Comprehensive biochemical insights into derivatives of PSA and glycosylation-specific changes in PSA: role in diagnosis and management of prostate cancer

Abstract

Background

Prostate cancer (PCa) is the most commonly diagnosed cancer in men worldwide, and prostate-specific antigen (PSA) test is currently the standard of choice for PCa diagnosis. Globally, several methods are employed for diagnosis of PCa, and each holds a unique role and importance. Serum PSA test and Magnetic Resonance Imaging (MRI) are widely used for diagnosis of PCa; however, MRI has certain limitations and disadvantages compared to PSA test. Moreover, growing evidences suggest that solely simple testing of serum PSA levels to diagnose PCa leads to widespread overdiagnosis and overtreatment. This is because simple PSA test has certain limitations, including lack of specificity, elevation in Benign Prostatic Hyperplasia (BPH) and inability to detect a significant number of PSA-negative tumors. Additionally, serum PSA levels do not directly correlate with higher grades and stages of PCa.

Main body

The purpose of this manuscript is to provide comprehensive and up-to-date knowledge, critical perspectives, and practical applications of derivatives of PSA and glycosylation-specific changes in PSA, with the aim of minimizing unnecessary and excessive biopsy in patients suspected of having PCa. Therefore, the use of various derivatives of PSA was focused here to increase the sensitivity and specificity of serum PSA for diagnosis of PCa. In addition to PSA levels, glycan structure of PSA changes with progression of PCa, which suggest that aberrant PSA glycosylation increases as PCa progresses. Differences in the glycan structure of PSA enable to distinguish normal PSA from cancerous origin, suggesting an important biochemical application of glycan structure of PSA in the diagnosis of PCa. Thus, present study highlights the role of derivatives of PSA and glycosylation-specific changes in PSA in the diagnosis and management of PCa. Additionally, this article emphasizes integration of glycan-based assays with PSA derivatives into multi-analyte panels, paving the way for more sensitive and specific diagnosis of PCa and aiding in the management of PCa.

Conclusion

The manuscript will provide extensive information to minimize superfluous biopsies, identify patients truly needing biopsy, differentiate BPH from PCa, distinguish indolent from aggressive PCa, enable early diagnosis and treatment at early-stage, monitor treatment response and early detection of recurrence.

Introduction

According to the National Cancer Institute (NCI) and the American Cancer Society (ACS), prostate cancer (PCa) is the most common type of cancer in men, excluding skin cancer, and is reported to be the second leading cause of cancer deaths (after lung cancer) among American men. However, till 2022, there were over 3.5 million men alive in the United States who had been diagnosed with PCa at some point in their lives [Key Statistics for Prostate Cancer: cancer.org/cancer/types/prostate-cancer/about/key-statistics.html (Last Revised: January 13, 2026)]. This data shows that with timely diagnosis and treatment of prostate cancer, patients can survive, as the diagnosis of clinically significant prostate cancer (csPCa) expands treatment options and improves outcomes. Currently, various methods are employed worldwide for the diagnosis of PCa, and each diagnostic method holds its own distinct role and significance. However, among all diagnostic methods, Magnetic Resonance Imaging (MRI) and prostate-specific antigen (PSA) test are the most widely used and popular techniques for diagnosing PCa. Many physicians, hemato-pathological and diagnostic laboratories rely on MRI results to diagnose this disease [1–3]. Although, MRI is a highly effective tool for diagnosing clinically significant PCa (csPCa) and avoiding unnecessary biopsies; however, it cannot, by itself, definitively confirm the disease. In addition to being susceptible to image distortion, being costly, and requiring the expertise of a radiologist, a major limitation is that small or low-grade tumors or those scattered throughout the entire prostate, may not be easily detected by MRI [4, 5]. Additionally, MRI cannot be used to diagnose PCa perfectly at an early stage; it has a suboptimal positive predictive value (PPV), and its false-negative rate is also high. Occasionally, csPCa may go undetected; MRI cannot distinguish between prostatitis and BPH, nor between BPH and PCa. Relying solely on a negative MRI result may lead to a delayed diagnosis of aggressive PCa [3, 4, 6, 7]. Furthermore, MRI alone cannot definitively diagnose PCa or accurately distinguish indolent from aggressive PCa, it is highly effective in identifying suspicious lesions within the prostate. Suspicious lesions, identified and highlighted via MRI, exhibit entirely distinct from that of healthy prostate tissue. While these areas carry a higher probability of harboring PCa, the imaging itself does not allow for the analysis of cellular structure. Rather than serving as a standalone diagnostic tool for PCa, advanced multiparametric mpMRI, aids in prioritizing PCa treatment by guiding targeted biopsies and avoiding unnecessary procedures [8–10]. Thus, compared to the PSA test, most methods used for the diagnosis of prostate cancer (PCa) have certain limitations and disadvantages (Fig. 1). Therefore, PSA test is currently the standard of choice for PCa diagnosis [11, 12].

Fig. 1.

Compared to PSA test, disadvantages of various methods used for diagnosis of prostate cancer. The illustration shows certain limitations and disadvantages (compared to PSA test) of the different types of tests (both invasive and non-invasive) currently used to diagnose prostate cancer, for which the PSA test is currently the standard of choice for diagnosing prostate cancer

There is growing evidence that the benefits of early detection and treatment of life-threatening PCa must be weighed against the potential risks at diagnosis. Simply, the total PSA test for PCa diagnosis has some limitations, because using PSA as a biomarker alone cannot distinguish between cancer, BPH, prostatitis, and changes in serum PSA level due to other prostate manipulations (such as cycling, DRE and catheterization). As, all of these conditions lead to elevated PSA levels, for which PSA test has a high rate of false positives, resulting in repetition of tests, unwanted biopsies and surgery [13]. The PSA test for PCa diagnosis is not able to detect a significant number of PSA-negative tumors and PSA levels do not directly correlate with increasing grades and stages of PCa. PSA screening cannot differentiate between BPH and malignant prostate disease for patients with total PSA (tPSA) in the range of 2 to 10 ng/mL, this is the so-called diagnostic ‘grey zone’ [14, 15]. Though, PSA is considered as a good marker for PCa diagnosis, but is limited due to its low tumor specificity. One-third of PCa patients are associated with normal PSA levels, whereas genetic polymorphisms can increase basal PSA levels in people without PCa [16, 17]. Black men without PCa have higher PSA levels than white or Hispanic men, reflecting the higher rate of PCa diagnosis in black men. However, the diagnostic accuracy of PSA for PCa in men of different ethnic groups is unknown, and current guidelines for PSA test interpretation do not take ethnicity into account [18]. PSA found in the serum of patients with PCa exists in various molecular forms termed as isoforms of PSA (Table 1.) [19, 20]. Early diagnosis of PCa using only isoforms of PSA test is insufficient and may not provide definitive results. Therefore, various derivatives of PSA are used to increase the sensitivity and specificity of serum PSA for diagnosis of PCa (Table 1.). Derivatives of PSA have proven to be valuable tool for improving the diagnosis of prostate cancer, potentiate the diagnostic efficiency of PSA for PCa, and potentially reduce the number of unnecessary biopsies, especially when total PSA levels are within the “borderline” or “gray zone” range [21]. Further, PSA levels alone cannot distinguish which tumors will remain indolent and which will progress to metastases, this often leads to unnecessary treatments in men with indolent PCa that do not affect the quality of life or shorten life expectancy [22]. Accordingly, important information and available findings on derivatives of PSA, which play an important role in the diagnosis of PCa, have been reviewed here. In addition, glycosylation-specific changes in PSA during progression of PCa, as well as critical assessments of emerging role of glycosylation-specific changes in PSA in diagnosis and management of PCa were depicted here. The purpose of this article is to provide comprehensive, up-to-date knowledge and critical perspectives to minimize unnecessary and excessive biopsy procedures; It may be possible to identify suitable patients who really need prostate biopsy; Early diagnosis and initiation of treatment at an early stage of the disease may be possible; Differentiation between patients with BPH, indolent PCa, and aggressive PCa will be possible.

Table 1.

Members of isoforms of PSA and derivatives of PSA

Isoforms of PSA (found in serum)

Derivatives of PSA

(I) Complexed PSA or Bound PSA (cPSA) 1. PSA-ACT complex [PSA-α1-antichymotrypsin (ACT)] 2. PSA-A2M complex [PSA-α2-macroglobulin (A2M)] 3. PSA-API complex [PSA -α1-protease inhibitor (API)] 4. PSA-PCI complex [PSA-Protein C inhibitor (PCI)] 5. PSA-PZP complex [PSA -pregnancy zone protein (PZP)] (II) Free PSA or Unbound PSA (fPSA) 1. ProPSA (native and truncated) a) Native i. ProPSA (pPSA) also known as [-7]ProPSA b. Truncated i. [− 1]proPSA or [− 1]pPSA, ii. [− 2]proPSA or [-2]pPSA or p2PSA iii. [− 4]proPSA or [− 4]pPSA iv. [− 5]proPSA or [− 5]pPSA 2. Benign PSA (BPSA) 3. Intact PSA (iPSA)

1. Prostate Health Index (PHI) 2. 4 K score 3. Prostate volume 4. PSA density (PSAD) 5. PSA transition zone density (PSATZD) 6. Complexed PSA density (cPSAD) 7. Complexed PSA transition zone density (cPSADTZ) 8. PSA velocity (PSAV) 9. PSA doubling time (PSADT) 10. Age-specific PSA (PSA and age)

To prepare this manuscript, four bibliographic databases were used: PubMed, Scopus, Web of Science, and Google Scholar, to conduct a comprehensive literature survey. Only peer-reviewed research and review articles written in English were included. No specific time frame was taken in the literature survey, and included articles describing derivatives of PSA as biomarker for diagnosis of PCa and role of glycosylation-specific changes in PSA in diagnosis of PCa. Additionally, manuscripts describing use of artificial intelligence in conjunction with derivatives of PSA and glycosylation-specific changes in PSA in the diagnosis of PCa were included. The title and abstract of each article were initially reviewed manually, and then full texts of all eligible articles were manually assessed before finally included in this manuscript to ensure they met the inclusion criteria. Articles those were not peer-reviewed, articles with insufficient methodological rigor or without full text, review articles lacking essential data, and articles that were not relevant to the topic were excluded. Thus, only specific and relevant articles were selected, compiled, analyzed, and described in this review article.

