- Journal List
- HHS Author Manuscripts
- PMC4254054
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
Alzheimers Dement. Author manuscript; available in PMC 2015 Nov 1.
Published in final edited form as:
Alzheimers Dement. 2014 Nov; 10(6): 630–636.
Published online 2014 Jun 28. doi:10.1016/j.jalz.2014.03.004
PMCID: PMC4254054
NIHMSID: NIHMS585196
PMID: 24985533
Keith A. Josephs, MD, MST, MSc,1,2 Joseph R. Duffy, PhD,3 Edythe A. Strand, PhD,3 Mary M. Machulda, PhD,4 Matthew L. Senjem, MS,5 Val J. Lowe, MD,7 Clifford R. Jack, Jr., MD,6 and Jennifer L. Whitwell, PhD6
Author information Copyright and License information PMC Disclaimer
The publisher's final edited version of this article is available at Alzheimers Dement
Abstract
Background
Apolipoprotein E epsilon 4 (APOE4) is a risk factor forβ-amyloid deposition in Alzheimer’s dementia. Its influenceon β-amyloid deposition in speech and language disorders, includingprimary progressive aphasia (PPA), is unclear.
Methods
One hundred and thirty subjects with PPA or speech apraxia underwentAPOE genotyping and Pittsburgh compound B (PiB) PET scanning. Therelationship between APOE4 and PiB status, as well as severity and regionaldistribution of PiB, was assessed.
Results
Forty-five subjects had an APOE4 allele and 60 subjects werePiB-positive. The odds ratio for a subject with APOE4 being PiB-positivecompared to a subject without APOE4 being PiB-positive was 10.2(4.4–25.5, p<0.0001). APOE4 status did not influenceregional PiB distribution or severity.
Conclusion
APOE4 increases the risk of β-amyloid deposition in PPA andspeech apraxia, but does not influence regional β-amyloiddistribution or severity.
Keywords: Apolipoprotein, Pittsburgh Compound B, primary progressive aphasia, logopenic aphasia, speech apraxia
1. Introduction
The presence of the apolipoprotein E epsilon 4 (APOE4) allele is a riskfactor for Alzheimer’s disease (AD)[1–3] and hence forβ-amyloid deposition. While β-amyloid deposition is usuallyassociated with episodic memory loss and Alzheimer’s dementia[4], patients with progressive speech or languagedisorders have also been reported to have AD or β-amyloid deposition.Patients with early and prominent deficits in language are generally diagnosed withone of three variants of primary progressive aphasia (PPA)[5]. The three variants include logopenic PPA (lvPPA) in whichpatients present with anomia, poor word retrieval in spontaneous speech, difficultyrepeating sentences and phonological errors, semantic PPA (svPPA) in which patientspresent with anomia and loss of word knowledge, and agrammatic PPA (agPPA) in whichpatients have difficulty with grammar and syntax, and can also have a motor speechdisorder known as apraxia of speech[6, 7]. In addition, patients with early andprominent deficits in speech in which the presenting disorder is dominated byapraxia of speech, or where apraxia of speech is the sole presenting feature[8], can be classified as progressive apraxia ofspeech (PAOS)[8, 9]. Hence progressive speech and language disorders can bebroadly classified as PPA and PAOS. Beta-amyloid deposition is strongly associatedwith lvPPA[10–12], but has also been observed to occur in patients withsvPPA[13], agPPA[11] and PAOS[8, 9], although these latter PPA variants and PAOSare usually associated with frontotemporal lobar degeneration pathologies[10, 14–16]. It is unclearwhether the APOE4 allele is a risk factor for the presence of β-amyloiddeposition in PPA or PAOS, or within the PPA variants. It is also unclear whetherAPOE4 influences the distribution or severity of β-amyloid deposition inthese patients. Understanding the relationship between the APOE4 genotype andβ-amyloid deposition in patients with speech and language disorders isimportant to better understand the underlying biological mechanisms that may accountfor pathological variability in these subjects.
The aim of this study was to use a large cohort of 130 patients with PPA orPAOS to determine the relationship between the APOE4 allele and β-amyloiddeposition. We hypothesized that the presence of the APOE4 allele would stronglyincrease the odds of β-amyloid deposition, but would not influenceβ-amyloid severity or distribution.
2. Materials and Methods
2.1. Subjects
Between February 2010 and February 2013 we consecutively recruitedsubjects with a progressive speech or language disorder who presented to theDepartment of Neurology, Mayo Clinic, Rochester MN (n=130). All 130 subjectsunderwent APOE genotyping as previously described[17, 18] andcompleted 11C Pittsburgh compound B (PiB) PET scanning fordetermination of β-amyloid status (see below).
All 130 subjects underwent detailed speech and language evaluations, aspreviously described[8], including theWestern Aphasia Battery[19] in which theAphasia Quotient is a measure of aphasia severity, and neurological testing thatincluded the Mini-Mental State Examination[20] as a measure of global cognitive impairment. Subjects wereclassified as PAOS, or as one of the three well-recognized PPA variants (agPPA,svPPA or lvPPA), based on qualitative and quantitative speech and language datawhich was influenced by the PPA consensus guidelines[5], and on recommended criteria for the diagnosis ofPAOS[8, 9]. Subjects who met criteria for PPA but could not be classifiedinto one of the three PPA variants were labeled as unclassified (ucPPA).
