An interventional study of AI-based prostate MRI analysis software in Prostate Cancer, sponsored by Fondazione Policlinico Universitario Agostino Gemelli IRCCS. Not yet recruiting at 1 site in Italy. Open to male participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-10-15.
Sponsored by Fondazione Policlinico Universitario Agostino Gemelli IRCCS · Not applicable, Interventional, and Diagnostic
The goal of this study is to learn how accurately an artificial intelligence (AI) software called Prostate Intelligence (Pi) detects clinically significant prostate cancer on prostate MRI in adults with a prostate gland who are referred for MRI because prostate cancer is suspected.
The main question it aims to answer is:
- How sensitive and specific is Pi, used on its own, in detecting clinically significant prostate cancer (Gleason grade 3+4 or higher) at the patient level?
Researchers will also compare Pi, the radiologist alone, and the radiologist assisted by Pi, to see whether AI assistance changes cancer detection.
Participants will:
About 226 participants will be enrolled at two Italian centres (Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, and IRCCS Ospedale San Raffaele, Milan).
Background and rationale Prostate cancer is the most common cancer in men in Europe. International guidelines recommend multiparametric MRI before biopsy, and findings are graded with PI-RADS v2.1. PI-RADS assessment varies with institution, reader experience and field strength, and inter-reader variability is substantial among less experienced radiologists. Prostate Intelligence (Pi, Lucida Medical Ltd) is an AI software that analyses prostate MRI and provides prostate volume, lesion detection/segmentation, an AI-Likert score and a preliminary report as a concurrent read to support the radiologist. This study evaluates the diagnostic performance of Pi in a prospective clinical setting.
Design Prospective, multicentre (national), interventional diagnostic accuracy study with a medical device. The software runs on servers of the participating institutions; DICOM data are de-identified and imaging data do not leave the institutional perimeter.
Reading workflow For each examination the radiologist first reports the MRI blinded to Pi; the report is locked and time-stamped in RIS/PACS. The Pi output is then made available and the radiologist produces a merged Pi-assisted report, recording PI-RADS v2.1 categories, lesion zone and whether each Pi finding is accepted or rejected, and adding any lesions considered missed by Pi.
Reference standard Histopathology from MRI-targeted biopsy (maximum two lesions, including discordant radiologist/Pi lesions) and, where available, radical prostatectomy. Participants who are not biopsied are followed for 12 months after MRI (PSA at 6 and 12 months plus clinical assessment) and classified as composite true negatives when there is no evidence of clinically significant cancer. Clinically significant prostate cancer (csPCa) is defined as Gleason grade 3+4 or higher.
Endpoints Co-primary: patient-level sensitivity and specificity of stand-alone Pi for csPCa (Gleason >=3+4).
Secondary (all exploratory): diagnostic performance of the Pi-assisted read at patient level for Gleason >=3+4 and >=4+3; comparison of Pi-assisted vs radiologist-alone and Pi vs radiologist-alone; proportion of csPCa detected and additional csPCa detection with Pi; lesion-level performance of the Pi-assisted read; exploratory cost-consequence analysis; subgroup performance in patients with radical prostatectomy histology.
Sample size 226 participants (150 at Gemelli, 76 at San Raffaele), based on 7.5% precision at a 5% significance level for an expected Pi sensitivity of 90% and specificity of 65%, with an assumed csPCa prevalence of 31% (R v4.4.2). Assumptions are conservative relative to published validation of the device. The study is powered for stand-alone accuracy only; comparative endpoints are exploratory.
Statistical analysis Sensitivity, specificity, PPV, NPV and ROC AUC with 95% CIs. AUCs of radiologist-alone, Pi-alone and Pi-assisted reads compared with the DeLong method, with bootstrapped two-sided 95% CIs. Lesion-level metrics (PI-RADS >=3 as positive) use generalised estimating equations to account for clustering. The primary analysis includes histologically verified participants and composite true negatives; pre-specified sensitivity analyses are restricted to histologically verified participants and a bounding analysis assigning unverified participants first to non-csPCa and then to csPCa. Missing data are tabulated and the primary analysis uses complete cases. Exploratory comparative p-values are reported unadjusted and Benjamini-Hochberg adjusted (5%). Centre, scanner model and field strength are recorded and explored as subgroups; no centre contributes more than 70% of participants. Reporting follows STARD.
Duration 27 months: 12 months enrolment and data collection, up to 12 months follow-up per participant, 3 months analysis.
Safety All adverse events are recorded; device incidents and deficiencies are reported to the National Competent Authority and the manufacturer under EU Regulation 2017/745 and Italian Legislative Decree 137/2022.
6,484 studies on the registry are indexed under Prostatic Neoplasms; 1,415 are open to participants now.
This study's planned enrollment of 226 is above the median of 58 across 4,908 interventional studies indexed under Prostatic Neoplasms.
Browse Prostatic Neoplasms studies →Fondazione Policlinico Universitario Agostino Gemelli IRCCS is the lead sponsor of 948 studies on the registry; 537 are open to participants now.
Counted across the registry records on this site, refreshed daily.
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
Exclusion Criteria:
Adults with a prostate gland referred for MRI for suspected prostate cancer. Each MRI is assessed by the radiologist alone (blinded to Pi), by Pi stand-alone, and by the radiologist assisted by Pi. Participants with suspicious lesions undergo MRI-targeted biopsy of up to two lesions (radiologist- and/or Pi-identified), with the order of lesion sampling randomised 1:1. Participants not biopsied are followed for 12 months after MRI with PSA at 6 and 12 months and clinical assessment.
