Non-linear mixed-effects modelling and population-based model selection for ¹³¹I kinetics in benign thyroid disease

Area of Science:

• Nuclear medicine
• Medical physics
• Pharmacokinetics

Background:

• Radioiodine (¹³¹I) therapy is crucial for treating benign thyroid diseases.
• Accurate dosimetry, specifically time-integrated activities (TIAs), is essential for effective and safe ¹³¹I treatment.
• Current dosimetry models may require refinement for improved accuracy.

Purpose of the Study:

• To develop and validate a novel mathematical model for precise TIA calculation in ¹³¹I therapy.
• To utilize the population-based model selection and non-linear mixed-effects (PBMS-NLME) method for this development.
• To compare the performance of the new model against existing standards, such as the European Association of Nuclear Medicine (EANM) Standard Operational Procedure (SOP).

Main Methods:

• Collected biokinetic data of ¹³¹I from 73 patients at multiple time points post-administration.
• Employed PBMS-NLME modeling to select the best sum-of-exponential function (SOEF) based on Akaike weights.
• Evaluated nine SOEFs, including the EANM SOP function, with repeated fittings to ensure optimal parameter identification.

Main Results:

• The PBMS-NLME analysis identified a specific SOEF as the optimal model, supported by approximately 100% Akaike weight.
• The selected PBMS-NLME model demonstrated superior performance in describing ¹³¹I biokinetics compared to the EANM SOP function with individual fitting.
• The best SOEF derived from PBMS-NLME incorporated an additional free parameter.

Conclusions:

• The developed mathematical model using PBMS-NLME offers enhanced accuracy for calculating TIAs in ¹³¹I therapy.
• The inclusion of an extra parameter in the PBMS-NLME model contributes to its improved predictive capability.
• This refined dosimetry approach holds promise for optimizing treatment outcomes in benign thyroid disease.