Diagnostic Performance of a Magnetic Resonance Imaging-directed Targeted plus Regional Biopsy Approach in Prostate Cancer Diagnosis: A Systematic Review and Meta-analysis

Take Home Message It has been shown that limiting systematic biopsies to the proximity of magnetic resonance imaging (MRI)-positive prostate lesions is a promising diagnostic biopsy approach. However, considering the high degree of heterogeneity between the studies included in our review, prospective clinical trials are needed to further investigate and optimize this MRI-directed targeted plus regional biopsy approach.


Introduction
For biopsy-naïve men, the European Association of Urology (EAU) guidelines on prostate cancer (PCa) diagnosis currently recommend upfront magnetic resonance imaging (MRI) and then an MRI-directed targeted biopsy (TBx) plus systematic biopsy (SBx) approach in MRI-positive cases [1]. With this recommended biopsy approach, both the ipsilateral and contralateral lobes are still biopsied in a predominantly random systematic fashion, which is unique in the diagnostic process for solid-organ cancers [2,3]. Most men benefit diagnostically from increased (perilesional/regional) sampling of the index lesion, while complementary SBx can increase unnecessary biopsy cores, potential harms, and patient burdens, and identify indolent cancers in men with false-positive MRI scans [4][5][6][7][8].
The concept of standard SBx is therefore widely disputed. MRI-directed biopsy approaches using TBx and only regional systematic biopsies (RBx) rather than standard SBx have been explored as an alternative approach to minimize biopsy cores, targeting errors, and grade migration. Use of such an MRI-directed TBx + RBx approach yields equivalent detection rates for significant PCa in comparison to the recommended MRI-directed TBx + SBx approach and reduces the number of biopsy cores and overdiagnosis rates [9][10][11][12]. Currently, there is no consensus on such an MRIdirected TBx + RBx approach and several definitions can be found in the literature, including focal saturation biopsy, perilesional biopsy, regional targeted biopsy, and targeted sector biopsy [3,[11][12][13]. Although different interpretations exist, the scope of these interpretations is broadly in line: limit SBx to the proximity of the MRI-positive lesion. In this systematic review, we investigate the diagnostic performance of MRI-directed TBx + RBx approaches in comparison to MRI-directed TBx alone and TBx + SBx approaches.

Objective
The aim of the review was to systematically evaluate the diagnostic performance of MRI-directed TBx + RBx approaches in detecting International Society of Urological Pathology (ISUP) grade group (GG) !2 PCa in comparison to MRI-directed TBx alone and TBx + SBx approaches in men suspected of having PCa and with positive MRI findings.

Study design
The study design in Fig. 1 shows the reference test (standard of care) using an MRI-directed TBx + SBx approach, index test 1 using an MRI-directed TBx + RBx approach, and index test 2 using an MRI-directed TBx approach. Given that there is currently no consensus on an MRI-directed TBx + RBx approach, different interpretations were included in this systematic review. RBx was defined as SBx in the proximity of the MRI-positive lesion, either via MRIdirected perilesional biopsy (biopsies around the index lesion), MRI-directed RBx (biopsies in the region of the index lesion), systematic ipsilateral biopsies (biopsies in the same lobe as the MRI-identified index lesion) or systematic sector biopsy (biopsies in the same sector as the MRIidentified index lesion). We excluded studies that failed to evaluate the diagnostic performance of TBx + RBx approaches in comparison to TBx alone and TBx + SBx. We also excluded letters, editorials, study protocols, case reports, brief correspondence articles, and conference abstracts because comprehensive information is needed to correctly assess study quality and the study results.

Data extraction
We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) process for reporting the studies included and excluded, with the recommended flow chart showing the number of papers identified and included or excluded at each stage ( Fig. 2) [14]. The abstract and full-text screening and subsequent data extraction were carried out by two reviewers (M.J.H. and M.F.S.) independently. Discrepancies between reviewers were resolved via discussion (M.J.H., M.F.S., and I.G.S.). A data extraction form was developed to collect information on study methodology, patient characteristics, and MRI, biopsy, and pathology protocols. Positive MRI was defined as identification of a lesion suspicious for GG !2 PCa on the MRI scan with a score of 3-5 on a 5-point scale (Likert or Prostate Imaging-Reporting and Data System [PI-RADS]) [15,16].

