Patients with unfavorable intermediate-risk (UIR) prostate cancer represent a heterogeneous group, comprised of individuals with distinct constellations of prostate cancer risk factors. As seen at the 2024 ASTRO Annual Meeting, the framework for clinical decision making not only includes classical risk factors but also extends to include genomic classifiers, prostate-specific membrane antigen positron emission tomography (PSMA PET), multiparametric magnetic resonance imaging (MRI), and artificial intelligence (AI)–based tools.

Following these proceedings, featured expert Oliver Sartor, MD, was interviewed by Conference Reporter Editor-in-Chief Tom Iarocci, MD. Dr Sartor’s clinical perspectives on these findings are presented here.

The classically defined UIR category is a particularly heterogeneous group that is comprised of people with various risk factors, whether this includes a Gleason score of 4 + 3 or 2 or 3 of the intermediate risk factors, such as a prostate-specific antigen (PSA) level of 10 to 20 ng/mL. We can classify these individuals in more detail with the genomic classifiers, which I think are going to be adopted more in the future, and I did see some polling of the audience at ASTRO 2024 suggesting that a fair number of people are already using these tools. During a presentation by Neil Desai, MD, Decipher was the most frequently used genomic classifier among the people polled in the audience, and Prolaris was also used by those in attendance. I think that Decipher might be able to play an important role in risk stratification that goes beyond the Gleason score.

And, very interestingly, the multimodal AI (MMAI) biomarker test (ie, ArteraAI) is being used by a small percentage of people, but this number will grow. The MMAI tool has some interesting classifications with regard to the responsiveness or nonresponsiveness to hormonal therapy, and it appears to have predictive value, which is exciting. Now, we are still learning how this applies to individual patients, but, when I classify patients with UIR disease, I want to think about transcriptomics and genomic classifiers. I am thinking more and more about MMAI in my classification scheme.

I also want to point out something that Daniel Spratt, MD, noted in his presentation at ASTRO 2024, and that is that we really need to incorporate imaging. I do get a PSMA PET for many of these patients because we can learn from it, and, often, we upstage the patient as a result, which can affect our treatment. Another very important element is the multiparametric MRI. I get an MRI for every single one of these patients. We can understand the local staging from an MRI even better than we can from a PSMA (eg, things such as extracapsular extension and seminal vesicle invasion). The ability of the PSMA PET to pick up metastatic lesions is really superior. For local lesions, we want to look at the MRI.

The bottom line is that imaging is improving with MRI and PSMA PET; genomics are improving, particularly with the genomic classifiers; and AI is now applicable. I do look at parameters that go beyond the typical Gleason score, number of positive biopsies, digital rectal examination, staging, and PSA. And I am extending those conventional parameters to make better decisions for my patients.

In terms of definitive radiotherapy for localized UIR prostate cancer, I think that there is an evolution toward shorter fractions. Mayo Clinic is using predominantly moderate hypofractionation. Now, when you are using stereotactic body radiotherapy (SBRT), I think that you have to be quite careful because there is the possibility of some additional acute and long-term toxicity, so I tend to want to see the longer-term follow-up because these patients are going to be living a long time.

We can also consider the possibility of using a brachytherapy boost, and that can either be high-dose rate or low-dose rate brachytherapy. In addition, the FLAME trial incorporated a microboost of external beam radiotherapy according to an MRI-defined dominant lesion. There are multifactorial analyses that would indicate that there are many different options available for patients in this category.

As seen at ASTRO 2024, there are a number of radiotherapy options for these patients, and I think that it is probably most appropriate to discuss these options with the patient and to be in your comfort zone. These are radiation oncology discussions, but, whether you consider moderate hypofractionation, conventional fractionation, SBRT, or using a boost with either brachytherapy or external beam radiotherapy, all of these are within the bandwidth of possibility and are supported, in my mind, by contemporary data.