DeepDyve requires Javascript to function. Please enable Javascript on your browser to continue.
Multiparametric magnetic resonance imaging ultrasound-guided fusion biopsy during active surveillance: A single-centre study Multiparametric magnetic resonance imaging ultrasound-guided fusion biopsy during active...
Röthlin, Kilian; Zamboni, Stefania; Moschini, Marco; Stucki, Patrick; Afferi, Luca; Baumeister, Philipp; Mattei, Agostino
2020-07-02 00:00:00
ARAB JOURNAL OF UROLOGY 2020, VOL. 18, NO. 3, 142–147 https://doi.org/10.1080/2090598X.2020.1749477 ONCOLOGY/ RECONSTRUCTION: ORIGINAL ARTICLE Multiparametric magnetic resonance imaging ultrasound-guided fusion biopsy during active surveillance: A single-centre study Kilian Röthlin, Stefania Zamboni, Marco Moschini , Patrick Stucki, Luca Afferi , Philipp Baumeister and Agostino Mattei Department of Urology, Luzerner Kantonsspital, Lucerne, Switzerland ABSTRACT ARTICLE HISTORY Received 30 December 2019 Objective: To analyse the role of multiparametric magnetic resonance imaging (mpMRI) Accepted 14 March 2020 ultrasound (US)-guided fusion biopsy (FB) in patients with low-risk prostate cancer (PCa) under active surveillance (AS). KEYWORDS Patients and methods: Our retrospective study included 47 patients under AS who consecu- Prostate cancer; tively underwent both FB and standard biopsy (SB), from May 2015 until November 2017. We multiparametric magnetic defined FB as a transrectal US-guided biopsy based on mpMRI. The primary endpoint was to resonance imaging/ assess the rate of concordance between FB and SB in terms of diagnostic yield, as well as the ultrasound fusion biopsy; 12- core template (standard) rate of Gleason Score upgrading/downgrading between the two techniques. Cohen’s kappa biopsy; active surveillance coefficient (κ) was applied to test the concordance between FB and SB. Results: The median (interquartile range [IQR]) follow-up was 20 (13–37) months. The median (IQR) number of cores taken was 13 (12–14) at SB and 4 (4–6) at FB. Overall, FB missed 12/47 (26%) PCa diagnoses compared to SB. There was concordance between SB and FB in 64% of the patients. The κ showed a perfect agreement between SB and FB for the detection of PCa with Gleason Score 4 + 4 and a weak concordance for negative biopsies (κ: 0.46) and for PCa with a Gleason Score 4 + 3 (κ: 0.54). There was Gleason Score upgrading at FB in two of 47 (4%) patients, whereas there was downgrading in three of 47 (6%) patients. Conclusion: In our present study, FB showed no superiority over SB for the detection of PCa. On the contrary, FB had a high rate of missed PCa compared to SB. Further studies are required to ascertain the role of FB in AS. Abbreviations: AS: active surveillance; FB: fusion biopsy; IL: index lesion; IQR: interquartile range; mpMRI: multiparametric MRI; (cs)PCa: (clinically significant) prostate cancer; PI-RADS: Prostate Imaging-Reporting and Data System; PRIAS: Prostate Cancer Research International Active Surveillance; ROI: region of interest; SB: standard biopsy Introduction and the superiority of mpMRI-targeted biopsy over the 12-core TRUS-guided SB in diagnosing PCa in Prostate cancer (PCa) is the most commonly diagnosed biopsy naïve men at clinical risk of PCa. Several retro- cancer in USA men, with an estimated 164 690 new spective series support the utility of mpMRI-based tar- cases in 2018 [1]. With the introduction of PSA screen- geted-fusion biopsies (FBs) in detecting clinically ing programmes, PCa incidence increased and this led significant PCa (csPCa) [12–16]. Moreover, a recent to an important debate about over-diagnosis and con- Cochrane Review [17] analysed the role of mpMRI in sequent potential over-treatment of PCa. Active sur- the repeat-biopsy