Coexistence of Amyotrophic Lateral Sclerosis and Alzheimer’s Disease: Case Report and Review of the Literature

INTRODUCTION

While the genetic, pathophysiological, and phenotypic link between amyotrophic lateral sclerosis (ALS) and frontotemporal degeneration (FTD) has been thoroughly acknowledged [1, 2], less is known about the association between ALS and Alzheimer’s disease (AD), although such a co-occurrence was reported as early as 1913, when Barrett described a patient with dementia and comorbid ALS, whose autoptic examination revealed that “all regions of the cortex showed numerous plaques and the peculiar neuro-fibril alterations described by Alzheimer and others [...]” [3].

In fact, reports on the co-existence of these two conditions can be traced throughout the nineteenth [4] and twentieth centuries [5], with recent evidence supporting a link at both neuropathological and neurochemical levels [6, 7]. Indeed, TDP-43 pathology, i.e., the neuropathological hallmark of ALS-FTD-spectrum disorders, can be found in up to 60% of AD cases [7], while approximately half of patients classified within the ALS-FTD spectrum have been shown to present with concomitant AD pathology [6]. Moreover, both postmortem AD pathology [6] and AD-specific neurochemical biomarkers [8, 9] have been reported to be associated with cognitive impairment in patients within the ALS-FTD spectrum.

The above being said, no synoptic study to date has focused on defining the clinical presentation of the co-occurrence of ALS and AD. Hence, within the present work, a case report on such a co-occurrence is first presented, followed by a literature review which aims at drawing conclusions on the main features of this disease association.

CASE REPORT

A 76-year-old woman developed memory and executive deficits, followed two months later by dysarthria and dysphagia; she came to clinical attention one year after the onset of the first symptoms. She had no family history of brain disorders. Her previous medical history was notable for arterial hypertension, dyslipidemia, and previous lumbar spine surgery.

On neurological examination, she was oriented, able to recall autobiographical memories, and aware of her cognitive decline. Mild anomia without further aphasic features was noted in spontaneous speech. She also had slight ideomotor apraxia. She was severely dysarthric and dysphagic, with tongue atrophy and fasciculations and a brisk jaw jerk. Lower facial and sternocleidomastoid muscles were bilaterally weak. The four limbs had preserved muscle mass; however, finger clumsiness in both hands and mild weakness of the right upper limb were noted. Deep tendon reflexes were normal in the upper limbs and decreased in the lower limbs. The snout reflex and glabellar sign were absent, but palmo-mental reflex was present on the right side. Plantar reflexes were normal bilaterally.

Formal neuropsychological testing revealed severe widespread cognitive deterioration characterized by impairment of executive functioning, selective attention, long-term memory, lexical retrieval and constructional praxis, while short-term memory abilities were relatively spared. Tests that heavily relied on verbal and motor responses (i.e., verbal fluency and Stroop tasks) could not be administered due to dysarthria and upper-limb deficits. The patient was unable to perform Part B of the Trail-Making Test. Psychodiagnostic evaluation suggested the presence of severe state and trait anxiety along with moderate depression.

Needle electromyography revealed chronic neurogenic changes in muscles of the bulbar region (right genioglossus and sternocleidomastoid and left masseter), upper and lower limbs, and abdominal wall, while signs of active denervation were found in the upper limbs. This was consistent with a diagnosis of possible ALS according to the revised El Escorial [10] and Awaji criteria [11] and with a diagnosis of ALS according to the more recent Gold Coast criteria [12]. Transcranial magnetic stimulation of the motor cortex disclosed normal central motor conduction time for the four limbs and bilaterally reduced duration of the cortical silent period for the upper limbs. Arterial blood gas analysis revealed mild hypoxic-hypercapnic respiratory insufficiency (PaO₂, 67 mmHg; PaCO₂, 48 mmHg), with increased standardized bicarbonate level (sHCO3^-; 31.3 mmol/l). On pulmonary function testing, forced vital capacity was slightly above the lower limit of the normal range (83% of predicted value).

