Cancer imaging : the official publication of the International Cancer Imaging Society · 2025
Development and validation of a novel clinical-radiological-pathological scoring system for preoperative prediction of extraprostatic extension in prostate cancer: a multicenter retrospective study.
СтатьяМногоцентровое исследованиеВалидационное исследованиеНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Yang Liqin, Jin Pengfei, Wang Ximing, Li Zhiping, Xu Huijing, Zhang Yongsheng, Cui Feng
Аннотация
OBJECTIVE: To develop and validate a multimodal scoring system integrating clinical, radiological, and pathological variables to preoperatively predict extraprostatic extension (EPE) in prostate cancer (PCa).
METHODS: This retrospective study included 667 PCa patients divided into a derivation cohort and two validation cohorts. Evaluated parameters comprised prostate-specific antigen density (PSAD), curvilinear contact length (CCL), lesion longest diameter (LD), National Cancer Institute EPE grade (NCI_EPE), International Society of Urological Pathology grade (ISUP), and other relevant variables. Independent predictors were identified through univariate and multivariate regression analysis to construct a logistic model. Coefficients from this model were then weighted to establish a scoring system. The predictive performance of the NCI_EPE, logistic model, and scoring system was systematically evaluated and compared. Finally, the scoring system was stratified into four distinct risk categories.
RESULTS: Multivariate analysis identified NCI_EPE, PSAD, CCL/LD, and ISUP as independent predictors of EPE. In the derivation and validation cohorts, the scoring system demonstrated robust predictive accuracy for EPE, with AUCs of 0.849, 0.830, and 0.847, respectively. These values outperformed the NCI_EPE (Derivation cohort: 0.849 vs. 0.750, P < 0.003, Validation cohort 1: 0.830 vs. 0.736, P = 0.138, Validation cohort 2: 0.837 vs. 0.715, P = 0.003) and were comparable to the logistic model (Derivation cohort: 0.849 vs. 0.860, P = 0.228, Validation cohort 1: 0.830 vs. 0.849, P = 0.711, Validation cohort 2: 0.837 vs. 0.843, P = 0.738). Decision curve analysis revealed higher net clinical benefit for both the scoring system and logistic model compared to the NCI_EPE. Risk stratification using the scoring system categorized patients into four tiers: low (0-3), intermediate-low (4-6), intermediate-high (7-9), and high risk (10-12) with corresponding mean EPE probabilities of 9.9%, 26.0%, 52.0%, and 85.0%. These probabilities closely aligned with observed pT3 incidences in the derivation and validation cohorts.
CONCLUSIONS: The scoring system provides enhanced predictive accuracy for EPE, preoperatively stratifying patients into distinct risk categories to facilitate personalized therapeutic strategies.
METHODS: This retrospective study included 667 PCa patients divided into a derivation cohort and two validation cohorts. Evaluated parameters comprised prostate-specific antigen density (PSAD), curvilinear contact length (CCL), lesion longest diameter (LD), National Cancer Institute EPE grade (NCI_EPE), International Society of Urological Pathology grade (ISUP), and other relevant variables. Independent predictors were identified through univariate and multivariate regression analysis to construct a logistic model. Coefficients from this model were then weighted to establish a scoring system. The predictive performance of the NCI_EPE, logistic model, and scoring system was systematically evaluated and compared. Finally, the scoring system was stratified into four distinct risk categories.
RESULTS: Multivariate analysis identified NCI_EPE, PSAD, CCL/LD, and ISUP as independent predictors of EPE. In the derivation and validation cohorts, the scoring system demonstrated robust predictive accuracy for EPE, with AUCs of 0.849, 0.830, and 0.847, respectively. These values outperformed the NCI_EPE (Derivation cohort: 0.849 vs. 0.750, P < 0.003, Validation cohort 1: 0.830 vs. 0.736, P = 0.138, Validation cohort 2: 0.837 vs. 0.715, P = 0.003) and were comparable to the logistic model (Derivation cohort: 0.849 vs. 0.860, P = 0.228, Validation cohort 1: 0.830 vs. 0.849, P = 0.711, Validation cohort 2: 0.837 vs. 0.843, P = 0.738). Decision curve analysis revealed higher net clinical benefit for both the scoring system and logistic model compared to the NCI_EPE. Risk stratification using the scoring system categorized patients into four tiers: low (0-3), intermediate-low (4-6), intermediate-high (7-9), and high risk (10-12) with corresponding mean EPE probabilities of 9.9%, 26.0%, 52.0%, and 85.0%. These probabilities closely aligned with observed pT3 incidences in the derivation and validation cohorts.
CONCLUSIONS: The scoring system provides enhanced predictive accuracy for EPE, preoperatively stratifying patients into distinct risk categories to facilitate personalized therapeutic strategies.