Parkinson’s disease, Lewy body dementia, and Alzheimer’s disease are considered part of the same disease spectrum because their symptoms and brain pathologies often overlap. But genetic data backing up this theory has been sparse. Now, a whole-genome-sequencing based, genome-wide association study puts some meat on its bones. In the February 15 Nature Genetics, researchers led by Sonja Scholz at the National Institute of Neurological Disorders and Stroke, Bethesda, Maryland, report five LBD risk loci. Three—GBA, APOE, and SNCA—were known; two—BIN1 and TMEM175—are new to LBD research. APOE and BIN1 are AD risk genes, as well, whereas variants in GBA, SNCA, and TMEM175 increase risk for PD. A larger sample will be needed to confirm the two new LBD risk loci, researchers said.

  • WGS identifies five LBD loci—two are new.
  • Three are known PD genes, two are AD genes.
  • People who carried AD variants died sooner; people with PD variants developed LBD at a younger age.

“Neuropathological features of multiple dementias co-existing in a person is known; these results suggest that does not occur by chance,” Philippe Amouyel, Institut Pasteur de Lille, France, wrote to Alzforum (full comment below). Julie Williams at Cardiff University, Wales, U.K., agreed. “This is pretty hard evidence that there are shared factors contributing to these three diseases,” she told Alzforum.

Memory loss and worsening cognition loom large in AD, LBD, and sometimes late-stage PD. People with PD lose motor function early on, and people with later-stage LBD do, as well. As for neuropathology, coexisting Lewy bodies, amyloid plaques, and neurofibrillary tau tangles bridge PD and AD in some people with LBD (Sep 2016 newsKon et al., 2020). “Stitching the three diseases together on a spectrum accounts for the clinical similarities,” said Clifton Dalgard, Uniformed Services University, Bethesda. Dalgard was a senior author on the study, along with Scholz, Bryan Traynor and Raphael Gibbs at the National Institute on Aging, Bethesda, Owen Ross at the Mayo Clinic, Jacksonville, Florida, and Adriano Chiò at the University of Turin, Italy.

LBDs encompass both dementia with Lewy bodies (DLB), which tends to present initially with cognitive symptoms, and Parkinson’s disease dementia (PDD), which begins as a movement disorder. A previous GWAS, led by Rita Guerreiro and Jose Bras, then at University College, London, linked GBA, APOE, and SNCA variants to DLB (Dec 2017 news; Dec 2015 news). In this new, whole-genome-sequencing (WGS) study, researchers combined DLB and PDD cases under the blanket term LBD. This was controversial: Several researchers were concerned that pooling the two may muddy the results.

Searching for Commonalities. Lewy body dementia GWAS hits were combined with expression data to identify risk variants, calculate genetic risk scores, and examine faulty pathways that underlie LBD, AD, and PD. [Courtesy of Chia et al., Nature Genetics, 2021.]

For this study, researchers combined samples from 2,981 people with clinically diagnosed LBD and from 2,173 healthy people of European ancestry from 27 sites in North America and 17 in Europe. Adding genome data from 1,016 volunteers of European ancestry from the NIA cohorts and the Accelerating Medicine Partnership–Parkinson’s Disease Initiative and 1,202 over age 80 in the Wellderly cohort (Erikson et al., 2016) generated a total of 4,391 controls available for analysis. LBDs are difficult to diagnose clinically due to their overlapping symptoms with other dementias. To get around that, the researchers selected for pathologically confirmed LBD cases, which accounted for 69 percent of the total.

Co-first authors Ruth Chia at NIA and Marya Sabir at NINDS sequenced each participant’s entire genome, ultimately analyzing 2,591 LBD cases and 4,027 controls. Whole-genome sequencing allows researchers to tease out both common and rare variants. “This allows us to detect associations with disease that we may have otherwise missed,” Dalgard explained.

The researchers then correlated variants with an LBD diagnosis, using a cutoff P value of less than 5 × 10−8. Only five risk loci prevailed (see image below).

Next, Chia and colleagues cross-checked the associations against an additional 970 LBD and 8,928 controls chosen from a previous cohort from Scholz and colleagues that included 468 DLB cases, 57 PDD cases, and 591 controls, and from the previous DLB GWAS cohort, which included 1,743 DLB cases and 4,454 controls (Sabir et al., 2019; Guerreiro et al., 2017). Among these samples, Chia found that only the associations between LBD and APOE and SNCA held. However, in a meta-analysis of the combined discovery and replication cohorts, all five loci strongly associated with LBD.

Next, the researchers searched for rare variants found in fewer than 1 percent of cases. Only GBA came up, meaning that both common and rare variants in this gene are linked to the disease (see image below).

