Clinical manifestations

Presentation and course

The clinical expression of Western Pacific ALS-PDC represents a spectrum (22). Some patients present with amyotrophic lateral sclerosis, whereas in others the parkinsonian and dementia symptoms predominate (54; 59). Features of both amyotrophic lateral sclerosis and parkinsonism-dementia complex are commonly seen in the same individual and in the same family, and similar neuropathologic findings support the argument that they are different clinical manifestations of the same neurodegenerative disease process (22).

Amyotrophic lateral sclerosis presents clinically as an asymmetric mixture of upper and lower motor neuron pathology (33; 32). Lower motor neuron symptoms include progressive muscle weakness, muscle atrophy, fasciculations, and respiratory muscle weakness. Upper motor neuron symptoms include hyperreflexia, spasticity, presence of a Babinski sign, and appearance of primitive reflexes or frontal release signs, including the palmomental, grasp, and snouting reflexes. Bulbar symptoms include dysarthria and dysphagia, and in some cases are the initial presenting symptoms. Extrapyramidal signs including tremor (13), rigidity (9), and bradykinesia (5) were present in 5.4% of a Guamanian population diagnosed with amyotrophic lateral sclerosis between 1950 and 1979 (76).

Parkinsonism-dementia complex is loosely defined as progressive dementia with extrapyramidal features of parkinsonism, frequently accompanied by amyotrophy and upper motor neuron signs (59). In a review of 293 patients, progressive dementia was the presenting feature in 30% of cases but ultimately occurred in 98% (76). Parkinsonism presented most commonly with bradykinesia, speech difficulty, and gait apraxia without festination. Masked facies is usually an early feature, and rigidity is common. Rapid alternating movements may be slowed, with a relative preservation of handwriting capability. The tremor of parkinsonism-dementia complex is usually 4 to 7 Hz in frequency and confined to the hands (59). A progressive supranuclear palsy-like syndrome with a supranuclear gaze palsy has also been described, with this syndrome thought to be a common phenotype (47; 90). Other symptoms have been reported. An unusual and distinctive retinopathy is observed among Chamorro people, but more commonly in Chamorros with ALS-PDC (59.7% of ALS-PDC cases vs. 24.7% of asymptomatic Chamorros in a case series of 918 Chamorros by Steele and colleagues), and the presence of retinopathy increases the risk for later developing ALS-PDC, suggesting that it may be part of the same degenerative process (11; 29; 92). Autonomic insufficiency, most often presenting as orthostatic hypotension, is also more prominent in Guamanian neurodegeneration when compared to idiopathic Parkinson disease (50). Similar to idiopathic Parkinson disease, marked olfactory deficits are common to ALS-PDC but rarely seen in amyotrophic lateral sclerosis (01).

Prognosis and complications

Survival rates vary in relation to the decade of onset: 15% of men and 29% of women who developed symptoms in the 1950s lived beyond 10 years, whereas only 3% of men and 13% of women who developed symptoms in the 1970s lived beyond 10 years. Death is most commonly associated with pulmonary or urinary tract infection (76).

Biological basis

Etiology and pathogenesis

As is the case with other neurodegenerative diseases, the definitive etiology of ALS-PDC is unknown despite intensive study. The leading hypothesis is an exposure to a disappearing environmental etiology. This is thought most likely to be an environmental toxin, with the notion that individuals exposed to higher doses of the toxin developed ALS-PDC at a relatively early age, and individuals exposed to a lesser dose developed neurodegenerative disease at a later date (102; 49). An environmental cause would best account for the sharp decline in disease prevalence, increasing age of onset, and change in disease phenotype. Migration studies also support an environmental etiology. ALS-PDC has been documented to have occurred with greater frequency among migrants to and from Guam. Garruto and colleagues estimated that the crude mortality rate from ALS-PDC for Chamorro migrants from Guam was at least three times higher than U.S. mortality rates from amyotrophic lateral sclerosis, yet lower than rates of nonmigrant Chamorros living in Guam (25). In addition, Garruto and associates documented a crude mortality rate for ALS-PDC among Filipino migrants to Guam that was six times higher than Filipino migrants living in the U.S., yet half the rate observed among Chamorros living on Guam during the same period (26). These data suggest that the exposure the migrants were subjected to occurred in Guam, and the latency period for disease development may be several years or even decades. A case of autopsy-proven ALS-PDC was reported in an immigrant family to the Kii peninsula (42).

