ArteraAI, the developer of multimodal artificial intelligence-based prognostic and predictive cancer tests, announced the validation of its first multimodal artificial intelligence (MMAI) digital pathology-based post-radical prostatectomy biomarker for stratifying risk of metastasis and identifying differential absolute benefit for the addition of hormone therapy to salvage radiation therapy in radical prostatectomy patients with biochemical recurrence (BCR).
Researchers successfully trained and validated the model for patients who experience BCR after radical prostatectomy surgery, which is strongly associated with the risk of the disease spreading. The data was presented by Dr. Todd Morgan from Michigan Medicine as an oral presentation at the 2024 AUA Annual Meeting, on Sunday, May 5, 2024.
A radical prostatectomy is the removal of the prostate in a patient with prostate cancer and has served as a primary treatment for this disease for many years. However, about 20-40% of patients who undergo the procedure eventually develop BCR, which is an indicator of the cancer returning. In this event, salvage radiotherapy is usually recommended and clinicians are also faced with the decision of whether to further intensify the treatment by combining the salvage radiotherapy with hormone therapy.
“There is a continued need to optimize treatment for men with BCR post-radical prostatectomy to avoid both over- and under-treatment,” said Morgan, urological surgeon and Chief of Urologic Oncology at Michigan Medicine. “In the near future, clinicians and patients may be able to use this digital pathology-based biomarker to better understand the risk of further disease progression and inform the treatment plan for these patients in particular.”
For the study, researchers aimed to develop a novel biomarker for patients with BCR post-radical prostatectomy using multimodal deep learning on digital histopathology to stratify the risk of distant metastasis. Data from two Phase 3 randomized trials, NRG/RTOG 9601 and 0534, were used for development and validation. Clinical and histopathological data were available for 1,855 of the 2,573 eligible patients (72.1%). The training cohort included 1,322 men and the validation cohort included 533 men. Digitized images of prostatectomy tumor samples were utilized to develop an MMAI model to estimate the risk of distant metastasis and identify benefits from adding hormone therapy to salvage radiation therapy.
“We’re thrilled to have had the opportunity for our data to be presented at AUA,” said Andre Esteva, Co-founder and CEO of ArteraAI. “These data exemplify how ArteraAI plans to extend its platform in the future for patients who have undergone a radical prostatectomy.”
Prior to this research, no AI-based models had been developed in cohorts of radical prostatectomy patients. This is the first MMAI digital pathology-based post-radical prostatectomy biomarker that has successfully been trained and validated using data from Phase 3 clinical trials of men post-prostatectomy to show prognostic capability for all tested endpoints as well as differential absolute benefit for the addition of hormone therapy to salvage radiation therapy.
Source: BusinessWire