18 F-FDG Is a Superior Indicator of Cognitive Performance Compared to 18 F-Florbetapir in Alzheimer’s Disease and Mild Cognitive Impairment Evaluation: A Global Quantitative Analysis

Abstract

Background:

¹⁸F-Fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography (PET) and ¹⁸F-florbetapir PET are approved neuroimaging biomarkers for the Alzheimer’s disease (AD) and mild cognitive impairment (MCI).

Objectives:

This study aims to compare the efficacy of ¹⁸F-FDG and ¹⁸F-florbetapir PET at evaluating the cognitive performance of patients with AD, MCI, and normal controls (NC).

Methods:

63 subjects (36 male/27 female, mean age = 68.3) including 19 AD, 23 MCI, and 21 NC underwent ¹⁸F-FDG and ¹⁸F-florbetapir PET imaging. A global quantification approach was applied on supra-tentorial, frontal, parieto-occipital, temporal, and cerebellar brain regions by calculating the global SUVmean ratios (GSUVr) as the weighted average of all regional SUVmean. ¹⁸F-FDG and ¹⁸F-florbetapir GSUVr of each region were subsequently correlated with the Mini-Mental State Examination (MMSE).

Results:

Subjects were studied in five categories as NC, MCI patients, AD patients, MCI and AD patients grouped together (MCI/AD), and a group including all the subjects (NC/MCI/AD). Both ¹⁸F-FDG and ¹⁸F-florbetapir could successfully detect subjects with dementia (p < 0.001). Studied in all regions and groups, the correlation analysis of ¹⁸F-FDG GSUVr with MMSE scores was significant in more regions and groups compared to that of ¹⁸F-florbetapir. We also demonstrated that the correlation of ¹⁸F-FDG GSUVr with MMSE is stronger than that of ¹⁸F-florbetapir in the supra-tentorial and temporal regions.

Conclusions:

This study reveals how ¹⁸F-FDG-PET global quantification is a superior indicator of cognitive performance in AD and MCI patients compared to ¹⁸F-florbetapir PET. Accordingly, we still recommend ¹⁸F-FDG-PET over amyloid imaging in the evaluation for AD and MCI.

Keywords

Alzheimer’s disease amyloid-β protein florbetapir mild cognitive impairment Mini-Mental State Examination positron emission tomography

INTRODUCTION

Alzheimer’s disease (AD), the most prevalent cause of dementia, is the sixth leading cause of death in the United States [1]. In 1984, The National Institute of Neurologic and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) established that patients must exhibit symptoms of dementia to be diagnosed with AD [2]. However, it has been shown that AD results in cognitive decline years before dementia is clinically diagnosed [3, 4]. Therefore, the National Institute on Aging and the Alzheimer’s Association (NIA-AA) guidelines in 2011 revised the definition of AD to include a continuum of pathologies beginning with asymptomatic neurological changes and continuing through full-onset dementia [2, 5].

The revised construct addresses not only the individuals suffering from Alzheimer’s dementia (AD), but also those individuals with mild cognitive impairment (MCI) who present subjective and/or objective cognitive complaints due to AD that are not severe enough to merit the diagnosis of dementia [2, 6].

While the 1984 criteria were mainly based on clinical judgment and exclusion of other pathologies of dementia, the revised NIA-AA guidelines advise that neuroimaging and the use of tissue biomarkers are essential for an AD diagnosis as well [2, 5]. Two of the most significant biomarkers found in AD are decreased cortical glucose metabolism and the accumulation of amyloid-β (Aβ) plaques, both of which can be measured using current imaging techniques with positron emission tomography (PET) [7 –9]. In fact, the NIA-AA’s neuropathologic guidelines in 2011 considered amyloid plaques, along with tau protein pathology in the form of neurofibrillary tangles (NFTs), to be essential for the diagnosis of AD [10].

The primary modality for quantifying glucose hypometabolism in the brain is ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET. It has been shown that ¹⁸F-FDG-PET can accurately diagnose AD patients with a sensitivity and specificity of about 90% and 80%, respectively [11, 12]. Specific ¹⁸F-FDG-PET patterns of hypometabolism could also distinguish AD patients from normal controls with 86– 99% sensitivity and 86– 98% specificity (98% accuracy, p < 0.001) [12 –14]. Although not specific to AD, decreased ¹⁸F-FDG-PET uptake has been shown to reflect impaired cognitive function and has been shown to correlate well with disease severity [13 , 16].

Several radiotracers have also been developed to analyze the deposition of amyloid plaques by binding to Aβ peptides within the plaques. ¹⁸F-florbetapir, ¹⁸F-florbetaben, and ¹⁸F-flutemetamol are second-generation amyloid tracers that have been approved by the Food and Drug Administration (FDA) and European regulatory authorities for clinical use under the names of Amyvid®, Neuraceq®, and Vizamyl® respectively. These ¹⁸F-labeled Aβ radiotracers bind specifically to Aβ fibrils with high affinity, have a longer half-life (110 min) than ¹¹C-Pittsburgh Compound B (PiB) (20 min), and are thus favored in the clinical setting [17]. Colloquially referred to as amyloid PET tracers, ¹⁸F-florbetapir, ¹⁸F-florbetaben, and ¹⁸F-flutemetamol are generally accepted to have similar uptake patterns and diagnostic power [18 –20]. ¹⁸F-florbetapir, in particular, boasts a sensitivity of 96% and a specificity of 100% for the detection of Aβ plaques in people who had autopsy within 1 year of PET imaging [21]. Moreover, ¹⁸F-florbetapir uptake correlates well with the postmortem presence and density of Aβ plaques [22, 23].

Studies have reported that some populations of AD patients have higher cortical ¹⁸F-florbetapir uptake compared to healthy controls [22]. However, while detecting Aβ plaques is one of the characteristic pathological findings of AD, studies of a large population of non-demented, cognitively healthy, elderly subjects has demonstrated significant Aβ burden in many of these individuals as well [24]. In fact, autopsy reports have shown that approximately 50% of subjects found with significant Aβ plaques were cognitively healthy at death [24]. Furthermore, studies using in vivo neuroimaging have shown that despite the strong association between amyloid burden and AD, the amount of amyloid deposition is not associated with inferior cognitive performance [24 –26]. These studies support the notion that the presence of Aβ plaques is necessary but not sufficient for the pathology of AD or even cognitive decline.