The study was approved by the Mayo Clinic institutional review board andall patients consented for enrolment into the study.
2.2. Imaging analysis
All PiB-PET scans were performed using a PET/CT scanner (GeneralElectric, Milwaukee, WI) operating in three-dimensional mode. Each subject wasinjected with approximately 614 MBq of PiB and after a 40 minute uptake period,a 20 minute PiB scan was obtained. All subjects also underwent MRI at 3.0 Tesla,which included a 3D magnetization prepared rapid acquisition gradient echo(MPRAGE) sequence, within two days of the PiB-PET scan.
A global PiB ratio[21] wascalculated for each subject in order to classify subjects as PiB-positive orPiB-negative. All PiB-PET images were co-registered to the MPRAGE for eachpatient, and the automated anatomical labeling atlas[22] was used to calculate median PiB uptake for thefollowing 6 cortical regions-of-interest: temporal lobe, parietal lobe,posterior cingulate/precuneus, anterior cingulate, prefrontal cortex, andoccipital lobe (left and right were combined for all regions). Median PiB uptakein each of the 6 regions was divided by median cerebellar uptake to createuptake ratios. A global cortical PiB retention summary was formed by calculatingmedian uptake ratio values across all 6 regions. Patients were classified asPiB-positive using a cortical-to-cerebellar (SUVR) ratio cut-point of 1.5[21].
In addition, a voxel-level comparison of PiB-PET regional distributionwas performed within all PiB-positive subjects. All voxels in the PiB-PET imagewere divided by median uptake of the cerebellum to form PiB uptake ratio images.The PiB-PET uptake ratio images were then normalized to a customized templateusing the normalization parameters from the MPRAGE normalization. Two-sidedt-tests were used to compare all the PiB-positive subjects with an APOE4 alleleand the PiB-positive subjects without an APOE4 allele to an age and gendermatched control cohort. The control cohort consisted of 30 healthy subjects thathad all undergone an identical PiB-PET and MRI acquisition and were allPiB-negative. Results were corrected for multiple comparisons using the familywise error (FWE) correction at p<0.05. Direct comparisons were alsoperformed between the APOE4 negative and positive disease groups, assesseduncorrected for multiple comparisons at p<0.001 with an extent thresholdof 100 voxels. These analyses were also repeated using only PiB-positive lvPPAsubjects given that the number of PiB-positive lvPPA subjects were large enoughfor analysis, and that the vast majority of PiB-positive PPA subjects were infact lvPPA. Age and gender was included as a covariate in all analyses.
2.3. Statistical analysis
Statistical analyses were performed utilizing JMP computer software (JMPSoftware, version 9.0.0; SAS Institute Inc., Cary, NC) with significanceassessed at p<0.05. Odds ratios and confidence intervals were calculatedusing logistic regression for PPA, PAOS, and for the secondary analysis for eachPPA variant. For the svPPA group, in order to calculate a conservative oddsratio, we had to artificially replace the 0 cell count with a count of 1, basedon published recommendation[23].Mann-Whitney-U test was used to compare SUVR ratios between APOE4 positivePiB-positive subjects and APOE4 negative PiB-positive subjects. Given the strongassociation between lvPPA and β-amyloid deposition, we performedadditional analyses for the lvPPA group. First, we compared the lvPPA group toall other PPA variants and PAOS. Secondly, within the lvPPA group, we compareddemographic and clinical features for all the PiB-positive subjects, stratifiedby APOE4 status (i.e. APOE4 positive PiB-positive lvPPA versus APOE4 negativePiB-positive lvPPA).
3. Results
Results for APOE4 and PiB status by clinical diagnosis are shown in Table 1. Of the 130 subjects with a progressivespeech and language disorder, 45 (34%) had at least one APOE4 allele whilethe remaining 85 did not. Sixty subjects (46%) were PiB-positive. Given anAPOE4 positive status a subject was more than 10X more likely to be PiB-positivethan if the APOE4 allele was not present (p<0.0001). Seven subjects wereAPOE 4/4 hom*ozygotes, and all 100% of these subjects were PiB-positive.After excluding the 53 lvPPA subjects, given an APOE4 positive status a subject was13X more likely to be PiB-positive than if the APOE4 allele was not present(OR=13.0; CI: 3.3–51.0; p<0.0001).
Table 1
Percentages of subjects with PAOS and PPA variants with the APOE4 allelestratified by PiB status
| Diagnosis | All | All | APOE4 + | APOE4 − | Odds ratio & p-values* | ||
|---|---|---|---|---|---|---|---|
| APOE4 + | PiB-positive | PiB-negative | PiB-positive | PiB-negative | PiB-positive | ||
| All subjects (n=130) | 45 (34%) | 60 (46%) | 9 (20%) | 36 (80%) | 61 (72%) | 24 (28%) | 10.2 (CI: 4.4 – 25.5); p<0.0001 |
| PAOS (n=39) | 6 (15%) | 7 (18%) | 3 (50%) | 3 (50%) | 29 (88%) | 4 (12%) | 7.3 (CI: 1.1 – 54.1); p=0.04 |
| PPA (n=91) | 39 (43%) | 53 (58%) | 6 (15%) | 33 (85%) | 32 (62%) | 20 (38%) | 8.8 (CI: 3.3 – 26.8); p<0.0001 |
| agPPA (n=12) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 12 (100%) | 0 (0%) | N/A |
| † svPPA (n=14) | 6 (43%) | 3 (21%) | 3 (50%) | 3 (50%) | 8 (100%) | 0 (0%) | 8.0 (CI: 2.2 – 203.3); p=0.01 |
| lvPPA (n=53) | 30 (57%) | 47 (89%) | 2 (7%) | 28 (93%) | 4 (17%) | 19 (83%) | 2.9 (CI: 0.5 – 22.8); p=0.22 |
| ucPPA (n=12) | 3 (25%) | 4 (33%) | 1 (33%) | 2 (67%) | 8 (89%) | 2 (11%) | 16.0 (CI: 0.8 – 73.0); p=0.07 |
Open in a separate window
*Odds ratio’s and p values are for subjects with APOE4 beingPiB-positive compared to subjects without APOE4 being PiB-positive.
