Abstract Code: IUC24413-79

Comparative Performance of AI vs Radiologists in Pre-Biopsy mpMRI Prostate Cancer Diagnosis: A Systematic Review and Meta-Analysis of Multi-Centre, Multi-Vendor Studies 

Uzair Khan¹, Aaditya Tiwari², Karran Bhagat³, Tulika Nahar⁴, Soumya Arun⁵, Aruni Ghose⁶, Abel Tesfai⁷, Alexandra Naranjo⁷, Pinky Kotecha⁵, Nicolas Omorphos⁸, Joecelyn Kirani Tan⁹, Maryam Hasanova¹⁰, Akash Maniam¹¹, Giuseppe Luigi Banna¹², Yüksel Ürün¹³, Swarupa Mitra¹⁴, Karan Jatwani¹⁵, Stergios Boussios¹⁶, Prasanna Sooriakumaran¹⁷, Benjamin Lamb¹⁸, Jeremy Teoh¹⁹, Antony Rix²⁰, Amy Rylance⁷, Balraj Dhesi²¹, Atif Khan²², Sola Adeleke²³

1. Early Cancer Institute, University of Cambridge, Cambridge, UK; 2. Princess Alexandra Hospital NHS Trust, Harlow, UK; 3. Royal Surrey NHS Foundation Trust, Guildford, UK; 4. Queen’s University, Belfast, UK; 5. Queen Mary University of London, London, UK; 6. Barts Cancer Centre, London, UK; 7. Prostate Cancer UK, London, UK; 8. University Hospitals of North Midlands NHS Trust, Stoke-on-Trent, UK; 9. University of Manchester, Manchester, UK; 10. OncoFlowTM, London, UK; 11. Caribbean Cancer Research Institute, Trinidad and Tobago; 12. Portsmouth Hospitals University NHS Trust, Portsmouth, UK; 13. Ankara University Cancer Research Institute, Ankara, Turkey; 14. Fortis Cancer Institute, Gurugram, India; 15. George Washington Cancer Centre, Washington DC, USA; 16. University of Ioannina, Ioannina, Greece; 17. University of Oxford, Oxford, UK; 18. North East London Cancer Alliance, London, UK; 19. Chinese University of Hong Kong Medical Centre, Hong Kong; 20. Lucida Medical Ltd, Cambridge, UK; 21. University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK; 22. Leeds Teaching Hospitals NHS Trust, Leeds, UK; 23. King’s College London, London, UK.

Background: Multiparametric MRI (mpMRI) is the cornerstone for diagnosing clinically significant prostate cancer (csPC) i.e., International Society of Urological Pathology (ISUP) Grade >=2. The European Association of Urology (EAU) recommends its upfront implementation in biopsy-naive patients to guide prostate biopsy decision making (PMID 38614820). However, significant inter- and intra-observer reporting variability affects patient outcomes. Our objective was to systematically review and evaluate the comparative performance of AI vs radiologists in diagnosing prostate cancer from pre-biopsy prostate mpMRI.

Methods: Our systematic review [PROSPERO ID: CRD420251037432] included MEDLINE, PMC, EMBASE, SCOPUS and COCHRANE databases. Searching primary research published in 2010 and beyond yielded 6,389 records. After screening 3,747 articles, 137 underwent full-text review, with 3 meeting inclusion criteria—2 from database searches and 1 from grey literature. A meta-analysis using R assessed the diagnostic performance of the AI models, with Area Under the Curve (AUC) as the primary outcome.

Results: 3 multi-centre, multi-vendor intervention studies on prostate mpMRI compared AI characterisation of csPC vs standard of care (SOC), i.e., >=2 radiologists performing Prostate Imaging-Reporting and Data Systems (PI-RADS) version >=2 scoring. A meta-analysis, including 552 pre-biopsy patients, yielded: pooled sensitivity 0.884 (95% CI: 0.75-0.98), specificity 0.681 (95% 0.51- 0.80) and AUC 0.837 (95% CI: 0.690-0.950) for the AI – principally supervised machine learning (ML) models.

Model 1 (PMID 40016318) had a 95% sensitivity, 67% specificity and was non-inferior to SOC (AUC 0.91 vs 0.95; p=0.044). Model 2 (PMID 37345961) reported 86-91% sensitivity, 64-75% specificity and was also non-inferior to SOC (comparable AUCs 0.82 – 0.86). Model 3 (PMID 33671533) showed 89% sensitivity and superiority over SOC (AUC 0.75 vs 0.47). Models 1 and 2 exhibited strong generalisability, with Model 2 aligning closely with PI-RADSv2 lesion characterisation. 

Conclusions: The mpMRI-directed prostate biopsy pathway increases csPC detection and decreases PC negative biopsy rates. Adopting this implies a significant time and labour-intensive radiology workforce pressure. Our meta-analysis demonstrates how AI PC diagnostic accuracy is comparable to radiologists. This aids standardisation, reduces diagnostic variability and radiologist workload.