Table 1

Demographic characteristics of the subjects

		NC	MCI	AD	MCI/AD	NC/MCI/AD	p
n		21	23	19	42	63	–
Age (SD)		65.7 (14.8)	65.7 (9.8)	74.3 (11.3)	69.6 (11.3)	68.3 (12.5)	One-way ANOVA^*
							0.044
Sex	M	9	16	11	27	36	χ ^2 *
	F	12	7	8	15	27	0.202
MMSE (SD)		30 (1.6)	26.4 (3.3)	20.3 (7.0)	23.6 (6.0)	25.5 (5.7)	Kruskall Wallis^*
							<0.001

^*Comparison is only made between NC, MCI, and AD groups. NC, normal control; MCI, mild cognitive impairment; AD, Alzheimer’s disease; MMSE, Mini-Mental State Examination.

Thus, although both cerebral glucose metabolism and Aβ plaque pathology are implicated in AD, a rise in plaque density does not necessarily imply a reduction in brain metabolism. While the presence of Aβ plaques has been detected years before the clinical manifestation of cognitive impairment, reduced ¹⁸F-FDG uptake has been shown to strongly indicate reduced cognitive function [16 , 27]. Given these nuances regarding amyloid presence, AD diagnosis, and cognitive function, new techniques for analyzing the relationships between ¹⁸F-florbetapir binding, cerebral glucose metabolism, and cognitive performance are necessary to determine the best practices for evaluating AD and its potential precursor, MCI.

In general, quantification of globally averaged whole brain and regional volume tracer uptake is a robust method for assessing disease burden. We have previously shown that such global quantitative analyses are effective and reliable methods of disease evaluation in numerous clinical contexts [28 –34]. As such, the purpose of this study is to compare the efficacy of ¹⁸F-FDG and ¹⁸F-florbetapir PET at discriminating between and evaluating the cognitive performance of normal controls, MCI, and AD patients using a global quantitative approach. Moreover, we are testing the hypothesis that a marker of functional neuronal activity, ¹⁸F-FDG, is a superior indicator of cognitive decline than markers of accumulation of amyloid aggregates such as ¹⁸F-florbetapir in the brain of human subjects.

METHODS

Subjects

The study was approved (No. 10F.476) by the Thomas Jefferson University Institutional Review Board. A total of 63 participants were recruited: 19 patients with a clinical diagnosis of AD, 23 patients with a diagnosis of MCI, and 21 normal control subjects (NC). Demographic data are presented in Table 1. Informed consent for human subject research was obtained from all subjects.

All subjects were evaluated with brain CT scan followed by PET scan with both ¹⁸F-florbetapir and ¹⁸F-FDG. At the time of scanning, all subjects were given a Mini-Mental State Examination (MMSE), with scores of 0 (severe dementia) to 30 (normal) to assess the overall cognitive performance.

The AD diagnosis was made on the basis of the criteria of the National Institute of Neurologic and Communicative Disorders and Stroke and the Alzheimer Disease and Related Disorders Association Work Group. In addition, AD patients had no other underlying disease process that could have caused dementia.

Subjects were given the diagnosis of MCI based on 1) complaint of memory or cognitive decline corroborated by an informant, 2) Clinical Dementia Rating global score of 0.5, 3) objective evidence of cognitive impairment or marginally normal cognition with a documented history of high cognitive performance, 4) no obvious neurological or medical cause for the impairment (i.e., encephalopathy, nephropathy, head trauma, or stroke), 5) the criteria for AD not being satisfied, and 6) essentially normal measures of activities of daily living.

NCs had no evidence of cognitive impairment by history, and no evidence of cognitive impairment as verified by neuropsychological testing. NCs had no history of psychiatric, neurological, or active medical conditions that would affect brain function or cognition. NCs were free of any medications that would notably affect brain function.

¹⁸F-FDG PET acquisition

Participants underwent ¹⁸F-FDG PET according to the Alzheimer’s Disease Neuroimaging Initiative protocol [35]. The head was fixed by a head holder, eyes and ears were open, and the correct body position was monitored by a technician throughout the study. Environment ambient noise was kept to minimum. Subjects were injected with approximately 185 MBq (5 mCi) of ¹⁸F-FDG. Scanning was initiated approximately 30 min after the administration of ¹⁸F-FDG. The acquisition time was 30 min, followed by a transmission scan for attenuation correction.

¹⁸F-florbetapir (Amyvid) PET acquisition

Participants also underwent ¹⁸F-florbetapir PET scan according to the prior protocols [36]. They were injected with approximately 370 MBq (10 mCi) of ¹⁸F-florbetapir and scanning was initiated approximately 50 min later. The acquisition time was 10 min, followed by a transmission scan for attenuation correction.

Image analysis and quantification

PET/CT images were converted to Digital Imaging and Communications in Medicine (DICOM) files for analysis which was performed using a dedicated image processing software (OsiriX software; Pixmeo SARL; Bernex, Switzerland) (see Fig. 1).

We developed a novel segmentation method to delineate cerebral regions for global quantification of the brain PET images. Supra-tentorial, frontal, parieto-occipital, temporal, and cerebellar (CL) regions of interest were outlined manually on 1 mm thick sagittal PET/CT fused image slices. Two lines were sketched manually from the posterior pole of the occipital lobe to the anterior pole of the frontal lobe and temporal lobe. Two perpendicular lines were drawn crossing the middle point of the prior ones. ROIs were drawn on every sagittal slide using inner skull margin portrayed on CT as the outer boundaries and perpendicular lines as well as lateral sulcus and tentorium as separators between regions (see Fig. 2). Subsequently, the cross-sectional (slice by slice) Standardized Uptake Values (SUVs) and the cross-sectional volumes were exported for each region.