setting, with a pooled sensitivity of veillance (AS) has been proven to be a safe and 0.91 (95% CI 0.83–0.95) and a pooled specificity of 0.37 effective strategy [2–4] for patients with low-risk PCa (95% CI 0.29–0.65) for International Society of [5,6]. Although 12-core TRUS-guided standard biopsy Urological Pathology (ISUP) Grade ≥2 PCas, although (SB) currently remains the ‘gold standard’ for diagnos- its role in AS is still under debate. For these reasons, the ing PCa [7], this technique samples ~1% of prostate aim of our present study was to analyse the impact and tissue. Consequently, the diagnostic accuracy of this role of FB in patients with low-risk PCa under AS. procedure is low and 10–40% of patients on AS have been upstaged by confirmatory targeted biopsies after initial TRUS-guided random SB [8–10]. Patients and methods The progress in multiparametric MRI (mpMRI) allows A total of 345 consecutive patients underwent both high-quality images of the prostate and improved PCa mpMRI/TRUS FB and SB in the same session, from identification. The recently released PRECISION trial May 2015 to November 2017, in a single tertiary care [11] supported the utility of mpMRI before biopsy CONTACT Stefania Zamboni Stefania.zamboni@libero.it Department of Urology, Luzerner Kantonsspital, Lucerne, Switzerland © 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ARAB JOURNAL OF UROLOGY 143 referral centre. A total of 47 patients were identified in AS was the confirmatory biopsy, with those follow- with low-risk PCa who underwent AS, and thus were ing considered as repeat biopsies. included in the study. All the biopsies were taken by a single experienced urologist. Unless otherwise indi- Outcomes of interest cated by the patient, biopsies were taken under loco- regional anaesthesia. Theprimary endpoint wastoassessthe rate of We defined SB as all TRUS-guided biopsies taken concordance between FB and SB in terms of diag- with a standardised sequence, i.e., six prostatic cores nostic yield, as well as the rate of Gleason Score taken from each side of the prostate, as previously upgrading/downgrading between the two techni- described [18]. We defined FB as every TRUS-guided ques. Moreover, we assessed the relationship biopsy taken based on previously mpMRI-defined pro- between the PI-RADS score of the IL and the static lesions, i.e., regions of interest (ROIs) at mpMRI of Gleason Score of the IL at FB, and we looked for the prostate. FB consisted of at least one biopsy taken predictors of missed PCa at FB. We used the results from every ROI. All the FBs were performed with the of theSBasameasuretodefine missed or not Artemis/Profuse® (Eigen, Grass Valley, CA, USA) plat- missed PCa diagnosis at FB and vice versa. form. The biopsy sequence consisted of SB followed by FB for all the patients included in this study. The reason for this is that we tried to avoid the operator Statistical analysis from being influenced to using the same biopsy track Descriptive statistics of categorical variables focused of the FB when performing the SB, as this could have on frequencies and proportions. Means, medians, and allegedly altered the diagnostic yield of the SB [19]. interquartile ranges (IQRs) were reported for continu- The mpMRI was performed with a 3.0-T scanner ously coded variables. The Cohen’s kappa coefficient (Achieva dStream, Philips Medical Systems, Best, the (κ) was applied to evaluate the concordance between Netherlands). Most of the mpMRIs were done at our the results of SB and FB. A univariable logistic regres- institution and were interpreted by a dedicated geni- sion model was used to assess the relationship tourinary radiologist, who was previously trained in the between independent variables and