Abstract
BACKGROUND.
In-gantry MRI-guided biopsy (MRGB) of the prostate has been shown to be more accurate than other targeted prostate biopsy methods. However, the optimal number of cores to obtain during in-gantry MRGB remains undetermined.
OBJECTIVE.
The purpose of this study was to assess the diagnostic yield of obtaining an incremental number of cores from the primary lesion and of second lesion sampling during in-gantry MRGB of the prostate.
METHODS.
This retrospective study included 128 men with 163 prostate lesions who underwent in-gantry MRGB between 2016 and 2019. The men had a total of 163 lesions sampled with two or more cores, 121 lesions sampled with three or more cores, and 52 lesions sampled with four or more cores. A total of 40 men underwent sampling of a second lesion. Upgrade on a given core was defined as a greater International Society of Urological Pathology (ISUP) grade group (GG) relative to the previously obtained cores. Clinically significant prostate cancer (csPCa) was defined as ISUP GG 2 or greater.
RESULTS.
The frequency of any upgrade was 12.9% (21/163) on core 2 versus 10.7% (13/121) on core 3 (p = .29 relative to core 2) and 1.9% (1/52) on core 4 (p = .03 relative to core 3). The frequency of upgrade to csPCa was 7.4% (12/163) on core 2 versus 4.1% (5/121) on core 3 (p = .13 relative to core 2) and 0% (0/52) on core 4 (p = .07 relative to core 3). The frequency of upgrade on core 2 was higher for anterior lesions (p < .001) and lesions with a higher PI-RADS score (p = .007); the frequency of upgrade on core 3 was higher for apical lesions (p = .01) and lesions with a higher PI-RADS score (p = .01). Sampling of a second lesion resulted in an upgrade in a single patient (2.5%; 1/40); both lesions were PI-RADS category 4 and showed csPCa.
CONCLUSION.
When performing in-gantry MRGB of the prostate, obtaining three cores from the primary lesion is warranted to optimize csPCa diagnosis. Obtaining a fourth core from the primary lesion or sampling a second lesion has very low yield in upgrading cancer diagnoses.
CLINICAL IMPACT.
To reduce patient discomfort and procedure times, operators may refrain from obtaining more than three cores or second lesion sampling.
Keywords: cores per lesion, in gantry, MRI-guided prostate biopsy, number of cores, prostate biopsy
Prostate cancer is the leading cancer diagnosis in men in the United States [1–3]. Traditionally, the diagnosis is made according to systematic prostate biopsy in which 12 cores are obtained from the prostatic parenchyma using ultrasound guidance [1, 4, 5]. This method has been shown to both underdetect clinically significant prostate cancer (csPCa) and overdetect indolent prostate cancer [6, 7]. Multiparametric MRI localizes suspicious lesions in the prostate, and MRI-guided biopsy (MRGB) of the prostate has emerged as a technique that both improves prostate cancer detection and minimizes the number of samples required [1, 6–11]. MRI may be used as a noninvasive method to identify men who are unlikely to have prostate cancer, allowing some to avoid biopsy [7, 10, 12]. PI-RADS version 2 (v2) is used to stratify MRI findings into low-risk (PI-RADS categories 1 and 2), intermediate-risk (PI-RADS category 3), and high-risk (PI-RADS categories 4 and 5) groups [13]. Patients with MRI results showing PI-RADS category 1 or 2 may avoid biopsy, taking into consideration their risk, clinical biomarkers, follow-up regimens, and the likelihood of yield from systematic prostate biopsy [13]. Current guidelines encourage a combination of systematic and targeted biopsy in biopsy-naïve patients with PI-RADS category 4 or 5 lesions, with the option to consider MRGB alone for PI-RADS category 3 lesions [13]. High-risk patients with a prior negative biopsy and a PI-RADS category 3–5 lesion are recommended to undergo MRGB alone or in combination with systematic biopsy [13].
Various MRGB techniques have emerged as alternatives to traditional transrectal ultrasound–guided biopsy, including cognitive, fusion, and in-gantry MRGB [1, 14]. Advantages of in-gantry MRGB over other methods are improved visualization of the needle trajectory in relation to the lesion and accurate documentation that the lesion was indeed sampled [1]. Although in-gantry MRGB is time-consuming relative to other targeting methods [1], it has been shown to be superior in accuracy and is particularly useful for targeting small and anterior lesions [15–17]. As with other MRGB techniques, in-gantry MRGB may be performed by the transrectal or transperineal route [18]. Despite the transperineal route conferring a lower infection risk and better access to the anterior prostate, the transrectal route remains more commonly performed [18, 19].
Although MRGB is associated with fewer cores compared with transrectal ultrasound–guided biopsy [1, 6, 10, 11], an optimal number of cores for MRGB is yet to be determined. A 2016 consensus statement recommended that at least two cores be obtained per lesion during MRGB [20]. Prior studies evaluated the value of obtaining varying numbers of cores during MRGB [8, 17, 21–26], although a paucity of studies have examined the value of obtaining more than two cores for in-gantry MRGB [26, 27]. Such insight is important for optimizing in-gantry MRGB to maximize the diagnosis of csPCa while reducing patient discomfort and procedure time [21, 28, 29]. Furthermore, to our knowledge, the value of sampling other lesions in addition to the most suspicious lesion during in-gantry MRGB has not yet been evaluated. If csPCa diagnosis does not increase by sampling a second lesion in addition to the most suspicious lesion on diagnostic MRI, then sampling the second lesion may be avoided, thereby further reducing procedure time.
The objective of this study was to evaluate the diagnostic yield of obtaining an incremental number of cores from the primary lesion and of sampling a second lesion during in-gantry MRGB of the prostate.
