The end of the single-marker era: multimarker algorithms and artificial intelligence in prostate cancer diagnosis—a narrative review

Introduction

Irrespective of the diagnostic pathway leading to the identification of prostate cancer (PCa), measurement of serum prostate-specific antigen (PSA) remains a central component of the diagnostic process. PCa is among the most diagnosed malignancies in men worldwide and represents a major cause of cancer-related morbidity and mortality (1). Since the early 1990s, the widespread adoption of PSA testing has profoundly transformed PCa detection, leading to a substantial increase in the identification of localized and low-risk tumors.

However, the extensive use of PSA screening has also exposed significant limitations of this biomarker. Several epidemiological studies have shown that the rise in PCa incidence following the introduction of PSA testing has not been accompanied by a proportional reduction in disease-specific mortality in many populations (2-5). This discrepancy largely reflects the enhanced detection of previously undiagnosed and frequently indolent tumors rather than a true increase in disease occurrence. Indeed, approximately half of newly diagnosed cases are estimated to represent low-risk lesions with limited biological aggressiveness and minimal likelihood of affecting patient survival (6). As a consequence, PSA-based screening has been associated with substantial overdiagnosis and overtreatment, with important repercussions for patients’ quality of life (1-4).

PSA received U.S. Food and Drug Administration (FDA) approval in 1994 for PCa screening in conjunction with digital rectal examination (DRE) (7). Nonetheless, concerns regarding its clinical implications emerged early. In 2001, Stamey who had previously demonstrated the superior sensitivity of PSA compared with prostatic acid phosphatase, warned about the potential long-term consequences of PSA-driven screening strategies (8). The debate intensified over the following decade and reached a turning point in 2012, when the U.S. Preventive Services Task Force recommended against routine PSA screening in the general population (9). Similarly, Richard Ablin, who first identified PSA in 1970 (10), later emphasized two fundamental limitations of the marker: its lack of diagnostic specificity and its inability to discriminate between indolent and clinically significant tumors.

Although risk-adapted management strategies such as active surveillance (AS) have been developed to mitigate overtreatment, the magnitude of unnecessary interventions remains considerable. Even when applying established risk-stratification systems such as the D’Amico criteria, an estimated 20–40% of men may still undergo treatment for tumors unlikely to threaten survival (2-4,11). These challenges reflect the marked clinical and biological heterogeneity of PCa, which ranges from indolent lesions to highly aggressive disease requiring immediate intervention.

In this context, the development of more accurate and personalized diagnostic strategies has become a major priority. Biomarkers are expected to play a key role in this transition (12-17), not only by improving early detection but, more importantly, by enabling the discrimination between indolent and clinically significant disease. The field is therefore progressively moving from a single-marker paradigm toward integrated diagnostic approaches that combine circulating biomarkers, imaging technologies, and computational models.

The aim of this review is to provide a clinically oriented overview of this transition. While numerous studies have examined PCa biomarkers or imaging modalities individually, a major unresolved challenge in contemporary diagnostics is the lack of integrated frameworks that combine multimarker algorithms, multiparametric MRI (mpMRI), and artificial intelligence (AI) within a unified clinical decision pathway. This review addresses this gap by synthesizing current evidence on multimarker tests, imaging, and AI-driven models, highlighting their complementary role in improving risk stratification, reducing unnecessary biopsies, and enabling more personalized diagnostic strategies beyond single-marker approaches. We present this article in accordance with the Narrative Review reporting checklist (available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-79/rc).

Methods

Literature search and study selection

This study was conducted as a narrative review. A comprehensive literature search was performed in PubMed/MEDLINE, Scopus, and Web of Science to identify relevant studies on PCa diagnostics, circulating biomarkers, multimarker algorithms, mpMRI, and AI-based diagnostic models.

The search covered the period from database inception to November 2025, with particular emphasis on studies published in the last 10–15 years to capture recent advances in precision diagnostics. The final search was conducted on November 15, 2025. Additional relevant studies were identified through manual screening of reference lists from key reviews, clinical guidelines, and seminal articles.

Studies were eligible for inclusion if they investigated circulating or molecular biomarkers for PCa detection or risk stratification, evaluated multimarker algorithms or integrated diagnostic models (biomarkers, mpMRI and/or AI), and reported clinically relevant outcomes such as diagnostic accuracy, area under the curve (AUC), sensitivity, or specificity. Original research articles, systematic reviews, meta-analyses, and clinical guidelines were considered eligible. Articles not written in English, conference abstracts without full text, editorials lacking scientific data, case reports, and studies not directly related to PCa diagnostics were excluded.

Study selection was performed through a two-step screening process consisting of title and abstract review followed by full-text evaluation. Screening was conducted independently by two authors with expertise in laboratory medicine and PCa biomarkers. Disagreements were resolved through discussion until consensus was reached.

Given the narrative nature of this review and the heterogeneity of study designs, findings were synthesized qualitatively with particular attention to diagnostic performance, clinical validation, integration with mpMRI, and emerging AI-driven multimarker models. The literature search strategy and study selection process are summarized in Table 1.

Limitations of conventional diagnostic tools: DRE, mpMRI, and prostate biopsy

Before the widespread adoption of multimarker strategies, the diagnostic pathway for suspected PCa has traditionally relied on DRE, mpMRI, and prostate biopsy (6,7). DRE is a simple and low-cost clinical tool, but its sensitivity is limited, particularly for small, anterior, or early-stage tumors, and it lacks sufficient accuracy as a standalone method for reliable PCa detection.

Consequently, DRE alone lacks the diagnostic accuracy required for reliable risk stratification and is primarily used in combination with PSA and other clinical parameters. mpMRI has significantly improved the detection of clinically significant PCa and is currently recommended in many diagnostic algorithms. It offers higher sensitivity than PSA alone for the identification of clinically significant disease and enables targeted biopsy through prostate imaging reporting and data system (PI-RADS)-guided lesion characterization. However, mpMRI also presents important limitations. Its specificity is variable, it requires substantial resources and specialized expertise, and its interpretation is subject to inter-reader variability. Moreover, equivocal findings (e.g., PI-RADS 3 lesions) represent a diagnostic “gray zone”, in which a relatively low proportion of patients harbor clinically significant cancer, yet many still undergo biopsy (12,18).

Prostate biopsy remains the reference standard for definitive diagnosis but is an invasive procedure associated with non-negligible risks, including infection, bleeding, pain, and psychological burden. In addition, systematic biopsy may lead to both underdiagnosis of clinically significant tumors and overdiagnosis of indolent lesions due to sampling limitations and tumor heterogeneity. The sensitivity and specificity of biopsy are also influenced by sampling strategy (systematic vs. targeted) and imaging guidance, and false-negative results can still occur even in the presence of clinically significant disease (18,19).

Taken together, these limitations highlight a key clinical challenge: the need to reduce unnecessary invasive procedures while maintaining accurate detection of aggressive tumors. In this context, circulating and multimarker biomarkers represent a valuable complementary tool, as they may improve pre-biopsy risk stratification, refine patient selection for mpMRI and biopsy, and ultimately minimize overdiagnosis, overtreatment, and procedure-related harms.