upstaging at FB reading of mpMRI and who had 2 years’ experience in compared to SB. Statistical significance was considered the assessment of the Prostate Imaging Reporting and at P < 0.05. Statistical analyses were performed using Data System (PI-RADS) score at study commencement. Stata 14 (Stata Corp., College Station, TX, USA). The mpMRIs performed in other centres were reviewed at our institution. The PI-RADS score [20] was used for grading the ROIs on the mpMRI. The index lesion (IL) Results was defined as the ROI with the highest PI-RADS score. If there was an equal PI-RADS score, the ROI with the Baseline characteristics larger diameter was defined as the IL. All the histo- Baseline characteristics of the 47 patients enrolled in pathological analyses were done at our institution and AS are reported in Table 1. The median (IQR) follow-up interpreted by dedicated genitourinary histopatholo- from the first diagnosis of PCa until the last study gists according to the Swiss Society for Pathology biopsy was 20 (13–37) months. At the time of confir- guidelines [21]. matory biopsy, the median (IQR) patient age was 64 (60–68) years, the median (IQR) PSA level was 5.67 AS (3.90–7.73) ng/mL, and the median (IQR) prostate volume was 50 (33–58) mL. Overall, six patients (13%) Of the 47 patients included in our study, 38 presented at the mpMRI harboured an IL with a PI-RADS score 1, with criteria for inclusion in AS according to the 18 (38%) an IL with a PI-RADS score 2, 12 (26%) an IL Prostate Cancer Research International Active with a PI-RADS score 3, nine (19%) an IL with a PI-RADS Surveillance (PRIAS) study: clinical stage T1/T2 PCa, score 4, and two (4.2%) an IL with a PI-RADS score 5. PSA level ≤10 ng/mL, PSA density <0.2 ng/mL/mL, The median (IQR) IL diameter was 12 (9–17) mm. one or two positive biopsy cores, and Gleason Score ≤6[22]. The remaining nine had low-risk PCa defined as Gleason Score ≤6 and clinical stage T1/T2, but did Biopsy results not strictly adhere to all the PRIAS inclusion criteria. Further we defined csPCa as Gleason Score ≥3+ 4. Biopsy results are reported in Table 2.The median Monitoring consisted of PSA measurements every (IQR) number of biopsy cores taken was 13 (12–14) 3months during the first year after biopsy and at SB and 4 (4–6) at FB. The FB detected PCa in every 6 months in the second year. Re-biopsies dur- 15 (32%) patients, whereas the SB detected PCa in ing AS were done annually, if not wished otherwise 26 (55%) patients. At FB the IL was positive for PCa in by the patients. The first biopsy after the enrolment 10 (21%) patients. 144 K. RÖTHLIN ET AL. Table 1. Baseline characteristics of the 47 AS patients. Relationship between the IL PI-RADS score and the Variable Value Gleason Score of the IL at FB Age, years Mean 63.5 The relationship between the IL PI-RADS score and Median (IQR) 64 (60–68) Gleason Score is reported in Figure 1. Only one patient PSA, ng/mL was diagnosed with an IL PI-RADS score 1; he had Mean 6.17 Median (IQR) 5.67 (3.9–7.73) a Gleason Score 3 + 3 at FB. The four patients with an PSA density, ng/mL/mL IL PI-RADS score 2 and the one with an IL PI-RADS score Mean 0.14 Median (IQR) 0.13 (0.10–0.14) 5 had negative FBs, whereas two of the 18 patients Prostate volume, mL with an IL PI-RADS score 3 had a Gleason Score 3 + 3; Mean 47 Median (IQR) 50 (33–58) the remaining 16 had negative FBs. Of the 15 patients Number of previous biopsies with an IL PI-RADS score 4, one had a Gleason Score Mean 2 Median (IQR) 2(1–3) 4 + 4, two a Gleason Score 4 + 3, five a Gleason Score GS previous biopsies, n (%) 3 + 3, and seven negative FBs. Negative 14 (30) GS 3 + 3 33 (70) Suspicious DRE, n (%) 4 (14) Clinical T stage, n (%) Predictors of missed PCa at FB