Optimal chemotherapy for leukemia: a model-based strategy for individualized treatment

Area of Science:

• Pharmacology
• Mathematical Modeling
• Pediatric Oncology

Background:

• Childhood Acute Lymphoblastic Leukemia (ALL) survival rates have improved, but long-term survivors face complications from chemotherapy.
• Current treatment protocols often involve standard doses, leading to toxicity or sub-optimal efficacy due to individual variations.

Purpose of the Study:

• To develop a model-based approach for personalizing 6-Mercaptopurine (6-MP) treatment in pediatric ALL.
• To incorporate pharmacogenomic variations into treatment personalization.
• To optimize 6-MP dosage for individual patients to maximize efficacy and minimize side effects.

Main Methods:

• Developed and validated semi-mechanistic mathematical models for 6-MP metabolism, red blood cell mean corpuscular volume (MCV), and leukopenia.
• Employed model reduction techniques using global sensitivity analysis to handle limited clinical data.
• Individualized patient models using reduced parameters and nonlinear model predictive control for dose optimization.

Main Results:

• The developed models accurately fit patient data and capture diverse clinical behaviors with minimal parameters.
• Model validation was successful using data from literature and a children's hospital.
• The approach demonstrated the potential for optimizing individual patient dosing.

Conclusions:

• Model-based, individualized treatment for childhood ALL using 6-MP is feasible with routine clinical measurements.
• This approach offers a promising alternative to standard dosing, aiming for optimal outcomes with reduced toxicity.
• Personalized dosing strategies can significantly improve the long-term health of ALL survivors.