Ipsilateral hemigland prostate biopsy may underestimate cancer burden in patients with unilateral mpMRI-visible lesions

Abstract

Purpose

To evaluate the cancer detection rates of reduced-core biopsy schemes in patients with unilateral mpMRI-visible intraprostatic lesions and to analyze the contribution of systematic biopsy cores in clinically significant prostate cancer (csPCa) detection.

Methods

212 patients with mpMRI-visible unilateral intraprostatic lesions undergoing MRI/TRUS fusion-guided targeted biopsy (TBx) and systematic biopsy (SBx) were included. Cancer detection rates of TBx + SBx, as determined by highest Gleason Grade Group (GG), were compared to 3 reduced-core biopsy schemes: TBx alone, TBx + ipsilateral systematic biopsy (IBx; MRI-positive hemigland), and TBx + contralateral systematic biopsy (CBx; MRI-negative hemigland). Patient-level and biopsy core-level data were analyzed using descriptive statistics with confidence intervals. Univariable and multivariable logistic regression analysis was conducted to identify predictors of csPCa (≥GG2) detected in MRI-negative hemiglands at p < 0.05.

Results

Overall, 43.4% (92/212) of patients had csPCa and 66.0% (140/212) of patients had any PCa detected by TBx + SBx. Of patients with csPCa, 81.5% had exclusively ipsilateral involvement (MRI-positive), 7.6% had only contralateral involvement (MRI-negative), and 10.9% had bilateral involvement. The csPCa detection rates of reduced-core biopsy schemes were 35.4% (75/212), 40.1% (85/212), and 39.6% (84/212) for TBx alone, TBx + IBx, and TBx + CBx, respectively, with detection sensitivities of 81.5%, 92.4%, and 91.3% compared to TBx + SBx.

Conclusion

Reduced-core prostate biopsy strategies confined to the ipsilateral hemigland underestimate csPCa burden by at least 8% in patients with unilateral mpMRI-visible intraprostatic lesions. The combined TBx + SBx strategy maximizes csPCa detection.

Graphical Abstract

Introduction

Prostate cancer (PCa) is the second leading cause of cancer-related mortality in biologically male individuals [1]. PCa diagnosis involves histological examination of prostate biopsies and classification using the Gleason scoring system, known as Gleason grade group (GG) [2]. Transrectal ultrasound (TRUS)-guided systematic prostate biopsy (SBx) is the most widely used prostate biopsy method which generally involves an average of 12-core biopsies from the apex, mid-gland, and base of each prostate hemisphere with medial and lateral sampling at each level [3]. Moreover, TRUS-guided SBx is considered a standardized method and does not vary among gland size or PCa suspicion [4]. While TRUS biopsy is generally safe with rare complications including hematuria, urinary tract infections, and sepsis [5–8], TRUS-guided SBx can undersample the prostate gland and additional biopsy cores increase risk of morbidity [8]. Additionally, the transperineal ultrasound (TPUS)-guided SBx method offers fewer complications, comparable PCa detection rates to TRUS [8, 9] and easier access to the anterior prostate gland [10]; however, TPUS may involve general anesthesia and increased postprocedural pain compared with TRUS [5]. Regardless of the approach, incorporating magnetic resonance imaging (MRI) into prostate biopsy methods via MRI/TRUS fusion-guided targeted biopsy (TBx) has significantly improved the detection of clinically significant prostate cancer (csPCa) [11–13]. Adoption of the most recent Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1) criteria for prostate multiparametric MRI (mpMRI) has added value by predicting the likelihood of cancer in identified intraprostatic lesions [12, 14, 15]. Combining MRI/TRUS fusion-guided TBx and SBx further maximizes the likelihood of detecting csPCa at the cost of increased overdiagnosis of clinically insignificant prostate cancer (low-grade PCa) and slightly higher risk of complications [11, 12, 16, 17].

