Targeted biopsy plus systematic biopsy versus targeted saturation biopsy plus systematic biopsy for the detection of clinically significant prostate cancer: A comparative study

ABSTRACT

Introduction:

Multiparametric magnetic resonance imaging (mpMRI)-guided targeted biopsy (TB) has improved the detection of clinically significant prostate cancer (csPCa). This study compared the diagnostic performance of TB versus targeted saturation biopsy (TS), both in addition to standard 12-core systematic biopsy (SB), for the detection of csPCa.

Methods:

This comparative study included men with suspected prostate cancer and mpMRI lesions (Prostate Imaging Reporting and Data System ≥3). Patients underwent either TB (4 cores per lesion) with SB or TS (4 targeted cores plus 4 perilesional cores) with SB. csPCa was defined as Gleason Grade Group ≥2. Detection rates were compared between groups. Continuous variables were compared using the Mann–Whitney U-test and categorical variables using the Chi-square or Fisher’s exact test. Logistic regression analysis was used to identify independent predictors of csPCa. Odds ratios, relative risk, and 95% confidence intervals (CIs) were calculated.

Results:

A total of 108 patients were analyzed (TB + SB: n = 65; TS + SB: n = 43). csPCa was detected in 56.9% of the TB + SB group and 72.1% of the TS + SB group (P = 0.110). The odds ratio (OR) for csPCa detection comparing TB + SB with TS + SB was 0.51 (95% CI: 0.22–1.17), indicating a clinical advantage with TS biopsy. TS missed no case identified by SB. Complication rates were comparable between groups (P = 0.472).

Conclusion:

Although the TS biopsy did not demonstrate statistically significant superiority over TB, the higher detection rate and favorable OR suggest a potential clinical benefit. Larger studies with appropriate power and balanced baseline characteristics are needed to confirm these observations.

INTRODUCTION

Accurate differentiation between clinically significant and insignificant prostate cancer remains central to contemporary prostate cancer management.

The standard 10–12 core TRUS biopsy, while long considered the diagnostic mainstay, is fundamentally compromised by its random sampling nature. This results in a well-documented fault that’s the under-detection of clinically significant tumors, particularly those situated in the anterior and transition zones, and the concurrent over-detection of clinically insignificant cancer (Gleason Grade Group 1), which leads to risk misclassification and potential overtreatment.[1]

Multiparametric magnetic resonance imaging (mpMRI) has transformed prostate cancer diagnostics by enabling lesion visualization and risk stratification, with standardization of reporting through the Prostate Imaging Reporting and Data System (PI-RADS). mpMRI serves as a powerful triage tool offering a high negative predictive value that can safely defer biopsy in selected patients.[2]

The introduction of MRI-guided biopsy methods, such as cognitive targeting, MRI-TRUS fusion, and in-bore MRI biopsy, has enabled direct sampling of lesions that appear suspicious on MRI. Level 1 evidence from pivotal trials such as PRECISION and MRI-FIRST has demonstrated that these targeted strategies yield superior detection rates for csPCa while simultaneously reducing the diagnosis of clinically insignificant disease compared with systematic biopsy (SB).[3,4] An analysis by Brisbane et al. provided a topographic rationale for template modification, demonstrating that 90% of csPCa foci reside within a 10-mm radius of the MRI-visible lesion epicenter.[5] They conceptualized this as a “penumbra” effect where a significant minority (26%) of csPCa cores were located outside the primary region of interest but within this immediate periphery. This evidence suggests that our current targeted strategies, often limited to 2–3 cores per lesion, may be insufficient to capture the full extent of the tumor. This has led to interest in targeted saturation biopsy (TS), which increases sampling within and around the index lesion.

However, increasing biopsy cores may increase morbidity without proportionate diagnostic benefit. This comparative study aimed to compare targeted biopsy (TB) and TS biopsy for the detection of clinically significant prostate cancer (csPCa).

MATERIALS AND METHODS

This was a comparative study with allocation concealment, conducted at a tertiary care center between April 2024 and December 2025 after institutional ethics committee approval (IEC letter No. TP (DM/M.Ch.) 25/2024/IEC/ABVIMS/RMLH/399, dated May 20^th, 2024).

