Assistant Professor of Radiology & Biomedical Imaging Yale University New Haven, Connecticut, United States
Disclosure(s):
Moses Wilks, PhD: No financial relationships to disclose
Recently, there has been a great increase in clinical use of targeted radiopharmaceutical treatments (TRTs), especially after the FDA approval of 177Lu-PSMA-617 (Pluvicto) for the treatment of metastatic castration-resistant prostate cancers. Despite variability in patient response and disease state, currently patients receive fixed doses at fixed dosing intervals. 177Lu-PSMA is a perfect test case for predictive dosimetry and dose personalization, as all subjects undergo pre-treatment PSMA-PET screening, and can undergo post-treatment SPECT/CT scans to periodically monitor dosimetry and treatment response. Personalization may allow for safely increasing or decreasing standard RPT doses, increasing patient response or reducing off-target treatment. It could also allow for better prediction of response, with improved eligibility criteria for 177Lu-RPT. In this talk, we will discuss machine-learning and deep-learning based methods for predictive dosimetry in RPT, with potential applications for dose personalization and prediction of long-term therapeutic response. Examples will focus on PSMA-RPT, but with an eye on generalizability to other theranostic-RPT applications.
