ARTICLE OPEN ACCESS

The SPID-GBA studySex distribution Penetrance Incidence and Dementia in GBA-PD

Letizia Straniero PhD Rosanna Asselta PhD Salvatore Bonvegna MD Valeria Rimoldi PhD

Giada Melistaccio MSc Giulia Solda PhD Massimo Aureli PhD Matteo Della Porta MD Ugo Lucca MSc

Alessio Di Fonzo MD PhD Anna Zecchinelli MD Gianni Pezzoli MD Roberto Cilia MD and

Stefano Duga PhD

Neurol Genet 20206e523 doi101212NXG0000000000000523

Correspondence

Dr Pezzoli

GianniPezzoliasst-pini-ctoit or

Prof Duga

Stefanodugahunimedeu

AbstractObjectiveTo provide a variant-specific estimate of incidence penetrance sex distribution and associationwith dementia of the 4 most common Parkinson disease (PD)-associated GBA variants weanalyzed a large cohort of 4923 Italian unrelated patients with primary degenerative parkin-sonism (including 3832 PD) enrolled in a single tertiary care center and 7757 ethnicallymatched controls

MethodsThe pE326K pT369M pN370S and pL444P variants were screened using an allele-specificmultiplexed PCR approach All statistical procedures were performed using R or Plink v107

ResultsAmong the 4 analyzed variants the pL444P confirmed to be the most strongly associated withdisease risk for PD PD dementia (PDD) and dementia with Lewy bodies (DLB) (odds ratio[OR] for PD 1563 95 confidence interval [CI] = 804ndash3037 p = 49710minus16 OR for PDD2957 95 CI = 1407ndash6213 p = 38610minus19 OR for DLB 1027 95 CI = 3138ndash3361 p =19110minus14) However an unexpectedly high risk for dementia was conferred by pE326K (ORfor PDD 480 95CI = 287ndash802 p = 21210minus9 OR for DLB 1224 95CI = 495ndash3024 p =57110minus8) which on the basis of the impact on glucocerebrosidase activity would be expectedto be mild The 15ndash21 male sex bias described in sporadic PD was lost in pT369M carriersWe also showed that PD penetrance for pL444P could reach the 15 at age 75 years

ConclusionsWe report a large monocentric study on GBA-PD assessing mutation-specific data on the sexdistribution penetrance incidence and association with dementia of the 4 most frequentdeleterious variants in GBA

These authors contributed equally

From the Department of Biomedical Sciences (LS RA VR GM GS MDP SD) Humanitas University Humanitas Clinical and Research Center (RA VR GS MDP SD)IRCCS Rozzano Fondazione Grigioni per il Morbo di Parkinson (SB AZ GP) Parkinson Institute (SB AZ GP RC) ASST ldquoGaetano Pini-CTOrdquo Department of MedicalBiotechnology and Translational Medicine (MA) University of Milan Laboratory of Geriatric Neuropsychiatry (UL) Istituto di Ricerche Farmacologiche Mario Negri IRCCS IRCCSFoundation Carsquo Granda Ospedale Maggiore Policlinico (ADF) Dino Ferrari Center Neuroscience Section Department of Pathophysiology and Transplantation University of Milanand Fondazione IRCCS Istituto Neurologico ldquoCarlo Bestardquo (RC) Milan Italy

Go to NeurologyorgNG for full disclosures Funding information is provided at the end of the article

The Article Processing Charge was funded by PRIN (Programmi di Ricerca Scientifica di Rilevante Interesse Nazionale Grant no 2017228L3J Program coordinator SD)

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 40 (CC BY-NC-ND) which permits downloadingand sharing the work provided it is properly cited The work cannot be changed in any way or used commercially without permission from the journal

Copyright copy 2020 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology 1

Parkinson disease (PD) is a neurodegenerative disordercharacterized by motor and nonmotor symptoms12 withdementia often contributing to poor life quality and reducedsurvival3 Besides aging epidemiologic studies reported ahigher risk of PD for males4

Among predisposing genes GBA (encoding the lysosomalbeta-glucocerebrosidase) represents the most commonlarge-effect genetic factor associated with PD in all analyzedpopulations56 Patients with heterozygous GBA mutationshave peculiar characteristics including earlier onset morerapid progression of motor impairment and greater risk fordementia and death7ndash9 Considering other synucleino-pathies GBA mutations were even more strongly associatedwith PD dementia (PDD) and dementia with Lewy bodies(DLB)1011

Among mutations reported in GBA 4 missense variants(pE326K pT369M pN370S and pL444P) account for82 of PD deleterious alleles12 Data on their pene-trance are disparate and sometimes conflicting especiallyfor variations having a lower impact on enzyme activity(pE326K and pT369M figure e-1 linkslwwcomNXGA327)813ndash18 Reasons for such discrepancies likelystem from (1) differences in the effect size of differentmutations (2) effects of agingsex on mutation-specificpenetrance and (3) population-specific differences inGBA variant frequencies (figure e-1 linkslwwcomNXGA327)

Here we propose an extension of our previous study11 bothin terms of number of patientscontrols and of analyzed GBAvariants (besides pN370S and pL444P we genotyped alsopE326K and pT369M) We aimed to define the prevalencepenetrance of the most common GBA-PD predisposing var-iants in a large monocentric PD cohort (Sex PenetranceIncidence and Dementia SPID-GBA study) especially in theview of ongoing trials which will hopefully indicate noveltreatments targeting GBA-PD19

MethodsStandard protocol approvals registrationsand informed consentsThis study was approved by the Ethics Committee of theFondazione IRCCS Carsquo Granda Ospedale Maggiore Policli-nico (study ID 483 March 8 2018) and conducted accordingto the Declaration of Helsinki All participants signed an in-formed consent