PSA as a diagnostic and prognostic biomarker of prostate cancer

PSA is the most common and reliable biomarker used to predict and monitor PCa because it is a blood-based biomarker used in all aspects of screening and patient care, including risk of recurrence, post-diagnosis surveillance and treatment follow-up [23]. However, PSA has limited specificity in differentiating PCa from BPH because PSA is released into the blood circulation in both conditions. Indeed, it is commonly reported that hyperplastic and neoplastic prostate tissue produces more PSA than malignant prostate tissue, but normal PSA secretion pathway is disrupted allowing PSA to be actively secreted into the extracellular space and ultimately escape into the blood circulation in PCa [24]. In the past, serum PSA levels were considered sufficient to determine PCa risk, with a PSA level of 4.0 ng/mL being accepted as a warning level at which a man should undergo a biopsy. However, this is no longer an accurate way to determine PCa risk. Recent findings suggest that, low PSA levels predict worse outcomes in high-grade PCa, but not in low- or intermediate-grade PCa [25]. There exists a linear correlation between PSA levels at initial diagnosis and outcome, particularly in patients with low- or intermediate-grade PCa [Gleason score (GS) ≤ 7], with higher PSA levels having worse outcomes. On the other hand, in patients with high-grade PCa (GS 8–10), even low PSA levels (≤ 2.5 µg/L) predict poor outcome, whereas approximately 6% of men with high grade of PCa have even low levels of PSA [25–28]. At initial diagnosis, a patient’s PSA level is broadly consistent with outcome, but considering PSA level alone has limited prognostic accuracy. Therefore, PSA levels are combined with tumor-specific clinical and histological factors to predict outcome [29, 30]. Comprehensive PSA screening has had a significant impact on the epidemiological profile of PCa. In the United States, the age-adjusted PCa mortality rate decreased by more than 53% during PSA screening period and proportion of patients with metastases at diagnosis decreased by 80% [31]. Recent studies have shown that cumulative excess incidence (CEI) is a measure of PSA screening intensity because there is a strong relationship between CEI and reduced PCa mortality that is independent of screening utilization. The Cumulative Excess Incidence (CEI) is a vital epidemiological and malignancy assessment metric. It quantifies the number of excess cases-those exceeding a baseline incidence rate-among the total disease occurrences identified through diagnostic methods like biopsies or biomarker (PSA) testing [32]. Furthermore, genetically adjusted PSA levels may more accurately detect aggressive PCa than unadjusted PSA levels, and polygenic PSA scores may be used to further personalize PCa screening for people of different ancestries [24].

Derivatives of PSA and their role in diagnosis of prostate cancer

Only PSA test is inadequate and inconclusive for early detection of PCa. Therefore, various derivatives of PSA have been used to improve the sensitivity and specificity of serum PSA for PCa diagnosis (Table 1.) [33]. In light of this, the introduction of PSA derivatives has been reported to improve the diagnostic efficiency of PSA for PCa, thereby reducing unnecessary biopsies in men with borderline PSA levels and helping to differentiate aggressive from indolent PCa. Various PSA derivatives, including Prostate Health Index (PHI) and 4 K score have been considered as biomarkers for PCa in recent years [34, 35].

Prostate Health Index (PHI) and its role in diagnosis of prostate cancer

Indices related to p2PSA, such as Prostate Health Index (PHI) and %p2PSA serve as better tumor markers for diagnosing aggressive PCa. The PHI is a combination of three forms of PSA (total PSA, free PSA, and p2PSA) used to derive a mathematical formula to get a single score that can be used for clinical decision making. The PHI represents probability of PCa and is calculated mathematically using the equation [(p2PSA/fPSA) × √tPSA] [36, 37]. Similarly, %p2PSA is a derivative of PSA and is calculated mathematically using the equation [(p2PSA/fPSA) × 100] without using tPSA value. Both PHI and %p2PSA are used to differentiate between PCa and BPH, because higher PHI and %p2PSA values indicate a higher probability of PCa. The primary goals of clinical use of PHI are to diagnose PCa grade and reduce unnecessary biopsies in men with borderline PSA. The accuracy of tPSA, p2PSA, %freePSA, and PHI in predicting PCa at biopsy is calculated by considering the receiver operating characteristic (ROC) curve, and area under the curve (AUC) of ROC for PHI is a measure of performance. ROC curve analysis shows the greatest interest for PHI, with AUC value of 0.77, compared to 0.71 for PSA and 0.65 for %fPSA [38]. PHI appears to be critically associated with high-grade aggressive PCa (GS ≥ 7) with AUC of 0.815, because the prognostic reliability of PHI for clinically aggressive PCa is higher than that of tPSA or %fPSA [39, 40]. The optimal cut-off value of PHI for detecting aggressive PCa is 24, and the specificity of PHI is 36.0% compared to 17.2% for tPSA and 19.4% for %fPSA with a sensitivity of 95.0%. Thus, determination of PHI can avoid 36–41% of unnecessary biopsies and 17–24% of over diagnosed and indolent PCa. However, considering the PHI cut-off point, results can vary by about 15–45%, indicating that the diagnosis is rarely failed if a biopsy is avoided [41]. Additionally, elevated %p2PSA and PHI are associated with more aggressive PCa. Therefore, %p2PSA and PHI value can be used to reduce overtreatment in indolent cases or to evaluate patients with ongoing PCa progression [42]. PHI is also useful for detecting clinically significant prostate cancer (csPCa) and can help avoid unnecessary biopsies in patients with negative MRI results [43].

The 4K score and its role in diagnosis of prostate cancer

The 4K score is a noninvasive indicator like PHI for predicting combativeness of PCa. The 4K score was conceptualized and developed in the laboratories of OPKO Health, Inc. (Miami, FL, USA) taking into account the levels of four kallikreins (4K): total PSA, free PSA, intact PSA, and hk2 [44]The 4K score is used with clinical data such as age, digital rectal examination (DRE) findings, and biopsy results. Accordingly, 4K score can be described as “the levels of total PSA, free PSA, intact PSA, and hK2 are combined into an algorithm together with the patient’s age, DRE status, and previous biopsy reports to diagnose and conclude the risk of a man with high grade of PCa [45]. Like PHI, 4K score provides greater accuracy in the diagnosis of advanced PCa than PSA or %fPSA because 4K score improves the diagnostic efficiency of csPCa compared to PSA alone [46–48]. Further, the use of 4K score is preferred in accuracy for predicting PCa, as AUC for diagnosing high-risk neoplasia increased from 0.87 to 0.90, concluding an approximately 60% reduction in biopsies in men with high PSA levels [47, 49]. If a risk value of 15% for advanced PCa is accepted and there is a risk of missing about 20% of advanced tumors, adopting the 4K score would prevent 44% of biopsies. In contrast, improved precision in identifying clinically significant PCa with an AUC of 0.82 (cutoff value ≥ 9%) avoids 31.2% of biopsies [50, 51]. Both PHI and 4K score are two important derivatives of PSA help in deciding whether to have prostate biopsy, potentially reduce unnecessary procedure. Studies have shown that both the tests perform similarly in predicting PCa risk, with neither clearly superior to the other. However, there exists some important differences between the PHI and 4K score (Table 2) [47, 48, 51].

Table 2.

