MRI-Based Machine Learning for Prostate Cancer

Overseen By Ileana Aldana

Age: 18+

Sex: Male

Trial Phase: Academic

Sponsor: University of Southern California

Disqualifiers: Prostate cancer, Benign prostate hyperplasia, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Compensation: Varies

Number of Visits: 10

Duration: Up to 3 months

What You Need to Know Before You Apply

What is the purpose of this trial?

This trial evaluates the effectiveness of an AI system in detecting prostate cancer using MRI images, compared to traditional radiologist methods. The AI may match experienced radiologists in identifying aggressive prostate cancer. The trial includes several rounds of MRI scans and prostate biopsies to compare results. It targets individuals who have undergone or will undergo a specific type of MRI and are scheduled for a prostate biopsy or prostate removal due to cancer. As an unphased trial, this study allows participants to contribute to cutting-edge research that could enhance prostate cancer detection.

Do I need to stop my current medications for the trial?

The trial information does not specify whether you need to stop taking your current medications. It's best to discuss this with the trial coordinators or your doctor.

What prior data suggests that this MRI-based machine learning approach is safe for detecting prostate cancer?

Research has shown that using AI in MRI scans holds promise for detecting prostate cancer. Studies indicate that AI can help doctors identify aggressive prostate cancers more accurately, allowing for earlier detection when treatment is more manageable. The AI system used in these studies excels at identifying prostate cancer and can also rule out cancer in low-risk cases, helping to avoid unnecessary treatments.

Although this AI system is neither a drug nor a device that interacts with the body, it plays a crucial role in diagnosing cancer. The focus is on its effectiveness in identifying cancer, rather than any direct safety concerns for participants. Overall, using AI in prostate cancer MRI scans is considered safe, as it poses no direct risks to patients. ^{1 2 3 4 5}

Why are researchers excited about this trial?

Researchers are excited about the MRI-Based Machine Learning approach for prostate cancer because it offers a cutting-edge alternative to the traditional methods like the standard biopsy and imaging techniques. This new method leverages artificial intelligence (AI) to enhance the accuracy of prostate cancer detection. By integrating machine learning models with MRI and ultrasound, it provides a more precise mapping of cancerous tissues, potentially reducing the number of unnecessary biopsies. This approach aims to improve diagnosis by using AI predictions, which could lead to earlier and more accurate identification of aggressive cancers, ultimately improving patient outcomes.

What evidence suggests that this trial's MRI-based machine learning approach could be effective for detecting prostate cancer?

This trial will evaluate the effectiveness of an MRI-based machine learning approach for detecting prostate cancer. Studies have shown that using MRI scans with machine learning can effectively detect prostate cancer. One AI system outperformed over 70% of human experts in diagnosing prostate cancer from these scans, demonstrating greater accuracy than most individuals. Another study found that deep learning, a type of AI that learns patterns, accurately identified cancer cases, though it occasionally missed some true cases. Overall, these findings suggest that AI can be a valuable tool in diagnosing prostate cancer, potentially matching or even surpassing human performance in some areas. Participants in this trial will undergo various sequences of MRI/TRUS and targeted prostate biopsies using different AI predictions and methods, as part of the different study arms. ^{2 6 7 8 9}

Who Is on the Research Team?

Andre Luis Abreu, MD

Principal Investigator

University of Southern California

Are You a Good Fit for This Trial?

This trial is for men who are suspected of having prostate cancer and are undergoing MRI scans to detect it. The study is open to those who have not yet had a biopsy or treatment for prostate cancer.

Inclusion Criteria

PROSTATE BIOPSY COHORT: Patients who provided with consent for the study

I am scheduled for a prostate biopsy using MRI/TRUS fusion.

I had a specific MRI before my prostate biopsy.

Exclusion Criteria

PROSTATE BIOPSY COHORT: Patients with previous PBx history

PROSTATE BIOPSY COHORT: MRI which was not interpreted by PIRADS

PROSTATE BIOPSY COHORT: MRI with significant artifact

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Treatment

Patients undergo MRI/TRUS followed by targeted prostate biopsies using various AI predictions and PIRADS, with up to 12 additional biopsies per standard of care

Up to 3 months

Multiple visits for MRI/TRUS and biopsies

Follow-up

Participants are monitored for safety and effectiveness after treatment

3 months

Follow-up visits at 10 days and 3 months

What Are the Treatments Tested in This Trial?

Interventions

• MRI-Based Machine Learning Approach

Trial Overview The PRIMER Trial is testing an AI-driven machine learning approach that analyzes MRI images against the traditional method where radiologists review MRIs and assign PIRADS scores to identify potential prostate cancer.

How Is the Trial Designed?

Group I: Group 2 (prostate mold evaluation) Experimental Treatment 2 Interventions

Patients have their removed prostate evaluated using a special mold on study. Prostate tissue is mapped and compared with the prostate cancer prediction on MRI generated by radiologists and AI reports. Patients may also undergo DRE on study.

Group II: Group 1 Arm VI (MRI/TRUS, DL AI, PIRADS, GL AI) Experimental Treatment 4 Interventions

Patients undergo MRI/TRUS followed by a targeted prostate biopsy using DL AI predictions. Patients then undergo a 2nd MRI/TRUS followed by a targeted prostate biopsy using PIRADS. Patients then undergo a 3rd MRI/TRUS followed by a targeted biopsy based on GL AI predictions. Finally, patients undergo up to 12 additional prostate biopsies per SOC. Patients may also undergo DRE on study.

