Frontal Lobe Degeneration in Adults with Down Syndrome and Alzheimer's Disease: A Review

Down syndrome (DS) is caused by a chromosomal abnormality. Most individuals have two copies of chromosome 21, whereas individuals with DS have those same two copies plus one extra copy. Consequently, individuals with DS present with various characteristics and symptoms, ranging from those indicative of the phenotype (including a transverse palmar crease, low muscle tone, protruding tongue, and short stature) to cognitive deficits, the vast majority of such individuals therefore having intellectual disability [1,2]. It has been estimated that DS occurs in approximately 1 in 732 infants in the United States [3].

Recent years have seen a considerable increase in the life expectancy of individuals with DS due to advances in biomedicine, the adoption of a series of preventive measures such as improvements of congenital heart diseases, and the emergence of targeted therapies, as well as early and ongoing interventions focused on key perceptual-motor behaviors. However, as such individual's age, many are affected by comorbidities that accelerate their cognitive and functional decline [4]. There is a proven genetic relationship between DS and early development of the neuropathological characteristics of Alzheimer's disease (AD) [5]. This is partly explained by the tripling (trisomy) of chromosome 21, which is involved in the overexpression of many proteins, including some related to the neurodegenerative process. For example, trisomy 21 can increase early deposition of the β-amyloid (Aβ) precursor protein and the production of superoxide dismutase 1, thus increasing oxidative stress [6,7]. Individuals with AD have progressive, irreversible neuropsychiatric degeneration that affects cognition and functionality [8]. Postmortem studies have shown that, after reaching 40 years of age, virtually all individuals with DS show neuropathological signs of AD, including senile plaques and neurofibrillary tangles [9,10]. Because of the pathological similarity of AD among individuals with DS in the fourth decade of life, DS in such individuals can be considered a model for the development of AD, even if the degree of neuropathological severity is not directly related to the clinical characteristics presented [11]. Neuroimaging studies have identified typical brain atrophy in patients with DS and AD [12], as well as indicators of dementia correlated with structural and functional findings in individuals with DS without clinical signs of AD but at the age of risk for its development [13]. Nevertheless, recent research indicates that many adults with DS can tolerate the deposition of Aβ with no effects on their cognitive functions [14].

Despite the proven link between DS and AD, pharmacological and nonpharmacological interventions are still under investigation [15], and diagnosing dementia in individuals with intellectual disabilities continues to be a challenge. That is because the vast majority of instruments used for the detection of dementia were developed from studies of individuals with normal intellectual development, being intended for application in the general population. The validity and reliability of such instruments can be questioned because they do not take into account the possibility of a low basal level of cognitive functioning [16]. Only a few instruments have been designed specifically for evaluating cognition in intellectually disabled individuals, and most of the tests currently used in order to detect dementia in such individuals have not been adapted for use in this population. Because of that instrumental limitation, the longitudinal comparison of individual performance continues to be the main diagnostic tool used in clinical practice [17] despite its high cost and the frequent lack of baseline measures preceding the cognitive complaints.

The clinical signs of AD can include behavioral symptoms, such as disinhibition, agitation, and apathy, as well as sleep disorders [18]. However, among individuals with DS and AD, a high prevalence of excessive agitation, sleep disorders, restlessness, and auditory hallucinations has been reported [19]. The results of cross-sectional and longitudinal studies suggest that changes in the personality and behavior of adults with DS can indicate the early stages of dementia prior or simultaneous to the onset of other typical AD symptoms such as cognitive decline [19,20,21,22,23]. Those initial alterations are comparable to the changes associated with frontotemporal dementia in the general population and are considered precursors of the development of AD in individuals with DS, indicating that the frontal and temporal lobes are the brain regions first affected in this population [23,24,25]. In fact, in the brains of individuals with DS, the deposition of Aβ initially occurs in the frontal and entorhinal cortices, subsequently spreading to other brain regions [26]. Some authors argue that this atypical presentation of AD in DS is due to preexisting abnormalities in brain development that would cause hypoplasia of the frontal lobe [22]. In general, individuals with DS perform more poorly in executive function tests than do those with intellectual disabilities of other etiologies [27]. In addition, longitudinal studies indicate that a decline in working memory is an early indicator of dementia in DS [28,29]. In individuals with DS, there are often volume reductions in the frontal lobe, limbic system, and brain stem [30], together with decreased cell density, particularly surrounding the projections of the corpus callosum, as well as evidence that the corpus callosum region responsible for neuronal projections to the frontal lobe is poorly developed [31]. Therefore, we can hypothesize that the decreased cognitive reserve in the frontal lobe contributes to the premature impairment of this region and the consequent initial appearance of symptoms of executive dysfunction and changes in behavior.

