Current strategies in the management of dementia with lewy bodies and future directions based on disease pathophysiology

Abstract

Dementia with Lewy bodies (DLB) is thought to be the second most common form of dementia after Alzheimer’s disease, and is characterised by a combination of cognitive, neuropsychiatric and motor symptoms. The present review seeks to discuss current strategies for the management of DLB, and future directions for novel disease-modifying therapies. Current best practice for the clinical management of DLB is based upon therapies that target specific symptom domains due to the lack of disease-modifying therapies. Cholinesterase inhibitors are the frontline treatment for treating cognitive decline in DLB, whereas the treatment of motor symptoms remains challenging due to poor response to dopaminergic therapies and the potential for exacerbation of neuropsychiatric features. There is emerging evidence suggesting a range of further pharmacological and non-pharmacological therapies may be effective in treating specific symptom domains of DLB, but further evidence is warranted to demonstrate their efficacy. A key challenge in the treatment of DLB is incomplete understanding of disease pathophysiology, which has limited attempts to develop disease-modifying therapies. In the present article, we discuss the multi-faceted nature of DLB neuropathology, from Lewy body pathology to mitochondrial dysfunction, and discuss therapies in development that target particular aspects of DLB neuropathology. In particular, we highlight antibody-based therapies to attenuate protein aggregation, compounds that enhance the generation of cellular energy and autophagy-enhancing agents as particular areas of promise. Furthermore, we discuss how optimal strategies for disease modification will be centred on agents that treat DLB neuropathology more holistically, and will be underpinned by a more complete understanding of the pathogenic events that underlie the full spectrum of pathological changes observed in the DLB brain.

Background

Dementia with Lewy bodies (DLB) is a significant and important cause of late life neurodegenerative dementia. It is neuropathologically characterised by the accumulation of the protein α-synuclein into spherical intraneuronal deposits termed Lewy bodies [1] and sits under the umbrella term of Lewy body disease which also includes individuals with Parkinson’s disease (PD) and Parkinson’s disease dementia (PDD). DLB and PDD overlap patho-aetiologically and from a treatment perspective; as clinical syndromes they are currently differentiated arbitrarily on the basis of the temporal onset of cognitive symptoms: PDD is a neurocognitive disorder that occurs in the context of established PD whereas DLB presents with cognitive impairment and the parkinsonism occurs contemporaneously or subsequent to cognitive impairment [2]. In the present review we focus primarily on DLB but take an agnostic perspective on either lumping or splitting these clinical syndromes instead including discussion where appropriate on both conditions where there may be a common aetiological or therapeutic perspective. This approach is also salient given recent efforts to accommodate both PD and DLB under a common biological framework which we discuss further below.

Although clinical estimates are thought to under-estimate the prevalence of DLB, neuropathological studies typically estimate the prevalence of DLB to be approximately 15–20% of all dementia cases and recent data using synuclein amplification assays (SAA) have suggested that abnormal α-synuclein is present in up to 23% of memory clinic patients with cognitive impairment [3, 4].

The clinical features of DLB are not the focus of the present review as they have been covered extensively in international consensus guidelines [2]. Briefly, DLB is a heterogeneous disorder associated with cognitive, motor and neuropsychiatric features. The cognitive profile of DLB typically implicates attention, executive function and visuospatial abilities, and episodic variations in attention and awareness over minutes or hours, a phenomenon known as cognitive fluctuations, are a core feature of DLB [2]. Complex visual hallucinations consisting of well-formed figures, such as animals or people, are another core feature of DLB, present in up to 80% of patients and are an especially distinctive feature of DLB relative to other forms of dementia [2, 5]. Motor features reminiscent of PD, including bradykinesia, rigidity and tremor, are also a core feature of DLB though they tend to be more symmetrical than in PD and respond less well to dopaminergic therapies [6, 7]. REM sleep behaviour disorder, the acting out of dreams during REM sleep is a further core feature of DLB and is thought to precede the onset of DLB by several years [8]. Other features associated with DLB include autonomic dysfunction and neuropsychiatric symptoms such as depression, anxiety and apathy.

