Development and External Validation of a Novel Nomogram to Predict the Probability of Pelvic Lymph-node Metastases in Prostate Cancer Patients Using Magnetic Resonance Imaging and Molecular Imaging with Prostate-specific Membrane Antigen Positron Emission Tomography

摘要

BACKGROUND: Preoperative assessment of the probability of pelvic lymph-node metastatic disease (pN1) is required to identify patients with prostate cancer (PCa) who are candidates for extended pelvic lymph-node dissection (ePLND). OBJECTIVE: To develop a novel intuitive prognostic nomogram for predicting pathological lymph-node (pN) status in contemporary patients with primary diagnosed localized PCa, using preoperative clinical and histopathological parameters, magnetic resonance imaging (MRI), and prostate-specific membrane antigen (PSMA) positron emission tomography (PET). DESIGN, SETTING, AND PARTICIPANTS: In total, 700 eligible patients who underwent robot-assisted radical prostatectomy and ePLND were included in the model-building cohort. The external validation cohort consisted of 305 surgically treated patients. Logistic regression with backward elimination was used to select variables for the Amsterdam-Brisbane-Sydney nomogram. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Performance of the final model was evaluated using the area under the receiver operating characteristic curve (AUC), calibration plots, and decision-curve analyses. Models were subsequently validated in an external population. RESULTS AND LIMITATIONS: The Amsterdam-Brisbane-Sydney nomogram included initial prostate-specific antigen value, MRI T stage, highest biopsy grade group (GG), biopsy technique, percentage of systematic cores with clinically significant PCa (GG ≥2), and lymph-node status on PSMA-PET. The AUC for predicting pN status was 0.81 (95% confidence interval [CI] 0.78-0.85) for the final model. On external validation, the Amsterdam-Brisbane-Sydney nomogram showed superior discriminative ability to the Briganti-2017 and Memorial Sloan Kettering Cancer Center (MSKCC) nomograms (AUC 0.75 [95% CI 0.69-0.81] vs 0.67 [95% CI 0.61-0.74] and 0.65 [95% CI 0.58-0.72], respectively; p < 0.05), and similar discriminative ability to the Briganti-2019 nomogram (AUC 0.78 [95% CI 0.71-0.86] vs 0.80 [95% CI 0.73-0.86]; p = 0.76). The Amsterdam-Brisbane-Sydney nomogram showed excellent calibration on external validation, with an increased net benefit at a threshold probability of ≥4%. CONCLUSIONS: The validated Amsterdam-Brisbane-Sydney nomogram performs superior to the Briganti-2017 and MSKCC nomograms, and similar to the Briganti-2019 nomogram. Furthermore, it is applicable in all patients with newly diagnosed unfavorable intermediate- and high-risk PCa. PATIENT SUMMARY: We developed and validated the Amsterdam-Brisbane-Sydney nomogram for the prediction of prostate cancer spread to lymph nodes before surgery. This nomogram performs similar or superior to all presently available nomograms.