The aim of this study was to investigate executive dysfunctions and changes in behavior associated with frontal lobe degeneration in AD in individuals with DS. To that end, we identified and analyzed relevant articles published in journals indexed for PubMed or PsycINFO.

We conducted systematic searches within the PubMed and PsycINFO databases. The principal search terms were ‘Down syndrome' and ‘dementia'. Those two terms were searched together with the terms ‘cognitive test', ‘cognitive task', ‘neuropsychological', ‘neuropsychology', and ‘neurocognitive'. Therefore, a total of five searches were made. We included studies meeting the following criteria: being a clinical trial or case study; including adults with DS who had been diagnosed with AD or showed signs of dementia; having employed cognitive assessments; having been published between January 2010 and May 2015, and being available in English, Spanish, Italian, or Portuguese. In evaluating the selected articles, we analyzed the following aspects: the tests employed; the neuropsychological functions investigated; the adaptability of the instruments for use in individuals with DS and AD; the results obtained for executive functions and for other frontal lobe functions.

We initially identified a total of 101 studies. We excluded 40 duplicates and analyzed the abstracts of the remaining 61 articles, 17 of which met the inclusion criteria. However, after detailed analysis of the articles, 3 were excluded: 1 was excluded because it was a postmortem analysis [32], and 2 were excluded because they did not address the diagnosis or signs of dementia [33,34], although 1 of those 2 did address the relationship between oxidative stress and a decline in memory [34]. Therefore, we reviewed 14 articles [12,20,35,36,37,38,39,40,41,42,43,44,45,46], all of which are profiled in table 1. Although we would have included studies available in English, Spanish, Italian, or Portuguese, from the 14 studies, only 1 was in Spanish, the other ones were all in English. A diagram illustrating the selection process is shown in figure 1.

The studies included in the analysis varied according to the main focus of the investigation. Of the 14 studies, only 2 focused on the cognitive assessment of individuals with DS and AD [20,36]. There were 11 studies in which the main focus was AD in DS [12,20,35,36,37,38,39,40,41,42,43]; there was 1 study in which the central theme was functional aspects of AD in adults with intellectual disability [44], and there were 2 studies that focused on aging individuals with DS but not necessarily on aspects related to dementia [45,46]. Of the 11 studies focusing on AD in DS, 2 included only subjects diagnosed with DS and AD [12,37], whereas the others involved control groups of individuals without AD, without DS, or even without intellectual disabilities. Two studies did not report the exact number of subjects showing signs of dementia or diagnosed with AD [35,46]. In one study [44], organizational problems resulted in 21% of the subjects going undiagnosed.