Brain magnetic resonance imaging (MRI) revealed mild widespread cortical atrophy and multiple small hyperintensities in the white matter of both hemispheres with frontal predominance, consistent with chronic cerebrovascular lesions. Electroencephalography showed predominant theta activity over the bilateral temporal-parietal regions. ¹⁸F-fluorodeoxyglucose positron emission tomography (¹⁸F-FDG PET) demonstrated left-greater-than-right reduction of glucose uptake in the posterior cingulate and temporal-parietal regions, whilst Aβ-PET with ¹⁸F-florbetapir disclosed brain Aβ amyloidosis (Fig. 1). Quantification of cerebrospinal fluid (CSF) biomarkers by means of enzyme-linked immunosorbent assay (ELISA) revealed an A + /T+/N+ profile [13], with low Aβ₄₂ (469 pg/ml; normal values, >647 pg/ml) [14] and high total tau (T-tau; 723 pg/ml; normal values, <500 pg/ml) [15] and phosphorylated tau levels (P-tau₁₈₁; 72 pg/ml; normal values, <61 pg/ml) [16]. Unfortunately, neither CSF levels of Aβ₄₀ (and hence the Aβ₄₂/Aβ₄₀ ratio) nor CSF or blood neurofilament levels could be determined. No mutations in the main ALS-associated genes (C9orf72, SOD1, TARDBP, and FUS) were found. The APOE genotype was ɛ3/ɛ4. We did not perform genetic analysis of APP, PSEN1, or PSEN2 genes.

OPEN IN VIEWER

Fig. 1

¹⁸F-FDG PET demonstrating left-greater-than-right reduction of glucose uptake in the posterior cingulate and temporal-parietal regions (left panel) and ¹⁸F-florbetapir PET showing brain Aβ amyloidosis (right panel). ¹⁸F-FDG, ¹⁸F-fluorodeoxyglucose. PET, positron emission tomography.

Based on the above findings, a diagnosis of ALS with concurrent AD was made. A summary of the patient’s clinical features is provided in Table 1. Over the following months, worsening weakness of the four limbs developed, and the patient died from cardio-respiratory arrest 22 months after onset of the first symptoms.

Informed consent was acquired from the patient described within the present study.

METHODS

A literature search was performed on April 12, 2023 in PubMed and Scopus databases by entering the following word string: (“amyotrophic lateral sclerosis” OR “motor neuron disease”) AND (“Alzheimer” OR “Alzheimer’s”). The field of search was the title. Additional records of potential interest were searched for within the reference lists of included articles. For a record to be included, it had to address patients with both 1) a clinical diagnosis of ALS/motor neuron disease (MND) and 2) a co-occurring in vivo (i.e., neurochemical or based on molecular imaging) or neuropathological diagnosis of AD. Group studies, case reports and case series were considered for inclusion. Non-original records and those without a full-text available were excluded. Records addressing the ALS-parkinsonism-dementia complex (ALS-PDC) of Guam or similar “overlap syndromes” [17] were excluded as well.

The following outcomes were extracted: 1) demographics, i.e., age at onset and sex; 2) disease duration; 3) genetics; 4) family history of brain disorders; 5) onset type, i.e., motor versus neuropsychological; 6) delay between the onset of motor and neuropsychological signs/symptoms; 7) bulbar signs at ALS onset and at referral; 8) cognitive domains/functions involved; 9) behavioral features; 10) positivity of AD-specific neurochemical/molecular imaging biomarkers; 11) morphological/functional features on brain imaging; and 12) neuropathological findings consistent with ALS and/or AD. Regarding point 11, we focused on the topography of atrophy (as shown by brain computed tomography (CT) or MRI), reduced cerebral blood flow (as shown by single-photon emission computed tomography (SPECT)), or cerebral hypometabolism (as shown by ¹⁸F-FDG PET). We subdivided the morphological findings into three categories: AD-like (predominant medial temporal atrophy and/or parieto-temporal reduction of blood flow or metabolism), FTD-like (predominantly frontal or frontotemporal atrophy and/or reduction of blood flow or metabolism), or unspecific (no topographical predominance of findings). In case of normal SPECT results (i.e., no abnormalities of cerebral blood flow) in the presence of slight brain atrophy, we deemed the findings as nonspecific. In case of topographically unspecific morphological results along with more specific functional findings (¹⁸F-FDG PET or SPECT), we classified the case according to the functional findings.