†To calculate odds ratio and CI for this group we artificially added1 APOE4 negative PiB-positive subject as a conservative option.
PPA = primary progressive aphasia; agPPA = agrammatic variant ofprimary progressive aphasia; lvPPA = logopenic variant of primaryprogressive aphasia; svPPA = semantic variant of primary progressiveaphasia; ucPPA = unclassified PPA; PAOS = progressive apraxia of speech; CI= confidence interval
3.1. PPA
Of the 91 PPA subjects, 39 (43%) had at least one APOE4 alleleand 53 (58%) were PiB-positive. Given an APOE4 positive status a PPAsubject was almost 9X times more likely to be PiB-positive than if the APOE4allele was not present (p<0.0001).
Within the PPA variants, the proportion of subjects with at least oneAPOE4 allele was highest in lvPPA (57%), followed by svPPA(43%), ucPPA (25%) and agPPA (0%). Similarly, theproportion of PiB-positive subjects was highest in lvPPA (89%), followedby ucPPA (25%), svPPA (21%), and agPPA (0%). Given anAPOE4 positive status, the svPPA and ucPPA subjects were more likely to bePiB-positive than if the APOE4 allele was not present (p=0.01 and p=0.07). ThelvPPA subjects did not have significant odds ratios. Six of the lvPPA subjectswere APOE 4/4 hom*ozygotes, and all were PiB-positive. All agPPA subjects wereAPOE4 negative and all were PiB-negative.
Subjects with lvPPA were more likely to be APOE4 positive than all otherspeech or language subjects combined (57% vs. 23%;p<0.0001). Subjects with lvPPA were also more likely to be PiB-positivecompared to all other speech or language subjects combined (89% vs.20%; p<0.0001). After accounting for APOE4 status, the oddsratio for a subject with lvPPA to be PiB-positive compared to a subject with anyof the other speech or language syndromes being PiB-positive was 33 (CI:11.7–113.6; p<0.0001). Demographic and clinical features of allsubjects by APOE4 status are shown in Table2. Notable observations include the fact that the PiB-positive lvPPAsubjects without an APOE4 allele were on average 10 years younger than the lvPPAPiB-positive subjects with an APOE4 allele (p=0.03). No other demographic orclinical differences were observed between the PiB-positive subjects stratifiedby APOE4 status.
Table 2
Demographics and clinical features of all subjects and by APOE4status
| Diagnosis | All | APOE4 + | APOE4 − | |||
|---|---|---|---|---|---|---|
| PiB-negative | PiB-positive | PiB-negative | PiB-positive | |||
| Age, years | All (n=130) | 69 (62–74) | 62 (59–68) | 70 (65–74) | 70 (65–75) | 66 (56–74) |
| PAOS (n=39) | 73 (64–78) | 67 (63–71) | 74 (67–74) | 73 (65–79) | 76 (70–78) | |
| PPA = (n=91) | 68 (62–73) | 61 (58–67) | 70 (65–73) | 69 (65–72) | 62 (56–72) | |
| agPPA (n=12) | 70 (65–72) | − | − | 70 (65–72) | − | |
| svPPA (n=14) | 69 (63–72) | 60 (59–61) | 70 (69–73) | 71 (69–73) | − | |
| lvPPA (n=53) | 68 (70–73) | 63 (60–65) | 70 (65–73) | 66 (63–68) | 60 (56–71) | |
| ucPPA (n=12) | 70 (67–73) | 70 | 66 (64–73) | 70 (65–73) | 76 | |
| Gender, % F | All (n=130) | 67 (52%) | 4 (44%) | 18 (50%) | 31 (51%) | 14 (58%) |
| PAOS (n=39) | 20 (51%) | 1 (33%) | 1 (33%) | 14 (48%) | 4 (100%) | |
| PPA = (n=91) | 47 (52%) | 3 (50%) | 17 (52%) | 17 (53%) | 10 (50%) | |
| agPPA (n=12) | 8 (67%) | − | − | 8 (67%) | − | |
| svPPA (n=14) | 7 (50%) | 2 (67%) | 1 (33%) | 4 (50%) | − | |
| lvPPA (n=53) | 26 (49%) | 0 (0%) | 14 (50%) | 2 (50%) | 10 (53%) | |
| ucPPA (n=12) | 6 (50%) | 1 (100%) | 2 (100%) | 3 (38%) | 0 (0%) | |
| Illness duration, | All (n=130) | 3.0 (2.0–4.5) | 2.0 (2.0–4.0) | 4.0 (3.0–5.0) | 3.0 (2.0–4.0) | 3.3 (2.4–4.3) |
| years | PAOS (n=39) | 3.5 (2.2–4.8) | 4.0 (3.8–5.0) | 4.0 (3.0–7.0) | 3.0 (2.0–4.8) | 3.5 (3.0–5.0) |
