Disease modiﬁcation and biomarker development in …A fundamental question in advancing Parkinson disease (PD) research is whether it represents one ... lecular and genetic discoveries,

VIEWS amp REVIEWS

Disease modification and biomarkerdevelopment in Parkinson diseaseRevision or reconstruction

Alberto J Espay MD MSc Lorraine V Kalia MD PhD Ziv Gan-Or MD PhD

Caroline H Williams-Gray BMBCh MRCP PhD Philippe L Bedard MD FRCPC Steven M Rowe MD MSPH

Francesca Morgante MD PhD Alfonso Fasano MD PhD Benjamin Stecher BA Marcelo A Kauffman MD PhD

Matthew J Farrer PhD Chris S Coffey PhD Michael A Schwarzschild MD PhD Todd Sherer PhD

Ronald B Postuma MD MSc Antonio P Strafella MD PhD FRCPC Andrew B Singleton PhD

Roger A Barker MBBS MRCP PhD Karl Kieburtz MD MPH C Warren Olanow MD FRCPC FRCP(HON)

Andres Lozano MD PhD Jeffrey H Kordower PhD Jesse M Cedarbaum MD Patrik Brundin MD PhD

David G Standaert MD PhD and Anthony E Lang MD FRCPC

Neurologyreg 202094481-494 doi101212WNL0000000000009107

Correspondence

Dr Espay

albertoespayucedu

AbstractA fundamental question in advancing Parkinson disease (PD) research is whether it represents onedisorder ormany Does each genetic PD inform a common pathobiology or represent a unique entityDo the similarities between genetic and idiopathic forms of PD outweigh the differences If aggregatesof α-synuclein in Lewy bodies and Lewy neurites are present in most (α-synucleinopathies) are theyalso etiopathogenically significant in each (α-synuclein pathogenesis) Does it matter that postmortemstudies in PD have demonstrated that mixed protein-aggregate pathology is the rule and pureα-synucleinopathy the exception Should we continue to pursue convergent biomarkers that arerepresentative of the diverse whole of PD or subtype-specific divergent biomarkers present in somebut absent inmost Have clinical trials that failed to demonstrate efficacy of putative disease-modifyinginterventions been true failures (shortcomings of the hypotheses which should be rejected) or falsefailures (shortcomings of the trials hypotheses should be preserved) Each of these questions reflectsa nosologic struggle between the lumperrsquos clinicopathologic model that embraces heterogeneity of onedisease and the splitterrsquos focus on a pathobiology-specific set of diseases Most important even if PD isnot a single disorder can advances in biomarkers and diseasemodification be revised to concentrate onpathologic commonalities in large clinically defined populations Or should our efforts be recon-structed to focus on smaller subgroups of patients distinguished by well-defined molecular charac-teristics regardless of their phenotypic classification Will our clinical trial constructs be revised totarget larger and earlier possibly even prodromal cohortsOr should our trials efforts be reconstructedto target smaller butmolecularly defined presymptomatic or postsymptomatic cohorts At theKrembilKnowledge Gaps in Parkinsonrsquos Disease Symposium the tentative answers to these questions werediscussed informed by the failures and successes of the fields of breast cancer and cystic fibrosis

RELATED ARTICLE

EditorialConsiderations for trialsof disease modifyingtreatments in Parkinsondisease

