Abstract
Amyotrophic lateral sclerosis (ALS) and frontotemporal dementia have a strong clinical, genetic, and pathological connection but association of ALS with Alzheimer’s disease (AD) is seldom reported. We report a series of 5 cases of AD associated with ALS. Our patients presented with cognitive deterioration with episodic memory impairment meeting criteria for AD. ALS occurred subsequently in all cases and its phenotype was not homogenous. Amyloid process was confirmed in four cases with cerebrospinal fluid biomarkers. One case underwent postmortem exam, demonstrating hallmarks lesions of both diseases. This series highlights that ALS-AD phenotype could be a specific underexplored entity.
Keywords
INTRODUCTION
Amyotrophic lateral sclerosis (ALS) is a sporadic or familial (around 10% of cases) fatal neurodegenerative condition that affects motor neurons in the primary motor cortex, brainstem, and spinal cord. While predominantly a motor disorder, ALS may be associated with cognitive impairment and behavioral changes [1]. Up to 15% of ALS patients meet formal criteria for a diagnosis of frontotemporal dementia (FTD) [2]. The clinical, neuropathological, and genetic overlapping observed favors the existence of an ALS-FTD spectrum. ALS and FTD share TDP-43 protein immunoreactive neuronal cytoplasmic inclusion areas of neurodegeneration [3]. The discovery of C9ORF72 gene hexanucleotide expansion in familial cases of ALS-FTD has also supported this link [4]. Less frequently, cognitive impairment associated to ALS can present as amnestic syndrome [5]. Neuroimaging studies with MRI have reported ALS-specific patterns of hippocampal atrophy [6]. Postmortem studies have also highlighted hippocampal involvement in ALS [7, 8]. Evidence of TDP-43 positive inclusions in hippocampal structures with neuronal loss and gliosis is in favor of possible disease-specific memory deficits related to ALS lesions. In association, pathological examinations in ALS patients with or without dementia have also described amyloid deposition in up to 50% of cases, suggesting that amyloid process could play a role [9]. Nevertheless, the relationship between these findings and clinical manifestations remains unclear. We report a series of cases of association of Alzheimer’s disease (AD) and ALS and discuss the mechanisms that could account for the underlying neurodegenerative process.
MATERIALS AND METHODS
Cases of defined AD associated to confirmed ALS were collected in French Memory Clinic and Research Centers. Diagnosis was made according to the NIA-AA work group including AD cerebrospinal fluid (CSF) biomarkers or pathological evidence for AD and to the revised El Escorial criteria for ALS [10, 11]. Six cases were collected: 5 with probable AD dementia with evidence of AD pathophysiological process and one case with definite AD (pathological evidence), all fulfilling ALS criteria. The cases originated from Lariboisière Fernand-Widal Hospital Paris, Lille University Hospital, Besançon University Hospital, and Rouen Normandie University Hospital. One case was excluded as the CSF biomarkers did not provide clear evidence of amyloid process.
Description of the cases includes personal and familial history, description of AD and ALS presentation, and outcome. All patients underwent detailed neuropsychological assessment at baseline. Neuropsychological assessment included global cognitive function assessment (Mini-Mental State Examination [MMSE] and/or Mattis Dementia scale test) and evaluation of specific domains: episodic memory (Free and Cued Selective Remining Test [12]), short-term memory (digit span), executive function (Frontal Assessment Battery at Bedside [13]), visuospatial (Rey–Osterrieth copying [14]), language (Dénomination Orale 80 test [15]), and praxies. Brain MRI was performed in all cases and cortical atrophy was rated visually as well as Scheltens and Fazekas scores for hippocampal atrophy and white matter lesions respectively [16, 17]. CSF was collected by lumbar puncture and Innotest® was performed to measure amyloid-β 42 (Aβ42) phospho-tau (p-tau) and total tau (t-tau) in all patients but one. The patient without CSF analysis underwent postmortem exam. Neuropathological assessment was performed according to the standard criteria for the diagnosis of neurodegenerative disease, including Braak staging for NFT [18], CERAD evaluation of amyloid load [19] and reporting of co-existing pathology. Written consent was provided for the use of anonymized clinical data for research purposes.
