Artificial intelligence (AI) is beginning to play a role in prostate cancer care, from helping predict which patients may benefit from additional treatment to supporting clinical decision-making. Prostate cancer specialist Dr. Andrew Hahn discusses the potential for AI to guide more personalized treatment decisions for patients.
Dr. Andrew Hahn is a Genitourinary Medical Oncologist and an Assistant Professor of Genitourinary Medical Oncology at The University of Texas MD Anderson Cancer Center. Learn more about Dr. Hahn.
Related Resources
Transcript
Katherine Banwell:
How could or how is artificial intelligence and other emerging technologies improving cancer detection, diagnosis, and treatment planning?
Dr. Andrew Hahn:
Like everywhere, it’s introducing into the field now. And I suspect in two or three years it’s going to massively change how we’re approaching. But prostate is fortunate in that I believe the first companion diagnosis that used artificial intelligence, ArteraAI, was approved for men with prostate cancer. So, what this test is doing in its initial approval is that for men who have localized prostate cancer, who are going to be getting radiation, there’s always a question of should we add a short duration of hormone therapy or not.
And someone’s radiation oncologist or oncologist could just order an ArteraAI score, which looks at the image of the biopsy that was obtained and some additional information and uses its AI algorithm to predict whether men would benefit from getting hormone therapy or not benefit.
So, what you end up doing is you take a whole group of men where we previously would have said let’s just give hormone therapy because these are their disease characteristics. And we know that they won’t benefit, so we don’t recommend it for those men. So, we avoid harm when there was no benefit that was going to be obtained. So, that’s one example.
That tool is also now being deployed in many different ArteraAIs. The tool is now being deployed in other settings in advanced prostate cancer to help us understand should we intensify or de-intensify our treatment. That’s a first example of what it’s capable of. But I suspect that’s just the tip of the iceberg.
I already described for everybody how complex the molecular characterization of metastatic prostate cancer is and how we’re trying to make decisions with that. You could imagine that all of those variables could be pulled into a multimodal tool that uses AI to help there.
AI is also helping in the clinic. In-clinic people are using tools such as OpenEvidence to help whenever there are questions that are difficult to answer. There are scribe tools that are helping improve efficiencies and workflows, so more patients can be seen quicker and more effectively. I think there are a lot of opportunities here. Again, probably just the tip of the iceberg.
Katherine Banwell:
It’s an exciting time.
Dr. Andrew Hahn:
Absolutely.