In a study conducted by De Vreese et al. [44], the focus was the validation of an instrument for the functional assessment of individuals with intellectual disabilities, with and without dementia. In the study conducted by Carr and Collins [36], the main objective was the longitudinal assessment of cognition in subjects with DS and AD. Lott et al. [37] and d'Orsi et al. [12] investigated the relationships between DS, AD, and epilepsy. Koran et al. [46] investigated the effect that age has on brain volume in adults with DS, in comparison with adults with Williams syndrome or normal development. In another longitudinal study, McCarron et al. [38] aimed to characterize the onset and evolution of dementia in DS. Esteba-Castillo et al. [35] conducted a validation study of a diagnostic tool for use in individuals with DS and AD. Chace et al. [39] investigated the relationships between DS, AD, and specific cytogenetic changes. Hanney et al. [40] conducted an experimental study of the use of memantine (an N-methyl D-aspartate receptor antagonist used in treating the symptoms of AD) in individuals with DS. Two studies explored the relationships between plasma Aβ, AD, and DS, one focusing on whether plasma Aβ level is associated with cognitive functions [41] and the other focusing on whether plasma Aβ level is associated with early-onset dementia [42]. Oliver et al. [43] examined the impact of behavioral changes in individuals with DS and AD, comparing it with that observed for individuals with DS and no dementia. Adams and Oliver [20] investigated the relationship between behavioral changes and deficits in cognitive functions associated with the frontal lobe. The study conducted by Iacono et al. [45] focused on the relationships between age, language, and DS.

The criteria used in order to diagnose AD in the selected studies are shown in table 1. Three studies did not consider a clinical diagnosis of dementia but identified the signs of dementia through the use of instruments designed specifically for use in intellectually disabled individuals and were therefore included in the analysis [20,36,46]. Carr and Collins [36] and Adams and Oliver [20] applied criteria for identifying signs of dementia based on specific, objective measures of cognitive deterioration, whereas Koran et al. [46] applied criteria based on objective, point measures of dementia, including a discussion of the role of apolipoprotein E4 and neuroimaging evidence of AD.

There were 11 studies that addressed the diagnosis of AD. Among those, the diagnosis of AD was based on clinical psychiatric evaluation, without the use of any specific gold standard, in one [43] and on the assessment of dementia in individuals with intellectual disabilities in another [44]. In the remaining 9 studies, various diagnostic criteria were used, as illustrated in table 1. All 9 of those studies applied at least one internationally recognized set of diagnostic criteria for AD. Five of the 9 studies used diagnostic criteria specific for individuals with intellectual disabilities. Of those same 9 studies, 2 applied 2 diagnostic criteria [12,38], 1 applied 3 [35], and 1 applied 4 [45]. Each of the 5 remaining studies applied only one set of diagnostic criteria.

The instruments used in the studies were codified and evaluated on the basis of the following aspects: whether or not they had been specifically designed or adapted for use in individuals with intellectual disabilities; whether or not they had been specifically designed or adapted for use in the evaluation of dementia; whether they had been administered directly to the subject or to an informant; whether the form in which they were administered was their original form or represented an adaptation specific to the study at hand, and which specific cognitive concepts were evaluated. These aspects are detailed in table 2.

Among the selected studies, a total of 44 instruments were cited. However, 2 of those instruments were actually one and the same, being referred to in some studies by its former name - the Dementia Questionnaire for Mentally Retarded Persons - and in others by its current name - the Dementia Questionnaire for Learning Disabilities (DLD). Therefore, the true total count was 43 different instruments. Of those, 31 were instruments for assessing cognitive function that were administered directly to the subject. One was a neuropsychological test battery, the Neuropsychological Assessment of Dementia in Adults with Intellectual Disabilities (NAID). In addition to the NAID, 7 of those 31 instruments were designed specifically for use in intellectual disability and AD. Another 2 of the 31 were specific for intellectual disabilities but not for AD. Conversely, 3 instruments were specific for AD but not for intellectual disability. The remaining 18 instruments for assessing cognitive function administered directly to the subject were not appropriate for use in individuals with intellectual disabilities or AD. Of the 18 instruments that were not appropriate for use in the population studied, 4 were instruments designed to assess children, although none of the selected studies included subjects under 18 years of age, and 3 were adapted specifically for use in the studies conducted. Only 1 study used qualitative instruments structured for use in research [45].

Of the 43 instruments employed in the selected studies, 12 were instruments that are administered to an informant. Of those 12, 5 were designed for use in intellectual disability and AD; 2 were designed specifically for use in AD but not intellectual disability; 2 were designed specifically for use in intellectual disabilities but not AD, and 3 were not designed specifically for use in intellectual disabilities or AD.