DESCRIPTION OF CASES
Cases are summarized Table 1.
Cases description
*CSF biomarkers were available for the patients N°1 to 4 measured with Innotest® (Normal ranges: Patient N° 1 to 3: Innotest®, cut-offs: Aβ42: N > 700 pg/mL, phospho-tau 60 < pg/mL<N, total tau < 340 pg/mL; normal ranges patient N°4: Innotest®, cut-offs: Aβ42: N > 700 pg/mL, phospho-tau 67 < pg/mL<N, total tau < 500 pg/mL). Patient N° 5 underwent postmortem examination. #Aβ40/Aβ42 ratio was available for Patient N°1 (normal < 20). Aβ42, amyloid beta 1–42; CSF, cerebrospinal fluid; FAB, frontal assessment battery; FCSRT, free and cued selective remining test; GRN, proganulin mutation; MMSE, Mini-Mental State Examination; MRI, magnetic resonance imaging; NA, non available; p-tau, phospho-tau; TBI, traumatic brain injury; t-tau, total tau; WM, white matter.
Case N°1
Patient N°1 was a 67-year-old woman referred in the memory clinic for a 2-year memory impairment and modifications in behavior. She had an history of tachycardia. There was no familial history of neurological disease. She exhibited irritability, apathy, and sweet food preference. MMSE scored at 23. Neuropsychological testing showed dysexecutive syndrome and alteration of social cognition. Brain MRI displayed isolated temporal atrophy Scheltens 3. Genotyping of APOE showed ɛ4/ɛ4 status. CSF biomarkers showed an AD profile with decreased Aβ42 and increased p-tau and t-tau. A diagnosis of behavioral AD was made, and she was started on anticholinesterase and thymoregulators regarding the behavioral symptoms. Five years later, she presented a dropped head followed by a bulbar syndrome associating dysarthria and swallowing impairment. Needle-electromyography found diffuse motoneuronal impairment. A diagnosis of ALS was made. C9ORF72 testing did not show expansion. Patient N°1 died of aspiration pneumoniae after a 2-year motor duration.
Case N°2
Patient N°2 was a 69-year-old woman, with a history of appendectomy. Familial history was profuse. Her mother had presented with dementia since the age of 75. Three of her brothers had a diagnosis of AD. One presented with associated dysarthria and swallowing impairment. At age 67, patient N°2 showed a progressive memory impairment. No behavioral change was associated. At cognitive evaluation, global cognition was altered (MMSE scored at 23 and Mattis at 119), as well as episodic memory. MRI showed marked hippocampal atrophy, associated to hypersignal in T2 gradient echo sequence of basal ganglia and cerebellar dentate nucleus. Core CSF biomarkers showed AD profile with decrease of Aβ42 and increase of p-tau and t-tau proteins. The patient was started on anticholinesterase for probable AD. Four years after AD onset, patient N°2 presented with progressive bulbar syndrome, dysarthria, and swallowing impairment. Needle-electromyography confirmed pure motor impairment in the 4 limbs and face with signs of acute processes. Analysis for C9ORF72 expansion was negative. Death occurred after a 2-year duration of ALS symptoms.
Case N°3
Patient N°3 was a 65-year-old man. His medical history counted a cholecystectomy, prostatic adenoma treated by anticholinergic medication, and brain traumatic injury 8 years prior to first symptom. No familial history was noted. First symptoms occurred at age 65 characterized by progressive memory impairment. No change in personality or behavior was noted. At first cognitive assessment, MMSE scored 30. First MRI at age 67 showed isolated white matter hyperintensities without atrophy. Biomarkers were pathological with decrease of Aβ42 and increase of p-tau and t-tau proteins, showing amyloid and neurodegenerative processes. Three years after the first cognitive signs, patient N°3 presented signs in favor of a spinal form of ALS. Needle-electromyography confirmed active denervation, with numerous fasciculations. Medulla MRI showed cervical vertebral arthropathy with no medullar lesion. Patient died two years after onset of motor symptoms.
