Stesura Seveso 401Archivio Italiano di Urologia e Andrologia 2022; 94, 4 ORIGINAL PAPER No conflict of interest declared. 7%, respectively (1, 2). After the diagnosis, a better stag- ing would allow both clinician and patient a better- informed choice and a better tailored treatment approach. Current recommendations from European Association of Urology (EAU) (3) and National Comprehensive Cancer Network (NCCN) (4) are for high-risk and VHR localized PCa patients to undergo computed tomography (CT) of abdomen and pelvis and whole-body bone scan (BSc). However, it is commonly known that CT and alternatives such as Magnetic Resonance Imaging (MRI) have a modest performance for nodal staging (5). More recent nuclear imaging such as Positron Emission Tomography (PET) with Choline or Prostate-Specific Membrane Antigen (PSMA) isotopes have been explored, but their utilization in staging localized PCa is not stan- dard (6, 7). PET-PSMA is currently regarded as the most promising radiotracer, but in most of the world it is more expensive and of more limited availability than PET- Choline (8). As such, in France, it is frequent that we find patients being staged with PET-Choline, exam considered in guidelines as having a role in biochemical recurrence. It has the advantage of being cheaper, easily available and reproducible (8). However, its role in staging of localized PCa is disputed, with some papers reporting interesting results in the subset of high risk and VHR patients (9, 10). In this work we explore the role of PET Choline realized in community non-centralized centers in node staging, when compared with CT findings and with the gold stan- dard of bilateral extended pelvic lymphadenectomy, in patients with VHR PCa. METHODS Our center is a high-ranking hospital in PCa care, receiv- ing patients mainly from Paris, but also from all France and the rest of the World. We maintain a prospectively updated database with all patients who undergo a radical prostatectomy in our center, including our study popula- tion. We explored this database for patients with VHR localized PCa who underwent PET Choline as a method of pre-operative staging, who were treated with Robotic Assisted Laparoscopic Prostatectomy (RALP) and Extended Pelvic Lymph Node Dissection (ePLND), between 2010 and Objectives: Localized very high-risk prostate cancer (VHR PCa) has long suffered from the inex-istence of good lymph node staging methods other than invasive surgery, as computed tomogra-phy has low sensitivity for nodal disease. With the rising use of positron emission tomography (PET), it is clinically meaningful to know its value for these patients. Our goal was to evaluate the real-life diag- nostic accuracy of PET Choline in nodal staging, comparing it with the gold standard of extended pelvic lymph node dissection (ePLND). Materials and methods: We reviewed data from a high-volume center, including patients with VHR PCa according to current NCCN guidelines who underwent community 18F-fluorocholine PET/CT; followed by ro-botic assisted laparoscopic prostatecto- my (RALP) and ePLND between 2010 and 2021. Results: We included 44 patients and 88 lymph node regions. Among those, 14/44 (31.8%) patients and 20/88 (22.7%) regions had nodal disease present on definitive pathology. In comparison with ePLND, we found a sensitivity of 64.3% (95% CI, 39.2-89.4%), specificity of 83.3% (95% CI, 70.0- 96.7%), PPV of 64.3% (95% CI, 39.2-89.4%), and NPV of 83.3% (95% CI, 70.0-96.7%) for nodal disease on a patient-based analysis; and sensitivity of 35.0% (95% CI, 14.1-60.0%), speci- ficity of 88.2% (95% CI, 80.6-95.9%), PPV of 46.7% (95% CI, 21.4-71.9%), and NPV of 82.2% (95% CI, 73.4-91.0%) on a region-based analysis. Conclusions: In our view 18F-fluorocholine PET/CT doesn’t meet the criteria to be a standard exam for pre-operative stag- ing for patients with VHR PCa, mostly due to its low sensitivity. However, other radiotracers should continue to be investigated in this setting. KEy WORDs: Positron-emission tomography; Prostatic neoplasms; Neoplasm staging. Submitted 23 October 2022; Accepted 30 October 2022 INTRODUCTION Adequately staging of very high-risk (VHR) localized prostate cancer (PCa) is of utmost