Patients who are diagnosed with low-grade PCa are often placed on active surveillance which involves longitudinal prostate-specific antigen (PSA) measurements and mpMRI, which can pose considerable cost and inconvenience for patients [18]. Additionally, patients may experience anxiety and distress due to the psychological burden of a cancer diagnosis and possibility of incorrect staging [19]. Overdiagnosis of low-grade PCa via TRUS SBx prompted consideration of alternative biopsy schemes that reduce the number of biopsies yet maintain high sensitivity for csPCa detection. Prior efforts have focused on lowering the number of biopsy cores to sample regions of the highest likelihood for csPCa [15, 20–23]. A prior study suggested that SBx may not contribute to relevant csPCa detection in a subset of patients who have PI-RADS 5 lesions [15]. Another study investigating the cancer detection yield of biopsy cores sampling both MRI-visible lesions (umbra) and perilesional areas (penumbra) found that most (> 90%) csPCa resided within 1 cm of MRI-visible lesions [23]. Other studies suggested that TBx accompanied by only ipsilateral SBx achieves similar csPCa detection and sensitivity when compared to TBx + SBx sampling of unilateral MRI-visible lesions [24–27]. Ultimately, minimizing biopsy cores without jeopardizing diagnostic accuracy would potentially improve patient care while decreasing costs and biopsy-related risks. However, prior studies did not use the well-established PI-RADS v2.1 scoring system for mpMRI and some excluded patients with multiple lesions. Thus, we sought to determine the cancer detection rate of 3 commonly mentioned reduced-core biopsy schemes, targeted biopsy (TBx) alone, TBx + ipsilateral systematic biopsy (IBx), and TBx + contralateral systematic biopsy (CBx) compared to the standard TBx + SBx strategy in patients with unilateral mpMRI-visible lesions.

Materials and methods

Patient population

This HIPAA-compliant study was conducted with IRB approval and is a retrospective analysis of a prospective dataset from a single center. The cohort included 654 consecutive patients with or without a prior diagnosis of PCa who provided written informed consent. The study consisted of patients who underwent PI-RADS v2.1-compliant mpMRI, and subsequent biopsy depending on mpMRI findings between April 2019 and March 2022. Exclusion criteria included prior history of PCa treatment, no mpMRI-visible lesions, the presence of bilateral and/or midline lesions on mpMRI, and patients who did not receive combined MRI/TRUS targeted and systematic biopsies. All patients satisfying the study requirements were considered eligible and were included in the final study cohort (Fig. 1).

Fig. 1.

Flowchart of patient selection criteria

mpMRI acquisition and prostate biopsy

Patients were scanned on a 3-T MRI (Ingenia Elition X, Class IIa; Philips Healthcare, Best, Netherlands) equipped with a surface coil (SENSE, Philips Healthcare, Best, Netherlands) for mpMRI evaluation; an endorectal coil (BPX-30, MEDRAD, Pittsburgh, Pennsylvania) was also used for 48 patients. Pulse sequences included T2-weighted imaging (T2W MRI), diffusion weighted imaging (DW MRI), apparent diffusion coefficient (ADC) map, and dynamic contrast enhanced imaging (DCE).

A genitourinary radiologist (> 15 years of experience in prostate MRI) interpreted each mpMRI and prospectively assigned a PI-RADS v2.1 score of 1–5 to each lesion. Maximum 2D lesion diameters were measured on axial MRI, and total prostate gland volumes were calculated by planimetric segmentation. Patients with unilateral mpMRI-visible intraprostatic lesions (1–3 lesion(s)) received subsequent combined MRI/TRUS fusion-guided targeted and systematic biopsies (median of 50 days) using a commercial platform (UroNav; Invivo, Gainesville, Florida). Biopsy sessions were conducted by either an interventional radiologist or a urologist (IR: 19 years, U: > 15 years of experience with MRI/TRUS fusion-guided biopsy). Targeted biopsies included a minimum of 2 cores per MRI-visible lesion with additional cores for lesions that impacted large regions of the prostate (median: 2; range 2–6). Each systematic biopsy included 12 cores per patient: 6 for each prostate hemigland (ipsilateral/contralateral). Biopsy specimens were processed, examined, and assigned a GG by one of the two expert genitourinary pathologists (> 15 years of experience). For each biopsy scheme, only the highest GG was considered. A threshold of GG2 (Gleason score of 3 + 4 = 7, favorable intermediate risk) was used to define csPCa, any PCa was defined as ≥ GG1 and low-grade PCa was defined as GG1 (Gleason score of 3 + 3 = 6). Moreover, an additional analysis was performed using csPCa as GG3 (Gleason score of 4 + 3 = 7, unfavorable intermediate risk) since some institutions find GG3 more clinically relevant and prefer using GG3 to define csPCa.

