Citations

AbstractBackgroundFamilial frontotemporal dementia (f‐FTLD) is typically caused by mutations in one of three genes: microtubule‐associated protein tau (MAPT), progranulin (GRN), and a repeat expansion in the chromosome 9 open reading frame 72 (C9orf72) gene. Accurate characterization of the natural history of each mutation is important for clinical prognostication and clinical trial design, and it could shed light on disease biology. Such models have not been thoroughly developed using participants that represent all disease stages, with longitudinal data. We characterized the trajectory of atrophy in each gene by using longitudinal voxel‐wise analyses of gray matter volume, and we assessed whether functional independence declined in tandem.MethodF‐FTLD participants (n=100) with a known mutation (MAPT+ (n=28), GRN+ (n=33), C9orf72+ (n=39)) were grouped according to disease stage (CDR®+NACC FTLD module). We included participants with at least two structural MRIs at a given disease stage: presymptomatic (CDR®+NACC‐FTLD=0, n=57), mild/questionable (CDR®+NACC‐FTLD=0.5, n=15), and symptomatic (CDR®+NACC‐FTLD ≥1, n=28). We fitted longitudinal linear mixed effects models to extract mean atrophy rates in each lobe compared to longitudinal imaging from family members without mutations (n=60). All results presented below were significant at p<.001.ResultUsing the left frontal lobe as an exemplar, in the presymptomatic stage, MAPT mutation carriers showed the greatest rate of atrophy compared to controls (88 mm3/year more volume loss than controls), followed by GRN+ (65 mm3/year) and then C9orf72+ (49 mm3/year). In the mild/questionable stage, MAPT+ showed a greater divergence from controls (374 mm3/year) than did GRN+ (107 mm3/year) or C9orf72+ (223 mm3/year). In the symptomatic stage, MAPT+ again lost volume at the fastest rate (2,099 mm3/year), followed by GRN+ (1,360 mm3/year). C9orf72 expansion carriers showed a much slower rate of volume loss (115 mm3/year). Similar patterns were observed for other brain regions. In contrast to the imaging results, C9orf72+ exhibited similar rates of functional decline compared to GRN+ and MAPT+ at all levels of disease severity.ConclusionThe primary f‐FTLD genes show divergent atrophy trajectories as a function of disease stage, with C9orf72 expansion carriers exhibiting a slow degeneration throughout the disease course, possibly due to unique pathophysiological mechanisms.

AbstractBackgroundData shared for research use must be de‐identified to protect participant confidentiality. A recent study has shown that face imagery from brain imaging studies could be used to re‐identify anonymous participants via automatic face recognition, linking them with all study‐collected protected health information. Here we measure how well existing software for removing MRI face imagery protects against this privacy threat.MethodWe recruited 153 participants with existing 3D FLAIR MRI and photographed their face from 5 varying angles. We used surf_ice to reconstruct each participant’s face from their MRI and tested whether free, publicly‐available face recognition software (Microsoft Azure) could match their photos to their correct MRI. We tested two existing programs for removing face voxels in MRI: mri_deface (FreeSurfer) and pydeface (python package), and one in‐house implementation. We observed that existing methods often left participants’ eyes intact while removing most of their nose/mouth/chin. This prevented Azure from detecting the reconstructions as faces, therefore preventing its attempt of face recognition. We hypothesized that the retained eye imagery alone could be sufficient for face recognition, if a face were detected. Therefore, we also reconstructed each de‐faced MRI after applying novel pre‐processing that replaced any removed/missing face voxels with those from our population‐average template, e.g. replacing missing noses/mouths/chins with population‐average noses/mouths/chins, before testing face recognition.ResultsWithout any MRI de‐facing, Microsoft Azure correctly matched 78% of participants’ photos to their MRI. After mri_deface and pydeface, 8% and 6% were still identified, respectively. After preprocessing the de‐faced MRI to replace missing areas with those from a population‐average template prior to face reconstruction, the face imagery remaining in the de‐faced images was enough to allow re‐identification of 23% and 20% of participants, respectively. With an in‐house face removal implementation that removes the whole face from every image, 0% of faces were re‐identified from the standard reconstructions, and 5% were identified from the “missing‐parts‐replacement” reconstructions.ConclusionsEven after applying popular MRI de‐facing software, Microsoft Azure face recognition automatically re‐identified up to 23% of participants. Better de‐facing software is needed to remove more of the face more reliably, without affecting brain measurements from de‐faced images.