importance, as staging may alter treatment choice. Even when patients are well selected for radical treatment, their biochemical recur- rence, metastasis, cancer specific mortality and all-cause mortality at 5 years can be as high as 69%, 20%, 5% and Accuracy of PET-choline in nodal staging of localized very high-risk prostate cancer Nuno Dias 1, 2, Gianmarco Colandrea 1, 3, Francisco Botelho 2, 4, Lara Rodriguez-Sanchez 1, Yann Barbé 1, Petr Macek 1, Xavier Cathelineau 1 1 Urology Department, Institut Mutualiste Montsouris, Paris, France; 2 Urology Department, São João Hospitalar and University Center, Porto, Portugal; 3 Unit of Urology, Division of Experimental Oncology, URI Urological Research Institute, IRCCS San Raffaele Scientific Institute, Vita-Salute San Raffaele University, Milan, Italy; 4 Institute for Life and Health Sciences Investigation, School of Medicine, Minho University, Braga, Portugal. DOI: 10.4081/aiua.2022.4.401 Summary Archivio Italiano di Urologia e Andrologia 2022; 94, 4 N. Dias, G. Colandrea, F. Botelho, L. Rodriguez-Sanchez, Y. Barbé, P. Macek, X. Cathelineau 402 2021. Patients who had a previous malignancy or history of prior hormonal therapy were excluded. Data collection Data was collected from our continuously updated data- base and patient records. We recorded age at time of sur- gery, pre-operative total PSA, clinical T staging, biopsy results, PET Choline results, anatomopathological study of prostatectomy and lymph node dissection specimens. The study has received approval by the institutional Ethics’ Committee. All research was conducted respecting the latest version of Helsinki’s declaration. Risk groups Patients were included if they fulfilled current NCCN guidelines (4) criteria for VHR patients: a) at least 1 of: cT3b-cT4, primary Gleason pattern 5, > 4 cores with Gleason Group 4 or 5; and/or b) at least 2 of: cT3a, Gleason Group 4 or 5, PSA > 20 ng/mL. PET Choline PET/CT exams were performed in various local centers in France, with images acquired using 18F-fluorocholine tracer being obtained according to institutional protocols, with images being reported by several different commu- nity radiologists. PET Choline imaging were acquired alongside a tho-raco-abdominal-pelvic CT scan. Images were not centrally reviewed. The diagnosis of pathological lymph nodes was based on the presence of increased uptake on PET images of the radiotracer, with exams being considered as positive if reported by the radiologist as positive or suspicious, irre- spective of standardized uptake value (SUV). CT adjunc- tive imaging was used only to locate the anomalies and not as a staging method by itself. surgery RALP + ePLND was performed in our institution by seven different assistants, with long experience in robotic sur- gery and RALP, with frequent participation of fellows under their direct supervi-sion. RALP was performed transperitoneally, and ePLND was performed as per guidelines standard (including node packets from com- mon iliac distal to the ureteric crossing, external iliac, internal iliac, and obturator origins), with nodes from each side being sent and analyzed separately (3). Other nodal regions such as presacral and retroperitoneal would only be removed in select cases where imaging was sus- picious of lymph node metastasis in those locations, with no cases present in our population. Histology Biopsy cores were analyzed in the institution where the biopsy was performed and not reviewed, except in doubt- ful cases. Prostatectomy and lymph node dissection prod- ucts were fixed in formalin and examined in our center by specialized uropathologists, in accordance with ISUP guidelines, with left and right templates being observed separately (11). statistics Statistical analysis was performed with IBM® SPSS® v27 software. Median and interquartile range are presented for continuous variables; frequencies are reported for cat- egorical variables. Characteristics of patients were com- pared using Chi-squared analysis for categorical variables and non-parametric Mann-Whitney U tests for continu- ous variables (age). Statistically significance in this study was set as p < 0.05. All reported p values are two-sided. Test characteristics (sensitivity, specificity, VPP, VPN) were calculated for both patient and region based analy- sis, with 95% confidence intervals. RESULTS We identified 2104 consecutive patients submitted to RALP, with 104 being VHR patients and 45 who had had PET Choline as part of their disease staging. One patient meeting the inclusion criteria was excluded for not hav- ing all information regarding PET Choline examen avail- able. Forty-four patients and eighty-eight lymph node regions were included in our analysis. Patient characteristics Patient characteristics are shown in Table 1. Median patient age was of 63 years old (IQR 59-68) and median total PSA of 15 (IQR 9.8-22.2) ng/mL, with 23 patients having a PSA > 15 ng/mL. All 44 patients had VHR PCa, Table 1. Patient characteristics. Total n = 44 Age, median (P25-P75) 63 (59-68) Total PSA, ng/mL, median (P25-75) 15 (9.8-22.2) Clinical T stage, n (%) T1c 4 (9) T2a 6 (14) T2b 4 (9) T2c 8 (18) T3a 17 (39) T3b 5 (11) Biopsy at our institution 15 (34) Biopsy ISUP, n (%) 1 5 (11) 2 6 (14) 3 8 (18) 4 16 (36) 5 9 (21) Number positive cores, median (P25-75) 6 (4-7) > 4 cores with ISUP ≥ 4, n (%) 17 (39) Pathological T stage, n (%) T2b 1 (2) T2c 7 (16) T3a 21 (48) T3b 15 (34) Prostatectomy ISUP, n (%) 2 8 (18) 3 22 (50) 4 4 (9) 5 10 (23) Nodal disease, n (%) 14 (32) Number of positive lymph-nodes, n (%) 0 30 (68) 1 4 (9) 2 1 (2) ≥ 3 9 (20) Number removed nodes, median (P25-75) 19 (11-26) 403Archivio Italiano di Urologia e Andrologia 2022; 94, 4 PET-choline in nodal staging of prostate cancer with 17 (39%) and 5 (11%) having cT3a and cT3b dis- ease, respectively. Overall, 9 (21%) patients had ISUP of 5, and 17 (39%) had more than 4 biopsy cores with ISUP ≥ 4. Histological results after RALP + ePLND showed most patients locally advanced disease, with 48% pre- senting with pT3a and 34% with pT3b disease; and 23% of patients had a Gleason Grade Group of 5. With a medi- an 19 (IQR 11-26) nodes removed per patient, 14 out of 44 (32%) of patients had nodal metastasis. Four (9%) patients had 1 positive lymph node, 1 (2%) had 2 posi- tive lymph nodes and 9 (20%) had ≥ 3 positive lymph nodes identified. 18F-fluorocholine PET/CT and ePLND Table 2 presents the cross results between 18F-fluoro- choline PET/CT and anatomopathological findings, in a patient and region base analysis. When compared with ePLND, 18F-fluorocholine PET/CT had a false negative rate of 17% and 18% for patient and region-based analy- sis, respectively; and a false positive rate of 50% and 53%, respectively. Table 3 shows the performance of 18F-fluorocholine PET/CT as a test for nodal staging. On a patient-based analysis (n=44); sensitivity was 64.3% (95% CI, 39.2- 89.4%), specificity 83.3% (95% CI, 70.0-96.7%), PPV 64.3% (95% CI, 39.2-89.4%), and NPV 83.3% (95% CI, 70.0-96.7%). On a region-based analysis (n=88); sensi- tivity was 35.0% (95% CI, 14.1-56.0%), specificity 88.2% (95% CI, 80.6-95.9%), PPV 46.7% (95% CI, 21.4- 71.9%), and NPV 82.2% (95% CI, 73.4-91.0%). Know- ing the correlation between biomarkers uptake and PSA, we performed the same analysis for patients with PSA > 15ng/mL and on a region basis; sensitivity was 50% (95% CI, 24-76%), with specificity of 84% (95% CI, 72-97%), PPV of 58% (95% CI, 30-86%) and NPV of 79% (95% CI, 66-93%). sub-group with ≥ 3 positive lymph nodes Nine (20%) patients had at least 3 positive lymph nodes at final pathology, with 6 of them having a positive 18F- fluorocholine PET/CT. For this subgroup, on a patient- based analysis (n=9); sensitivity was 66.7% (95% CI, 35.9-97.5%), specificity 77.1% (95% CI, 63.2-91.1%), PPV 42.9% (95% CI, 16.9-68.8%), and NPV 90.0% (95% CI, 79.3-100.0%). DISCUSSION The importance of a good nodal staging in VHR PCa patients is three ways. First, patients with node positive disease should be well identified, since these patients may benefit from adjuvant hormonal therapy and/or radio- therapy after surgery, and because they have worse out- comes in biochemical free survival, metastasis free sur- vival and overall survival (2). Second, when discussing treatment options with a patient before deciding on a rad- ical treatment, we should be able to provide him with the best information possible, in order