| PPA = (n=91) | 3.0 (2.0–4.0) | 2.0 (1.6–2.0) | 4.0 (3.0–5.0) | 2.8 (1.9–4.0) | 3.3 (2.0–4.3) | |
| agPPA (n=12) | 2.5 (1.4–3.6) | − | − | 2.5 (1.4–3.6) | − | |
| svPPA (n=14) | 4.0 (2.3–5.0) | 1.5 (1.3–3.3) | 8.0 (6.0–11.0) | 3.5 (2.8–4.3) | − | |
| lvPPA (n=53) | 3.5 (2.0–5.0) | 2.0 (2.0–2.0) | 4.0 (3.0–5.0) | 2.5 (2.0–3.0) | 3.5 (2.3–4.5) | |
| ucPPA (n=12) | 2.0 (1.4–3.0) | 2.0 | 2.5 (2.3–2.8) | 1.8 (1.0–3.3) | 2.0 | |
| MMSE (/30) | All (n=130) | 27 (23–29) | 27 (26–29) | 24 (15–27) | 29 (27–30) | 25 (21–28) |
| PAOS (n=39) | 29 (28–30) | 29 (28–29) | 27 (25–28) | 29 (28–30) | 29 (28–29) | |
| PPA = (n=91) | 25 (22–28) | 27 (25–28) | 23 (15–27) | 28 (26–29) | 24 (20–25) | |
| agPPA (n=12) | 29 (24–29) | − | − | 29 (24–29) | − | |
| svPPA (n=14) | 28 (26–29) | 28 (27–29) | 16 (15–23) | 29 (28–29) | − | |
| lvPPA (n=53) | 24 (15–27) | 27 (26–27) | 24 (15–26) | 27 (19–28) | 24 (18–26) | |
| ucPPA (n=12) | 28 (24–28) | 14 | 26 (24–27) | 28 (27–29) | 21 | |
| WAB AQ (/100) | All (n=130) | 88 (78–95) | 90 (73–96) | 83 (73–88) | 93 (85–96) | 84 (72–89) |
| PAOS (n=39) | 96 (87–97) | 96 (77–97) | 83 (82–89) | 96 (93–97) | 87 (82–91) | |
| PPA = (n=91) | 85 (73–92) | 90 (77–94) | 83 (73–84) | 90 (79–93) | 83 (63–89) | |
| agPPA (n=12) | 84 (72–89) | − | − | 84 (72–89) | − | |
| svPPA (n=14) | 92 (80–95) | 95 (84–97) | 64 (54–79) | 92 (89–94) | − | |
| lvPPA (n=53) | 83 (73–88) | 90 (90–90) | 83 (73–87) | 77 (66–81) | 83 (62–89) | |
| ucPPA (n=12) | 93 (90–95) | 55 | 94 (94–95) | 94 (92–95) | 78 | |
Open in a separate window
Data shown as median and interquartile range
PPA = primary progressive aphasia; agPPA = agrammatic variant ofprimary progressive aphasia; lvPPA = logopenic variant of primaryprogressive aphasia; svPPA = semantic variant of primary progressiveaphasia; ucPPA = unclassified PPA; PAOS = progressive apraxia of speech
3.2. PAOS
Of the 39 PAOS subjects, six (15%) had at least one APOE4 alleleand seven (18%) were PiB-positive. Given an APOE4 positive status, aPAOS subject was more than 7X more likely to be PiB-positive than if the APOE4allele was not present (p=0.04). One PAOS subject was APOE 4/4 hom*ozygous andwas PiB-positive. The APOE4 allele frequency in the PAOS group was significantlydifferent from the frequency observed in PPA (43%) (P=0.002) but was notdifferent from the APOE4 allele frequency in PPA when the lvPPA subgroup wasexcluded from PPA (9/38=24%) (p=0.36).
3.3. Imaging findings
In the voxel-level analyses, the regional distribution of PiB-PET uptakewas very similar in the PiB-positive subjects with and without an APOE4 allele,with widespread PiB-PET uptake observed in prefrontal cortex, temporoparietallobes, and posterior cingulate/precuneus in both groups compared to controls(Figure 1). No differences wereobserved between the APOE4 positive and negative groups on direct comparison.The median (inter-quartile range) global PiB ratio was 2.1 (2.0–2.3) inthe PiB-positive APOE4 negative subjects and 2.1 (1.9–2.3) in thePiB-positive APOE4 positive subjects, with no difference observed across thegroups (p=0.74). Similarly, no differences were observed when the analyses werelimited to only lvPPA subjects.
Figure 1
Voxel-level maps of PiB-PET uptake in PiB-positive APOE4 negative andPiB-positive APOE4 positive subjects compared to controls
Results are shown after correction for multiple comparisons atp<0.05. Renders were generated using the BrainNet Viewer (http://www.nitrc.org/projects/bnv/).
4. Discussion
This study demonstrates that the APOE4 allele is associated withβ-amyloid deposition in subjects with speech and language disorders,including PPA, but APOE4 does not appear to affect the severity and regionaldistribution of β-amyloid deposition in these subjects.