Statistical analysis

Patient-level and core-level pathology data were obtained to evaluate differences in cancer detection rates between the reference standard, TBx + SBx, and 3 hypothetical reduced-core biopsy schemes, namely targeted biopsy alone, targeted biopsy + ipsilateral systematic biopsy, and targeted biopsy + contralateral systematic biopsy; henceforth, these reduced-core biopsy schemes are referred as TBx alone, TBx + IBx, and TBx + CBx, respectively (Fig. 2). Briefly, patient-based and biopsy core-based descriptive statistics evaluated differences in cancer detection rates per patient in the entire prostate, each hemigland, and in targeted lesions; standard deviations and confidence intervals (95%) were estimated by random sampling with replacement from n = 2000 bootstrapped samples on the patient-level. P-value statistical tests were not valid to compare cancer detection rates determined by highest GG since each reduced-core biopsy scheme was a component of TBx + SBx. Univariable and multivariable logistic regression analyses were conducted to identify predictors of csPCa missed by TBx (SBx: csPCa, TBx: GG1 PCa or benign) and csPCa confined to the MRI-negative hemigland (CBx: csPCa, TBx + IBx: GG1 PCa or benign). Univariable features with p < 0.1 were used to determine the relevant features to be included in the multivariable analysis, and p < 0.05 was considered statistically significant.

Fig. 2.

Illustration of different biopsy schemes including the reference standard (TBx + SBx), targeted biopsy (TBx, dark red), ipsilateral systematic biopsy (IBx, light red), and contralateral systematic biopsy (CBx, light red). Median number of biopsy cores are shown below the corresponding schemes

Results

Patient-level and core-level csPCa and any PCa detection rate

Overall, 212/654 patients (median age of 67 years; median PSA of 6.48 ng/mL) with MRI-visible unilateral lesion(s) were included, and a total of 293 remarkable lesions [and 212 whole glands] were sampled by combined MRI/TRUS fusion-guided TBx + SBx (TBx: 663 total cores; SBx: 2544 total cores, 1272 per sextant). Patient demographics and clinical characteristics, mpMRI details, and highest GG biopsy findings per patient are presented in Table 1. Patient-level cancer detection rates for csPCa and any PCa were 43.4% (92/212) and 66.0% (140/212), respectively. Of the 92 patients with csPCa, most (81.5%, 75/92) demonstrated only ipsilateral hemigland involvement as illustrated in Fig. 3. However, 10.9% (10/92) of patients had bilateral involvement and 7.6% (7/92) demonstrated csPCa confined to the contralateral hemigland (Table 2), and the percentage of Gleason pattern 4 and cancer core lengths were variable in these 17 patients (Table 3). Additionally, patients with ≥ GG3 showed a similar pattern of both ipsilateral hemigland and bilateral involvement (Supplementary Table 1). Of the 140 patients with any PCa, 58.6% (82/140) were sampled exclusively in the ipsilateral hemigland, whereas 32.1% (45/140) had bilateral involvement and 9.3% (13/140) were confined to the contralateral hemigland (Table 2).

Table 1.

Characteristics of patient cohort

Characteristics and demographics	Patients(n)	Median (IQR)
Number of unilateral intraprostatic lesions		1 (1–2)
Any lesion(s)	212
1 lesion	145	-
2 lesions	53	-
3 lesions	14	-
Prior prostate biopsy		-
Yes	147	-
No	65	-
Prior prostate biopsy with PCa		-
Yes	117	-
No	30	-
Ethnicity		-
White	169	-
African American	24	-
Asian	9	-
American Indian	2	-
Multiple	2	-
Unknown	6	-
Age at biopsy (years)		67 (61–72)
PSA at mpMRI (ng/mL)		6.48 (4.30–9.83)
Prostate Volume (cc)		62.5 (44–88.3)
PSAD (ng/mL/cc)		0.091 (0.064–0.150)
Greatest 2D lesion dimension (cm)		1.2 (0.8–1.5)
Time between MRI to biopsy (days)		50 (16–89)
Highest PI-RADS score		4 (3–4)
1	1	-
2	39	-
3	54	-
4	75	-
5	43	-
Highest Gleason Grade Group		1 (0–2)
Benign	72	-
1	48	-
2	50	-
3	17	-
4	13	-
5	12	-

