Multiple linear analysis methods for the quantification of irreversibly binding radiotracers

Area of Science:

• Nuclear medicine
• Radiochemistry
• Pharmacokinetics

Background:

• Gjedde-Patlak graphical analysis (GPGA) is a standard method for quantifying irreversible radioligand uptake.
• GPGA has limitations in estimating net accumulation (K(in)) for certain tracers.

Purpose of the Study:

• To introduce and evaluate Multiple Linear Analysis for Irreversible Radiotracers (MLAIR) as an alternative to GPGA.
• To compare the performance of MLAIR with GPGA using both simulated and real positron emission tomography (PET) data.

Main Methods:

• Derivation of two multiple linear regression models (MLAIR1 and MLAIR2) from the two-tissue compartment model with irreversible binding.
• Computer simulations to assess bias and uncertainty of K(in) estimates.
• Application to real [(11)C]MeNTI PET data and comparison with nonlinear least squares (NLS) and GPGA.

Main Results:

• MLAIR1 yielded less biased K(in) estimates but higher uncertainty with noisy data.
• MLAIR2 demonstrated increased robustness against variability but with increased bias.
• MLAIR2 parametric images showed superior quality and improved statistical power for voxelwise comparisons over GPGA.
• Both MLAIR methods correlated well with NLS estimates on real PET data.

Conclusions:

• MLAIR, particularly MLAIR2, presents advantages over GPGA for quantifying irreversible radiotracers.
• MLAIR2 offers improved image quality and statistical power, making it a valuable alternative for PET data analysis.
• The MLAIR approaches provide a robust and effective alternative for pharmacokinetic modeling in nuclear medicine.