Methods
Patients
This institutional review board–approved, HIPAA-compliant,, single-institution, retrospective study included 131 consecutive patients who underwent in-gantry MRGB from October 2016 through May 2019. Written informed consent was waived by the institutional review board. Each patient had at least one PI-RADS v2 category 3, 4, or 5 lesion on prior diagnostic MRI and subsequently underwent in-gantry MRGB. For the sequential core analysis, three patients with four lesions were excluded because the order of the cores within a lesion was not recorded. An additional eight lesions that had only a single core were excluded without further patient exclusions (one of two lesions in five patients [five lesions], one of three lesions in one patient, and two of three lesions in one patient), leaving a final sample of 128 patients. A total of 128 patients had at least two cores per lesion (163 lesions), 114 patients had at least three cores per lesion (121 lesions), and 52 patients had at least four cores per lesion (52 lesions) (Fig. 1A). For the multiple lesion analysis, all 131 patients were included with no exclusions. The lesion analysis included a total of 175 lesions in 131 patients; 91 patients underwent biopsy of a single lesion, and 40 patients underwent biopsy of at least two lesions (two lesions biopsied in 36 patients, three lesions biopsied in four patients) (Fig. 1B). A total of 57 patients were included in two prior studies exploring MR fingerprinting of prostate cancer [30, 31].
Fig. 1—
Study cohort.
A, Flowchart shows sequential core analysis cohort.
B, Flowchart shows multiple lesion analysis cohort.
Diagnostic MRI Interpretation
All prebiopsy standard-of-care diagnostic MRI examinations were performed on 3-T MRI scanners (Verio [Siemens Healthineers], Skyra [Siemens Healthineers], or Achieva [Philips Healthcare]). The MRI protocol and lesion interpretations were based on PI-RADS v2 recommendations [32]. Examinations were interpreted before in-gantry biopsy by one of five board-certified body radiologists with 19–35 years of experience. The interpreting radiologist was aware of clinical details and serum PSA levels. At the time of original clinical interpretation, each lesion was characterized by PI-RADS v2 category, diameter, and location based on the prostate sector map [32]. On referral for biopsy, the examinations were reviewed by the radiologist performing the biopsy (see the In-Gantry Biopsy Technique section), and any discrepancy in MRI interpretation was discussed at a weekly multidisciplinary meeting, at which time a consensus interpretation was reached. These final prospectively determined interpretations were used for the purposes of this investigation, and the images were not rereviewed. No minimum size was required for a lesion to undergo MRGB, and the smallest biopsied lesion measured 0.45 cm. The reported lesion locations were classified in terms of anterior versus posterior half of the prostate, zone (peripheral zone or transition zone), and level of the prostate (apex; apex and mid prostate; apex, mid prostate, and base; mid prostate; mid prostate and base; or base).
In-Gantry Biopsy Technique
All in-gantry biopsies were performed on a 3-T MRI scanner (Verio or Skyra) with a dedicated MRI-compatible biopsy device (DynaTRIM system, Invivo). Per institutional protocol, patients were provided 1 g of ceftriaxone intramuscularly for antibiotic prophylaxis before the biopsy. The biopsies were performed by one of three body and interventional radiologists (V.G. [n = 116], I.J.P. [n = 3], and D.A.N. [n = 12], with 19, 10, and 30 years of experience, respectively) using an MRI-compatible 18-gauge beveled-tip needle with 20-mm throw. Both 150- and 175-mm-length needles were used according to the lesion location. Patients were placed in the prone position, and the rectal wall was numbed using lidocaine gel before the procedure. The biopsy system was mounted on a base plate fitted to the MRI table. The disposable needle sleeve attached to the arm of the biopsy system was inserted transrectally. Biopsy planning sequences were obtained using a 6-channel body matrix coil and a 24-channel spine coil. The acquisition parameters and the sequence durations are provided in Table S1, which can be viewed in the AJR electronic supplement to this article, available at doi.org/10.2214/AJR.20.24918.
The biopsy procedure was planned using assisted planning software (DynaLOC, Invivo). The lesion locations were confirmed on axial T2-weighted turbo spin-echo sequence and standard echo-planar DWI. Serial T2-weighted sagittal and coronal oblique images were obtained to monitor the needle path and subsequent adjustments. Images were obtained to confirm the needle position within the central portion of the lesion before taking each sample (Fig. 2). A minimum of two cores were taken for each lesion, and third and fourth cores were taken at the discretion of the operator. Factors contributing to a decision to obtain a greater number of cores per lesion included uncertainty that the lesion was adequately sampled and heterogeneity of the lesion on T2-weighted images or the ADC map. Patient discomfort contributed to the decision not to obtain additional cores. For large lesions, the first core targeted the center of the lesion, and subsequent cores targeted other parts of the lesion, including low ADC regions. For small lesions, the center of the lesion was targeted for each core. For patients under active surveillance at the time of biopsy, the most suspicious lesion on MRI was targeted, even if it differed from their known cancerous lesion. The cores for each lesion were numbered sequentially and placed in separate vials. In patients with multiple targeted lesions, the lesion with the higher PI-RADS category was targeted first. In patients with two or more targeted lesions that had the same PI-RADS category, the larger of these lesions was targeted first.
Fig. 2—
75-year-old man with International Society of Urological Pathology grade group 1 prostate cancer who was under active surveillance and underwent in-gantry MRI-guided prostate biopsy.
A, ADC map shows lesion (arrow).
B–D, Oblique coronal turbo spin-echo T2-weighted MR images show needle position in relation to lesion before biopsy for planning (B) and during biopsy of first (C) and second (D) cores of lesion to guide biopsy needle.
Histologic Analysis and Definition of Upgrades
Biopsy results were described according to the 2014 International Society of Urological Pathology (ISUP) grade group (GG) classification system [33]. ISUP GG 0 was used for no cancer, ISUP GG 1 or higher for any prostate cancer, and ISUP GG 2 or higher for csPCa. An upgrade in ISUP GG from the first to second core of a lesion was defined as the second core ISUP GG being higher than the first core ISUP GG. An upgrade from core 2 to core 3 was defined as the third core ISUP GG being higher than the maximum ISUP GG of either of the first two cores. An upgrade from core 3 to core 4 was defined as the fourth core ISUP GG being higher than the maximum ISUP GG of any of the first three cores. An upgrade from a patient’s first lesion to second lesion was defined as the ISUP GG being greater for the second targeted lesion than for the first targeted lesion. Upgrades from ISUP GG 0 to ISUP GG 1 (no cancer to clinically insignificant cancer), from ISUP GG 0 to ISUP GG 1 or greater (no cancer to any cancer), from ISUP GG 0 to ISUP GG 2 or greater (no cancer to csPCa), from ISUP GG 1 to ISUP GG 2 or greater (clinically insignificant cancer to csPCa), from ISUP GG 1 or less to ISUP GG 2 or greater (no cancer or clinically insignificant cancer to csPCa), and from ISUP GG 2 or less to ISUP GG 3 or greater (no or low-grade cancer to high-grade cancer, defining high-grade cancer as Gleason score 4 + 3 or higher) were summarized. Because a single lesion could fulfill multiple criteria for upgrade, the presence of any upgrade for a given lesion was also summarized.