Fig.1

PET imaging using ¹⁸F-FDG and ¹⁸F-florbetapir in Alzheimer’s disease (AD), mild cognitive impairment (MCI) patients, and normal controls (NC).

Fig.2

The segmentation method. 1) Supra-tentorial (ST) cerebellar (CL) regions: The Inner margin of the skull portrayed on CT was used as a guide to delineate the outer boundaries and tentorium was the separator between supra-tentorial and cerebellar regions. 2) Frontal (F), parieto-occipital (PO) and temporal (T) regions: Two lines were sketched from the posterior pole of the occipital lobe to the anterior pole of the frontal lobe (a) and temporal lobe (b). Two perpendicular lines (c, d) were drawn crossing the middle point of lines A and B. Frontal, parieto-occipital and temporal regions were outlined by drawing manual ROIs using line C as the separator between PO and F regions and d as the separator between T and PO regions, tentorium as the separator between supra-tentorial and cerebellar regions and lateral sulcus as the separator between F and T regions. The Inner margin of the skull was again used as a guide to define the outer borders of the regions.

Mean standardized uptake value (SUVmean) was the measure of choice for the global quantification approach. The global metabolic activity and global amyloid burden were measured by calculating a global standardized uptake value (GSUVmean) as the weighted average of all regional SUVmean values.

GSUVmean = Weighted Average $SUVmean = \frac{\sum [Cross - sectional SUVmean \times Cross - sectional PET Volume]}{\sum Cross - sectional PET Volume}$

GSUVmean of each region was then normalized to that of the cerebellar region to obtain a standardized uptake value ratio (SUVr). The cerebellar uptake value was selected as a reference since it is not supposed to be affected substantially by the course of AD [37]. The global metabolic activity and the global amyloid burden were referred to as the Global SUV ratio (GSUVr). ${GSUV}_{r} = \frac{GSUVmean}{GSUVmean of Cerebellar Region}$ (1)

Fig.3

¹⁸F-FDG and ¹⁸F-florbetapir GSUVr in Alzheimer’s disease (AD), mild cognitive impairment (MCI), and normal control (NC) groups. Both scans detect dementia in patients with known clinical AD or MCI, but neither tracer was able to distinguish between MCI and AD. Units are shown in GSUVr referenced to cerebellar uptake (95% CI).

Data analysis

Group age difference was evaluated with analysis of variance (ANOVA), followed by Bonferroni post hoc tests to assess for significance. The Chi-square test was used for sex differences between the AD, MCI, and NC groups and the Kruskal-Wallis test was used to assess differences between the MMSE score. Correlation analysis for ¹⁸F-FDG GSUVr and ¹⁸F-florbetapir GSUVr with MMSE scores yielded Pearson’s product moment correlation coefficient(r). Steiger’s Z test was applied to determine the relative strength between two correlation coefficients when the correlation analysis was significant with both ¹⁸F-FDG and ¹⁸F-florbetapir GSUVr. Results were considered significant at p < 0.05. Statistical analyses were performed with IBM SPSS Statistics Ver.22.

RESULTS

As shown in Table 1, subjects were grouped into five categories: Normal Controls (NC), Mild Cognitive Impairment (MCI), Alzheimer’s Disease patients (AD), MCI plus AD patients together (MCI/AD), and all the subjects included together (NC/MCI/AD). Analyses revealed no statistically significant group differences in terms of sex (p < 0.001). There was only a negligible age difference between the NC and AD group (p = 0.044), but age was not significantly different when NC was compared to MCI or when MCI was compared to AD.

Supra-tentorial ¹⁸F-FDG GSUVr in AD and MCI groups was significantly lower than in the NC group (p < 0.001). Likewise, AD and MCI patients had significantly higher ¹⁸F-florbetapir GSUVr than normal controls (p < 0.001). The difference between the MCI and AD groups was not significant for either ¹⁸F-FDG (p = 0.188) or ¹⁸F-florbetapir (p = 0.392; see Fig. 3).

For the globally averaged supra-tentorial region, ¹⁸F-FDG GSUVr showed a strong positive correlation with MMSE scores for the AD, MCI/AD, and NC/MCI/AD groups. Positive correlations were also seen within the NC and MCI groups, although they were not statistically significant. By contrast, ¹⁸F-florbetapir GSUVr patterns showed a significant negative correlation with the NC/MCI/AD group only. But this correlation was significantly weaker than that for ¹⁸F-FDG based on Steiger’s Z test results (Z = 2.798, p = 0.003). For the NC and MCI groups, ¹⁸F-florbetapir showed no statistically significant correlation in either group (see Table 2 and Fig. 4).

Fig.4

Correlation analysis for supra-tentorial ¹⁸F-FDG and ¹⁸F-florbetapir GSUVr with MMSE score.

Table 2

Correlation analysis for supra-tentorial ¹⁸F-FDG and ¹⁸F-florbetapir GSUVr with MMSE score in different groups

		NC	MCI	AD	MCI/AD	NC/MCI/AD
¹⁸F-FDG	r	0.285	0.295	0.592^**	0.511^**	0.614^**
	p	0.210	0.172	0.008	0.001	<0.001
¹⁸F-Forbetapir	r	0.348	0.143	– 0.036	– 0.05	– 0.291^*
	p	0.122	0.515	0.884	0.753	0.021

^*Correlation is significant at the 0.05 level (2-tailed). ^**Correlation is significant at the 0.01 level (2-tailed).

Table 3

Correlation analysis for regional ¹⁸F-FDG and ¹⁸F-florbetapir GSUVr with MMSE score in different groups

¹⁸F-FDG		Frontal	Parieto-occipital	Temporal
NC	r	0.151	0.131	0.493^*
	p	0.514	0.570	0.023
MCI	r	0.462^*	0.174	0.286
	p	0.027	0.428	0.187
AD	r	0.676^**	0.437	0.604^**
	p	0.002	0.062	0.006
MCI/AD	r	0.552^**	0.409^**	0.479^**
	p	0.000	0.007	0.001
NC/MCI/AD	r	0.606^**	0.537^**	0.578^**
	p	0.000	0.000	0.000

¹⁸F-florbetapir		Frontal	Parieto-occipital	Temporal
NC	r	0.048	0.213	0.307
	p	0.837	0.354	0.175
MCI	r	0.275	– 0.057	0.110
	p	0.204	0.781	0.616
AD	r	– 0.048	– 0.347	– 0.342
	p	0.845	0.145	0.152
MCI / AD	r	– 0.133	– 0.345^*	– 0.309^*
	p	0.401	0.025	0.046
NC/MCI/AD	r	– 0.227	– 0.374^**	– 0.291^*
	p	0.074	0.003	0.021

^*Correlation is significant at the 0.05 level (2-tailed). ^**Correlation is significant at the 0.01 level (2-tailed).

