Model‐Informed Reinforcement Learning for Enabling Precision Dosing Via Adaptive Dosing

Clinical Pharmacology & Therapeutics, Год журнала: 2024, Номер 116(3), С. 619 - 636

Опубликована: Июль 11, 2024

Precision dosing, the tailoring of drug doses to optimize therapeutic benefits and minimize risks in each patient, is essential for drugs with a narrow window severe adverse effects. Adaptive dosing strategies extend precision concept time-varying treatments which require sequential dose adjustments based on evolving patient conditions. Reinforcement learning (RL) naturally fits this paradigm: it perfectly mimics decision-making process where clinicians adapt administration response evolution monitoring. This paper aims investigate potentiality coupling RL population PK/PD models develop algorithms, reviewing most relevant works field. Case studies were integrated within algorithms as simulation engine predict consequences any action have been considered discussed. They mainly concern propofol-induced anesthesia, anticoagulant therapy warfarin variety anticancer differing administered agents and/or monitored biomarkers. The resulted picture highlights certain heterogeneity terms approaches, applied methodologies, degree adherence clinical domain. In addition, tutorial how problem should be formulated key elements composing framework (i.e., system state, agent actions reward function), could enhance approaches proposed readers interested delving Overall, integration into RL-framework holds great promise but further investigations advancements are still needed address current limitations applicability methodology requiring adaptive strategies.

Язык: Английский

---

Machine Learning Methods for Precision Dosing in Anticancer Drug Therapy: A Scoping Review

Clinical Pharmacokinetics, Год журнала: 2024, Номер 63(9), С. 1221 - 1237

Опубликована: Авг. 17, 2024

In the last decade, various Machine Learning techniques have been proposed aiming to individualise dose of anticancer drugs mostly based on a presumed drug effect or measured biomarkers. The aim this scoping review was comprehensively summarise research status use for precision dosing in therapy. This conducted accordance with interim guidance by Cochrane and Joanna Briggs Institute. We systematically searched databases Medline (via PubMed), Embase Library articles reviews including results published after 2016. Results were reported according Preferred Reporting Items Systematic Reviews Meta-Analyses extension Scoping (PRISMA-ScR) checklist. A total 17 relevant studies identified. 12 included studies, Reinforcement methods used, Classical, Deep, Double Deep Conservative Q-Learning Fuzzy Learning. Furthermore, classical compared terms their performance an artificial intelligence platform parabolic equations used guide prospectively retrospectively, albeit only limited number patients. Due significantly different algorithm structures, meaningful comparison between approaches not possible. Overall, emphasises clinical relevance optimisation, as many algorithms shown promising enabling model-free predictions potential maximise efficacy minimise toxicity when standard protocols.

Язык: Английский

---

Precision Dosing in Presence of Multiobjective Therapies by Integrating Reinforcement Learning and PK‐PD Models: Application to Givinostat Treatment of Polycythemia Vera

CPT Pharmacometrics & Systems Pharmacology, Год журнала: 2025, Номер unknown

Опубликована: Май 5, 2025

Язык: Английский

---

Drug Delivery to the Respiratory System: Novel Approaches and Therapeutics

Methods in pharmacology and toxicology, Год журнала: 2025, Номер unknown, С. 303 - 334

Опубликована: Янв. 1, 2025

Язык: Английский

---

Leveraging machine learning in limited sampling strategies for efficient estimation of the area under the curve in pharmacokinetic analysis: a review

European Journal of Clinical Pharmacology, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 21, 2024

Язык: Английский

---