Limited sampling models for doxorubicin pharmacokinetics.

Area of Science:

• Pharmacokinetics and Pharmacodynamics
• Oncology Drug Metabolism

Background:

• Doxorubicin is a widely used antineoplastic agent.
• The relationship between doxorubicin's area under the concentration-time curve (AUC) and its toxicity or efficacy remains unclear.
• Limited sampling strategies are emerging as a viable method for AUC estimation in pharmacodynamic studies.

Purpose of the Study:

• To develop and validate a model for estimating doxorubicin's AUC using limited plasma concentration data.
• To assess the feasibility of using two timed plasma concentrations to accurately predict doxorubicin AUC.
• To establish a reliable method for AUC estimation to facilitate future pharmacodynamic research.

Main Methods:

• Utilized pharmacokinetic data from 26 sarcoma patients and 5 breast cancer/unknown primary patients.
• Developed a predictive model using stepwise multiple regression on a training dataset (15 patients).
• Validated the model on two independent test datasets (11 sarcoma patients and 5 other patients).

Main Results:

• A regression model was established: AUC = 17.39 C2 + 163 C48 - 111.0 [dose/(50 mg/m2)].
• The model demonstrated good predictive accuracy with a mean predictive error (MPE) of 4.7% and root mean square error (RMSE) of 12.4% on the first test set.
• The model achieved an MPE of 4.5% and RMSE of 9.2% on the second test set, confirming its robustness. An additional model was developed for a 3-day dosing schedule.

Conclusions:

• Doxorubicin's AUC following bolus administration can be reliably estimated using just two plasma concentration measurements.
• The developed limited sampling model provides an accurate and feasible method for AUC estimation.
• This approach will aid in conducting pharmacodynamic studies and potentially optimizing doxorubicin therapy.