The present article aims to summarise the current state of the art regarding the diagnosis and clinical management of DLB, in addition to emerging evidence regarding the pathophysiology of DLB and how these may inform the therapeutic management strategies of the future.

Main body

Diagnosis of DLB

International consensus diagnostic criteria take a predominantly clinical approach to the diagnosis of DLB and considers four core clinical features which strongly support a diagnosis of DLB along with a wider range of supportive clinical features which increase the likelihood of the diagnosis of DLB, although in themselves are not as specific (Table 1). As diagnostic criteria have iteratively developed, there has been increasing clarity on the inclusion of biomarkers and the most recent version of the criteria include both indicative (strongly favouring a diagnosis of DLB) and supportive (helping support a diagnosis of DLB but have less diagnostic specificity) biomarkers [2].

Table 1.

Core and supportive clinical features of DLB [2]

Core clinical features	Description
Fluctuating cognition	Variations in attention and awareness
Visual hallucinations	Recurrent and well formed visual hallucinations
REM sleep behaviour disorder	Acting out dreams
Parkinsonism	One or more of the following: bradykinesia, rest tremor or rigidity
Supportive clinical features	Description
Neuroleptic sensitivity	Severe sensitivity to antipsychotics, particularly neuroleptics
Autonomic dysfunction	Constipation, orthostatic hypotension, urinary incontinence
Additional features	Falls, syncope, hypersomnia, hyposmia, hallucinations in other modalities, delusions, apathy, anxiety, depression

Refinements to the clinical diagnostic criteria have increased sensitivity and specificity and there is also research criteria for prodromal DLB [8]; however, in clinical practice there is considerable under-diagnosis and under-recognition of DLB [9]. Nevertheless, although there is considerable clinical under-recognition of DLB, the implementation of the consensus diagnostic toolkit has been demonstrated to increase diagnostic rates of DLB but there continues to be a significant mismatch between the expected prevalence of DLB and the proportion of dementia cases receiving such a diagnosis [10]. Therefore, there continues to be a pressing unmet need to develop diagnostic biomarkers that enable measurement of the underlying pathology of DLB to improve clinical recognition of DLB.

SAA represent a potential paradigm shift in how we diagnose synucleinopathies, including DLB. SAA can identify α-synuclein pathology using bodily fluids, based on the capacity of disease-associated α-synuclein to induce the misfolding and aggregation of the native protein in significant abundance to be detectable using fluorescent dyes such as thioflavin [11]. The ability to detect disease-associated α-synuclein pathology in bodily fluids using SAA creates the possibility to detect neuropathology intra vitam, overcoming the considerable limitation of neuropathological assessment being the only method to definitively confirm the presence of cerebral α-synucleinopathy. Furthermore, SAA provides a more proximal diagnostic biomarker of the key pathology of abnormal synuclein in contrast to DaT scan or MIBG imaging which are indirect biomarkers of the neurodegenerative impact of synuclein disease and which may be moderated by other pathologic processes. It is notable that recent data using SAAs have suggested that abnormal α-synuclein is present in up to 23% of memory clinic patients with cognitive impairment which is aligned with neuropathological estimates of the prevalence of DLB to be approximately 15–20% of all dementia cases [3, 4].

