Quantification of [11C]PBR28 data after systemic lipopolysaccharide challenge

Background Lipopolysaccharide (LPS) is a classic immune stimulus. LPS combined with positron emission tomography (PET) 18 kDa translocator protein (TSPO) brain imaging provides a robust human laboratory model to study neuroimmune signaling. To evaluate optimal analysis of these data, this work compared the sensitivity of six quantification approaches. Methods [C-11]PBR28 data from healthy volunteers (N = 8) were collected before and 3 h after LPS challenge (1.0 ng/kg IV). Quantification approaches included total volume of distribution estimated with two tissue, and two tissue plus irreversible uptake in whole blood, compartment models (2TCM and 2TCM-1k, respectively) and multilinear analysis-1 (MA-1); binding potential estimated with simultaneous estimation (SIME); standardized uptake values (SUV); and SUV ratio (SUVR). Results The 2TCM, 2TCM-1k, MA-1, and SIME approaches each yielded substantive effect sizes for LPS effects (partial eta(2) = 0.56-0.89, ps <. 05), whereas SUV and SUVR did not. Conclusion These findings highlight the importance of incorporating AIF measurements to quantify LPS-TSPO studies.