MRI, AI, and Urine Tests Reshape Prostate Cancer Detection and Treatment

New studies and technologies presented in 2026 are reshaping prostate cancer care, from screening and diagnosis to treatment selection. The developments include broader use of MRI before biopsy, a urine test for active surveillance, and artificial intelligence tools that predict treatment benefit and radiation dose. Together, they aim to reduce unnecessary biopsies and treatments while better personalizing care for men with prostate cancer.

Prostate cancer is the most frequently diagnosed cancer in the U.S., with more than 333,000 new cases each year, but only about 4% of cases end up being lethal and the 5-year survival rate is about 99%. The standard two-step diagnostic process—a PSA blood test followed by systematic biopsy if PSA is elevated—is effective at detecting cancer but may be too effective: up to 70% of new diagnoses are low-grade and in no need of treatment, yet as many as 50% or more of men with clinically insignificant disease get treated in the U.S., according to data from the American Urological Association Quality Registry. A good number of prostate cancer experts argue that these low-grade lesions “shouldn’t even be called a cancer,” said a professor of urology at the University of California, Los Angeles. In the last 15 years, MRI scans of the prostate have emerged as the preferred test after elevated PSA and before a potential biopsy, and between 30% and 50% of patients can avoid a biopsy by doing an MRI. Pre-biopsy MRI is the standard of care in Europe, Canada, Australia, and the U.K., and has been included in U.S. guidelines since 2020; the National Comprehensive Cancer Network strongly recommends it, while the American Urological Association and ASCO give it a conditional recommendation even though they acknowledge it is backed by high-level evidence. Despite this, MRIs were used ahead of biopsies in only about a third of U.S. cases in 2022. The 2018 PRECISION trial showed more cancer is found and fewer men diagnosed with clinically insignificant disease if an MRI is done before biopsy. “It’s one of the world’s worst global scandals that the richest country in the world denies [most] of its men access to something that is proven at level one evidence and that the rest of the world has adopted,” said a professor of interventional oncology at University College London.

A separate study published in the Southern Medical Journal found that very few middle-aged men are discussing prostate cancer screening with their doctor, even though guidelines recommend this talk for all men between 55 and 69. Only about 6% of men had a documented discussion with their primary care doctor. Researchers reviewed medical records for 600 men aged 45 to 69 treated at Medical University of South Carolina family medicine clinics in 2019 and 2020. When a discussion did take place, nearly 72% of all men and 85% of Black men received a PSA test, whereas only about a third (36%) got a PSA test when no discussion occurred. Screening doubled among men who talked it over with their doctor compared with those who didn’t. “One of the biggest takeaways was the surprisingly low rate of documented shared decision-making conversations,” said the lead researcher, a family medicine doctor at MUSC. He noted that newer tools like non-invasive prostate MRI and active surveillance have changed the landscape, and a recent uptick in prostate cancer cases means it is more important than ever for men to discuss screening.

Another study, published online April 28 in The Journal of Urology, found that a urinary biomarker test could reduce the need for biopsies and serial imaging in men undergoing active surveillance for low-grade prostate cancer. Researchers at Vanderbilt University Medical Center developed and validated a non-digital rectal examination urine test called MyProstateScore 2.0-Active Surveillance (MPS2-AS). In 330 patients with grade group 1 cancer scheduled for an active surveillance biopsy, the test provided a higher area under the curve than multiparametric MRI for upgrading to both GG≥3 (0.82 versus 0.73) and GG≥2 (0.74 versus 0.64). Prebiopsy MPS2-AS would have avoided nearly two-thirds (64%) of unnecessary biopsies while failing to detect only 3.2% of GG≥3 upgrades, compared with use of Prostate Imaging Reporting and Data System ≥3, which would have failed to detect 18% of GG≥3 upgrades and avoided 50% of unnecessary biopsies. On biopsy, 9.4% of patients upgraded to GG≥3 and 37% to GG≥2. The test’s performance was consistent across clinically pertinent subgroups, including Black and non-Black patients. “For patients undergoing monitoring of low-grade prostate cancer, these findings suggest that use of the urine test can reduce the need for invasive biopsies without compromising prompt detection of higher-grade cancers that require treatment,” said a researcher at Vanderbilt University Medical Center.

A post-hoc biomarker analysis of two STAMPEDE phase 3 trials, published in Annals of Oncology on June 5, 2026, evaluated whether a multimodal artificial intelligence (MMAI) model using digital pathology and clinical variables could predict benefit from abiraterone in patients with non-metastatic very high-risk prostate cancer. The model integrates digitized hematoxylin and eosin-stained prostate biopsy images with clinical variables including age, PSA, and clinical tumor stage, and had previously been validated as a prognostic biomarker. Among 1,137 patients (583 assigned to long-term androgen deprivation therapy and 554 assigned to long-term androgen deprivation therapy with abiraterone-based treatment) with a median follow-up of 6.1 years, adding abiraterone improved 5-year metastasis-free survival from 77% to 83% in the overall cohort. In patients classified as MMAI very high-risk, abiraterone was associated with a significant improvement in metastasis-free survival (HR 0.47; 95% CI 0.31–0.70), increasing the estimated 5-year metastasis-free survival from 62% to 81%. In contrast, patients classified as MMAI standard high-risk had limited evidence of benefit from adding abiraterone (HR 0.83; 95% CI 0.63–1.09), with estimated 5-year metastasis-free survival of 82% versus 84%.

Another AI approach, presented at the Society of Nuclear Medicine and Molecular Imaging 2026 Annual Meeting, aims to predict radiation dose before therapy in metastatic castration-resistant prostate cancer. Researchers at University Hospital Southampton and the University of Southampton developed a machine learning model that uses pre-therapy 18F-PSMA PET/CT scans to estimate absorbed doses to tumors and healthy organs for 177Lu-PSMA radiopharmaceutical therapy. In a proof-of-concept study of nine patients with mCRPC contributing 57 tumors, 36 salivary glands, and 18 kidneys, the model combined uptake-based PET metrics, radiomic features, and clinical biomarkers to predict post-therapy dosimetry. If validated in larger studies, the approach may improve patient selection and support better decision-making during pre-treatment assessment.

Separately, The Institute of Cancer Research, London, and The Royal Marsden NHS Foundation Trust announced a licensing agreement with MedTech company Mint Medical to offer AI-powered software for monitoring bone disease in advanced prostate cancer and multiple myeloma. The AI algorithm, developed by researchers and clinicians at the ICR and Royal Marsden, has been integrated into Mint Medical’s mint Lesion software, which uses diffusion-weighted MRI to pinpoint areas of bone disease and assess whether the disease is getting better or worse with treatment. The agreement allows Mint Medical to offer the software to healthcare providers globally.