Magnetic Resonance Imaging-Based Artificial Intelligence in Predicting Prostate Cancer Biochemical Recurrence: Systematic Review and Meta-Analysis

Clinical Snapshot

Key Findings

• P Value: Internal vs. external validation: sensitivity P=0.73; specificity P>0.99; AUC P=0.53; DOR P=0.98 — no statistically significant differences

• Effect Size: Internal validation: pooled sensitivity 0.80, specificity 0.83, AUC 0.86. External validation: pooled sensitivity 0.82, specificity 0.83, AUC 0.84. No statistically significant difference between internal and external validation cohorts across all metrics.

• Primary Outcome: Diagnostic performance of MRI-based AI models for predicting biochemical recurrence of prostate cancer

• Nnt Or Sensitivity: Best-performing models by metric: Medical Net and XGBoost achieved highest sensitivity and AUC; multiple kernel learning and SVM achieved highest specificity. No NNT calculable from available data. Sensitivity-specificity trade-off identified in integrated radiomic-clinical models.

• Confidence Interval: Internal validation: sensitivity 95% CI 0.73–0.86 (PI 0.48–0.99); specificity 95% CI 0.77–0.89 (PI 0.49–1.00); AUC 95% CI 0.83–0.89 (PI 0.74–0.99). External validation: sensitivity 95% CI 0.72–0.91 (PI 0.54–0.99); specificity 95% CI 0.71–0.92 (PI 0.49–1.00); AUC 95% CI 0.79–0.90 (PI 0.70–0.98).

Clinical Application

Currently not feasible for routine clinical implementation. Barriers include: lack of standardised MRI acquisition protocols across institutions, absence of prospectively validated commercial AI tools for BCR prediction, significant intermodel variability precluding a single recommended platform, and the need for local model validation before deployment. Research feasibility in academic centres with existing MRI radiomics pipelines is reasonable. In Australia, multiparametric MRI (mpMRI) for prostate cancer is Medicare-rebatable (MBS Item 63541 and related items) for staging and active surveillance. However, no TGA-approved AI tool specifically for MRI-based BCR prediction is currently listed on the ARTG. The RACGP and Urological Society of Australia and New Zealand (USANZ) guidelines do not yet incorporate AI-based BCR prediction tools into standard care pathways. PSA-based BCR definitions (Phoenix criteria for radiotherapy; 0.2 ng/mL post-prostatectomy) remain the clinical standard. This meta-analysis supports the research agenda for Australian prospective multicentre studies (e.g., through the Prostate Cancer Foundation of Australia research network) but does not yet justify PBS or MBS funding for AI-based BCR prediction tools. Clinicians should be aware that the wide prediction intervals suggest individual patient predictions from current models carry substantial uncertainty. Adult males with confirmed prostate cancer undergoing radical prostatectomy or radiotherapy in centres with access to multiparametric MRI and computational infrastructure for AI model deployment. Most applicable to intermediate- and high-risk PCa where BCR risk stratification influences adjuvant treatment decisions.

Abstract

BACKGROUND: Artificial intelligence (AI) has emerged as a promising tool for prostate cancer (PCa) risk stratification and outcome prediction. However, current studies often lack multicenter external validation, have limited sample sizes, present significant intermodel variability, and face overfitting concerns. OBJECTIVE: This study aimed to comprehensively evaluate the diagnostic performance of magnetic resonance imaging (MRI)-based AI models in predicting biochemical recurrence (BCR) of PCa. METHODS: Systematic searches were conducted in the PubMed, Embase, Web of Science, and Cochrane Library databases up to January 13, 2026. Studies were included that involved participants diagnosed with PCa, used MRI-based AI for predicting BCR, and had clearly defined reference standards. The quality of the included studies was assessed using the PROBAST+AI tool. A bivariate random effects model was used to pool sensitivity, specificity, and area under the curve (AUC) statistics. RESULTS: A total of 28 studies were included, with 2623 patients in internal validation and 1134 patients in external validation. Diagnostic contingency tables were reconstructed from published performance metrics for most studies, while others were extracted from receiver operating characteristic curves due to the lack of direct reporting. In the internal validation set, pooled sensitivity was 0.80 (95% CI 0.73-0.86; prediction interval [PI] 0.48-0.99), specificity was 0.83 (95% CI 0.77-0.89; PI 0.49-1.00), and AUC was 0.86 (95% CI 0.83-0.89; PI 0.74-0.99). In the external validation set, pooled sensitivity was 0.82 (95% CI 0.72-0.91; PI 0.54-0.99), specificity was 0.83 (95% CI 0.71-0.92; PI 0.49-1.00), and AUC was 0.84 (95% CI 0.79-0.90; PI 0.70-0.98). No statistically significant differences were observed between internal and external validation in sensitivity (P=.73), specificity (P>.99), AUC (P=.53), or diagnostic odds ratio (P=.98). Medical Net and Extreme Gradient Boosting achieved the highest sensitivity and AUC, whereas multiple kernel learning and support vector machine had the highest specificity. Subgroup and meta-regression analyses suggested that AI method, model type, timing of MRI acquisition, and treatment modality may contribute to heterogeneity. CONCLUSIONS: This meta-analysis innovatively realizes the quantitative direct comparison of MRI-based AI model performance between internal and external validation cohorts for PCa BCR prediction. It comprehensively evaluates AI performance across diverse PCa treatment modalities and integrates machine learning and deep learning approaches. For the field, it identifies key sources of performance heterogeneity (eg, MRI acquisition timing and treatment modality) and quantifies the sensitivity-specificity trade-off in integrated radiomic-clinical models, advancing the systematic understanding of MRI-based AI for BCR prediction. In real-world practice, it provides actionable guidance to prioritize pretreatment MRI for AI model development and clinical BCR assessment and underscores the urgent need for standardized imaging protocols and prospective multicenter studies, laying a foundation for the safe clinical translation of these AI tools as adjunctive decision support instruments.

References

1. 1.Jin, Y., Yuan, T., & Chen, Z. (2026). Magnetic resonance imaging-based artificial intelligence in predicting prostate cancer biochemical recurrence: Systematic review and meta-analysis. Journal of Medical Internet Research. https://doi.org/10.2196/85360