Study participants

PatientsWe enrolled 4923 unrelated patients who contributed to theParkinson Institute Biobank (wwwparkinsonitbiobanca) ofMilan Italy from 2002 to 2015 regardless of their sex familyhistory age at onset or other clinical characteristics Apartfrom 26 patients having frontotemporal dementia (FTD) allthe others received a diagnosis of primary degenerative par-kinsonism (PKS) by expert neurologists 3832 fulfilled thecriteria for PD 95 for DLB 216 for multiple system atrophy(MSA) 199 for progressive supranuclear palsy (PSP) and 56for corticobasal degeneration (CBD) Diagnosis of PD wasmade according to the UKBrain Bank criteria20 and diagnosisof DLB was made for those fulfilling the 1-year rule betweenthe dementia onset and PKS21 Patients with suspect of sec-ondary PKS were excluded A total of 323 patients who didnot meet the validated diagnostic criteria of clinically probablefor the various forms of primary (degenerative) PKS (ie PDvs DLB vs MSA vs PSP vs CBD vs FTD) and at the sametime who fulfilled the diagnostic criteria of a clinically possiblePKS for more than one among the etiologies listed abovewere classified in the category of undefined PKS

For each participant the following demographic and clinicaldata were collected sex place of birth age at onset (the age atwhich the patient noticed the first motor symptom) diseaseduration (calculated on the basis of the last examination) anddementia status (diagnosed according to the criteria of theDiagnostic and Statistical Manual of Mental Disorders DSM-IV)7 whenever full neuropsychological testing was availablethe Movement Disorder Societyndashrecommended diagnosticcriteria were applied22 For each patient a detailed familyhistory was collected and patients were classified as familial ifat least 1 relative among their 1st- or 2nd-degree relatives hada diagnosis of PD Only 1 proband was considered for eachfamily Among patients with PD those who developed de-mentia during the course of the disease were defined asPDD22 All clinical data were independently reviewed by anadditional experienced neurologist (RC or SB)

Among the 3832 patients fulfilling the criteria for PD 30were previously screened for several PD-related genes such asLRRK2 (pG2019S pR1441CG and pI2020L) ParkinPINK1 DJ1 and SNCA The 129 carriers of mutations inthese genes were excluded from subsequent analyses

ControlsOur control cohort consisted of 7757 subjects deriving from3 different Italian series In particular a total of 1625 controls

GlossaryCBD = corticobasal degeneration CI = confidence interval DLB = dementia with Lewy body FTD = frontotemporaldementia GBA = glucocerebrosidaseMSA = multiple system atrophy OR = odds ratio PD = Parkinson disease PDD = PDdementia PKS = parkinsonism

2 Neurology Genetics | Volume 6 Number 6 | December 2020 NeurologyorgNG

were recruited among partners and caregivers of patients withPD at the Parkinson Institute ofMilan They were not affectedby neurodegenerative disorders and denied any family historyfor movement disorders in first-degree relatives

A total of 4016 individuals along with information abouttheir sex age ancestry and variant calls from whole-exomesequencing data were collected as part of the MyocardialInfarction Genetics Consortium (MIGen cohort)23

A total of 2116 subjects deriving from the Health amp Anemiastudy2425 were also recruited Health amp Anemia is a pro-spective population-based observational study (performedbetween years 2003 and 2017) of all residents in the munic-ipality of Biella (Italy) aged 65 years or older aimed to in-vestigate the clinical consequences of anemia in theelderly2425 Information on sex age and ancestry was avail-able for all the analyzed subjects

GBA mutation screeningDNA was extracted from peripheral blood (Maxwell-16 Sys-tem Promega Madison WI or Chemagic-Star workstationHamilton ON Canada)

The mutational screening for the 4 most common GBA ge-netic defects (pE326K p T369M pN370S and pL444Plegacy names) in cases and controls was performed by amultiplex allele-specific PCR approach designed to avoid thecoamplification of the highly homologous GBAP1 pseudo-gene26 All variants were confirmed by conventional Sangersequencing All PCRs were performed on 10 ng of genomicDNA using the GoTaq DNA Polymerase (Promega) Directsequencing of PCR products was performed using the BigDyeTerminator Cycle Sequencing Ready Reaction Kit v11(Thermo Fisher Scientific Carlsbad CA) and analyzed on anABI-3500 Genetic Analyzer (Thermo Fisher Scientific) Alloligonucleotides used in the genotyping and validation stepswere purchased from Sigma (St Louis MO)

Statistical analysisAll statistical procedures were performed using R (r-projectorg) or Plink v107 (zzzbwhharvardeduplink)

For eachGBA variant standard case-control analyses on allelefrequency data were performed with χ2 statistics (Fisher exacttest) using the minor allele as reference p values are pre-sented as noncorrected for multiple testing

The association of variables with the presence ofGBA variantsin patients was evaluated by testing for differences betweencarriers and noncarriers using either a χ2 test for categoricaldata or the Student t test for continuous data (variables wereanalyzed as quantitative factors after having controlled theirnonsignificant departure from linearity) The effect of eachfactor was expressed as the odds ratio (OR) and 95 confi-dence interval (CI) Unadjusted ORs were calculated using alogistic regression model which included only the factor of

interest adjusted ORs were obtained using a model includingthe variable of interest plus all variables found significant inthe first stage of analysis

Cumulative incidence was calculated and visualized combin-ing data for different variantsphenotypesex in a graphicalrepresentation that we named Vinicunca plot (after theamazing rainbow mountains of Peru) using the incidence Rpackage Penetrance was calculated according to27 using PDprevalence data of the Italian population28

This work conforms the STrengthening the REporting of Ge-netic Association Studies guidelines (goodreportsorgstregasee supplemental materials linkslwwcomNXGA327)

Data availabilityAny data not published within the article will be shared byrequest from any qualified investigator as anonymized data

ResultsDistributionof the4mostcommonPD-associatedGBA variants in cases and controlsWe screened a total of 4923 patients and 7757 controls forthe presence of the 4 most common PD-associated GBAvariants (pE326K pT369M pN370S and pL444P thelatter including carriers of the RecNcil allele) Overall 308patients and 170 controls were carriers of at least 1 GBAvariant (carrier frequency 67 and 22 respectively) Onlypatients with a single variant in GBA were considered forfurther analyses