to allow the best indi- vidual decision. In VHR patients in particular, the rate of node positive disease has been estimated to be as high as 37% (2). Third, when surgery for VHR patients is per- formed, it entails not only radical prostatectomy but also an ePLND, which increases surgical time and risk of peri- operative complications, and has not been shown to lead to better patient outcomes (12). As such, if an accurate enough non-invasive staging method were easily available, it could lead to better iden- tifications of node positive patients before intervention, best individualized information before treatment deci- sions and eventual nonrealization of a procedure with sig- nificant comorbidities but no increased benefit to patients (12). Current recommended staging of localized VHR PCa is CT and whole-body bone scan (3, 4). CT scan relies mainly on size criteria, location and contrast enhance- ment to classify lymph nodes as suspicious or not. However, CT scan has been extensively reported as hav- ing a poor performance for nodal staging, with sensitivity depending on the prostate cancer risk, but remaining low even for high-risk patients (13). Knowing this, previous studies have searched for adequate methods for non-inva- sive PCa staging. Those alternatives should be sensitive, specific, clinically useful, easily accessible and repro- ducible (14). Previous studies have explored the role of PET Choline and PET PSMA as staging imaging for local- ized PCa, mostly in a trial setting. Van den Bergh et al. (9) performed a prospective study with 75 patients with negative CT scans and lymph node extension risk assessed to be between 10-35% with Partin tables and compared the performance of pre-operative staging with 11C-Choline PET and pelvic MRI with super-extended pelvic lymphadenectomy results. They reported a low performance of both PET and MRI, main- Table 3. Test performance on patient and region-based analysis. 18F PET-Choline Values Patient-based analysis Patients with positive LN 14/44 (31.8%) Sensitivity 64.3% [39.2-89.4%] Specificity 83.3% [70.0-96.7%] PPV 64.3% [39.2-89.4%] NPV 83.3% [70.0-96.7%] Region-based analysis Regions with positive LN 20/88 (22.7%) Sensitivity 35.0% [14.1-56.0%] Specificity 88.2% [80.6-95.9%] PPV 46.7% [21.4-71.9%] NPV 82.2% [73.4-91.0%] Table 2. Two-by-two tables presenting cross-results for nodal disease of ePLND and 18F PET-Choline, on patient and region-based analysis. 18F PET-Choline Positive ePLND Negative ePLND Patient-based analysis Total, n 14 30 Positive/suspicious, n (%) 9 (64) 5 (17) Negative, n (%) 5 (36) 25 (83) Region-based analysis Total, n 20 68 Positive/suspicious, n (%) 7 (35) 8 (12) Negative, n (%) 13 (65) 60 (88) Archivio Italiano di Urologia e Andrologia 2022; 94, 4 N. Dias, G. Colandrea, F. Botelho, L. Rodriguez-Sanchez, Y. Barbé, P. Macek, X. Cathelineau 404 ly at the expenses of a low sensitivity (PET: 19% and 8% on patient and region-based analysis; MRI: 36% and 10% on patient and region-based analysis), which lead them to suggest the absence of benefit in this patient cohort. Schiavina et al. (10) reported their retrospective experi- ence with 11C-Choline PET in a cohort of intermediate, high and VHR patients. They report low sensitivities for the detection of nodal disease both on medium and high- risk patients, with 17% and 40% in a region-based analy- sis, respectively. However, in a subgroup of 28 VHR patients where 50% harbored nodal disease, the test per- formance was better, with sensitivity, specificity, positive predictive value and negative predictive value of 71%, 93%, 91% and 76%, respectively. In our sub-analysis of 9 patients with ≥ 3 positive lymph nodes on high-quality ePLND, 18F-fluorocholine PET/CT on a patient-based analysis had sensitivity, specificity, PPV and NPV of 66.7%, 77.1, 42.9, and 90.0%, respec- tively. This was similar to our results in the overall VHR PCa population, with a high NPV pointing to a potential role as means of excluding ≥ 3 positive lymph nodes dis- ease. As the number of patients in this group was low (n=9) it would be interesting to see multicenter results in this subgroup, as the presence of ≥ 3 positive lymph nodes has been suggested as an independent risk factor for both biochemical recurrence and metastatic progres- sion, with these patients benefiting from adjuvant treat- ment after surgery (15). More recently, some studies have shown promise of good results with PSMA PET/CT staging in the treatment-naïve high-risk patients. Hope et al. (16) performed a trial which enrolled 764 patients with intermediate and high-risk PCa being considered for prostatectomy at a single insti- tution, who performed 68Ga-PSMA-11 PET/CT prior to intervention. 