Partial volume correction in longitudinal tau PET studies: is it really needed?

Area of Science:

• Neuroimaging
• Nuclear Medicine
• Alzheimer's Disease Research

Background:

• Non-specific binding in off-target regions affects [¹⁸F]flortaucipir (FTP) tau PET quantification.
• Partial volume correction (PVC) methods can mitigate these effects but may introduce noise.
• The impact of PVC on longitudinal FTP studies requires further investigation.

Purpose of the Study:

• To evaluate the performance of common PVC techniques for longitudinal FTP imaging.
• To assess the influence of PVC on the quantification of FTP tau PET.
• To determine the effectiveness of PVC in detecting longitudinal changes and predicting outcomes.

Main Methods:

• Utilized data from 247 Alzheimer's Disease Neuroimaging Initiative participants with longitudinal FTP-PET, amyloid-beta PET, and MRI.
• Applied Meltzer’s, region-based voxel-wise (RBV), and iterative Yang (iY) PVC techniques to FTP-PET scans.
• Analyzed longitudinal FTP SUVR rates, off-target binding correlations, group differences (A-/A+, CU/CI), and prediction of atrophy and cognitive decline.

Main Results:

• Hemispheric white matter and CSF showed the highest correlation with longitudinal FTP SUVR rates; voxel-wise PVC minimized this.
• RBV PVC demonstrated the best performance in reducing off-target binding correlations.
• PVC enhanced group differences between cognitively unimpaired and impaired individuals and strengthened the prediction of longitudinal brain atrophy and cognitive decline.

Conclusions:

• Longitudinal FTP SUVR quantification is influenced by off-target regions, notably white matter and CSF.
• Voxel-wise PVC techniques effectively reduce these off-target effects.
• PVC offers a modest but significant benefit for longitudinal group studies, impacting sample size and analysis methodology.