Comparison between PHI and 4 K score

Parameters	Prostate Health Index (PHI)	4K score	Ref.
Description	A blood test used to diagnose prostate cancer in men with PSA levels between 2 and 10 ng/mL	A blood test used to diagnose aggressive, high-grade prostate cancer (Gleason score ≥ 7)	[37, 50]
Factors considered	A combination of total PSA, free PSA, and p2PSA are used to derive a mathematical formula to get a single score	This was conceptualized by considering the levels of four kallikreins (4 K): total PSA, free PSA, intact PSA and hk2	[37, 45]
Method of calculation	This is a mathematical model and calculated using the equation [(p2PSA/fPSA) × √tPSA]	This is an algorithm developed using four kallikreins (4 K) taking into account the patient’s age, DRE results and previous biopsy reports (if any)	[37, 40, 45]
Ranges of AUC for csPCa	0.66–0.76	0.67–0.77	[40]
Use	It is used to decide whether a prostate biopsy is needed, potentially reducing the number of unnecessary biopsies	It is used to identify men who are at high risk of developing aggressive prostate cancer and who might benefit from a biopsy	[41, 46–49]
FDA Approved	Yes	No, but this is a CLIA#-certified test	[40]
Complementary information	It focuses on overall risk of prostate cancer	It focuses on the risk of aggressive and high-grade prostate cancer	[37, 45]
Cost	This is currently less expensive in the USA (approximately 80–130 USD)	This is currently more expensive in the U.S. (approximately 600–760 USD)	[40]

(^#CLIA: Clinical Laboratory Improvement Amendments)

Prostate volume and its role in diagnosis of prostate cancer

Elevated serum tPSA and fPSA levels are associated with increased prostate volume, which was observed when evaluating tPSA levels, fPSA levels and prostate volume in asymptomatic patients undergoing routine screening without a history of PCa. Among men with PSA levels ≥ 4 ng/mL during follow-up period, small prostate volume is the most important factor for early detection of PCa [52]. Prostate volume has been reported to be an independent predictor of PCa. In men with PSA levels of 10-19.9 ng/ml, the PCa rate is 40.6% when prostate volume is less than 60 ml, whereas the PCa rate is 15.1% when prostate volume is 60 ml or greater. Similarly, in men with PSA levels of 20–50 ng/mL, the PCa rate is 65.1% when prostate volume is less than 60 mL, while PCa rate is 26.8% when prostate volume is 60 ml or greater [53]. Thus, a larger prostate may have a protective effect against PCa, so that patients with a larger prostate have a better prognosis [54]. A study of patients with total serum PSA ranging from 0.22 to 7.0 ng/mL further confirmed the correlation between PSA and prostate volume, suggesting that PSA levels and prostate volume increase with age [55]. Patients with low prostate volume should be diagnosed with proper care for malignancy and prostate volume must be considered before deciding to perform a biopsy. This is because prostate volume plays an important role in diagnosing the grade of PCa in patients whose PSA levels are in the gray zone and in patients whose PSA levels are elevated in the range of 10.1–30 ng/ml [56]. In addition, serum PSA levels can appropriately determine prostate volume, which can be used for therapeutic purposes, especially in medical management. The outcome of pharmacotherapy for BPH has been shown to depend on baseline prostate volume and serum PSA assessment may be used to improve patient management even without an accurate and reliable assessment of prostate volume [57].

PSA density (PSAD) and its role in diagnosis of prostate cancer

In asymptomatic individuals, PSA alone is not sensitive or specific enough to be useful in the routine diagnosis of PCa. To improve the accuracy of serum PSA, PSA density (PSAD) is used, which is the ratio of serum PSA to prostate volume. PSAD is said to be more accurate and specific than PSA levels alone in detecting PCa [58]. Although, PCa produces more PSA per volume of tissue than BPH, PSA density (PSAD) determination usually requires an MRI or transrectal ultrasound scan report (TRUS), and PSAD measurements adjust the PSA value for prostate volume [59–61]. Using PSAD to select men for MRI to diagnose PCa could significantly reduce harm to patients by reducing overdiagnosis and unnecessary biopsies and by reducing the number of MRI scans required [62]. In clinical studies, mean PSAD of PCa was reported to be 0.581, whereas mean PSAD of BPH was 0.044 (p < 0.002), and no BPH patient had a PSAD greater than 0.117. Therefore, PSAD may be useful in differentiating BPH from PCa [63]. At moderate serum PSA levels, PSAD can be used for diagnosis and treatment of PCa because it provides unique information about the need for biopsy and likelihood of PCa [64]. PSAD can be considered as a parameter to improve the diagnosis of PCa at serum PSA values < 4 ng/mL and an improved parameter over %fPSA at tPSA levels < 4 ng/mL in PCa diagnosis with a significantly higher AUC for PSAD (0.739) compared to %fPSA (0.667). However, PSAD did not perform better than %fPSA when tPSA levels were in the range of 4 to 10 ng/mL. Thus, different cut-off values for PSAD (e.g., 0.05 in the 2 to 4 ng/mL tPSA range, 0.1 in the 4 to 10 ng/mL tPSA range, and 0.19 in the 10 to 20 ng/mL tPSA range) are considered to achieve a sensitivity of 95% [65].

The diagnosis rate of clinically significant PCa in men with PSAD 0.09–0.19 ng/mL is significantly higher when prostate volume is less than 33 ml. Accordingly, PSAD is a superior parameter for diagnosis of PCa compared to PSA only for clinically significant patients undergoing TRUS-guided biopsy, and PSAD can be favourably used to distinguish BPH and PCa [63, 66]. PSAD is also useful in predicting treatment outcome in men undergoing radical prostatectomy for clinically localized PCa, as low PSAD values are 90% accurate in predicting surgical success and high PSAD values are 67% accurate in predicting failure [67]. When TRUS results are equivocal in patients with PSA levels between 2.5 and 20.0 ng/mL, PSAD may guide biopsy decisions as the best predictor of PCa [68]. PSAD is a cost-effective and clinically simple biomarker that is preferable to PSA alone and is therefore recommended as a parameter to improve the specificity of PCa detection in men with normal DRE and median PSA values [69]. However, in young patients, PSAD has a high negative predictive value (NPV) but a low positive predictive value (PPV). A reverse trend is observed with increasing age due to higher prevalence. In older patients, PSAD threshold may be less useful in ruling out high-grade PCa, but the clinical implications of diagnosis require case-by-case evaluation [70].

PSA transition zone density (PSATZD) and its role in diagnosis of prostate cancer

The ratio between absolute value of PSA and transitional zone volume of prostate (TZV) is designated as PSA density of the transitional zone (PSATZD) [71, 72]. Diagnosing PCa at intermediate PSA levels using PSAD has its limitations because of conflicting results from multiple case studies. To improve the diagnosis of PCa using PSATZD, a comparative study of BPH and PCa with a serum PSA level of less than 10 ng/ml can be performed. Therefore, PSATZD is used as a routine tool by urologists to predict PCa in men with PSA levels of 4–10 ng/ml [73, 74]. Additionally, PSATZD is considered as a desirable and potential parameter in reducing the number of unnecessary biopsies in patients with serum PSA levels of 2.6–10.0 ng/mL [71]. Using PSATZD cut-off value as a biopsy index, duplicate biopsies can be reduced without missing any case of PCa in the PSA range of 2.1 to 4.0 ng/ml [75]. Furthermore, PSATZD can increase the sensitivity for detecting csPCa by 2–5% at commonly used specificity levels, and PSATZD is considered superior to conventional PSAD in predicting clinically significant PCa [76]. PSATZD is more effective in prostates larger than 30 cc, and %fPSA is more effective in prostates smaller than 30 cc, so use of %fPSA in combination with PSATZD increases the detection rate of PCa [77]. PSATZD is positively correlated with cancer dedifferentiation compared to PSAD and is considered a better surrogate marker for diagnosis of aggressive PCa. In men with suspected PCa, PSATZD at a threshold level of 0.22 ng/ml may be helpful in risk stratification [78]. Thus, both PSAD and PSATZD show significant predictive value in differentiating PCa and BPH. However, the predictive value of PSATZD is sufficiently accurate in stepwise logistic regression analysis [79–81].

Complexed PSA density (cPSAD) and complexed PSA transition zone density (cPSATZD) and their role in diagnosis of prostate cancer

Complexed PSA density (cPSAD) is a clinical metric used to assess the likelihood of malignancy in PCa diagnosis. It is calculated by dividing the level of complexed prostate-specific antigen (cPSA) in the blood by the total volume of the prostate. Mathematically, it is expressed as follows: cPSAD = [Serum cPSA (ng/mL) / Prostate Volume (cm³)] [82]. Similarly, complexed PSA transition zone density (cPSATZD) is an advanced clinical derivative used as a clinical diagnostic biomarker to screen for PCa and differentiate it from BPH. cPSATZD is calculated by dividing the patient’s serum cPSA level by the prostate transition zone volume (TZV). That is, cPSATZD = [Serum cPSA (ng/mL) / Prostate Transition Zone Volume (cm³)] [83]. Complexed PSA (cPSA) is more accurate than tPSA and f/t PSA in the diagnosis of PCa, as fPSA is a component of f/c PSA and may provide complementary value to cPSA [20, 84]. cPSAD and cPSATZ (Transition Zone cPSA) can be considered superior indicators for diagnosing PCa compared to cPSA. cPSAD is reported to be better than other PSA in early detection of PCa in men with total PSA between 2.5 and 20ng/ml. However, if serum cPSA and cPSAD are measured, additional fPSA testing is not necessary [82, 85]. cPSAD and cPSATZD have best specificity compared to cPSA and %fPSA. cPSA, cPSAD, and cPSATZD improve the detection of PCa in men with tPSA of 2.5 to 10.0 ng/mL, especially in those with tPSA of 2.5 to 4.0 ng/mL. cPSA, cPSAD, and cPSATZD were found to improve the specificity of PSA in early detection of PCa when PSA and its derivatives were evaluated together with prostate volume and prostate transition zone volume [83, 86]. Furthermore, cPSAD and cPSATZD are reported to demonstrate superior performance compared to conventional total PSA density in distinguishing between BPH and PCa [82, 83]. Since BPH develops almost exclusively within and expands from the transition zone, the transition zone volume (TZV) increases significantly in BPH condition. Thus, estimation of cPSATZD determines whether the PSA level is exactly proportional to benign growth (BPH) or it is abnormally elevated due to PCa [83].