277 patients ended up receiving radical prostatectomy with extended lymph node dissection, in which the PET sensitivity, specificity, positive predictive value and negative predictive value were of 40% (95% CI, 34-46%), 95% (95% CI, 92-97%), 75% (95% CI, 70- 80%) and 81% (95% CI, 76-85%), respectively. Hofman et al. (17) published the results of probably the most important study so far, the proPSMA trial. This was a multicenter randomized controlled trial in 10 Australian hospitals. 302 patients with high-risk PCa were randomly selected for either usual imaging or 68Ga- PSMA-11 PET/CT. PSMA PET performed better than CT scan, and very well, with an 85% (95% CI, 74-96%) sen- sitivity and 91% (95% CI, 85-97%) specificity for nodal disease. Also, although patients who underwent PSMA PET as staging method didn’t undergo classical staging previously, the exam results are reported to have changed the clinical preferred options in 28% of patients, with 14% being directed from curative to palliative treatment, 7% with a change in radiotherapy technique and 7% with a change in surgical technique. Our work analyzed a real-life experience of a region and country where availability of 68Ga-PSMA-11 PET/CT is scarce and 18F-fluorocholine is widespread. As such, many of our patients don’t have the option for the newest but less accessible radiotracer. For that reason, it was important for us to understand the value caretakers are providing VHR PCa patients regarding staging imaging. It comes with the limitations of being a retrospective study, which includes patients operated in a single center, with imaging being performed in many centers. The inclusion of only patients who underwent RALP + ePLND may have selected patients with less advanced localized disease stages, more amenable for radical prostatectomy as a treatment choice; and also excluded metastatic patients and radiotherapy patients who didn’t have sur- gery. The analysis of 18F-fluorocholine PET/CT accuracy was also only made by patient and side, and using every lymph node region separately, as our current technique of ePLND retrieves all nodes of a side all together in a sin- gle packet. Reviewing our data is important to understand what real life results we get when using 18F-fluorocholine PET/CT in VHR patients, and we found that even though the exam specificity and NPV were > 80%, its low sensitivity both on patient (64.3%) and side (35.0%) based analysis are not enough to consider the exam as a “game changer”, even in the VHR PCa group. After the first data on new imaging modalities emerged, some studies started suggesting the wide-spread study of new tracers, such as 177Lu-PSMA (18) and government approval of PET/CT for staging (19), which should facili- tate the increasing the availability and decreasing the costs of those exams. As such, efforts to advance the knowledge in the field and maintain low-cost accessibility to exams considered clin- ically useful should be pursued continuously. CONCLUSIONS We conclude that 18F-fluorocholine PET/CT as staging for patients with VHR localized PCa doesn’t meet the criteria to be a standard exam for pre-operative staging. We there- fore believe new radiotracers such as the 68Ga-PSMA-11 should continue to be assessed in this setting. Governments and health services should also make efforts to ensure that new meth-ods of imaging with clinical util- ity have a reasonable financial cost and availability. REFERENCES 1. Sundi D, Wang VM, Pierorazio PM, et al. Very-high-risk localized prostate cancer: definition and outcomes. Prostate Cancer Prostatic Dis. 2013; 17:57-63. 2. Sundi D, Tosoian JJ, Nyame YA, et al. Outcomes of Very High- Risk Prostate Cancer After Radical Prostatectomy: Validation Study From 3 Centers. Cancer. 