From PSA to multiple biomarkers: an overview of the evolution in PCa detection

Serum PSA levels are commonly interpreted within defined clinical ranges. Values below 4 ng/mL have traditionally been considered within the normal range, although clinically significant PCa may still be present in a subset of men. PSA concentrations between 2 and 10 ng/mL are generally referred to as the “gray zone”, a range associated with substantial diagnostic uncertainty in which the probability of PCa increases but diagnostic specificity remains limited. In this context, the free-to-total PSA ratio (f/t PSA) has been introduced to refine risk stratification: lower ratios are more suggestive of malignancy, whereas higher ratios are more frequently associated with benign prostatic conditions such as benign prostatic hyperplasia. However, evidence indicates that only about 56% of cases with an f/t PSA ratio below 10% are associated with malignancy, highlighting the limited reliability of biopsy decisions based solely on this parameter (20). Consequently, clinical decision-making typically relies on a combination of total PSA (tPSA) level, f/t PSA ratio, DRE findings, patient age, and additional risk factors, underscoring the need for more accurate diagnostic strategies.

In response to these limitations, current research efforts are directed toward identifying biomarkers capable of reducing overdiagnosis while maintaining the sensitivity required to detect clinically significant tumors, thereby enabling therapeutic decisions that better align tumor aggressiveness with treatment intensity (1,7,13,14,21). A clear trend has emerged toward the development of multimarker panels that combine multiple analytes to overcome the intrinsic limitations of single-marker approaches, an evolution largely driven by the experience with PSA-based screening (21). Several circulating biomarker panels have been developed following this principle, including the prostate health index (PHI), 4Kscore, Stockholm3, Proclarix, and IsoPSA, all of which incorporate different molecular forms of PSA within integrated diagnostic algorithms (12,17,22).

The rationale for these approaches derives from the biochemical complexity of PSA. PSA is a serine protease physiologically involved in semen liquefaction and circulates in different molecular forms, primarily free PSA (fPSA) and PSA complexed with protease inhibitors such as α1-antichymotrypsin (ACT). fPSA itself comprises several inactive fragments and precursor forms, including proPSA isoforms such as [-2]proPSA, which have been shown to be more strongly associated with prostate malignancy (23,24). In the presence of cancer, disruption of the normal glandular architecture alters PSA processing and secretion, leading to a relative increase in precursor and complexed forms and a decrease in fPSA (23). These molecular differences provide the biological basis for diagnostic approaches that combine multiple PSA-derived biomarkers rather than relying on tPSA alone.

PHI

To overcome the limited diagnostic performance of the free-to-total PSA ratio, the PHI was developed. The PHI is a composite mathematical algorithm that integrates three PSA isoforms, tPSA, fPSA, and the [-2]proPSA isoform (p2PSA), into a single risk score designed to improve the specificity of PCa detection, particularly in the PSA gray zone. The PHI score is calculated using the formula:

PHI=p2PSAfPSA×tPSA[1]

where p2PSA and fPSA are expressed in the same units and tPSA represents total serum PSA concentration. By combining the relative increase of the cancer-associated p2PSA isoform with the inverse relationship between fPSA and malignancy, and scaling the result by tPSA, the algorithm provides a more accurate estimation of the probability of clinically significant PCa than tPSA or the f/t PSA ratio alone. This integrative approach reflects the biological shift in PSA isoform distribution observed in malignant prostate tissue, where higher levels of p2PSA and lower proportions of fPSA are typically detected (23).

The PHI received approval from the U.S. Food and Drug Administration (FDA) in 2010 for use in men aged over 50 years with a negative DRE and PSA concentrations within the so-called “gray zone” (2–10 ng/mL) (23). Since its approval, multiple studies have evaluated the diagnostic performance of PHI in comparison with the f/t PSA ratio. Notably, a meta-analysis published in 2014 demonstrated that, among men with PSA levels in the gray zone, the likelihood of detecting a positive prostate biopsy is significantly higher when PHI is used instead of the f/t PSA ratio (20).

Moreover, PHI has been shown to correlate significantly with tumor aggressiveness, as assessed in radical prostatectomy specimens (25). These findings were subsequently confirmed by Tosoian et al. In a study including a large cohort of 1,663 patients (26), PHI demonstrated superior discrimination for clinically significant prostate cancer (csPCa) compared with tPSA and the f/t PSA ratio, with reported AUC values ranging approximately from 0.70 to 0.77. Similarly, meta-analytic data have shown that PHI significantly improves the prediction of positive biopsy and csPCa detection, with higher specificity than PSA alone while maintaining high sensitivity. Moreover, several studies report that the implementation of PHI-based thresholds may reduce unnecessary biopsies by up to 25–40% without substantially compromising the detection of clinically significant disease (20,23,27).

Subsequent studies, including recent investigations, have consistently confirmed these observations. In a cohort of 77 patients undergoing active surveillance, higher PHI values were significantly associated with disease progression (25).

By 2017, systematic reviews of the literature consistently demonstrated that the routine use of PHI could substantially reduce the number of unnecessary prostate biopsies without compromising the detection of clinically significant cancers (28,29). Consistent with this evidence, PHI has been incorporated into the European Association of Urology (EAU) guidelines since 2015 and continues to be recommended in current clinical practice guidelines (30).

4kscore

This test integrates measurements of four kallikreins, tPSA, fPSA, intact PSA, and human kallikrein 2 (hK2) with clinical variables, including age, DRE findings, and prior biopsy status, to generate a composite risk score that estimates the probability of harboring aggressive PCa (31).

Evidence from the literature indicates that the 4Kscore is significantly associated with the presence of distant metastases (32). Comparative studies indicate that both PHI and the 4Kscore show similar sensitivity for the detection of clinically significant PCa, while differences in specificity are modest and context-dependent, with both tests consistently demonstrating higher specificity than tPSA alone and enabling a meaningful reduction in unnecessary biopsies without substantially compromising sensitivity (33).

Stockolm 3

More recently, the Stockholm3 test, developed at the Karolinska Institute in Stockholm, has been introduced as a multivariable diagnostic tool. This assay integrates clinical variables, including age, family history, prior biopsy status, prostate volume, and DRE findings, with genetic information derived from 232 single-nucleotide polymorphisms (SNPs; including variants in IL4, MGMT, AKT, among others) and six plasma protein biomarkers. Notably, four of these proteins correspond to the kallikreins included in the 4Kscore, complemented by β-microseminoprotein (MSMB) and macrophage inhibitory cytokine-1 (MIC1) (34).

The diagnostic performance of the Stockholm3 test was demonstrated in a large study published in European Urology involving approximately 60,000 men, which reported an AUC of 0.75, significantly higher than that of PSA for the detection of high-grade PCa. The same study also underscored the clinical utility of Stockholm3, showing that, compared with PSA alone, its use substantially reduces unnecessary biopsies while preserving the detection of clinically significant disease (35).