The study was initially designed as a randomized controlled trial, with allocation concealment using the SNOSE technique. However, as evidenced by the significant baseline imbalances observed between groups (PSA, PSA density, and PI-RADS distribution), SNOSE allocation did not achieve the statistical equivalence expected of true randomization. The study is therefore more accurately characterized as a comparative study with allocation concealment and is described as such throughout this manuscript.

Participants

Men aged ≥40 years with elevated PSA (≥4 ng/mL), abnormal digital rectal examination, and mpMRI lesions scored PI-RADS ≥3 were included. Exclusion criteria were prior prostate biopsy, active infection, coagulopathy, contraindications to MRI, and previous prostate surgery.

Allocation and interventions

Allocation was performed using the SNOSE technique targeting a 1:1 ratio. The actual enrolled ratio was approximately 1.5:1 (TB + SB: n = 65 versus TS + SB: n = 43), representing a deviation from the planned balance, which is acknowledged as a study limitation. Participants were allocated to one of two groups.

• TB + SB group: TB (4 cognitive MRI–TRUS fusion cores per lesion) followed by standard 12-core SB

• TS + SB group: TS biopsy (4 targeted cores plus 4 perilesional cores) followed by standard 12-core SB.

The operating urologist was not blinded to group allocation as the biopsy templates differed between groups. However, the pathologist conducting the histopathological assessment was blinded to the biopsy technique used. All biopsies were performed under local anesthesia. csPCa was defined as a Gleason Grade Group ≥2. The primary objective was to compare the csPCa detection rates between the TB + SB and TS + SB groups. Secondary objectives were to compare postbiopsy complication rates between the two groups and to identify independent predictors of csPCa using logistic regression analysis.

Hypothesis

Within a combined biopsy strategy that includes standard SB in both arms, we hypothesized that there is no significant difference in csPCa detection between TB + SB and TS + SB. The study does not address whether perilesional cores can substitute for SB, as that question requires a separate comparison of TB + SB versus TB + perilesional biopsy without systematic cores.

Sample size was calculated using the formula for comparing two independent proportions (α = 0.05, two-tailed; power = 80%). Based on csPCa detection rates reported by Saner (42.4% with conventional TB and 67.1% with target saturation biopsy), a clinically meaningful difference of approximately 24.7% points was assumed.[6] This yielded a required sample size of 60 patients per group (total = 120). Accounting for a 10% dropout rate, the total required sample was approximately 134 patients (67 per group). A total of 108 patients were enrolled (TB + SB: n = 65; TS + SB: n = 43), falling short of the required target, particularly in the TS + SB arm. The study is underpowered, this is a limitation. The observed 15.17% difference in csPCa detection may not have reached statistical significance, partly due to insufficient sample size.

Statistical analysis

Statistical analysis was performed using SPSS version 23 (IBM Corp., Armonk, NY, USA). Continuous variables were compared using the Mann–Whitney U-test. Categorical variables were compared using the Chi-square or Fisher’s exact test as appropriate. Univariable and multivariable binary logistic regression were performed to identify independent predictors of csPCa; predictor variables included age, PSA, prostate volume, PSA density, and PI-RADS score (PI-RADS 3 as reference category). Results are reported as odds ratios (OR) with 95% confidence intervals. Relative risk and absolute risk difference with 95% confidence intervals were also calculated. McNemar’s test was used for within-group paired comparisons with SB. All enrolled patients who underwent biopsy were included in the analysis (complete-case analysis). P < 0.05 was considered statistically significant. We confirm the availability of and access to all original data reported in this study. The authors confirm that they have reviewed the manuscript and take full responsibility for the contents in the article. The authors have not used AI or AI assisted technologies, including large language learning models, in preparation of the manuscript.

RESULTS

Baseline characteristics

A total of 108 patients were enrolled and analyzed (TB + SB: n = 65; TS + SB: n = 43). The overall study cohort comprised predominantly elderly men (mean age 68.56 ± 8.10 years; 61.1% aged >65 years). Baseline characteristics and PI-RADS distribution are summarized in Table 1 and Figure 1.

Table 1.

