New AI Tool Predicts Prostate Cancer Treatment Success
A recent study suggests that using artificial intelligence to measure tumor size could reduce treatment failures by more than 70%.
prostate cancer
A new artificial intelligence (AI)-powered software, Unfold AI, has demonstrated its ability to estimate tumor volume, which is strongly linked to biopsy outcomes following focal cryoablation. According to a study conducted by researchers at University of California, Los Angeles (UCLA) and published in BJUI Compass, larger tumor volumes are inversely correlated with treatment success rates.1
This innovative technology underscores the potential of AI to enhance treatment outcomes for prostate cancer patients. By providing physicians with tools to better predict which patients are most likely to benefit from partial gland cryoablation—a minimally invasive procedure targeting localized prostate tumors—AI could revolutionize personalized treatment planning.
Developed collaboratively by UCLA and Avenda Health, Unfold AI has proven effective in accurately measuring prostate tumor volume. This capability allows clinicians to identify patients with a higher likelihood of successful treatment outcomes, paving the way for more tailored and effective care strategies in community oncology settings.
“By using AI to measure the size of a man’s prostate tumor more precisely, we can better predict who is likely to be cured with focal therapies like partial gland cryoablation,” said Wayne Brisbane, MD, assistant professor of urology at the David Geffen School of Medicine at UCLA, member of the UCLA Health Jonsson Comprehensive Cancer Center, and first author of the study, in a press release.1
Developed collaboratively by UCLA and Avenda Health, Unfold AI has proven effective in accurately measuring prostate tumor volume. The technology analyzes data from MRI scans and biopsies and is able to produce a detailed, three-dimensional map of the prostate tumor, helping doctors to see the true size and boundaries of a patient's cancer more accurately. The AI was approved by the FDA in 2022.2
In the study, 204 men with prostate cancer of grade groups 2 to 4 were enrolled at UCLA from 2017 to 2022. Magnetic resonance imaging (MRI)-guided biopsy (MRGB) was conducted at time of diagnosis and at 6 and 18 months following partial gland ablation.
Researchers retrospectively estimated tumor volume for each patient. Then, the AI-derived tumor volume was compared with traditional baseline variables to assess its correlation with a successful primary end point, defined as the absence of Gleason grade 2 or greater cancer on follow-up MRGB at 6 months. Secondary end points assessed were MRGB results at 18 months and failure-free survival, which was measured by the absence of metastasis or the need for salvage whole-gland therapy.
The study also used receiver operating characteristic curves and multivariate analysis to evaluate the statistical significance of these findings.
Findings showed there to be a successful primary outcome in 77.7% of patients. Key factors associated with successful ablation included the percentage of pattern 4 cancer and tumor volume, with areas under the curve (AUCs) of 0.60 and 0.73, respectively. Gleason grade was not correlated with treatment success, with an AUC of 0.51.
A tumor volume of 1.5 cc was the optimal threshold, balancing sensitivity (55.8%) and specificity (85.7%) at the 6-month follow-up. Additionally, tumor volume showed to have a strong association with secondary outcomes, including biopsy results at 18 months and failure-free survival.
Further, multivariate analysis identified tumor volume as the most significant predictor of biopsy success at both 6 and 18 months, with adjusted odds ratios of 6.1 and 4.2, respectively. The study also suggested that using a tumor volume threshold smaller than 1.5 cc as a criterion for partial gland ablation could potentially prevent 72% of treatment failures.
Following these promising results, experts plan to evaluate the AI in larger, multicenter trials to further confirm the studies results.
“With Unfold AI, doctors now have a method to determine the volume of cancer within a prostate tumor,” said Leonard Marks, MD, professor and deKernion Endowed Chair in Urology at the David Geffen School of Medicine at UCLA, member of the UCLA Health Jonsson Comprehensive Cancer Center, and senior author of the study, in the press release.1 “Such a method has not been previously available. It's important because tumor volume is a major determinant of treatment success or failure. Using AI to predict tumor volume and shape gives a clearer picture and could help choose better candidates for focal cryotherapy.”
REFERENCES
Artificial intelligence tool helps predict who will benefit from focal therapy for prostate cancer. News release. UCLA Health. March 18, 2025. Accessed March 18, 2025. https://tinyurl.com/32zjpfaw
Brisbane WG, Priester AM, Nguyen AV, et al. Focal therapy of prostate cancer: Use of artificial intelligence to define tumour volume and predict treatment outcomes. BJUI Compass. 2024;6(1):e456. Published 2024 Nov 28. doi:10.1002/bco2.456