The effectiveness of SAA to detect α-synucleinopathy, coupled with the central role ascribed to α-synuclein pathology in DLB, has led to the recent development of biological frameworks to define Lewy body diseases anchored primarily on the presence of seed-competent α-synuclein on SAA [12, 13]. Aside from the fact that such frameworks typically subsume DLB into a wider identity centred around PD, the ability of SAA to identify distinct disease states is predicated upon seed-competent α-synuclein pathology being relatively specific for primary Lewy body diseases; however, Lewy body pathology is a common feature of other conditions, particularly Alzheimer’s disease, where it occurs in over 50% of cases [14]. Our studies in post-mortem tissue of paediatric lysosomal storage disorders cases have identified seed-competent α-synuclein pathology that gives a positive response on SAA, despite these disorders resulting from the accumulation of lipids rather than proteins [15–17]. Furthermore, recent data from a large cohort of clinically diagnosed DLB cases (n = 191) found that CSF α-synuclein SAA was not evident in 28.3% of the sample [18]. Further work clearly is needed to clarify the meaning of these findings whether this is due to loss of sensitivity from an analytic perspective, or is something that relates to the intrinsic sensitivity of SAA to detect brain alpha-synucleinopathy when present in lower abundance or localised to particular brain regions [19]. Clinical misdiagnosis in cohorts of DLB is another possibility and it may be that SAA is more sensitive for the detection of particular subgroups of DLB with more evident parkinsonism, REM sleep symptoms and worse hyposmia [18]. Therefore, whilst SAA is likely to be an important tool for the diagnosis of DLB in future, further validation will be needed especially against neuropathologically confirmed cohorts and it seems likely that, practically, its use will be in combination with existing clinical assessment toolkits and supportive biomarkers.

Management of DLB

The evidence base for the management of DLB has been covered elsewhere [20]. DLB management is primarily symptomatic, in part, to the current lack of disease-modifying therapies. Pharmacological treatment for DLB typically is primarily focused on the management of cognitive and neuropsychiatric features with acetylcholinesterase inhibitors, such as donepezil, rivastigmine and galantamine. This class of drugs are based on ameliorating cholinergic deficits in DLB patients, which are thought to be the result of degeneration of the primary cholinergic nuclei in the brain of DLB patients, such as the nucleus basalis of Meynert and the pedunculopontine nucleus [21–23]. Acetylcholinesterase inhibitors have generally been demonstrated to give significant, albeit modest, benefits in improving cognitive function and reducing hallucination severity in DLB patients; these effects may be sustained and associate with increased survival [24–26].

The NMDA receptor antagonist, memantine, has also been trialled in DLB. The prevailing mechanistic view has been that NMDA antagonism for DLB is predicated on attenuating excitotoxicity resulting from abnormalities in glutamatergic neurotransmission that leads to a state of excitotoxicity. Other secondary mechanisms of action have been proposed including, for example, cholinergic and dopaminergic modulation, as well as synaptic NMDA actions although how these translate into clinically meaningful impacts on symptoms still remains unclear [27]. Memantine has been demonstrated to be well-tolerated but with only a modest impact on cognition and behaviour and evidence for its efficacy is mixed [28–30].

Neuropsychiatric features, including visual hallucinations, apathy, and delusions, are especially pronounced in DLB compared to other neurodegenerative disorders, and have been demonstrated to have a particularly marked effect on patient quality of life and caregiver burden [31]. Management of neuropsychiatric features is typically targeted and weighed against the potential for adverse events; anti-psychotic medications can induce a severe and even fatal reaction in DLB patients and are thus typically only used with caution and as a second line treatment [20]. Quetiapine is often used given it has the least side effects although evidence for its efficacy is lacking [32]. Clozapine has an evidence base in PD psychosis but has not been trialled in DLB and may come with unacceptable side effects in older frail patients. It is important to emphasise that some antipsychotics should generally be avoided in DLB patients, such as risperidone, olanzapine and aripiprazole, due to risk of deterioration, including increased parkinsonism [33]. Aripiprazole may help neuropsychiatric symptoms but evidence is based on case series also carries attendant risks similar to other antipsychotics for DLB [34, 35].

Pimavanserin, a serotonin inverse agonist has been shown to be more effective and better tolerated than quetiapine in the treatment of hallucinations, delusions and paranoia but is not widely available outside the United States [36, 37]. Benzodiazepines are best avoided in the management of DLB due to their exacerbation of functional decline although clonazepam is sometimes used to manage significant REM sleep behaviour symptoms [38]. Melatonin is sometimes recommended first line for REM sleep behaviour symptoms given its better side effect profile compared to clonazepam; however there is no trial evidence to support its efficacy in DLB [39].