Baseline characteristics of study population – targeted biopsy + systematic biopsy versus targeted saturation + systematic biopsy

Variable	TB + SB (n=65)	TS + SB (n=43)	P
Mean age±SD (years)	70.1±7.8	67.4±8.5	0.436
Age >65 years, n (%)	42 (64.6)	24 (55.8)	0.387
PSA (ng/mL), median (IQR)	10.20 (8.66–12.60)	12.00 (9.10–18.55)	0.042*
Prostate volume (mL), median (IQR)	61.00 (46.00–70.00)	55.00 (44.50–64.50)	0.164
PSA density, median (IQR)	0.17 (0.14–0.25)	0.22 (0.16–0.34)	0.033*
PI-RADS 3, n (%)	21 (32.3)	11 (25.6)	0.024*
PI-RADS 4, n (%)	28 (43.1)	17 (39.5)	0.024*
PI-RADS 5, n (%)	16 (24.6)	15 (34.9)	0.024*

*P<0.05 is considered statistically significant. Mann–Whitney U-test for continuous variables; Chi-square test for PI-RADS distribution. IQR=Interquartile range, TB=Targeted biopsy, TS=Targeted saturation, SB=Systematic biopsy, PI-RADS=Prostate Imaging Reporting and Data System, SD=Standard deviation, PSA=Prostate-specific antigen

Figure 1.

This figure compares key clinical variables between the two groups using Mann-Whitney U-test. TB = Targeted biopsy, SB = Systematic biopsy, TS = Targeted saturation, PSA = Prostate-specific antigen

Statistically significant differences were observed between groups in PSA (median 10.2 vs. 12.0 ng/mL, P = 0.042), PSA density (0.17 vs. 0.22, P = 0.033), and PI-RADS distribution (P = 0.024), with the TS + SB group harboring more PI-RADS 5 lesions (34.9% vs. 24.6%). Age and prostate volume were comparable between groups. These baseline imbalances are important confounders of the between-group comparison and are discussed later.

Diagnostic performance

csPCa (Gleason Grade Group ≥2) was detected in 37 patients (56.9%) in the TB + SB group and 31 patients (72.1%) in the TS + SB group. Both targeted approaches detected more csPCa than SB alone; TB identified 17 additional csPCa cases missed by SB, whereas TS biopsy identified 11 such additional cases. One csPCa detected on SB was missed by TB, while none was missed by the TS biopsy.

The absolute risk difference (TS minus TB) was −15.17% (95% CI: −33.19 to +2.85). The relative risk of csPCa detection comparing TB with TS was 0.79 (95% CI: 0.60 to 1.05), and the OR was 0.51 (95% CI: 0.22 to 1.17, P = 0.154 by Fisher’s exact test). Although the confidence interval crosses 1.0 and statistical significance was not achieved on Chi-square (P = 0.110) or Fisher’s exact test (P = 0.154), an OR of 0.51 represents a clinically meaningful effect size in uro-oncology. Comparative outcomes are presented in Table 2.

Table 2.

Comparative diagnostic performance of targeted biopsy + systematic biopsy versus targeted saturation + systematic biopsy

Outcome	TB + SB (n=65)	TS + SB (n=43)	Statistical result
csPCa detected by SB alone, n (%)	21 (32.3)	20 (46.5)	-
csPCa detected by targeted technique only (SB-negative), n (%)	17 (26.2)	11 (25.6)	-
Total csPCa detected, n (%)	37 (56.9)	31 (72.1)	P=0.110
csPCa missed by targeted technique, n (%)	1 (1.5)	0	-
McNemar test (vs. SB reference), P	0.0004	0.0025	Both significant
Cohen’s kappa with SB	0.47 (moderate)	0.50 (moderate)	-
OR (TB vs. TS)	-	-	0.51 (95% CI 0.22 to 1.17)
Relative risk (TB vs. TS)	-	-	0.79 (95% CI 0.60 to 1.05)
Absolute risk difference	-	-	−15.17% (95% CI −33.19 to +2.85)

Chi-square and Fisher’s exact tests for between-group comparison. McNemar’s test for within-group paired comparison with systematic biopsy. P<0.05 considered statistically significant. csPCa=Clinically significant prostate cancer, TB=Targeted biopsy, TS=Targeted saturation, SB=Systematic biopsy, OR=Odds ratio, CI=Confidence interval

Complications

Postbiopsy complications, including hematuria, rectal bleeding, fever, and urinary retention, were comparable between the two groups (P = 0.472), confirming that the additional perilesional cores in the TS + SB strategy did not increase procedural morbidity. Hematuria was the most common adverse event noted in both groups. Complication data are presented in Table 3 and Figure 2.

