Introduction: Several tools have been created to support clinician’s diagnosis decision for Alzheimer’s disease (AD). These tools make use of the hypometabolism pattern as seen in 2-[18F]-fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) scans. Studies have shown, however, that regional cerebral blood flow (rCBF) images derived from 11C-labelled Pittsburgh Compound B (PIB) PET scans may be good surrogates for FDG scans. The aim of this study was to explore whether rCBF images, derived from dynamic PIB scans, could be used for diagnosis of AD using the PALZ tool from PMOD. Subjects and Methods: Nine patients diagnosed with AD, ten with mild cognitive impairment, and eleven healthy controls underwent static FDG and dynamic PIB PET scans. Three sets of images were generated, all normalized to cerebellar grey matter values: from the FDG images, (1) standardized uptake value ratios (SUVR); and from the dynamic PIB images, (2) the ratio of tracer influx relative to reference region (R1) (through the simplified reference tissue model 2), and the time-weighted average of initial frames (ePIB; with time interval of (3) 20-130s, and (4) 1-8min). The images were analysed with the PALZ tool from PMOD (v3.9, PMOD Technologies LLC), providing an AD score. SUVR scores were used
as a reference to compare with rCBF scores. Correlations of the methods were drawn, and receiver operating curves (ROC) were made to explore sensitivity and specificity outcomes. Results: Both R1 and ePIB(20-130s) presented a good correlation when compared to SUVR (R²=0.78 and 0.75, respectively), with linear regression slopes of 0.74 and 0.71, while ePIB(1-8min) showed a worse correlation (R²=0.46), and a slope of 0.44. All ROC curves
exhibited a similar shape, with an area under the curve (AUC) of 0.95 for ePIB(20-130s), 0.94 for R1, 0.93 for SUVR, and 0.87 for ePIB(1-8min), and no statistically significant differences were found between curves. However, R1 and ePIB(1-8min) showed to have the highest sensitivity, of 100%, while SUVR and ePIB(20-130s) presented 89%. Meanwhile, SUVR had the highest specificity, of 76%, while R1 had a specificity of 67%, and both ePIB, 57%.Conclusion: Due to its high sensitivity and the fact that its ROC curve was not statistically significantly different from those seen with FDG SUVR, rCBF images suggested to be a good surrogate for FDG scans for diagnostic purposes. Therefore, rCBF (using R1) seems to be a valid method for the diagnosis of AD patients.