PSA velocity (PSAV) and PSA doubling time (PSADT) and their role in diagnosis of prostate cancer

PSA velocity is the change in PSA levels over time and is studied in relation to PSA levels. To determine PSAV, at least three consecutive tests must be performed on samples over 18 to 24 months. According to National Comprehensive Cancer Network, PSAV of 0.35 ng/mL/year, when PSA level is ≤ 2.5 ng/mL and PSAV of 0.75 ng/mL/year, when PSA level is 4–10 ng/mL, are considered to be suspicious for PCa. However, in clinically significant advanced PCa, PSAV slightly improves the predictive accuracy of elevated PSA or DRE, but significantly increases the number of men for whom biopsy should be recommended [87]. PSA kinetics is important in understanding the prognosis of advanced or recurrent PCa. But PSA kinetics provides no information in men who have not undergone prostate treatment [88]. Therefore, it is meaningless to use PSA velocity and/or PSA doubling time (PSADT) to diagnose PCa or to analyze the prognosis in men before treatment. In locally recurrent and advanced PCa, PSA velocity is suggested to have prognostic value for metastasis-free and cancer-specific survival. This suggests that patients with advanced or recurrent PCa with rapid PSADT should be offered more aggressive treatment immediately [88, 89]. PSADT is retrospectively consistent with pathological and clinical findings, reflecting clinical importance of PSADT procedure and subsequent radical prostatectomy [90]. Diagnosis of PCa using PSA velocity may improve accuracy compared to considering PSA value alone. The elevation of PSA greater than 0.75 ng/mL/year is a concern for patients with PSA levels in the range of 4–10 ng/mL. Early PSAV (EPSAV) can be used to diagnose and refer patients for early prostate biopsy, and to determine if men can be safely monitored with less frequent PSA tests [91]. The standard concept of PSAV confuses disease severity with the interval from onset to detection. Therefore, other approaches are needed to apply to integrate information on PSA kinetics that could ultimately improve and transform appropriate detection and treatment of PCa [92].

Age-specific PSA (PSA and age) and its role in diagnosis of prostate cancer

There exists a correlation between PSA and male age, as serum PSA levels increase with male age. Therefore, normal range for PSA is higher in older men. Natural enlargement of the prostate is thought to contribute in part to this increase, which coincides with increased permeability of the prostate, allowing more PSA to be released into the bloodstream. The stromal compartment does not express PSA, but integrity of the epithelial barrier in BPH nodules might be compromised by loss of cell junctions, allowing PSA to leak into the stromal compartment, promoting BPH [93]. However, because men tend to develop BPH as they age, an age-adjusted scale has been developed. According to the Medescape reference updated on June 01, 2023, the age-adjusted scale of PSA is: (a) 0-2.5 ng/ml: normal for a man with 40–49 years, (b) 0-3.5 ng/ml: normal for a man with 50–59 years, (c) 0-4.5 ng/ml: normal for a man with 60–69 years, (d) 0-6.5 ng/ml: normal for a man with 70–79 years. Therefore, application of age-specific reference ranges can overcome some of the shortcomings of PSA testing. Age-specific reference ranges not only improve the sensitivity of PSA for detecting treatable tumors in the prostate in young men when normal PSA levels are low, but also improve the specificity of PSA testing by raising the normal PSA threshold in older men. Consideration of age can adjust the interpretation of PSA levels to account for the higher prevalence of BPH and subsequent non-life-threatening cancers in a population of older men. In this way, many unnecessary biopsies can be avoided [94]. Men older than 70 years are more likely to die from PCa at all PSA levels than younger men, suggesting that PCa remains a major problem among older men [95]. Surprisingly, more than 40% of PCa cases in black men are undetected by tests using traditional age-specific reference ranges, which maintain 95% specificity. Therefore, maintaining sensitivity at 95%, an alternative approach can be used for black men with an acceptable reduction in specificity [96]. In younger patients, age-specific PSA significantly increases the diagnostic sensitivity and also increases number of biopsies by 45%, whereas age-specific PSA increases the detection rate of clinical PCa by 20–60% in patients aged 60 years or older. Therefore, age is not an important parameter by itself [97]. Additionally, the relationship between PSA levels and age is not entirely clear. Age-based criteria are more discriminatory, meaning that higher-risk cancer levels are overlooked [98]. The utility of combining PSA with other tests at low levels suggests that the minimum risk is associated with discontinuing screening at ages 67–71 years, when PSA is low [99]. A PSA test may be needed between ages 40 and 54 if a man has at least one first-degree relative (such as a father or brother) with PCa and two or more extended relatives with PCa. However, a family history of PCa may not determine a patient’s PSA value at the time of diagnosis [100].

Glycosylation of PSA

Glycosylation is one of the most common types of post-translational modifications (PTMs) of proteins. There are two major mechanisms for cancer-related structural changes in glycans: incomplete (or truncated) synthesis and de novo synthesis. In the early stages of tumorigenesis, an incomplete biosynthetic pathway occurs due to defect in the biosynthesis of typical glycans present in normal epithelial cells. Incomplete biosynthesis leads to the synthesis of truncated structures, such as short-chain O- N-acetylgalactosamine (O-GalNAc) glycans (Tn, T, sialyl-Tn, and sialyl-T antigens). In contrast, neo-synthesis occurs extensively in advanced stages of cancer and is associated with induction of some genes involved in the glycosylation pathway, resulting in de novo expression of antigens such as Lewis blood group-related antigen (Le) and its sialylated antigens counterparts: sialyl Lewis A (sLe^a) and sialyl Lewis X (sLe^x) antigens [101]. Aberrant glycosylation in cancer occurs due to deregulated expression of glycosylation enzymes in cancer cells, and glycan alterations in PCa include changes in PSA glycosylation, tumor hypersialylation, truncated O-glycans, and increased core fucosylation [102]. PSA consists of approximately 8% (by weight) carbohydrates in the form of N-linked oligosaccharides (N-glycans) that occupy a single glycosylation site at Asn69 [103]. However, PSA has three potential sites for O-glycosylation [104]. PSA possess core 1 O-glycan and core 2 O-glycan along with sialyl Lewis X (sLe^X) in patients with malignant PCa but not in normal prostate tissue [105]. Differences in varying degrees of sialylation result in PSA charge heterogeneity [106]. Different glycosylation variants of PSA have been reported in human seminal fluid with same enzymatic activity. The major variant with an isoelectric point (pI) of 6.9 constitutes approximately 90% of the PSA, while minor variant has a pI of 7.2. At higher pI, glycosylation patterns of PSA variants have high levels of sialylation. On sialidase treatment, both PSA variants acquire identical isoelectric point (pI = 7.7) [103, 106]. PSA variants with lower pI, 56 N-glycans are associated, whereas PSA variants with higher pI, 57 N-glycans are observed [103]. Normal PSA glycans with high and low pI fractions are sialylated biantennary complex structures. PSA fractions with higher pI are monosialylated, whereas PSA fractions with lower pI are disialylated [103, 107]. Most of the PSA N-glycans are core-fucosylated and presence of GalNAc residues decreases as pI of the PSA fraction increases [103, 107, 108]. PSA variants with higher pI were found to have more core-fucosylated and/or disialylated glycans compared to PSA variants with lower pI. Additionally, highly branched glycan structures, such as tri- or tetra-antennary structures are more enriched in PSA variants with a higher pI than PSA with a lower pI [103, 107, 109]. Since progression of PCa causes significant changes in glycoproteome, tumor-secreted glycoproteins can reflect glycosylation-specific changes in cancer cells and serve as potential biomarkers.

Glycosylation-specific changes in PCa cells can modify protein function and play a role in biological processes such as cell signaling, adhesion, migration and cell metabolism. Changes in glycan profiles are often considered “hallmarks of cancer”, so cancer-related glycan modifications may aid in more accurate cancer diagnosis. Various glycosylation-specific changes occur in PCa including increased sialylation and fucosylation, increased O-GlcNAcylation [addition of an O-linked β-N-acetylglucosamine (O-GlcNAc) moiety], branched N-glycans, cryptic N-glycans, truncated O-glycans, presence of polysialic acid, N-acetylglucosamine and Lewis antigens.

Diagnosis of prostate cancer by detecting glycosylation-specific changes in PSA

Differences in PSA-derived carbohydrates obtained from serum or prostate tissue of normal and PCa patients and use of these carbohydrate differences may serve as potential diagnostic markers for PCa [110]. To develop a more reliable test than serum PSA test for PCa diagnosis, several studies have been conducted to investigate whether cancer-specific glycan signature of PSA could be used to differentiate men with PCa from those with BPH [111]. Understanding the glycan structure of PSA from normal, BPH and cancer origins may help to improve PSA assays for PCa diagnosis, because aberrant glycosylation has been reported to occur in cancer cells, but not in normal cells [102, 112]. Rapid detection methods can be used to analyze glycosylation-specific changes in PSA [101, 113]. Glycan structure of PSA changes with progression of PCa and increasing aberrant glycosylation of PSA has been reported with progression of cancer [102, 105, 114]. Differences in the PSA glycan structure make it possible to distinguish PSA from normal and PCa origin. Important glycosylation-specific changes in PSA considered in the diagnosis of PCa include sialylation, fucosylation, mannosylation, sialyl Lewis antigen and LacdiNAc (GalNAc-GlcNAc).

