut-off value that can facilitate the early detection of PC with high sensitivity and specificity in healthy men has not yet been defined. In addition, the ideal age at which to initiate or discontinue PSA testing, and the appropriate frequency of testing remain unclear. Two recent randomized trials evaluating the effect of PSA-based screening on mortality reduction reported conflicting results [6, 7]. Together, these findings prompted the United States Preventive Services Task Force to recommend against the use of PSA-based screening in 2012 [8]. Nevertheless, as there are no other reliable PC biomarkers to replace PSA, PSA screening remains the first-line assay for PC detection. As this approach is the subject of much debate and controversy, there is an unmet need for the identification of novel biomarkers with high sensitivity and specificity for detecting PC and predicting aggressive disease. Recently identified putative PC biomarkers are described in Table 1.

Serum PSA predominantly exists in a complex with α-1-antichymotrypsin. Levels of unbound PSA, referred to as free PSA (fPSA), are calculated using the following formula fPSA = total PSA − α-1-antichymotrypsin-bound PSA. %fPSA is associated with a greater specificity for detecting PC in men with a total PSA between 4 and 10 ng/mL [9, 10]. Three isoforms of fPSA are found in the serum: (1) proPSA, (2) intact PSA, and (3) benign PSA. There are also several truncated isoforms of proPSA, including [−2]proPSA, [−5]proPSA, and [−7]proPSA. [−2]proPSA is the predominant proPSA isoform in tumor extracts [11], suggesting that it has the potential to play a role in the early detection of PC and the prediction of aggressive disease [12] (Fig. 1).

Cancer-associated glycan aberrations are frequently observed in tumors. The majority of tumor markers, including PSA, are glycoproteins that have glycosylation sites in their amino acid sequence. Importantly, each glycan exhibits specific cancer-associated carbohydrate aberrations compared with its normal counterpart, and these aberrations can be detected using specific monoclonal antibodies or lectin.

Circulating tumor cells (CTCs) are tumor cells that have shed into the peripheral circulation from solid malignancies. CTCs have generally been considered surrogates for metastatic cells, and CTCs levels in patients undergoing treatment have proven to be a response marker with a strong independent prognostic value [42]. In addition, CTCs have been characterized as a ‘‘liquid biopsy” that can facilitate the real-time monitoring of therapeutic efficacy, and the identification of therapeutic targets and resistance mechanisms [43].

Recently, the presence of the AR splice variant 7 (AR-V7) in CTCs was shown to be strongly associated with resistance to anti-AR therapy [44]. The presence of AR-V7 in CTCs was associated with poor overall survival, with a hazard ratio (HR) of 6.9 [95 % confidence interval (CI) 1.7–28.1, P = 0.002] in patients receiving enzalutamide cohort and an HR of 12.7 (95 % CI 1.3–125.3, P = 0.006) in patients receiving abiraterone cohort. The same group also reported that the presence of AR-V7 did not significantly influence the effect of taxane treatment [45]. More recently, Scher et al. reported similar results from a prospective study of 191 blood samples from patients with metastatic CRPC. Although the proportion of patients with AR-V7-positive CTCs in this study was relatively low (34/191 samples, 18 %), AR-V7 positive patients were resistant to AR inhibition therapy, were on therapy for less time, and experienced a reduction in radiographic progression-free survival and overall survival compared with patients with AR-V7-negative CTCs [46]. However, there are substantial limitations to the clinical application of AR-V7 monitoring as only half of patients have detectable CTCs, CTSs cannot be stored, and the procedure is costly and can only be conducted in specialized labs. Although AR-V7 appears to be a promising prognostic factor, additional studies and technological innovations are needed to facilitate its widespread clinical use.

Although serum PSA has been used as a prostate cancer biomarker for several decades, studies demonstrating its limitations have incited much controversy and debate. The ideal PC biomarker would be capable of distinguishing PC from benign prostate conditions and differentiating between aggressive and indolent tumors. Recent technological innovations have led to substantial improvements in biomarker screening assays. An overview of the emerging prostate cancer biomarkers described in this review are summarized in Fig. 3. Specifically, two new tests designed to help determine the need for prostate biopsy have recently been approved by the US FDA. Currently, the incorporation of new biomarkers appears to be a promising approach for improving the sensitivity and specificity of PSA assays. However, comparative studies evaluating these various biomarker assays are still needed, as are studies with larger sample sizes and improvements in cost effectiveness. Large-scale, prospective trials can help evaluate the utility of these new approaches in multiple clinical contexts. The use of novel biomarkers that can serve as alternatives to PSA appears to be a promising approach to improving risk assessment strategies and has the potential to improve outcomes in patients with PC.