Among several possible environmental causes, genotoxins found in cycad plants have garnered the most attention (84). Anthropologist Marjorie Whiting initially proposed the cycad plant Cycas circinalis as the likely cause of the disease. Cycads are used for medicinal purposes in the Kii peninsula and West Papua, whereas the dietary habits of indigenous Chamorro included the consumption of cycad seeds that were typically ground into a flour to make tortillas (88; 98). Cycad seeds contain two toxins of interest: cycasin (methylazoxymethanol-β-d-glucoside), which is both neurotoxic and carcinogenic, and the neurotoxic nonprotein amino acid β-N-methylamino-l-alanine (BMAA).

Methylazoxymethanol (MAM) is the aglycone of cycasin and a potent alkylating agent that inflicts damage to DNA; this damage triggers cell signaling and DNA repair by O6-methylguanine methyltransferase (MGMT) (87; 82). The MGMT enzyme is encoded by the MGMT gene, which has been implicated as cause of an inherited form of Alzheimer disease. Additionally, it has been shown that cortical MGMT is reduced in both Alzheimer disease and Guam PDC (39). In addition to being an alkylating agent, methylazoxymethanol is also readily converted to formaldehyde at physiologic temperatures. Formaldehyde has poorly understood acute and chronic neurotoxic effects, and its relationship to ALS-PDC is unclear; however, formaldehyde is a major component of tobacco smoke that has been strongly linked to sporadic amyotrophic lateral sclerosis (96). Spencer and colleagues proposed that altered expression of short, noncoding segments of RNA, micro-RNA in the setting of DMA damage within nondividing cells such as neurons, is a potential mechanism of both MAM and formaldehyde-induced neurotoxicity (82). Cycasin occurs in high concentrations in cycad plants and is usually removed during the elaborate soaking process used to prepare cycad seeds for consumption; residual content of cycasin in cycad flour did appear to correlate with risk of ALS-PDC in Chamorros (102; 38; 73). The retinal and cerebellar pathology identified in ALS-PDC may also be related to in utero exposure to cycasin, based on animal studies revealing similar pathology (84).

β-Methylamino-l-alanine (BMAA) is acutely neurotoxic via multiple mechanisms: binding to NMDA receptors, binding of glutamate receptor 5, and induction of oxidative stress. Nitrosated BMAA has also been shown to be an alkylating agent, creating another hypothesis for its mechanism of neurotoxicity (71). The role of BMAA in ALS-PDC was initially refuted because of the minute concentrations of measurable free BMAA that remained in prepared cycad flour, and animal models that demonstrated that very large quantities of BMAA were required to produce acute toxicity (88; 18). However, the BMAA hypothesis was revived after Cox and Sacks postulated that protein-bound BMAA could be biomagnified through the food chain; consequently, exposure to larger quantities of BMAA might occur through consumption of flying foxes and other animals that feed on cycads (15; 05). Flying foxes were a delicacy consumed more commonly by Chamorro males; WWII brought money and guns to Guam, which facilitated the capture and consumption of the foxes to the point of extinction by the mid-1970s. Concurrent extinction of the foxes and of the cycad plants by a scale insect correlated with the decline in the prevalence of neurodegenerative disease on Guam (13). Secondly, a mechanism for subacute and chronic toxicity of BMAA may exist via protein synthesis. L-BMAA can be misincorporated into proteins during protein synthesis by being erroneously substituted for L-serine, which could ultimately trigger protein misfolding, accumulation of protein aggregates, and apoptosis (21). It is important to note that during the Japanese occupation of Guam in WWII, the Chamorro natives were malnourished and mistreated (03); preexisting malnutrition and protein deficiency could result in the misincorporation of BMAA at a higher frequency, producing higher rates of disease or earlier onset of disease as was the case in the 1940s and 1950s. This mechanism occurs in the neurologic disease lathyrism, where a similar nonprotein amino acid (BOAA) acquired though chickpea consumption becomes neurotoxic in the setting of malnutrition (85). Lastly, several neuropathological studies have supported the link between BMAA and ALS-PDC (60). Spencer and colleagues described pathological changes in macaques dosed with BMAA (86). Cox and colleagues reported inducing neurofibrillary tangles comprised of tau in the brains of vervets who had been fed BMAA-injected fruit (14). Protein-bound BMAA has been detected in the brain tissue of Chamorro patients who died of ALS-PDC.