Among synucleinopathies a gradient in carrier frequency isobservable going from 258 in DLB to 67 in PD (112in PDD) to 37 in MSA increased carrier frequenciescompared with controls were also observed for the tauo-pathies CBD FTD PSP and for undefined PKSs (table 1)

Comparing the carrier frequency of the 4 variants in indi-viduals affected by the different synucleinopathies in patientswith PD the 4 genetic variations have similar frequencies(range 13ndash21) while in PDD and evenmore so in DLBthe pL444P and the pE326K are overrepresented

Association of GBA variants with PD PDD andDLB the weak pE326K is not so weakCompared with the allele frequencies observed in controlsall variants had an allelic OR gt 156 for PD (range156ndash1563 corrected for age and sex table 2) As expectedthe pL444P shows by far the strongest association withdisease risk for all analyzed phenotypes (up to OR = 1027095 CI = 3138ndash3361 in DLB) except for MSA FTD andCBD where no pL444P carrier was found Among nonndashGD-associated variants pE326K has an OR for PDD (48095 CI = 287ndash802) and DLB (1224 95 CI =495ndash3024) similar to those of pN370S This is confirmed

NeurologyorgNG Neurology Genetics | Volume 6 Number 6 | December 2020 3

by the observation that pE326K is significantly associatedalso with undefined PKSs

Genotype-phenotype correlations in PDpT369M is the only variant not associatedwithdementia but showing a sex biasGiven the relatively low number of carriers with other diagnoses(table 1) we compared the clinical phenotype between GBAcarriers (considering all variants together or stratified accordingto the 4 variants) and noncarriers only in PD (table 3)

This analysis highlighted that

1 There is no significant sex unbalance among carriers ofpE326K pN370S and pL444P compared with non-carriers whereas pT369M shows a significant skewingtoward females (OR = 053 95 CI = 030ndash094)

2 An earlier age at onset is significantly associated only withthe 2 more severe mutations (pN370S and pL444P539 and 51 years respectively against 583 years innoncarriers)

3 Positive family history is significantly associated only withpL444P (OR = 23 95 CI = 136ndash389)

4 A higher risk of dementia was significantly associated notonly with the 2 more severe mutations but also with thepE326K variant (OR = 283 95 CI = 169ndash472)

Mutation- sex- and disease-specificcumulative incidence the sex mattersVinicunca plots in figure 1 report the calculated cumulativeincidence by variant and sex in PD According to the strongerassociation of pN370S and pL444P with earlier age at onsetat 40 years 70 of patients with GBA-PD were carriers ofthese variants Conversely at age 70 years the contribution of

the 2 nonndashGD-associated variants (pE326K and pT369M)was proportionally increased (from 30 to 43) For 3 of the 4analyzed variants a plateau was visible at an age ranging from71 to 75 years with the pL444P reaching the plateau first(figure 1A)

Looking at the sex distribution of patients carrying the dif-ferent variants a bimodal curve was observed for thepT369M variant with a first plateau reached at age 52 yearsand a second increase in disease prevalence observable after57 years Of interest males carrying the variant reached theplateau at age 66 years whereas the incidence in femalescontinued to increase up to 80 an effect that contributes tothe increased OR for female sex described above for thisvariant (figure 1B) Moreover considering all genetic defectstogether the mean age at onset in females was unexpectedlylower than in males (5324 plusmn 109 vs 5528 plusmn 109) This issignificantly different (p = 00074) from what we observed inour PD cohort of noncarriers (males have an anticipateddisease onset 5759 plusmn 108 vs 5934 plusmn 105)

Looking at the different analyzed phenotypes (figure 2 A andB) a sex bias was observed in patients with PDD and DLBwith GBA variants with a larger fraction of male patients withdementia at all ages (percentage of males 49 64 and 79 inPD only PDD and DLB respectively p = 00037 figure 2B)Finally the distribution of variant frequencies in PDD is moresimilar to that of DLB than that of the PD-only group(figure 2C)

Age-dependent penetrance and relative riskin GBA-PDThe age-dependent penetrance was estimated using pop-ulation allele frequencies (from our cohort of 7755 Italian

Table 1 GBA mutation carriers according to clinical diagnosis

Diagnosis Subjects n

GBA carriers n ()

Total pE326K pT369M pN370S pL444P

PD 3691 248 (67) 61 (16) 49 (13) 76 (21) 62 (17)

PDD 807 90 (112) 26 (32) 11 (14) 26 (32) 27 (33)

DLB 93 24 (258) 7 (74) 4 (42) 6 (63) 7 (74)

MSA 216 8 (37) 4 (19) 3 (14) 1 (05) mdash

CBD 56 3 (54) mdash 2 (36) 1 (18) mdash

FTD 26 2 (77) 1 (38) 1 (38) mdash mdash

PSP 199 9 (45) 2 (10) 3 (15) 3 (15) 1 (05)

Undefined PKS 323 14 (43) 6 (19) 2 (06) 4 (12) 2 (06)

Controls 7755 170 (22) 55 (07) 61 (08) 43 (06) 11 (01)

Abbreviations CBD = corticobasal degeneration DLB = dementia with Lewy body FTD = frontotemporal dementia MSA = multiple system atrophy PD =Parkinson disease PDD = Parkinson with dementia PKS = parkinsonism PSP = progressive supranuclear palsyInformation on a total of 308 heterozygous patients and 170 controls is reported (collectively or stratified according to the pE326K pT369M pN370S orpL444P) Data are reported also for patients with PD classified as PDD