2.6.
Assessment of study quality The quality of the studies was reviewed according to the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) criteria [17].

Statistical analysis
The primary outcome was detection of GG !2 PCa. To synthesize the results, we performed a meta-analysis using RevMan version 5.3.3 (Cochrane Collaboration, London, UK). Three comparisons were analyzed ( Fig. 1 Regional biopsy (perilesional or ipsilateral or sector) Fig. 1 -Definition of an MRI-directed targeted plus regional biopsy (TBx + RBx) approach: TBx with additional perilesional, ipsilateral, or sector biopsies. MRI = magnetic resonance imaging.
A value of p < 0.05 was considered to indicate statistical significance. All statistical analyses were performed using SPSS for MacOS version 27 (IBM, Armonk, NY, USA).

Evidence synthesis
Nine potentially eligible studies were identified by the first reviewer (M.J.H.) and seven potentially eligible studies were identified by the second reviewer (M.F.S.); an agreement rate of 78% was observed between the reviewers. Ultimately, eight studies were eligible for inclusion: three prospective cohort studies [9,11,18] and five retrospective cohort studies [3,10,12,13,19]. All the studies included investigated the performance of their proposed TBx + RBx approach relative to both TBx + SBx and TBx alone. Seven studies performed MRI/transrectal ultrasound (TRUS) fusion biopsies, whereas one study performed in-bore MRI-guided biopsies followed by systematic TRUS-guided biopsies (Supplementary Table 1). This systematic review comprises a cumulative of 2603 MRI-positive men, all suspected of having GG !2 PCa who underwent prostate biopsies using TBx + SBx approaches (  Table 3).

IdenƟficaƟon of studies via databases and registers
Detection of GG !2 PCa using MRI-directed TBx was visually best using three to four biopsy cores (Supplementary Fig. 1B). However the number of biopsy cores was not significantly associated with GG !2 PCa detection (p = 0.07; B = 5.75, 95% CI À0.76 to 12.26).

3.4.
Publication bias, study quality, and heterogeneity Publication bias was assessed via funnel plot analysis (Supplementary Fig. 2). We found no strong evidence of publication bias on graphical inspection. Most of the individual studies followed the Standards for Reporting of Diagnostic Accuracy (STARD) guidelines [20] and prostate Standards of Reporting for MRI-Targeted biopsy studies (START) guidelines [21]. We reviewed the reports according to the QUADAS-2 criteria [17]. The risk of bias was assessed in a qualitative manner. The overall quality of the individual diagnostic studies should not be considered as high. We concluded that the overall methodological quality of the studies was moderate. Supplementary Fig. 3 provides a detailed description and analysis.
We did not identify heterogeneity between study outcomes (Fig. 3). The recruitment of men with a suspicion of PCa was focused on positive MRI findings. These men all underwent biopsies. Patient characteristics (age, prostatespecific antigen levels, and prostate volume) were representative of the general PCa patient population.