When evaluating individual regions, the MMSE score significantly correlated with the temporal lobe ¹⁸F-FDG GSUVr for the NC group and the frontal lobe ¹⁸F-FDG GSUVr for the MCI group. In the AD group, the ¹⁸F-FDG GSUVr for both frontal and temporal regions showed a significant positive correlation with MMSE score. In the MC/AD and NC/MCI/AD groups, the ¹⁸F-FDG GSUVr of all three regions were significantly correlated to the MMSE score. The positive correlation in AD, MCI/AD, and NC/MCI/AD were stronger than those in the NC and MCI groups. Conversely, ¹⁸F-florbetapir GSUVr only showed a significant negative correlation with MMSE score in the parieto-occipital and temporal regions for the MCI/AD and NC/MCI/AD groups. No further correlations were observed in the other regions or groups. The correlation analysis in the NC and MCI groups surprisingly indicated a trend toward positive correlation with MMSE score (See Table 3 and Fig. 4). The Steiger’s Z test indicated that both ¹⁸F-FDG and ¹⁸F-florbetapir GSUVr have a significant correlation with the MMSE score, only the correlation of ¹⁸F-FDG GSUVr of temporal region of NC/MCI/AD group is significantly stronger than that of ¹⁸F-florbetapir (Z = 2.006, p = 0.022) and in all other regions the correlation is not actually significantly stronger (see Table 4).

Table 4

Relative strength analysis with Steiger’s Z test in the regions where both ¹⁸F-FDG and ¹⁸F-florbetapir GSUVr indicated significant correlation with MMSE score

		Parieto-occipital	Temporal
MCI/AD	Z	0.406	0.885
	p	0.343	0.188
NC/MCI/AD	Z	1.346	2.006^*
	p	0.089	0.022

^*Significant at the 0.05 level (2-tailed).

DISCUSSION

In this study, we employed a global quantitative approach in order to assess the global metabolic activity and global amyloid burden of the entire cerebrum, as well as notable individual brain regions. We have previously used similar methods of global quantitative analysis to assess other neurological disorders such as temporal lobe epilepsy [29, 30], as well as diseases pertaining to the cardiovascular [31, 32], pulmonary [28], and gastrointestinal systems [33]. In this study, we also used computed tomography (CT) as a structural guide for a novel segmentation method in order to improve the quantitative accuracy of molecular brain imaging. As a result, the findings described in this paper represent the globally averaged tracer uptake of the entire affected tissue, and thus likely encompass the true burden of the disease.

Our primary finding is that both ¹⁸F-FDG and ¹⁸F-florbetapir were able to effectively discriminate demented subjects (MCI or AD) from normal controls when used in a global quantitative analysis of the supra-tentorial region. ¹⁸F-FDG uptake showed a characteristic decrease in the MCI and AD groups as compared to normal controls—indicative of global cerebral hypometabolism. This finding is well established in the literature and provides further evidence for the high efficacy of ¹⁸F-FDG in the clinical setting of dementia [20]. ¹⁸F-florbetapir also exhibited characteristic uptake patterns of increased tracer binding in demented patients, consistent with previous studies reporting the relatively high sensitivity and specificity of amyloid PET tracers, including ¹⁸F-florbetapir, at detecting the presence of plaques [20, 38]. However, as previously mentioned, the presence of amyloid plaques is necessary but not sufficient to diagnose AD. In fact, the Food and Drug Administration (FDA) has not approved ¹⁸F-florbetapir for making the positive diagnosis of AD, but only for the purpose of excluding AD from the differential diagnosis when ¹⁸F-florbetapir PET scans are negative [39]. Negative amyloid PET results are indicated to be more useful in diagnostic decision-making than positive results, confirming its primary use as an adjunct test for ruling out AD [40]. Moreover, Amyloid PET tracers are unable to detect NFT pathology—another critical biomarker involved in AD identified by the updated NIA-AA guidelines. As such, these results confirm that ¹⁸F-florbetapir cannot be used on its own to diagnose AD [10 , 42].

We also found that neither tracer was able to differentiate between MCI and AD. It has been reported by numerous studies that MCI populations exhibit a bimodal distribution of amyloid PET tracer uptake with the majority of patients resembling the uptake patterns of either AD patients or normal controls [43 –45]. One literature review reported that the bimodal distribution is skewed such that the majority of MCI patients have amyloid binding similar to that of AD patients, which is consistent with our results [25]. Although some studies have reported significant differences in amyloid PET tracer binding between MCI and AD populations [46], these results are difficult to interpret based on the heterogeneous nature of MCI [42]. Several studies have noted that amyloid binding patterns are different among patients with differing forms of MCI [13 , 47]. Similarly, in a review of the literature, Smailagic et al. found that ¹⁸F-FDG, as a single test, lacks enough accuracy to identify MCI subjects who may develop AD [48]. As such, comparative results between AD and MCI populations as a whole may differ significantly based on exclusion criteria and diagnostic thresholds.