IQR interquartile range, PSA prostate-specific antigen, PSAD prostate-specific antigen density, PI-RADS prostate imaging reporting and data system

Fig. 3.

A patient with a prostate-specific antigen (PSA) level of 1.8 ng/mL has a left-sided (unilateral) lesion in the mid peripheral zone. The lesion appears hypointense (arrows) on T2W MRI in axial (a), sagittal (b), and coronal (c) planes. The lesion displays early enhancement (arrow) on DCE (d), and diffusion restriction (arrows) on both high-b value (b = 1500 s/mm²) DW MRI (e), and ADC map (f). The lesion was assigned to PI-RADS category 4. MRI/TRUS fusion-guided (targeted) biopsy of the lesion yielded a GG2 cancer, similarly, ipsilateral (left-sided) systematic biopsy cores were positive for GG2 cancer; however, contralateral systematic biopsy cores were benign

Table 2.

Distribution of PCa in different regions of the prostate gland based on highest GG for patients receiving combined targeted and systematic biopsy

Gland involvement	csPCa (≥ GG2)		a ny PCa (≥ GG1)
	All patients (n = 212)	Patients with csPCa (n = 92)	All patients (n = 212)	Patients with any PCa (n = 140)
Any	43.4% (92/212)	100% (92/92)	66.0% (140/212)	100% (140/140)
Bilateral	4.7% (10/212)	10.9% (10/92)	21.2% (45/212)	32.1% (45/140)
Ipsilateral only	35.4% (75/212)	81.5% (75/92)	38.7% (82/212)	58.6% (82/140)
Contralateral only	3.3% (7/212)	7.6% (7/92)	6.1% (13/212)	9.3% (13/140)
No PCa involvement	56.6% (120/212)	-	34.0% (72/212)	-

csPCa clinically significant prostate cancer, GG Gleason Grade Group, PCa prostate cancer

Table 3.

Systematic biopsy cores associated with patients who had MRI-negative hemigland (contralateral) csPCa involvement

Patient groups	Patient no.	CBx Highest GG	CBx Gleason pattern 4a (%)	CBx cancer core length (mm)	IBx Highest GG	IBx Gleason pattern 4a (%)	IBx cancer core length (mm)
Patients with contralateral only csPCa (n = 7)	1	GG2	5	1.5	GG1	0	9.0
	2	GG2	10	2.0	Benign	0	0
	3	GG2	10	3.0	GG1	0	4.0
	4	GG2	15	1.0	Benign	0	0
	5	GG2	20	2.0	Benign	0	0
	6	GG2	30	2.7	GG1	0	3.5
	7	GG3	> 50	4.5	Benign	0	0
Patients with bilateral csPCa (n = 10)	1	GG2	< 5	0.5	GG3	60	1.0
	2	GG2	10	0.7	GG1	0	1.0
	3	GG2	10	7.0	GG2	20	0.7
	4	GG2	20	2.0	GG2	30	0.6
	5	GG2	30	1.0	GG2	30	2.5
	6	GG3	> 50	4.5	GG4	100	10.0
	7	GG3	> 50	15.5	Benign	0	0
	8	GG3	80	4.0	GG2	10	6.0
	9	GG4	100	5.0	GG4	100	12.0
	10	GG4	100	2.0	GG5b	< 50	6.0

CBx contralateral systematic biopsy, csPCa clinically significant prostate cancer, IBx ipsilateral systematic biopsy, GG Gleason Grade Group

^aAll GG4 detected by systematic biopsy was Gleason score 4 + 4 = 8

^bGG5 detected in IBx was Gleason score 5 + 4, thus < 50% of Gleason pattern 4 was present