High Five. Multiple common variants (red dots) at five loci survived statistical analysis in an LBD GWAS (top). Known LBD risk genes are green, new ones are black. GBA also associated with LBD in a search for rare variants (bottom). [Courtesy of Chia et al., Nature Genetics, 2021.]

All five genes previously have been tied to other neurodegenerative diseases. ApoE and BIN1, which encodes an endosomal trafficking protein, are two of the strongest risk genes for late-onset AD. GBA, encoding lysosomal enzyme β-glucocerebrosidase; TMEM175, encoding a lysosomal potassium channel; and the α-synuclein gene SNCA are PD risk genes. There was no genetic synergy between AD and PD risk, meaning that disease alleles independently affected LBD risk.

The five identified genes appear to point to the lysosomal pathway as a nexus for LBD pathology. ApoE and GBA variants mishandle lipids, impairing their metabolism and clearance from lysosomes (Lee et al., 2021). TMEM175 stabilizes lysosomes and reduces buildup of phosphorylated α-synuclein (Jinn et al., 2017). In people, the rs6733839 Bin1 SNP, the variant identified in the current study, regulates microglial function (Nov 2019 news). In rat neurons, lack of Bin1 revs up endocytosis and promotes tau misfolding and aggregation (Oct 2016 news). In mice, Bin1 deficiency impairs synaptic transmission and spatial memory (Mar 2020 news). 

“Neurodegenerative disease risk genes commonly impair essential cellular pathways, such as protein degradation and endosomal trafficking, that are crucial in many age-related diseases,” Scholz noted. Dalgard agreed. “Both BIN1 and TMEM175 are involved with either eating up protein trash in the brain or processing trash that has been eaten up,” he told Alzforum.

Chia, Sabir, and colleagues wondered if having the BIN1 or TMEM175 variant correlated with amyloid plaque or neurofibrillary tangle load. They counted plaques in 700 of the LBD cases using the CERAD criteria, and tangles in 1,459 cases using postmortem Braak staging. There was no association for the TMEM175 variant; however, people who carried the rs6733839-T BIN1 SNP had more tangles than those without the variant. “An association between BIN1 and tangles in the brain was also reported for AD cases, which seems to reinforce the role of this interaction in dementia,” wrote Jean-Charles Lambert, Institut Pasteur de Lille, France (full comment below).

To dig deeper into how the five loci were driving LBD risk, the researchers turned to expression data. They used co-localization statistics called Coloc to ask if higher risk and altered gene expression could both be traced back to the same variant (Giambartolomei et al., 2014). They examined expression quantitative trait loci (eQTL) data from two datasets. eQTLGen draws on blood-expression data from 31,684 healthy people, and PsychENCODE catalogs brain eQTLs from 1,387 healthy people and those with autism, bipolar disorder, and schizophrenia. Only eQTLs at the TMEM175 and the SNCA loci emerged; each regulated their respective gene’s expression in the blood and brain. Curiously, the SNCA eQTL regulates an antisense transcript that the authors found to be specifically expressed in neurons. The authors believe the LBD variant increases SNCA antisense expression, suppressing α-synuclein production. “This insight may open up a new therapeutic avenue for modifying disease risk by fine-tuning SNCA expression,” Scholz wrote to Alzforum.

To probe the variants for effects on cognition, the scientists calculated a polygenic risk score based on the five WGS hits and 117 others that fell just under the cutoff of statistical significance. In the 214 cases for whom the authors had Clinical Dementia Rating scores, they saw that those with the highest risk scores had the most severe dementia.

Where’s the Damage? LBD variants (dots) and their biological processes (orange), cellular functions (blue), molecular functions (green), and pathways from WikiPathways (pink). A higher enrichment percent means more variants feed into a given pathway; bigger dots denote a stronger link between variant and pathway. [Courtesy of Chia et al., Nature Genetics, 2021.]

Previously, researchers had found that Parkinson’s patients who had many genetic risk factors for AD were more likely to develop aggressive PD with dementia than those without AD genetic risk (Sandor et al., 2021). Here, Chia found that people who had a higher AD risk score died sooner after their LBD diagnosis, and at a younger age. "Genetic risk factors for Alzheimer's result in quicker decline in both Parkinson's and now Lewy body dementia," Williams told Alzforum. "We need to know more about these mechanisms of action because they span across different dementias."

Grouping the LBD-linked variants into molecular pathways, Chia, Sabir, and colleagues found that most of the five definite and 117 possible variants were involved in Aβ and endocytosis regulation, tau binding, and protein-lipid complexes (see image at right).