4 Neurology Genetics | Volume 6 Number 6 | December 2020 NeurologyorgNG

Table 2 Allele frequency differences of GBA variants between cases and controls

Diagnosis Variants

Allele frequency Uncorrected analysis Adjusted analysis

Cases Controls OR (95 CI) p OR (95 CI) p

PD All 0034 0011 318 (261ndash387) 189E-33 365 (294ndash445) 125E-31

pE326K 00083 00035 234 (163ndash337) 262E-06 248 (167ndash370) 799ndash06

pT369M 00066 00039 169 (116ndash247) 00057 156 (102ndash239) 0039

pN370S 0010 00028 374 (257ndash544) 138E-13 459 (307ndash686) 122E-13

pL444P 00084 000071 1193 (628ndash2267) 513E-22 1563 (804ndash3037) 497E-16

PDD All 0056 0011 539 (415ndash700) 369E-45 589 (438ndash791) 570E-32

pE326K 0016 00035 460 (288ndash735) 255E-12 480 (287ndash802) 212E-09

pT369M 00068 00039 1738 (091ndash331) 0089 116 (051ndash261) 073

pN370S 0016 00028 589 (361ndash961) 833E-16 668 (386ndash1154) 103E-11

pL444P 0017 000071 2397 (1187ndash4842) 986E-39 2957 (1407ndash6213) 386E-19

DLB All 013 0011 1353 (859ndash2132) 387E-48 1753 (999ndash3076) 174E-23

pE326K 0038 00035 1099 (494ndash2446) 174E-13 1224 (495ndash3024) 571E-08

pT369M 0022 00039 557 (201ndash1547) 000021 402 (116ndash1398) 0029

pN370S 0032 00028 1199 (504ndash2852) 774E-13 163 (574ndash4628) 157E-07

pL444P 0038 000071 5510 (2112ndash1438) 167E-49 1027 (3138ndash3361) 191E-14

MSA All 0019 0011 172 (084ndash353) 013 181 (087ndash374) 011

pE326K 0003 00035 263 (095ndash728) 0054 266 (095ndash749) 0064

pT369M 00069 00039 177 (055ndash567) 033 186 (058ndash604) 030

pN370S 00023 00028 084 (012ndash607) 086 088 (012ndash651) 090

pL444P mdash 000071 mdash mdash mdash mdash

FTD All 0039 0011 365 (088ndash1514) 0056 385 (090ndash1654) 0070

pE326K 0019 00035 551 (075ndash4058) 0059 558 (074ndash4234) 0097

pT369M 0019 00039 497 (068ndash3651) 0080 512 (068ndash3871) 011

pN370S mdash 00028 mdash mdash mdash mdash

pL444P mdash 000071 mdash mdash mdash mdash

CBD All 0027 0011 251 (079ndash799) 011 188 (045ndash787) 039

pE326K mdash 00035 mdash mdash mdash mdash

pT369M 0018 00039 461 (111ndash1907) 0021 259 (035ndash1929) 035

pN370S 00089 00028 324 (044ndash2374) 022 371 (049ndash2828) 021

pL444P mdash 000071 mdash mdash mdash mdash

PSP All 0023 0011 211 (107ndash416) 0027 232 (115ndash443) 0019

pE326K 00050 00035 145 (035ndash584) 063 147 (035ndash620) 060

pT369M 00075 00039 192 (060ndash616) 026 198 (060ndash651) 026

pN370S 00075 00028 273 (084ndash884) 0080 325 (096ndash1101) 0059

pL444P 00025 000071 355 (046ndash2756) 016 522 (064ndash4264) 012

Undefined PKS All 0022 0011 202 (117ndash351) 0011 212 (117ndash384) 0014

pE326K 00093 00035 263 (113ndash614) 0020 277(114ndash674) 0024

Continued

NeurologyorgNG Neurology Genetics | Volume 6 Number 6 | December 2020 5

controls) which are in line with those observed amongSouthern Europeans (table 1 figure e-1 linkslwwcomNXGA327) as well as age-dependent PD prevalence esti-mates for the Italian population28 The most striking resultsconcern pL444P which reaches a penetrance of 1507above 75 years All the other 3 variants show a comparablepenetrance ranging from 215 to 472 for pT369M andpN370S respectively (gt75 years table 4)

Age-dependent increasing penetrance is paralleled by de-creasing relative risk estimates for all variants In particularpL444P shows a peak in relative risk at an age comprisedbetween 35 and 44 years (RR = 751 95 CI = 591ndash954)whereas above 75 years the relative risk for this mutation isnull (table 4)

DiscussionAlthough there is consensus on the fraction of patients withPD bearing GBA defects (5ndash10)6 the effect of eachvariant changes according to different authors with pE326Kbeing reported as having the highest or the lowest ORamong the 4 most frequent PD-associatedGBA variants2930

Moreover pL444P was reported to confer an OR rangingfrom 26218 to 81730 This is likely due to population-specific effects31 and technical difficulties in genotypingthese variants which is also present in the GBAP1 pseudo-gene In this frame we tackled a large monocentric study onthe penetrance incidence and association with dementia ofthe 4 most frequent GBA variants in PD The main strengthsof this work are the meticulous clinical characterization ofpatients all followed in a single center the ethnical matchbetween cases and controls and the accurate genotyping ofvariants which were all analyzed by allele-specific PCR andconfirmed by Sanger sequencing Moreover recent literatureon the predisposing role ofGBA variants on PD suggests thatnonndashGD-causing variants (pE326K and pT369M) mightbe associated with a second genetic defect pointing to anoligogenic model for PD predisposition1718 further com-plicating the interpretation of the specific role of singlevariants We therefore chose to focus as much as possible onsingle-allele carriers excluding from the analysis all GBA-

positive patients and controls homozygous or compoundheterozygous for one of the analyzed GBA variants as well asall carriers of PD-causing mutations in known PD genesHaving taken these considerations into account we couldcalculate an unbiased estimate of the true risk penetrancecumulative incidence and sex-specific effects associated withthe most frequent PD-related variants in GBA