The number of positive biopsy cores per patient with csPCa and any PCa was reported for TBx and separated for SBx based on IBx and CBx hemigland constituents. Among patients with csPCa (n = 92), the median number of ≥ GG2 positive IBx cores was 1 (mean 1.4, range 0–6), the median number of ≥ GG2 positive CBx cores was 0 (mean 0.2, range 0–4), and the median number of ≥ GG2 positive TBx cores was 2 (mean 1.9, range 0–7). Regarding patients with any PCa (n = 140), the median number of ≥ GG1 positive IBx cores was 1 (mean 1.8, range 0–6), the median number of ≥ GG1 positive CBx cores was 0 (mean 0.6, range 0–5), and the median number of ≥ GG1 positive TBx cores was 2 (mean 2.0, range 0–8).

Comparison of reduced-core biopsy schemes to the TBx + SBx reference standard

Among the 3 reduced-core biopsy schemes, patient-level analysis determined that csPCa detection rates would be 35.4% (75/212), 40.1% (85/212), and 39.6% (84/212) for patients sampled by TBx alone, TBx + IBx, and TBx + CBx, respectively (Table 4). Relative to the reference standard, the sensitivity of each alternative scheme was 81.5% (75/92), 92.4% (85/92), and 91.3% (84/92) for TBx alone, TBx + IBx, and TBx + CBx, respectively (Table 5). Hypothetically, combining TBx with only one-sided SBx sextants would have missed 7.6% (7/92) and 8.7% (8/92) of csPCa diagnoses in the contralateral and ipsilateral hemiglands, respectively. Although TBx alone sampled a reasonable proportion of csPCa (81.5%, 75/92), omission of SBx would have undersampled 18.5% (17/92) of csPCa cases missed by TBx, of which 8 were involved in the ipsilateral hemigland, 7 in the contralateral hemigland, and 2 had bilateral involvement (Fig. 4). Conversely, the TBx + IBx scheme would have detected ≥ GG3 in nearly every patient (97.6%, 41/42) compared to the TBx + SBx reference standard (Supplementary Table 2, Supplementary Fig. 1). With regard to any PCa, TBx alone had the lowest detection rates (51.4%), followed by TBx + IBx (59.9%), and TBx + CBx (61.8%) (Table 4). The higher PCa detection rate observed by TBx + CBx may be attributed to increased GG1 in the contralateral hemigland (19.3% [41/212]). In contrast, GG1 detection rates noted in only TBx and IBx cores were 16% (34/212) and 18.9% (40/212), respectively. Given the higher GG1 detection rate by SBx (23.6% [50/212]), complementing SBx or its components (IBx or CBx) with TBx decreased the overall GG1 detection rate.

Table 4.

Cancer detection rate (CDR) with standard deviation and 95% confidence intervals of csPCa and any PCa sampled by TBx + SBx and by reduced-core biopsy schemes

PCa involvement	csPCa (≥ GG2)		any PCa (≥ GG1)
Biopsy Scheme	CDR (% total)	SD (95% CI)	CDR (% total)	SD (95% CI)
TBx + SBx (refer-ence standard)	43.4% (92/212)	3.4% (36.3,49.5)	66.0% (140/212)	3.2% (59.0,71.7)
TBx + IBx	40.1% (85/212)	3.3% (32.5,45.3)	59.9% (127/212)	3.3% (50.0,62.7)
TBx + CBx	39.6% (84/212)	3.4% (33.5,46.2)	61.8% (131/212)	3.3% (55.6,67.9)
TBx alone	35.4% (75/212)	3.2% (27.8,40.6)	51.4% (109/212)	3.4% (40.1,53.3)

CBx contralateral systematic biopsy, CDR cancer detection rate, CI confidence interval, GG Gleason Grade group, IBx ipsilateral systematic biopsy, TBx targeted biopsy, SBx systematic biopsy, SD standard deviation

Table 5.

