Correspondence 1032 www.thelancet.com/neurology Vol 17 December 2018 Marialuisa Quadri and colleagues’ recent report 1 in The Lancet Neurology on the involvement of LRP10 variants in various α-synucleinopathies is important because it provides a potential novel gene for dementia with Lewy bodies, a disease for which causative genes have been difficult to identify. In an attempt to replicate these results, we mined whole-exome sequencing data from a cohort of 1040 patients with dementia with Lewy bodies and 1422 controls. Patients, who were identified through the International Dementia with Lewy Bodies Genetics Consortium, were included if they met the McKeith criteria 2 of intermediate or high likelihood for a neuropathological diagnosis of dementia with Lewy bodies; samples without neuro- pathological information were diagnosed according to established clinical criteria. 2 Controls were derived from two sources: the 1958 British Birth cohort 3 (n=985) and the Healthy Exomes database 4 (n=437). Sequencing of these samples led to the identification of 25 variants, including 14 occurring only in controls and eight occurring ex- clusively in patients (appendix). Two of the variants identified by Quadri and colleagues in two Dutch individuals were present in our data (Ala212Serfs*17 and Asn517del), in one person each. The individuals with these variants in our study, who were from The Netherlands Brain Bank, also match those in the study by Quadri and colleagues on sex, age at death, and disease duration. Given the extreme rarity of these variants (they are absent from the Genome Aggregation Database 5 ) and the low probability of these individuals having the same diagnosis as in the previous study, it is very likely that our patients are the same as those described by Quadri and colleagues. 1 Additionally, we identified two protein-truncating variants (Gln67* and Arg554* in one LRP10 in α-synucleinopathies We read with interest the Article by Marialuisa Quadri and colleagues 1 in The Lancet Neurology, in which they suggested LRP10 mutations as a cause of Parkinson’s disease, Parkinson’s disease dementia, and dementia with Lewy bodies. As part of the International Parkinson’s Disease Genomics Consortium, we analysed LRP10 in exome sequencing data from 2835 individuals with Parksinson’s disease and 5343 individuals without known neurological dis- ease (appendix). We also analysed exome sequencing data from 111 individuals with pathologically confirmed dementia with Lewy bodies and another 233 individuals without known neurological disease (appendix). In the exomes from people with Parkinson’s disease and the first set of controls, after applying filtering criteria (appendix), we identified 92 variants in LRP10, including 17 synonymous, 72 non-synonymous, and three loss-of-function variants; the loss-of-functon variants were present in one patient and two controls (appendix). None of these variants were reported in the original Article, 1 and no enrichment among cases compared with controls was found. To assess the collective association of these variants with Parkinson’s disease, we did gene- based collapsing analyses, finding no evidence of an association between rare variant burden and Parkinson’s disease (appendix). Exome analysis of individuals with dementia with Lewy bodies and corresponding controls found three non-synonymous variants in one patient and two controls (appendix), none of which were reported by Quadri and colleagues. 1 Gene-based aggregation tests in the individuals with dementia with Lewy bodies and controls also showed no association with the disease (appendix). In summary, we found no evidence to support a pathogenic role for LRP10 mutations in Parkinson’s disease or dementia with Lewy bodies. Although our series of patients with Parkinson’s disease is, to our knowledge, the largest exome dataset in which LRP10 has been studied, our study was probably underpowered, with a power of around 80% to detect rare variants in Parkinson’s disease, and less than 25% power for dementia with Lewy bodies, at an α of 0·01. 2 Against this argument is the number of variants described by Quadri and colleagues: 1 they report mutations not only in the index family but also in several other, unrelated individuals. Additionally, the variants might be population-specific. Our population was mainly of northern European ancestry; however, the original report identified variants in several different populations, suggesting that, if LRP10 is a causal gene, mutations are not ethnically confined. In conclusion, our data suggest that caution needs to be applied when interpreting LRP10 as being a disease-causing gene in Parkinson’s disease or dementia with Lewy bodies. We declare no competing interests. Demis A Kia, Marya S Sabir, Sarah Ahmed, Joanne Trinh, *Sara Bandres-Ciga, on behalf of the International Parkinson’s Disease Genomics Consortium† sara.bandresciga@nih.gov Department of Molecular Neuroscience, UCL Institute of Neurology, Queen Square, London, UK (DAK); Neurodegenerative Diseases Research Unit, National Institute of Neurological Disorders and Stroke (MSS, SA), and Laboratory of Neurogenetics, National Institute on Aging (SA, SB-C), National Institutes of Health, Bethesda, MD, USA; and Institute of Neurogenetics, University of Lübeck, Lübeck, Germany (JT) †Members listed in the appendix. 1 Quadri M, Mandemakers W, Grochowska MM, et al. LRP10 genetic variants in familial Parkinson’s disease and dementia with Lewy bodies: a genome-wide linkage and sequencing study. Lancet Neurol 2018; 17: 597–608. 2 Wu B, Pankow JS. On sample size and power calculation for variant set-based association tests. Ann Hum Genet 2016; 80: 136–43. See Online for appendix For the Healthy Exomes database see https://www. alzforum.org/exomes/hex See Online for appendix For the Genome Aggregation Database see http://gnomad. broadinstitute.org/