Daytime drowsiness is common in DLB. First line, good sleep hygiene is advocated. Psychostimulants like armodafinil have been trialled in small studies and might help but carry potential side effects and more research is needed to recommend them. Other issues, such as sleep apnoea, may respond to continuous positive airways pressure treatment and should be assessed for by specialist sleep services.

Although antidepressants have been used in the clinical management of DLB relatively few studies have investigated their efficacy, though existing evidence from a systematic review indicates the serotonin selective reuptake inhibitor (SSRI), citalopram, may lead to deterioration of neuropsychiatric features whilst an alternative SSRI, paroxetine, has shown some benefits in case studies [40]. However, given that paroxetine has significant anti-cholinergic activity, caution should be exercised in its use in DLB [41].

The management of motor symptoms in DLB can be challenging, primarily due to DLB patients responding less consistently and effectively to dopaminergic therapies than PD cases, combined with the risk of exacerbating neuropsychiatric features, in particular, hallucinations [42, 43]. Therefore, when dopaminergic therapies are used, they should be at the lowest effective dose and combined with careful monitoring of the patients for deterioration of neuropsychiatric symptoms. Levodopa monotherapy is generally preferred over dopamine agonists due to the view that dopamine agonists may exacerbate neuropsychiatric features of DLB [44]. More recently the anti-convulsant drug zonisamide has been recommended as an adjunct treatment in DLB. This drug has complex pharmacodynamics that includes dopamine agonism, has, in double blind RCTs, demonstrated efficacy in improving parkinsonism in DLB without significant worsening of neuropsychiatric features [45, 46].

Dysautonomia, including but not limited to, orthostatic hypotension, urinary urgency, constipation, etc. has significant impacts on the quality of life of people living with DLB [47]. Unfortunately, there is a paucity of research and therapeutic trials in DLB in this key symptom domain and evidence for the management of these symptoms typically draws on work in PD; coverage of this is beyond the scope of the current review although we would draw the reader’s attention to other reviews and guidelines provide advice on how to manage autonomic symptoms in DLB [39].

The past decade has heralded an increase in studies evaluating the efficacy of non-pharmacological interventions in DLB. Although evidence continues to be collected as to the efficacy of such interventions, most trials to date have been performed in individuals with PDD, where interventions such as cognitive rehabilitation and physical therapy have led to improvements in cognitive and motor symptoms, respectively [48]. Studies in DLB, whilst frequently reporting benefits are often small in sample size and low quality which makes it difficult to make specific non-pharmacological recommendations [49].

Electroceuticals have been advocated as another potential therapeutic route [50]. There have been tentative open label findings of improvements with transcranial direct current stimulation (tDCS) in attention and visuoperceptual function and a small case study (n = 6) suggesting transcranial magnetic stimulation may ameliorate depression [51]. However, two randomised sham-controlled trials of transcranial direct current stimulation in DLB and PDD patients did not demonstrate any significant improvements in cognition or hallucinations [52, 53]. Electroconvulsive therapy may help with depression and psychosis in DLB although data is limited to case series [54, 55]. Deep brain stimulation of cholinergic nuclei has also been evaluated in DLB with no consistent improvements in cognitive measures although 3/5 patients treated have a reduction in neuropsychiatric symptoms [56]. Further work is needed to evaluate this approach, however.

Overall, the evidence base for symptomatic management in DLB is piecemeal with the strongest evidence for cholinesterase inhibitors and decreasing levels of evidence for other pharmacological and non-pharmacological approaches. This relative lack of evidence should, however, not abrogate a focus on treating symptoms in DLB; holistic care approaches which take a multi-symptom perspective, for example the Diamond-Lewy management toolkit appear to have demonstrable benefits for patients and their care-givers in global outcomes and in reducing care-giver burden and depression [10]. Critically, going forward, however, is the urgent need to develop drugs which modify the underlying disease process and thus the disease course and trajectory. In the next section, we will discuss the pathophysiology of DLB and expand upon how further understanding of disease aetiology may foster a new era of disease-modifying therapies.