Detection sensitivity with standard deviation and 95% confidence intervals of reduced-core biopsy schemes compared to TBx + SBx in patients with csPCa and any PCa determined by highest GG

PCa Involvement	csPCa (≥ GG2)		any PCa (≥ GG1)
Reduced-core Biopsy Scheme	Sensitivity relative to TBx + SBx	SD (95% CI)	Sensitivity relative to TBx + SBx	SD (95% CI)
TBx + IBx	92.4% (85/92)	3.1% (83.7,96.0)	90.7% (127/140)	3.0% (80.1,92.0)
TBx + CBx	91.3% (84/92)	2.8% (87.1,97.8)	93.6% (131/140)	1.9% (90.4,97.9)
TBx alone	81.5% (75/92)	4.2% (71.2,87.5)	77.9% (109/140)	3.9% (64.4,79.6)

CBx contralateral systematic biopsy, CDR cancer detection rate, CI confidence interval, GG Gleason grade group, IBx ipsilateral systematic biopsy, TBx targeted biopsy, SBx systematic biopsy, SD standard deviation

Fig. 4.

Distribution of disease involvement on patients harboring clinically significant prostate cancers (csPCa) which were not detected by targeted biopsy (TBx, red) (n = 17). Among these patients, 8 of them were detected by ipsilateral systematic biopsy (IBx, gray), 7 were detected by contralateral systematic biopsy (CBx, blue), and 2 were detected by both IBx and CBx systematic biopsies

Univariable and multivariable analyses for csPCa underestimated by TBx alone and TBx + IBx

Logistic regression of relevant clinical variables (age, prostate-specific antigen density [PSAD], prior biopsy history, prior positive biopsy history) and radiological variables (prostate volume, maximum 2D diameter and binarized PI-RADS score [≥ 4 vs < 4] of index lesions, number of lesions [single vs multiple]) was analyzed to predict the presence of ≥ GG2 csPCa missed by TBx or missed by TBx + IBx. Univariable and multivariable analyses revealed that a large prostate volume (odds ratio [OR] 1.01, p = 0.016) and lower index lesion maximum 2D diameter (OR 0.32, p = 0.045) were predictive for csPCa detected by SBx but underestimated by TBx (Table 6). Although both a large prostate volume and the presence of a lower PI-RADS category (< 4) index lesion were relevant univariable features of exclusively contralateral csPCa (TBx + IBx underestimated), these factors were not predictive in the multivariable analysis (p > 0.05).

Table 6.

Predictive variables of MRI-invisible csPCa involvement

Variables	csPCa missed by TBx but detected by SBx (Npatient = 212, Nevent= 17)						csPCa missed by TBx + IBx but detected by CBx (Npatient = 212, Nevent = 7)
	Univariable			Multivariable			Univariable			Multivariable
	OR	95% Cl	p-value	OR	95% Cl	p-value	OR	95% Cl	p-value	OR	95% Cl	p-value
Age	1.02	0.96–1.09	0.515	-	-	-	1.05	0.94–1.17	0.377	-	-	-
PSAD	0.01	0–18.6	0.229	-	-	-	0.01	0–1180	0.443	-	-	-
Prostate volume	1.01	1–1.02	0.017*	1.01	1–1.02	0.016*	1.02	1–1.03	0.023*	1.01	1–1.03	0.055
Maximum 2D diametera	0.34	0.11–0.99	0.048*	0.32	0.1–0.97	0.045*	0.24	0.04–1.4	0.112
Binarized PI-RADS score (≥4 or<4)a	0.51	0.19–1.39	0.187	-	-	-	0.12	0.01–1.02	0.052	0.14	0.02–1.25	0.079
Binarized no. of lesions (single or multiple)	0.83	0.3–2.36	0.733	-	-	-	1.06	0.22–6.14	0.861	-	-	-
Prior Bx	0.6	0.22–1.67	0.331	-	-	-	1.11	0.21–5.87	0.903	-	-	-
Prior positive Bx	0.7	0.26–1.89	0.484	-	-	-	1.09	0.24–4.97	0.916	-	-	-

Bx biopsy, TBx targeted biopsy, SBx systematic biopsy, CBx contralateral systematic biopsy, IBx ipsilateral systematic biopsy, PSAD prostate-specific antigen density, PI-RADS prostate imaging reporting and data system

