Identification of the Optimal Candidates for Nodal Staging with Extended Pelvic Lymph Node Dissection among Prostate Cancer Patients Who Underwent Preoperative Prostate-specific Membrane Antigen Positron Emission Tomography. External Validation of the Memorial Sloan Kettering Cancer Center and Briganti Nomograms and Development of a Novel Tool.
Giorgio Gandaglia,Francesco Barletta,Daniele Robesti,Simone Scuderi,Pawel Rajwa,Juan Gomez Rivas,Laura Ibanez,Timo F. W. Soeterik,Lorenzo Bianchi,Luca Afferi,Claudia Kesch,Christopher Darr,Hongqian Guo,Junlong Zhuang,Fabio Zattoni,Wolfgang Fendler,Giancarlo Marra,Armando Stabile,Daniele Amparore,Nicolai A. Huebner,Alexander Giesen,Steven Joniau,Riccardo Schiavina,Eugenio Brunocilla,Agostino Mattei,Fabrizio Dal Moro,Jesus Moreno Sierra,Francesco Porpiglia,Maria Picchio,Roderick van den Bergh,Shahrokh F. Shariat,Francesco Montorsi,Alberto Briganti
DOI: https://doi.org/10.1016/j.euo.2023.05.003
IF: 8.208
2023-01-01
European Urology Oncology
Abstract:Background: Although the therapeutic role of extended pelvic lymph node dissection (ePLND) in patients with prostate cancer (PCa) is still under debate, this procedure is recommended for staging purposes in selected cases. Nomograms for predicting lymph node invasion (LNI) do not account for prostate-specific membrane antigen (PSMA) positron emission tomography (PET) imaging, which is characterized by a high negative predictive value for nodal metastases.Objective: To externally validate models predicting LNI in patients with miN0M0 PCa at PSMA PET and to develop a novel tool in this setting.Design, setting, and participants: Overall, 458 patients with miN0M0 disease undergoing radical prostatectomy (RP) and ePLND at 12 centers between 2017 and 2022 were identified.Outcome measurements and statistical analyses: Available tools were externally validated using calibration plots, the area under the receiver operating characteristic curve (AUC), and decision curve analyses to assess calibration, discrimination, and the net benefit. A novel coefficient-based model was developed, internally validated, and compared with available tools.Results and limitations: Overall, 53 patients (12%) had LNI. The AUC was 69% for the Briganti 2012, 64% for the Briganti 2017, 73% for the Briganti 2019, and 66% for the Memorial Sloan Kettering Cancer Center nomogram. Multiparametric magnetic resonance imaging stage, biopsy grade group 5, the diameter of the index lesion, and the percentage of positive cores at systematic biopsy were independent predictors of LNI (all p <= 0.04). Internal cross-validation confirmed a coefficient-based model with AUC of 78%, better calibration, and a higher net benefit in comparison to the other nomograms assessed. Use of a 5% cutoff would have spared 47% ePLND procedures (vs 13% for the Briganti 2019 nomogram) at the cost of missing only 2.1% LNI cases. The lack of central review of imaging and pathology represents the main limitation.Conclusions: Tools for predicting LNI are associated with suboptimal performance for men with miN0M0 PCa. We propose a novel model for predicting LNI that outperforms available tools in this population. Patient summary: Tools currently used to predict lymph node invasion (LNI) in prostate cancer are not optimal for men with negative node findings on PET (positron emission tomography) scans, leading to a high number of unnecessary extended pelvic lymph node dissection (ePLND) procedures. A novel tool should be used in clinical practice to identify candidates for ePLND to reduce the risk of unnecessary procedures without missing LNI cases.(c) 2023 European Association of Urology. Published by Elsevier B.V. All rights reserved.