The pathophysiology of DLB: α-synuclein

The underlying cause of DLB is not known; however, the accumulation of the protein α-synuclein is thought to play a central role [1]. α-synuclein has been implicated in Lewy body diseases as duplications/triplications in its gene, SNCA, are associated with autosomal dominant PD, variants in SNCA increase risk of idiopathic and familial PD, drugs that repress α-synuclein transcription are associated with reduced risk of PD and α-synuclein is a major component of Lewy bodies [57–60]. Lewy bodies, large spherical deposits immunoreactive for α-synuclein, are the characteristic neuropathological feature associated with DLB. Widespread Lewy body pathology is necessary for a neuropathological diagnosis of DLB, which has underpinned the assumption that they play a central role in disease pathogenesis [1, 2].

Despite the focus on Lewy bodies in DLB, even regions with substantial cell loss, such as the substantia nigra, do not evidence associations between the abundance of Lewy bodies and the degree of neuronal loss or phenotypic severity [61, 62]. Furthermore, we have previously reported neuronal loss in regions of the thalamus with no Lewy body pathology, in contrast to preserved neuronal populations in adjacent regions with Lewy bodies, and others have reported neuronal loss and/or molecular alterations in neurons without Lewy bodies, suggesting a poor relationship between regions that manifest Lewy body pathology and those with neuronal loss [63–65]. The original staging scheme for Lewy body pathology, devised by Braak and colleagues, assumed that neocortical Lewy bodies would associate with clinical dementia; however, this work was performed without clinical data and subsequent studies have demonstrated significant numbers of PD cases with neocortical Lewy bodies but no evidence of clinical dementia, and vice versa [66]. Taken together, despite continued focus on Lewy bodies as central drivers of DLB and other Lewy body diseases, there is little neuropathological or molecular evidence to support an association with neuronal loss or phenotypic severity, in Lewy body diseases.

It is important to note that poor associations between Lewy bodies and meaningful clinical or neuropathological variables does not exclude the potential for α-synuclein to be a central driver of DLB. It has been suggested that a poorly-defined group of multimers that may precede the formation of higher-order α-synuclein aggregates, typically termed “oligomers”, could be the primary toxic agent in DLB and related Lewy body diseases [1]. Novel methods, such as α-synuclein proximity ligation assay, have been employed to detect oligomeric α-synuclein, and typically identify pathology not observed with standard α-synuclein immunohistochemistry [67]. There is limited evidence that α-synuclein oligomers may be better correlates of cognitive dysfunction in DLB than Lewy body pathology; however, this is an under-studied area and further studies are required to better characterise and define oligomers, and evaluate their role in DLB [68].

α-synuclein is extensively post-translationally modified via phosphorylation and truncation and, in DLB, there is a particular increase in the abundance of α-synuclein phosphorylated at serine 129 (pS129) [69]. The precise role of pS129 is not clear, with studies ranging from those demonstrating a physiological role in regulating excitation/inhibition in the neuronal pre-synaptic terminal to a potentially protective role by inhibiting aggregation [70, 71]. Although a number of other post-translational modifications beyond pS129 have been described for α-synuclein, some of which appear to regulate α-synuclein aggregation in vitro, the extent to which these occur in individuals with DLB and thus play a role in disease remains unclear [72].

Genome-wide association studies indicate SNCA variants are associated with risk of DLB, and broadly suggest overlap with PD (GBA1 and TMEM175) and Alzheimer’s disease risk genes (APOE and BIN1) [73]. The overlap with Alzheimer’s disease and PD corresponds well with the neuropathology of DLB, where significant levels of concomitant Alzheimer-type pathology are a frequent observation [74, 75]. The levels of concomitant Alzheimer-type pathology vary widely across DLB cases, typically lying at an intermediate stage based on consensus criteria for the neuropathological reporting of Alzheimer’s disease; however, up to 30% of DLB cases have a high level of Alzheimer-type neuropathological change, a level sufficient for a neuropathological diagnosis of Alzheimer’s disease [74, 76, 77]. The presence of Alzheimer-type pathology in DLB associates with a more amnestic cognitive profile, more rapid disease progression and higher mortality, suggesting it is not an epiphenomenon but a key contributor to phenotypic severity [78]. Therefore, although Lewy body pathology defines DLB, Alzheimer-type pathology is a common concomitant feature that is associated with a worse prognosis.