RESULTS

The study selection process, pursuant to the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [18], is shown in Supplementary Figure 1. Starting from N = 149 unique records initially identified, N = 13 were actually included (case reports: N = 11; case series: N = 2), which addressed N = 30 patients (males, N = 13 (43.3%); females, N = 17 (56.7%); mean age at onset, 63.8±9.7 years; range, 33-80 years) (Table 2).

In N = 24 cases (80.0%), the diagnosis of AD was based on neuropathology. Among the N = 6 cases without available neuropathological examination, AD diagnosis was supported by CSF biomarkers and Aβ-PET in one patient, by Aβ-PET alone in one patient, and by CSF biomarkers alone in 4 patients. Regarding the retrospective series of ALS patients with concomitant AD neuropathology of Hamilton and Bowser [19], we excluded two cases (both males; ages, 68 and 73 years, respectively) who had only sparse Aβ plaques, no neuritic plaques, and no relevant neurofibrillary tangle pathology (i.e., they had Braak stage 0). In the N = 24 cases with neuropathological evidence of AD, the autopsy findings also confirmed the diagnosis of ALS. Of note, however, the patient described by McCluskey et al. had no TDP-43 inclusions, but rather tau pathology in both the motor cortex and spinal cord motor neurons [20].

Among those patients for whom information on neurological family history was reported (N = 14), N = 9 (64.3%) had no family history of neurological disorders, whereas N = 5 (35.7%) had a positive family history (dementia, AD, and AD associated with possible bulbar MND, N = 1; AD and ALS, N = 1; AD, N = 1; dementia associated with progressive motor disability, N = 1; vascular dementia, N = 1).

For N = 13 patients, information on the chronological relationship between the onset of motor and cognitive symptoms was available. Of these, N = 9 (69.2%) had a neuropsychological onset and N = 2 (15.4%) a motor onset, whereas in N = 2 (15.4%) patients, motor and cognitive symptoms appeared at the same time. For patients with neuropsychological onset, the mean time interval between onset and appearance of motor symptoms was 34.2±22.3 months (range, 12-84 months), while the opposite intervals for the two patients with motor onset were “few months” and 60 months, respectively. Mean disease duration (N = 28) was 39.0±30.1 months (range, 5-144 months). Information on the site of onset of ALS was provided for N = 17 patients. In N = 7 of these (41.2%), motor impairment first manifested in the bulbar region, whereas the onset was spinal in only N = 6 patients (35.3%); in N = 4 patients (23.5%), ALS symptoms appeared simultaneously in the bulbar and spinal segments. Excluding the N = 12 patients for whom no data about presence/absence of bulbar signs were available, among the other N = 18 such signs were present in the majority (N = 15, i.e., 83.3%). Notably, two patients were actually diagnosed with primary lateral sclerosis (PLS) rather than ALS properly. Genetic analyses were performed in only N = 5 patients (C9orf72, N = 3; APOE, N = 2; MAPT, TARDBP and SOD1, N = 1; GRN, N = 1). No pathogenic (i.e., ALS- or FTD-causing) mutations were found, while the APOE genotypes in the two patients tested were ɛ3/ɛ3 and ɛ3/ɛ4, respectively.

Morphological brain imaging was available in N = 15 patients (MRI only, N = 12; CT only, N = 1; MRI and CT, N = 2). Among these patients, N = 7 also had functional imaging (¹⁸F-FDG PET, N = 2; SPECT for cerebral blood flow, N = 5). The topographical pattern of brain atrophy or hypofunction was AD-like in N = 8 (53.3%), FTD-like in N = 1 (6.7%), and unspecific in N = 6 (40.0%).

With regard to cognitive features, N = 7 (23.3%) ALS patients showing AD pathology at autopsy were described as cognitively unimpaired. On the contrary, some patients were merely reported to have dementia (N = 3), aphasia (N = 1), or both (N = 3), without further detailing impaired cognitive functions. As to the remaining patients (N = 16), for whom more specific cognitive data were available, memory and executive/attentive deficits were the most represented (N = 12, i.e., 75.0%, and N = 10, i.e., 62.5%, respectively), followed by disorientation (N = 6, i.e., 37.5%) and language dysfunction (N = 4, i.e., 25.0%). On the other hand, apraxic and agnosic features were reported only in N = 2 (12.5%) and N = 1 (6.2%) patients, respectively.

