Can relative flow derived from dynamic ¹¹C-PIB scans replace FDG scans in Alzheimer disease PET studies?

Aim: Different PET biomarkers are used for the differential diagnosis of neurodegenerative diseases such as Alzheimer’s disease (AD). β-amyloid plaque deposition can be assessed by ¹¹C-Pittsburgh Compound B (PIB) while ¹⁸F-fluorodeoxyglucose (FDG) is used to assess cerebral glucose metabolism, which can be an indicator for synaptic dysfunction. Although FDG is a biomarker for glucose consumption, its distribution is strongly affected by cerebral perfusion. The aim of the study is to explore whether relative blood flow (rCBF) data, extracted from dynamic ¹¹C-PIB studies, can be used as a surrogate for relative FDG distributions. Methods: Thirty subjects (10 AD, 10 mild cognitive impairment (MCI) patients, and 10 healthy controls (HC)) underwent static ¹⁸F-FDG (45-60 min p.i.) and 70 min dynamic ¹¹C-PIB PET scans. T1-MRI scans were used for anatomical labelling of brain regions. Parametric ¹¹C-PIB rCBF images (R1: tracer delivery relative to the cerebellum) were obtained using the simplified reference tissue model 2 (SRTM2, PMOD v3.8). ¹⁸F-FDG images were normalized to the cerebellar grey matter values (SUVr). The Hammers atlas was used to analyze regional correlations between R₁ and SUVr. We used SPM12 to compare SUVr and R₁ statistical parametric maps between AD and HC. Results: The regional analysis showed strong correlations between R1 and SUVr for AD, MCI and HC (group average R² of 0.89 (AD), 0.90 (MCI) and 0.84 (HC)). Both R₁ and SUVr showed similar statistical parametric maps for AD and HC which are in agreement with previous reports in literature. However, SPM based on R₁ showed a smaller number of significant clusters involved than with the SPM based on FDG SUVr. In addition, the clusters found in R₁ were smaller in size and T-value. For example, in the left inferiolateral parietal lobe FDG SUVr showed 2297 voxels(mean(T)=5.4) versus 999 voxels for R₁ (mean(T)=4.7). Conclusion: Relative cerebral perfusion images (R₁) obtained from ¹¹C-PIB dynamic PET acquisitions are strongly correlated with normalized ¹⁸F-FDG images in all 3 groups and largely show the same SPM pattern in a group comparison. However, the observed group differences for R₁ were smaller and less significant which may be related to the lower signal-to-noise ratio of the R₁ images. These preliminary results suggest that dynamic ¹¹C-PIB PET can not only be used to assess β-amyloid plaque deposition, but that relative rCBF derived from these scans can in some cases be used to replace FDG thereby obviating the need to perform a second tracer (FDG) study.