Introduction

Prostate cancer (PCa) is the most commonly diagnosed and second-leading cause of cancer-specific mortality among men in the United States [1]. Patients are screened for PCa with serum prostate-specific antigen (PSA), which offers value in both diagnosis and risk stratification [2]. Recent technological improvements have altered diagnosis and staging strategies for the disease, including multiparametric magnetic resonance imaging (mpMRI), and prostate-specific membrane antigen (PSMA) positron emission tomography (PET) [3, 4]. However, prostate needle biopsy remains the standard diagnostic test [5, 6]. Increasing evidence has supported the use MRI- ultrasound-guided fusion prostate biopsy to improve the accuracy of Gleason score (GS) findings and thus PCa risk stratification [7–9]. Nonetheless, prostate biopsy only samples a small proportion of the prostate and is prone to error. Despite refinements in biopsy technique, patients commonly experience GS discordance and upgrade between biopsy and final pathology on radical prostatectomy (RP). GS upgrading (GSU), defined as an increase in Gleason grade group (GGG) by ≥ 1 or in combined GG by ≥ 1, may occur in as many as 35% of patients who ultimately undergo RP [10]. Modifications to the International Society of Urological Pathology standards for PCa grading have also instituted changes to the reporting of proportions of high-grade disease and tertiary patterns and aggregate reporting of MRI-targeted biopsies to better convey disease risk [11].

The accuracy of GGG assessments from biopsy has assumed greater importance in recent years due to the increasing utilization of conservative treatment options, as well as differences in the intensity and modality of treatment. In particular, disease misclassification at biopsy may cause patients to be inappropriately identified as candidates for conservative management. Such patients who are undergraded are more likely to have adverse pathological features, including extra-prostatic extension and biochemical recurrence [12]. GSU, even after adjusting for known preoperative variables, is a strong predictor of biochemical recurrence after local treatment [13]. Additionally, undergrading of clinically significant PCa could lead to missed escalation of treatment such as combination androgen deprivation therapy-radiation therapy, or omission of pelvic lymph node dissection at the time of RP, further impacting survival. Conversely, overgrading of GSs can lead to overtreatment of less aggressive PCa cases.

Because of its important link to prognosis, researchers have attempted to characterize risk of GSU. Proper identification of patients at risk of GSU could prompt second-level tests such as additional biopsy, prostate MRI, or genomic testing, and may allow for the development of more personalized PCa treatment decisions. The objective of this review is to discuss the risk factors for GSU upon prostatectomy.

Evidence acquisition

A comprehensive literature search for eligible records was conducted from inception until February 2024. All studies were identified and selected from the PubMed database. We employed a broad search strategy with major search terms that included: “Gleason score”, “upgrade^*,” “downgrade^*”, “risk”, “PSA”, “MRI”, “multivariate analysis”, “odds ratio”, etc. which were searched in all fields. The literature search was restricted to the English language and did not include any year restrictions.

Studies were eligible for inclusion if they were (1) original research with experimental design, (2) in the field of urologic oncology, (3) focused on identifying risk factors for GSU, (4) included uni- or multivariate statiscially analysis. Major exclusion criteria included reviews, studies without corresponding full-text such as conference abstracts, ongoing studies, and studies with sample sizes fewer than 50 patients. The final included studies were evaluated in qualitative synthesis. In total, 48 studies were included in this literature review.

Patient demographics and clinical factors

PCa is a spectrum of heterogenous diseases, some of which are so indolent that they may never impact survival in many patients while others may significantly limit life expectancy. Because of this diversity, there has been a push to detail which demographic factors and other patient clinical factors are valuable in aiding the diagnosis of clinically significant PCa. Among these, increasing age, larger body mass index (BMI), race, and lower prostate volume have often been linked to various clinical end points, including incidence, stage at presentation, biochemical recurrence, and mortality [14, 15]. These risk factors have also been examined as potential risk factors for Gleason upgrading. Table 1 examines identified studies that found significant associations between demographic and clinical factors and GSU.

Age has been well described as a risk factor for GSU. In a retrospective analysis of 3,571 men with GG1 disease on transrectal ultrasound (TRUS)-guided biopsy who underwent prostatectomy, Leeman et al. [16] found older age to be a significant predictor of upgrade on surgical pathology (OR 1.05, 95% CI 1.01–1.08, P = 0.005). This complements the findings of a similar analysis by Gershman et al. [17] which, in a retrospective analysis included 1,836 patients with GG1 disease who had prostatectomy. The study reported a similar risk of upgrade on prostatectomy per increased year of age (OR 1.05, 95% CI 1.03–1.07, P < 0.001) [17]. However, on subgroup analysis of only those patients age > 60, this association is considerably strengthened (OR 2.31, 95% CI 1.50–3.54, P < 0.001) [17]. In an analysis of 7,643 paired prostate biopsy and RP specimen recorded in the Institutional Urology Prostate Cancer Database, Epstein et al. [18] found that men who were upgraded at prostatectomy were on average 2 years older. However, despite the large evidence of an age-upgrade association, the magnitude of this effect size was small, limiting the finding’s clinical utility. The introduction of MRI into the PCa diagnostic pathway and further refinements in biopsy technique may reduce lead-time bias and rates of misdiagnosis at initial biopsy with more targeted biopsies. Mazzone et al. [19] examined a cohort of 424 patients receiving systematic and targeted biopsy and found that age was not a significant risk factor for upgrading when MRI-guided biopsy techniques were utilized.