Knowing pathways that are affected in multiple forms of dementia can inform clinical trial design and facilitate drug repurposing. “We hope shared gene variants and disease pathways could be targeted by disease-modifying therapies,” Scholz said. Other agree. “Including subgroups of LBD patients in AD and PD drug trials would uncover if they can also benefit,” Amouyel wrote. “If you have preventative measures that can help brain tissue recycle trash better, then we could prevent all three diseases at once,” Dalgard added.

Chia and colleagues deposited their GWAS data on NCBI’s database of Genotypes and Phenotypes.—Chelsea Weidman Burke


  1. In this paper, Chia et al. described the analysis of an impressive volume of whole-genome-sequencing (WGS) data with the main objective to characterize genetic determinants of Lewy body dementia. They report five genome-wide significant signals, three already known (GBA, SNCA and APOE) and two new to LBD (BIN1 and TMEM175).

    This paper is based on classical approaches, particularly well-performed without unnecessary analyses as we can sometimes see in such genetic papers. The results are very solid, with replication in independent populations and sensitivity analyses in pathologically defined LBD. This latter point is particularly relevant since the two new loci are already known to be associated with the risk of other neurodegenerative diseases.

    In a way, it's as if the authors had a Ferrari, i.e., WGS, but decided to bridle it to a Ford to do reasonable analyses at first, genome-wide association (GWAS) and whole-exome sequencing (WES). This is definitely not a criticism. There is no doubt that these data will be particularly valuable to decipher more deeply the genetics of LBD, especially structural variants as mentioned by the authors and will be a remarkable resource for all researchers working on neurodegenerative diseases.

    These genetic data seem to support that LBD lies on the continuum of Parkinson’s disease and Alzheimer’s disease. Beyond APOE, SNCA, and GBA, TMEM175 is a genetic risk factor for PD, and BIN1 is for AD. In particular, BIN1 is the second genetic factor of Alzheimer's disease after APOE in terms of association level (Seshadri et al., 2010) and Chia et al. also report that the sentinel variant associated with LBD (the same as the one associated with AD), is also associated with neurofibrillary tangle pathology in the brains of LBD cases. This latter observation is of particular interest since this association was also reported in the brains of AD cases (Chapuis et al., 2013). This seems to reinforce the role of the BIN1-tau interaction in pathophysiological processes (Sottejeau et al ., 2015; Sartori et al, 2019). 


    . Genome-wide analysis of genetic loci associated with Alzheimer disease. JAMA. 2010 May 12;303(18):1832-40. PubMed.

    . Increased expression of BIN1 mediates Alzheimer genetic risk by modulating tau pathology. Mol Psychiatry. 2013 Nov;18(11):1225-34. Epub 2013 Feb 12 PubMed.

    . Tau phosphorylation regulates the interaction between BIN1's SH3 domain and Tau's proline-rich domain. Acta Neuropathol Commun. 2015 Sep 23;3:58. PubMed.

    . BIN1 recovers tauopathy-induced long-term memory deficits in mice and interacts with Tau through Thr348 phosphorylation. Acta Neuropathol. 2019 Oct;138(4):631-652. Epub 2019 May 7 PubMed.

  2. This is important work which explains the connections of dementia with Lewy bodies with Alzheimer’s and Parkinson’s diseases, as indeed indicated by the coexistence of Lewy bodies, amyloid plaques, and neurofibrillary tangles in dementia with Lewy bodies.

    The results of this GWAS study will contribute to understanding the mechanism behind the origin of dementia with Lewy bodies and could in the end lead to mechanism-based therapies.

  3. Among dementia, Lewy body disease (LBD) is the second most common type of progressive dementia after Alzheimer's disease. As in most dementias, LBD causes a progressive decline in cognitive function. However, compared to AD, several symptoms are more specific of LBD: visual hallucinations, Parkinson's disease (PD)-like signs, and a younger age at onset.

    The main feature of LBD is the abnormal accumulation of α-synuclein protein into specific masses, the Lewy bodies. Interestingly, α-synuclein is also associated with Parkinson's disease and patients with Lewy bodies in their brains also have plaques and tangles associated with AD. More generally, the necropsy of aged people finds accumulations of neuropathologies that account for the association between age and dementia. Thus, among the 1,362 autopsied participants of three community-based clinico-pathologic cohorts, 44 percent had a clinical dementia diagnosis (Power et al., 2018). In this study, the pathways involving amyloid/tau, neocortical Lewy bodies, and TDP-43/hippocampal sclerosis were interdependent, attributable to the importance of Aβ plaques. Age-related increases in dementia risk could be attributed to accumulation of multiple pathologies, each of which contributes to dementia risk.

