Image derived input function for [18F]-FEPPA: application to quantify translocator protein (18 kDa) in the human...

Area of Science:

• Neuroimaging
• Radiochemistry
• Nuclear Medicine

Background:

• Automatic blood sampling system (ABSS) is the gold standard for input function (IF) extraction in [18F]-FEPPA PET imaging.
• Accurate IF is crucial for quantifying translocator protein (TSPO) in the human brain.
• Image-based IF extraction methods offer potential alternatives to invasive ABSS.

Purpose of the Study:

• To compare the performance of two image-based IF extraction methods (carotid signal [CS] and independent component analysis [ICA]) against the ABSS gold standard.
• To evaluate the reliability of these methods for quantifying TSPO in various human brain regions.

Main Methods:

• Collected [18F]-FEPPA PET scans from 18 healthy volunteers.
• Extracted IFs using direct delineation of the internal carotid signal (CS-IF) and independent component analysis (ICA-IF).
• Compared estimated total distribution volume (V(T)) values derived from CS-IF and ICA-IF to reference V(T) from ABSS-IF across multiple brain regions.

Main Results:

• ICA-IF demonstrated higher reliability than CS-IF for V(T) estimation across all evaluated brain regions.
• Frontal cortex analysis showed a strong regression slope for ICA-IF (r² = 0.91, p<0.05) compared to CS-IF (r² = 0.71, p<0.05).
• ICA-IF provided more consistent and accurate TSPO quantification compared to CS-IF.

Conclusions:

• Image-based IF extraction using ICA is a reliable alternative to ABSS for [18F]-FEPPA PET.
• ICA-IF improves the accuracy and consistency of TSPO quantification in the human brain.
• This method enhances the utility of PET imaging for studying TSPO in neurological research.