"I think the main take-home message from our study is that this is a really powerful prognostic prediction tool," says Charles Parker, MD.
In this video, Charles Parker, MD, shares the take-home message from the study “External validation of a digital pathology-based multimodal artificial intelligence (MMAI)-derived model in high-risk localized (M0)/metastatic (M1) prostate cancer (PCa) starting androgen deprivation therapy (ADT) in the docetaxel (Doc) or abiraterone (AAP) phase III STAMPEDE trials,” which was presented at the 2023 European Society for Medical Oncology Annual Congress in Madrid, Spain. Parker is a clinical research fellow and PhD student at UCL Medical College in London, the United Kingdom.
What is the take-home message of this study for the practicing urologist?
I think the main take-home message from our study is that this is a really powerful prognostic prediction tool, which can be implemented alongside current clinical decision-making to provide additional information to help drive those difficult conversations with patients about where you're headed with treatment. What we found in this study is we've benchmarked the current clinical decision-making tools, which, broadly speaking, in the non-metastatic patients, is nodal status, so whether the disease has spread to the lymph nodes...and then in the metastatic setting, what was shown to be prognostic in previous work is what we call metastatic disease burden. So this is using the same definitions as the CHAARTED and LATITUDE trials to divide patients into low-volume metastatic disease and high-volume metastatic disease. For the M0 patients, we used those clinical stratifiers to look at what the estimated prostate cancer-specific mortality was, and then applied the MMAI score generated by this study and looked at how these patients could then be re stratified. What we found was that you could identify the highest risk group of the patients in the node-negative setting, had a similar outcome to the lowest risk MMAI score patients in the next category up, so in the node-positive setting, and likewise, in the metastatic setting, we found that adding the score to the M1 lows identified a subset of patients whose behavior was quite close to the lower scores of the M1 high, so we're able to identify patients who, under traditional clinical decision-making, might be missed, to some extent, who potentially have got much higher prognostic outcome risks than you'd otherwise suspect.
This transcription was edited for clarity.