Alzheimer disease (AD) is a neurodegenerative disease notorious for its chronic and debilitating course, and it is strongly correlated with cognitive impairments and behavioral disturbances. Among the several pathophysiological causes of AD, cerebral retention of toxic amyloid-β (Aβ) peptides in soluble or intraneuronal forms has been repeatedly suggested, with amyloid plaques known to provoke synaptic injury and neurodegeneration associated with AD.¹

Introduction of amyloid positron emission tomography (PET) has contributed to the unraveling of the underlying pathological mechanisms of AD.² To date, the most widely used and studies Aβ PET tracer has been, 11C-labelled Pittsburgh compound-B (PiB).³ However, considering short half-life of carbon radioisotope (20 min), use of PiB is largely restricted, hindering it from active application in clinical settings.⁴ In these regards, 18F-labelled Aβ PET tracers longer half-life of 110 min (florbetaben, florbetapir, flutemetamol), have been discovered and gained US Food and Drug Administration approval for use in clinical practice.⁴ Among the 3 F labeled Aβ tracers, florbetaben (FBB) was the first 18F-labeled Aβ tracer studied in humans.⁴ Several prior works showed high sensitivity and specificity (90% sensitivity and 80% specificity) in discriminating AD from the normal controls.⁵ However, as compared to the studies using PiB, there is no comparison study on cerebral retention pattern between the cognitively intact elderly according to their Aβ retention status. As Aβ protein is known to accumulate for 10-15 years before the emergence of cognitive impairments in AD, measuring Aβ retention pattern in the cognitively normal older adults would help to understand more sophisticated pathophysiological process underpinning AD trajectory.⁶ In this study we used whole brain voxel wise analysis to measure Aβ retention pattern in cognitively normal older adults. As compared to region of interest (ROI), the benefits of the voxel wise analysis is to measure whole brain Aβ without a priori consumption. We hypothesized that the retention patterns will differ between FBB (+) and (−) group in the posterior cingulate (PCC), precunes (PRC), anterior cingulate (ACC), lateral temporal (LT) and frontal cortex (FRC) as in the previous voxel based study using the PiB.⁷

Table 1 showed the demographic data of the study participants. There was no significant differences between the FBB (+) and the FBB (−) group in demographic and the cognitive and behavioral measurements. In the group analysis, The FBB (+) group showed significantly higher amyloid retention pattern in the ACC, PRC, middle cingulate and PCC (Figure 1, Supplementary Table 1 in the online-only Data Supplement, FWE corrected p<0.05). However, these results were expanded to ACC, PRC, PCC FRC, and LT compared with the FBB (−) group when the level of significant adjusted to false discovery rate p<0.05 (Supplementary Figure 1 online-only Data Supplement). There was no significant correlation between the cognitive function scores measured by CERAD-K and the amyloid retention in the FBB (+) group. The detailed coordinates of group differences showed in the Supplemental Table 1 (online-only Data Supplement).

To our knowledge, this the first article to report results from whole brain voxel-based analysis of Aβ deposition utilizing ¹⁸F-Florbetaben on cognitively normal elderly subjects. According to our results, several notable brain regions in FBB (+) group demonstrated significantly higher Aβ retention pattern when compared with that of FBB (−) group.

Due to a short history of 18F-labeled ligands in amyloid imaging research, there have been relatively few articles on voxel-based analysis of cerebral Aβ deposition using 18F-labeled ligands. To list a few, a recent article utilizing ¹⁸F-Florbetapir discussed the potential role interaction between Aβ deposition and phosphorylated tau played on the decline of metabolism certain brain regions.¹¹ Such interaction was most evident in the temporal cortex, frontal cortex and both cingulate cortices.¹¹ Our results are in line with the aforementioned results, where amyloid retention was most evident the both ACC, PRC, PCC, FRC and LT. Studies on major depressive disorder patients with mild cognitive impairment (MCI) and Down syndrome and the pattern of Aβ retention in these two cohorts explored with ¹⁸F-Florbetapir yielded some meaningful results that accentuated the diagnostic utility of amyloid imaging in the prevention and early intervention of dementia.¹²¹³ As with ¹⁸F-Flutemetamol, a recent research utilized the ligand in discriminating MCI subjects into amyloid positive and negative groups, exploring possible differential factor between two groups including cortical metabolism and hippocampal volume by adopting voxel-wise surface maps.¹⁴

The brain regions with prominent Aβ deposition in FBB (+) group were regions that have been known to comprise the default mode network (DMN),¹⁵ which is frequently suggested to be areas of great connectivity where incipient Aβ deposition takes place.¹⁶ Despite the aforementioned Aβ burden in the DMN, which plays an integral role in the self-referential processing, attention and working memory,¹⁷ there was no significant correlation between the cognitive function scores and Aβ retention in our study, deviant from our original hypothesis. This phenomenon could be explained by relatively preserved brain reserve and cognitive function in this group.⁶¹⁸ Moreover, this group of participants could be subjects with high basal cerebral metabolism that has strength to withstand the adverse effects of Aβ retention.¹⁸

Our study has several limitations that must be taken into consideration. Our sample sizes were small, making our results difficult to generalize. Moreover, extension of the study to include data on apolipoprotein E genotyping and each regional standard uptake value ratio could have added more strength in our results. Further studies including the aforementioned variables could increase validity and objectivity. Lastly, although validated to a certain level proving accuracy in detecting a Aβ deposition,¹⁹ additional standardized data on ¹⁸F-Florbetaben could be of help for more active utilization of this ligand in the clinical setting.

In summary, our study results are in line with previous findings where Aβ deposition mainly affected brain regions with active connectivity at rest. Moreover, further exploration on the relationship between brain reserve and amyloid burden will unravel underlying pathologic mechanism that precedes actual manifestation of AD symptoms.