Statistical Analysis
Lesions sampled by two cores, three cores, and four cores were summarized in terms of prostate volume, lesion diameter, lesion location, and PI-RADS v2 category. This analysis by number of cores included all biopsied lesions in the study sample, whether the first, second, or third lesion biopsied in a given patient. First and second biopsied lesions were also summarized in terms of these characteristics; third biopsied lesions were excluded from the formal statistical assessments that were performed based on biopsy order. Categoric variables were reported as numbers and percentages, and continuous variables were reported as medians and interquartile ranges (IQRs). Wilcoxon sum rank and Kruskal-Wallis tests were used to compare continuous data between groups, whereas the chi-square test was used for categoric data. The 95% CIs were calculated for the percentage of lesions with an upgrade. Univariable and multivariable logistic regression analyses were used to evaluate the associations of lesion size and location with upgrading. For the purpose of regression analysis, lesions within both the peripheral zone and transition zone were excluded (n = 3), and ISUP GGs 3, 4, and 5 were combined into one group (ISUP GG ≥ 3). Potential trends in upgrade frequency across PI-RADS v2 categories 3, 4, and 5 and across the number of sampled cores were examined using the exact Cochran-Armitage test with increasing and decreasing alternative hypotheses, respectively. All statistical analyses were performed on R software (version 4.0.2, R Foundation for Statistical Computing). A p value of less than .05 was considered statistically significant.
Results
Patient and Lesion Characteristics
The characteristics of the 128 patients included in the final analysis are summarized in Table 1. The median age was 68 years (IQR, 63–72 years). The median PSA was 7.9 ng/mL (IQR, 5.7–10.6 ng/mL), median PSA density was 0.14 ng/mL/mL (IQR, 0.08–0.25 ng/mL/mL), and median prostate volume was 59.1 g (IQR, 39.4–83.7 g). A total of 63.3% (81/128) of patients had at least one prior negative prostate biopsy, 7.0% (9/128) had no prior prostate biopsy, and 29.7% (38/128) were under active surveillance for known prostate cancer at the time of in-gantry MRGB. Table 2 summarizes the characteristics of sampled lesions stratified by varying numbers of cores. A higher number of lesion core samples was associated with anterior lesion location (anterior location in 21.4%, 40.6%, and 50.0% of lesions sampled by two, three, and four cores, respectively; p = .02). A higher number of cores was also associated with a higher PI-RADS v2 category (PI-RADS category 5 in 11.9%, 31.9%, and 38.5% of lesions sampled by two, three, and four cores, respectively; p = .01). The number of cores was not associated with prostate volume or lesion diameter, zone, or level (all p > .05).
TABLE 1:
Characteristics of Patients Who Underwent Lesion Biopsy
| Characteristic | ≥ 2 Cores (n = 128) | ≥ 3 Cores (n = 114) | ≥ 4 Cores (n = 52) |
|---|---|---|---|
| Age (y) | 68 (63–72) | 68 (63–72) | 66 (62–70) |
| PSA (ng/mL) | 7.9 (5.7–10.6) | 7.9 (5.7–10.8) | 9.0 (6.3–12.5) |
| PSAD (ng/mL/mL) | 0.14 (0.08–0.25) | 0.15 (0.08–0.28) | 0.16 (0.10–0.29) |
| Time between MRI and biopsy (mo) | 2.0 (1.0–4.0) | 2.0 (1.3–4.0) | 2.0 (1.0–4.0) |
| Prostate volume (g) | 59.1 (39.4–83.7) | 57.1 (38.7–83.4) | 57.1 (36.0–88.7) |
| ISUP GG from in-gantry MRGB (maximum across lesions) | |||
| Negative | 60 (46.9) | 48 (42.1) | 21 (40.4) |
| 1 | 19 (14.8) | 18 (15.8) | 9 (17.3) |
| 2 | 25 (19.5) | 23 (20.2) | 11 (21.2) |
| 3 | 12 (9.4) | 13 (11.4) | 5 (9.6) |
| 4 | 6 (4.7) | 6 (5.3) | 4 (7.7) |
| 5 | 6 (4.7) | 6 (5.3) | 2 (3.8) |
| PI-RADS v2 score of first biopsied lesion | |||
| 3 | 36 (28.1) | 30 (26.3) | 12 (23.1) |
| 4 | 48 (37.5) | 42 (36.8) | 20 (38.5) |
| 5 | 44 (34.4) | 42 (36.8) | 20 (38.5) |
| Previous prostate biopsy status | |||
| Active surveillance | 38 (29.7) | 35 (30.7) | 16 (30.8) |
| No prior biopsy | 9 (7.0) | 8 (7.0) | 4 (7.7) |
| Prior negative biopsy | 81 (63.3) | 71 (62.3) | 32 (61.5) |
Note—Values are the median with interquartile range in parentheses or the number of patients with the percentage in parentheses. PSAD = PSA density, ISUP = International Society of Urological Pathology, GG = grade group, MRGB = MRI-guided biopsy, v2 = version 2.