The BMAA hypothesis may not be unique to Guam. Studies have shown that in addition to the ALS-PDC cases, BMAA has been detected in the brain tissue in patients from North America with sporadic Alzheimer disease and amyotrophic lateral sclerosis but not in healthy controls (60; 53; 66). Diverse taxa of ubiquitous cyanobacteria capable of producing BMAA are found in eutrophic freshwater and marine water bodies and are an increasing environmental hazard in several parts of the world (55). Consequently, there is potential for BMAA to accumulate in aquatic food chains, resulting in human exposure outside of Guam (07; 10).

However, BMAA as the cause for ALS-PDC has been disputed. Chernoff and colleagues have detailed the gaps in the data and argued against the hypothesis (12). Among the stated arguments include questions regarding the relationship between neurodegeneration and BMAA found in human brain tissue, techniques used to quantify BMAA in tissues or other substances, the process of BMAA dosing or administration in existing animal models, and the evidence for the protein-misincorporation mechanism of BMAA neurotoxicity. They also provided concerns regarding the flying fox theory put forth by Cox and Sacks, including the methods used to estimate the flying fox population numbers that were then used as a correlation with the decline in ALS-PDC (12). In addition, Cox and Sacks measured tissue concentrations of BMAA in dehydrated museum specimens of flying foxes, using techniques that could lead to false positive results (12; 24). In 2021, Dunlop and colleagues published a rebuttal to Chernoff’s 2017 paper in support of the role of BMAA in neurodegeneration (19).

Other environmental causes of ALS-PDC have been considered. All three Western Pacific regions also shared a unique volcanic mineral environment characterized by severely low levels of calcium (Ca2+) and magnesium (Mg2+), coupled with high levels of bioavailable transition metals in the soil and drinking water (49; 100). Aluminum and selenium are also present in washed cycad seed flour, particularly when the plants are grown in the shade (52). It has been hypothesized that CA2+ and MG2+ deficiency and an excess of aluminum, manganese, and iron might lead to the deposition of trace elements in the CNS. Eventually, this could cause neuronal degeneration through oxidative stress (100; 101). High concentrations of aluminum have indeed been identified within neurofibrillary tangles in ALS-PDC (67). The above findings could also be linked to specific mechanisms of gene-environment interactions, including through SOD1 and TRPM7 (30; 101).

A purely genetic origin has been disputed for a number of reasons: (1) all three Western Pacific populations with ALS-PDC were genetically distinct but had consistent clinical and neuropathological findings; (2) a paucity of disease in other Mariana islands and nearby Filipino populations of similar genetic ancestry; (3) a shift in disease expression from amyotrophic lateral sclerosis–predominant to parkinsonism-dementia complex–predominant; and (4) mathematical constructs that refuted a purely genetic basis (70; 69; 04). However, there have been genetic associations identified. Linkage studies in Guam have found significant evidence for loci on chromosome 12 and 17, suggesting that one or more of these loci may confer genetic susceptibility to ALS-PDC (80) or from mitochondrial DNA mutations (75). Genetic sequencing among Chamorros identified PINK1 p.L347P, heterozygous DCTN1 p.T54I, FUS p.P431L, and a CAG repeat within HTT as pathogenic mutations linked to neurodegenerative disease (91). In the Kii peninsula, about 20% of amyotrophic lateral sclerosis patients carry a C9orf72 expansion, a genetic mutation responsible for some genetic forms of amyotrophic lateral sclerosis and frontotemporal lobar dementia (36); however, this does not explain all cases of ALS-PDC in Kii or elsewhere (17).

Pathological findings in ALS-PDC have been well described. The patterns of anterior horn cell loss and chromatolysis in Western Pacific ALS-PDC are similar to the spinal cord changes of patients with classic amyotrophic lateral sclerosis (31). For example, the transactive response (TAR) DNA binding protein 43 (TDP-43) is a major component of ubiquitinated inclusions in ALS-PDC as well as sporadic amyotrophic lateral sclerosis and frontotemporal lobar dementia (FTD) (02; 62). However, the presence of neurofibrillary tangles comprised of pathologic tau isoforms in the brain and spinal cord of ALS-PDC patients separates this disorder from classic amyotrophic lateral sclerosis (31; 37; 64; 93; 57). The distribution of neurofibrillary degeneration in subcortical structures and brainstem are similar to progressive supranuclear palsy, though the tau subtypes differ (90). Hippocampal sclerosis has been reported (65). Histologically, the aberrantly phosphorylated tau in Alzheimer disease and in ALS-PDC are structurally similar but vary in location within the cortical layers (35; 08). Yamazaki and colleagues have identified novel hyperphosphorylated tau-positive structures found almost exclusively in ALS-PDC brains but not amyotrophic lateral sclerosis, Alzheimer disease, or Parkinson brains, possibly representing a neuropathologic marker of this particular tauopathy (99). Similar to chronic traumatic encephalopathy, thorn-shaped astrocytes in the depth of cortical sulci have been described, though the study lacked information regarding previous head trauma in the patients (43).