As for behavioral changes, when assessed (N = 18 patients), disinhibition was the most frequent one (N = 5, i.e., 27.8%), followed by apathy (N = 4, i.e., 22.2%) and depression/anxiety (N = 4, i.e., 22.2%). Pseudobulbar affect was reported in N = 3 patients (16.7%), whereas perseveration, irritability and aggressiveness were present in N = 2 (11.1%), N = 1 (5.5%), and N = 1 (5.5%) patients, respectively. Conversely, N = 7 (38.9%) patients were allegedly free of behavioral dysfunctions.

DISCUSSION

We have reported a case of ALS associated with AD and reviewed the relevant literature. Our review suggests the following concepts on the coexistence of these two disorders: 1) the mean age at onset was similar to that of ALS [21] but lower than that of AD [22]; 2) conversely, the slight female predominance was more in agreement with the epidemiology of AD [23] than with that of ALS (showing a clearly higher incidence in males [24]); 3) the appearance of cognitive deficits often preceded the onset of ALS (namely, in more than two-thirds of patients for whom the chronological order could be retrieved), with highly heterogeneous time intervals between neuropsychological and motor onset (range, 12-84 months), whereas motor-only or bimodal (i.e., cognitive+motor) onset was uncommon; 4) onset of ALS in the bulbar region (alone or in combination with the spinal one) was frequent (i.e., almost two-thirds of patients for whom information on the site of onset was available); 5) MND mostly presented as ALS [25]; 6) disease duration (mean, 39.0 months) was, as expected, consistent with that of ALS [21] and much shorter than that of AD [22]; 7) the cognitive phenotype was mostly that of classical, amnestic-predominant AD [26], but executive deficits and language dysfunctions, both possibly pertaining to the frontotemporal spectrum [2], were moderately represented (with the former being more frequent than the latter); 8) behavioral disturbances were rather frequent and resembled those common to both the ALS-FTD spectrum [2] and AD [26]; 9) in the majority of cases, the pattern of brain involvement was consistent with AD, featuring predominant atrophy in the medial temporal lobe and/or reduction of blood flow or metabolism in parietal-temporal cortices; 10) ALS/FTD-related genetic mutations appeared not to be a major etiology.

Therefore, the clinical features of the ALS-AD case herewith reported are mostly consistent with the framework outlined above, except for the fact that the time interval between the onset of cognitive and motor deficits was remarkably shorter (namely, 2 months) when compared to the mean delay of previous cases.

Besides providing a synoptic glance at the coexistence of ALS and AD, the present study also indirectly adds to the long-lasting debate on whether such a co-occurrence is merely fortuitous or reflects etiopathogenetic determinants leading to both ALS and AD [5]. Actually, the finding of an overall lack of unique features that could have specifically distinguished a putative “ALS-AD syndrome” from both ALS and AD alone supports, at least from a clinical standpoint, the notion of such a co-occurrence being likely coincidental. Indeed, albeit clinically distinct, ALS and AD share a set of environmental risk factors [27–29] and underlying pathophysiological mechanisms [29–32] associated with neurodegeneration. Consistently, “neurodegenerative overlap syndromes” have been described [17, 33–35]. Additionally, it has to be borne in mind that AD is the leading cause of dementia in individuals aged≥65 years [22]: hence, it might be postulated that, from an epidemiological perspective, such a co-occurrence might be most likely age-related. It should also be considered that the coexistence of multiple proteinopathies in the same patients is a well-known and age-related phenomenon, with a large retrospective series (N = 766 autopsies) reporting Aβ pathology in more than one-third of ALS cases and tau pathology (Braak stage III or higher) in 18% of ALS cases [36]. Interestingly, however, in 12 (i.e., 40%) of the 30 cases reviewed in our work the age at clinical onset of the first disease (whether ALS or AD) was < 65 years. Among the reviewed cases, several ALS patients with neuropathological evidence of AD were not reported to have overt cognitive impairment. Based on this, one could hypothesize that AD pathology may not rarely be an incidental finding in ALS patients, not determining mental deterioration. However, it should be noted that for most of these patients no detailed cognitive information was available, as formal neuropsychological testing was rarely performed [19]. More generally, the lack of a standardized neuropsychological assessment is a limitation of our work. It will be important in the future to include detailed neuropsychological testing results in studies on the coexistence of ALS and AD, especially in those focused on the neuropathological demonstration of AD in ALS cohorts. This is indeed facilitated by the increased attention on cognitive abnormalities in ALS (due to the established link with FTD [1]) and the connected increased practice of neuropsychological testing of ALS patients, at least with screening batteries such as the Edinburgh Cognitive and Behavioural ALS Screen (ECAS) [37]. Such investigations will contribute to elucidating the biological and clinical significance of possible coexisting AD pathology in ALS. The existence of a common biological pathway selectively leading to both ALS and AD has been suggested by previous work [38–40], also on a genetic basis [41, 42]. Shared genetic factors predisposing to both neurodegenerative diseases and positively selected through the course of human evolution have also been hypothesized [43]. However, as focusing on clinical reports, the present work does not allow to test such etiopathogenic hypotheses, which, therefore, deserve to be further explored in future studies.