AbstractBackgroundCharacterization of the changes that occur throughout the transition from asymptomatic to symptomatic for those with familial frontotemporal dementia is essential for planning future disease‐modifying trials.MethodWe identified four related microtubule associated protein tau (MAPT) N279K mutation carriers who have phenoconverted from the asymptomatic to the symptomatic state. Key features were as follows: Case 1 (female) initial exam age 36, onset age 44, current age 49; Case 2 (male) initial exam age 39, onset age 43, current age 48; Case 3 (female) initial exam age 41, onset age 47, current age 53; Case 4 (male) initial exam age 36, onset age 37, current age 46. MRI and neuropsychological data were analyzed.ResultStructural MRI was analyzed using Tensor Based Morphometry Symmetric Diffeomorphic Image Normalization (TBM‐SyN) methodology to quantify lobar volumes, which was then analyzed using a Mixed Effects Model to determine the annual change in lobar volumes. The phenoconverters had a greater estimated annual rate of cortical atrophy in temporal (p <0.001), frontal (p <0.01) and parietal (p <0.05) lobes compared to non‐carriers, with the rate of annual change in symptomatic individuals double that of asymptomatic individuals. Annual change in the parietal and occipital lobes was also analyzed and found to be similar to healthy controls of a similar age. Raw scores on neuropsychological measures were converted to z‐scores for analysis. Category fluency, letter fluency, and Trails A and B were all impaired (z‐score <‐1.5) following phenoconversion, but performance on these and other measures was variable across these four in the late asymptomatic state.ConclusionFor each of the four phenoconverters, structural MRI changes preceded symptom onset, whereas neuropsychological performance was more variable and occurred in closer proximity to the time of phenoconversion. Decrements in verbal fluency and cognitive flexibility were observed earlier than other cognitive domains. These findings suggest that changes in frontal and temporal volume on MRI precede symptom onset, with regional brain volumes most consistently tracking the evolution from the asymptomatic to symptomatic state in N279K MAPT mutation carriers. Supported by: AG063911, AG045390, NS092089, AG016976, AG016574, AG062677.

AbstractBackgroundThe LEFFTDS Consortium involves investigators at 8 centers in North America evaluating individuals in kindreds with mutations in microtubule associated protein tau (MAPT), progranulin (GRN), or chromosome 9 open reading frame 72 (C9orf72) genes using a standardized battery of measures. Participants are evaluated annually. We sought to determine possible correlations between FTLD disease progression and neuropsychiatric features.MethodThe Neuropsychiatric Inventory Questionnaire (NPI‐Q) scale is an informant‐based instrument that measures the presence and severity of 12 neuropsychiatric features in patients with dementia. The NPI‐Q was completed at each LEFFTDS visit. The individual NPI questions and total score were analyzed in those with CDR® Dementia Staging Instrument PLUS National Alzheimer Coordinating Center Frontotemporal Lobar Degeneration (CDR®+ NACC FTLD) ratings of 0 (i.e., clinically normal), 0.5 (i.e., questionably/minimally abnormal), 1, 2, and 3.ResultThe 472 participants enrolled to date with a baseline (Visit 1) assessment have the following characteristics: 261 (55.5%) female, mean age 48 (range 18‐80) years, mean education 16 (range 11‐22) years, and 275 participants with a CDR®+NACC FTLD=0 (58%) and known genetic status. Participants with a CDR®+NACC FTLD=0 plus a genetic mutation were compared to those without a mutation and found to have increased severity of apathy (mean score of 0.116 versus 0.035, p=0.0302). Regardless of the specific mutation, those with a CDR®+NACC FTLD=0.5 were reported to have increased severity of apathy, disinhibition, and irritability. For most domains, mean NPI‐Q severity scores increase between groups up to CDR®+NACC FTLD=2. Depression increases in severity up to CDR®+NACC FTLD=1, whereas apathy severity continues to increase with increasing disease severity through CDR®+NACC FTLD=3. Twenty participants converted from CDR®+NACC FTLD 0 to >0 between their baseline and most recent visit; apathy severity almost tripled in these participants. For those that converted, disinhibition, irritability, and NPI total score were different between baseline and most recent visit (Wilcoxon signed rank test, p<0.05).ConclusionKey neuropsychiatric features measured by the NPI‐Q demonstrate differences in apathy severity between clinically normal mutation vs non‐mutation carriers. Key domains differ at different levels of FTLD severity and show dynamic changes with increasing level of FTLD severity. Supported by AG063911, AG045390, NS092089, AG016976.