Diagnosis of prostate cancer by detecting sialylation and fucosylation changes in PSA

Alterations in the degree of core fucosylation and sialylation of PSA N-glycans have been reported in PCa patients compared to healthy individuals or patients with BPH. PSA derived from cancer appears to have a greater number of N-linked branch structures compared to PSA from benign prostate tissues [105]. Therefore, impaired glycosylation (primarily sialylation and fucosylation) of PSA is a prostate cancer-specific biomarker that can differentiate PCa from BPH [111, 115]. Recent studies have reported that there are two novel types of PSAs that can be used as biomarkers for PCa diagnosis. One is high molecular weight with highly branched N-glycans only seen in cancer, and the other is low molecular weight without N-glycans. In addition, multiantennary N-glycan PSA and non-N-glycosylated PSA are rarely present in normal seminal plasma, but are found abundantly in seminal plasma of PCa patients [116]. Important changes in glycosylation observed in PCa are elevated sialylation levels and overexpression of cancer-associated sialoglycans. Abnormal sialylation is associated with tumor growth, invasion, improved cell survival and onslaught of metastasis [105]. Analysis of glycosylation-specific changes in PSA can differentiate between individuals with BPH and with PCa, because serum glycoproteins of man with PCa has higher levels of α 2,3-linked sialic acid compared to serum glycoproteins of man with BPH [115]. Increased concentrations of the α 2,3-sialylated isomer of PSA have been observed as a hallmark of aggressive PCa [113, 117, 118]. Changes in the PSA glycosylation pattern are involved in neoplastic progression, as decrease in the percentage of α2,6-sialic acid is associated with PCa combativity. Sialylated glycoforms of serum PSA containing GalNAc residues are significantly increases in patients with aggressive PCa, whereas the disialylated core fucosylated biantennary structures with α2,6- sialic acid, (major PSA glycoforms of normal PSA) are greatly reduces in aggressive PCa [119]. Both free PSA and ACT conjugated PSA (PSA-ACT) have been reported in serum from patients with metastatic PCa. Both are particularly characterized by sialylated biantennary glycan structures, but presence of several multiantennary complex structures has also been observed [120]. Serum PSA from men with PCa has higher concentrations of α2,3-sialic acid bound to terminal galactose residues compared to PSA from normal, healthy individuals. This is further confirmed by looking at the results of one study suggesting that Maackia amurensis lectin (specific for α 2,3-linked sialic acid) binds to serum PSA from men with PCa more potently than to serum PSA from healthy individuals [107, 114, 120]. The applicability of α2–3 sialic acid PSA in differentiating benign or low-risk disease from high-risk PCa has been confirmed, which is proven to be a considerable promising parameter for stratifying patients who require treatment and those who tend to have an indolent course [118, 121]. In a population-based clinical study, detection of PSA glycans with α 2,3-linked sialic acid and combined with PHI in a cohort of 79 patients has been reported with 100% sensitivity and 94.7% specificity. The proposed assay was found to be superior to simple PSA assay in differentiating BPH, indolent PCa and aggressive PCa [121]. A screening method based on glycosylation-specific changes in PSA has been developed that can differentiate the α2,3- and α2,6-linked sialic acid isomers of PSA. In this method, purified PSA samples are analyzed by capillary electrophoresis-electrospray ionization coupled to mass spectrometry (CE-ESI-MS) to distinguish between α2,3- and α2,6-sialoglycopeptide isomers of PSA with high probability for PCa diagnosis [122].

Furthermore, unlike normal PSA glycans which are sialylated, all PSA glycans in LNCaP cells are neutral and have higher fucose content. About 10–15% of the fucose structures of PSA glycans from PCa are α1,2-linked to galactose to form an H2 epitope, however this structure is not present in normal PSA [107]. The concentration of fucosylated biantennary structure of the disialylated core with α 2,6-linked sialic acid in the PSA glycan are significantly decreased in aggressive PCa [119]. PSA core fucose ratio cutoff value of 0.86 can differentiate high-risk PCa from BPH with a sensitivity of 90% and a specificity of 95% [113]. A quantitative lectin immunoassay using Lens culinaris agglutinins (LCA) and Aleuria aurantia lectin (AAL) was developed to evaluate the levels of fucosylated glycans of PSA in the serum of PCa patients with different Gleason scores. Both the LCA and AAL immunoassays confirmed elevated levels of fucosylated serum PSA in PCa patients, which were simultaneously observed to correlate with higher Gleason scores. Thus, determination of the fucosylation form of PSA can be considered as an important biomarker to differentiate between aggressive and indolent PCa [123]. Significant reduction of core fucose and increased percentage of α2,3-sialic acid in PSA in high-risk PCa patients differentiates patient with PCa from BPH, and patient with indolent PCa from aggressive PCa [113]. Therefore, a model has been designed to detect advanced PCa using α2,3-sialylated PSA and α1,6-fucosylated PSA simultaneously, which were analyzed by a microfluidic electrophoresis system using two different lectins, MAA and PhoSL [124].

Diagnosis of prostate cancer by detecting mannosylation changes in PSA

Mannosylation of PSA changes as PCa develops, and these changes can be used to diagnose and stage PCa. PSA mannose levels decrease as PCa progresses; therefore, PSA mannose levels can be used as a diagnostic biomarker. Lectin-based assays have been reported that PSA mannose levels were lower in patients with PCa than in patients with BPH [114, 125, 126]. Additionally, an independent study reported that high-mannose structures in Asn-301 are present in prostatic fluid at different stages of PCa [127]. In aggressive PCa, the enzyme α-mannosidase 1 A is mislocalized within the Golgi apparatus. This leads to incomplete processing of N-glycans, resulting in high-mannose structures to accumulate on the cell surface and on secreted glycoproteins such as PSA. Therefore, higher levels of high-mannose glycans in PSA are associated with aggressive PCa. This change serves as a potential biomarker to differentiate aggressive from indolent PCa [128, 129].

Diagnosis of prostate cancer by detecting Sialyl Lewis antigen changes in PSA

Changes in sialylated glycans, including sialyl Lewis X (sLe^X), have been associated with PCa and may be used in the diagnosis of PCa. The presence of Lewis X antigen group in PSA has been confirmed, which can be used as a biomarker for the diagnosis of PCa [116]. Over expression of GCNT1 is observed in PCa and is associated with higher levels of core 2 O-sLe^X in PSA. Higher rates of O-linked glycosylation with sLe^X are observed in serum PSA from patients with malignant PCa compared to healthy men. In PCa, an increase in the level of sLe^X in PSA has been detected. Elevated levels of sLe^X on core 2 O-glycans expressed on PSA are associated with increased metastatic potential in PCa [105]. The sLe^X may influence PCa progression through its binding to E-selectin, evading Natural Killer (NK) cell immunity and promoting bone metastasis. Elevated sLe^X levels in PCa are associated with poor prognosis, aggressive and hormone-resistant PCa [128]. Therefore, sLe^X expression in PSA is considered a candidate prognostic marker. Similarly, sLe^a expression levels in PSA increase in correlation with the aggression of PCa cells (LNCaP< DU145) and tissue, but are absent in normal prostate cells (RWPE-1) and tissues. The sLe^a-containing glycoproteins (e.g., PSA) are predominantly expressed in aggressive PCa tissues and are thought to contribute to metastatic properties [130].

Diagnosis of prostate cancer by detecting truncated O-glycans change in PSA

The incomplete biosynthetic pathway typical of early tumor stages results from defective synthesis of typical glycans present in normal epithelial cells, resulting in the synthesis of truncated structures such as short-chain O-GalNAc glycans (Tn, T antigens, sialyl-Tn and sialyl-T antigen) [101]. Truncated O-glycans attract carbohydrate-binding proteins, which may contribute to tumor growth and metastasis. In PCa tissues, truncated O-glycans are found to be associated with changes in PSA glycosylation, where biosynthesis of O-linked glycans is intiated by GALNT7, a glycosyltransferase. Therefore, upregulation of GALNT7 levels in PCa cells enhances tumor growth and promotes cell invasion [102]. Increased expression of truncated O-glycans in PSA is a hallmark of PCa and is associated with poor prognosis of PCa. Additionally, truncated O-glycans in PSA are expressed at higher levels in PCa compared to BPH due to changes in PSA glycosylation. Thus, change in truncated O-glycans in PSA can be used to diagnosis and distinguish PCa from BPH [101, 102, 105].