Another limitation of our work is that many reviewed cases were studied in a time preceding the description of TPD-43 pathology in ALS in 2006 [44]. This applies especially to the retrospective neuropathological series of Hamilton and Bowser [19]. Therefore, we do not have a comprehensive and detailed view of motor neuron pathology in ALS cases with associated AD. This point is further highlighted by the report of McCluskey et al. of tau pathology in the motor cortex and spinal cord motor neurons, in the absence of TDP-43 inclusions, in a patient with ALS and AD not harboring MAPT mutations [20]: based on this, we cannot exclude that further ALS cases with associated AD included in our review did not actually feature TDP-43 pathology.

Finally, the present work highlights that, in ALS patients with dementia lacking unequivocal features of FTD, additional examinations besides clinical assessment should be performed in order to unravel possible AD-related pathophysiological processes. This stance would also be in agreement with neuropathological evidence indicating that AD pathology often contributes to cognitive deficits in ALS [6] as well as with reports showing that hippocampal pathology, and thus amnestic features, are not rare in ALS [45]. Most importantly from a clinical standpoint, the recognition of AD as the cause of neuropsychological disturbances in a patient with ALS would be important for the purpose of proper management.

In our case we did not measure CSF Aβ₄₀, therefore we cannot provide the value of the Aβ₄₂/Aβ₄₀ ratio. This could have allowed to make the neurochemical diagnosis of AD even more certain [46]. However, the level of P-tau₁₈₁ was clearly elevated, which can be considered quite specific for AD [16], and, perhaps most relevantly, the results of ¹⁸F-FDG and Aβ-PET strongly supported the diagnosis of AD. Moreover, we were not able to provide CSF or blood levels of neurofilaments (NFs; light chain (NFL) or phosphorylated heavy chain (pNFH)). This would have resulted in a neurochemical estimation of the rate (or extent) of neuroaxonal degeneration [47], which would have been particularly interesting considering that ALS and AD generally determine a marked and a slighter elevation, respectively, of NF levels in biological fluids [48]. One could speculate that NFs would have been found to be increased to the levels commonly observed in ALS because of the predominance of neuroaxonal degeneration associated with motor neuron loss. It will be interesting to quantify CSF and/or blood NF levels in future patients suffering from both diseases concomitantly. This issue raises a further point of practical relevance. It is known that the presence of co-pathologies influences neurochemical biomarkers, with, for instance, patients with Lewy body disease (LBD) and a CSF A + T+ profile displaying higher levels of synaptic and neuroaxonal degeneration biomarkers compared to LBD patients with an A-T- profile [49]. This would presumably apply also to cases in which one pathological entity has already come to clinical expression while the other pathology is still in a presymptomatic phase (but probably already determines neurochemical changes). This possibility should be kept in mind when interpreting biomarker levels for diagnostic/prognostic purposes as well as in the context of clinical trials targeting specific proteinopathies. This point might be important not only regarding CSF or blood levels of the unspecific neuroaxonal degeneration biomarker NFL but also plasma levels of P-tau₁₈₁, which are an accurate blood biomarker for AD but have also been demonstrated to be increased in ALS [50, 51], possibly being associated with LMN pathology [52].