A subsequent multiethnic trial (“SEPTA/STHLM3 in a Multiethnic Cohort”, 2024), enrolling more than 2,000 men from diverse ethnic backgrounds, further confirmed that Stockholm3 maintains non-inferior sensitivity compared with PSA for the detection of clinically significant PCa, while demonstrating substantially superior specificity (36). These findings indicate that the test effectively reduces unnecessary prostate biopsies and the detection of indolent, low-grade tumors without compromising the identification of clinically significant disease. However, early validation studies were conducted predominantly in Northern European populations, limiting the immediate generalizability of the results. Although more recent investigations incorporating multiethnic cohorts have strengthened the evidence base and improved external validity, the full applicability of the test across all ethnic groups and clinical settings has yet to be definitively established. For diagnostic tests in PCa, clinical validation should therefore systematically include men from diverse ethnic backgrounds. This requirement reflects the well-documented interethnic variability in baseline biomarker levels, disease incidence and aggressiveness, as well as differences in the distribution of genetic risk variants, all of which may influence test performance and clinical utility (37).

The use of ethnically representative cohorts is therefore essential to establish accurate diagnostic cut-offs, minimize methodological bias, and ensure the reliable translation of results into routine clinical practice across diverse populations.

Proclarix

Proclarix is a blood-based biomarker test developed to support clinical decision-making regarding prostate biopsy. The assay integrates thrombospondin-1 (THBS1), cathepsin D (CTSD), tPSA, fPSA, and patient age into a composite risk score that estimates the probability of csPCa (38).

The novel biomarkers THBS1 and CTSD were originally identified through a genetics-guided biomarker discovery strategy targeting the PI3K/PTEN signaling pathway, which plays a central role in PCa development and progression (39).

The analytical and clinical performance of Proclarix was initially established in retrospective studies and subsequently validated in the large, prospective, multicenter PROPOSe trial (Prospective Proclarix Outcome Study). This study enrolled 457 men undergoing prostate biopsy with PSA levels between 2 and 10 ng/mL, a normal DRE, and a prostate volume ≥35 cm³ (40).

In clinical evaluation (CE) validation studies, Proclarix demonstrated a sensitivity of 90% and a negative predictive value (NPV) of 95% for the detection of csPCa (Grade Group ≥2). Subsequent analyses revealed a strong correlation with Likert scores derived from mpMRI, highlighting the test’s ability to effectively stratify patients with indeterminate imaging findings (Likert 3). In this specific subgroup, Proclarix achieved 100% sensitivity and 100% NPV, with a specificity of 34%, enabling approximately one-third of patients to avoid unnecessary biopsies without compromising diagnostic accuracy (40).

The PROPOSe study confirmed these findings in real-world clinical practice, demonstrating sensitivity and NPV exceeding 90%, with even higher performance when biopsies were guided by mpMRI (97% sensitivity and 96% NPV). Notably, Proclarix was significantly superior to the f/t PSA ratio in avoiding unnecessary biopsies (22% vs. 14%), achieving the primary study endpoint with statistical significance (P=0.004) (38).

IsoPSA

Most currently available tumor biomarkers, particularly protein-based assays, classify results as normal or abnormal based exclusively on biomarker concentration in body fluids. IsoPSA represents a paradigm shift by moving beyond quantitative assessment to focus on the molecular structure of PSA, specifically detecting conformational changes that differentiate PSA produced by malignant prostatic cells from that derived from benign tissue (41). IsoPSA is designed to detect structurally distinct isoforms of PSA associated with malignant transformation. Unlike conventional PSA-based assays that quantify total protein concentration, IsoPSA characterizes the molecular composition of PSA, as tumor-derived and benign PSA molecules exhibit distinct conformational properties. In this context, diagnostic information is derived not from the amount of PSA present, but from the structural features of the protein that reflect its biological origin (41-50).

The key analytical parameter of the assay, denoted as K, is calculated according to the following formula:

K=PSAtotalbottom−PSAfreebottomPSAtotaltop−PSAfreetop[2]

From a clinical perspective, IsoPSA has demonstrated superior diagnostic accuracy compared with both tPSA and the f/t PSA ratio for predicting positive prostate biopsy results and identifying csPCa (41).

IsoPSA is based on a two-phase aqueous system in which the various PSA isoforms and complexes present in serum partition between the upper and lower phases according to their structural characteristics and protein-protein interactions. Immunometric quantification of total and fPSA in each phase is subsequently combined to calculate the K ratio, which functions as a ratiometric index rather than a direct measure of PSA concentration.

Importantly, the difference between total and fPSA within each phase does not represent a quantification of the PSA-α1-antichymotrypsin (PSA-ACT) complex, but instead reflects the distinctive distribution of PSA isoforms. Although both K values and serum PSA concentrations generally increase in the presence of PCa, they capture biologically distinct and complementary information.

The optimal K cut-off is derived through statistical modeling to maximize diagnostic performance within specific clinical settings. Alternatively, K can be transformed into a normalized risk score (KR-HG), scaled from 0 to 100, which provides a clinically intuitive estimate of the probability of high-grade PCa.

From a clinical standpoint, IsoPSA offers several practical advantages in the management of men with suspected PCa. By enhancing risk discrimination in patients with elevated but nonspecific PSA levels, it facilitates the identification of individuals more likely to harbor clinically significant disease. IsoPSA may also support biopsy decision-making in settings where mpMRI is unavailable or when imaging findings are equivocal, such as in cases with indeterminate PI-RADS scores. In addition, its ability to predict the long-term risk of clinically significant PCa enables a more personalized approach to patient follow-up, including optimization of biopsy timing and surveillance strategies. Based on these characteristics, IsoPSA has been proposed as an effective triage tool for identifying patients with PSA values >4 ng/mL who should be prioritized for referral to mpMRI (43,45,47).

The current landscape of the new biomarkers

Beyond PSA-derived indices and established multimarker panels, several emerging single biomarkers are currently being investigated to further refine PCa risk stratification (13). These include urine-based molecular assays targeting cancer-specific transcripts such as PCA3, TMPRSS2-ERG fusion transcripts, HOXC6, and DLX1, as well as exosomal RNA signatures (12,15,16,51). Unlike traditional PSA derivatives, these biomarkers aim to capture tumor-specific molecular alterations, potentially improving biological specificity for clinically significant disease. Although many of these assays have demonstrated promising diagnostic performance in selected cohorts, their integration into routine clinical practice still requires broader external validation, cost-effectiveness assessment, and confirmation across diverse populations.

Among circulating biomarker tests currently available in clinical practice (Table 2), diagnostic performance for detecting csPCa is broadly comparable. However, important differences exist in terms of cost, accessibility, and clinical implementation. From a health economics perspective, cost-effectiveness represents a critical determinant for clinical adoption. In this context, the PHI is often considered one of the most accessible and economically sustainable options, as it can be easily integrated into routine laboratory workflows while providing meaningful improvements in risk stratification. In contrast, more complex algorithm-based tests such as 4Kscore, Stockholm3, and Proclarix involve greater analytical complexity and higher economic burden, factors that may limit their widespread implementation in resource-constrained settings.

