PT  - JOURNAL ARTICLE
AU  - Susan De Santi
AU  - Ana Catafau
AU  - John Seibyl
AU  - Santiago Bullich
TI  - Robustness of 18F-Florbetaben SUVR cutoff quantification across reference regions and standards of truth
DP  - 2016 May 01
TA  - Journal of Nuclear Medicine
PG  - 458--458
VI  - 57
IP  - supplement 2
4099  - http://jnm.snmjournals.org/content/57/supplement_2/458.short
4100  - http://jnm.snmjournals.org/content/57/supplement_2/458.full
SO  - J Nucl Med2016 May 01; 57
AB  - 458Objectives Thresholds of 18F-florbetaben (FBB) standardized uptake value ratios (SUVRs) to categorize scans as positive or negative between 1.39 [1] and 1.48 [2] have been reported using cerebellar grey matter (GCER) as the reference region (RR) and histopathology or clinical diagnosis as the standard of truth (SoT). The aims of this study were: 1) to generate FBB SUVR cutoff values for different RRs using visual assessment as SoT, and 2) to validate them across different clinical populations and SoTs.Methods 1) SUVR cutoff was generated using FBB scans from a sample of 143 subjects (69.5±7.5 yrs; n=75 Alzheimer’s disease (AD), n=68 healthy volunteers (HVs)) who were assessed by 3 blinded independent readers as positive and negative. A composite SUVR (mean of frontal, occipital, parietal, lateral temporal and posterior and anterior cingulate regions) was calculated for different RRs: cerebellar gray matter (GCER), whole cerebellum (average of white and grey matter) (WCER), pons (PONS), and subcortical white matter (SWM). A SUVR cutoff value for each RR was generated using Receiver Operating Characteristic analysis, visual assessment as the SoT, and highest accuracy as the selection criteria. 2) SUVR cutoffs for each RR were validated in two different samples and SoTs: Sample a) n=78 end-of life subjects (80.1±10.4 yrs) who underwent autopsy after death were included. Histopathological confirmation of the presence or absence of neuritic beta-amyloid plaques was performed using Bielschowsky silver stain and immunohistochemistry. Cutoff validation was validated using histopathology and visual assessment (VA) results from 8 independent blinded readers as SoTs; and Sample b) n=233 additional subjects (72.6±7.3 yrs; n=118 HVs, n=115 AD) using VA from the same 8 readers and clinical diagnosis as SoT. Sensitivity, specificity and accuracy (and their 95% confidence intervals (CI)) were used as performance metrics for the SUVR cutoff validation.Results Composite SUVR cutoff generated values for each RR were 1.43 (GCER); 0.96 (WCER); 0.78 (PONS); 0.71 (SWM). Accuracy values were high and very consistent across SUVR cutoffs and VA methods and SoTs (range 83-94% for histopathology, 80-87% for clinical diagnosis and 86-97% for visual assessment). When using histopathology as SoT, no statistically significant differences in sensitivity, specificity and accuracy were found across RRs.Conclusions These results indicate the robustness of 18F-FBB SUVR cutoff quantification. SUVR cutoff performance was high for all RRs, independently of the SoTs. The lack of differences across RRs and the fact that GCER has similar non-displaceable activity as the target region support the use of GCER as the biologically most appropriate RR for 18F-FBB scan quantification. [1] Barthel et al., Lancet Neurol 2011; 10: 424-35 [2] Sabri et al., Alzheimers Dement. 2015;11(8):964-74