In summary, despite the important role ascribed to Lewy bodies in DLB, there is relatively little evidence that they are associated with the neurodegeneration and impairment that is thought to underlie clinical manifestation and progression, perhaps suggesting oligomers or other α-synuclein conformers could be more centrally involved in driving DLB. However, there is clear evidence that Alzheimer-type pathology is a common feature of DLB, where it is associated with accelerated decline and early mortality.

The pathophysiology of DLB: beyond proteopathy

Although there is considerable focus on the role of α-synuclein in DLB, a number of other pathological changes also occur in DLB that may contribute to the symptoms experienced by patients. Cells have a number of mechanisms to prevent the accumulation of misfolded proteins such as α-synuclein, including degradation via the ubiquitin-proteasome system or the autophagy-lysosome system [79]. However, when the accumulation of misfolded proteins cannot be controlled via these processes, misfolded proteins are trafficked through the microtubule network to the perinuclear microtubule organising complex (MOC) and formed into an aggresome, a perinuclear structure to facilitate the containment and degradation of misfolded proteins via the chaperones and proteases present within the MOC [80]. A key outstanding question is why these multiple lines of defence against misfolded proteins go awry in Lewy body diseases, and whether this is indicative of more significant and widespread cellular impairment than the misfolding of α-synuclein in isolation.

Mitochondrial dysfunction has long been implicated in Lewy body diseases, primarily based on evidence from the PD literature demonstrating pathogenic variants in mitochondrial genes, particularly those involved in mitophagy, underlie familial forms of PD [81]. There is much less literature reporting mitochondrial dysfunctional in DLB specifically, though transcriptomics studies have indicated mitochondrial dysfunction and oxidative stress in post-mortem DLB brain tissue [82]. Furthermore, we and others have reported reduced expression of components of complex I of the mitochondrial respiratory chain in neurons of the nucleus basalis of Meynert and substantia nigra in DLB [22, 83]. A key question is whether mitochondrial dysfunction is a contributor to DLB or the consequence of another insult, such as α-synucleinopathy. We have previously reported Lewy body pathology is common in the brains of older individuals with primary mitochondrial diseases, suggesting that mitochondrial dysfunction and age predispose towards the development of Lewy body pathology [84]. However, our previous neuropathological observations in individuals with primary mitochondrial diseases demonstrates the striking vulnerability of regions that are resilient to neurodegeneration and Lewy body pathology in DLB, such as the cerebellum and primary visual cortex, perhaps arguing against a primary mitochondrial aetiology for DLB [65, 85, 86]. Oligomeric α-synuclein has been reported to impair complex I-dependent respiration and alter mitochondrial membrane permeability in vitro, suggesting α-synuclein pathology directly impairs mitochondrial function [87, 88]. Taken together, mitochondrial dysfunction is associated with DLB though it remains unclear if it is a cause or consequence of the primary disease process.

Inflammation has also been reported to occur in DLB, primarily due to observations in model systems [89]. Studies in patient-derived plasma suggest an increase in inflammatory cytokines in DLB, particularly early-stage cases, though the extent to which peripheral plasma is a proxy of cerebral changes is debatable, and post-mortem evaluation of DLB brain tissue does not typically indicate evidence of inflammation [90, 91]. It is tempting for one to speculate that DLB may be marked by an early inflammatory reaction that subsequently subsides over time, perhaps due to an infection precipitating disease onset. Previous studies have identified that α-synuclein has anti-microbial properties and is significantly up-regulated in conditions of cerebral infection, perhaps suggesting a mechanism underlying early inflammation in DLB that subsides over time [92–94]. Therefore, inflammation continues to be a topical area of research and is frequently posited as a therapeutic target for DLB [91].