Additionally, we report that global cerebral hypometabolism as detected by ¹⁸F-FDG is a much stronger indicator of cognitive impairment than is increased ¹⁸F-florbetapir uptake. We found that ¹⁸F-FDG uptake correlated significantly with the MMSE score for the AD, MCI/AD, and NC/MCI/AD groups. By contrast, ¹⁸F-florbetapir uptake showed a significant negative correlation with the MMSE score only in the pooled NC/MCI/AD group. This disparity is in high accordance with numerous studies indicating that amyloid PET is inadequate for assessing disease severity in patients with MCI and AD, as well as cognitive function in general [25 , 49– 53]. In fact, one longitudinal study of AD patients failed to find a significant change in amyloid PET activity over time, despite a significant decline in overall MMSE scores [54]. Other studies have shown that there is no significant difference in PiB binding between amnestic and non-amnestic MCI patients—suggesting that amyloid PET is not a good indicator of phenotypic variation [47]. By contrast, many studies have shown that suppressed ¹⁸F-FDG uptake does correlate effectively with cognitive decline [44 , 55]. Another longitudinal study found a correlation between decreasing ¹⁸F-FDG uptake over time and declining MMSE scores across normal controls, MCI patients, and AD patients [56]. While it is true that some studies have reported modest correlations between amyloid PET and cognitive performance, such findings remain controversial and lack sufficient theoretical basis [20 , 58].

In an analysis of individual brain regions, we found strong correlations in multiple patient groups between MMSE scores and ¹⁸F-FDG uptake in both the frontal and temporal regions, as well as somewhat weaker correlations in the parieto-occipital region (Table 3). This finding suggests that the severity of AD symptoms correlates well with increasingly severe neuronal dysfunction in these areas of the brain. One explanation of this finding is that the frontal and temporal regions continue to deteriorate throughout the progression of the disease at a steeper rate than the parieto-occipital regions. Of course, this pattern is expected as the frontal and temporal lobes preside over executive function and memory—two of the primary cognitive abilities affected in AD. These patterns are in congruence with previous studies showing that atrophy of the medial temporal lobe is one of the most consistent findings on MRI within AD populations [13 , 59]. Because it is known that atrophy appears on MRI much later than hypometabolism is detected by ¹⁸F-FDG-PET, atrophy of the temporal lobe suggests a correlation between lesions within this region and severe morbidity. It has also been found that the presence of hypometabolism in the frontal and temporal regions may be able to differentiate between AD and MCI in some populations [59, 60]. Thus, marked decline in these regions is likely associated with disease progression.

The regional analysis using ¹⁸F-florbetapir revealed no significant correlations between declining MMSE scores and tracer uptake within the frontal region. This is an important finding as pathological studies have reported that the frontal lobe is a critical region for Aβ plaque deposition [61]. Additionally, although we found mild correlations between MMSE scores and amyloid binding within the temporal and parieto-occipital regions, they were weaker than those found using ¹⁸F-FDG. In agreement with these findings, data from autopsy studies have shown that frontal and temporal region neocortical Aβ plaque densities do not correlate with the severity of the disease [50, 62]. It is true that other studies have found some correlations between plaque presence and cognitive decline [63, 64]. However, as we found, these correlations typically do not describe the extent of disease progression as well as the hypometabolism detected by ¹⁸F-FDG, and are likely due to the high association of plaque deposition in AD patients. One explanation presented by Landau et al. is that Aβ plaque presence is associated with the initial stages of the disease, whereas continuous declining metabolism persists throughout its entirety [64]. Still, given the high prevalence of Aβ plaques among cognitively healthy populations as well [65], it is difficult to see the utility of amyloid PET imaging for assessing the cognitively impaired. Additionally despite many studies that state amyloid imaging has an additive value in making a diagnosis of AD [66], we know the autopsy findings and other potential indicators fit poorly with amyloid PET, Høilund-Carlsen et al. surprisingly believe this statement is primarily based on circular reasoning and, hence, misleading and this examination has no role in the diagnosis of AD [67].

The relatively small sample size for each clinical group presents a limitation to this study. Another limitation is the lack of clinical follow-up to assess which radiotracer better predicts further cognitive decline in the MCI patients. Additionally, further studies should assess the efficacy of amyloid PET and ¹⁸F-FDG at evaluating the transition from MCI to AD.

Conclusion

As a result of our findings, we conclude that ¹⁸F-FDG is a superior indicator of cognitive decline in AD and MCI populations. Given the lack of correlation between MMSE score decline and ¹⁸F-florbetapir uptake in the frontal region—a region that is supposed to possess high amyloid burden—we believe that current Aβ imaging technology is a modest prognostic tool for assessing this form of dementia and would not be suitable as a stand-alone method of diagnosis for AD. By contrast, ¹⁸F-FDG-PET allows direct imaging of neuronal dysfunction and correlates well with the phenotypic changes associated with cognitive impairment. As such, we recommend a greater emphasis on applications of ¹⁸F-FDG PET imaging for the management and treatment of AD and MCI.

DISCLOSURE STATEMENT

Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/19-0220r2).

References

Taylor

, Greenlund

, McGuire

, Lu

, Croft

(2017) Deaths from Alzheimer’s Disease - United States, 1999-2014. MMWR Morb Mortal Wkly Rep 66, 521–526.

Alzheimer’s Association (2017) 2017 Alzheimer’s disease facts and figures. Alzheimers Dement 13, 325–373.

Barker

, Luis

, Kashuba

, Luis

, Harwood

, Loewenstein

, Waters

, Jimison

, Shepherd

, Sevush

, Graff-Radford

, Newland

, Todd

, Miller

, Gold

, Heilman

, Doty

, Goodman

, Robinson

, Pearl

, Dickson

, Duara

(2002) Relative frequencies of Alzheimer disease, Lewy body, vascular and frontotemporal dementia, and hippocampal sclerosis in the State of Florida Brain Bank. Alzheimer Dis Assoc Disord 16, 203–212.

Wilson

, Segawa

, Boyle

, Anagnos

, Hizel

, Bennett

(2012) The natural history of cognitive decline in Alzheimer’s disease. Psychol Aging 27, 1008–1017.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–944.

Petersen

, Caracciolo

, Brayne

, Gauthier

, Jelic

, Fratiglioni

(2014) Mild cognitive impairment: A concept in evolution. J Intern Med 275, 214–228.

