Mathematical Biomarkers of Adaptive Therapy Outcomes in Prostate Cancer

Area of Science:

• Oncology
• Mathematical Biology
• Biostatistics

Background:

• Adaptive therapy, an evolution-based strategy, delays resistance in prostate cancer by managing tumor burden.
• Patient responses to adaptive therapy are heterogeneous, necessitating biomarkers for personalized treatment scheduling.

Purpose of the Study:

• To develop and validate mathematical biomarkers predicting time to progression (TTP), mean daily dose, and overall survival (OS) from first-cycle prostate-specific antigen (PSA) dynamics.
• To assess the performance of these novel biomarkers against traditional PSA metrics.

Main Methods:

• Retrospective analysis of longitudinal data from 53 patients with castrate-sensitive (CSPC) and castrate-resistant (mCRPC) prostate cancer.
• Development of a 2-population differential equation model to describe tumor growth dynamics.
• Derivation of mechanism-based mathematical biomarkers (adaptive therapy score, expected TTP, expected mean daily dose) from initial-cycle PSA kinetics.

Main Results:

• The adaptive therapy score derived from first-cycle PSA dynamics was prognostic for prolonged TTP in both CSPC and mCRPC cohorts.
• In mCRPC patients, the adaptive therapy score and expected TTP were significantly associated with prolonged OS, unlike standard PSA metrics.
• The novel biomarkers demonstrated superior predictive performance compared to traditional phenomenological PSA metrics.

Conclusions:

• Mechanism-based mathematical biomarkers from initial-cycle PSA dynamics accurately predict patient-specific outcomes and survival.
• These biomarkers outperform traditional PSA monitoring and can inform personalized adaptive therapy protocols.
• The developed metrics offer a decision support framework for stratifying patients and optimizing treatment schedules.