When GBA variants are ranked according to the predictedimpact on enzyme activity13ndash15 a gradient in the conferredrisk is evident not only for PD but also for PDD and DLB forpT369M pN370S and pL444P (figure e-2 linkslwwcomNXGA327) Conversely pE326K appears to have a moresevere impact than expected Indeed in our study pE326Kand pL444P are the 2 variants with the highest OR for PDD(OR = 480 95 CI = 287ndash802 p = 21210minus9 and OR =2957 95 CI = 1407ndash6213 p = 38610minus19 respectively)This is further confirmed by an OR for pE326K of 1224(95 CI = 495ndash3024 p = 57110minus8) for DLB and by therecently reported frequency of pE326K in a large study onDLB among Caucasians (39)32 almost overlapping withthe pE326K allele frequency found in our DLB cohort(38) However in their study Orme and colleagues32 failedto report the frequency of pL444P the most strongly asso-ciated with DLB in our cohort (OR = 1027 95 CI =3138ndash3361 p = 19110minus14)

Reliable penetrance estimates are hugely important particu-larly for individuals undergoing predictive genetic testing andin the view of the clinical trials on GBA-PD Family-basedmethods to estimate penetrance have been used but theyhave ascertainment bias A practical alternative to estimatepenetrance is to use orthogonal population-based methods27

To this aim a good estimate of allele frequency even in thecase of uncommon variants is crucial However despite thevast amount of information in databases (Exome AggregationConsortium ExAC and Genome Aggregation DatabasegnomAD) these data are scarcely available in case-controlmatched populations The situation is even more complicatedfor GBA as the presence of the GBAP1 pseudogene compli-cates genotyping thus reducing the availability of reliableinformation from exome data This is why we decided togather a large number of ethnically matched controls to have a

Table 2 Allele frequency differences of GBA variants between cases and controls (continued)

Diagnosis Variants

Allele frequency Uncorrected analysis Adjusted analysis

Cases Controls OR (95 CI) p OR (95 CI) p

pT369M 00031 00039 079 (019ndash322) 074 082 (019ndash344) 078

pN370S 00062 00028 224 (080ndash626) 011 207 (061ndash701) 024

pL444P 00031 000071 438 (097ndash1978) 0036 723 (150ndash349) 0014

Abbreviations CI = confidence interval CBD = corticobasal degeneration DLB = dementia with Lewy body PD = Parkinson disease MSA = multiple systematrophy OR = odds ratio PDD = Parkinson with dementia PKS = parkinsonism PSP = progressive supranuclear palsyp Values lt 005 and the corresponding ORs and 95 CI are indicated in bold In the adjusted analyses values are corrected for age and sex

6 Neurology Genetics | Volume 6 Number 6 | December 2020 NeurologyorgNG

Table 3 Analysis of demographic and general clinical features of 3691 patients with PD according to GBA mutation status

Feature Noncarriers (n = 3433)

All variants (n = 248)

OR (95 CI)adj (95 CI)

padj p

Sex (Male [n]) 6030 (2076) 5444 (135) 079 (061ndash102) 0069

Age at onset (mean plusmn SD) 5824 plusmn 1078 5435 plusmn 1096 139 (124-156)140 (124ndash158)

248E-08294E-08

PD family history ( [n]) 1938 (653) 2869 (70) 167 (125-224)156 (116ndash210)

00005100030

Disease duration (mean plusmn SD) 1355 plusmn 689 1405 plusmn 729 101 (099ndash103) 029

Dementia ( [n]) 2082 (717) 3629 (90) 217 (165-284)225 (171ndash2977)

245-08901ndash09

Feature

pE326K (n = 61) pT369M (n = 49) pN370S (n = 76) pL444P (n = 62)

OR (95 CI) p OR (95 CI) pOR (95 CI)adjOR (95CI)

padj p

OR (95 CI)adjOR (95CI)

padj p

Sex (Male [n]) 6230(38)

109(064ndash181)

0762 4490(22)

053(030ndash094)

00305 5395(41)

077 (049ndash122) 0258 5484(34)

080 (048ndash131) 0378

Age at onset (mean plusmn SD) 5675 plusmn1188

110(088ndash139)

0396 5622 plusmn1065

120(093ndash156)

0149 5392 plusmn1048

147(120ndash181)149 (122ndash185)

00001850000108

5103 plusmn987

185(147ndash232)181 (144ndash232)

174E-07466E-07

PD family history ( [n]) 2000(12)

104(055ndash197)

0898 2766(13)

159(084ndash304)

0156 2800(21)

162 (097ndash271) 00637 3871(24)

263(157ndash442)230 (136ndash389)

0000248000199

Disease duration (mean plusmnSD)

1386 plusmn680

102(098ndash1059)

0404 1452 plusmn697

102(098ndash106)

0339 1369 plusmn733

100 (097ndash104) 0865 1386 plusmn680

101 (097ndash105) 0736

Dementia ( [n]) 4355(27)

283(169ndash472)

750E-05 2245(11)

110(056ndash216)

078 3421(26)

198(122ndash320)208 (128ndash337)

000545000296

4355(27)

293(176ndash488)314 (187ndash525)

337E-05138E-05

Abbreviations CI = confidence interval OR = odds ratio PD = Parkinson diseaseValues are expressed as (and counts) or as mean plusmn SD Significant values are in boldIf significantly different in the unadjusted analysis variables were also analyzed in amultivariate context throughmultivariate logistic regression (adjusting for all covariates that resulted significantly different between carriersand noncarriers in the crude analysis) with the presence of GBA variants as dichotomous response variable

Neurolo

gyorgN

GNeurologyG

enetics|

Volume6N

umber

6|

Decem

ber

20207

reliable estimate of the Italian population frequencies in in-dividuals without PD for the 4 most common PD-associatedGBA variants