TABLE 2:
Lesion Characteristics Stratified by Number of Lesion Core Samples
| Characteristic | All Lesions (n = 163) | 2 Cores (n = 42) | 3 Cores (n = 69) | 4 Cores (n = 52) | p |
|---|---|---|---|---|---|
| Prostate volume (g) | 61.8 (40.8–86.5) | 71.4 (54.3–87.3) | 58.6 (42.9–77.2) | 57.1 (36.0–88.7) | .14 |
| Lesion diameter (cm) | 1.2 (0.9–1.7) | 1.1 (0.8–1.3) | 1.2 (0.9–1.7) | 1.3 (0.9–1.7) | .34 |
| Anterior location | 63 (38.7) | 9 (21.4) | 28 (40.6) | 26 (50.0) | .02 |
| Zone | .73 | ||||
| Both TZ and PZ | 3 (1.8) | 0 (0) | 1 (1.4) | 2 (3.8) | |
| PZ | 74 (45.4) | 20 (47.6) | 31 (44.9) | 23 (44.2) | |
| TZ | 86 (52.8) | 22 (52.4) | 37 (53.6) | 27 (51.9) | |
| Prostate level | .65 | ||||
| Apex | 43 (26.4) | 9 (21.4) | 21 (30.4) | 13 (25.0) | |
| Apex, mid prostate | 17 (10.4) | 2 (4.8) | 8 (11.6) | 7 (13.5) | |
| Apex, mid prostate, base | 3 (1.8) | 0 (0) | 1 (1.4) | 2 (3.8) | |
| Mid prostate | 57 (35.0) | 18 (42.9) | 23 (33.3) | 16 (30.8) | |
| Base, mid prostate | 9 (5.5) | 4 (9.5) | 2 (2.9) | 3 (5.8) | |
| Base | 34 (20.9) | 9 (21.4) | 14 (20.3) | 11 (21.2) | |
| PI-RADS v2 score | .01 | ||||
| 3 | 58 (35.6) | 23 (54.8) | 23 (33.3) | 12 (23.1) | |
| 4 | 58 (35.6) | 14 (33.3) | 24 (34.8) | 20 (38.5) | |
| 5 | 47 (28.8) | 5 (11.9) | 22 (31.9) | 20 (38.5) |
Note—Values are the median with interquartile range in parentheses or the number of lesions with the percentage in parentheses. TZ = transition zone, PZ = peripheral zone, v2 = version 2.
Sequential Core Analysis
Among the 163 lesions sampled by at least two cores, 23.3% (38/163) and 28.8% (47/163) of cores 1 and 2 showed csPCa, respectively. Among the 121 lesions sampled by at least three cores, 27.3% (33/121), 35.5% (43/121), and 34.7% (42/121) of cores 1, 2, and 3 showed csPCa, respectively. Among the 52 lesions sampled by at least four cores, 25.0% (13/52), 36.5% (19/52), 34.6% (18/52), and 30.8% (16/52) of cores 1, 2, 3, and 4 showed csPCa, respectively (Fig. 3).
Fig. 3—
Sequential core analysis. Bar graph shows distribution of biopsy results showing negative findings and International Society of Urological Pathology (ISUP) grade group (GG) 1, 2, and 3 or higher cancer in each core. Lesions are stratified according to number of cores sampled. Values represent number of lesions with given histologic result for each core number.
Among the 163 lesions sampled by at least two cores, core 2 yielded 21 (12.9%) upgrades relative to core 1 (Table 3, Fig. 4): 12 (7.4%) upgrades from no cancer or clinically insignificant cancer to csPCa; nine (5.5%) upgrades from no cancer to any cancer; three (1.8%) upgrades from no cancer to clinically insignificant cancer; six (3.7%) upgrades from no cancer to csPCa; six (3.7%) upgrades from clinically insignificant cancer to csPCa; and four (2.5%) upgrades from no or low-grade cancer to high-grade cancer (ISUP GG ≥ 3). Among the 121 lesions with at least three cores, core 3 yielded 13 (10.7%) upgrades relative to the first two cores (Table 3, Fig. 4): five (4.1%) upgrades from no cancer or clinically insignificant cancer to csPCa; two (1.7%) upgrades from no cancer to any cancer; two (1.7%) upgrades from no cancer to clinically insignificant cancer; zero (0%) upgrades from no cancer to csPCa; five (4.1%) upgrades from clinically insignificant cancer to csPCa; and five (4.1%) upgrades from no or low-grade cancer to high-grade cancer. Among the 52 lesions with four cores, core 4 yielded one (1.9%) upgrade relative to the first 3 cores (Table 3, Fig. 4); this was a PI-RADS category 5 lesion that was upgraded from ISUP GG 2 on the first three cores to ISUP GG 3 on core 4.
TABLE 3:
Pathology Upgrades for Each Core Number
| Upgrade, Final Grade | Upgrades on Core 2 Relative to Core 1 (n = 163 Lesions With ≥ 2 Cores) | Upgrades on Core 3 Relative to First 2 Cores (n = 121 Lesions With ≥ 3 Cores) | Upgrades on Core 4 Relative to First 3 Cores (n = 52 Lesions With ≥ 4 Cores) | |||
|---|---|---|---|---|---|---|
|
| ||||||
| n | % (95% CI) | n | % (95% CI) | n | % (95% CI) | |
| ISUP GG 0 to 1 (no cancer to clinically insignificant cancer) | 3 | 1.8 (0.5–5.7) | 2 | 1.7 (0.3–6.4) | 0 | — |
| Final grade | ||||||
| GG 1 | 3 | 2 | — | |||
| ISUP GG 0 to ≥ 1 (no cancer to any cancer) | 9 | 5.5 (2.7–10.5) | 2 | 1.7 (0.3–6.4) | 0 | — |
| Final grade | ||||||
| GG 1 | 3 | 2 | — | |||
| GG 2 | 6 | — | ||||
| ISUP GG 0 to ≥ 2 (no cancer to clinically significant cancer) | 3.7 (1.5–8.2) | 0 | — | 0 | — | |
| Final grade | ||||||
| GG 2 | 6 | — | — | |||
| ISUP GG 1 to ≥ 2 (clinically insignificant cancer to clinically significant cancer) | 6 | 3.7 (1.5–8.2) | 5 | 4.1 (1.5–9.9) | 0 | — |
| Final grade | ||||||
| GG 2 | 6 | 5 | — | |||
| ISUP GG ≤ 1 to ≥ 2 (no or clinically insignificant cancer to clinically significant cancer) | 12 | 7.4 (4–12.8) | 5 | 4.1 (1.5–9.9) | 0 | — |
| Final grade | ||||||
| GG 2 | 12 | 5 | — | |||
| ISUP GG ≤ 2 to ≥ 3 (clinically significant cancer or lower to high-grade cancer) | 4 | 2.5 (0.8–6.6) | 5 | 4.1 (1.5–9.9) | 1 | 1.9 (0.1–11.6) |
| Final grade | ||||||
| GG 3 | 2 | 5 | 1 | |||
| GG 4 | 2 | 0 | 0 | |||
| Any | 21a | 12.9 (8.3–19.2) | 13b | 10.7 (6.1–18) | 1 | 1.9 (0.1–11.6) |
| Final grade | ||||||
| GG 1 | 3 | 2 | 0 | |||
| GG 2 | 12 | 5 | 0 | |||
| GG 3 | 2 | 5 | 1 | |||
| GG 4 | 3 | 1 | 0 | |||
| GG 5 | 1 | 0 | 0 | |||
Note—Dash (—) indicates not applicable. ISUP = International Society of Urological Pathology, GG = grade group.