Table 3.

Distribution of postbiopsy complications between targeted biopsy + systematic biopsy and targeted saturation + systematic biopsy groups

Complication	TB + SB, n (%)	TS + SB, n (%)	Total, n (%)	P
Fever	3 (9.4)	1 (4.5)	4 (7.4)	0.472
Hematuria	20 (62.5)	15 (68.2)	35 (64.8)	-
Rectal bleeding	7 (21.9)	3 (13.6)	10 (18.5)	-
Urinary retention	2 (6.2)	3 (13.6)	5 (9.3)
Total	32 (100.0)	22 (100.0)	54 (100.0)	-

Fisher’s exact test for overall comparison. P<0.05 is considered statistically significant. TB=Targeted biopsy, TS=Targeted saturation, SB=Systematic biopsy

Figure 2.

Comparison of postbiopsy complications. TB = Targeted biopsy, SB = Systematic biopsy, TS = Targeted saturation

DISCUSSION

In this study, standard SB detected prostate cancer in 58.33% of patients overall, but only 36.11% had clinically significant disease, and the remainder carried low-grade GG1 pathology. This is a well-recognized limitation of the blind, template-based approach. The PRECISION trial reported a csPCa detection rate of 26% with SB,[3] and Krausewitz et al. reported an overall detection rate of 60%,[7] both consistent with our findings and reflecting the fundamental inadequacy of sampling without imaging guidance.

TB addressed this limitation. Using four cognitive MRI-TRUS fusion cores, TB + SB detected csPCa in 56.9% of patients and reduced GG1 detection to 6.1%, a gain over SB alone. McNemar’s test (P = 0.0004) confirms that TB detected 17 additional csPCa cases that SB would have missed, and this is not a chance finding. Cohen’s kappa of 0.47 between TB and SB reflects moderate agreement, which tells us that the two techniques are picking up overlapping but not identical disease, an argument for keeping both in the biopsy protocol rather than abandoning one. This aligns with Jiang et al., who reported a 48% csPCa detection rate with TB,[8] and with data from the PAIREDCAP trial, where cognitive fusion TB achieved a detection rate of around 47% and software-assisted fusion TB around 60%.[7] The gap between those two figures is relevant to understanding our own results since we used cognitive rather than software fusion throughout.

TS biopsy achieved the highest detection rate in our cohort at 72.1%, and it missed none of the csPCa cases that SB had identified, a sensitivity that TB could not match (one SB-positive case was missed by TB). McNemar’s test (P = 0.0025) again confirms that the incremental yield of TS over SB is statistically real. Saner et al. demonstrated that wider perilesional sampling identifies cancers lying just outside the MRI-defined target, improving sensitivity at the cost of some specificity,[6] and our kappa of 0.50 between TS and SB is consistent with this, suggesting that even with saturation sampling, there remains a subset of disease that only SB captures.

The comparison between TB + SB and TS + SB is where the interpretation becomes more nuanced. The 15.17%-point difference in csPCa detection (56.9% vs. 72.1%) did not reach statistical significance in either the Chi-square test (P = 0.110) or the Fisher’s exact test (P = 0.154). However, dismissing this finding as simply negative would be a mistake. The OR of 0.51 (95% CI: 0.22 to 1.17) means that patients in the TB + SB group had roughly half the odds of a csPCa-positive biopsy compared to those in the TS + SB group. In uro-oncology, where missing a significant tumor carries the consequences of delayed treatment, understaging or losing the curative window, an OR of 0.51 is not a number to set aside simply because the confidence interval crosses 1.0. The more honest explanation for the nonsignificant P value is that the TS + SB arm enrolled only 43 patients, and an underpowered study that cannot reject the null hypothesis is not the same as a study that has proven equivalence.