Other ultrastructural inclusion bodies seen in both Alzheimer disease and ALS-PDC include alpha-synuclein inclusions (Lewy bodies) and Hirano bodies. Lewy bodies are seen throughout ALS-PDC brains, particularly in the substantia nigra and the spinal cord. Hirano bodies, eosinophilic rod-shaped structures found in a variety of neurodegenerative diseases, are found in large numbers near pyramidal neurons in the hippocampus and are less frequent in Alzheimer disease and frontotemporal lobar dementia (34; 49). In contrast, plaques comprised of β-amyloid are less frequent in ALS-PDC (28; 95) and are similar in content to cognitively intact normal individuals (77). Clinical phenotype may depend on the predominant neuropathology; in a group of 18 Kii peninsula patients with ALS-PDC, all patients had tau pathology, but subgroups with TDP-43 dominant pathology clinically had amyotrophic lateral sclerosis, and those with synuclein-dominant pathology had parkinsonism dementia complex. Another subgroup identified as tau-dominant had atypical parkinsonism and dementia, some of which also had motor neuron disease (58).

Imaging studies support pathological findings. Volumetric MRI has demonstrated that hippocampal formation volumes are significantly reduced in ALS-PDC and correlated clinically with memory function and neuropathologically with the site of intense neurofibrillary tangle deposition (68). Striatal 18F-6-fluorodopa uptake, a marker for dopaminergic pathology, was significantly reduced in all the patients with ALS-PDC and was reduced in the patients with amyotrophic lateral sclerosis to a lesser degree (81). PET imaging utilizing [11C]PBB3 as a tau-binding ligand has been presented in a series of Kii peninsula ALS-PDC patients (78). The finding of an obvious decrease in cerebral blood flow of the frontal and temporal lobes in patients with amyotrophic lateral sclerosis and parkinsonism-dementia complex with or without cerebral atrophy supports the concept that the two conditions are different phenotypes of a single frontotemporal tauopathy (41).

Epidemiology

The incidence of ALS-PDC in Guam was highest in 1954 to 1955, when the prevalence of this disorder reached 420 per 100,000 population (03; 44; 74; 49). During this time, it is estimated that ALS-PDC resulted in one fourth of Chamorro deaths. The incidence of ALS-PDC declined over the subsequent five decades to seven per 100,000 Guamanians for amyotrophic lateral sclerosis in 1989, whereas parkinsonism-dementia complex remained somewhat higher at 22 per 100,000 (97). This suggests that the cause of the disease was environmental, possibly related to the westernization of Guam and subsequent ethnographic changes (76).

Epidemiologically, Kii ALS-PDC differs from Guamanian ALS-PDC in its continuing high incidence and prevalence rates of amyotrophic lateral sclerosis in the Kii area after the 1990s, in contrast to the marked decline in high incidence rates in Guam (79; 68; 40). In addition, the high rates of familial occurrence suggest the existence of some genetic abnormalities in Kii ALS-PDC, including the high prevalence of the C9orf72 expansion (36; 61). In West Papua, rates of ALS-PDC have declined but continue to remain elevated compared to global amyotrophic lateral sclerosis rates, with an estimated crude prevalence rate of amyotrophic lateral sclerosis in 2007 of 73 to 133 per 100,000 population, and that of ALS-PDC to be 53 to 98 per 100,000 (63).

Patients with predominant amyotrophic lateral sclerosis are affected earlier in life than those with parkinsonism-dementia complex, and both conditions develop insidiously, with death usually occurring within 10 years of diagnosis. Men are affected earlier and more frequently in both disorders, and the male to female ratio is approximately 2:1 (49). The age of onset of ALS-PDC ranged from 20 to 72 years, and during the interval from 1950 to 1979, the mean age of onset had increased by 4.5 years in men and 3.4 years in women (76). Similar to other industrialized nations, the frequency of women affected did increase (from 38.2% to 44.0% over this 30-year study) with no significant changes in the constellation of symptoms with respect to gender or to the decade of presentation (76).

