Quantitative PET Imaging in Drug Development: Estimation of Target Occupancy

Area of Science:

• Neuroscience
• Pharmacology
• Medical Imaging

Background:

• Positron emission tomography (PET) is crucial for drug development, especially in central nervous system research.
• A key application of PET is quantifying drug occupancy at molecular targets like receptors and transporters.
• Current methods for occupancy estimation rely on linear mathematical models and dual-scan PET imaging.

Purpose of the Study:

• To introduce novel maximum likelihood (ML) estimation approaches for determining drug target occupancy using PET imaging.
• To address the challenge of defining the covariance matrix in PET occupancy analysis, considering regional variance and covariance.
• To compare the performance of new ML methods against conventional linear approaches using simulated and human data.

Main Methods:

• Development of maximum likelihood estimation algorithms for PET target occupancy.
• Definition of covariance matrices accounting for regional binding measure variances and covariances.
• Comparison of ML methods with standard linear methods using PET imaging data from simulations and human studies.

Main Results:

• Simulations demonstrated reduced variance and bias with ML methods when assumptions were met.
• The novel ML methods provide a foundation for improved PET covariance models.
• Differences between methods were less pronounced in small human datasets, highlighting the need for robust models.

Conclusions:

• Maximum likelihood methods offer a statistically rigorous framework for PET drug occupancy quantification.
• Accurate covariance matrix definition is critical for minimizing bias and variance in occupancy studies.
• These advancements are expected to enhance the precision and reliability of PET-based drug development assessments.