Page 467

From the Department of Neurology (AJE) James J and Joan A Gardner Family Center for Parkinsonrsquos Disease and Movement Disorders University of Cincinnati OH Edmond J SafraProgram inParkinsonrsquos Disease (LVK) Krembil Research Institute TorontoWesternHospital Ontario Canada Department of Neurology andNeurosurgery (ZG-O)Montreal NeurologicalInstitute andDepartmentofHumanGenetics (ZG-O)McGillUniversityMontrealQuebec CanadaDepartmentofClinicalNeurosciences (CHW-G) JohnvanGeest Centre forBrainRepairUniversity of Cambridge UK Division of Medical Oncology and Hematology (PLB) Princess Margaret Cancer Centre University Health Network and Department of Medicine (PLB)University of Toronto Ontario Canada Departments of Medicine (SMR) Pediatrics (SMR) Cell Developmental and Integrative Biology (SMR) and Neurology (DGS) University ofAlabama at Birmingham Department of Clinical and Experimental Medicine (FM) University of Messina Italy Institute of Molecular and Clinical Sciences (FM) St Georgersquos University ofLondon UK Edmond J Safra Program in Parkinsonrsquos Disease (AF APS AEL) Morton and Gloria Shulman Movement Disorders Clinic Toronto Western Hospital University of TorontoKrembil Research Institute (AF APS AEL) Tomorrow Edition (BS) Toronto Ontario Canada Consultorio y Laboratorio de Neurogenetica (MAK) Centro Universitario de NeurologıaldquoJose Marıa Ramos Mejıardquo y Division Neurologıa Hospital JM Ramos Mejıa Facultad de Medicina UBA Programa de Medicina de Precision y Genomica Clinica (MAK) Instituto deInvestigaciones enMedicina Traslacional Facultad deCiencias Biomedicas Universidad Austral-CONICET Buenos Aires Argentina Clinical Genomics Program (MJF) Norman Fixel Institutefor Neurological Diseases McKnight Brain Institute University of Florida Clinical and Translational Science Institute Gainesville Clinical Trials Statistical amp Data Management Center (CSC)University of Iowa College of Public Health Iowa City Department of Neurology (MAS) Massachusetts General Hospital Boston Michael J Fox Foundation for Parkinsonrsquos Research (TS)New York City NY Department of Neurology (RBP) Montreal General Hospital Quebec Research Imaging Centre (APS) Campbell Family Mental Health Research Institute Centre forAddiction and Mental Health University of Toronto Ontario Canada Laboratory of Neurogenetics (ABS) National Institute on Aging NIH Bethesda MD Department of Clinical Neuro-sciences (RAB) JohnvanGeest Centre for Brain RepairWT-MRCCambridge StemCell InstituteUniversity of CambridgeUK Clinical TrialsCoordinationCenter (KK) University ofRochesterMedical Center NY Department of Neurology and Neuroscience (CWO) Mount Sinai School of Medicine New York NY Clintrex LLC (CWO) Sarasota FL Division of Neurosurgery (AL)Krembil Neuroscience Institute Toronto Western Hospital Ontario Canada Department of Neurological Sciences ( JHK) Rush University Medical Center Chicago IL Coeruleus ClinicalSciences ( JMC) Woodbridge CT and Center for Neurodegenerative Science (PB) Van Andel Institute Grand Rapids MI

Go to NeurologyorgN for full disclosures Funding information and disclosures deemed relevant by the authors if any are provided at the end of the article

Copyright copy 2020 American Academy of Neurology 481

Copyright copy 2020 American Academy of Neurology Unauthorized reproduction of this article is prohibited

Phenotypic refinements aside Parkinson disease (PD)remains largely defined as it was clinically gt200 years ago andpathologically 100 years ago Technical advances in the ex-ploration and analysis of pathobiological processes have beenused to validate rather than to question this clinicopathologicconstruct As a result the whole has become the sum of manyand increasing parts To order the increasing complexity intoan etiopathologic whole bioinformatic tools integrate mo-lecular and genetic discoveries connecting synaptic dysfunc-tion failed vesicular trafficking mitochondrial dysfunctionimpaired proteostasis α-synuclein (α-syn) pathobiology andneuroinflammation as pieces in a single puzzle1 In the ther-apeutic realm promising molecular interventions have beenadministered to patients in the absence of relevant measuresof target engagement that is bioassays of relevant molecularpathways rescued by such interventions Clinical trials aimedat slowing or halting progressive PD disability are undertakenwith the hypothesis that a molecular modification will affectcommon neurodegeneration pathways in most patients withthe clinical diagnosis of PD Consequently disease modifi-cation has remained elusive during the 4 decades in which wehave been trying with hypotheses and disease constructslargely unaltered

The fundamental question is whether PD is one disorder withvarying clinical manifestations influenced by a number of ge-netic and environmental factors that need to be revisedthereby allowing current approaches to understanding patho-genesis and advancing disease modification therapies orreconstructed to frame PD as a syndrome in which pathologiccommonalities are subordinate to unique etiologic or patho-physiologic entities thereby forcing new approaches to ad-vancing our understanding of the disease and developingsuccessful disease modification Understanding the answer tothis questionwill have far-reaching implications in terms of trialdesign disease modeling biomarker development therapeutictargeting and so on A key question is whether the negativeclinical trials of putative disease-modifying interventions havebeen true failures (failure of the hypothesis negative outcomesdeserve to be trusted) or false failures (failure of the trialmethodology hypotheses deserve another chance)