Albert

, Dekosky

, Dickson

, Dubois

, Feldman

, Fox

, Gamst

, Holtzman

, Jagust

, Petersen

, Snyder

, Carrillo

, Thies

, Phelps

(2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 270–279.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

, Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

Knopman

, DeKosky

, Cummings

, Chui

, Corey-Bloom

, Relkin

, Small

, Miller

, Stevens

(2001) Practice parameter: Diagnosis of dementia (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 56, 1143–1153.

10.

Hyman

, Phelps

, Beach

, Bigio

, Cairns

, Carrillo

, Dickson

, Duyckaerts

, Frosch

, Masliah

, Mirra

, Nelson

, Schneider

, Thal

, Thies

, Trojanowski

, Vinters

H V

, Montine

(2012) National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement 8, 1–13.

11.

Silverman

DHS

, Small

, Chang

, Lu

, de Aburto

MAK

, Chen

, Czernin

, Rapoport

, Pietrini

, Alexander

, Schapiro

, Jagust

, Hoffman

, Welsh-Bohmer

, Alavi

, Clark

, Salmon

, de Leon

, Mielke

, Cummings

, Kowell

, Gambhir

, Hoh

, Phelps

(2001) Positron emission tomography in evaluation of dementia. JAMA 286, 2120.

12.

Patwardhan

, McCrory

, Matchar

, Samsa

, Rutschmann

(2004) Alzheimer disease: Operating characteristics of PET–a meta-analysis. Radiology 231, 73–80.

13.

Mosconi

, Tsui

, Herholz

, Pupi

, Drzezga

, Lucignani

, Reiman

, Holthoff

, Kalbe

, Sorbi

, Diehl-Schmid

, Perneczky

, Clerici

, Caselli

, Beuthien-Baumann

, Kurz

, Minoshima

, de Leon

(2008) Multicenter dtandardized 18F-FDG PET diagnosis of mild cognitive impairment, Alzheimer’s disease, and other dementias. J Nucl Med 49, 390–398.

14.

Mosconi

(2005) Brain glucose metabolism in the early and specific diagnosis of Alzheimer’s disease: FDG-PET studies in MCI and AD. Eur J Nucl Med Mol Imaging 32, 486–510.

15.

Hyman

(2011) Amyloid-dependent and amyloid-independent stages of Alzheimer disease. Arch Neurol 68, 1062–1064.

16.

Drzezga

, Grimmer

, Riemenschneider

, Lautenschlager

, Siebner

, Alexopoulus

, Minoshima

, Schwaiger

, Kurz

(2005) Prediction of individual clinical outcome in MCI by means of genetic assessment and (18)F-FDG PET. J Nucl Med 46, 1625–1632.

17.

Dupont

, Santiago Ribeiro

, Guilloteau

, Arlicot

(2017) β-amyloid PET neuroimaging: A review of radiopharmaceutical development. Méd Nucl 41, 27–35.

18.

Wolk

, Zhang

, Boudhar

, Clark

, Pontecorvo

, Arnold

(2012) Amyloid imaging in Alzheimer’s disease: Comparison of florbetapir and Pittsburgh compound-B positron emission tomography. J Neurol Neurosurg Psychiatry 83, 923–926 .

19.

Landau

, Breault

, Joshi

, Pontecorvo

, Mathis

, Jagust

, Mintun

(2013) Amyloid-imaging with Pittsburgh Compound B and Florbetapir: Comparing radiotracers and quantification methods. J Nucl Med 54, 70–77.

20.

Rice

, Bisdas

(2017) The diagnostic value of FDG and amyloid PET in Alzheimer’s disease—A systematic review. Eur J Radiol 94, 16–24.

21.

Mintun

, Joshi

, Pontecorvo

, Lu

, Krautkramer

, Carpenter

, Bedell

, Fleisher

, Skovronsky

, Beach

, Reiman

, Doraiswamy

, Sadowsky

, Coleman

, Clark

, Schneider

, Sabbagh

, Arora

, Flitter

(2012) Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-β plaques: A prospective cohort study. Lancet Neurol 11, 669–678.

22.

Wong

, Rosenberg

, Zhou

, Kumar

, Raymont

, Ravert

, Dannals

, Nandi

, Brasic

, Ye

, Hilton

, Lyketsos

, Kung

, Joshi

, Skovronsky

, Pontecorvo

(2010) In vivo imaging of amyloid deposition in Alzheimer disease using the radioligand 18F-AV-45 (Florbetapir F 18). J Nucl Med 51, 913–920.

23.

Clark

(2011) Use of Florbetapir-PET for imaging β-amyloid pathology. JAMA 305, 275.

24.

Aizenstein

, Nebes

, Saxton

, Price

, Mathis

, Tsopelas

, Ziolko

, James

, Snitz

, Houck

, Bi

, Cohen

, Lopresti

, DeKosky

, Halligan

, Klunk

(2008) Frequent amyloid deposition without significant cognitive impairment among the elderly. Arch Neurol 65, 1509–1517.

25.

Rowe

, Ng

, Ackermann

, Gong

, Pike

, Savage

, Cowie

, Dickinson

, Maruff

, Darby

, Smith

, Woodward

, Merory

, Tochon-Danguy

, O’Keefe

, Klunk

, Mathis

, Price

, Masters

, Villemagne

(2007) Imaging β-amyloid burden in aging and dementia. Neurology 68, 1718–1725.

26.

Jack

, Knopman

, Jagust

, Shaw

, Aisen

, Weiner

, Petersen

, Trojanowski

(2010) Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol 9, 119–128.

27.

Minoshima

, Giordani

, Berent

, Frey

, Foster

, Kuhl

(1997) Metabolic reduction in the posterior cingulate cortex in very early Alzheimer’s disease. Ann Neurol 42, 85–94.

28.

Abdulla

, Salavati

, Saboury

, Basu

, Torigian

, Alavi

(2014) Quantitative assessment of global lung inflammation following radiation therapy using FDG PET/CT: A pilot study. Eur J Nucl Med Mol Imaging 41, 350–356.

29.

Peter

, Houshmand

, Werner

, Rubello

, Alavi

(2016) Applications of global quantitative18F-FDG-PET analysis in temporal lobe epilepsy. Nucl Med Commun 37, 223–30.

30.