Group III: Group 1 Arm V (MRI/TRUS, DL AI, GL AI, PIRADS) Experimental Treatment 4 Interventions

Patients undergo MRI/TRUS followed by a targeted prostate biopsy using DL AI predictions. Patients then undergo a 2nd MRI/TRUS followed by a targeted prostate biopsy based on GL AI predictions. Patients then undergo a 3rd MRI/TRUS followed by a targeted biopsy using PIRADS. Finally, patients undergo up to 12 additional prostate biopsies per SOC. Patients may also undergo DRE on study.

Group IV: Group 1 Arm IV (MRI/TRUS, GL AI, DL AI, PIRADS) Experimental Treatment 4 Interventions

Patients undergo MRI/TRUS followed by a targeted prostate biopsy using GL AI predictions. Patients then undergo a 2nd MRI/TRUS followed by a targeted prostate biopsy based on DL AI predictions. Patients then undergo a 3rd MRI/TRUS followed by a targeted biopsy using PIRADS. Finally, patients undergo up to 12 additional prostate biopsies per SOC. Patients may also undergo DRE on study.

Group V: Group 1 Arm III (MRI/TRUS, GL AI, PIRADS, DL AI) Experimental Treatment 4 Interventions

Patients undergo MRI/TRUS followed by a targeted prostate biopsy using GL AI predictions. Patients then undergo a 2nd MRI/TRUS followed by a targeted prostate biopsy using PIRADS. Patients then undergo a 3rd MRI/TRUS followed by a targeted biopsy based on DL AI predictions. Finally, patients undergo up to 12 additional prostate biopsies per SOC. Patients may also undergo DRE on study.

Group VI: Group 1 Arm II (MRI/TRUS, PIRADS, DL AI, GL AI) Experimental Treatment 4 Interventions

Patients undergo MRI/TRUS followed by a targeted prostate biopsy using PIRADS. Patients then undergo a 2nd MRI/TRUS followed by a targeted prostate biopsy based on DL AI predictions. Patients then undergo a 3rd MRI/TRUS followed by a targeted biopsy based on GL AI predictions. Finally, patients undergo up to 12 additional prostate biopsies per SOC. Patients may also undergo DRE on study.

Group VII: Group 1 Arm I (MRI/TRUS, PIRADS, GL AI, DL AI) Experimental Treatment 4 Interventions

Patients undergo MRI/TRUS followed by a targeted prostate biopsy using PIRADS on study. Patients then undergo a 2nd MRI/TRUS followed by a targeted prostate biopsy based on GL AI predictions. Patients then undergo a 3rd MRI/TRUS followed by a targeted biopsy based on DL AI predictions. Finally, patients undergo up to 12 additional prostate biopsies per SOC. Patients may also undergo DRE on study.

Who Is Running the Clinical Trial?

University of Southern California

Lead Sponsor

Trials

956

Recruited

1,609,000+

National Cancer Institute (NCI)

Collaborator

Trials

14,080

Recruited

41,180,000+

Citations

1. nature.com nature.com/articles/s41391-025-00997-2

Machine learning-based MRI imaging for prostate cancer ... Machine learning-based MRI imaging demonstrated good diagnostic value for both benign/malignant prostate cancer and csPCa. The pooled ...

2. clinicaltrials.gov clinicaltrials.gov/study/NCT07162194

Study Details | NCT07162194 | MRI-Based Machine ... If a suspicious area is seen in the MRI, the radiologist assigns it a PIRADS score. This stands for Prostate Imaging Reporting and Data System.

3. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC11592433/

Systematic Review of AI-Assisted MRI in Prostate Cancer ... The AI-based system outperformed more than 70% of ordinary readers in the MRI-based diagnosis of prostate cancer. Hosseinzadeh et al., 2022 [30] ...

4. pubs.rsna.org pubs.rsna.org/doi/abs/10.1148/radiol.232788

MRI-based Deep Learning Algorithm for Assisting ... A commercial MRI-based deep learning algorithm for prostate cancer detection showed greater positive predictive value, despite its lower sensitivity.

5. sciencedirect.com sciencedirect.com/science/article/pii/S0010482525007917

Incorporating clinical data and uncertainty in MRI deep ... A bimodal AI model is proposed that combines MRI and clinical data for prostate cancer diagnosis. Integrating clinical features with imaging significantly ...

6. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC12154491/

A Narrative Review of Artificial Intelligence in MRI-Guided ... AI-based methods have demonstrated promising potential to enhance multiple stages of the MRI-guided prostate cancer PCa diagnostic pathway. The ...

7. nature.com nature.com/articles/s41391-025-01014-2

Clinical impact of MRI-based risk calculators for prostate ... Ultimately, MRI-based RCs could reduce overtreatment and enhance the detection of aggressive cancers at earlier and more treatable stages, ...

8. jamanetwork.com jamanetwork.com/journals/jamanetworkopen/fullarticle/2835296

AI-Assisted vs Unassisted Identification of Prostate Cancer ... AI assistance was associated with a statistically superior improvement in detecting csPCa, increasing the area under the receiver operating characteristic ...

9. sciencedirect.com sciencedirect.com/science/article/pii/S0720048X24005060

Assessment of a fully-automated diagnostic AI software in ... The AI tool mdprostate shows high accuracy in prostate cancer detection and grading using MRI. •. Allows to accurately rule out prostate cancer in low-risk ...