The Krembil Knowledge Gaps in PD Symposium held inToronto from April 24 to 26 2019 tackled this fundamentalconceptual challenge if PD is not a single disorder canadvances in biomarkers and disease modification be revised toconcentrate on commonalities of pathogenic mechanisms in

large populations (assuming biologic convergence) or dothey need to be reconstructed for application to smallersubgroups of patients distinguished by well-defined molec-ular characteristics with some mechanistic understanding ofetiopathogenesis and a relevant bioassay to measure it Thisreport summarizes the deliberations at that meeting andacknowledges the unresolved tension between the revisionistand the reconstructionist approach to PD and its treatment

Key knowledge gaps in 2019 Revisionor reconstructionAt the core the key question is whether PD is one disease withmany facets each contributing to explaining the whole ormany diseases sharing some elements with one another butwith distinctive pathogenic elements As a field we have livedwith this many-diseasesone-disease tension by reasoningthat both can be correct (figure 1) The road ahead for bio-marker development and clinical trial efforts toward diseasemodification depends on whether PD is a multinodal diseaseprocess connected by α-syn aggregation or a collection ofindependent pathologic processes that are largely if not en-tirely independent The former is the more alluring modelmolecular genetic forms of parkinsonism are consideredpieces of the same puzzle driving the creation of key researchconstructs (table 1)

Knowledge gap 1 Does the heterogeneity of PDrepresent a disease spectrum with differentclinical subtypes and rates of progression ordistinct molecular entities each a separatedefinable diseaseThe former premise has been a dominant force Adopting theview that PD is a spectrum supports efforts at increasing ormerging ongoing cohorts to maximize scale and deployingimproved analytic processes to the accumulating data (re-vision) However the data-drivenndashgenerated subtypes thusfar created are not stable over time23 have not beenreproducible45 and do not predict unique patterns of ag-gregation of α-syn or other pathologies67 Moreover meas-urements of putative aggregated protein-based biomarkershave shown substantial signal overlap between patients andcontrols and between disease phenotypes and replication ofresults across cohorts is lacking8 Combining cohorts an-chored on clinical criteria and dependent on confirmation ofdopamine deficit is a strategy likely to run into the same

GlossaryAD = Alzheimer disease α-syn = α-synuclein CF = cystic fibrosis CFTR = CF transmembrane conductance regulatorDIAN-TU = Dominantly Inherited Alzheimer Network Trials Unit ER = estrogen receptor GCase = glucocerebrosidase HER2 =human epidermal growth factor receptor 2 I-SPY 2 = Investigation of Serial Studies to Predict Your Therapeutic Responsethrough Imaging and Molecular Analysis 2 IL = interleukin LRRK2 = leucine-rich repeat kinase 2 PD = Parkinson diseaseSNCA = synuclein SURE-PD3 = Study of Urate Elevation in Parkinsonrsquos Disease

482 Neurology | Volume 94 Number 11 | March 17 2020 NeurologyorgN


validation and replication difficulties because the problemmay not be one of statistical power but rather of underlyingassumptions As our biomarker armamentarium broadensa reversal of the developmental direction from biomarker tophenotype stratifying on the basis of biological clusters oftherapeutic significance rather than on phenotypic classifica-tion will require an overhaul of these efforts (reconstruction)

Knowledge gap 2 Will past negative clinicaltrials be replaced by future successes iftreatments are delivered to a larger and earliercohort or to smaller biologically definedpopulations prodromal or manifestingThe conventional view is that failure of clinical trials to meetefficacy endpoints has 3 primary causes shortcomings of theclinical trials themselves (eg wrong dosage or insufficientblood-brain barrier penetration inadequate target engagementinsensitive outcome measures too short a follow-up period)cohorts too advanced for the intervention of interest to beeffective (once symptoms appear it may be too late) and poortranslatability from animal models to human trials Proposalsfor revision are aimed at studying larger and earlier cohorts andcreating better animal models Proposals for reconstruction areaimed at recruiting more selective cohorts whether pre-symptomatic or postsymptomatic identified by biomarkersreflecting specific disease mechanisms ensuring that putativetherapies are tested only in those most likely to respond

Lessons from other fieldsOther fields of medicine have had their share of failures inbiomarker development and clinical trials of disease-modifyinginterventions Did they struggle with moving away from a sin-gle-disease model Potentially informative lessons from breastcancer and the monogenic disease cystic fibrosis (CF) werediscussed at the Krembil Knowledge Gaps Symposium

Cystic fibrosisDespite the discovery of the CF transmembrane conductanceregulator (CFTR) gene in 1989 treatments for CF were ex-clusively symptomatic targeting the secondary effects of dys-function in the CFTR protein involved in the secretion ofchloride and bicarbonate and the regulation of water in the lung