Four of the instruments employed were designed to assess overall cognitive integrity and to identify dementia by revealing evidence of a cognitive decline. Two instruments were designed to assess intelligence. In terms of specific cognitive domains, memory and language were investigated in 11 of the instruments; attention and executive function were investigated in 7; praxis was investigated in 4; orientation was investigated in 5; visual-motor integration and perception were investigated in 2; spatial ability was investigated in 2, and construction, general knowledge, conceptualization, calculation, and abstract thinking and problem solving were investigated in 1.

Although few of the selected studies focused on cognitive assessment, most provided that assessment in the secondary analysis of the data or did not present further details of individuals' cognitive performance. Six of the studies reported objective data for the subjects with dementia but not necessarily for all of the instruments used [12,20,36,37,38,43], whereas 5 studies reported only some general results for the sample as a whole, without distinguishing between the subjects with and without dementia [35,40,44,45,46]. Therefore, the data are related to cognition in adults with DS, across a wide age range, with or without cognitive decline. In one study, the results of the cognitive assessments were available only as supplemental data [46]. Three other studies reported no objective data from cognitive assessments [39,41,42]. In another 3 studies, it was necessary to refer to earlier studies by the same group in order to determine which cognitive assessment instruments were used [39,42,43].

The neuropsychological aspects of frontal lobe functions include executive functions (such as planning, inhibitory control, working memory, and abstract thinking), attention, and behavior. In this review, we identified 3 studies that employed instruments specific for the assessment of executive functions [20,43,45]. Eight of the selected studies used screening instruments specific for the assessment of attention [35,37,38,39,40,41,42,45]. In the study conducted by Esteba-Castillo et al. [35], abstract thought was evaluated with the Cambridge Cognitive Examination adapted for persons with Down syndrome or other intellectual disabilities. Working memory was evaluated in 2 studies [20,39], although quantitative data were reported in only 1 of the 2, that conducted by Adams and Oliver [20]. Those authors found significant differences between the subjects with and without dementia, over a 16-month period, in terms of executive functions [t(28) = 3.3, p = 0.01], cognitive decline being observed only in the subjects with dementia.

Ten studies employed research protocols that included the study of behavior, using a total of 5 instruments to assess behavior and adaptive skills. Two of those studies investigated only aspects of adaptive ability [37,45], whereas the other 8 evaluated aspects of behavioral disorders [12,20,35,36,40,43,44,46]. One of those 8 evaluated only aggressiveness [36], the remaining studies evaluating behavioral changes (excesses and deficits) through the use of specific instruments.

Adams and Oliver [20] used neuropsychological tests to examine the relationship between a decline in executive functioning and behavioral changes. The authors concluded that significant changes in behavior, whether excesses or deficits, were associated with a decline in executive functioning, although not in memory. They also stated that even in the early stages of cognitive deterioration, it is possible to identify specific changes in behavior that are not present in individuals without signs of dementia. They suggested that this pattern of change and decline is indicative of the process of frontal lobe neurodegeneration. The results of the study conducted by Oliver et al. [43] also showed that behavioral changes (excesses and deficits) are more common in aging DS patients with AD than in those without. That partly explains the fact that DS is more difficult to manage in adult patients with AD than in those who are younger or are of the same age but show no signs of dementia. In addition, d'Orsi et al. [12] stated that behavioral symptoms are often misdiagnosed, the early symptoms of dementia being mistaken for psychiatric symptoms of other causes in 50% of cases. In the study conducted by Carr and Collins [36], DS patients considered to be in the early stages of dementia showed no increase in aggression or other behavioral problems, although that study had limitations in terms of data collection. In a study involving DS patients, with and without AD, who were treated with memantine or received a placebo, Hanney et al. [40] found no differences among the groups in terms of adaptive skills.

Only 2 of the selected studies employed neuroimaging [12,46]. Using magnetic resonance imaging (MRI) data, d'Orsi et al. [12] demonstrated diffuse brain atrophy in all DS patients with AS, although the authors did not provide further details of the MRI findings. Koran et al. [46] also employed MRI and found that the effect of age was greatest in the frontal regions of the brain (especially the orbital part of the inferior frontal gyrus) and in the parietal lobe.