Includes one upgrade from ISUP GG 3 to 4 and one upgrade from ISUP GG 4 to 5.
Includes one upgrade from ISUP GG 3 to 4.
Fig. 4—
Bar graph shows number of lesions with and without histologic upgrade on second, third, and fourth cores.
The lower frequency of any upgrade on core 3 (10.7%) relative to core 2 (12.9%) was not significant (p = .29). However, the lower frequency of any upgrade on core 4 (1.9%) relative to core 3 was significant (p = .03). The lower frequency of upgrade to csPCa on core 3 (4.1%) relative to core 2 (7.4%) was not significant (p = .13), nor was the lower frequency of upgrade to csPCa on core 4 (0%) relative to core 3 (p = .07). The frequency of upgrade increased with increasing PI-RADS category on core 2 (upgrade in 6.9%, 10.3%, and 23.4% for PI-RADS categories 3, 4, and 5, respectively; p = .007) and on core 3 (upgrade in 2.9%, 9.1%, and 19.0% for PI-RADS categories 3, 4, and 5, respectively; p = .01) but was not associated with increasing PI-RADS category on core 4 (p = .13) (Table 4).
TABLE 4:
Frequency of Any Upgrade Stratified by Core Number and PI-RADS v2 Score
| PI-RADS v2 Category | Core 2 Relative to Core 1 (n = 163 Lesions With ≥ 2 Cores)a | Core 3 Relative to First 2 Cores (n = 121 Lesions With ≥ 3 Cores)b | Core 4 Relative to First 3 Cores (n = 52 Lesions With ≥ 4 Cores)c | |||
|---|---|---|---|---|---|---|
|
| ||||||
| Upgrade | 95% CI | Upgrade | 95% CI | Upgrade | 95% CI | |
| 3 | 4/58 (6.9) | 2.2–17.5 | 1/35 (2.9) | 0.1–16.6 | 0/12 (0) | 0–30.1 |
| 4 | 6/58 (10.3) | 4.3–21.8 | 4/44 (9.1) | 3.0–22.6 | 0/20 (0) | 0–20.0 |
| 5 | 11/47 (23.4) | 12.8–38.4 | 8/42 (19.0) | 9.1–34.6 | 1/20 (5.0) | 0.3–26.9 |
Note—Values are the number of lesions with percentages in parentheses. v2 = version 2.
p = .007.
p = .01.
p = .13.
Table S2, which can be viewed in the AJR electronic supplement to this article, available at doi.org/10.2214/AJR.20.24918, compares the diameter and location of lesions upgraded and not upgraded on cores 2 and 3. Lesion diameter was not significantly associated with any upgrade on core 2 (p = .15) or on core 3 (p = .49). A total of 81.0% of lesions upgraded on core 2 were anterior compared with 32.4% of lesions not upgraded on core 2 (p < .001). However, upgrade on core 3 was not associated with anterior versus posterior location (p > .99). Upgrade on core 3 was associated (p = .01) with lesion level (e.g., upgrade on core 3 for 38.5% of lesions at the apex; 23.1% of lesions at the apex and mid prostate; and 15.4% of lesions at the apex, mid prostate, and base vs 27.1%, 11.2%, and 0.9%, respectively, of lesions not upgraded on core 3). Upgrade on core 2 was associated with transition zone location (OR, 3.15; 95% CI, 1.10–9.08) and anterior location (OR, 9.18; 95% CI, 2.92–28.87) in univariable regression analysis and with only anterior location (OR, 7.35; 95% CI, 2.15–25.16) in multivariable regression analysis (Table S3, which can be viewed in the AJR electronic supplement to this article, available at doi.org/10.2214/AJR.20.24918). Upgrade on core 3 was only associated with apical location in univariable (OR, 5.42; 95% CI, 1.41–20.87) and multivariable regression analyses (OR, 6.33; 95% CI, 1.42–28.26) (Table S3).
Second Lesion Analysis
Table 5 summarizes the characteristics of the first and second biopsied lesions in the 40 patients who underwent MRGB of at least two lesions. A total of 13 (32.5%), 17 (42.5%), and 10 (25.0%) of the first lesions compared with 25 (62.5%), 14 (35.0%), and one (2.5%) of the second lesions were classified as PI-RADS categories 3, 4, and 5, respectively (p = .003). The median number of cores obtained was higher for the first than for the second lesion (3.0 vs 2.0 cores, respectively; p < .001). The two groups were not statistically different in terms of the other characteristics assessed (all p > .05), including lesion diameter (median, 1.1 cm [IQR, 0.9–1.5 cm] for the first lesion vs 0.9 cm [IQR, 0.8–1.3 cm] for the second lesion; p = .30).
TABLE 5:
Lesion Characteristics Stratified by Lesion Biopsy Order
| Characteristic | First Lesion (n = 40) | Second Lesion (n = 40) | p |
|---|---|---|---|
| Prostate volume (g) | 70.2 (52.1–91.2) | 70.2 (52.1–91.2) | > .99 |
| Lesion diameter (cm) | 1.1 (0.9–1.5) | 0.9 (0.8–1.3) | .30 |
| Anterior | 15 (37.5) | 9 (22.5) | .22 |
| Zone | .28 | ||
| Both TZ and PZ | 2 (5.0) | 0 (0) | |
| PZ | 21 (52.5) | 19 (47.5) | |
| TZ | 17 (42.5) | 21 (52.5) | |
| Prostate level | .86 | ||
| Apex | 9 (22.5) | 12 (30.0) | |
| Apex, mid prostate | 2 (5.0) | 2 (5.0) | |
| Apex, mid prostate, base | 0 (0) | 0 (0) | |
| Mid prostate | 16 (40.0) | 17 (42.5) | |
| Base, mid prostate | 2 (5.0) | 1 (2.5) | |
| Base | 11 (27.5) | 8 (20.0) | |
| PI-RADS v2 score | .003 | ||
| 3 | 13 (32.5) | 25 (62.5) | |
| 4 | 17 (42.5) | 14 (35.0) | |
| 5 | 10 (25.0) | 1 (2.5) | |
| No. of cores per lesion | 3.0 (2.0–3.0) | 2.0 (2.0–3.0) | < .001 |
Note—Values are the median with the interquartile range in parentheses or the number of patients with percentage in parentheses. TZ = transition zone, PZ = peripheral zone, v2 = version 2.