Discussion
The current review synthesizes the evidence on MRIdirected TBx + RBx approaches for PCa diagnosis in MRI-positive men. Our analyses show noninferiority to the currently recommended MRI-directed TBx + SBx approach [1]. On the basis of these analyses, an MRI-directed TBx + RBx approach could potentially be a future alternative to current recommendations.
Although previous studies have shown that MRI-directed TBx is a promising biopsy strategy [22,23], there are still substantial differences in GG !2 PCa detection when compared to saturation template biopsy [13,[24][25][26][27]. Additionally found cancers are often in sextants adjacent to MRIpositive lesions, while systematic sampling of normalappearing nonadjacent sectors does not alter risk stratification in the majority of cases. The main reasons for missing these systematically found GG !2 cancers are imprecise lesion registration (underestimation of true tumor volume) and targeting errors due to MRI and ultrasound/cognitive fusion inaccuracies [28][29][30][31]. Additional SBx is therefore still advocated to improve GG !2 PCa detection, but at the expense of a higher overall number of biopsy cores [32,33]. The proposed MRI-directed TBx + RBx approach may overcome these differences in GG !2 PCa detection, as imprecise lesion registration and targeting errors are covered by the RBx cores.
MRI-directed TBx + RBx substantially improves lesion detection, tumor characterization, and tumor volume estimation, and also reduces the total number of biopsy cores and false-positive MRI results, potentially reducing procedure time and pathologist workload. Owing to the inclusion of three studies in which a 24-core SBx template was used, the median number of biopsy cores was particularly high. The exact framing of an MRI-directed TBx + RBx approach still needs to be defined. The index lesion size, PI-RADS score, biopsy history, lesion location, and number of lesions can potentially influence this approach and should therefore be taken into account (ie, fewer biopsy cores may be considered for larger lesions and more biopsy cores for smaller lesions).
Reducing the number of biopsies by limiting SBx to the proximity of the MRI-positive lesion could potentially also reduce overdiagnosis rates. The studies included suggest a reduction of 5-19% in detection of GG 1 PCa, as RBx reduce the chance of finding indolent PCa in men with falsepositive MRI scans [9,18,19]. At the same time, this reduction in biopsy cores did not significantly impact on histopathological concordance [12]. However, a substantial number of men are still at risk of grade migration; biopsy cores underestimate or overestimate the true Gleason score. As shown by Raman et al [12], use of a TBx + RBx approach increased the risk of grade inflation when compared to performing TBx or SBx alone. TBx and RBx may result in a grade shift [34,35]. The 2019 ISUP consensus conference on PCa grading has overcome some of the issues raised, by aggregating Gleason scores for biopsy cores [36]. The strategy for histopathology analysis of perilesional biopsy cores needs to be further explored, as it touches not only on tumor characterization (sampling heterogeneity) and related grade shifts but also on targeting errors and volume estimations [37]. The major strength of this diagnostic meta-analysis is its focus on studies using biopsy sampling via MRI-directed TBx, RBx, and SBx in the same patient. However, this meta-analysis has several limitations that may reduce the strengths of the conclusions. Demonstrated differences in many of the variables in conducting MRI and biopsies may all contribute to study heterogeneity. Since there is no consensus on how to perform MRI-directed TBx + RBx, the definition of regional sampling differed greatly between the studies included. Both the number and location of biopsy cores were different in the studies, limiting the generalizability of the proposed approach, so the results need to be regarded with caution. Although it affects our efforts to synthesize the available evidence, heterogeneity does offer opportunities for increasing our knowledge and provides relevant additional information on the topic of interest. In addition, this retrospective study presents a diagnostic biopsy approach and focuses on reducing biopsy cores. On the basis of the studies included it remains unknown how an MRI-directed TBx + RBx approach impacts therapeutic choices. Whether it is preferable to decrease the number of biopsy cores in the diagnostic setting or improve prostatic mapping for treatment planning in cases in which prostate cancer is present still needs to be assessed. We did not register our review protocol before the article search process, and we therefore mention this as a limitation. Finally, eight studies were included in this meta-analysis and were predominantly retrospective in nature. Moderate quality on critical appraisal (QUADAS-2) limits the strengths of the conclusions that can be drawn.

Conclusions
This systematic review synthesized the evidence on the diagnostic performance of MRI-directed TBx + RBx approaches. In comparison to the current MRI-directed TBx + RBx approach recommended, a nonsignificant difference in detection of GG !2 PCa was observed, although the total number of biopsies was significantly reduced. Considering the high degree of heterogeneity between the studies included, the findings should be interpreted with caution. Future prospective clinical trials are needed to further investigate and optimize this MRI-directed TBx + RBx approach and to substantiate the argument to obviate standard SBx from the current recommended diagnostic workup for MRI-positive men. Other: None.
Financial disclosures: Marinus J. Hagens certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.