Case N°4
Patient N°4 was an 82-year-old male, retired from the car industry. Personal history associated carpal tunnel syndrome, surgery for bilateral inguinal hernia, and rotator cuff tear. Familial history was unremarkable. He presented with a one-year history of progressive cognitive impairment.
MMSE scored at 22. Neuropsychological testing showed on first ground attention deficit and dysexecutive syndrome. MRI showed bilateral hippocampal atrophy Scheltens 3 and white matter lesions scored Fazekas 2. CSF exam showed decreased Aβ42 protein, limit p-tau protein and normal t-tau, in favor of prodromal AD. He later installed diffuse motor deficit with diffuse atrophy and fasciculations and pyramidal syndrome. Needle-electromyography was in favor of ALS. Death occurred at one-year of diagnosis.
Case N°5
Patient N°5 was a 76-year-old female, who had worked as a physical education teacher. She had presented depression treated by a serotonin-norepinephrine reuptake inhibitor. Her father had received a diagnosis of AD at the age of 70. At age 61, she presented a memory complaint with a first assessment at age 73 that showed cognitive performance within normal range. Memory impairment then progressed with loss of autonomy. At age 76, she presented with swallowing impairment, sialorrhea, and dysarthria, followed by general weakness. A diagnosis of bulbar ALS was made after a needle-electromyography objectified 1rst and 2nd motor neuron impairment. A cognitive assessment at age 77 showed predominant episodic memory and visual memory impairment. The patient died one year after onset of motor symptoms. Postmortem examination was performed and showed both pathologies in favor of ALS and AD: anterior horn atrophy with motor neuron loss with positive anti-TDP43 inclusions, Braak IV neurofibrillary tangles, and CERAD C amyloid load as well as alpha-synuclein deposits (Fig. 1).

Brain postmortem examination of Patient N°5. A) Lateral view of the left fixed hemisphere showing mild and global cortical atrophy. B) Coronal section of the left fixed hemisphere through the lateral geniculate body showing moderate hippocampal atrophy. C) Microscopic aspect of the temporal cortex after hematoxylin-eosin staining showing mild neuronal loss and some senile plaques. D) Microscopic aspect of the temporal cortex after phosphorylated-tau protein immunostaining showing abundant dystrophic neurites and neuritic plaques. E) Multiples focal amyloid deposits are revealed by Aβ immunostaining in the temporal cortex. F) Microscopic aspect of the hypoglossal nerve nucleus after hematoxylin-eosin staining severe neuronal loss and spongiosis. G) TDP-43 immunostaining showed several filamentous “skein-like” inclusions in remaining neurons of the hypoglossal nerve nucleus. H) Alpha-synuclein immunostaining revealed Lewy bodies and neurites in several limbic areas such as anterior cingular cortex presented here. Scale bar for all microscopic images: 50μm.
DISCUSSION
We report the association of AD and ALS in 5 patients. In four cases, biomarkers objectified AD process and needle-electromyography motoneuron disease. Neuropathological examination demonstrated pathological hallmarks of both diseases in one case. This phenotype has been seldom reported. One case of clinical AD with positive CSF biomarkers followed by ALS at age 60 has been reported by Farid et al. [20]. Nijboer et al. reported the case of a patient with ALS that presented with posterior cortical atrophy syndrome, that could indicate associated amyloid pathology [21]. Our cases presented as late-onset AD (mean age: 69 years old). All demonstrated amnestic syndrome of hippocampal profile, evaluated on the Free and Cued Selective Reminding Test, that has been shown to distinguish efficiently AD from FTD [22]. Cognitive impairment described in ALS typically affects foremost executive functions. In the memory impairment observed in pure ALS phenotype, it is mostly the encoding process that appears deficient [23]. In our series, patients displayed evolutive memory deficits, whereas patients with ALS have been shown to present only modest decline of memory impairment during disease course [24]. Patient N°1 met the criteria for possible bvFTD but a diagnosis of frontal variant of AD was made, prompted by the neuropsychological assessment and AD CSF signature. The 4 patients with CSF analysis displayed a A + T+profile, which makes it less likely that this observation is solely related to physiological modification in CSF biomarkers that can be observed in aging. High t-tau levels (above 1000 pg/mL) were observed in two cases in favor of important axonal neurodegeneration. Onset of motor impairment occurred after cognitive impairment in all cases (mean delay: 3.4). The same chronology was observed by Farid et al. [20]. ALS phenotype in our cases was heterogenous: the presentation was bulbar for three patients, spinal and diffuse for one patient respectively. Duration of motor symptoms was less than 3 years in all cases. In ALS cohorts, older age is associated with increased probability of bulbar form and poor prognosis [25].