Negative 14 (30) cT1a 6 (13) In our present AS patient cohort, none of the factors cT1b 0 cT1 c 24 (51) tested at univariable analyses were predictors of cT2 3 (6.4) missed PCa at FB (all P > 0.05). Results are reported in mpMRI suspicious score, n (%) PI-RADS 1 6 (13) Table 4. PI-RADS 2 18 (38) PI-RADS 3 12 (26) PI-RADS 4 9 (19) PI-RADS 5 2 (4.2) Discussion Diameter index lesion, mm Mean 14 Although FB has been found to be superior to the 12- Median (IQR) 12 (9–17) Time between mpMRI and biopsy, days core TRUS-guided SB in diagnosing PCa in biopsy naïve Mean 63 men at risk of csPCa, sparse data exists regarding its Median (IQR) 42.5 (30–84) specific role in AS patients. For this reason, we sought GS, Gleason Score. to analyse the impact and role of FB in patients with low-risk PCa under AS. Our primary endpoint was to Table 2. The FB and SB results in the 47 AS patients. assess the concordance between FB and SB and the Variable SB FB rate of missed PCa diagnoses at FB. We found that FB Number of biopsy cores missed 26% of PCa diagnoses. Furthermore, FB missed Mean 13 5 Median (IQR) 13 (12–14) 4(4–6) five of 10 csPCa, whereas SB missed only one of them. Number of positive biopsy cores This result is consistent with that reported by Ma et al. Mean 1 1 Median (IQR) 1(0–2) 0(0–1) [23], who found a rate of missed csPCa for FB of 65% Number of patients diagnosed with 26 (55) 15 (32) and 19% for SB. Two other studies that analysed con- PCa, n (%) Extend of tumor involvement per firmatory biopsies in AS made similar observations biopsy core (%) concerning missed csPCa rates [24,25]. On the con- Mean 11 26 trary, multiple studies have reported the superiority Median (IQR) 5(2–15) 25 (4–40) Positive IL, n (%) – 10 (21) of FB over SB in detecting csPCa and reduced missing rates in the diagnostic setting [16–18]. However, in these investigations the whole cohort was divided Concordance between FB and SB, upgrading/ into two different study arms, specifically patients downgrading of Gleason Score at FB who underwent either SB + FB or FB + SB were eval- There was concordance between SB and FB in 30/47 uated by two different blinded operators. On the con- patients (64%). Cohen’s κ (Table 3) showed perfect trary, in our present study the operator was not agreement between SB and FB for the detection of blinded to the mpMRI results, as the SB was always PCa Gleason Score 4 + 4 and a weak concordance for taken before the FB it is possible that the surgeon took negative biopsies (κ: 0.46) and for PCa Gleason Score more biopsies during SB near the known ROIs. The 4+3(κ: 0.54). There was upgrading of the Gleason diverging results between our present study and Score at FB in two (4%) of the 47 patients, whereas those mentioned above could be related to the differ- there was downgrading in three (6%). FB and SB ence in the study designs. missed five of 10 and one of 10 patients with csPCa, In our present study, only the ILs with PI-RADS respectively. scores ≥4 were associated with the detection of PCa ARAB JOURNAL OF UROLOGY 145 Table 3. Number of cases graded at SB and at concurrent FB and Cohen’s κ assessing the concordance between SB and FB for the 47 AS patients. mpMRI/FB, n Negative GS 3 + 3 GS 3 + 4 GS 4 + 3 GS 4 + 4 GS 4 + 5 GS 5 + 4 Total SB, n Negative 20 1000 –– 21 κ 0.46 GS 3 + 3 10 6 010 –– 17 κ 0.24 GS 3 + 4 2 2 1 00 –– 5 κ 0.24 GS4+3 0 102 0 –– 3 κ 0.54 GS4+4 0 0001 –– 1 κ 1 GS 4 + 5 – ––– ––– – GS 5 + 4 – ––– ––– – Total 32 10 1 3 1 –– – GS, Gleason Score. Figure 1. Relationship between PI-RADS score and Gleason-Score (GS) of the (IL) in the 47 AS patients. Table 4. Univariable logistic regression analysis assessing the However, this finding cannot be generalised to all the predictors of missed PCa at