Peter

, Houshmand

, Werner

, Rubello

, Alavi

(2016) Novel assessment of global metabolism by 18F-FDG-PET for localizing affected lobe in temporal lobe epilepsy. Nucl Med Commun 37, 882–887.

31.

Basu

, Hoilund-Carlsen

, Alavi

(2012) Assessing global cardiovascular molecular calcification with 18F-fluoride PET/CT: Will this become a clinical reality and a challenge to CT calcification scoring? Eur J Nucl Med Mol Imaging 39, 660–664.

32.

Beheshti

, Saboury

, Mehta

, Torigian

, Werner

, Mohler

, Wilensky

, Newberg

, Basu

, Langsteger

, Alavi

(2011) Detection and global quantification of cardiovascular molecular calcification by fluoro-18-fluoride positron emission tomography/computed tomography-A novel concept. Hell J Nucl Med 14, 114–120.

33.

Saboury

, Salavati

, Brothers

, Basu

, Kwee

, Lam

MGEH

, Hustinx

, Louis

, Torigian

, Alavi

(2014) FDG PET/CT in Crohn’s disease: Correlation of quantitative FDG PET/CT parameters with clinical and endoscopic surrogate markers of disease activity. Eur J Nucl Med Mol Imaging 41, 605–614.

34.

Saboury

, Parsons

, Moghbel

, Rubello

, Brothers

, Torigian

, Werner

, Houshmand

, Basu

, Lam

MGEH

, Alavi

(2016) Quantification of aging effects upon global knee inflammation by18F-FDG-PET. Nucl Med Commun 37, 254–258.

35.

Langbaum

JBS

, Chen

, Lee

, Reschke

, Fleisher

, Alexander

, Foster

, Michael

, Koeppe

, Jagust

, Reiman

; Alzheimer’s Disease Neuroimaging Initiative (2010) Categorical and correlational analyses of baseline fluorodeoxyglucose positron emission tomography images from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Neuroimage 45, 1107–1116.

36.

Clark

, Schneider

, Bedell

, Beach

, Bilker

, Mintun

M a

, Pontecorvo

, Hefti

, Carpenter

, Flitter

, Krautkramer

, Kung

, Coleman

, Doraiswamy

, Fleisher

, Sabbagh

, Sadowsky

, Reiman

, Zehntner

, Skovronsky

(2011) Use of florbetapir-PET for imaging beta-amyloid pathology. JAMA 305, 275–283.

37.

Dukart

, Mueller

, Horstmann

, Vogt

, Frisch

, Barthel

, Becker

, Möller

, Villringer

, Sabri

, Schroeter

(2010) Differential effects of global and cerebellar normalization on detection and differentiation of dementia in FDG-PET studies. Neuroimage 49, 1490–1495.

38.

Morris

, Chalkidou

, Hammers

, Peacock

, Summers

, Keevil

(2016) Diagnostic accuracy of 18F amyloid PET tracers for the diagnosis of Alzheimer’s disease: A systematic review and meta-analysis. Eur J Nucl Med Mol Imaging 43, 374–385.

39.

Yang

, Rieves

, Ganley

(2012) Brain amyloid imaging — FDA approval of Florbetapir F18 Injection. N Engl J Med 367, 885–887.

40.

Fantoni

, Chalkidou

, O’Brien

, Farrar

, Hammers

(2018) A systematic review and aggregated analysis on the impact of amyloid PET brain imaging on the diagnosis, diagnostic confidence, and management of patients being evaluated for Alzheimer’s disease. J Alzheimers Dis 63, 783–796.

41.

Kepe

, Moghbel

, Långström

, Zaidi

, Vinters

H V.

, Huang

S-CC

, Satyamurthy

, Doudet

, Mishani

, Cohen

, Høilund-Carlsen

, Alavi

, Barrio

, Långströmd

, Zaidi

, Vinters

, Huang

S-CC

, Satyamurthy

, Doudet

, Mishani

, Cohen

, Høilund-Carlsen

, Alavi

, Barrio

(2013) Amyloid-β positron emission tomography imaging probes: A critical review. J Alzheimers Dis 36, 613–631.

42.

Aschenbrenner

, Gordon

, Benzinger

TLS

, Morris

, Hassenstab

(2018) Influence of tau PET, amyloid PET, and hippocampal volume on cognition in Alzheimer disease. Neurology 91, e859–e866.

43.

Forsberg

, Engler

, Almkvist

, Blomquist

, Hagman

, Wall

, Ringheim

, Långström

, Nordberg

(2008) PET imaging of amyloid deposition in patients with mild cognitive impairment. Neurobiol Aging 29, 1456–1465.

44.

Jack

, Lowe

, Senjem

, Weigand

, Kemp

, Shiung

, Knopman

, Boeve

, Klunk

, Mathis

, Petersen

(2008) 11C PiB and structural MRI provide complementary information in imaging of Alzheimer’s disease and amnestic mild cognitive impairment. Brain 131, 665–680.

45.

Rabinovici

, Jagust

(2009) Amyloid imaging in aging and dementia: Testing the amyloid hypothesis in vivo. Behav Neurol 21, 117–128.

46.

Johnson

, Sperling

, Gidicsin

, Carmasin

, Maye

, Coleman

, Reiman

, Sabbagh

, Sadowsky

, Fleisher

, Murali Doraiswamy

, Carpenter

, Clark

, Joshi

, Lu

, Grundman

, Mintun

, Pontecorvo

, Skovronsky

(2013) Florbetapir (F18-AV-45) PET to assess amyloid burden in Alzheimer’s disease dementia, mild cognitive impairment, and normal aging. Alzheimers Dement 9, S72–S83.

47.

Wolk

, Price

, Saxton

, Snitz

, James

, Lopez

, Aizenstein

, Cohen

, Weissfeld

, Mathis

, Klunk

, DeKoskym

(2009) Amyloid imaging in mild cognitive impairment subtypes. Ann Neurol 65, 557–568.

48.

Smailagic

, Vacante

, Hyde

, Martin

, Ukoumunne

, Sachpekidis

(2015) 18F-FDG PET for the early diagnosis of Alzheimer’s disease dementia and other dementias in people with mild cognitive impairment (MCI). Cochrane Database Syst Rev CD010632.