Diagnosis of prostate cancer by detecting LacdiNAc (GalNAc-GlcNAc) changes in PSA

LacdiNAc-glycosylated PSA is significantly superior to conventional PSA testing in identifying patients with csPCa [131]. The level of GalNAc is up to 65% in PSA glycans of PCa origin (LNCaP), whereas it is present only up to 25% in normal PSA glycans [107]. Expression of ST6GalNAc1 is upregulated in primary prostate tumors, whereas it is relatively downregulated in metastatic PCa. These results are consistent with “reprogramming” of androgen receptor (AR) genomic action during PCa progression, indicating that ST6GalNAc1 may have a short-lived role in tumor progression [132, 133]. Concentrations of serum PSA with sialylated glycans carrying the GalNAc moieties GalNAcβ1→4GlcNAc (LacdiNAc) have been found to be increased in aggressive PCa [107]. Although occurrence of the GalNAcβ1→4GlcNAc (LacdiNAc) structure is rarely found in normal individuals, it is significantly increased in PCa. Several studies have reported the appearance of LacdiNAc clusters in PSA using Wisteria floribunda agglutinin (WFA) [134–136]. Further, Trichosanthes japonica agglutinin-II (TJA-II) interacts with glycans terminated at GalNAcβ1→4GlcNAc (LacdiNAc) in addition to α-1,2-linked fucose residues. Based on this, it was reported that the binding of serum PSA from man with PCa to TJA-II was significantly increased compared to BPH, suggesting that LacdiNAc expression increased in PSA glycan in PCa patient [137]. Significantly higher ratios of TJAII-bound PSA to total PSA have been reported in the serum of men with PCa compared to men with BPH, which is equal to 8.3 ± 5.6% and 1.0 ± 0.55%, respectively. Therefore, PSA binding ratios in different individuals with TJA-II can be used to differentiate between BPH patients and PCa patients [125, 138]. The β4 binding of N-acetyl galactosamine to N-acetyl glucosamine to form LacdiNAc units (GalNAcβ1→4GlcNAc) results in the terminal modification of N-glycan. β4-N-acetyl galactosaminyltransferases (β4GalNAcTs) - β4GalNAcT3 and β4GalNAcT4 are involved in the biosynthesis of human LacdiNAc motifs. N-glycans of PSA purified from semen plasma of healthy individuals and from LNCaP cells show the LacdiNAc structure, however the occurrence of LacdiNAc structures are observed to be increased in the PSA of cancer cells [107, 109]. Interestingly, β4GalNAcT4 is observed to increase in PCa, whereas no change is observed for β4GalNAcT3 [134, 138]. Overexpression of LacdiNAc epitope in PSA from men with PCa has been shown to correlate with increased enzyme activity of PSA, for which PSA enhances PCa growth rate [101, 138–140]. Partially sialylated one LacdiNAc unit and an elevated level of similar biantennary complex-type glycans have been found in the PSA of man with PCa, whereas glycans of PSA from man with BPH predominantly contain terminally sialylated, complex-type biantennary N-glycans [101, 134]. Therefore, glycoprofiling of PSA can be used to analyze glycosylation-specific changes in PSA for accurate diagnosis and prognosis of PCa [103, 125].

Thus, detection of glycosylation-specific changes in PSA can be used to diagnose patients with PCa. Furthermore, by detecting glycosylation-specific changes in PSA, it is possible to distinguish between healthy individuals and men with prostate diseases (benign or malignant), differentiate between patients with PCa and those with BPH, and distinguish between patients with high-risk PCa and those with indolent PCa (Intermediate risk / Low risk). Details are described in Tables 3, 4 and 5.

Table 3.

Differentiation between normal verses cancerous prostate based on glycan analysis of PSA

Glycosylation-specific changes which distinguishes normal and cancerous prostate	Ref.
1. Normal PSA glycans are sialylated, whereas cancerous PSAs are all neutral and have higher fucose content 2. In normal PSA glycans, GalNAc present only up to 25% of the structure, but in cancerous states, GalNAc present up to 65% 3. Compared to normal PSA glycans, decreased sialic acid content and increased fucosylation and N-acetylglucosamine content are observed in cancerous PSA glycans	[107]
4. Increase in O-linked glycosylation with sialyl lewis X (SLex) antigen is observed in PSA from malignant prostate compared to adjacent normal prostate tissues 5. Higher core 2 O-linked SLex is observed in PSA glycans of malignant prostate tissue compared to normal prostate tissue 6. Serum PSA glycans possess core 1 O-glycan and core 2 O-glycan with sialyl Lewis X (sLeX) in patients with malignant prostate cancer but not in normal prostate tissue	[105]
7. Higher concentrations of α 2, 3-linked sialic acid linked to the terminal galactose residues are observed in serum PSA glycans from man with prostate cancer compared to normal and healthy individuals	[107, 120]
8. High molecular weight PSA with highly branched N-glycans are observed in prostate cancer only, whereas low molecular weight PSA without branched N-glycans are found in normal prostate 9. Multiantennary N-glycan PSA and non-N-glycosylated PSA are almost absent in normal seminal plasma but are abundantly found in seminal plasma of prostate cancer patients	[118]
10. The occurrence of GalNAc1→4GlcNAc (LacdiNAc) structures are elevated significantly in prostate cancer, although hardly found in normal individuals	[134, 135]
11. Compared to normal, β4GalNAcT4 increases in prostate cancer, but no changes are observed for β4GalNAcT3	[134, 138]
12. In aggressive PCa patients Serum PSA sialylated glycoforms containing GalNAc residues are largely increases, whereas the disialylated core fucosylated biantennary structures with α2,6-SA are markedly reduce in aggressive PCa	[119]

Lower values mean, lower than normal healthy men (same age and ethnicity), and higher values mean, higher than normal healthy men (same age and ethnicity)

Table 4.

Differentiation between Malignant Prostate Cancer and Benign Prostatic Hyperplasia (BPH) based on glycan analysis of PSA

Glycosylation-specific changes which distinguishes cancerous prostate and prostate with BPH	Ref.
1. Significantly higher concentration of α2,3-linked sialic acid is observed in serum PSA of prostate cancer patients compared to serum of BPH patients	[107, 114, 115, 120]
2. The binding of TJA-II to serum PSA of prostate cancer patients is significantly increases compared to BPH due to increased expression of LacdiNAc on PSA glycans in prostate cancer patients	[137]
3. Partially sialylated one LacdiNAc unit and higher level of similar biantennary complex-type glycans are found in serum PSA of prostate cancer patients, whereas, terminally sialylated, complex-type biantennary N-glycans are predominately present in PSA of BPH patients	[101, 134]
4. Significantly lower proportions of α2,6-sialic acid are observed in the serum PSA of prostate cancer patients compared to BPH patients	[113, 118, 121]

Lower values mean, lower than normal healthy men (same age and ethnicity), and higher values mean, higher than normal healthy men (same age and ethnicity)

Table 5.

Differentiation between aggressive prostate cancer (High Risk) verses indolent prostate cancer (Intermediate risk / Low risk) based on glycan analysis of PSA

Glycosylation-specific changes which distinguishes patients with high risk prostate cancer from indolent prostate cancer (Intermediate risk / Low risk)	Ref.
1. Higher concentration of the α 2,3-sialylated isomer of PSA are observed in aggressive prostate cancer (considered as a hallmark) compared to indolent (intermediate or low risk)	[113, 117, 118]
2. In PSA glycans from aggressive PCa patients, the sialylated glycoforms of serum PSA containing GalNAc residues are significantly increases, whereas the disialylated core fucosylated biantennary structures with α2,6- sialic acid (which are the major PSA glycoforms of normal PSA) are greatly reduced, compared to indolent prostate cancer 3. Concentration of sialylated glycans of serum PSA, bearing GalNAc moieties GalNAc1→4GlcNAc (LacdiNAc) are increases and the concentration of disialylated core fucosylated biantennary structures with α 2,6-linked sialic acid in glycans of PSA are significantly decreases in aggressive prostate cancer compared to indolent prostate cancer.	[119]
4. Expression of ST6GalNAc1 is down regulated in metastatic and aggressive prostate cancer, but upregulated in indolent condition	[132, 133]
5. In patients with high-risk prostate cancer (aggressive prostate cancer), a significant decrease in core fucose and an increase in the α2,3-sialic acid percentage of PSA distinguish them from indolent prostate cancer	[113]
6. Percentage of α2,6-sialic acid is significantly decrease in glycans of serum PSA from high-risk PCa patients compared to indolent PCa	[113, 118, 121]
7. In aggressive PCa, elevated levels of high-mannose glycans in PSA are observed, differentiating it from indolent PCa	[128, 129]

Lower values mean, lower than normal healthy men (same age and ethnicity), and higher values mean, higher than normal healthy men (same age and ethnicity)

Diagnosis of prostate cancer using a combination of derivatives of PSA and glycosylation-specific changes in PSA

PCa is a slow-growing cancer and PSA testing has only been available for decades for diagnosis of PCa. The development of standardized and sensitive analytical methods for measuring and analyzing PSA glycosylation patterns is critical for clinical implementation [122]. This is because there is no specific threshold value for derivative of PSA that can perfectly confirm (guaranty) the presence or absence of prostate cancer. Although, some derivatives, such as PHI, can help predict the presence of clinically significant PCa, their ability to perfectly distinguish between aggressive and non-aggressive PCa is not absolute, which can lead to missing aggressive PCa or unnecessary treatment of indolent PCa [40, 141–143]. Therefore, combining PSA glycosylation analysis with other derivatives of PSA such as prostate volume, PSA density, and [-2]proPSA may further improve diagnostic accuracy and prognostic prediction. A study developed an “SF index” that combines α2,3-sialylated PSA and α1,6-fucosylated PSA to more accurately discriminate high-grade PCa from indolent PCa compared with using PSA alone [124, 144]. Combining PSA glycan information with other biomarkers can significantly improve the accuracy of prostate cancer prediction. Integration of PSA glycoform analysis with the PHI provides higher specificity for predicting aggressive PCa compared with traditional PSA testing alone [37, 128, 142].

