AI Can Change Your Treatment Plan—and Maybe Your Life

What’s the secret to the perfect treatment plan for prostate cancer (PCa)? It’s YOU, it’s YOUR tumor and YOUR anatomy. No two patients have identical disease in identical bodies. Thus, the perfect treatment plan is the one tailor-made for YOUR PCa.

Therefore, it’s essential that your doctor has as much accurate information as possible about the contours (location, size, shape) and risk level of your individual PCa. The tools your doctor has are clinical features based on PSA, biopsy, genomic analysis (if one was done), and multiparametric MR (mpMRI)I. When all information is pooled, it gives your doctor what he needs to suggest one or more treatment options for you. For many patients, their menu of options will include focal therapy.

Focal therapy is the most tailored treatment available because it selectively destroys the PCa tumor (plus an extra preemptive safety margin) while preserving healthy tissue. The key to an effective focal treatment is accurately identifying each patient’s clinical features to determine if his tumor is amenable to a focal ablation.

Artificial Intelligence can help

A May 8, 2024 press release announced, “Avenda Health, an AI healthcare company creating the future of personalized prostate cancer care, unveils the results of a new UCLA study examining the role of AI in identifying tumor margins and predicting the efficacy of focal therapy treatments.” Avenda Health has developed an AI software platform called Unfold AI designed to support treatment decision-making.

This platform was tested by a team led by researchers at UCLA. The results by Mota, et al. (2024) were published in the July issue of the Journal of Urology.^[i] The study involved 50 prostatectomy cases that were retrospectively determined to have been focal therapy candidates based on their pre-surgery clinical factors and their post-surgery prostate specimens. (The actual prostate specimens were used as the “ground truth” to compare image readers’ identification of tumors).

The readers included 7 urologists and 3 radiologists from 5 institutions. Their experience levels varied from 2-23 years. Each reader examined the 50 MRI prostate scans and biopsy reports. According to the study, “First, readers defined cancer contours cognitively, manually delineating tumor boundaries to encapsulate all clinically significant disease. Then, after ≥ 4 weeks, readers contoured the same cases using AI software.”^[ii]

The readers’ cognitive contours and AI-assisted contours were compared with the actual tumor boundaries from the surgical specimen slides. The comparisons were used to evaluate the accuracy of the cognitive vs. AI-assisted contours. What do you think they found? If you believe that AI improved identification of tumor size and shape, you are correct. The research demonstrated significant improvement in contouring and achieving negative tumor margins (no tumor extending beyond the imaging contour). The authors’ optimism over more accurate patient management was justified.

A June 18 Renal&Urology news item summarized a survey report saying that “…the AI software led urologists to recommend focal therapy more often and radical prostatectomy less frequently,” a clear indication that AI has a clear role to play in helping doctors and patients change from a whole-gland treatment to a less aggressive focal therapy. Hopefully, this bodes well for future PCa patients who are drawn to the idea of focal treatment.

NOTE: This content is solely for purposes of information and does not substitute for diagnostic or medical advice. Talk to your doctor if you are experiencing pelvic pain, or have any other health concerns or questions of a personal medical nature.

[i] Mota SM, Priester A, Shubert J, Bong J et al. Artificial Intelligence Improves the Ability of Physicians to Identify
Prostate Cancer Extent. J Urol. 2024 Jul;212(1):52-62.
[ii] Ibid.