The APOE4 allele was associated with an increased risk of β-amyloiddeposition across all subjects. The risk was, in fact, highest in subjectshom*ozygous for the APOE 4/4 allele, since all these subjects had β-amyloiddeposition on PiB-PET. Interestingly, the association between APOE4 andβ-amyloid deposition was also found for most of the clinical groups that aretypically associated with FTLD pathologies, i.e. svPPA, ucPPA and PAOS. This findingdemonstrates that patients with any of these syndromic variants are at increasedrisk of having β-amyloid deposition in the brain if they have an APOE4allele; a finding that would affect patient management and treatment strategies,particularly if amyloid imaging is unavailable. These findings do not, however,imply that β-amyloid deposition is the primary pathological processaccounting for the presenting syndrome. Instead, β-amyloid deposition mayrepresent a secondary pathology in these subjects. In fact, in a recent case report,a PPA patient that was PiB-positive was found to have FTLD pathology, as well asβ-amyloid deposition[24]. Therefore,the APOE4 allele may be increasing the risk of β-amyloid being co-deposited,as opposed to increasing the risk of AD being the primary pathology accounting forthese syndromes. Tau assessment either via CSF analysis or tau-PET imaging could behelpful in this regard. Our findings also fit with the fact that the APOE4 allelehas been shown to be associated with β-amyloid deposition in diseasescharacterized pathologically by FTLD-tau, as well as FTLD characterized bydeposition of the TAR DNA binding protein of 43kDa (TDP-43)[17, 25]. The PAOS, agPPAand svPPA syndromes are indeed most commonly pathologically characterized by tau andTDP-43 deposition, respectively[10, 14–16].
The association between APOE4 and β-amyloid deposition in agPPA wasdifficult to assess since all agPPA subjects were PiB-negative and all were APOE4negative. The findings indicate however that agPPA subjects are much less likely tohave β-amyloid deposition. It is possible that the absence ofβ-amyloid deposition in agPPA is a direct result of the fact that none ofthe agPPA subjects had an APOE4 allele. It is, however, unclear why none of the 12agPPA subjects had an APOE4 allele, since the APOE4 allele occurs in approximately25–30% in the healthy population[26–28]. It is possiblethat the absence of the APOE4 allele in the agPPA was due to a relatively smallsample size.
This is the first study to report the APOE4 allele frequency in patientswith PAOS. Interestingly, the frequency of APOE4 was low, with only 15% ofPAOS subjects having an APOE4 allele. With that said, however, PAOS is stronglyassociated with tau pathology, with almost 100% of such subjects having taupathology; with the majority having progressive supranuclear palsy pathology[14, 15,29]. Although our PAOS sample size wasless than half the sample size of our PPA group, the APOE4 allele frequency in PAOSwas significantly lower than the frequency observed in PPA. This difference washowever driven by the lvPPA subgroup.
The APOE4 allele was most frequent in the lvPPA variant. Similarly, lvPPAhad the highest frequency of positive PiB-PET scans with almost 90% beingpositive, similar to previous reports[11,30]. Interestingly, there was an almostequal chance of being PiB-positive whether the subject did, or did not have, anAPOE4 allele, although of note was the fact that 100% of the APOE 4/4hom*ozygotes were PiB-positive. Therefore, the striking association of lvPPA withβ-amyloid deposition remained strong even after taking into account APOE4.It appears that although we cannot entirely exclude APOE4 as having a role inβ-amyloid deposition in lvPPA, there may be another unknown factor orfactors, which is playing a role. Of note is the fact that the PiB-positive lvPPAsubjects without an APOE allele were unusually young. While it is possible thatolder age played a role in the PiB-positive status of some of the non-lvPPA subjectswithout an APOE4 allele, older age cannot explain the high frequency ofPiB-positivity in APOE4 negative lvPPA subjects and the lack of association withAPOE4. Given the young age at evaluation, and even younger age at onset, it would bereasonable to postulate a genetic factor playing some type of a role in thesesubjects. One limitation of the study is the lack of screening for dominantlyinherited Alzheimer’s or FTLD genes[31], although none of our PiB-positive subjects had a positive familyhistory. We have however recently screened the six PiB-negative lvPPA subjects forFTLD gene mutations and identified a progranulin gene mutation in three of the sixsubjects[32].
Our finding that APOE4 was associated with increased odds of havingβ-amyloid deposition persisted within the PPA cohort as a whole. Thisfinding differs from another study that did not observe an association between APOE4and the presence of Alzheimer’s disease in 31 PPA subjects[33]. While it is possible that this other studysuffered from a lack of power, the APOE4 frequencies were slightly different betweenstudies (32% versus 43%) and the outcome measures also differedacross studies with our study using β-amyloid deposition measured on PiB-PETand the previous study diagnosing Alzheimer’s disease by assessing thepresence of both β-amyloid and tau on autopsy.
