Diagnostic performance of prediction models for extraprostatic extension in prostate cancer: a systematic review and meta-analysis

Abstract

Purpose In recent decades, diverse nomograms have been proposed to predict extraprostatic extension (EPE) in prostate cancer (PCa). We aimed to systematically evaluate the accuracy of MRI-inclusive nomograms and traditional clinical nomograms in predicting EPE in PCa. The purpose of this meta-analysis is to provide baseline summative and comparative estimates for future study designs. Materials and methods The PubMed, Embase, and Cochrane databases were searched up to May 17, 2023, to identify studies on prediction nomograms for EPE of PCa. The risk of bias in studies was assessed by using the Prediction model Risk Of Bias ASsessment Tool (PROBAST). Summary estimates of sensitivity and specificity were obtained with bivariate random-effects model. Heterogeneity was investigated through meta-regression and subgroup analysis. Results Forty-eight studies with a total of 57 contingency tables and 20,395 patients were included. No significant publication bias was observed for either the MRI-inclusive nomograms or clinical nomograms. For MRI-inclusive nomograms predicting EPE, the pooled AUC of validation cohorts was 0.80 (95% CI: 0.76, 0.83). For traditional clinical nomograms predicting EPE, the pooled AUCs of the Partin table and Memorial Sloan Kettering Cancer Center (MSKCC) nomogram were 0.72 (95% CI: 0.68, 0.76) and 0.79 (95% CI: 0.75, 0.82), respectively. Conclusion Preoperative risk stratification is essential for PCa patients; both MRI-inclusive nomograms and traditional clinical nomograms had moderate diagnostic performance for predicting EPE in PCa. This study provides baseline comparative values for EPE prediction for future studies which is useful for evaluating preoperative risk stratification in PCa patients. Critical relevance statement This meta-analysis firstly evaluated the diagnostic performance of preoperative MRI-inclusive nomograms and clinical nomograms for predicting extraprostatic extension (EPE) in prostate cancer (PCa) (moderate AUCs: 0.72–0.80). We provide baseline estimates for EPE prediction, these findings will be useful in assessing preoperative risk stratification of PCa patients. Key points • MRI-inclusive nomograms and traditional clinical nomograms had moderate AUCs (0.72–0.80) for predicting EPE. • MRI combined clinical nomogram may improve diagnostic accuracy of MRI alone for EPE prediction. • MSKCC nomogram had a higher specificity than Partin table for predicting EPE. • This meta-analysis provided baseline and comparative estimates of nomograms for EPE prediction for future studies. Graphical Abstract

Deep Learning Enhances Detection of Extracapsular Extension in Prostate Cancer from mpMRI of 1001 Patients

“…This underscores the critical need for accurate preoperative ECE prediction to tailor surgical plans and ensure safe margin resection without compromising functional outcomes. Despite the validation of clinical nomograms for ECE prediction 7,8,20,21 , these were developed prior to the widespread adoption of mpMRI, highlighting a gap in leveraging modern imaging in surgical planning.…”

“…While predictive nomograms based on clinicopathological attributes as well as MRI have been previously developed and validated 7,8 , the intricate anatomy of the prostate and the subtle contrast between cancerous and healthy tissues present formidable challenges for ECE detection on magnetic resonance imaging (MRI) images. In the ever-evolving landscape of medical imaging and diagnosis, machine learning (ML) and deep learning (DL) algorithms have emerged as revolutionary tools with widespread applications [9][10][11][12] .…”