Comparing the penetrance of the 4 variants pL444P stands outas the mutation conferring the higher PD risk with a lifelong (at75 years) penetrance of 15 This is not far from the penetrancereported for the LRRK2 pG2019S mutation among AshkenaziJews (26 to age 80 years)33 Of interest the risk for PD wassimilar among male and female pG2019S carriers whereas it isquite higher inmales among noncarriers33 This is similar to whatwe observed in GBA carriers of pN370S pL444P and evenmore so of pT369M for which the OR of being female whenaffected and carrier of the variant reached statistical significance(OR = 189 95 CI = 106ndash333) Taken together these ob-servations suggest that in mutation carriers the contribution of adeleterious variant in LRRK2 or inGBA is the major cause of PDin both sexes overriding the differential environmental exposureandor hormonal influences that have been proposed to explainthe higher prevalence of PD in men34 This evidence is furthercorroborated by the finding that the mean age at onset in femalecarriers of aGBA variant was almost 2 years anticipated compared

withmen at difference with what reported in sporadic PD wherefemales have a comparable or even later onset4

We are aware of the potential limitations of our study whichcannot rule out the possibility that rare deleterious variants inGBA or other mutations in PD-causing genes might have animpact on variant penetrance However our study design isfocused to estimate the disease penetrance which can only bereliably calculated on frequent variants In addition consid-ering the ethnic-specific differences in genetic variation fre-quencies we chose to limit our analysis to Italians whohowever show variant frequencies similar to those ofSouthern Europeans and Caucasians so that our calculatedage-related penetrance and cumulative incidence might bereasonably extended to an average Caucasian Finally patientsreferred to a tertiary care center are likely to be enriched insevere cases this selection bias is counterbalanced by the highaccuracy of clinical diagnoses

The complex nature of PD etiology its long presymptomaticphase and its heterogeneous clinical manifestations presentmajor challenges for formulating successful personalized

Figure 1 Overall and sex-specific age-related cumulative incidences of PD according to GBA genotype

(A) Vinicunca plot showing the cumulative incidence of PD inGBA-mutated patients Each colored area represents the individual contribution of each analyzedGBA variant (B) Cumulative incidence of PD according to genotype (heterozygosity for pE326K pT369M pN370S and pL444P) and sex (the number ofmaleand female individuals is presented in table 3) In all cases cumulative incidence is reported aspercentage (Y axis) and age reported in 20-year intervals on theX axis PD = Parkinson disease

8 Neurology Genetics | Volume 6 Number 6 | December 2020 NeurologyorgNG

Figure 2 Age-related cumulative incidence of PD PDD and DLB in male and female carriers of a GBA variant

Vinicunca plot showing the cumulative incidence of theGBA-associated synucleinopathies inGBA-mutated patients (A) and according to diagnosis and sex (B)In all cases cumulative incidence is reported as percentage (Y axis) and age reported in 20-year intervals on the X axis (C) Pie charts report the distribution ofthe 4 analyzedGBA variants in patients with respect to clinical diagnosis p Values were calculated using the χ2 test DLB = dementia with Lewy body PD only =Parkinson disease without dementia PDD = Parkinson with dementia

Table 4 Overall variant-specific penetrance and relative risk of GBA-PD

Age (y)

Penetrance (95 CI)

All variants pE326K pT369M pN370S pL444P

35 0010 (00034ndash0027) 00055 (00007ndash0041) 00060 (000070ndash0041) 0017 (00025ndash010) 0033 (00016ndash065)

45 0034 (0021ndash0054) 0021 (00079ndash0055) 0018 (00070ndash0047) 0034 (0014ndash0082) 020 (0055ndash074)

55 0089 (0061ndash013) 0062 (0029ndash013) 0038 (0016ndash0086) 0090 (0046ndash018) 048 (016ndash14)

65 039 (028ndash054) 028 (015ndash053) 023 (012ndash042) 044 (024ndash080) 140 (055ndash360)

75 151 (110ndash200) 113 (065ndash190) 084 (048ndash150) 190 (110ndash330) 542 (230ndash1300)

ge75 390 (300ndash510) 296 (180ndash490) 215 (130ndash360) 472 (280ndash790) 1507 (650ndash3400)

Relative risk (95 CI)

le34 383 (222ndash661) 236 (070ndash799) 236 (070ndash799) 485 (234ndash1006) 633 (276ndash1453)

35ndash44 372 (288ndash480) 253 (139ndash460) 222 (121ndash409) 317 (195ndash517) 751(591ndash954)

45ndash54 156 (141ndash173) 154 (127ndash186) 150 (111ndash204) 142 (117ndash173) 168 (151ndash187)

55ndash64 133 (123ndash143) 125 (105ndash148) 123 (105ndash145) 145 (141ndash150) 128 (109ndash152)

65ndash74 154 (126ndash189) 147 (103ndash208) 107 (064ndash178) 193 (142ndash264) 219 (160ndash300)

ge75 094 (037ndash239) 181 (067ndash494) 065 (01ndash425) 0 0

Abbreviations CI = confidence interval PD = Parkinson diseaseFor penetrance calculations prevalence for the Italian population at different ages has been obtained from Pupillo et al28

NeurologyorgNG Neurology Genetics | Volume 6 Number 6 | December 2020 9

disease-modifying therapies Hence the identification ofGBAmutation carriers combined with a thorough understanding ofthe risk conferred by the different genetic defects alone and inthe context of other genetic and nongenetic risk factors ismandatory to select subjects who could benefit from GBA-targeted therapies currently in clinical trials

AcknowledgmentLuigi Onorati and Francesca Cancellieri are acknowledged fortheir invaluable assistance and technical support The authorsthank (in alphabetical order) Drs Margherita CanesiFrancesca Del Sorbo Claudio B Mariani Nicoletta MeucciGiorgio Sacilotto Silvana Tesei and Michela Zini of theParkinson Institute for patient referral Finally the authorsthank all patients and their relatives for their contributionThe DNA samples from patients were obtained from theParkinson Institute Biobank (parkinsonitbiobanca) mem-ber of the Telethon Network of Genetic Biobanks (bio-banknetworktelethonit) and supported by FondazioneGrigioni per il Morbo di Parkinson Milan Italy