2019; 125:391-397. 3. Mottet N, Cornford P, van den Bergh RCN, et al. EAU, EANM, ESTRO, ESUR, ISUP, SIOG: Guidelines on prostate cancer. 2022. 4. NCCN Clinical practice guidelines in oncology: prostate cancer. Version 3.2022, consulted 23rd April 2022. Available from: https://www.nccn.org/professionals/physician_gls/pdf/prostate.pdf 5. Hövels AM, Heesakkers RAM, Adang EM, et al. The diagnostic accuracy of CT and MRI in the staging of pelvic lymph nodes in patients with prostate cancer: a meta-analysis. Clin Radiol 2008; 63:387-95. 6. Umbehr MH, Muntener M, Hany T, et al. The role of 11C- choline and 18F-fluorocholine positron emission tomography (PET) and 405Archivio Italiano di Urologia e Andrologia 2022; 94, 4 PET-choline in nodal staging of prostate cancer PET/CT in prostate cancer: a systematic review and meta-analysis. Eur Urol 2013; 64:106-17. 7. Evangelista L, Guttilla A, Zattoni F, et al. Utility of choline positron emission tomography/computed tomography for lymph node involvement identification in intermediate- to high-risk prostate can- cer: a systematic literature review and meta-analysis. Eur Urol 2013; 63:1040-8. 8. Alberts I, Mingels C, Zacho HD, et al. Comparing the clinical per- formance and cost efficacy of 68Ga-PSMA-11 and 18F-PSMA-1007 in the diagnosis of recurrent prostate cancer: a Markov chain deci- sion analysis. Eur J Nucl Med Mol Imaging. 2022; 49:4252-4261. 9. van den Bergh L, Lerute E, Haustermans K, et al. Final analysis of a prospective trial on functional imaging for nodal staging in patients with prostate cancer at high risk for lymph node involve- ment. Urol Oncol. 2015; 33:109.e23-31. 10. Schiavina R, Bianchi L, Bianchi FM, et al. Preoperative staging with 11C-Choline PET/CT is adequately accurate in patients with very high-risk prostate cancer. Clin Genitourin Cancer. 2018; 16:305-312.e1. 11. Saramatunga H, Montironi R, True L, et al. International Society of Urological Pathology (ISUP) Consensus Conference on Handling and Staging of Radical Prostatectomy Specimens. Working group 1: specimen handling. Mod Pathol. 2011; 24:6-15. 12. Bianchi L, Gandaglia G, Fossati N, et al. Pelvic lymph node dis- section in prostate cancer: indications, extent and tailored approach- es. Urologia. 2017; 84:9-19. 13. Briganti A, Abdollah F, Nini A, et al. Performance characteris- tics of computed tomography in detecting lymph node metastasis in contemporary patients with prostate cancer treated with extended pelvic lymph node dissection. Eur Urol. 2012; 61:1132-8. 14. Balboaca SD. Medical diagnostic tests: a review of test anatomy, phases and statistical treatment of data. Comput Math Methods Med. 2019; 2019:1891569. 15. Stolzenbach LF, Knipper S, Mandel P, et al. Oncological out- comes of pathologically organ-confined, lymph node-positive prostate cancer after radical prostatectomy. Urol Oncol. 2021; 39:234.e1- 234.e7. 16. Hope T, Eiber M, Armstrong W, et al. Diagnostic accuracy of 68Ga-PSMA-11 PET for pelvic nodal metastasis detection prior to radical prostatectomy and pelvic lymph node dissection. JAMA Oncol. 2021; 7:1635-1642. 17. Hofman M, Lawrentschuk N, Francis RJ, et al. Prostate specific membrane antigen PET-CT in patients with high-risk prostate can- cer before curative-intent surgery or radiotherapy (proPSMA): a prospective, randomized, multi-centre study. Lancet. 2020; 395:1208-1216. 18. Farolfi A, Calderoni L, Mattana F, et al. Current and emerging clinical applications of PSMA PET diagnostic imaging for prostate cancer. J Nucl Med. 2021; 62:596-604. 19. Afaq A, Ell PJ, Bomanji JB. Is it time to fund routine NHS usage of PSMA PET CT? Nucl Med Commun. 2019; 40:975-979. Correspondence Nuno Dias, MD (Corresponding Author) nunodiasds@gmail.com Gianmarco Colandrea,MD colandrea.gianmarco@hsr.it Lara Rodriguez-Sanchez rodriguezsanchezlara@gmail.com Yann Barbé yann.barbe@imm.fr Petr Macek, MD petr.macek@imm.fr Xavier Cathelineau, MD, PhD xavier.cathelineau@imm.fr Urology Department, Institut Mutualiste Montsouris, Paris (France) 42 Bd Jourdan; 75014 Paris (France) Francisco Botelho, MD, MSCE francisco.botelho@gmail.com Urology Department, São João Hospitalar and University Center, Porto (Portugal)