Upgrade on the second lesion occurred in one of the 40 patients (2.5%). In this patient, both lesions were classified as PIRADS category 4 and showed csPCa. The upgrade was from ISUP GG 2 for the first lesion to ISUP GG 3 for the second lesion. No upgrade (0/4) was identified in any third targeted lesion.
Discussion
This study evaluated the diagnostic yield of obtaining multiple cores per lesion and of sampling a second lesion after the most suspicious lesion for in-gantry MRGB. Cores 2 and 3 of a lesion resulted in an upgrade to csPCa relative to the prior cores in 7.4% and 4.1% of lesions, respectively. In comparison, cores 2 and 3 resulted in an upgrade from no cancer to clinically insignificant cancer in 1.8% and 1.7% of lesions, respectively. No lesion was upgraded to any cancer or to csPCa on core 4. Biopsy of a second lesion resulted in only a single upgrade. On the basis of these findings, three cores of the primary lesion are sufficient for accurate diagnosis, whereas a fourth core provides little additional utility. Targeting a second lesion also adds little diagnostic value if whole-gland therapy is planned.
A higher frequency of histologic upgrade was associated with anterior lesion location and increasing PI-RADS v2 category. Specifically, anterior lesions showed a significantly higher frequency of upgrade on the second (but not the third) core. Anterior prostate cancers are difficult to detect by systematic biopsy and are recognized as more effectively sampled by targeted MRGB [34]. Nonetheless, the observed higher frequency of upgrade for anterior lesions indicates difficulties in sampling anterior lesions even with in-gantry MRGB. Reaching anterior lesions may be challenging because of the need to avoid overshooting the lesion across the capsule and the longer lengths of tissue to traverse to reach the lesion. In addition, lesions with higher PI-RADS v2 categories showed a significantly higher frequency of upgrade on both the second and third cores. The reasons for this observation are unclear, but it indicates a particular importance of obtaining three cores for PI-RADS v2 category 4 and 5 lesions.
Prior studies examined the diagnostic value of increasing the number of cores obtained during MRGB [8, 17, 21–24, 26, 27]. A study from 2016 found an increase in csPCa diagnosis with in-gantry MRGB when increasing from one to two cores [27]; however, the study did not assess the impact on prostate cancer diagnosis when increasing from two to three cores. The authors found a frequency of upgrade of 11% on the second core, similar to the frequency of any upgrade of 12.9% in the current study [27]. However, in the earlier study, the determination of any upgrade included upgrades that do not carry clinical significance, such as from Gleason score 4 + 5 to 5 + 4. In addition, the 2016 study reported an upgrade frequency of only 2.3% from no cancer or clinically insignificant cancer to csPCa with the addition of a second core, which is lower than our reported frequency of 7.4% [27]. The authors interpreted the 2.3% upgrade frequency as minor and suggested that one core per lesion is acceptable [27]. The discrepancy in upgrade frequency between the earlier and cur rent study is likely due to a difference in threshold for csPCa (Gleason score 4 + 3 in the previous study vs Gleason score 3 + 4 in the cur rent study). A separate prospective study from 2021 sought to deter mine the optimal number of cores to obtain with in-gantry MRGB [26]. Five cores were obtained per lesion, and rates of csPCa (ISUP GG ≥ 2) were evaluated for the first, third, and fifth cores. The frequency of csPCa increased from 26% to 44% for the first to the third core and from 44% to 52% from the third to the fifth core. However, that study only included PI-RADS category 4 or 5 lesions and excluded PI-RADS category 3 lesions in the reported results [26]. PI-RADS category 3 le sions were reported separately and had much lower frequencies of upgrading with sequential cores, similar to the findings in the cur rent study. The results of the 2021 study suggest that up to five cores may be necessary. The current study suggests that fewer cores are needed than previously supported, and up to three cores should be obtained during in-gantry MRGB.
Other studies have examined the utility of additional biopsy cores during fusion and cognitive MRGB. A 2019 study examined the impact of increasing cores in fusion MRGB and reported that almost 25% of csPCa diagnoses were missed with only two cores per lesion compared with five cores per lesion, further illustrating the inadequacy of a two-core approach [23]. Another study from 2019 reported an increase in csPCa diagnoses with cognitive MRGB when the number of cores per lesion was increased from one to three and from three to five [21]. The authors also found an increase in clinical significance with increasing number of cores, consistent with the findings for in-gantry MRGB in the current study. However, few studies have directly evaluated fusion versus in-gantry MRGB, and caution is warranted when comparing the two methods [1]. In particular, more cores overall may be needed for both fusion and cognitive MRGB than for in-gantry MRGB because of the potential for registration error and operator error, respectively [1], thereby potentially explaining why those two earlier studies found diagnostic benefit for up to five cores for fusion and cognitive MRGB.
The increase in ISUP GG when the number of cores was increased from one to two and from two to three in the current study could impact the treatment of patients with prostate cancer by causing a shift from active surveillance to more aggressive treatment protocols [35]. According to National Comprehensive Cancer Network guidelines, the recommended management of prostate cancer differs substantially between ISUP GG 1 and ISUP GG 2 or 3 [35]. We observed that core 3 had an upgrade frequency of 4.1% from ISUP GG 1 or lower to 2 or higher and from ISUP GG 2 or lower to 3 or higher. These upgrades increase the likelihood of a patient being recommended for invasive treatment, such as radical prostatectomy, rather than active surveillance [35]. The single upgrade that we observed on core 4 was from ISUP GG 2 or lower to 3 or higher and thus also has implications for clinical management [35].