Study fundingThis work was supported by PRIN (Programmi di RicercaScientifica di Rilevante Interesse Nazionale Grant no2017228L3J program coordinator S Duga) and by theFondazione Grigioni per il Morbo di Parkinson Milan ItalyL Straniero was supported by fellowships from the Fonda-zione Veronesi and Fondazione Grigioni per il Morbo diParkinson

DisclosureThe authors report no disclosures Go to NeurologyorgNGfor full disclosures

Publication historyReceived by Neurology Genetics July 6 2020 Accepted in final formAugust 31 2020

References1 Lees AJ Hardy J Revesz T Parkinsonrsquos disease Lancet 20093732055ndash20662 Mendoza-Velasquez JJ Flores-Vazquez JF Barron-Velazquez E Sosa-Ortiz AL Illi-

gens BW Siepmann T Autonomic dysfunction in α-synucleinopathies Front Neurol201910363

3 Hanagasi HA Tufekcioglu Z Emre M Dementia in Parkinsonrsquos disease J Neurol Sci201737426ndash31

4 Georgiev D Hamberg K Hariz M Forsgren L Hariz GM Gender differences inParkinsonrsquos disease a clinical perspective Acta Neurol Scand 2017136570ndash584

5 Sidransky E Nalls MA Aasly JO et al Multicenter analysis of glucocerebrosidasemutations in Parkinsonrsquos disease N Engl J Med 20093611651ndash1661

6 Migdalska-Richards A Schapira AH The relationship between glucocerebrosidasemutations and Parkinson disease J Neurochem 2016139(suppl 1)77ndash90

7 Cilia R Tunesi S Marotta G et al Survival and dementia in GBA-associated Par-kinsonrsquos disease the mutation matters Ann Neurol 201680662ndash673

8 Iwaki H Blauwendraat C Leonard HL et al Genetic risk of Parkinson disease andprogression an analysis of 13 longitudinal cohorts Neurol Genet 20195e348

9 Stoker TB Camacho M Winder-Rhodes S et al Impact of GBA1 variants on long-term clinical progression and mortality in incident Parkinsonrsquos disease J NeurolNeurosurg Psychiatry 202091695ndash702

10 Nalls MA Duran R Lopez G et al A multicenter study of glucocerebrosidase mu-tations in dementia with Lewy bodies JAMA Neurol 201370727ndash735

Appendix Authors

Name Location Contribution

LetiziaStranieroPhD

Humanitas UniversityMilan Italy

Designed andconceptualized the studyanalyzed the data anddrafted the manuscript forintellectual content

RosannaAsselta PhD

Humanitas UniversityMilan Italy

Was in charge of thestatistical analysis of thedata and revised themanuscript for intellectualcontent

SalvatoreBonvegnaMD

Fondazione IRCCS IstitutoNeurologico ldquoCarlo BestardquoMilan Italy

Interpreted the data andrevised the manuscript forintellectual content

Appendix (continued)

Name Location Contribution

ValeriaRimoldiPhD

Humanitas UniversityMilan Italy

Performed genotyping anddrew the figures

GiadaMelistaccioMSc

Humanitas UniversityMilan Italy

Performed genotyping

Giulia SoldaPhD

Humanitas UniversityMilan Italy

Interpreted the data andrevised the manuscript forintellectual content

MassimoAureli PhD

Department of MedicalBiotechnology andTranslational MedicineUniversity of Milan MilanItaly

Participated in themanuscript writing andcritical reviewing

MatteoDella PortaMD

Humanitas UniversityMilan Italy

Major role in control cohortcollection

Ugo LuccaMSc

Laboratory of GeriatricNeuropsychiatry Istituto diRicerche FarmacologicheMario Negri IRCCS MilanItaly

Major role in control cohortcollection

Alessio DiFonzo MDPhD

IRCCS Foundation CarsquoGranda OspedaleMaggiore Policlinico MilanItaly

Participated in themanuscript writing andcritical reviewing

AnnaZecchinelliMD

Parkinson Institute ASSTldquoGaetano Pini-CTOrdquo MilanItaly

Obtained samples andclinical data from patients

GianniPezzoli MD

Fondazione Grigioni per ilMorbo di Parkinson MilanItaly

Interpreted the data andrevised themanuscript andcoordinated the study

RobertoCilia MD

Fondazione IRCCS IstitutoNeurologico ldquoCarlo BestardquoMilan Italy

Obtained samples andclinical data from patientsand revised themanuscript

StefanoDuga PhD

Humanitas UniversityMilan Italy

Interpreted the datarevised the manuscriptand coordinated the entirestudy

10 Neurology Genetics | Volume 6 Number 6 | December 2020 NeurologyorgNG

11 Asselta R Rimoldi V Siri C et al Glucocerebrosidase mutations in primary parkin-sonism Parkinsonism Relat Disord 2014201215ndash1220

12 Malek N Weil RS Bresner C et al Features of GBA-associated Parkinsonrsquos disease atpresentation in the UK tracking Parkinsonrsquos study J Neurol Neurosurg Psychiatry201889702ndash709

13 Malini E Grossi S Deganuto M et al Functional analysis of 11 novel GBA alleles EurJ Hum Genet 201422511ndash516

14 Alcalay RN Levy OA Waters CC et al Glucocerebrosidase activity in Parkinsonrsquosdisease with and without GBA mutations Brain 2015138(pt 9)2648ndash2658

15 Sanchez-Martinez A Beavan M Gegg ME Chau KY Whitworth AJ Schapira AHParkinson disease-linked GBA mutation effects reversed by molecular chaperones inhuman cell and fly models Sci Rep 2016631380

16 Berge-Seidl V Pihlstroslashm L Maple-Groslashdem J et al The GBA variant E326K isassociated with Parkinsonrsquos disease and explains a genome-wide association signalNeurosci Lett 201765848ndash52