^*p<0.05

^aObtained from the index lesion on mpMRI

Discussion

The findings of this study suggest that disease burden for most patients with csPCa (≥ 80%) is confined to the same (ipsilateral) hemigland as the unilateral MRI-visible lesion regardless of the number of lesions (single: 82.5%, 47/57; multiple: 80.0%, 28/35). However, a surprising number of patients with csPCa (17/92, 18.5%) had disease involved in the opposite (contralateral) hemigland. Of this subset of patients, 7/17 had csPCa exclusively in the contralateral hemigland, most (85.7%, 6/7) of which had MRI-visible lesions with PI-RADS score ≤ 3. Further, of the 10/17 patients with bilateral csPCa, 70% (7/10) had MRI-visible lesions with PI-RADS score ≥ 4. Although pathological findings in the contralateral hemigland usually (14/17 patients) presented as a single core of csPCa with small length (median of 2 mm), a significant proportion of patients (8/17) had a core with ≥ 30% Gleason pattern 4. These results suggest that a consequential number of patients with MRI-negative csPCa would be missed or undersampled with the TBx + IBx scheme, thus the TBx + SBx strategy is the best option for csPCa detection.

Commonly used in combination with prostate mpMRI, MRI/TRUS fusion-guided TBx and SBx protocols are widely employed and are considered the standard in many institutions; however, biopsy protocols do vary. Our data may help clarify the added value of non-targeted biopsies in patients with unilateral intraprostatic lesions, thus offering more informed choices of biopsy strategy. For instance, multiple studies investigating csPCa detection in patients with unilateral MRI-visible lesions demonstrated that substituting hemigland TBx + IBx for whole gland TBx + SBx would be feasible for most patients (> 95%) since MRI-invisible csPCa was unusual [24–27]. In this setting, patients may benefit from fewer biopsies while receiving an accurate diagnosis of PCa staging. One particular study by Nakanishi et al. [24] suggested that patients with unilateral intraprostatic lesions who had any of 3 predictive variables including age ≥ 75 years, PI-RADS score ≥ 4, and PSAD ≥ 0.3 were at increased risk of harboring contralateral csPCa and would benefit from combined TBx + SBx. In the absence of these risk factors, the TBx + IBx hemigland scheme was sufficient. A limitation of this and other similar studies is variable classification criteria for mpMRI interpretation such as Likert or PI-RADS v2.0, number of unilateral lesions (single or multiple), method of prostate biopsy (MRI/TRUS vs MRI/TPUS), and definition of csPCa (≥ GG2, GG1 with > 50% positive cores, or ≥ 5 mm core length) [24–27]. Thus, it is difficult to summarize the existing data with regard to reduced-core biopsy strategies.

Our study defined csPCa as ≥ GG2 which is a generally accepted definition; however, we did not account for cancer core length, percent of positive cores, or percent of Gleason pattern 4 in our analysis. Based on combined biopsy findings, cancer detection rates were 66.0% (140/212) and 43.4% (92/212) for patients with any PCa and csPCa, respectively, which are comparable to similar patient populations [25]. Applying the Nakanishi model [24] to our cohort would have underestimated contralateral csPCa burden in 2.8% (6/212) of all patients biopsied by the TBx + IBx hemigland scheme. In contrast, our multivariable analysis determined that patients who had a large prostate volume (p = 0.016) and small index lesion diameter (p = 0.045) were at increased risk of harboring csPCa that would be missed by TBx but detected by SBx. Although statistically significant, large prostate volume probably has no practical implication because the effect size is small (OR 1.01). However, the likelihood of being sampled only by SBx was greater in patients with smaller index lesion diameters (OR 0.32). Furthermore, our multivariable analysis found no predictors of csPCa detected by CBx but missed by TBx + IBx. Thus, there is little evidence that it is possible to predict the presence of contralateral csPCa in patients with MRI-visible unilateral lesions.