That said, the baseline imbalances between groups complicate any straightforward interpretation of the detection difference. The TS + SB group entered the study with a higher median PSA (12.0 vs. 10.2 ng/mL, P = 0.042), higher PSA density (0.22 vs. 0.17, P = 0.033), and a greater proportion of PI-RADS 5 lesions (34.9% vs. 24.6%, P = 0.024). PSA density and PI-RADS category are the strongest predictors of csPCa, shown consistently by Washino et al.,[9] Filson et al.,[10] and Corcoran et al.[11] and also confirmed in our own regression analysis. A group with higher PSA density and more PI-RADS 5 lesions will detect more csPCa, regardless of biopsy technique. Hence, the 15% absolute difference in detection rates between arms likely reflects this preexisting difference in disease burden rather than a pure technique effect. Future studies using this comparison must use randomization by PSA density and PI-RADS category from the outset.

The detection rate of 72.1% with TS is lower than that reported in software-fusion-based saturation studies. Radtke et al. achieved 95%,[12] and Hansen et al. 91%.[13] This difference is expected, and it does not diminish the validity of our findings. All those studies used software-assisted MRI-TRUS fusion platforms, which offer more precise co-registration of MRI targets with real-time ultrasound. We used cognitive fusion, which is more operator-dependent but also far more widely available, particularly in resource-limited tertiary care settings like ours. The detection rates we report are therefore more representative of what a standard urology team without access to fusion software can realistically achieve with saturation sampling.

Several recent studies lend important contemporary context to these findings. He et al. showed that targeted plus regional biopsy matches TB + SB detection with fewer total cores.[14] The PERI-PRO trial (n = 380) demonstrated noninferiority of targeted plus perilesional biopsy versus combined TB + SB.[15] Bastos et al. (n = 617) found that contralateral systematic cores add minimal yield once perilesional sampling is already performed,[16] and Tarim et al. (n = 803) demonstrated that TB + RB is essential for optimizing csPCa detection in PI-RADS 4 lesions smaller than 9.25 mm, while TB alone may be adequate for larger PI-RADS 4 lesions, PI-RADS 5 lesions, and nonperipheral zone lesions, supporting an individualized rather than uniform biopsy strategy.[17] The EAU guidelines now recognize targeted plus perilesional biopsy as an accepted strategy. Taken together, these data confirm that the perilesional tissue zone is the relevant sampling territory.

Regarding safety, the complication profiles of both strategies were comparable (P = 0.472). Hematuria was the most common adverse event occurring in 62.5% of TB + SB and 68.2% of TS + SB patients, consistent with Drost et al.[2] and the PRECISION trial,[3] which both identify hematuria as a frequent complication of transrectal biopsy. Rectal bleeding was seen in 21.9% of TB + SB patients and 13.6% of TS + SB patients within the 14%–22% range reported by PRECISION.[3] Fever rates were low in both groups (9.4% vs. 4.5%) with no significant difference. The practical implication is that taking four additional perilesional cores per lesion did not increase patient morbidity.

Limitations

The principal limitations of this study are: (1) underpowered: the enrolled sample (n = 108 total; TS + SB n = 43) fell short of the calculated requirement of 60 per group (134 total with 10% dropout), particularly in the TS + SB arm; (2) baseline imbalances in PSA, PSA density and PI-RADS distribution could have influenced detection differences to technique; (3) the allocation ratio deviated from the planned 1:1 to approximately 1.5:1; (4) use of cognitive rather than software-assisted fusion may limit detection precision. These limitations should be considered while interpreting the results.

CONCLUSION

The null hypothesis that there is no significant difference in csPCa detection between TB + SB and TS + SB was not rejected (72.1% vs. 56.9%, P = 0.110).w However, the clinically meaningful OR of 0.51 and the absence of any missed SB-positive case with TS + SB indicate that this nonrejection reflects insufficient statistical power rather than true equivalence between the strategies. Thus, increasing the number of biopsy cores may provide meaningful clinical diagnostic benefit. PSA density and PI-RADS category were the strongest independent predictors of clinically significant disease in this cohort. Both biopsy strategies were well tolerated with comparable complication rates. Adequately powered prospective studies with balanced baseline characteristics are required to draw definitive conclusions.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.