Prevention

At this time, no method of prevention exists. If specific toxins are identified as the definitive cause of ALS-PDC, then potentially avoiding sources of these toxins would potentially prevent ALS-PDC cases.

Based on the interest of BMAA as a potential cause of ALS-PDC and neurodegenerative disease, there has been research into methods of prevention of this toxicity. Dunlop and colleagues demonstrated that L-BMAA can substitute for L-Serine during in vitro protein synthesis using human cell lines and that this misincorporation can be inhibited by the presence of L-serine (21). Building on these findings, Cox and colleagues found that vervets (Chlorocebus sabaeus) given oral BMAA for 140 days developed neurofibrillary tangles and sparse beta amyloid deposits in the cerebral cortex, but the density of neurofibrillary tangles was significantly attenuated by coadministration of equal amounts of L- serine (14). Similarly, BMAA-fed vervets have also been found to develop spinal cord pathology with anterior horn cell degeneration and demyelination of corticospinal tracts; coadministration of L-serine was again found to attenuate these neuropathological changes (16). Cai and colleagues demonstrated that angiopoietin‐1 and C16 peptide reduced autophagy and apoptosis and improved cognition, motor function, and electrophysiologic measures in a BMAA fed rat model of ALS-PDC; these benefits were amplified by the coadministration of L-serine (09).

Though initial interest in L-serine as a potential therapeutic agent centered on competitive inhibition of BMAA substitution during protein synthesis, additional research has called into question whether this is the primary mechanism underlying L-serine’s putative neuroprotective properties. Evidence against the amino acid substitution hypothesis includes the following: (1) the ratio of protein-bound to total BMAA found in brain tissues in animal models has been found to be unaffected by coadministration of L-serine (14); and (2) coadministration of L-alanine has no protective effects on BMAA-induced neurotoxicity (van Onselen eta al 2020). Other potential mechanisms that have been proposed involve the influence of L-serine on antioxidant and proteostatic pathways (20; 72).

There has been some speculation that BMAA exposure may increase the risk of sporadic amyotrophic lateral sclerosis, and a phase I study of L-serine in amyotrophic lateral sclerosis reported a dose-dependent reduction in slope of ALSFRS-R scores when comparing to historical controls (48; 06). L-serine is also being investigated as a potential therapeutic agent in Alzheimer disease (56).

Differential diagnosis

Confusing conditions

Differential diagnosis of ALS-PDC consists of other primary and secondary forms of parkinsonism, dementia, and amyotrophic lateral sclerosis (ALS). Notably, clinical or subclinical parkinsonism in amyotrophic lateral sclerosis has been well described. Although rare, there has also been recognition of other neurodegenerative “overlap” syndromes with features of motor neuron disease, parkinsonism, and dementia (83). These have been identified in geographical or familial clusters. Farnikova and colleagues reported the coexistence of parkinsonism, dementia, and an upper and lower motor neuron syndrome typical of amyotrophic lateral sclerosis, confirmed by EMG findings, in four patients from the Czech Republic (23). A family with "disinhibition-dementia-parkinsonism-amyotrophy complex" linked to chromosome 17 has also been described (51). A cluster of ALS-parkinsonism was discovered on the Caribbean Island of Guadeloupe. Lannuzel and colleagues reported a 4.5-fold increased risk of amyotrophic lateral sclerosis, and ALS-PDC occurred in 23.8% of cases (45). It is unclear if there is any connection, genetic or environmental, with the known clusters in the Western Pacific.

Diagnostic workup

As with most neurodegenerative disorders, the diagnosis of ALS-PDC is purely based on clinical presentation, with imaging and laboratory evaluations performed to exclude other causes. EMG, however, would be essential to confirm the presence of motor neuron disease.

Management

Management of ALS-PDC consists primarily of supportive care.

Outcomes

Survival rates vary in relation to the decade of onset: 15% of men and 29% of women who developed symptoms in the 1950s lived beyond 10 years, whereas only 3% of men and 13% of women who developed symptoms in the 1970s lived beyond 10 years. Death is most commonly associated with pulmonary or urinary tract infection (76).