The definition of csPCa in this study is important and directly relates to the results. Considering csPCa as ≥ GG2, TBx + SBx detected 7.6% (7/92) more csPCa compared to the TBx + IBx hemigland approach implying that TBx + SBx is superior. On the contrary, for csPCa defined as ≥ GG3, TBx + SBx would have detected only 2.4% (1/42) more patients compared to TBx + IBx (97.6%, 41/42) indicating that an ipsilateral hemigland biopsy scheme might be sufficient for such patients with unilateral MRI-visible lesions. This is important for clinical centers defining csPCa as ≥ GG3 and prioritizing ≥ GG3 cancer detection. However, defining csPCa in this manner forestalls the possibility of detecting contralateral GG2 lesions with pathological features associated with higher mortality such as cribriform pattern and/or intraductal carcinoma [28, 29]. Therefore, by cautiously defining csPCa as ≥ GG2, the benefit of combined TBx + SBx outweighs the risk of potentially missing csPCa using the ipsilateral hemigland scheme.

Although detecting the most aggressive PCa within a gland remains the purpose of a prostate biopsy, management decisions can be based on the laterality of PCa to determine which patients are eligible for focal therapies such as focal laser ablation (FLA) [30]. Per guidelines, only patients with confirmed unilateral GG2 or GG3 PCa are candidates for FLA, and combined TBx + SBx provides the best assessment [31]. One study determined that 37.5% (21/56) of patients with unilateral MRI-visible intraprostatic lesions and TBx-confirmed ≤ GG2 PCa had additional contralateral csPCa (≥ GG2) sampled by SBx and, therefore, would be ineligible focal therapy candidates [32]. However, prior evidence suggests that complementing MRI fusion-guided TBx with SBx can still underestimate disease burden assessed by whole mount histopathology [33]. For instance, > 40% of patients (range 66–185) who underwent TBx + SBx with confirmed unilateral csPCa and subsequent radical prostatectomy actually had bilateral disease at whole mount histopathology and would, therefore, fail to meet the requirements for FLA [34–36].

There are some limitations of this study. First, we did not use stringent PI-RADS category thresholds and included patients who had unilateral lesions with low PI-RADS scores (PI-RADS ≤ 2), which may not be biopsied at other centers. This might have affected our cancer detection rates and comparisons between biopsy schemes given that targeted biopsy samples csPCa more accurately for higher PI-RADS scores [15]. Second, we included patients with more than one unilateral lesion which increased the likelihood of multiple sampling bias (i.e., TBx and IBx may have sampled the same lesion); thus, we did not perform core-level analysis comparing these regions. Third, we also included patients with large unilateral lesions (median diameter of 1.2 cm) such that it was possible for some lesions to have microscopic extension into the MRI-negative hemigland. Fourth, we excluded patients with unilateral lesions who received MRI/TPUS biopsy due to differences in SBx sampling locations compared to MRI/TRUS biopsy. Fifth, we included few patients (n = 2) with large prostate volumes (> 200 cc) who underwent standard 12-core TRUS SBx. Future studies with larger cohorts may further interrogate the impact of prostate volume by stratifying patients based on gland size. As discussed above, we defined csPCa as ≥ GG2 which is widely adopted; nonetheless, we acknowledge that csPCa definitions vary among institutions and may impact clinical decision-making [2, 37]. Additionally, we do not have surgical pathology as a validation method since most of our patients did not undergo prostatectomy. Moreover, our limited number of patients for logistic regression warrants a larger validation cohort to evaluate predictors of csPCa confined to the MRI-negative hemigland. Lastly, our analyses were limited to descriptive statistics due to the reduced-core biopsy schemes being constituents of the TBx + SBx reference standard, thus we could not perform statistical tests including modeling approaches comparing PCa detection between techniques.

In conclusion, hypothetical prostate biopsy schemes using fewer biopsy cores resulted in lower cancer detection rates compared to TBx + SBx. Recommending ipsilateral hemigland prostate biopsy, although sufficient for many patients (92.0%) with unilateral MRI-visible lesions, would have underestimated 8.0% of patients with ≥ GG2 csPCa. More evidence is needed to identify patients with lower risk of contralateral csPCa before the ipsilateral hemigland biopsy strategy can be considered. Combined TBx and SBx should be standard practice in similar patient cohorts until such evidence is validated.