AbstractBackgroundIt is important to determine the natural history of sporadic and familial frontotemporal lobar degeneration (FTLD) and generate clinical, neuropsychological, neuroimaging and biofluid data for planning disease‐modifying trials.MethodAs part of the ARTFL LEFFTDS Longitudinal Frontotemporal Lobar Degeneration (ALLFTD; U19 AG063911) protocol, investigators at 19 centers in North America will enroll 2100 participants with FTLD over the next 5 years beginning in early 2020.ResultAs of 1/20/20, the ARTFL/LEFFTDS (A/L) Consortium had enrolled 1832 participants, including 850 in kindreds with familial FTLD (239 associated with mutations in MAPT, 192 in GRN, 370 in C9orf72, 4 with mutations in both GRN and C9orf72, and 45 with a mutation in a different gene or no mutation in any known FTLD‐associated gene). Over 500 participants have undergone 2 or more annual visits to date. MRI has been performed in 1105. Biofluid samples have also been collected, with blood (DNA, plasma, serum, mRNA, PBMC) in 1413 and CSF in 303. Over 60 manuscripts using A/L data or samples have been published to date. Five clinical trials involving A/L and ALLFTD participants are in progress or planned. The longitudinal arm in ALLFTD will enroll 500 of existing A/L and 600 future participants for annual assessments with similar methodology to A/L. An additional 1000 FTLD patients will undergo focused one‐time clinical evaluations and biofluid collection.ConclusionThe data/samples from already‐enrolled and planned participants in ALLFTD and findings published to date underscore the utility of evaluating FTLD subjects. The absence of identifiable mutations in some with familial FTLD suggests that other genes are yet to be discovered. ALLFTD data will inform clinical trial design, and many participants will be eligible for future trials. The key data and samples in ALLFTD are available to interested investigators worldwide. Supported by: AG063911, AG045390, NS092089, AG016976, AG21886.