FB. PI-RADS score ≤3 cases due to the few patients found Variable OR (95% CI) P with this type of lesion. Moreover, the necessity of Age, years 1.00 (0.89–1.12) 0.9 taking FB from PI-RADS score 3 lesions has not yet PSA level, ng/mL 1.00 (0.88–1.14) 0.9 been clarified [21]; to this end, the PRECISION trial Prostate volume, mL 0.98 (0.93–1.03) 0.5 Positive DRE 3.16 (0.36–27.5) 0.3 avoided taking cores for ROI with PI-RADS scores of Median diameter of IL, mm 0.77 (0.55–1.08) 0.1 <3 [11]. According to Hauth et al. [26], it would be OR, odds ratio. reasonable and practicable to carry out a mpMRI- based follow-up for PI-RADS score 3 lesions instead with Gleason Scores ≥3+4at FB.So,itmay be of taking biopsies in a diagnostic setting. In our pre- possible to omit FB in AS patients with ILs with PI- sent study, we did not find any predictors of missed RADS scores of ≤3 on mpMRI without missing csPCa. PCa at FB. Unfortunately, due to the small study 146 K. RÖTHLIN ET AL. population, our logarithmic correlation analysis was [8] Bonekamp D, Jacobs MA, El-Khouli R, et al. Advancements in MR imaging of the prostate: from diag- only performed with univariable analysis. nosis to interventions. RadioGraphics. 2011;31:677–703. Our present study had several limitations, which are [9] Stamatakis L, Siddiqui MM, Nix JW, et al. Accuracy of mainly related to its retrospective nature. Secondly, the multiparametric magnetic resonance imaging in con- SB and FB were performed by the same urologist, who firming eligibility for active surveillance for men with was not blinded to the distribution and to the PI-RADS prostate cancer. Cancer. 2013;119:3359–3366. [10] Nahar B, Katims A, Barboza MP, et al. Reclassification rates scores of the patients’ ROIs; this knowledge may have of patients eligible for active surveillance after the addi- influenced his choice regarding the core taking in the tion of magnetic resonance imaging-ultrasound fusion SB. Thirdly, our AS population was small and hetero- biopsy: an analysis of 7 widely used eligibility criteria. geneous: a quarter of patients did not strictly meet AS Urology. 2017;110:134–139. criteria according to the PRIAS study [22]. Furthermore, [11] Kasivisvanathan V, Rannikko AS, Borghi M, et al. MRI- because of the small number of patients analysed, it targeted or standard biopsy for prostate-cancer diagnosis. N Engl J Med. 2018;378:1767–1777. was not possible to calculate a multivariable logarith- [12] Schoots IG, Roobol MJ, Nieboer D, et al. Magnetic mic regression. Lastly, another weakness was the com- resonance imaging–targeted biopsy may enhance parison of FB and SB against each other, instead of the diagnostic accuracy of significant prostate cancer comparing the two biopsy methods with the gold detection compared to standard transrectal ultra- standard of a prostatectomy specimen. sound-guided biopsy: a systematic review and meta- analysis. Eur Urol. 2015;68:438–450. [13] Siddiqui MM, Rais-Bahrami S, Turkbey B, et al. Comparison of MR/ultrasound fusion–guided biopsy Conclusion with ultrasound-guided biopsy for the diagnosis of According to our present study, FB was characterised prostate cancer. JAMA. 2015;313:390–397. [14] Valerio M, Donaldson I, Emberton M, et al. Detection of by a high rate of missed PCa compared to the SB in AS clinically significant prostate cancer using magnetic patients. Further prospective studies are required to resonance imaging-ultrasound fusion targeted biopsy: ascertain the role of FB in patients with PCa under AS. a systematic review. Eur Urol. 2015;68:8–19. [15] Porpiglia F, Manfredi M, Mele F, et al. Diagnostic pathway with multiparametric magnetic resonance imaging ver- Disclosure statement sus standard pathway: results from a randomized