To our knowledge, previous studies have not examined the utility of sampling second lesions during in-gantry MRGB. The single upgrade in a second lesion in our study occurred in a patient with two lesions with identical PI-RADS v2 categories. An ISUP 2019 consensus conference reported 67% consensus that a global Gleason score is sufficient to direct patient treatment in the setting of multifocal tumors [36]. Thus, a decision to refrain from sampling second lesions may be considered given the very low rate of upgrade from second lesions and to avoid prolonging the MRGB procedure time if whole-gland therapy is planned. On the other hand, sampling a second lesion may be advised when focal therapy is planned. Focal therapy is an emerging technique used to treat localized, low-volume, low-to-intermediate risk prostate cancer, most commonly with focal cryotherapy or high-intensity focused ultrasound [37], and is most commonly used to treat only the most suspicious lesion [37].
Limitations of our study include its retrospective design, the lack of prostatectomy results, and the lack of follow-up for patients with negative biopsy results. Furthermore, the sample size is small, particularly for the analysis of second lesions. The primary analyses of the incremental value of sequential cores per lesion pooled such cores from first, second, and third lesions. Because of the retrospective design, a standard approach was not used to determine the number of cores obtained per lesion, and radiologists obtained third and fourth cores at their discretion. Of note, PI-RADS categories were not used in determining the number of cores obtained per lesion. The biopsies were performed by operators with varying levels of experience, which could have contributed to the variation in the number of cores per lesion. Further, 116 of the 131 biopsies were performed by a single operator; future studies with a greater number of operators should be performed to confirm our present observations. The biopsy cores were evaluated by more than one pathologist as part of routine clinical care, and repeat pathology review was not performed. The study sample was heterogeneous, with only 7.0% of patients undergoing their first prostate biopsy, thereby limiting generalizability to such patients. Although MRGB is currently recommended after a negative prostate biopsy [26], interest in implementing MRGB in biopsy-naïve patients is growing [38].
In conclusion, during in-gantry MRGB of the prostate, obtaining three cores from the primary lesion is warranted to optimize csPCa diagnosis. Obtaining a fourth core from the primary lesion or sampling a second lesion had very low yield in upgrading cancer diagnoses in our study, and operators may refrain from these options to decrease procedure times. Future prospective studies with larger sample sizes and a multicenter design should be performed to confirm these findings.
Supplementary Material
HIGHLIGHTS.
Key Finding
In 128 men undergoing in-gantry MRGB of 163 prostate lesions, frequency of upgrade from ISUP GG ≤ 1 to ≥ 2 was 7.4% on core 2, 4.1% on core 3, and 0% on core 4 of a given lesion. Sampling a second lesion resulted in an upgrade in only one patient.
Importance
Obtaining only three cores from the primary lesion during in-gantry MRGB may be sufficient for high diagnostic yield while reducing patient discomfort and procedure times.
Acknowledgments
Supported by NIH grant R01CA208236–04 and Early Postdoc.Mobility fellowship grant P2SKP3_178132 from the Swiss National Science Foundation (V.C.O.).
V. C. Obmann received Early Postdoc. Mobility fellowship grant P2SKP3_178132 from the Swiss National Science Foundation. The remaining authors declare that they have no disclosures relevant to the subject matter of this article.
Footnotes
The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Based on a presentation at the International Society for Magnetic Resonance in Medicine and Society for MR Radiographers and Technologists 2020 virtual annual meeting.
An electronic supplement is available online at doi.org/10.2214/AJR.20.24918.
References
- 1.Verma S, Choyke PL, Eberhardt SC, et al. The current state of MR imaging-targeted biopsy techniques for detection of prostate cancer. Radiology 2017; 285:343–356 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Brawley OW. Prostate cancer epidemiology in the United States. World J Urol 2012; 30:195–200 [DOI] [PubMed] [Google Scholar]
- 3.American Cancer Society website. Cancer facts & figures 2018. www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-figures-2018.html. Accessed July 2019 [Google Scholar]
- 4.Litwin MS, Tan HJ. The diagnosis and treatment of prostate cancer: a review. JAMA 2017; 317:2532–2542 [DOI] [PubMed] [Google Scholar]
- 5.Uriburu-Pizarro F, Kasivisvanathan V, Puech P, Villers A. Pre-biopsy MRI as an adjunct for cancer detection in men with elevated PSA and no previous biopsy. Transl Androl Urol 2017; 6:387–394 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kasivisvanathan V, Rannikko AS, Borghi M, et al. ; PRECISION Study Group Collaborators. MRI-targeted or standard biopsy for prostate-cancer diagnosis. N Engl J Med 2018; 378:1767–1777 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Drost FH, Osses DF, Nieboer D, et al. Prostate MRI, with or without MRI-targeted biopsy, and systematic biopsy for detecting prostate cancer. Cochrane Database Syst Rev 2019; 4:CD012663 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Porpiglia F, De Luca S, Passera R, et al. Multiparametric magnetic resonance/ultrasound fusion prostate biopsy: number and spatial distribution of cores for better index tumor detection and characterization. J Urol 2017; 198:58–64 [DOI] [PubMed] [Google Scholar]
- 9.Elkhoury FF, Simopoulos DN, Marks LS. Targeted prostate biopsy in the era of active surveillance. Urology 2018; 112:12–19 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ahmed HU, El-Shater Bosaily A, Brown LC, et al. ; PROMIS study group. Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study. Lancet 2017; 389:815–822 [DOI] [PubMed] [Google Scholar]
- 11.Murphy G, Haider M, Ghai S, Sreeharsha B. The expanding role of MRI in prostate cancer. AJR 2013; 201:1229–1238 [DOI] [PubMed] [Google Scholar]