AbstractBackgroundDementia with Lewy Bodies (DLB) is the second most common cause of degenerative dementia but the evolution of the core clinical features of DLB has not been clearly defined. The objective of this study was to identify the temporal relationship between each core feature from the estimated onset of cognitive symptoms.MethodParticipants include 485 patients with probable DLB who were evaluated at the Mayo Clinic Alzheimer Disease Research Center with neurologic and neurocognitive assessment. Each core clinical feature was compared using Kaplan‐Meier (KM) curves with the sentinel event representing age at which each core feature developed, or time from estimated cognitive onset.ResultThe sample was disproportionately male (76%). Patients were followed longitudinally, with 84% followed until death. Mean MMSE at the last evaluation was 18.3 ± 7.8. KM curves for age of onset and cumulative incidence for DLB core features as time from cognitive onset are provided. REM‐sleep behavior disorder (RBD) developed earliest and in 77% of our sample with a median KM survival estimate of 69 years (95% CI: 67.6–70.8). RBD was present in greater than 50% of patients at cognitive symptom onset. The median KM survival estimate for cognitive symptom onset was 70.2 years (95% CI: 69.3‐71.0). Parkinsonism developed in 89% with a median KM survival estimate of 73.7 years (95% CI: 72.7‐74.7). 76% had fluctuations with a similar median KM survival estimate of 74.6 years (95% CI: 73.7‐75.8). Visual hallucinations (VH) were present in 70% emerging at a median KM survival estimate of 80.2 years (95% CI: 79.5‐ 81.3). Cumulative incidence estimates for median time from cognitive onset to parkinsonism, fluctuations and VH were 2 years (95% CI: 2.0‐2.5), 2.3 years (95% CI: 2.0‐3.2) and 4.3 years (95% CI: 4.0‐4.8 years), respectively.ConclusionIn this DLB cohort, RBD was most likely to precede cognitive onset. The evolution of the other core features after cognitive onset was parkinsonism and fluctuations and then VH. While VH occur about 4 years after cognitive symptom onset, parkinsonism and fluctuations occur within the first 2 years.

AbstractBackgroundPractice effects can obscure interpretation of longitudinal cognitive performance. Reliable change indices (RCIs) provide data for determining a significant change, typically for a single repeat assessment, but multiple follow‐up visits are common clinically and in research cohorts. Further, regression‐based or linear mixed effects models improve prediction of future cognitive decline and have higher correlations with neuroimaging biomarkers than RCI approaches. We present norms for change for the Auditory Verbal Learning Tests (AVLT) with conditional normative z‐scores for up to 7 repeat assessments, and demonstrate application of this method to individuals with preclinical Alzheimer’s disease (AD).MethodNormative Sample: Robust conditional norms were derived from 1001 adults ages 50‐90 from the Mayo Clinic Study of Aging (MCSA) who completed the AVLT across 4‐8 visits approximately 15 months apart and remained cognitively unimpaired (CU) across all available visits. Linear mixed effects models were used to compute a conditional normative z‐score for AVLT 30‐minute recall and sum of trials. Models included adjustment for baseline performance, number of prior test exposures, demographic factors (age, sex, education), and interactions (baseline*time; sex*time). Test Sample: We used a preclinical AD sample with positive amyloid and tau imaging (n= 27 CU A+T+; Jack et al. 2017) expected to have a higher rate of intra‐individual cognitive decline relative to CU A‐T‐ individuals (n = 268) to compare mean performances of conditional normative z‐scores to typical age‐corrected Mayo’s Older Americans Normative Studies (MOANS) norms. Participants had at least 1 and up to 4 follow‐up visits, with their most recent AT data used for classification.ResultAVLT performance based on typical norms did not differ across A+T+ and A‐T‐ groups. Specifically, across 2‐4 follow‐up visits, the typical norms mean z‐score was similar across groups (p=0.945; see Table 1). However, the conditional norms mean z‐score that accounts for practice effects and typical trajectories of AVLT performance was lower in the A+T+ relative to the A‐T‐ group for 30‐minute recall (p=0.055) and sum of trials (p=0.060).ConclusionConditional normative methods that account for practice effects are more sensitive to preclinical AD than typical norms, and show promise for identifying transitional cognitive decline.