Table 1 Key research constructs Differences between revision and reconstruction

Researchconstruct Support for revision Support for reconstruction

Copathology PD is an α-synucleinopathy other pathologies must informPD variability α-syn causes or contributes to copathologies(eg seeding)

Mixed pathology is the rule pureα-syn pathology is the exceptioncopathologies are independent and not secondary to α-syn

Resilience Oligomeric species of α-syn are toxic even if they or theaggregates they become poorly correlate withneurodegeneration some regions have greater resistance tothe toxic effects of α-syn

Surviving occurs because of rather than despite α-syn aggregatesthey are absent in certain genetic parkinsonisms andoverexpressed in neurologically normal oldest-old individuals

Prodrome Clinical trials have failed because once symptoms appear itis too late to intervene earlier is always better

Prodromal PD is also a syndrome it can evolve into PD multiplesystem atrophy or dementia with Lewy bodies earlier is betteronly if molecularly subtyped

Phenoconversion Appearance of motor features define PD cluster of nonmotorsymptoms yields a high risk for meeting threshold for disease

There is no clear biological separation between prodrome anddisease or between nonmotor and motor PD

Abbreviations α-syn = α-synuclein PD = Parkinson diseaseBecause PD can diagnosed only be when certain motor features are present and its pathology should include only aggregated α-syn we have wrestled withPD risk andmodifiers through the constructs of prodrome (any nonmotor features appearing before cardinalmotor features) copathologymixed pathology(protein aggregates besides α-syn) resilience (surviving beyond a normal lifespan despite the presence of toxic α-syn in the brain) and phenoconversion (themoment when motor features meeting current criteria for diagnosis appear in individuals with prediagnostic nonmotor features)

Figure 1 Venn diagrams of the revision vs reconstructionmodels for PD

Each numbered circle represents a distinct biological entity or disease Lefta model warranting reconstruction While the diseases are related to one an-other they are mostly unique most biological elements are not sharedAccording to thismodel if α-synuclein (α-syn) is common to all ormost forms ofPD it cannot be pathogenic to each genetic Parkinson disease (PD) subtypesinform about those subtypes and not others Right a model best served byrevision All diseases havemore in commonwith one another their uniquenessis mostly theoretical given the large overlap This model does not discounta potential important role forα-syn inmany or all of the diseases a therapy thatmightwork for a genetic PDsubtypemaywork for others Themodel also allowsthe explanation that PDmay involvemultiple pathways and if we can fix one ofthem we could delay sporadic PD by years

NeurologyorgN Neurology | Volume 94 Number 11 | March 17 2020 483


sweat glands pancreas and gastrointestinal tract Depending onwhether each of the asymp1900 CFTR mutations impaired proteinproduction its tridimensional functional shape ion transportthrough the channel door or other aspects of the proteinfunction CF was split into 6 major classes The first disease-modifying medication ivacaftor was found to depend on thepresence of at least 1 G551D CFTR mutation assayed withepithelial cells in vitro (not animal models) which excluded 95of every 100 patients with CF A promising phase 2 trial (n = 39)9 of only these genotyped and assay-positive patients was fol-lowed by a successful phase 3 placebo-controlled trial (n = 84 ineach of 2 arms)10 The first Food and Drug Administrationapproval of a disease-modifying intervention for CF occurred 23years after the discovery of theCF gene Although it was effectivein only 4 to 5 of patients with CF it represented a proof ofconcept that genetically subtyped diseases could be modifiablewith class-specific therapies From this small but definite successthe pool of responders has been increased in selected geneticand class subtypes with the use of double- and triple-drugcombinations Critically the Therapeutic Development Pro-gram of the Cystic Fibrosis Foundation was instrumental inderisking pharmaceutical industry high-throughput drug dis-covery programs and early-phase clinical trials

Key developmental milestones in CFThe 3 developmental milestones were identification of mu-tation class-specific response to small molecules the use ofhighly predictive cell-based screening and biomarker plat-forms that avoided animal models of efficacy and multidrugcombination of potentiator and corrector therapies with po-tential for synergism that was evaluated from the outset TheCF experience provides an example of how even a single-genedisease was successfully tackled once it was divided intoidentifiable molecular subtypes mechanistically defined withcell-based biomarkers

Breast cancerAfter a randomized clinical trial in 1981 showed that radicalmastectomy yielded no benefits over quadrantectomy(breast-preserving surgery)1112 breast cancer began to besplit into clusters of different pathobiological entities In-dividual patient data from randomized trials allowed therecognition that the benefit of tamoxifen was limited only towomen with estrogen receptor (ER)ndashpositive tumorswhether premenopausal or postmenopausal13 Incrementalimprovements in survival required the development of otherdrug treatments for ER-positive breast cancers includingaromatase inhibitors selective ER degraders CDK46inhibitors and mammalian target of rapamycin inhibitors