Furthermore, although MRI alone cannot definitively confirm the diagnosis of PCa, when combined with the results of other diagnostic tests performed for the detection of PCa, a definitive and accurate diagnosis can be possible. Derivatives of PSA and glycosylation-specific changes in PSA contribute to provide a more comprehensive risk assessment when combined with conventional clinical parameters such as imaging (MRI) and biopsy results [145]. Similarly, integrating the 4 K score with clinical variables (age, DRE status, and history of any prior biopsies) and PI-RADS scores derived from prostate MRI creates a powerful and highly accurate algorithm for diagnosing csPCa. This synergistic approach significantly enhances diagnostic accuracy and prevents unnecessary invasive prostate biopsies [146, 147]. An illustration showing flow diagram (Fig. 2) was proposed here. The flow diagram illustrates the proposed model for diagnosing PCa and differentiating between PCa patients and patients with BPH. This model is based on evaluating PSA levels, isoforms of PSA, derivatives of PSA, and glycosylation-specific changes in PSA, and also utilizes, prostate MRI results and MRI-transrectal ultrasound (TRUS) fusion biopsy data. The proposed model suggests that if a definitive and accurate diagnosis can be established by analyzing results from PSA isoform assays, PSA derivatives, glycan analysis of PSA, and prostate MRI, then painful procedures-such as standard biopsy or MRI-TRUS fusion biopsy-could be avoided. Thus, this novel model can reduce unnecessary and unwanted biopsies and improve the accurate diagnosis of csPCa. Additionally, incorporating glycan anlysis of PSA along with results of the isoforms of PSA test and derivatives of PSA, this model enables the differentiation between BPH, indolent and aggressive PCa (Fig. 2).

Fig. 2.

Proposed model for prostate cancer diagnosis and management using isoforms of PSA, derivatives of PSA and glycosylation-specific changes in PSA. The flow diagram describes the proposed model for prostate cancer diagnosis and the way of management of prostate cancer patients based on evaluation of PSA level and isoforms of PSA levels, derivatives of PSA and glycosylation-specific changes in PSA. MRI result, MRI-TRUS fusion biopsy result are also included. For the diagnosis of cancerous prostate (suspected), an MRI scan should be performed prior to the MRI-TRUS fusion biopsy, following a 4 K score analysis. If prostate cancer is definitively confirmed, the MRI-TRUS fusion biopsy may be omitted. Glycan analysis of PSA presented in Table 3. can be used for differentiation between normal verses cancerous prostate. Glycan analysis of PSA presented in Table 4. can be used for differentiation between malignant prostate cancer and benign prostatic hyperplasia (BPH). Similarly, glycan analysis of PSA presented in Table 5. can be used for differentiation between aggressive prostate cancer (High Risk) verses indolent prostate cancer (Intermediate risk / Low risk). The → represents only flow from one step to another step irrespective of colour

Recent advances in PCa diagnosis utilizing derivatives of PSA and glycosylation-specific changes in PSA

Advances in PCa diagnosis focus on improving the accuracy of the PSA test in combination with other methods to reduce overdiagnosis and overtreatment. Remarkable advancements include the integration of PSA testing with imaging techniques such as MRI and PSMA PET/CT. Using MRI with PSA helps to improve diagnosis of PCa early when it is suspected while PSMA-based screening offers high sensitivity and specificity for detection of localized and potentially extraprostatic disease. Similarly, integration of biomarkers, such as circulating tumor cells (CTCs), with PSA may improve diagnostic performance, specifically in the “PSA gray area”. The use of miRNAs can help differentiate between BPH from PCa, and can distinguish aggressive disease from indolent [148–150]. Diagnosis of PCa is evolving, moving from simply relying on total PSA levels to utilizing multimodal AI-based systems that integrate PSA derivatives and specific changes in PSA glycosylation. This approach can improve the ability to distinguish between BPH, indolent PCa, and aggressive PCa, potentially improving risk stratification and reducing unnecessary biopsies. Artificial intelligence, particularly machine learning (ML) and deep learning (DL), is used to synthesize this complex and high-dimensional data to create personalized and risk-based diagnostic tools [151–155]. AI models can integrate PSA data, imaging, genetic information, pathological data, and clinical factors to provide a more accurate and personalized risk assessment than clinical assessment or MRI alone. Identification of specific genetic changes like changes in DNA methylation patterns or signaling pathways may help to diagnose PCa and predict outcome. Analysis of the complex interactions between these data points, AI can better predict the probability of clinically significant and non-clinically significant PCa [154, 156, 157]. The integration of these markers leads to a multi-omics approach combining proteomics (PSA), glycoproteomics, and genomics data, creating “liquid biopsy” programs that provide highly accurate diagnoses and prognostic predictions. However, a challenge remains in the need for vast, diverse, and standardized datasets to ensure algorithmic accuracy and prevent bias [153]. Therefore, development of methods may allow for a more comprehensive and accurate assessment of PCa by developing algorithms that combine derivatives of PSA with glycosylation-specific changes in PSA [158]. An illustration (Fig. 3) was presented here which describes the integration of glycan-based assays with derivatives of PSA into multi-analyte panels paves the way for more sensitive and specific detection of PCa. Automation of these techniques on high-throughput platforms is crucial for implementing this complex analysis in routine clinical practice [128, 159]. The diagram schematically illustrates the process of performing complex analyses in routine clinical practice by integrating glycosylation-specific changes in PSA obtained through various glycan-based analytical methods, and derivatives of PSA obtained through various analytical methods and mathematical calculations (Fig. 3).

Fig. 3.

This figure illustrates integration of glycan based assays of PSA with derivatives of PSA to differentiate BPH, indolent and aggressive PCa. The illustration shows different methods for analysis of glycosylation-specific changes in PSA and estimation of the amount of different types of glycosylation-specific changes in PSA. The figure also illustrates different methods for analysis of derivatives of PSA and estimation of the amount of different types of derivatives of PSA. Integration of glycosylation-specific changes in PSA with derivatives of PSA using multi-analyte panels and translation to high-throughput automated platforms for AI based analysis to obtain results to clinically differentiate BPH, indolent and aggressive PCa

Derivatives of PSA and glycosylation-specific changes in PSA in the management of prostate cancer

PSA is described as a biomarker because it is objectively measured as an indicator of an individual having a normal biological state, a pathogenic state, or a pharmacological response to a treatment intervention [160]. In addition to the use of PSA for PCa diagnosis, PSA estimation is performed to evaluate a patient’s response to first-line therapy, such as surgery, radiation therapy endocrine therapy and for long-term follow-up of the patient to detect recurrence or progression of the disease. Laboratory medicine, specifically through the use of PSA, plays an important role in the early diagnosis and management of patients at later stage of the PCa [161]. One of the biggest challenges for doctors is diagnosing a PCa patient who needs treatment, especially when the disease is treatable. Therefore, increase in diagnosis of PCa by using PSA and DRE of moderately differentiated organ obstruction cancer are motivating for further reductions in prostate cancer-related mortality. PSA test will continue to become an essential component in assessing response to radical and endocrine therapy, and will provide important prognostic information in the follow-up of these patients to detect relapse/progression [162, 163]. Patient outcomes depend primarily on early diagnosis, but randomized clinical trials for PCa screening have yielded conflicting results. Overtreatment of early-stage PCa leads to unnecessary biopsy and active treatment for elderly patients with indolent PCa has become a serious problem. Evaluating PSA alone or as part of a derivative test such as PHI is one of the keystone of current PCa management philosophy. The use of PSA as a biomarker in PCa, can be used as a model for integrating biomarkers into management strategies for other cancers [164, 165]. The PSA test has significantly improved survival and management of PCa. However, knowledge about the correlation between PSA assessment and disease outcome is still lacking. The emergence of new data may help further refine our understanding of the use of PSA in the detection, staging and treatment of PCa patients [31, 166]. Thus, it can be said that PSA derivatives play a crucial role in the management of PCa by acting as a primary diagnostic tool, monitoring disease progression after treatment, detecting disease recurrence and allowing for early intervention and the development of individualized treatment plans. Additionally, critical assessments of glycosylation-specific changes in PSA contribute to the management of PCa by providing a more accurate method to differentiate BPH from indolent and aggressive PCa, and monitoring disease progression and treatment response.

Derivatives of PSA and glycosylation-specific changes in PSA in monitoring progression and metastasis of prostate cancer

In the above sections, the roles of derivatives of PSA and glycosylation-specific changes in PSA in the diagnosis of PCa, monitoring progression and aggressiveness of PCa, differentiating patients with PCa from BPH, and distinguishing between indolent and aggressive PCa were described extensively. In addition to the detail description described above, some more specific details were described here to support the notion that derivatives of PSA and glycosylation-specific changes in PSA play a crucial role in monitoring the progression and metastasis of PCa.