Several hypotheses on the physiopathology underlying this phenotype can be made, supported by pathologic, biochemical, and genetic arguments. In the case reported by Farid et al., functional imaging by positron emission tomography (PET) including FDG-PET, amyloid PET, and neuroinflammation 11C-DED PET, demonstrated respectively frontal and temporal hypometabolism, cerebral amyloidosis and astrocytosis consistent with the double neurodegenerative process [20]. Postmortem examination in case N°5 retrieved lesions of both diseases. Existence of amyloid deposits in ALS patients is common [9, 26]. In neuropathological series, prevalence of significant amyloid pathology ranged from 22 to 50% [9, 26]. In Coan et al., 22% of 46 sporadic ALS patients met the CERAD criteria for pathological AD [26]. Up to 16.6% of a series of 18 sporadic ALS patients exhibited positive amyloid PET [27]. The contribution of the amyloid deposits to the cognitive impairment remained unclear in those different studies however [9, 27]. It is now established that co-pathologies are highly prevalent in neurodegenerative diseases. Co-pathologies were observed in up to half of cases in various neurodegenerative disorders and the highest in AD [28]. Aging, proteopathic seeding effects, and genetic factors appeared as significant drivers of these co-occurrence. TDP-43 has been characterized as a main hallmark of ALS. Numerous studies have indicated TDP-43 deposits in AD brain observing a colocalization of TDP-43 with Aβ42 and tau [28]. In vitro, TDP-43 issued from AD brain could be used as seed agent to trigger tau aggregation [29]. TDP-43 was also shown to have a strong catalytic effect in vitro on Aβ aggregation [30].
Association of PD, ALS, and dementia has been reported in the Guam island [31]. Typical pathology was described as association of neurofibrillary tangles with a tau distribution similar to that of AD and in cases presenting with ALS and dementia, extensive neurofibrillary tangles in the hippocampus with amyloid pathology [32, 33]. Familial circumscription of the disease supports a genetic hypothesis and single nucleotide polymorphisms in the coding region of tau were identified and reported to constitute significant risk factors [34].
In our cohort, Patient N°2 presented with an history of dementia in 4 first-degree relatives, suggesting a genetic character. Two patients were screened for C9ORF72 expansion but no pathogenetic variant was observed. C9ORF72 expansion could have accounted for the association of cognitive impairment, behavioral changes and motoneuron disease, as amyloidopathy may be found in carriers and it has been shown to be a risk factor for AD [35, 36].
We finally cannot exclude the possibility that the co-occurrence of ALS with AD in our series was coincident but the clinical similarities between the reported cases and existing evidence on possible underlying physiopathological mechanisms prompt for deeper investigations. Moreover, we calculated that, with an incidence of ALS in France evaluated at 2.5/100,000 [37], as the patients were included over 5 years in a total cohort of circa 12,000 AD patients followed in the different centers [38], theoretical number of cases in our cohort would be estimated at 1.25 case over 5 years. Thus, we can suspect a higher occurrence of ALS in our cohort than predicted, suggesting a link between the two diseases. Further studies will be needed to better estimate the prevalence of the association. Patients reported in our series were issued from a memory clinic setting; study of patients in facilities specialized in ALS could provide more insights.
In conclusion, these observations demonstrate that the association of AD and ALS might constitute an underexplored specific phenotype. The mechanistic links between both diseases remain unknown. Nevertheless, recognizing the breadth of neurodegenerative disorders is necessary to ensure proper diagnosis and to guide further inquiry into their mechanisms.
DISCLOSURE STATEMENT
Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/21-5226r1).