AbstractBackgroundNeuropsychiatric symptoms (NPS) are associated with the risk of incident mild cognitive impairment (MCI) and dementia. However, the association between NPS and global and domain‐specific cognitive trajectories among older adults free of dementia remains unclear.MethodWe conducted a longitudinal study derived from the population‐based Mayo Clinic Study of Aging, including individuals aged ≥ 50 years who were free of dementia. Participants completed the Neuropsychiatric Inventory Questionnaire at baseline and underwent neuropsychological testing assessing four cognitive domains (i.e., memory, language, attention, visuospatial skills) every 15 months. We calculated linear mixed‐effect models with random subject‐specific intercepts and slopes to examine the association between presence of twelve different NPS (i.e., delusions, hallucinations, agitation, depression, anxiety, euphoria, apathy, disinhibition, irritability, motor behavior, nighttime behavior, appetite) at baseline and trajectories for individual yearly change in global and domain‐specific cognitive z‐scores. The models were adjusted for age, sex, education, and whether or not this was the first time the cognitive battery had been administered.ResultThe sample included 5081 persons (51% males; 4439 cognitively unimpaired, 642 with MCI). The median [IQR] age was 74 [67, 81] years, and the median [IQR] education was 14 [12, 16] years. Regardless of NPS status, participants declined in cognition over time, on average. However, presence of NPS was significantly associated with an increased decline in cognition, i.e., the annualized change in global cognition z‐scores for participants with compared to without the given NPS ranged from ‐0.018 (p = 0.011) for irritability to ‐0.16 (p = 0.001) for hallucinations. These associations between NPS and annual decline in global cognition were significant for all NPS except euphoria and disinhibition. Similar results were seen for the individual cognitive domains, e.g., participants with depression, apathy and nighttime behavior had a greater longitudinal decline in all domain specific z‐scores, whereas presence of delusions and anxiety was associated with a more pronounced decline in language, attention and visuospatial skills.ConclusionNeuropsychiatric symptoms are associated with greater longitudinal decline of both global and domain‐specific cognitive functions in community‐dwelling older adults free of dementia. More research is needed to confirm these preliminary findings.

AbstractBackgroundThe numeric clinical staging scheme described by the NIA‐AA workgroup proposes evidence of subtle decline on longitudinal cognitive testing as one option to define transitional cognitive decline for individuals in the Alzheimer’s continuum. Few studies have compared longitudinal and cross‐sectional definitions of subtle objective cognitive impairment (sOBJ). The study aim was to compare the diagnostic accuracy of cross‐sectional sOBJ and longitudinal objective cognitive decline (ΔOBJ) for preclinical Alzheimer’s disease (AD) based on Cogstate One Card Learning (OCL) performance.MethodWe included Mayo Clinic Study of Aging cognitively unimpaired (CU) participants who were A+T+ by amyloid and tau PET assessment (n=32) and A‐T‐ (n=211); all had at least two follow‐up visits. sOBJ was defined as performance ≤ ‐1SD on OCL accuracy using age‐corrected normative data. ΔOBJ was measured using two different methods: (1) within subjects’ standard deviation (WSD) using a ≤ ‐1 z‐score for change to define decline based on Cogstate’s normative change data, and (2) extracted subject‐specific slopes from a linear mixed effects model (LME) which can be interpreted as approximate annual change; decline was defined as slope < 10%ile among a reference sample of CU participants aged 50‐65 at baseline (N=732).ResultssOBJ and ΔOBJ‐WSD both showed low sensitivity in detection of A+T+ (9.4% and 12.5%, respectively). Total AUC values did not differ significantly (p=.31) across sOBJ (AUC = .64, CI = .53‐.75) and ΔOBJ‐WSD (AUC = .53, CI = .42‐.65) methods. Sensitivity of the ΔOBJ‐LME method to predict A+T+ was improved (37.5%) when using a conventional cutoff of <10%ile slope and was further improved (56%) when using an optimal derived cutoff equivalent to <17%ile slope within the reference group. Like the sOBJ method, the ΔOBJ‐LME method differentiated groups better than chance (AUC = .69, CI = .58‐.80). Despite increased sensitivity, total AUC values were not significantly different across sOBJ and ΔOBJ‐LME methods (p=.34). However, total AUC of ΔOBJ‐LME was significantly better than ΔOBJ‐WSD (p=.03).ConclusionsΔOBJ may be more sensitive to preclinical AD than sOBJ and may serve as a helpful method for identifying at risk CU individuals if advanced statistical methods are used for defining change.