PHI also represents the only circulating biomarker assay that is both FDA-approved and CE-marked and for which dedicated studies have evaluated diagnostic performance across different ethnic groups, including individuals of African and Asian ancestry (37,52-57). This aspect is particularly relevant in PCa, a disease characterized by marked ethnic disparities in incidence and aggressiveness, with men of African ancestry showing a higher risk of aggressive disease and men of Asian ancestry generally presenting lower incidence rates (58). Consistent with this observation, our preliminary study conducted at the University of Naples Federico II in a cohort of men of African origin suggested distinct PHI performance patterns (37), further supporting the need for ethnicity-specific validation of circulating biomarkers in PCa diagnostics.

Current EAU guidelines reference all five circulating biomarker tests, summarizing the available clinical evidence for each (30). For screening and early detection in asymptomatic men with PSA levels between 3 and 20 ng/mL and a negative DRE, the guidelines strongly recommend the use of mpMRI, whereas biomarkers currently receive a weaker recommendation.

Despite its central role in the diagnostic pathway, mpMRI also presents important limitations. It is a resource-intensive modality requiring specialized infrastructure and expertise, and its diagnostic performance remains partly operator-dependent. Evidence from the PROMIS cohort demonstrated that mpMRI may miss small-volume or low-grade tumors (18). Furthermore, PI-RADS 3 lesions represent an imaging “gray zone”, with clinically significant PCa detected in fewer than 15% of cases, frequently leading to unnecessary biopsies. Even among PI-RADS 4 and 5 lesions, a proportion of patients harbor indolent tumors, whereas some individuals with negative mpMRI findings may still present with clinically significant disease (19).

These limitations highlight that imaging alone is insufficient to optimize biopsy decision-making. Consequently, current research increasingly supports integrated diagnostic strategies in which circulating biomarkers are used either prior to or in combination with mpMRI to refine patient selection for biopsy. Triage-based models, such as biomarker-first approaches (e.g., PHI-to-refine-MRI strategies), have demonstrated the potential to reduce unnecessary imaging and invasive procedures while maintaining diagnostic sensitivity for clinically significant tumors. In this framework, biomarkers and imaging should be viewed as complementary tools within a multi-step diagnostic pathway, an approach that may be further enhanced by the incorporation of AI-based decision support systems (59).

Combined approach: biomarkers and imaging

In this context, the use of circulating biomarkers may offer valuable complementary support. Among available biomarkers, PHI represents the most extensively validated and cost-effective candidate for integration with mpMRI and AI-based models (60,61).

Accordingly, over recent years, several research groups have investigated the performance of PHI in comparison with mpMRI, as well as its combined use with mpMRI, including analyses stratified by PI-RADS category. These studies were designed to define the optimal integration and clinical positioning of PHI within the diagnostic pathway of men with suspected PCa.

Initial evidence was provided by a study of 279 men undergoing repeat prostate biopsy, which demonstrated that the addition of PHI to mpMRI, unlike tPSA, significantly improved diagnostic performance. Specifically, the combined approach enhanced both overall PCa detection [AUC =0.71 (95% CI: 0.61–0.76) vs. 0.64 for mpMRI alone] and the identification of clinically significant disease (AUC =0.75 vs. 0.64) (62). The combined use of PHI and mpMRI has been shown to achieve a significantly greater reduction in unnecessary prostate biopsies compared with either PHI or mpMRI alone (50% vs. 35%, P<0.001). Notably, up to 50% of biopsies could be avoided when biopsy referral is restricted to patients with PI-RADS scores between 3 and 5 and PHI values >30. In a large study of 345 men undergoing initial prostate biopsy, Tosoian et al. (26) at Johns Hopkins University first demonstrated that PHI is able to identify csPCa even among patients with negative (PI-RADS 1) or equivocal (PI-RADS 3) mpMRI findings. These observations have since been consistently confirmed by subsequent studies (63).

We further demonstrated that PHI density, defined as the ratio of PHI to prostate volume, enhances the detection of clinically significant tumors in men with negative or indeterminate mpMRI findings. Consistently, Druskin et al. reported that PHI density effectively discriminates csPCa in patients with PI-RADS 1 (negative mpMRI) and PI-RADS 3 (equivocal mpMRI) lesions (64).

More recently, additional studies have reported that the combined use of PHI and mpMRI further improves the discrimination of csPCa in patients with PI-RADS 4 and 5 lesions (65).

Based on this body of evidence, the combined use of PHI and mpMRI has been explored as a strategy to predict grade reclassification in men undergoing active surveillance (AS). This integrated approach has the potential to reduce the frequency of surveillance biopsies, thereby improving patient adherence to AS protocols, enhancing quality of life, and decreasing healthcare costs associated with long-term monitoring. A retrospective study by Schwen et al. (66) demonstrated that the combined use of PHI and mpMRI increased the NPV for biopsy reclassification in patients undergoing active surveillance to 98%, outperforming both PSA density combined with mpMRI (95.4%) and mpMRI alone (91.6%). Accordingly, the integration of PHI and mpMRI has the potential to substantially reduce the number of surveillance biopsies, limit procedure-related morbidity, and ultimately improve adherence to active surveillance protocols. Notably, the combined use of PHI and mpMRI has been shown to improve the preoperative prediction of extracapsular extension (ECE) of PCa, thereby supporting more accurate preoperative counseling and informing the selection of nerve-sparing surgical strategies (67). This is particularly relevant given that nerve-sparing techniques are critical for preserving erectile function and minimizing postoperative morbidity. Overall, the available evidence indicates that PHI and mpMRI are complementary diagnostic tools, and that their combined use prior to prostate biopsy results in improved diagnostic accuracy compared with either modality alone (68).

Consistent with these findings, recently developed nomograms for the identification of csPCa routinely incorporate PHI in combination with PI-RADS scoring (69).

The integration of PHI with PI-RADS scoring within nomograms is supported by the complementary biological and imaging information provided by these parameters. While PI-RADS reflects the anatomical and radiological suspicion of csPCa, PHI captures the underlying tumor-related biochemical activity through PSA isoform profiling. Their combined use within predictive models improves risk stratification by increasing diagnostic accuracy, particularly in patients with equivocal mpMRI findings (e.g., PI-RADS 3 lesions), and enhances the identification of clinically significant disease while reducing unnecessary biopsies. From a clinical perspective, this integrative approach supports more personalized decision-making by refining biopsy selection and optimizing the pre-biopsy diagnostic pathway.

AI-driven approach

It is increasingly evident that the integration of biomarkers and imaging will be further enhanced by AI. Indeed, several published models already incorporate PHI within AI-based frameworks, highlighting the potential of AI-assisted risk stratification to further refine pre-biopsy clinical decision-making (70).