    Several questions remained: Does the co-existence of neuropathological features of different dementias in one individual occur by chance, or does the clinical expression of dementia vary according to the predominance of one of these neuropathological features over others? One way to try to answer these questions can be approached by genetics. Here, Chia et al. have performed a whole-genome sequencing on 2,981 LBD cases and 2,173 neurologically healthy controls. A replication study for the GWAS analyses was also performed in 970 LBD cases and 8,928 controls. This GWAS identified five independent genome-wide-significant loci that influence risk for developing LBD. Among these, three were already known as LBD genes, the GBA (Glucocerebrosidase) already associated with PD, the famous AD APOE gene encoding apolipoprotein E, and the famous PD gene SNCA encoding α-synuclein. They added two new genes to this list: another famous AD gene, BIN1 (bridging integrator 1) and the TMEM175 gene on chromosome 4p16.3, a known PD risk locus.

    The risk scores for AD and PD of the subjects included in the population sample of this study, derived from the very rich collection of GWAS already published in the literature, were associated with LBD disease status, and with age at death, age at onset, and the duration of illness observed among LBD cases. Individuals diagnosed with LBD had a 66 percent increased genetic risk for developing AD and a 20 percent increased genetic risk for developing PD. The AD genetic risk score was also found to be significantly associated with an earlier age of death in LBD and shorter disease duration. Conversely, the PD genetic risk score was associated with an earlier age at onset among patients diagnosed with LBD. No evidence of interaction between the genetic risk scores of AD and PD in the LBD cohort was detected, implying that AD and PD risk variants were independently associated with LBD risk.

    These results suggest that the co-existence of neuropathological features of dementias in one individual does not occur by chance. Dementia syndrome may be explained by a combination of various pathological processes (vascular, amyloid, tau, α-synuclein) mixed in a subtle dosage offering a wide range of clinical features for LBD patients. This discovery will have also an implication on the treatment strategy, as more and more drugs will be available for these different dementia pathways. Indeed, it would be now interesting to include subgroups of LBD patients in AD and PD drug trials to test if they can also benefit of these new treatments.


    . Combined neuropathological pathways account for age-related risk of dementia. Ann Neurol. 2018 Jul;84(1):10-22. Epub 2018 Jun 26 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Tau Deepens Cognitive Trouble in Lewy Body Diseases
  2. First Genome-Wide Association Study of Dementia with Lewy Bodies
  3. Genetics of DLB: Setting Up to Fill a Mostly Empty Canvas
  4. Cell-Specific Enhancer Atlas Centers AD Risk in Microglia. Again.
  5. Lack of BIN1 Sows Tau Trouble for Neurons
  6. Alzheimer’s Gene BIN1 Promotes Synaptic Transmission

Paper Citations

  1. . Neuropathology of Lewy body disease: Clinicopathological crosstalk between typical and atypical cases. Neuropathology. 2020 Feb;40(1):30-39. Epub 2019 Sep 9 PubMed.
  2. . Whole-Genome Sequencing of a Healthy Aging Cohort. Cell. 2016 May 5;165(4):1002-11. Epub 2016 Apr 21 PubMed.
  3. . Assessment of APOE in atypical parkinsonism syndromes. Neurobiol Dis. 2019 Jul;127:142-146. Epub 2019 Feb 21 PubMed.
  4. . Investigating the genetic architecture of dementia with Lewy bodies: a two-stage genome-wide association study. The Lancet Neurol, Dec 15, 2107.
  5. . Defective Lysosomal Lipid Catabolism as a Common Pathogenic Mechanism for Dementia. Neuromolecular Med. 2021 Mar;23(1):1-24. Epub 2021 Feb 7 PubMed.
  6. . TMEM175 deficiency impairs lysosomal and mitochondrial function and increases α-synuclein aggregation. Proc Natl Acad Sci U S A. 2017 Feb 28;114(9):2389-2394. Epub 2017 Feb 13 PubMed.
  7. . Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet. 2014 May;10(5):e1004383. Epub 2014 May 15 PubMed.
  8. . Universal latent axes capturing Parkinson’s patient deep phenotypic variation reveals patients with a high genetic risk for Alzheimer’s disease are more likely to develop a more aggressive form of Parkinson’s. bioRxiv. February 18, 2021

External Citations

  1. Accelerating Medicine Partnership–Parkinson’s Disease Initiative
  2. eQTLGen
  3. PsychENCODE
  4. NCBI’s database of Genotypes and Phenotypes

Further Reading

Primary Papers

  1. . Genome sequencing analysis identifies new loci associated with Lewy body dementia and provides insights into its genetic architecture. Nat Genet. 2021 Mar;53(3):294-303. Epub 2021 Feb 15 PubMed.