ARTI-MPC – Activity, Risk & Therapy-Intensity in Metastatic Prostate Cancer

About This Grant

PROJECT SUMMARY Metastatic prostate cancer (MPC) disproportionately affects older adults (OAs), significantly impairing their functional independence due to severe toxicities associated with androgen receptor signaling inhibitors (ARSIs). Despite guidelines recommending geriatric assessments (GAs), clinical dosing of ARSIs remains largely subjective, neglecting critical variations in frailty and physical function. Consequently, many OAs experience preventable adverse outcomes, underscoring an urgent need for individualized, evidence-based dosing strategies. My long-term goal is to transform MPC treatment for older men by developing precision dosing models that mitigate toxicity while preserving functional independence. My overall objectives include validating ARSI dose intensity and wearable sensor-derived metrics, particularly daily step counts, and using these metrics to inform predictive toxicity modeling. These objectives are designed to yield practical dosing guidelines tailored to individual frailty levels and functional trajectories. The rationale is that real-time, objective biomarkers like wearable-derived step counts combined with ARSI dose intensity can significantly enhance the predictive accuracy of ARSI-related toxicity, overcoming limitations of traditional, static assessments. Leveraging detailed data from two ongoing prospective studies, ProsGATE and DaroStep, my approach integrates serial geriatric assessments, wearable-derived activity metrics, and precise ARSI exposure records. Specifically, I will first determine if higher monthly ARSI doses predict an increased 12-month risk of severe toxicity and functional decline. Next, I will validate whether baseline step counts and significant early declines in activity independently predict adverse treatment outcomes. Finally, I will apply advanced statistical methods, including penalized spline-based and Bayesian additive regression tree models, to define personalized safe- dose windows for ARSIs utilizing GA and/or step defined strata. The significance of this research is profound, offering a scalable model to guide ARSI dosing, substantially reducing severe toxicities among over 120,000 older Americans with MPC. By pioneering the integration of wearable technology with rigorous statistical modeling, this proposal sets a benchmark for innovation in geriatric oncology. This K23 award will provide essential training through structured mentorship by experts in geriatrics (Dr. Huisingh-Scheetz), prostate oncology (Dr. Szmulewitz), Bayesian biostatistics (Dr. Ji), longitudinal modeling (Yan Che), and clinical pharmacology (Dr. Gobburu). This mentorship, combined with advanced coursework, will equip me with skills crucial for digital biomarker development, nonlinear dose-toxicity modeling, and precision clinical trial design. Ultimately, this award will facilitate my transition into an independent physician-scientist role, laying the foundation for future R01-level trials aimed at optimizing cancer therapy for older adults.