In a prospective study involving 344 men with PSA levels between 2 and 10 ng/mL, an artificial neural network (ANN) integrating [−2]proPSA, tPSA, fPSA, cathepsin D, and thrombospondin-1 demonstrated improved performance in the detection of csPCa (71).

The ANN model achieved sensitivity and specificity values of approximately 66–68% for csPCa, significantly outperforming PHI and Proclarix when used as standalone tests. Notably, this combinatorial AI-based approach demonstrated superior discrimination of high-grade disease (Gleason score ≥7) compared with conventional logistic regression models. Model performance remained robust across different training and validation sample sizes and was only marginally influenced by patient age. Collectively, these findings support the integration of AI-driven multimarker models to enhance risk stratification and enable more personalized decision-making at the time of initial PCa diagnosis.

The AI-based models described were generally developed using retrospective and prospective clinical cohorts ranging from approximately 100 to several hundred patients, integrating multidimensional inputs including clinical variables (age, PSA, prostate volume), biomarker data (e.g., PHI and PSA isoforms), and mpMRI parameters such as PI-RADS scores. In validation settings, these models consistently demonstrated improved discrimination for csPCa, with reported AUC values around 0.75–0.85, alongside higher sensitivity for high-grade disease detection while maintaining acceptable specificity compared with single-marker approaches.

In our previous study (72), we demonstrated that integrating PHI and mpMRI through ANN models yields a composite approach with superior accuracy for the identification of csPCa confirmed by histopathology. Moreover, the combined model achieved the highest sensitivity, whereas PHI alone exhibited the highest specificity for the detection of high-grade PCa.

To our knowledge, this is the first study to estimate the presence of aggressive PCa confirmed at radical prostatectomy (RP) using a combinatorial ANN model integrating PHI and mpMRI data. In this framework, the ANN combining PHI and mpMRI achieved the highest sensitivity (0.80) for the detection of csPCa, whereas PHI alone demonstrated the highest specificity (0.73). This enhanced sensitivity is consistent with our previous observations showing that PHI and PHI density (PHID) are able to identify csPCa even in patients with PI-RADS 1–2 lesions and positive biopsy results.

Recent studies further support the value of AI-based integrative approaches. For example, in a cohort of 131 PCa patients, a machine learning–based nomogram incorporating MRI data showed improved discrimination for clinically significant tumors following the inclusion of PHI, thereby confirming the synergistic benefit of biomarker–imaging integration within AI-driven predictive models (73).

Future perspectives

In the coming years, PCa biomarkers and imaging modalities should be integrated within a complementary, interconnected diagnostic framework rather than applied in a competitive manner, with the goal of optimizing patient selection for further diagnostic assessment and therapeutic intervention. ANN-based approaches are particularly well suited to support this integration, as they enable the harmonization of heterogeneous data sources and enhance the accuracy and consistency of clinical decision-making.

Within the diagnostic–therapeutic pathway of PCa, the synergistic implementation of AI, nanotechnologies, capable of improving analytical efficiency and reducing costs (74-78), and structured health literacy initiatives that actively engage and empower patients, is expected to drive the paradigm shift originally envisioned by Ablin (Figure 1). The “paradigm shift originally envisioned by Ablin” refers to the transition from indiscriminate PSA-based screening toward a more selective, risk-adapted diagnostic strategy capable of distinguishing indolent from clinically significant PCa. In his later reflections on PSA screening, Ablin emphasized the need to move beyond PSA as a stand-alone test and to develop more biologically informed approaches that could reduce overdiagnosis and overtreatment while preserving the benefits of early detection (79-83). In this context, the integration of multimarker algorithms, imaging modalities, and AI-driven predictive models represents a concrete realization of this shift toward precision-oriented PCa diagnostics. This evolution toward precision-oriented, patient-centred care will facilitate more personalized therapeutic strategies, improve quality of life, and minimize overtreatment and its associated morbidity.

Figure 1 Integrated framework for the stratification of prostate cancer aggressiveness. The figure illustrates a multidimensional approach to PCa management, distinguishing indolent PCa (iPCa) and clinically significant PCa (csPCa). Biomarkers represent the central pillar for biological risk stratification and are integrated with artificial intelligence (AI)-driven analytical models, nanotechnology-based diagnostic platforms, imaging and health literacy-oriented patient engagement. Together, these components support personalized risk assessment, improved diagnostic accuracy, and informed clinical decision-making along the iPCa-csPCa continuum. This figure is an original conceptual illustration created for this manuscript by the authors. Graphical elements were generated using Servier Medical Art (licensed under Creative Commons Attribution 3.0 Unported License). AI, artificial intelligence; csPCa, clinically significant PCa; iPCa, indolent PCa; PCa, prostate cancer.

Limitations of the evidence

This review provides a comprehensive and clinically oriented synthesis of the evolving landscape of PCa diagnostics, integrating circulating biomarkers, multimarker algorithms, mpMRI, and AI within a unified translational framework. A multidisciplinary perspective was adopted, bridging laboratory medicine, urology, and precision diagnostics, with particular emphasis on clinically actionable tools such as PHI, 4Kscore, Stockholm3, IsoPSA, and Proclarix. In addition, the analysis focuses on real-world clinical applicability, including biopsy decision-making, risk stratification, and strategies aimed at reducing unnecessary invasive procedures.

However, several limitations should be acknowledged. As a narrative review, the study does not follow a fully systematic selection process and may therefore be subject to selection and publication bias. Moreover, the available literature is characterized by substantial methodological heterogeneity in study design, patient selection, cohort size, biopsy indications, and definitions of csPCa, which may affect the comparability of reported diagnostic performance metrics. Many validation studies are based on retrospective or single-center cohorts and predominantly involve selected populations, potentially limiting external validity across diverse ethnic and clinical settings. Variability in reference standards (systematic versus targeted biopsy, mpMRI-based pathways) and differences in PSA ranges may further influence reported sensitivity, specificity, and AUC estimates. Finally, the rapidly evolving nature of biomarker research and AI-driven diagnostic models highlights the need for larger prospective studies, multiethnic cohorts, and standardized study designs to strengthen the robustness and generalizability of the current evidence base.

Conclusions

Reliance on a single biomarker for PCa diagnosis within the PSA “gray zone” is increasingly inadequate. Future diagnostic strategies should instead rely on the integrated assessment of multiple biomarkers across complementary analytical platforms, including emerging molecular markers, to enable more accurate and personalized risk stratification in this challenging PSA range.

The integration of biomarker panels with multiparametric imaging and AI-based models is expected to define a multi-layered, patient-centered diagnostic pathway that maximizes accuracy while minimizing unnecessary procedures and patient burden. While progress toward reducing overdiagnosis and overtreatment is most evident in high-income countries, significant disparities persist in low- and middle-income countries (LMICs), where limited access to advanced diagnostics and follow-up infrastructure continues to drive suboptimal care.

Ultimately, reducing overdiagnosis and overtreatment in PCa requires a paradigm shift toward precision medicine, incorporating risk-adapted screening, imaging-guided diagnostics, biomarker-based stratification, and shared decision-making. To fully realize this transition, healthcare systems must update clinical guidelines, promote equitable access to innovation, and strengthen patient and public education.