We found no evidence that the APOE4 allele influences the severity ofβ-amyloid deposition measured by the global PiB ratio, or the distributionof β-amyloid deposition, in subjects that are PiB-positive. These findingswere consistent across all PiB-positive subjects and within the lvPPA group only. Atypical distribution of β-amyloid deposition was observed both in subjectswith and without the APOE4 allele, with greatest PiB-PET uptake observed in theprefrontal cortex, temporoparietal lobes, and posterior cingulate/precuneus. Thistopographic pattern concurs with the distribution of β-amyloid depositionobserved at autopsy in AD[34], and with thedistribution of PiB-PET uptake typically reported in subjects withAlzheimer’s dementia[21, 35] and lvPPA[11, 12, 36, 37]. Therefore,while APOE4 may increase the odds of developing β-amyloid deposition, onceβ-amyloid deposition is present the APOE4 allele does not appear toinfluence the spread or amount of β-amyloid deposition. Previous imagingstudies in Alzheimer’s dementia have reported conflicting findingsconcerning whether the APOE4 allele influences the degree of β-amyloiddeposition in subjects that are PiB-positive, with some showing increasedβ-amyloid deposition associated with the APOE4 allele[38], while others found decreased β-amyloid depositionassociated with the APOE4 allele[39, 40], and others found no differences accordingto APOE4 genotype[41, 42]. Pathological studies have similarly observed conflictingfindings with some not observing any relationship between the burden ofβ-amyloid senile plaques and the APOE4 allele in AD[43], and others finding a greater burden of plaques in APOE4carriers[44, 45]. Discrepancies across studies may be due to heterogeneousclinical and pathological cohorts and methodological differences. Our study is,however, the first to assess this issue in PiB-positive subjects with speech andlanguage disorders, and in subjects specifically with lvPPA, and in this cohortAPOE4 does not appear to be associated with the severity of β-amyloiddeposition.
Our findings suggest that determining the APOE4 allele status in subjectswith PPA and PAOS may help to shed light on why β-amyloid deposition isobserved in some subjects, despite an expected isolated FTLD pathology. Autopsystudies will, however, be needed to determine whether the β-amyloiddeposition reflects a primary or secondary pathology in these cases.
Research in Context
Systematic review
We performed a pubmed search for articles published in English withsearch terms “apolipoprotein”, “amyloid”,“Pittsburgh Compound B”, “Alzheimer’sdisease”, “apraxia of speech” and“aphasia”, in order to identify manuscripts that had assessedapolipoprotein genotyping and β-amyloid deposition in speech andlanguage disorders and Alzheimer’s disease.
Interpretation
This is the first study to show that the apolipoprotein e4 (APOE4)allele increases the risk of β-amyloid deposition in primary progressiveaphasia and progressive apraxia of speech. We also show that APOE4 does notinfluence the severity or distribution of β-amyloid deposition insubjects in which β-amyloid deposition is present. Previous studies havereported conflicting results regarding the relationship between APOE4 and theseverity of β-amyloid deposition, and have focused only on subjects withAlzheimer’s dementia.
Future directions
Autopsy studies will be needed to determine whether β-amyloiddeposition present in speech and language subjects reflects a primary orsecondary pathology.
Acknowledgement
This study was supported by NIH grant R01 DC010367 (PI Josephs).
Abbreviations
| AD | Alzheimer’s disease |
| PPA | primary progressive aphasia |
| PAOS | progressive apraxia of speech |
| lvPPA | logopenic variant of PPA |
| svPPA | semantic variant of PPA |
| agPPA | agrammatic variant of PPA |
| ucPPA | unclassified PPA |
| PiB | Pittsburgh Compound B |
| PET | positron emission tomography |
| APOE4 | apolipoprotein e4 |
| MPRAGE | 3D magnetization prepared rapid acquisition gradient echo |
| FTLD | frontotemporal lobar degeneration |
Footnotes
Publisher's Disclaimer: This is a PDF file of an uneditedmanuscript that has been accepted for publication. As a service to our customerswe are providing this early version of the manuscript. The manuscript willundergo copyediting, typesetting, and review of the resulting proof before it ispublished in its final citable form. Please note that during the productionprocess errors may be discovered which could affect the content, and all legaldisclaimers that apply to the journal pertain.