pro- spective study in biopsy-naïve patients with suspected No potential conflict of interest was reported by the authors. prostate cancer. Eur Urol. 2017;72:282–288. [16] Radtke JP, Schwab C, Wolf MB, et al. Multiparametric magnetic resonance imaging (MRI) and MRI–transrec- ORCID tal ultrasound fusion biopsy for index tumor detection: correlation with radical prostatectomy specimen. Eur Marco Moschini http://orcid.org/0000-0002-3084-2458 Urol. 2016;70:846–853. Luca Afferi http://orcid.org/0000-0002-5472-1933 [17] Drost FH, Osses D, Nieboer D, et al. Prostate MRI, with or without MRI-targeted biopsy, and systematic biopsy for detecting prostate cancer. Cochrane Database Syst References Rev. 2019;4:CD012663. [18] Bjurlin MA, Taneja SS. Standards for prostate biopsy. [1] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Curr Opin Urol. 2014;24:155–161. Cancer J Clin. 2018;68:7–30. [19] Elkhoury FF, Felker ER, Kwan L, et al. Comparison of [2] Klotz L, VespriniD,SethukavalanP,etal. Long-term targeted vs systematic prostate biopsy in men who are follow-up of a large active surveillance cohort of patients biopsy naive: the prospective assessment of image with prostate cancer. J Clin Oncol. 2015;33:272–277. registration in the diagnosis of prostate cancer [3] Bokhorst LP, Valdagni R, Rannikko A, et al. A decade of (PAIREDCAP) study. JAMA Surg. 2019;154:811. [Epub active surveillance in the PRIAS study: an update and ahead of print]. evaluation of the criteria used to recommend a switch [20] Weinreb JC, Barentsz JO, Choyke PL, et al. PI-RADS to active treatment. Eur Urol. 2016;70:954–960. prostate imaging–reporting and data system: 2015, [4] Moschini M, Fossati N, Sood A, et al. Contemporary man- version 2. Eur Urol. 2016;69:16–40. agement of prostate cancer patients suitable for active [21] Bubendorf L, Diener PA, Grobholz R, et al. surveillance: a North American population-based study. Qualitätsrichtlinien SGPath Prostata. Schweiz Ges Für Eur Urol Focus. 2018;4:68–74. Pathol. 2017 [cited 2018 Jul 31]. Available from: http:// [5] Moschini M, Carroll PR, Eggener SE, et al. Low-risk www.sgpath.ch/docs/QRL/QRL_SGPath_Prostata_ prostate cancer: identification, management, and 2017.pdf outcomes. Eur Urol. 2017;72:238–249. [22] Bul M, Zhu X, Valdagni R, et al. Active surveillance for [6] Mottet N, Bellmunt J, Briers E, et al. EAU - ESTRO - ESUR low-risk prostate cancer worldwide: the PRIAS study. -SIOG guidelines on guidelines on prostate cancer. Eur Urol. 2013;63:597–603. 2017. Available from: https://uroweb.org/wp-content [23] Ma TM, Tosoian JJ, Schaeffer EM, et al. The role of /uploads/09-Prostate-Cancer_2017_web.pdf multiparametric magnetic resonance imaging/ultra- [7] Elkhoury FF, Simopoulos DN, Marks LS. Targeted pros- sound fusion biopsy in active surveillance. Eur Urol. tate biopsy in the era of active surveillance. Urology. 2017;71:174–180. 2018;112:12–19. ARAB JOURNAL OF UROLOGY 147 [24] Hu JC, Chang E, Natarajan S, et al. Targeted prostate restratification in active surveillance of low-risk pros- biopsy to select men for active surveillance: do the tate cancer: a prospective multicenter cohort study. Epstein criteria still apply? J Urol. 2014;192:385–390. Invest Radiol. 2014;49:165–172. [25] Hoeks CM, Somford DM, van Oort IM, et al. Value of [26] Hauth E, Jaeger H, Hohmuth H, et al. Follow-up MR 3-T multiparametric magnetic resonance imaging and imaging of PI-RADS 3 and PI-RADS 4 prostate lesions. magnetic resonance-guided biopsy for early risk Clin Imaging. 2017;43:64–68.
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png
Arab Journal of Urology
Taylor & Francis
http://www.deepdyve.com/lp/taylor-francis/multiparametric-magnetic-resonance-imaging-ultrasound-guided-fusion-EnYWoXgueU