Acknowledgments

The authors gratefully acknowledge colleagues in Urology and Laboratory Medicine for their insightful discussions.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-79/rc

Peer Review File: Available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-79/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-79/coif). D.T. serves as an unpaid editorial board member of Journal of Laboratory and Precision Medicine from September 2025 to December 2027. D.T. reports receiving honoraria for scientific lectures, educational grants, and advisory board activities from Fujirebio, Novartis, and Roche, outside the submitted work. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Al Barajraji M, Coscarella M, Svistakov I, et al. Beyond PSA: The Future of Prostate Cancer Diagnosis Using Artificial Intelligence, Novel Biomarkers, and Advanced Imagery. Life (Basel) 2025;15:1508. [Crossref] [PubMed]
2. Daskivich TJ, Leppert JT. Overtreatment in Prostate Cancer: An Overused Term-Reply. JAMA Intern Med 2025;185:901-2. [Crossref] [PubMed]
3. Daskivich TJ, Luu M, Heard J, et al. Overtreatment of Prostate Cancer Among Men With Limited Longevity in the Active Surveillance Era. JAMA Intern Med 2025;185:28-36. [Crossref] [PubMed]
4. Daskivich TJ, Leppert JT. Overtreatment of Prostate Cancer in Men With Limited Life Expectancy-A Global Phenomenon? JAMA Netw Open 2025;8:e258576. [Crossref] [PubMed]
5. Croswell JM, Kramer BS, Crawford ED. Screening for prostate cancer with PSA testing: current status and future directions. Oncology (Williston Park) 2011;25:452-60.
6. Cooperberg MR, Broering JM, Kantoff PW, et al. Contemporary trends in low risk prostate cancer: risk assessment and treatment. J Urol 2007;178:S14-9. [Crossref] [PubMed]
7. Mutalip IW, Barth AF, Sellers GS, et al. From bench to bedside: the evolution and future of prostate-specific antigen testing. Med Oncol 2025;42:486. [Crossref] [PubMed]
8. Stamey TA, Teplow DB, Graves HC. Identity of PSA purified from seminal fluid by different methods: comparison by amino acid analysis and assigned extinction coefficients. Prostate 1995;27:198-203. [Crossref] [PubMed]
9. Moyer VA. U.S. Screening for prostate cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2012;157:120-34.
10. Ablin RJ, Soanes WA, Bronson P, et al. Precipitating antigens of the normal human prostate. J Reprod Fertil 1970;22:573-4. [Crossref] [PubMed]
11. Ziglioli F, Granelli G, Cavalieri D, et al. What chance do we have to decrease prostate cancer overdiagnosis and overtreatment? A narrative review. Acta Biomed 2019;90:423-6.
12. Assani KD, Pierce JM, Galloway LA, et al. Blood- and urine-based biomarkers for the detection of clinically significant prostate cancer: a contemporary review. Curr Opin Urol 2025;35:590-6. [Crossref] [PubMed]
13. Lopez-Valcarcel M, Lopez-Campos F, Zafra-Martín J, et al. Biomarkers in Localized Prostate Cancer: From Diagnosis to Treatment. Int J Mol Sci 2025;26:7667. [Crossref] [PubMed]
14. Min K, Lin Q, Qiu D. Precision medicine in prostate cancer: individualized treatment through radiomics, genomics, and biomarkers. Cancer Imaging 2025;25:116. [Crossref] [PubMed]
15. Plas S, Melchior F, Aigner GP, et al. The impact of urine biomarkers for prostate cancer detection-A systematic state of the art review. Crit Rev Oncol Hematol 2025;210:104699. [Crossref] [PubMed]
16. Ravi N, John NA, Taranikanti M. Serum and Urine Biomarkers for Prostate Cancer: A Mini Review. Mymensingh Med J 2025;34:598-603.
17. Robinson HS, Lee SS, Barocas DA, et al. Evaluation of blood and urine based biomarkers for detection of clinically-significant prostate cancer. Prostate Cancer Prostatic Dis 2025;28:45-55. [Crossref] [PubMed]
18. Brown LC, Ahmed HU, Faria R, et al. Multiparametric MRI to improve detection of prostate cancer compared with transrectal ultrasound-guided prostate biopsy alone: the PROMIS study. Health Technol Assess 2018;22:1-176. [Crossref] [PubMed]
19. Kasivisvanathan V, Rannikko AS, Borghi M, et al. MRI-Targeted or Standard Biopsy for Prostate-Cancer Diagnosis. N Engl J Med 2018;378:1767-77. [Crossref] [PubMed]
20. Bruzzese D, Mazzarella C, Ferro M, et al. Prostate health index vs percent free prostate-specific antigen for prostate cancer detection in men with “gray” prostate-specific antigen levels at first biopsy: systematic review and meta-analysis. Transl Res 2014;164:444-51. [Crossref] [PubMed]
21. Asawa G, Basu B, Basu S. The promise of liquid biopsies in prostate cancer: a potential point-of-care modality for precision oncology. Expert Rev Mol Diagn 2025;25:801-16. [Crossref] [PubMed]
22. Ahamed Y, Hossain M, Baral S, et al. The research progress on diagnostic indicators related to prostate-specific antigen gray-zone prostate cancer. BMC Cancer 2025;25:1264. [Crossref] [PubMed]
23. Balk SP, Ko YJ, Bubley GJ. Biology of prostate-specific antigen. J Clin Oncol 2003;21:383-91. [Crossref] [PubMed]
24. Kumar A, Mikolajczyk SD, Hill TM, et al. Different proportions of various prostate-specific antigen (PSA) and human kallikrein 2 (hK2) forms are present in noninduced and androgen-induced LNCaP cells. Prostate 2000;44:248-54. [Crossref] [PubMed]
25. Aktas S, Yücetaş U, Yücetaş E, et al. The efficacy of prostate health index (PHI) in predicting pathological progression in low-risk localized prostate cancer cases under active surveillance protocol. World J Urol 2025;43:431. [Crossref] [PubMed]
26. Tosoian JJ, Druskin SC, Andreas D, et al. Use of the Prostate Health Index for detection of prostate cancer: results from a large academic practice. Prostate Cancer Prostatic Dis 2017;20:228-33. [Crossref] [PubMed]
27. Catalona WJ, Partin AW, Sanda MG, et al. A multicenter study of [-2]pro-prostate specific antigen combined with prostate specific antigen and free prostate specific antigen for prostate cancer detection in the 2.0 to 10.0 ng/ml prostate specific antigen range. J Urol 2011;185:1650-5. [Crossref] [PubMed]