17 Mallett V Ross JP Alcalay RN et al GBA pT369M substitution in Parkinson diseasepolymorphism or association A meta-analysis Neurol Genet 20162e104

18 Goldstein O Gana-Weisz M Cohen-Avinoam Det al Revisiting the non-Gaucher-GBA-E326K carrier state is it sufficient to increase Parkinsonrsquos disease risk MolGenet Metab 2019128470ndash475

19 Mullin S Smith L Lee K et al Ambroxol for the treatment of patients with Parkinsondisease with and without glucocerebrosidase gene mutations a nonrandomizednoncontrolled trial JAMA Neurol 202077427ndash434

20 Hughes AJ Daniel SE Kilford L Lees AJ Accuracy of clinical diagnosis of idiopathicParkinsonrsquos disease a clinico-pathological study of 100 cases J Neurol NeurosurgPsychiatry 199255181ndash184

21 McKeith IG Dickson DW Lowe J et al Diagnosis and management of dementia withLewy bodies third report of the DLB Consortium Neurology 2005651863ndash1872

22 Barton B Grabli D Bernard B et al Clinical validation of Movement Disorder Society-recommended diagnostic criteria for Parkinsonrsquos disease with dementia Mov Disord201227248ndash253

23 Do R Stitziel NO Won HH et al Exome sequencing identifies rare LDLR andAPOA5 alleles conferring risk for myocardial infarction Nature 2015518102ndash106

24 Riva E Tettamanti M Mosconi P et al Association of mild anemia with hospitali-zation and mortality in the elderly the Health and Anemia population-based studyHaematologica 20099422ndash28

25 Tettamanti M Lucca U Gandini F et al Prevalence incidence and types of mildanemia in the elderly the ldquoHealth and Anemiardquo population-based study Haemato-logica 2010951849ndash1856

26 Straniero L Rimoldi V Melistaccio G et al A rapid and low-cost test for screening themost common Parkinsonrsquos disease-related GBA variants Parkinsonism Relat Disord202080138ndash141

27 Minikel EV Vallabh SM Lek M et al Quantifying prion disease penetrance usinglarge population control cohorts Sci Transl Med 20168322ra9

28 Pupillo E Cricelli C Mazzoleni F et al Epidemiology of Parkinsonrsquos disease apopulation-based study in primary care in Italy Neuroepidemiology 20164738ndash45

29 Blauwendraat C Bras JM Nalls MA Lewis PA Hernandez DG Singleton AB Codingvariation in GBA explains the majority of the SYT11-GBA Parkinsonrsquos disease GWASlocus Mov Disord 2018331821ndash1823

30 Ran C Brodin L Forsgren L et al Strong association between glucocere-brosidase mutations and Parkinsonrsquos disease in Sweden Neurobiol Aging 201645212

31 Zhang Y Shu L Sun Q et al Integrated genetic analysis of racial differences ofcommon GBA variants in Parkinsonrsquos disease a meta-analysis Front Mol Neurosci20181143

32 Orme T Hernandez D Ross OA et al Analysis of neurodegenerative disease-causing genes in dementia with Lewy bodies Acta Neuropathol Commun 202085

33 Marder K Wang Y Alcalay RN et al Age-specific penetrance of LRRK2 G2019S inthe Michael J Fox Ashkenazi Jewish LRRK2 Consortium Neurology 20158589ndash95

34 Marras C Saunders-Pullman R The complexities of hormonal influences andrisk of Parkinsonrsquos disease Mov Disord 201429845ndash848

NeurologyorgNG Neurology Genetics | Volume 6 Number 6 | December 2020 11

DOI 101212NXG000000000000052320206 Neurol Genet

Letizia Straniero Rosanna Asselta Salvatore Bonvegna et al GBA-PD

The SPID-GBA study Sex distribution Penetrance Incidence and Dementia in

This information is current as of October 20 2020

reserved Online ISSN 2376-7839Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology All rightsan open-access online-only continuous publication journal Copyright Copyright copy 2020 The Author(s)

is an official journal of the American Academy of Neurology Published since April 2015 it isNeurol Genet

ServicesUpdated Information amp

httpngneurologyorgcontent66e523fullhtmlincluding high resolution figures can be found at

References httpngneurologyorgcontent66e523fullhtmlref-list-1

This article cites 34 articles 10 of which you can access for free at

Subspecialty Collections

httpngneurologyorgcgicollectionprevalence_studiesPrevalence studies

mhttpngneurologyorgcgicollectionparkinsons_disease_parkinsonisParkinsons diseaseParkinsonism

ntiahttpngneurologyorgcgicollectionparkinsons_disease_with_demeParkinsons disease with dementia

httpngneurologyorgcgicollectionincidence_studiesIncidence studiesfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpngneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpngneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

reserved Online ISSN 2376-7839Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology All rightsan open-access online-only continuous publication journal Copyright Copyright copy 2020 The Author(s)

is an official journal of the American Academy of Neurology Published since April 2015 it isNeurol Genet

ServicesUpdated Information amp

httpngneurologyorgcontent66e523fullhtmlincluding high resolution figures can be found at

References httpngneurologyorgcontent66e523fullhtmlref-list-1

This article cites 34 articles 10 of which you can access for free at

Subspecialty Collections

httpngneurologyorgcgicollectionprevalence_studiesPrevalence studies

mhttpngneurologyorgcgicollectionparkinsons_disease_parkinsonisParkinsons diseaseParkinsonism

ntiahttpngneurologyorgcgicollectionparkinsons_disease_with_demeParkinsons disease with dementia

httpngneurologyorgcgicollectionincidence_studiesIncidence studiesfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpngneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpngneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

reserved Online ISSN 2376-7839Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology All rightsan open-access online-only continuous publication journal Copyright Copyright copy 2020 The Author(s)

is an official journal of the American Academy of Neurology Published since April 2015 it isNeurol Genet