All the 14 studies in this review included the investigation of at least 1 of the following aspects: cognitive functions related to the frontal lobe, behavior changes or AD features on neuroimaging. Nevertheless, the association of frontal lobe functions and dementia was beyond the scope of the majority of them. Summarizing the results of their investigations in this area: 4 studies found results that pointed out a direct association between components of frontal lobe functions and signs of dementia [12,20,43,46], 3 other studies did not find a direct relation but still considered changes in behavior as a possible confounder of AD symptoms [35,36,40], and the other 7 studies had inconclusive results on the subject.

Although recent data indicate that the onset of AD in individuals with DS is characterized by symptoms typical of early frontotemporal degeneration, there have been few studies using cognitive measures that might confirm such an association. Although executive dysfunction is one of the classic early symptoms of frontotemporal degeneration [48,49], only a few recent studies on AD and DS have employed instruments designed specifically for the assessment of executive functions. The investigation of cognitive aspects related to the frontal lobe is still an emerging field in neuropsychological research, which could explain the paucity of studies in the area.

In this systematic review of the literature on AD in DS, we found only 1 study focusing specifically on cognitive and behavioral aspects related to the frontal lobe. The study conducted by Adams and Oliver [20] was the first to demonstrate an association between behavioral changes and a decline in executive functions in individuals with DS showing signs of cognitive deterioration. In addition, we found no studies that investigated the relationship between symptoms of frontal lobe degeneration and evidence of structural changes in the brain. Among other aspects related to frontal lobe degeneration, some studies have evaluated nonadaptive changes in the behavior of individuals with DS and AD, indicating that there is an association between changes in behavior and dementia in this population. Despite the limited number of studies in this area, the cognitive findings [20], the neuropathological evidence [46], and the fact that behavioral change is a possible confounder of the symptoms typical of AD [12,20,35,36,40,43,46] indicate that the frontal lobe is one of the first brain regions affected in dementia.

In the selected studies, most of the instruments used in the assessment of neurocognitive functions were not adapted for use in individuals with intellectual disabilities or dementia. The lack of appropriate instruments might explain the fact that so many instruments are employed in this area, which leads to methodological problems that have a negative effect on the development of new knowledge, whether for research or clinical practice.

The fact that the studies evaluated here employed so many different diagnostic criteria, and that some studies applied more than one set of diagnostic criteria, whereas others considered only cognitive deficits suggestive of dementia, underscores the idea that making a diagnosis of AD is a complex and difficult proposition in this population.

In this review, we encountered a number of difficulties. There was a lack of data related to objective measures of cognitive functioning and to the instruments used. In addition, our sample was quite heterogeneous in that the assessment of cognitive function was a secondary objective in the vast majority of the selected studies, and there was therefore little or no information on the related variables. Furthermore, we found few studies evaluating dementia in DS, which forced us to modify the exclusion criteria and include studies whose samples included subjects with signs of dementia and not exclusively those diagnosed with it. Moreover, given the heterogeneity of the information obtained and the scarcity of data from cognitive assessments, it was not possible to conduct a meta-analysis to integrate data from independent studies and thus devise a common measure.

Given the increasing life expectancy of individuals with DS, together with the proven genetic relationship between DS and AD, as well as the neuropathological and psychiatric features of the symptomatology of dementia in such individuals, there is a need for further studies to clarify the neuropsychological and pathological profile of aging in this population. Studies that broaden the discussion of the cognitive reserve hypothesis as a possible explanation for the differences observed among individuals with DS in terms of the clinical presentation of dementia [20,22] could be developed with the aid of measures of cognition and behavior that allow brain morphology and function to be jointly evaluated. Characterization of the unique features of the clinical symptoms of AD in individuals with DS could facilitate the diagnosis of AD and inform decisions regarding its treatment, as well as stimulate discussion on the creation of a classification system for the specific clinical features of AD in DS.

This study was supported by Fundação de Amparo a Pesquisa do Estado de São Paulo grant 2013/11571-9.