28. Lamy PJ, Allory Y, Gauchez AS, et al. Prognostic Biomarkers Used for Localised Prostate Cancer Management: A Systematic Review. Eur Urol Focus 2018;4:790-803. [Crossref] [PubMed]
29. Sharma S, Zapatero-Rodríguez J, O’Kennedy R. Prostate cancer diagnostics: Clinical challenges and the ongoing need for disruptive and effective diagnostic tools. Biotechnol Adv 2017;35:135-49. [Crossref] [PubMed]
30. Santiago I, Rivas JG. Re: EAU-EANM-ESTRO-ESUR-ISUP-SIOG Guidelines on Prostate Cancer. Eur Urol 2026;89:100-2. [Crossref] [PubMed]
31. Vickers AJ, Cronin AM, Aus G, et al. A panel of kallikrein markers can reduce unnecessary biopsy for prostate cancer: data from the European Randomized Study of Prostate Cancer Screening in Göteborg, Sweden. BMC Med 2008;6:19. [Crossref] [PubMed]
32. Punnen S, Pavan N, Parekh DJ. Finding the Wolf in Sheep’s Clothing: The 4Kscore Is a Novel Blood Test That Can Accurately Identify the Risk of Aggressive Prostate Cancer. Rev Urol 2015;17:3-13.
33. Nordström T, Vickers A, Assel M, et al. Comparison Between the Four-kallikrein Panel and Prostate Health Index for Predicting Prostate Cancer. Eur Urol 2015;68:139-46. [Crossref] [PubMed]
34. Grönberg H, Adolfsson J, Aly M, et al. Prostate cancer screening in men aged 50-69 years (STHLM3): a prospective population-based diagnostic study. Lancet Oncol 2015;16:1667-76. [Crossref] [PubMed]
35. Ström P, Nordström T, Aly M, et al. The Stockholm-3 Model for Prostate Cancer Detection: Algorithm Update, Biomarker Contribution, and Reflex Test Potential. Eur Urol 2018;74:204-10. [Crossref] [PubMed]
36. Vigneswaran HT, Eklund M, Discacciati A, et al. Stockholm3 in a Multiethnic Cohort for Prostate Cancer Detection (SEPTA): A Prospective Multicentered Trial. J Clin Oncol 2024;42:3806-16. [Crossref] [PubMed]
37. Ferro M, De Cobelli O, Crocetto F, et al. Lower PHI, [-2]proPSA/fPSA and testosterone/estradiol ratios in healthy black men: preliminary results and potential implications in prostate cancer clinical management. Minerva Urol Nephrol 2025;77:52-6. [Crossref] [PubMed]
38. Klocker H, Golding B, Weber S, et al. Development and validation of a novel multivariate risk score to guide biopsy decision for the diagnosis of clinically significant prostate cancer. BJUI Compass 2020;1:15-20. [Crossref] [PubMed]
39. Wang R, Qu Z, Lv Y, et al. Important Roles of PI3K/AKT Signaling Pathway and Relevant Inhibitors in Prostate Cancer Progression. Cancer Med 2024;13:e70354. [Crossref] [PubMed]
40. Morote J, Triquell M, Campistol M, et al. Stratifying the initial prostate cancer suspicion to avoid magnetic resonance exams by sequencing men according to serum prostate-specific antigen, digital rectal examination and the prostate-specific antigen density based on digital rectal prostate volume category. BJUI Compass 2023;4:266-8. [Crossref] [PubMed]
41. Klein EA, Partin A, Lotan Y, et al. Clinical validation of IsoPSA, a single parameter, structure-focused assay for improved detection of prostate cancer: A prospective, multicenter study. Urol Oncol 2022;40:408.
42. Wang B, Benidir T, Dortche K, et al. IsoPSA density improves risk stratification and biopsy decision-making for clinically significant prostate cancer. Urol Oncol 2026;44:70.
43. Abdallah N, Campbell RA, Benidir T, et al. Low Baseline IsoPSA Index is Associated With a Prolonged Low Risk of Clinically Significant Prostate Cancer Diagnosis in Men With an Elevated PSA. Urology 2025;201:69-75. [Crossref] [PubMed]
44. Scovell JM, Stovsky M, Partin A, et al. IsoPSA Performance Characteristics are Unaffected by 5-Alpha Reductase Inhibitors or Alpha-Blockers: Results From the IsoPSA Validation Study. Urology 2023;175:132-6. [Crossref] [PubMed]
45. Benidir T, Lone Z, Wood A, et al. Using IsoPSA With Prostate Imaging Reporting and Data System Score May Help Refine Biopsy Decision Making in Patients With Elevated PSA. Urology 2023;176:115-20. [Crossref] [PubMed]
46. Scovell JM, Hettel D, Abouassaly R, et al. IsoPSA® Reduces Provider Recommendations for Biopsy and Magnetic Resonance Imaging in Men with Total Prostate Specific Antigen ≥4 ng/ml: A Real-World Observational Clinical Utility Study. Urol Pract 2022;9:173-80. [Crossref] [PubMed]
47. Benidir T, Hofmann M, Lone Z, et al. Elevated IsoPSA Selects for Clinically Significant Prostate Cancer Without a Preference for Any Particular Adverse Histopathologic or Radiographic Feature. Urology 2022;168:150-5. [Crossref] [PubMed]
48. Lotan Y, Stovsky M, Rochelle R, et al. Decision Analysis Model Comparing Cost of IsoPSA™ vs Repeat Biopsy for Detection of Clinically Significant Prostate Cancer in Men with Previous Negative Findings on Biopsy. Urol Pract 2021;8:40-6. [Crossref] [PubMed]
49. Stovsky M, Klein EA, Chait A, et al. Clinical Validation of IsoPSA™, a Single Parameter, Structure Based Assay for Improved Detection of High Grade Prostate Cancer. J Urol 2019;201:1115-20. [Crossref] [PubMed]
50. Klein EA, Chait A, Hafron JM, et al. The Single-parameter, Structure-based IsoPSA Assay Demonstrates Improved Diagnostic Accuracy for Detection of Any Prostate Cancer and High-grade Prostate Cancer Compared to a Concentration-based Assay of Total Prostate-specific Antigen: A Preliminary Report. Eur Urol 2017;72:942-9. [Crossref] [PubMed]
51. Crocetto F, Musone M, Chianese S, et al. Blood and urine-based biomarkers in prostate cancer: Current advances, clinical applications, and future directions. J Liq Biopsy 2025;9:100305. [Crossref] [PubMed]
52. Ye C, Ho JN, Kim DH, et al. The Prostate Health Index and multi-parametric MRI improve diagnostic accuracy of detecting prostate cancer in Asian populations. Investig Clin Urol 2022;63:631-8. [Crossref] [PubMed]
53. Babajide R, Carbunaru S, Nettey OS, et al. Performance of Prostate Health Index in Biopsy Naïve Black Men. J Urol 2021;205:718-24. [Crossref] [PubMed]
54. Chiu PK, Ng CF, Semjonow A, et al. A Multicentre Evaluation of the Role of the Prostate Health Index (PHI) in Regions with Differing Prevalence of Prostate Cancer: Adjustment of PHI Reference Ranges is Needed for European and Asian Settings. Eur Urol 2019;75:558-61. [Crossref] [PubMed]
