Model-based optimization of G-CSF treatment during cytotoxic chemotherapy

Area of Science:

• Oncology
• Pharmacology
• Computational Biology

Background:

• Granulocyte-colony stimulating factor (G-CSF) is crucial for managing chemotherapy-induced leukopenia.
• Optimal G-CSF dosing, timing, and patient-specific factors remain debated.
• Current G-CSF use lacks standardized, predictive scheduling.

Purpose of the Study:

• To develop a comprehensive model integrating biological and clinical data for granulopoiesis under chemotherapy.
• To predict and optimize G-CSF (filgrastim, pegfilgrastim) treatment schedules.
• To validate model predictions with clinical trial data.

Main Methods:

• Integrated data on granulopoiesis cell kinetics, chemotherapy effects, and G-CSF pharmacokinetics/pharmacodynamics.
• Developed a computational model to simulate leukocyte time courses across >70 therapy scenarios.
• Applied the model to devise optimized G-CSF schedules for diverse clinical situations.

Main Results:

• Model predictions for alternative G-CSF schedules were validated by clinical trials.
• Proposed optimized G-CSF modifications for specific chemotherapy regimens (BEACOPP escalated, ETC, CHOP-14).
• Identified risk-adapted schedules for elderly patients with aggressive non-Hodgkin's lymphoma.

Conclusions:

• Established a predictive model for human granulopoiesis under chemotherapy.
• The model enables comparison and optimization of G-CSF schedules.
• Recommends initiating G-CSF later in the cycle and continuing until chemotherapy completion for patient benefit.