BMI has also been associated with GSU at prostatectomy. Among a sample of 959 urban-residing males, Vora et al. [20] reported a positive association between BMI and GSU (OR 1.04, P = 0.02). Freedland et al. [21], examining the Shared Equal Access Regional Cancer Hospital database, found that obese men (BMI ≥ 30 kg/m²) had 1.89 times greater odds of GSU when compared to non-obese (BMI < 25 kg/m²) (P = 0.003). More recently, this association was also seen independent of biopsy technique [22]. Several explanations for this observed association have been offered. First, high BMI may impair proper patient positioning and needle trajectory, making more likely that biopsy insufficiently samples lesions. As such, some studies have suggested that these patients receive more extended biopsy schemes to overcome sampling issues [23]. Additionally, the association between obesity and GSU mirrors the association between obesity and PCa prognosis, likely through similar mechanisms. Obesity is closely related to PCa aggressiveness and PCa-related mortality; several mechanisms have been suggested to underly this including tumor proliferation via the insulin-like growth factor-1 (IGF-1) pathway and oxidative stress-related inflammatory signaling [24, 25].

Race has also been examined as a risk factor for GSU at the time of RP. However, the role of race in GSU is conflicting and with limited reports. Vora et al. [20] used both univariate and multivariate analysis to identify significant predictors of GSU while controlling for clinical parameters. On multivariate analysis, African American men had 1.74 times increased odds of upgrade at the time of prostatectomy. Additionally, a study conducted by Sundi et al. [26] also concluded that African American men diagnosed with very low-risk PCa were more likely to experience upgrade and had worse pathologic outcomes than Caucasian patients (27.3% versus 14.4%, P < 0.001). More recently, however, two studies examining the National Cancer Database and Surveillance, Epidemiology, and End Results (SEER) database, each containing samples of more than 10,000 patients, did not observe a significant association between Black versus White race and GSU [27, 28]. These findings were consistent among individuals with GS 3+3 and GS 3+4 cancer at biopsy. As such, it remains unclear what role GSU plays in this relationship. Recent advances in understanding of racial disparities in PCa epidemiology have suggested that much of these differences stem from differential access to care at each level, from screening to diagnosis and treatment [29]. It is possible that in some studies, differences in timeliness and quality of care may contribute to upgrade. Regardless, further research is needed to better characterize this association and its potential causes.

Several studies have examined the influence of prostate size on risk of GSU. An early study by Uzzo et al. [30] suggested that larger glands were associated with disease under-sampling, resulting in an increased risk of GSU. The Prostate Cancer Prevention Trial confirmed these findings among 369 patients with PSA < 10 ng/mL who underwent RP at their institution; TRUS volume was associated with GG4 pattern or greater on biopsy but not at prostatectomy [31]. However, in more recent years, several studies have suggested an inverse relationship—patients with smaller glands are more likely to be upgraded [17, 32–35]. Davies et al. [33] examined 1,251 patients with D’Amico low risk disease and found that men with prostate volumes at the 25th percentile (36 cm³) were 50% more likely to experience upgrade than men with prostate volumes at the 75th percentile (58 cm³). A later study conducted by Gershman et al. [17] found increased risk for combined GS7 or greater disease with decreasing prostate weight even after controlling for age and PSA among a sample of 1,836 patients with GS6 on initial prostate biopsy. Additionally, Pierorazio et al. [34] found that while smaller volume was associated with upgrading, larger volume was associated with downgrading at prostatectomy.

Several theories have been offered to explain varied findings between small prostates and GSU. In the last decade, advancement in biopsy technology and technique have made possible more targeted and extended biopsy schemes that can improve sampling accuracy in larger glands. Additionally, there is increasing evidence to suggest that PCa found in smaller glands may harbor distinct, more aggressive biology leading to adverse outcomes. One study has suggested that PCa found in the smaller glands are more likely to be androgen-independent, a marker of advanced disease manifestation [35, 36]. Additionally, larger prostates could also serve as physical barriers for the growth of PCa, reducing the risk that tumor extends beyond the gland [37]. However, some studies have suggested that the association between small prostate gland size and adverse clinical outcomes is due to lead-time bias, as improved PSA-based detection is better at identifying PCa in larger prostates, which generally secrete more PSA [38]. However, Freedland et al. [35] point out that this association persists even after exclusion of cT1c biopsy PCa, cases most likely to be subject to lead-time bias. Ultimately, prostate size may affect Gleason upgrading, but more studies are needed to better establish the mechanism underlying this relationship.

