Comparison of SUVr metrics in PET with ¹⁸F-Flutemetamol using the SPM and CPET software packages

Abstract

Alzheimer's disease (AD) can be characterised by the deposition of β-amyloid (βA) plaques in the brain. Positron emission tomography with computed tomography (PET-CT) with ¹⁸F-Flutemetamol allows the anatomical-physiological identification and quantification of the βA plaques deposition through the Standard Uptake Value ratio (SUVr). These scans are typically quantified using paid software tools such as Comprehensive PET Analysis (CPET), economic motivations led to the development of the software Statistical Parametric Mapping (SPM) to replace CPET. The aim of this study was to compare the quantification of βA plaque density by both software programmes in patients with suspected AD, using metrics based on SUVr. A retrospective cross-sectional study was carried out with a non-probabilistic convenience sample of 30 ¹⁸F-Flutemetamol PET-CT images in adult patients with suspected AD. All images were previously anonymized, ensuring participant confidentiality. As this study used fully anonymized secondary data, no additional ethical concerns were raised.

Statistical analysis was carried out using the SPSS software and the agreement between the software programmes ranged from poor (K=0.294) to very good (K=0.865).

In terms of diagnosis, the SPM showed a sensitivity of 82.35% and specificity of 81.25%, with an agreement rate of 90%, a false positive rate of 0% and a false negative rate of 18.75%. Currently, SPM should not replace CPET, as their agreement disparity can affect the diagnosis of patients. SPM could prove promising as a complementary or alternative tool to CPET if its methodology and accuracy improve.