- 12.Hosseiny M, Shakeri S, Felker ER, et al. 3-T multiparametric MRI followed by in-bore MR-guided biopsy for detecting clinically significant prostate cancer after prior negative transrectal ultrasound-guided biopsy. AJR 2020; 215:660–666 [DOI] [PubMed] [Google Scholar]
- 13.Padhani AR, Barentsz J, Villeirs G, et al. PI-RADS Steering Committee: the PI-RADS multiparametric MRI and MRI-directed biopsy pathway. Radiology 2019; 292:464–474 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Aminsharifi A, Gupta RT, Tsivian E, Sekar S, Sze C, Polascik TJ. Reduced core targeted (RCT) biopsy: combining multiparametric magnetic resonance imaging–transrectal ultrasound fusion targeted biopsy with laterally-directed sextant biopsies—an alternative template for prostate fusion biopsy. Eur J Radiol 2019; 110:7–13 [DOI] [PubMed] [Google Scholar]
- 15.van Luijtelaar A, Bomers J, Fütterer J. A comparison of magnetic resonance imaging techniques used to secure biopsies in prostate cancer patients. Expert Rev Anticancer Ther 2019; 19:705–716 [DOI] [PubMed] [Google Scholar]
- 16.Wegelin O, van Melick HHE, Hooft L, et al. Comparing three different techniques for magnetic resonance imaging-targeted prostate biopsies: a systematic review of in-bore versus magnetic resonance imaging-transrectal ultrasound fusion versus cognitive registration—is there a preferred technique? Eur Urol 2017; 71:517–531 [DOI] [PubMed] [Google Scholar]
- 17.Grummet J How to biopsy: transperineal versus transrectal, saturation versus targeted, what’s the evidence? Urol Clin North Am 2017; 44:525–534 [DOI] [PubMed] [Google Scholar]
- 18.Penzkofer T, Tuncali K, Fedorov A, et al. Transperineal in-bore 3-T MR imaging-guided prostate biopsy: a prospective clinical observational study. Radiology 2015; 274:170–180 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Grummet J, Pepdjonovic L, Huang S, Anderson E, Hadaschik B. Transperineal vs. transrectal biopsy in MRI targeting. Transl Androl Urol 2017; 6:368–375 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Rosenkrantz AB, Verma S, Choyke P, et al. Prostate magnetic resonance imaging and magnetic resonance imaging targeted biopsy in patients with a prior negative biopsy: a consensus statement by AUA and SAR. J Urol 2016; 196:1613–1618 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zhang M, Milot L, Khalvati F, et al. Value of increasing biopsy cores per target with cognitive MRI-targeted transrectal US prostate biopsy. Radiology 2019; 291:83–89 [DOI] [PubMed] [Google Scholar]
- 22.Leyh-Bannurah SR, Kachanov M, Beyersdorff D, et al. Minimum magnetic resonance imaging-ultrasound fusion targeted biopsy cores needed for prostate cancer detection: multivariable retrospective, lesion based analyses of patients treated with radical prostatectomy. J Urol 2020; 203:299–303 [DOI] [PubMed] [Google Scholar]
- 23.Lu AJ, Syed JS, Ghabili K, et al. Role of core number and location in targeted magnetic resonance imaging-ultrasound fusion prostate biopsy. Eur Urol 2019; 76:14–17 [DOI] [PubMed] [Google Scholar]
- 24.Hansen NL, Barrett T, Lloyd T, et al. Optimising the number of cores for magnetic resonance imaging-guided targeted and systematic transperineal prostate biopsy. BJU Int 2020; 125:260–269 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Tu X, Lin T, Cai D, Liu Z, Yang L, Wei Q. The optimal core number and site for MRI-targeted biopsy of prostate? A systematic review and pooled analysis. Minerva Urol Nefrol 2020; 72:144–151 [DOI] [PubMed] [Google Scholar]
- 26.Tracy CR, Flynn KJ, Sjoberg DD, Gellhaus PT, Metz CM, Ehdaie B. Optimizing MRI-targeted prostate biopsy: the diagnostic benefit of additional targeted biopsy cores. Urol Oncol 2021; 39:193.e1–193.e6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Schimmöller L, Quentin M, Blondin D, et al. Targeted MRI-guided prostate biopsy: are two biopsy cores per MRI-lesion required? Eur Radiol 2016; 26:3858–3864 [DOI] [PubMed] [Google Scholar]
- 28.Rouvière O, Schoots IG, Mottet N; EAU-EANM-ESTRO-ESUR-SIOG Prostate Cancer Guidelines Panel. Multiparametric magnetic resonance imaging before prostate biopsy: a chain is only as strong as its weakest link. Eur Urol 2019; 75:889–890 [DOI] [PubMed] [Google Scholar]
- 29.D’Agostino D, Romagnoli D, Giampaoli M, et al. “In-bore” MRI-guided prostate biopsy for prostate cancer diagnosis: results from 140 consecutive patients. Curr Urol 2020; 14:22–31 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Panda A, O’Connor G, Lo WC, et al. Targeted biopsy validation of peripheral zone prostate cancer characterization with magnetic resonance fingerprinting and diffusion mapping. Invest Radiol 2019; 54:485–493 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Panda A, Obmann VC, Lo WC, et al. MR fingerprinting and ADC mapping for characterization of lesions in the transition zone of the prostate gland. Radiology 2019; 292:685–694 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Weinreb JC, Barentsz JO, Choyke PL, et al. PI-RADS Prostate Imaging–Reporting and Data System: 2015, version 2. Eur Urol 2016; 69:16–40 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Epstein JI, Egevad L, Amin MB, Delahunt B, Srigley JR, Humphrey PA; Grading Committee. The 2014 International Society of Urological Pathology (ISUP) consensus conference on Gleason grading of prostatic carcinoma: definition of grading patterns and proposal for a new grading system. Am J Surg Pathol 2016; 40:244–252 [DOI] [PubMed] [Google Scholar]
- 34.Ouzzane A, Puech P, Lemaitre L, et al. Combined multiparametric MRI and targeted biopsies improve anterior prostate cancer detection, staging, and grading. Urology 2011; 78:1356–1362 [DOI] [PubMed] [Google Scholar]
- 35.Mohler JL, Kantoff PW, Armstrong AJ, et al. Prostate cancer, version 2.2014. J Natl Compr Canc Netw 2014; 12:686–718 [DOI] [PubMed] [Google Scholar]
- 36.van Leenders GJLH, van der Kwast TH, Iczkowski KA. The 2019 International Society of Urological Pathology consensus conference on prostate cancer grading. Eur Urol 2020; 44:87–99 [DOI] [PubMed] [Google Scholar]
- 37.Marshall S, Taneja S. Focal therapy for prostate cancer: the current status. Prostate Int 2015; 3:35–41 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Schoots IG, Padhani AR, Rouvière O, Barentsz JO, Richenberg J. Analysis of magnetic resonance imaging–directed biopsy strategies for changing the paradigm of prostate cancer diagnosis. Eur Urol Oncol 2020; 3:32–41 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.