Impact and future directions

In the modern PSA era, the number of patients diagnosed with low-risk PCa (clinical T1c or T2a or GS ≤ 6), has risen substantially [75]. To limit overtreatment in patients with clinically indolent PCa, less invasive management strategies including active surveillance have been increasingly used to delay definitive management [76]. However, as was most notably discovered by D’Amico et al. [77], roughly 40% of men with GS ≤ 6 at biopsy may have occult, higher-grade disease at the time of prostatectomy. The management and prognosis of PCa is heavily driven by GS. Despite improvement in biopsy approaches, difference between biopsy and postoperative pathology remains an enduring source of uncertainty in the management of localized PCa [78]. Gleason upgrading can lead to unnecessary delays in treatment while downgrading can lead to overtreatment, impacting a patient’s quality of life and healthcare expenditure [69]. As such, understanding factors that underly the discrepancy between biopsy and prostatectomy GS is an area for continued refinement.

Appreciating the possibility of Gleason undegrading contributes to the personalized discussion of PCa decision making. For example, patients with initial “low risk” PCa but with significant risk factors for upgrading including high PSA values and high suspicion MRI scans should be counseled about their risks for disease undersampling. As such, close evaluation including confirmatory testing and repeat prostate biopsy in the setting of risk factors may be warranted.

Improvements in image-guided biopsy modalities can also reduce sampling error that miss higher Gleason grade PCa. Prostate biopsy has traditionally been guided with conventional ultrasound. The use of mpMRI and MRI-TRUS fusion biopsy has led to improvements in the diagnostic assessment for clinically significant PCa, but possibilities of undersampling remain as noted by the presence of GSU on final pathology, especially for PI-RADS 4–5 lesions [79, 80]. Conversely, the addition of mpMRI to the PCa diagnostic pathway has caused risk inflation, often doubling the proportion of National Comprehensive Cancer Network (NCCN) high-risk patients identified at biopsy [81]. In fact, the use of MRI-targeted biopsy has been associated with increased proportions of downgrading on final pathology, possibly due to sampling bias in heterogenous lesions [82].

Because of these limitations, advancements in imaging have been offered and are being evaluated to improve the accuracy of clinical disease sampling. For example, micro-ultrasound (microUS) technology represents a novel technology that may further improve diagnostic accuracy of prostate biopsy. Compared to TRUS with 6–9 Mhz, microUS operates at 29 Mhz and provides a threefold improvement in spatial resolution [83]. Indeed, several studies have shown that the use of microUS alone is non-inferior to MRI-TRUS fusion biopsy for clinically significant Pca (csPCa) detection [84]. Additionally, microUS may also display improved sensitivity while retaining similar specificity when compared to MRI-TRUS [85–88]. Large randomized controlled trials such as the optimization of prostate biopsy–micro-ultrasound versus MRI (OPTIMUM) trial are ongoing to assess the effectiveness of MRI/microUS fusion biopsy [89]. To our knowledge, no study has yet examined rates of GSU in patients biopsied via microUS when compared to other imaging guided techniques.

In addition to advances in imaging, genomic classifiers, such as the Decipher classifier (GenomeDx Biosciences, Vancouver, BC, Canada), have recently been developed as a tissue-based platform to evaluate the expression of 22 genes reflecting pathways involved in cellular proliferation, differentiation, immune modulation, and androgen-receptor signaling [90]. The Decipher classifier has been shown to provide robust predictions of disease outcome among patients with high-risk PCa [90]. However, little is known of Decipher’s ability to estimate the trajectory of untreated low or favorable-risk PCa. One study conducted by Press et al. [91] has shown that elevated Decipher score is associated with short-term biopsy upgrading among patients on active surveillance. Another study by Sheng et al. [92] report that Decipher can independently predict upgrading at prostatectomy. As such, the Decipher genomic classifier may represent a novel approach to predicting GSU and prognostic value in PCa, but larger retrospective studies are necessary to confirm these findings.

Conclusions

Risk factors for GSU from initial biopsy to prostatectomy have been identified including PSA, PSAD, imaging findings and clinical factors such as age and BMI. Pathologic Gleason upgrading highlights the persistence of sampling error during biopsy, even in the era of image guided prostate biopsy. Identifying patients at higher risk for GSU is an important component of counseling, particularly for patients with favorable risk PCa undergoing initial management with active surveillance, and can inform the need for confirmatory testing.