(UroToday.com) The 2024 American Society of Clinical Oncology (ASCO) annual meeting featured a session on prostate cancer, and a presentation by Dr. Philip Sutera discussing the validation of a digital pathology-based multimodal artificial intelligence model in oligometastatic castration-sensitive prostate cancer (CSPC). Oligometastatic CSPC is a state of limited metastatic disease, and randomized trials have demonstrated improvements in progression-free survival in patients with oligometastatic CSPC treated with metastasis-directed therapy. However, clinical outcomes remain heterogeneous and response to metastasis-directed therapy is variable, raising the need for prognostic/predictive biomarkers. A multimodal artificial intelligence biomarker (ArteraAI Prostate Test) was recently trained using data from patients with localized prostate cancer and found to be prognostic.1,2 The multimodal architecture is composed of two parts: (i) a tower stack to parse a variable number of digital histopathology slides, and (ii) another tower stack to merge the resultant features and predict binary outcomes:
At the 2024 ASCO annual meeting, Dr. Sutera and colleagues evaluated this biomarker in oligometastatic CSPC.
This was an international multi-institution retrospective review of 222 men with oligometastatic CSPC who were evaluated with multimodal artificial intelligence scoring. The primary objective was to compare overall survival between patients with high- and low-multimodal artificial intelligence score (stratified by median score). Overall survival was defined as the time from diagnosis of oligometastatic CSPC to death of any cause, calculated with the Kaplan-Meier method and compared using the log-rank test and Cox regression. The secondary objective was to evaluate multimodal artificial intelligence score as predictive for metastasis-directed therapy treatment effect in a subset of patients enrolled in the STOMP3 and ORIOLE4 randomized clinical trials. Given too few overall survival events for this subset, the authors evaluated multimodal artificial intelligence for metastasis-free survival, defined as time of randomization to development of a new metastasis or death of any cause and analyzed using Cox regression. An interaction test was performed between treatment arm and multimodal artificial intelligence score.
The median follow-up of the surviving patients was 38.0 months. Patients with high multimodal artificial intelligence (>0.527) were found to have higher PSA at diagnosis (9.61 vs 5.95 ng/mL, p = 0.005), higher Gleason score (69.4% vs 39.6% Grade Group ≥ 4, p < 0.001), more likely to have de novo metastatic disease (28.2% vs 8.1%, p < 0.001), and more likely to have bone metastases (55.5% vs 39.6%, p = 0.019). Patients with a high multimodal artificial intelligence score had a significantly worse overall survival (HR 7.68, 95% CI 1.62-36.49; p = 0.01) with a median overall survival of 108 months versus “not reached” (p = 0.004):
In the STOMP and ORIOLE subset (n = 51; median follow-up 61 months), multimodal artificial intelligence was not prognostic for metastasis-free survival (HR 1.24, 95% CI 0.64-2.43, p = 0.52). Multimodal artificial intelligence however was predictive for metastasis-directed therapy benefit as patients with high (HR 0.32, 95% CI 0.12-0.90; p = 0.03), but not low (HR 1.59, 95% CI 0.63-4.04; p = 0.33) multimodal artificial intelligence demonstrated improvement in metastasis-free survival when treated with metastasis-directed therapy (p-interaction = 0.02):
Dr. Sutera concluded his presentation by discussing the validation of a digital pathology-based multimodal artificial intelligence model in oligometastatic CSPC with the following take home messages:
- This study showed for the first time that the ArteraAI multimodal artificial intelligence biomarker is prognostic for overall survival in patients with oligometastatic CSPC
- Furthermore, multimodal artificial intelligence appears to be predict benefit of metastasis-directed therapy with high multimodal artificial intelligence demonstrating a greater improvement in metastasis-free survival following metastasis-directed therapy over observation
- Further work in validating these findings is warranted to allow for greater personalization in the management of patients with oligometastatic CSPC
Presented by: Philip A. Sutera, MD, Johns Hopkins University, Baltimore, MD
Written by: Zachary Klaassen, MD, MSc – Urologic Oncologist, Associate Professor of Urology, Georgia Cancer Center, Wellstar MCG Health, @zklaassen_md on Twitter during the 2024 American Society of Clinical Oncology (ASCO) Annual Meeting, Chicago, IL, Fri, May 31 – Tues, June 4, 2024.
References:
- Esteva A, Feng J, van der Wal D, et al. Prostate cancer therapy personalization via multi-modal deep learning on randomized phase III clinical trials. NPJ Digit Med. 2022 Jun 8;5(1):71.
- Spratt DE, Tang S, Sun Y, et al. Artificial Intelligence Predictive Model for Hormone Therapy Use in Prostate Cancer. NEJM Evid 2023;2(8).
- Ost P, Reynders D, Decaestecker K, et al. Surveillance of metastasis-directed therapy for oligometastatic cancer recurrence: A prospective, randomized, multicenter phase II trial. J Clin Oncol. 2018 Feb 10;36(5):446-453.
- Phillips R, Shi WY, Deek M, et al. Outcomes of Observation vs Stereotactic Ablative Radiation for Oligometastatic Prostate Cancer: The ORIOLE Phase 2 Randomized Clinical Trial. JAMA Oncol 2020 Mar 26;6(5):650-659.