Device: AI-based prostate MRI analysis software
Prostate Intelligence (Pi, Lucida Medical Ltd) is an artificial intelligence software that analyses prostate MRI and provides prostate volume, lesion detection and segmentation, an AI-Likert score and a preliminary report. It runs on the participating institutions' servers on de-identified DICOM data. Each MRI is first reported by the radiologist blinded to Pi (report locked and time-stamped); Pi output is then revealed and the radiologist issues a merged Pi-assisted report (PI-RADS v2.1), accepting or rejecting each Pi finding. Up to two lesions per participant (radiologist- and/or Pi-identified) undergo MRI-targeted biopsy, with sampling order randomised 1:1. Non-biopsied participants are followed for 12 months (PSA at 6 and 12 months, clinical assessment).
Also known as: Prostate Intelligence (Pi)
Patient-level sensitivity of Pi stand-alone for clinically significant prostate cancer (Gleason score >=3+4)
Proportion of participants with clinically significant prostate cancer (csPCa, Gleason score \>=3+4 / ISUP grade group \>=2) correctly identified as positive by Prostate Intelligence (Pi) stand-alone at a threshold fixed before study start. Reference standard: histopathology from MRI-targeted biopsy or radical prostatectomy, or composite true negative status (no csPCa on 12-month clinical and PSA follow-up) for non-biopsied participants. Reported with 95% confidence interval; NPV, PPV and ROC AUC also reported.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Patient-level specificity of Pi stand-alone for clinically significant prostate cancer (Gleason score >=3+4)
Proportion of participants without clinically significant prostate cancer (csPCa, Gleason score \>=3+4 / ISUP grade group \>=2) correctly identified as negative by Prostate Intelligence (Pi) stand-alone at a threshold fixed before study start. Reference standard: histopathology from MRI-targeted biopsy or radical prostatectomy, or composite true negative status (no csPCa on 12-month clinical and PSA follow-up) for non-biopsied participants. Reported with 95% confidence interval.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Patient-level sensitivity and specificity of the Pi-assisted radiologist read for csPCa (Gleason score >=3+4)
Sensitivity and specificity of the radiologist's merged report after review of the Pi output, for csPCa defined as Gleason score \>=3+4. Reference standard: histopathology from MRI-targeted biopsy or radical prostatectomy, or composite true negative status (no csPCa on 12-month clinical and PSA follow-up) for non-biopsied participants. Exploratory.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Patient-level sensitivity and specificity of the Pi-assisted radiologist read for Gleason score >=4+3 cancer
Sensitivity and specificity of the Pi-assisted read for prostate cancer of Gleason score \>=4+3 (ISUP grade group \>=3). Reference standard: histopathology from MRI-targeted biopsy or radical prostatectomy, or composite true negative status (no csPCa on 12-month clinical and PSA follow-up) for non-biopsied participants. Exploratory.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Diagnostic performance of Pi-assisted read versus radiologist-alone read (ROC AUC)
Difference in patient-level area under the ROC curve between the Pi-assisted read and the radiologist-alone read (blinded to Pi), for csPCa (Gleason score \>=3+4). Compared with the DeLong method; two-sided 95% CIs by bootstrapping. Sensitivity, specificity, PPV and NPV also compared. Exploratory.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Diagnostic performance of Pi stand-alone versus radiologist-alone read (ROC AUC)
Difference in patient-level area under the ROC curve between Pi stand-alone and the radiologist-alone read for csPCa (Gleason score \>=3+4). Compared with the DeLong method; two-sided 95% CIs by bootstrapping. Exploratory.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Proportion of csPCa detected by Pi-assisted and Pi stand-alone reads versus radiologist-alone read
Proportion of participants with csPCa (Gleason score \>=3+4) detected by the Pi-assisted read and by Pi stand-alone, compared with the proportion detected by the radiologist-alone read. Exploratory.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Lesion-level sensitivity and positive predictive value of the Pi-assisted read for csPCa
Lesion-level sensitivity and PPV of the Pi-assisted read for csPCa (Gleason score \>=3+4), with PI-RADS \>=3 considered positive; lesions matched to histopathology. Generalised estimating equations account for clustering of multiple lesions within a participant. Exploratory.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Additional csPCa detected with Pi compared with radiologist-alone read
Number and proportion of csPCa (Gleason score \>=3+4) lesions/participants detected only through Pi (Pi-identified lesions not reported by the radiologist alone) and confirmed on histopathology. Exploratory.
Time frame: Index test at baseline MRI; reference standard up to 12 months after MRI
Diagnostic performance of Pi stand-alone and Pi-assisted read against radical prostatectomy histopathology
Sensitivity, specificity and lesion-level detection of Pi stand-alone and the Pi-assisted read in the subgroup of participants who undergo radical prostatectomy, using whole-mount prostatectomy histopathology as the reference standard. Exploratory subgroup analysis.
Time frame: Up to 12 months after MRI
Plan to share: No — Individual participant data will not be shared with external researchers. Under the approved consent to data processing, personal data may be communicated only between the participating centres where necessary to conduct the study, and will not be disseminated except in anonymous and/or aggregate form. Imaging data are de-identified and processed on institutional servers. Aggregate results will be published in peer-reviewed journals according to STARD.
No publications or documents are linked to this record.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
This study is not yet recruiting, as verified in Oct 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
Fondazione Policlinico Universitario Agostino Gemelli IRCCS →