The autophagy-lysosomal pathway has been implicated in Lewy body diseases more broadly, primarily as a number of genes encoding proteins involved in the autophagy-lysosomal pathway are associated with risk of PD and DLB [95]. DLB cases manifest alterations in autophagy markers in plasma from DLB cases, and the accumulation of autophagosomes and autophagic mitochondria in DLB neurons, consistent with deficient autophagy in DLB [96, 97]. Studies characterising the ultra-structure of Lewy bodies report they contain apparently damaged organelles which, combined with the observation that Lewy bodies are immunoreactive for markers of the regulated encapsulation of misfolded proteins, aggresomes, is consistent with the sequestration of waste into Lewy bodies as a response to deficient autophagic clearance of cargo [98, 99]. Many risk genes for DLB are specifically involved in the catabolism of a class of lipids termed sphingolipids, the best known of which is GBA1, which encodes the lysosomal glucocerebrosidase [73, 100]. We have also previously reported α-synuclein with putative pathological properties in the brain tissue of children with Krabbe disease and metachromatic leukodystrophy, sphingolipid storage disorders caused by Lewy body disease risk genes, suggesting a direct link between deficient sphingolipid homeostasis and α-synuclein aggregation [16, 17]. It is not clear how dysfunction of the lysosomal enzymes encoded by these genes could contribute to the disease process of DLB and related Lewy body diseases; however, we have previously hypothesised that this could reflect the combination of deficient lysosomal clearance and the accumulation of lipid substrates that cross-seed α-synuclein aggregation [101]. It is notable that dyshomeostasis of sphingolipids has previously been reported to underlie the diversity of pathological processes implicated in Lewy body disease, such as deficient autophagy, inflammation and mitochondrial dysfunction, and thus altered sphingolipid metabolism could provide a parsimonious explanation for the range of pathological processes reported in the brain in Lewy body disease [102, 103]. However, further functional work is necessary to understand the relative contribution of distinct pathological processes of DLB more broadly, and multi-disciplinary studies evaluating the full spectrum of pathological changes in DLB are likely to be fruitful in developing therapies that treat the disorder, rather than specific facets of it.

New horizons in the therapeutic management of DLB

The considerable body of work aiming to understand the pathogenesis of DLB, as summarised above, is likely to open up new fronts in the development of disease-modifying therapies for DLB. To some extent, such efforts have already begun with the development of therapies aiming to ameliorate α-synuclein aggregation, albeit in the context of PD, such as prasinezumab; however, this therapy has not shown meaningful benefits in a Phase II clinical trial beyond an exploratory analysis in a post-hoc sub-group analysis [104, 105]. It is striking that relatively few clinical trials are underway targeting α-synuclein given the central role ascribed to it in the pathogenesis of Lewy body diseases, even in the relatively more active therapeutic pipeline of PD [106]. In part, this may be due to the typically intracellular nature of α-synuclein pathology in contrast to the primarily extracellular amyloid-β, where a number of antibodies have been developed for Alzheimer’s disease [107]. Nevertheless, further therapies targeting α-synuclein aggregation seem a likely development in Lewy body disease in the future, and given the prominent role ascribed to this protein in DLB it seems prudent to evaluate their efficacy in DLB in addition to PD.

As noted above, α-synuclein pathology is not the only proteopathic lesion typically identified in DLB brains, and concomitant Alzheimer-type pathology is also a frequent occurrence and appears to contribute to accelerated cognitive decline. The relatively recent development of antibody-based therapies targeting amyloid-β have demonstrated a remarkable capacity to reduce the burden of cerebral amyloid-β pathology [108, 109]. Given the association between concomitant Alzheimer-type pathology and accelerated cognitive decline in DLB, amyloid antibodies would seem to be a logical option for future evaluation regarding DLB disease modification. However, it is important to note that amyloid antibodies have only modest effects in leading to clinical improvement in Alzheimer’s disease and have known risks that would have to be balanced against the probability of acquiring clinically meaningful improvements.

