Risk analysis of adverse pathological changes in cT2cN0M0 prostate cancer after robot-assisted radical prostatectomy: Results from a population-based study.

Objective: To explore clinicopathological predictors of adverse pathological changes (APCs) (upgrading, upstaging, and positive surgical margin [PSM]) after robot-assisted radical prostatectomy (RARP) in clinical tumor stage 2c (cT2c) prostate cancer (PCa) patients. Methods: From January 2018 to December 2022, cT2cN0M0 PCa patients who underwent prostate biopsies and subsequent RARP at the Peking University First Hospital with an interval between biopsy and RARP of ≤90 days were included. Univariable and stepwise multivariable logistic regression analyses were performed to identify independent risk factors associated with APCs. Nomograms were constructed based on these predictive models. The performance of the nomograms was evaluated by receiver operating characteristic curves, decision curve analyses, and calibration plots. Results: A total of 423 eligible cT2cN0M0 PCa patients were included. The rates of upgrading, upstaging, and PSM in our cohort were 33%, 51%, and 35%, respectively. The stepwise multivariate logistic analysis suggested that PSA density and the percentage of positive cores in systematic biopsy were significantly associated with the occurrence of APCs. The score of the Prostate Imaging Reporting and Data System, PSA density, and the International Society of Urological Pathology grade group (IGG) of needle-biopsy specimens (or clinical IGG [cIGG]) were significantly associated with upgrading. The PSA density, percentage of positive cores in systematic biopsy, and largest tumor percentage in all cores of each patient (LTP) were significantly associated with upstaging. The PSA density and LTP were significantly associated with the PSM. Based on these results, four nomograms were developed. Receiver operating characteristic curves, decision curve analyses, and calibration plots implied that the nomograms exhibited excellent accuracy. Conclusion: The predictive models we developed could help to identify high-risk PCa early, and optimize clinical decisions of cT2cN0M0 PCa patients.