Conflicts of interest: none
References
1. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene dose of apolipoprotein E type 4 allele and the risk ofAlzheimer’s disease in late onset families. Science. 1993;261:921–923. [PubMed] [Google Scholar]
2. Gomez-Isla T, West HL, Rebeck GW, Harr SD, Growdon JH, Locascio JJ, et al. Clinical and pathological correlates of apolipoprotein E epsilon4 in Alzheimer’s disease. Ann Neurol. 1996;39:62–70. [PubMed] [Google Scholar]
3. Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of age, sex, and ethnicity on the association betweenapolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE andAlzheimer Disease Meta Analysis Consortium. JAMA : the journal of the American Medical Association. 1997;278:1349–1356. [PubMed] [Google Scholar]
4. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer's disease: report of theNINCDS-ADRDA Work Group under the auspices of Department of Health and HumanServices Task Force on Alzheimer's Disease. Neurology. 1984;34:939–944. [PubMed] [Google Scholar]
5. Gorno-Tempini ML, Hillis AE, Weintraub S, Kertesz A, Mendez M, Cappa SF, et al. Classification of primary progressive aphasia and itsvariants. Neurology. 2011;76:1006–1014. [PMC free article] [PubMed] [Google Scholar]
6. Duffy J. Apraxia of Speech in degenerative neurologicdisease. Aphasiology. 2006;20:511–527. [Google Scholar]
7. McNeil MR, Robin RA, Schmidt RA. Apraxia of speech: definition and differentialdiagnosis. In: McNeil MRE, editor. Clinical management of sensorimotor speech disorders. New York: Thieme; 2009. [Google Scholar]
8. Josephs KA, Duffy JR, Strand EA, Machulda MM, Senjem ML, Master AV, et al. Characterizing a neurodegenerative syndrome: primary progressiveapraxia of speech. Brain. 2012;135:1522–1536. [PMC free article] [PubMed] [Google Scholar]
9. Josephs KA, Duffy JR, Strand EA, Machulda MM, Senjem ML, Lowe VJ, et al. Syndromes dominated by apraxia of speech show distinctcharacteristics from agrammatic PPA. Neurology. 2013;81:337–345. [PMC free article] [PubMed] [Google Scholar]
10. Mesulam M, Wicklund A, Johnson N, Rogalski E, Leger GC, Rademaker A, et al. Alzheimer and frontotemporal pathology in subsets of primaryprogressive aphasia. Ann Neurol. 2008;63:709–719. [PMC free article] [PubMed] [Google Scholar]
11. Rabinovici GD, Jagust WJ, Furst AJ, Ogar JM, Racine CA, Mormino EC, et al. Abeta amyloid and glucose metabolism in three variants of primaryprogressive aphasia. Ann Neurol. 2008;64:388–401. [PMC free article] [PubMed] [Google Scholar]
12. Leyton CE, Villemagne VL, Savage S, Pike KE, Ballard KJ, Piguet O, et al. Subtypes of progressive aphasia: application of the InternationalConsensus Criteria and validation using beta-amyloid imaging. Brain : a journal of neurology. 2011;134:3030–3043. [PubMed] [Google Scholar]
13. Knibb JA, Xuereb JH, Patterson K, Hodges JR. Clinical and pathological characterization of progressiveaphasia. Ann Neurol. 2006;59:156–165. [PubMed] [Google Scholar]
14. Deramecourt V, Lebert F, Debachy B, Mackowiak-Cordoliani MA, Bombois S, Kerdraon O, et al. Prediction of pathology in primary progressive language andspeech disorders. Neurology. 2010;74:42–49. [PubMed] [Google Scholar]
15. Josephs KA, Duffy JR, Strand EA, Whitwell JL, Layton KF, Parisi JE, et al. Clinicopathological and imaging correlates of progressive aphasiaand apraxia of speech. Brain. 2006;129:1385–1398. [PMC free article] [PubMed] [Google Scholar]
16. Josephs KA, Hodges JR, Snowden JS, Mackenzie IR, Neumann M, Mann DM, et al. Neuropathological background of phenotypical variability infrontotemporal dementia. Acta Neuropathol. 2011;122:137–153. [PMC free article] [PubMed] [Google Scholar]
17. Josephs KA, Tsuboi Y, Cookson N, Watt H, Dickson DW. Apolipoprotein E epsilon 4 is a determinant for Alzheimer-typepathologic features in tauopathies, synucleinopathies, and frontotemporaldegeneration. Arch Neurol. 2004;61:1579–1584. [PubMed] [Google Scholar]
18. Crook R, Hardy J, Duff K. Single-day apolipoprotein E genotyping. Journal of neuroscience methods. 1994;53:125–127. [PubMed] [Google Scholar]
19. Kertesz A. Western Aphasia Battery (Revised) San Antonio, Tx: PsychCorp; 2007. [Google Scholar]
20. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method forgrading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–198. [PubMed] [Google Scholar]
21. Jack CR, Jr., Lowe VJ, Senjem ML, Weigand SD, Kemp BJ, Shiung MM, et al. 11C PiB and structural MRI provide complementary information inimaging of Alzheimer's disease and amnestic mild cognitiveimpairment. Brain. 2008;131:665–680. [PMC free article] [PubMed] [Google Scholar]
22. Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F, Etard O, Delcroix N, et al. Automated anatomical labeling of activations in SPM using amacroscopic anatomical parcellation of the MNI MRI single-subjectbrain. Neuroimage. 2002;15:273–289. [PubMed] [Google Scholar]
23. Agresti A, Coull BA. Approximate is Better than “Exact” for IntervalEstimation of Binomial Proportions. The American Statistician. 1998;52:119–126. [Google Scholar]
24. Caso F, Gesierich B, Henry M, Sidhu M, LaMarre A, Babiak M, et al. Nonfluent/agrammatic PPA with in-vivo cortical amyloidosis andPick’s disease pathology. Behavioural neurology. 2013;26:95–106. [PMC free article] [PubMed] [Google Scholar]
25. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, Chou TT, et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration andamyotrophic lateral sclerosis. Science. 2006;314:130–133. [PubMed] [Google Scholar]
26. Harrington CR, Louwagie J, Rossau R, Vanmechelen E, Perry RH, Perry EK, et al. Influence of apolipoprotein E genotype on senile dementia of theAlzheimer and Lewy body types. Significance for etiological theories ofAlzheimer's disease. The American journal of pathology. 1994;145:1472–1484. [PMC free article] [PubMed] [Google Scholar]
27. St Clair D, Norrman J, Perry R, Yates C, Wilco*ck G, Brookes A. Apolipoprotein E epsilon 4 allele frequency in patients with Lewybody dementia, Alzheimer's disease and age-matchedcontrols. Neuroscience letters. 1994;176:45–46. [PubMed] [Google Scholar]
28. Gearing M, Schneider JA, Rebeck GW, Hyman BT, Mirra SS. Alzheimer's disease with and without coexistingParkinson's disease changes: apolipoprotein E genotype andneuropathologic correlates. Neurology. 1995;45:1985–1990. [PubMed] [Google Scholar]
29. Josephs KA, Boeve BF, Duffy JR, Smith GE, Knopman DS, Parisi JE, et al. Atypical progressive supranuclear palsy underlying progressiveapraxia of speech and nonfluent aphasia. Neurocase. 2005;11:283–296. [PubMed] [Google Scholar]
30. Leyton CE, Villemagne VL, Savage S, Pike KE, Ballard KJ, Piguet O, et al. Subtypes of progressive aphasia: application of the InternationalConsensus Criteria and validation using beta-amyloid imaging. Brain. 2011;134:3030–3043. [PubMed] [Google Scholar]
31. Goldman JS. New approaches to genetic counseling and testing forAlzheimer's disease and frontotemporal degeneration. Current neurology and neuroscience reports. 2012;12:502–510. [PMC free article] [PubMed] [Google Scholar]
32. Josephs KA, Duffy JR, Strand EA, Machulda MM, Vemuri P, Senjem ML, et al. Progranulin-associated PiB-negative logopenic primary progressiveaphasia. Journal of neurology. 2014;261:604–614. [PMC free article] [PubMed] [Google Scholar]
33. Rogalski EJ, Rademaker A, Harrison TM, Helenowski I, Johnson N, Bigio E, et al. ApoE E4 is a susceptibility factor in amnestic but not aphasicdementias. Alzheimer disease and associated disorders. 2011;25:159–163. [PMC free article] [PubMed] [Google Scholar]
34. Braak H, Braak E. Neuropathological stageing of Alzheimer-relatedchanges. Acta Neuropathol. 1991;82:239–259. [PubMed] [Google Scholar]
35. Rowe CC, Ng S, Ackermann U, Gong SJ, Pike K, Savage G, et al. Imaging beta-amyloid burden in aging and dementia. Neurology. 2007;68:1718–1725. [PubMed] [Google Scholar]
36. Lehmann M, Ghosh PM, Madison C, Laforce R, Jr., Corbetta-Rastelli C, Weiner MW, et al. Diverging patterns of amyloid deposition and hypometabolism inclinical variants of probable Alzheimer's disease. Brain. 2013;136:844–858. [PMC free article] [PubMed] [Google Scholar]
37. Whitwell JL, Lowe VJ, Duffy JR, Strand EA, Machulda MM, Kantarci K, et al. Elevated occipital beta-amyloid deposition is associated withwidespread cognitive impairment in logopenic progressiveaphasia. Journal of neurology, neurosurgery, and psychiatry. 2013;84:1357–1364. [PMC free article] [PubMed] [Google Scholar]
38. Drzezga A, Grimmer T, Henriksen G, Muhlau M, Perneczky R, Miederer I, et al. Effect of APOE genotype on amyloid plaque load and gray mattervolume in Alzheimer disease. Neurology. 2009;72:1487–1494. [PubMed] [Google Scholar]
39. Ossenkoppele R, van der Flier WM, Zwan MD, Adriaanse SF, Boellaard R, Windhorst AD, et al. Differential effect of APOE genotype on amyloid load and glucosemetabolism in AD dementia. Neurology. 2013;80:359–365. [PubMed] [Google Scholar]
40. Lehmann M, Ghosh PM, Madison C, Karydas A, Coppola G, O’Neil JP, et al. Greater medial temporal hypometabolism and lower cortical amyloidburden in ApoE4-positive AD patients. Journal of neurology, neurosurgery, and psychiatry. 2013 [PMC free article] [PubMed] [Google Scholar]
41. Villemagne VL, Pike KE, Chetelat G, Ellis KA, Mulligan RS, Bourgeat P, et al. Longitudinal assessment of Abeta and cognition in aging andAlzheimer disease. Ann Neurol. 2011;69:181–192. [PMC free article] [PubMed] [Google Scholar]
42. Rowe CC, Ellis KA, Rimajova M, Bourgeat P, Pike KE, Jones G, et al. Amyloid imaging results from the Australian Imaging, Biomarkersand Lifestyle (AIBL) study of aging. Neurobiology of aging. 2010;31:1275–1283. [PubMed] [Google Scholar]
43. Landen M, Thorsell A, Wallin A, Blennow K. The apolipoprotein E allele epsilon 4 does not correlate with thenumber of senile plaques or neurofibrillary tangles in patients withAlzheimer's disease. Journal of neurology, neurosurgery, and psychiatry. 1996;61:352–356. [PMC free article] [PubMed] [Google Scholar]
44. Tiraboschi P, Hansen LA, Masliah E, Alford M, Thal LJ, Corey-Bloom J. Impact of APOE genotype on neuropathologic and neurochemicalmarkers of Alzheimer disease. Neurology. 2004;62:1977–1983. [PubMed] [Google Scholar]
45. Nagy Z, Esiri MM, Jobst KA, Johnston C, Litchfield S, Sim E, et al. Influence of the apolipoprotein E genotype on amyloid depositionand neurofibrillary tangle formation in Alzheimer'sdisease. Neuroscience. 1995;69:757–761. [PubMed] [Google Scholar]