Attenuating mitochondrial dysfunction has been a major theme of PD research, and a number of clinical trials are currently on-going targeting energy homeostasis. Nicotinamide riboside, a precursor of the essential mitochondrial oxidative phosphorylation cofactor, nicotinamide adenine dinucleotide (NAD+), is currently in Phase II clinical trial in PD [110]. Although no efficacy data is currently available, future indications of improvement in PD could suggest this therapy may also be effective in DLB given mitochondrial dysfunction is thought to be a feature of DLB. A recent study has indicated that terazosin, a drug that promotes ATP production through glycolysis, is associated with lower risk of developing DLB [111]. One could speculate that enhancing ATP production through glycolysis could compensate for diminished ATP production due to mitochondrial dysfunction, and may suggest this could be an effective treatment for DLB. However, it is important to emphasise that it is currently unclear whether mitochondrial dysfunction is a cause or consequence of DLB; thus, it is not clear whether improvement of mitochondrial function will be truly disease-modifying. Nevertheless, the close association between mitochondrial dysfunction and Lewy body disease more broadly would suggest this direction is worth further exploration.

Although proteopathy and mitochondrial dysfunction are common features of DLB, relatively few studies have sought to understand pathological changes that could underlie both. A plausible factor underlying both features is deficiencies in the autophagy-lysosomal pathway, as this would plausibly lead to concentration-dependent aggregation of α-synuclein and mitochondrial dysfunction through reduced mitophagy [112]. Although enhancing autophagy has been suggested as a therapeutic strategy in DLB, supported by positive results from pre-clinical studies using autophagy promoting agents such as rapamycin, it remains untested [113–115]. Given the evidence that autophagy is deficient in DLB and that its restoration can attenuate cell death and phenotype in vivo, future studies may wish to investigate whether rapamycin or other autophagy enhancers are beneficial in the treatment of DLB. In our recent study, we identified a number of commonly used drugs, such as celecoxib and memantine, which activate autophagy and could thus be repurposed to target autophagy in DLB [116].

A key aspiration must be to understand the mechanisms that underlie the full spectrum of pathological changes observed in DLB, with a view to targeting all pathological changes rather than one or two in isolation. As noted above, a number of Lewy body disease risk genes encode lysosomal enzymes responsible for the degradation of sphingolipids. Although most of these risk genes have been identified in PD, perhaps due to larger numbers of samples increasing study power, GBA1 encoding glucocerebrosidase is amongst the strongest risk factors for DLB [117]. A Phase II clinical trial targeting glucocerebrosidase with the mucolytic ambroxol has recently commenced [118]. Given that a number of other sphingolipid catabolic enzymes have been implicated in Lewy body disease more broadly, including SMPD1, GALC and ARSA, future studies may wish to determine whether targeting the function of their protein products, or the prosaposins that activate many of these enzymes, is a useful therapeutic strategy in DLB.

Conclusions

The clinical management of DLB is currently framed around targeting specific symptom domains to improve patient and caregiver quality of life. Such interventions are modestly effective but there is a pressing need for disease-modifying drugs to treat the underlying cause of the disease and alleviate symptoms more holistically. A key barrier to the development of disease-modifying drugs is lack of understanding about the cause of DLB and, by extension, appropriate methods through which it can be attenuated. A number of therapies are currently being explored to treat either proteopathy, mitochondrial dysfunction or deficient autophagy in isolation. However, a more complete understanding of the pathological processes that underlie the full spectrum of changes observed in the DLB brain is likely to herald the development of more effective therapies by treating the disease holistically, rather than particular manifestations.

Author contributions

DE and JPT jointly conceived and prepared the manuscript together.

Funding

DE is funded by an Alzheimer’s Research UK Senior Fellowship (ARUK-SRF2022A-006). This work was supported by the NIHR Newcastle Biomedical Research Centre (BRC), a partnership between Newcastle Hospitals NHS Foundation Trust, Newcastle University, and Cumbria, Northumberland and Tyne and Wear NHS Foundation Trust and is funded by the National Institute for Health and Care Research (NIHR). The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.