49.

Newberg

, Arnold

, Wintering

, Rovner

, Alavi

(2012) Initial clinical comparison of 18F-Florbetapir and 18F-FDG PET in patients with Alzheimer disease and controls. J Nucl Med 53, 902–907.

50.

Bennett

, Cochran

, Saper

, Leverenz

, Gilley

, Wilson

(1993) Pathological changes in frontal cortex from biopsy to autopsy in Alzheimer’s disease. Neurobiol Aging 14, 589–596.

51.

Reed

, Weiner

, Jagust

, Madison

, Marchant

, DeCarli

, Chui

(2011) Cerebrovascular disease, beta-amyloid, and cognition in aging. Neurobiol Aging 33, 1006.e25–1006.e36.

52.

Pike

, Ellis

, Villemagne

, Good

, Chételat

, Ames

, Szoeke

, Laws

, Verdile

, Martins

, Masters

, Rowe

(2011) Cognition and beta-amyloid in preclinical Alzheimer’s disease: Data from the AIBL study. Neuropsychologia 49, 2384–2390.

53.

Mosconia

, Bertia

, Glodzika

, Pupib

, De Santia

, De Leon

(2011) Pre-clinical detection of Alzheimer’s disease using FDG-PET, with or without amyloid imaging. Alzheimers Dement 6, 843–854.

54.

Engler

, Forsberg

, Almkvist

, Blomquist

, Larsson

, Savitcheva

, Wall

, Ringheim

, Långström

, Nordberg

(2006) Two-year follow-up of amyloid deposition in patients with Alzheimer’s disease. Brain 129, 2856–2866.

55.

Edison

, Archer

, Hinz

, Hammers

, Pavese

, Tai

, Hotton

, Cutler

, Fox

, Kennedy

, Rossor

, Brooks

(2007) Amyloid, hypometabolism, and cognition in Alzheimer disease: An [11C]PIB and [18F]FDG PET study. Neurology 68, 501–508.

56.

Ossenkoppele

, Tolboom

, Foster-Dingley

, Adriaanse

, Boellaard

, Yaqub

, Windhorst

, Barkhof

, Lammertsma

, Scheltens

, Van Der Flier

, Van Berckel

BNM

(2012) Longitudinal imaging of Alzheimer pathology using [11C]PIB, [18F]FDDNP and [18F]FDG PET. Eur J Nucl Med Mol Imaging 39, 990–1000.

57.

Dumurgier

, Hanseeuw

, Hatling

, Judge

, Schultz

, Chhatwal

, Blacker

, Sperling

, Johnson

, Hyman

, Gómez-Isla

(2017) Alzheimer’s disease biomarkers and future decline in cognitive normal older adults. J Alzheimers Dis 60, 1451–1459.

58.

Moghbel

, Saboury

, Basu

, Metzler

, Torigian

, Långström

, Alavi

(2012) Amyloid-β imaging with PET in Alzheimer’s disease: Is it feasible with current radiotracers and technologies? Eur J Nucl Med Mol Imaging 39, 202–208.

59.

Nestor

, Fryer

, Smielewski

, Hodges

(2003) Limbic hypometabolism in Alzheimer’s disease and mild cognitive impairment. Ann Neurol 54, 343–351.

60.

De Santi

, De Leon

, Rusinek

, Convit

, Tarshish

, Roche

, Tsui

, Kandil

, Boppana

, Daisley

, Wang

, Schlyer

, Fowler

(2001) Hippocampal formation glucose metabolism and volume losses in MCI and AD. Neurobiol Aging 22, 529–539.

61.

Altmann

, Ng

, Landau

, Jagust

, Greicius

(2015) Regional brain hypometabolism is unrelated to regional amyloid plaque burden. Brain 138, 3734–3746.

62.

Mann

DMA

, Marcyniuk

, Yates

, Neary

, Snowden

(1988) The progression of the pathological changes of Alzheimer’s disease in frontal and temporal neocortex examined both at biopsy and at autopsy. Neuropathol Appl Neurobiol 14, 177–195.

63.

Blessed

, Tomlinson

, Roth

(1968) The association between quantitative measures of dementia and of senile change in the cerebral grey matter of elderly subjects. Br J Psychiatry 114, 797–811.

64.

Landau

, Mintun

, Joshi

, Koeppe

, Petersen

, Aisen

, Weiner

, Jagust

(2012) Amyloid deposition, hypometabolism, and longitudinal cognitive decline. Ann Neurol 72, 578–586.

65.

Paroni

, Bisceglia

, Seripa

(2019) Understanding the amyloid hypothesis in Alzheimer’s disease. J Alzheimers Dis 68, 493–510.

66.

Brendel

, Schnabel

, Schönecker

, Wagner

, Brendel

, Meyer-Wilmes

, Unterrainer

, Schildan

, Patt

, Prix

, Ackl

, Catak

, Pogarell

, Levin

, Danek

, Buerger

, Bartenstein

, Barthel

, Sabri

, Rominger

(2017) Additive value of amyloid-PET in routine cases of clinical dementia work-up after FDG-PET. Eur J Nucl Med Mol Imaging 44, 2239–2248.

67.

Høilund-Carlsen

, Barrio

, Gjedde

, Werner

, Alavi

(2018) Circular inference in dementia diagnostics. J Alzheimers Dis 63, 69–73.

18 F-FDG Is a Superior Indicator of Cognitive Performance Compared to 18 F-Florbetapir in Alzheimer’s Disease and Mild Cognitive Impairment Evaluation: A Global Quantitative Analysis

Abstract

Background:

Objectives:

Methods:

Results:

Conclusions:

Keywords

INTRODUCTION

METHODS

Subjects

18F-FDG PET acquisition

18F-florbetapir (Amyvid) PET acquisition

Image analysis and quantification

Data analysis

RESULTS

DISCUSSION

Conclusion

DISCLOSURE STATEMENT

References

¹⁸F-FDG PET acquisition

¹⁸F-florbetapir (Amyvid) PET acquisition