Background: [11C]PIB PET is a marker of amyloid in dementia. The majority of studies have either used Logan analysis to generate relative distribution volumes (DVR) with cerebellar uptake as a reference of non-specific binding or target to cerebellar ratios as the preferred methods of analysis. However, cerebellar amyloid may be present in genetic Alzheimer’s disease (AD) and prion diseases. In this study we assess whether the pons can be used as a reliable reference region for the analysis of [11C]PIB PET.

Methods: 12 AD subjects age 65±4.5 yrs and MMSE 21.4±4 and 10 control subjects had [11C]PIB PET with arterial blood sampling. Object maps were created by segmenting individual MRIs and spatially transforming the grey matter images into standard stereotaxic MNI space and then superimposing a probabilistic atlas. Cortical [11C]PIB uptake was assessed by ROI (region of interest) analysis. Regional DVRs were generated with the Logan method. Additionally, 60-90 min target to cerebellar ratios (RATIOCER) and 60-90 min target to pons ratios (RATIOPONS) were computed. Using SPSS we calculated T-scores, p values, percentage increases and intraclass correlations (ICC) for the different regions. Results: All the analytical methods were able to differentiate AD from controls (p<0.001). RATIOCER and RATIOPONS showed increased [11C]PIB uptake in AD compared to controls that was higher than seen using an arterial input function. All methods had a very high ICC; RATIOCER performed best closely followed by RATIOPONS.

Conclusions: This study shows that 60-90 min target to pons RATIOs can be used as a reliable method of analysis in [11C]PIB studies where cerebellum is not appropriate without the use of invasive arterial input measurements in clinical studies.