A rapid increase in PSA levels after radical prostatectomy indicates metastatic disease, and a high PHI score shows a strong correlation with the presence of aggressive tumors and the likelihood of metastatic PCa [20, 39]. Similarly, higher PSAD value indicates more aggressive and there is a greater risk of PCa that it is not confined to a single organ. A rapid increase in PSADT after prostatectomy is more likely to indicate metastatic PCa than a gradual increase [89]. It was observed that higher levels of derivatives of PSA and glycosylation-specific changes in PSA (e.g., PHI, %p2PSA, PSA α2,3-sialylation) associated with PCa aggressiveness showed a significant positive correlation with higher Gleason scores (generally ≥ 7) and more advanced stages of PCa [119]. Combining derivatives of PSA and glycosylation-specific changes in PSA with Gleason score and clinical stage information can improve the accuracy of predicting tumor aggressiveness, helping clinicians choose between active surveillance and more aggressive treatment options [113, 167, 168]. Glycosylation-specific changes in PSA, measured using advanced techniques such as mass spectrometry and lectin-based assays, provide valuable prognostic information and can help identify patients at risk of metastasis [101, 169].

Derivatives of PSA and glycosylation-specific changes in PSA in deciding treatment plan of a patient with prostate cancer

Many discrepancies in PSA concentrations due to different assays may influence treatment decisions [31, 170, 171]. Several improvements of the PSA test have been proposed to increase the specificity and reduce number of unnecessary prostate biopsies. Ongoing large multicenter screening and outcome studies provide important information to further improve the use of PSA in the detection and treatment of PCa [172]. By understanding a patient’s tumor-specific glycosylation profile, physicians can develop treatment strategies to each patient’s needs [173, 174]. Derivatives of PSA such as PHI can help determine the most appropriate treatment strategy for each patient. Glycosylation-specific changes in PSA can aid in risk stratification and influence treatment decisions, such as choosing between active surveillance, prostatectomy, or radiation therapy [40, 175, 176]. Combination of elevated PHI/p2PSA ratio and patterns of glycosylation-specific changes in PSA increases the confidence of clinicians in recommending a biopsy [101, 113, 177]. In patients undergoing active surveillance, specific changes in PSA derivatives and PSA glycosylation may help identify patients, whose disease is progressing to a more aggressive form, thus necessitating a change in treatment plan [101, 113, 141].

Derivatives of PSA and glycosylation-specific changes in PSA in monitoring response to therapy

Monitoring PSA levels during follow-up after treatment is crucial and can serve as an indicator of treatment success or the likelihood of recurrence. Both derivatives of PSA and glycosylation-specific changes in PSA play a crucial role in monitoring response to prostate cancer therapy, as they provide higher specificity and prognostic information than total PSA alone. A significant decrease in PSA levels after treatment (surgery, radiation therapy) indicates successful treatment, while no decrease or an increase suggests treatment resistance or disease persistent. Glycosylation-specific changes in PSA are emerging as more specific and sensitive tools than total PSA for monitoring treatment effectiveness. If a particular glycosylation-specific change in PSA is associated with a particular response to treatment, then changes in that pattern could use as a biomarker to monitor the effectiveness of therapy [101, 128]. Assays specifically targeting cancer-related glycan patterns have showed high accuracy in distinguishing between treated and untreated patients and demonstrated superior performance compared to tPSA levels alone in monitoring treatment effectiveness. One “Glycan Index Assay” (GIA) demonstrated significantly higher AUC values than tPSA in treatment monitoring. Analysis of derivatives of PSA and glycosylation-specific changes in PSA can provide crucial information about the aggressiveness and biological behavior of remaining or recurrent cancer cells, allowing for better monitoring of treatment response and improved clinical decision-making [101, 169, 178]. PSAV and PSADT are crucial metrics in the management of PCa, particularly for patients on active surveillance or following primary treatment (radical prostatectomy or radiation), as a rapid increase in PSAV and PSADT values indicate disease progression [179].

Derivatives of PSA and glycosylation-specific changes in PSA in monitoring recurrence of prostate cancer

Total PSA testing is not interchangeable and could potentially have serious implications for clinical practice. Continuously determined PSA levels are so important to detect disease recurrence that persistent elevations are termed “biochemical relapse” and have important implications for the management of PCa patients [31, 170, 171]. An increase in PSA levels after treatment also indicates biochemical recurrence of PCa. Glycosylation-specific changes in PSA are more specific and sensitive tools than total PSA for predicting the risk of recurrence in PCa patients, because which patients are at higher risk of recurrence can be predicted by studying glycosylation-specific changes in PSA [101, 128]. PSAV and PSADT values provide the most useful information after treatment or during active surveillance for detecting recurrence, because a rapid increase in PSA levels in these patients indicates the need for further investigation or treatment. Following curative treatments such as prostatectomy, PSA levels should decrease to undetectable levels, and detectable levels of free PSA indicate residual disease. Subsequent biochemical recurrence (BCR), where PSA levels rise again, is typically the first sign of recurrence [88, 179]. Post treatment analysis of derivatives of PSA (e.g., PSADT and PHI) and glycosylation-specific changes in PSA allows for more accurate identification of PCa patients with clinically significant and aggressive recurrence, enabling more informed treatment decisions. Increased GCNT1 expression is associated with PCa progression and is a predictive factor for disease relapse after radical prostatectomy [101, 113]. A shorter PSADT after treatment (e.g., less than 6 months) indicates more aggressive recurrence and a poorer prognosis, often leading to early intervention with salvage therapy [180].

Future prospectus and conclusion

Currently, the possibility of early detection of PCa is increasing due to generalization of PSA screening, for which the number of patients with diagnosable metastatic disease is decreasing [160, 181, 182]. Because of the inherent limitations of PSA as a biomarker, the PCa biomarker field is advancing rapidly. The possibility of future use of various factors such as blood biomarkers (e.g. PHI, 4 K score, etc.), histological and clinical criteria, and gene expression profiles in some special cases are likely to be used to determine tumor aggressiveness and predict patient outcome [183]. To reduce overtreatment, approaches such as active surveillance can be used to separate diagnosis and treatment of aggressive, indolent and very-low-risk PCa. Introduction of new methods are necessary to improve the diagnosis of high-risk prostate cancer and to accurately identify which men will benefit from treatment and which will not, for example through active surveillance. Recent advances in multiparametric MRI, new biomarkers, and machine learning (ML) models promise to transform the diagnosis of PCa [184, 185]. Our understanding of the PCa biology has grown tremendously in recent years due to advent of advanced technologies in the fields of genomics and proteomics. As a result, PCa biomarker research has expanded with the discovery of several promising alternative biomarkers for PCa diagnosis [186]. Diagnosis and management of PCa have changed significantly over the past few decades. However, more innovative achievements are expected in the near future for diagnosis, prognosis and management of PCa. This manuscript presented a comprehensive elucidation highlighting the role of derivatives of PSA and glycosylation-specific changes in PSA in the diagnosis and management of PCa. Therefore, present study is expected to contribute to a better understanding of the emerging role of PSA as a biomarker for PCa diagnosis. So that number of unnecessary and inappropriate biopsies can be minimized, patients eligible for prostate biopsy can be identified, early diagnosis and treatment initiation at an early stage of disease will be possible. In addition, the ability to distinguish between BPH, indolent PCa and aggressive PCa will allow for more reliable and effective treatment planning. Monitoring treatment response and early detection of recurrence in patients with PCa can be possible.

Abbreviations

PCa

Prostate cancer

PSA

Prostate-specific antigen

BPH

Benign Prostatic Hyperplasia

tPSA

Total PSA

cPSA

Bound or complexed PSA

fPSA

Free or uncomplexed PSA

pPSA

ProPSA

BPSA

Benign PSA

iPSA

Inactive PSA

CEI

Cumulative excess incidence

PHI

Prostate Health Index

ROC

Receiver operating characteristic

AUC

Area under the curve

4K score

Four kallikreins score

FDA

Food and drug administration

CLIA

Clinical laboratory improvement amendments

DRE

Digital rectal examination

csPCa

Clinically significant PCa

GS

Gleason Score

PSAD

PSA density

TRUS

Transrectal ultrasound

NPV

Negative predictive value

PPV

Positive predictive value

PSATZD

PSA transition zone density

TZV

Transitional zone volume of prostate

cPSAD

Complexed PSA density

cPSADTZ

Complexed PSA transition zone density

PSAV

PSA velocity

EPSAV

Early PSA velocity

PSADT

PSA doubling time

O-GalNAc

O- N-acetylgalactosamine

sLe^a

Sialyl Lewis A

sLe^x

Sialyl Lewis X

pI

Isoelectric point

LCA

Lens culinaris agglutinins

AAL

Aleuria aurantia lectin

WFA

Wisteria floribunda agglutinin

BCR

Biochemical recurrence

PI-RADS

Prostate imaging-reporting and data system

Author contributions

LD conceived idea of this study, performed literature search and analysis, prepared the figures and tables, and wrote the manuscript.

Funding

Open access funding provided by Siksha 'O' Anusandhan (Deemed To Be University). No funding was used for this study.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.