55. Hsieh PF, Li WJ, Lin WC, et al. Combining prostate health index and multiparametric magnetic resonance imaging in the diagnosis of clinically significant prostate cancer in an Asian population. World J Urol 2020;38:1207-14. [Crossref] [PubMed]
56. Ng CF, Chiu PK, Lam NY, et al. The Prostate Health Index in predicting initial prostate biopsy outcomes in Asian men with prostate-specific antigen levels of 4-10 ng/mL. Int Urol Nephrol 2014;46:711-7. [Crossref] [PubMed]
57. Tan LG, Tan YK, Tai BC, et al. Prospective validation of %p2PSA and the Prostate Health Index, in prostate cancer detection in initial prostate biopsies of Asian men, with total PSA 4-10 ng ml-1. Asian J Androl 2017;19:286-90. [Crossref] [PubMed]
58. Wang A, Van Blarigan EL, Cheng I, et al. Race and Ethnicity, Lifestyle, Diet, and Survival in Patients With Prostate Cancer. JAMA Netw Open 2025;8:e2460785. [Crossref] [PubMed]
59. Satturwar S, Parwani AV. Artificial Intelligence-Enabled Prostate Cancer Diagnosis and Prognosis: Current State and Future Implications. Adv Anat Pathol 2024;31:136-44. [Crossref] [PubMed]
60. Agnello L, Vidali M, Salvaggio G, et al. Prostate Health Index (PHI) as a triage tool for reducing unnecessary magnetic resonance imaging (MRI) in patients at risk of prostate cancer. Clin Biochem 2024;127.
61. Kim L, Boxall N, George A, et al. Clinical utility and cost modelling of the phi test to triage referrals into image-based diagnostic services for suspected prostate cancer: the PRIM (Phi to RefIne Mri) study. BMC Med 2020;18:95. [Crossref] [PubMed]
62. Gnanapragasam VJ, Burling K, George A, et al. The Prostate Health Index adds predictive value to multi-parametric MRI in detecting significant prostate cancers in a repeat biopsy population. Sci Rep 2016;6:35364. [Crossref] [PubMed]
63. Sekine Y, Fujizuka Y, Nakazawa S, et al. Utility of Combining Prostate Health Index and Magnetic Resonance Imaging for the Diagnosis of Prostate Cancer. Int J Urol 2025;32:658-63. [Crossref] [PubMed]
64. Ferro M, Crocetto F, Bruzzese D, et al. Prostate Health Index and Multiparametric MRI: Partners in Crime Fighting Overdiagnosis and Overtreatment in Prostate Cancer. Cancers (Basel) 2021;13:4723. [Crossref] [PubMed]
65. Wang J, Chen M, Guo S, et al. Development and validation of biopsy free nomograms for predicting clinically significant prostate cancer in men with PI-RADS 4 and 5 lesions. Sci Rep 2025;15:2506. [Crossref] [PubMed]
66. Schwen ZR, Mamawala M, Tosoian JJ, et al. Prostate Health Index and multiparametric magnetic resonance imaging to predict prostate cancer grade reclassification in active surveillance. BJU Int 2020;126:373-8. [Crossref] [PubMed]
67. Huang YP, Lin TP, Shen SH, et al. Combining prostate health index and multiparametric magnetic resonance imaging may better predict extraprostatic extension after radical prostatectomy. J Chin Med Assoc 2023;86:52-6. [Crossref] [PubMed]
68. Lee IT, Hou CM, Vo TTT, et al. Optimizing Prostate Cancer Care: Clinical Utility of the Prostate Health Index. Prostate 2025;85:1357-68. [Crossref] [PubMed]
69. Zhou Y, Fu Q, Shao Z, et al. Nomograms Combining PHI and PI-RADS in Detecting Prostate Cancer: A Multicenter Prospective Study. J Clin Med 2023;12:339. [Crossref] [PubMed]
70. Vidiyala N, Parupathi P, Sunkishala P, et al. Artificial intelligence: a new era in prostate cancer diagnosis and treatment. Int J Pharm 2025;683:126024. [Crossref] [PubMed]
71. Gentile F, La Civita E, Ventura BD, et al. A Neural Network Model Combining [-2]proPSA, freePSA, Total PSA, Cathepsin D, and Thrombospondin-1 Showed Increased Accuracy in the Identification of Clinically Significant Prostate Cancer. Cancers (Basel) 2023;15:1355. [Crossref] [PubMed]
72. Gentile F, La Civita E, Della Ventura B, et al. A Combinatorial Neural Network Analysis Reveals a Synergistic Behaviour of Multiparametric Magnetic Resonance and Prostate Health Index in the Identification of Clinically Significant Prostate Cancer. Clin Genitourin Cancer 2022;20:e406-10. [Crossref] [PubMed]
73. Luo Z, Li J, Wang K, et al. Study of AI algorithms on mpMRI and PHI for the diagnosis of clinically significant prostate cancer. Urol Oncol 2025;43:527.
74. Lim WY, Chin WJ, Packirisamy G, et al. Advancing Prostate Cancer Diagnostics: The Potential of Paper-Based Biosensors for Point-of-Care Testing. Clin Chim Acta 2026;584:120872. [Crossref] [PubMed]
75. Akbari Nakhjavani S, Tokyay BK, Soylemez C, et al. Biosensors for prostate cancer detection. Trends Biotechnol 2023;41:1248-67. [Crossref] [PubMed]
76. Sarkar S, Gogoi M, Mahato M, et al. Biosensors for detection of prostate cancer: a review. Biomed Microdevices 2022;24:32. [Crossref] [PubMed]
77. Sharifianjazi F, Jafari Rad A, Bakhtiari A, et al. Biosensors and nanotechnology for cancer diagnosis (lung and bronchus, breast, prostate, and colon): a systematic review. Biomed Mater 2021;
78. Singh S, Gill AAS, Nlooto M, et al. Prostate cancer biomarkers detection using nanoparticles based electrochemical biosensors. Biosens Bioelectron 2019;137:213-21. [Crossref] [PubMed]
79. Ablin RJ. The need for personalized therapy and companion diagnostics in prostate cancer. Biomark Med 2011;5:281-3. [Crossref] [PubMed]
80. Ablin RJ. The United States Preventive Services Task Force recommendation against prostate-specific antigen screening--point. Cancer Epidemiol Biomarkers Prev 2012;21:391-4. [Crossref] [PubMed]
81. Ablin RJ, Haythorn MR. Screening for prostate cancer: Controversy? What controversy? Curr Oncol 2009;16:1-2.
82. Ablin RJ, Sartor O, Bennett CL. Invited commentary: Prostate cancer: doing less might be more. Comment on “Severity of comorbidity and non-prostate cancer mortality in men with early-stage prostate cancer”. Arch Intern Med 2010;170:1397-9.
83. Brett AS, Ablin RJ. Prostate-cancer screening--what the U.S. Preventive Services Task Force left out. N Engl J Med 2011;365:1949-51.