In parallel identification of the human epidermal growthfactor receptor 2 (HER2) led to successful clinical trials ofHER2-targeted drug therapies such as trastuzumab (Her-ceptin) pertuzumab (Perjeta) and trastuzumab-DM1 (Kad-cyla) in the 20 of women with aggressive HER2-positivebreast cancers14 but only after post hoc analyses of initiallynegative trials revealed the presence of a potential responsive

population Breast cancer is now divided into severalpathology-molecular subtypes with distinct response totreatments relapse patterns and long-term survival15

Key developmental milestones in breast cancerThe 3 developmental milestones were recognition of themolecular basis for disease heterogeneity subtyping based onmolecular testing and histopathology and clinical trials withtargeted drug treatments based on molecular subtype Thisexperience demonstrates that a single negative trial sufficed toreject the hypothesis that breast cancer was a single but het-erogeneous disease

Conceptual roadblocks in the futureapplication of these lessons to PDShould pathology remain the gold standard forPD diagnosis and biomarker developmentConceptually PD is considered an α-synucleinopathy How-ever there is no dose-dependent correlation between Lewypathology and cell loss or clinical features In fact studies havefound either no16 or negative correlations17 between Braakstage and cell density in the substantia nigra FurthermoreLewy pathology is not necessary for nigral degeneration asdemonstrated by the appreciable proportion of leucine-richrepeat kinase 2 LRRK2- and Parkin-related cases with sub-stantia nigra neuron loss and motor deficits consistent withPD but without Lewy bodies18 Also challenging the primacyof α-syn pathology PD is usually associated with abnormalaccumulation of several different types of misfolded proteinsthe exclusive presence of α-syn pathology is the exception notthe rule19ndash21 Lastly some studies suggest that proteinaggregates may even be compensatory in a brain under mo-lecular stress and that with the plausible exception ofα-synndashrelated genetic PD subtypes (SNCA-PD) α-syn ag-gregation is neither necessary nor sufficient for neuro-degeneration or clinical parkinsonism22

Key roadblocksDevelopment of biomarkers of diagnostic susceptibility anddetection based on α-syn and other aggregated proteinsmay be insufficient for assessing disease trait or stateProtein aggregates could be related to convergent pro-tective strategies rather than specific pathogenetic mecha-nisms23 Definitive evidence of a direct causal relationshipbetween Lewy pathology and neurodegeneration inhumans is lacking

Should pathology be the target oftherapeutic developmentIf the extent of the contribution of α-syn aggregation to thepathogenesis of most subtypes of PD can be questioned thentargeting it in full length or through specific epitopes ofmaximum antigenicity prediction24 may be insufficient or ir-relevant outside of SNCA-PD Upregulation of α-syn (andother inflammatory markers such as Toll-like receptor 2



interleukin [IL]-6 and tumor necrosis factor) has beendocumented in mice or rats in response to infections orinflammation2526 Within this framework α-syn aggregationmay behave not unlike an acute-phase reactant increasing inconcentration during an inflammatory state

Aggregated β-amyloid has been targeted in Alzheimer disease(AD) trials over the past 2 decades Apart from recent in-triguing but unpublished results suggesting efficacy of adu-canumab all 32 phase 2 and 3 antiamyloid clinical trialsincluding 6 on prodromal cohorts have been negative despiteachieving expected target engagement that is demonstratingsignificant reduction in CSF- or PET-based aggregatedβ-amyloid Critically nearly 40 of the trials (largely but notexclusively of ɣ secretase and β-secretase inhibitors) havedemonstrated worsening of cognition and in half of theseacceleration of hippocampal atrophy in the treated groupcompared to placebo Fourteen clinical trial participants withpathology-confirmed AD who received β-amyloid42 immu-nization (AN1792 Elan Pharmaceuticals Dublin Ireland)progressed to moderate or severe dementia despite extensiveplaque removal in all regions examined at postmortem27

While β-amyloid aggregation and α-syn aggregation are struc-turally and functionally different targets 2 main arguments maypredict future failures of targeting α-synwith approaches similarto those used for β-amyloid both pathologies often co-occur asdownstream events in sporadic AD and PD and proteinopathyburden does not correlate with symptom severity In sum theseproteins represent pathologic markers but may not themselvesbe pathogenic (etiologic)

Key roadblocksα-Synndashbased biomarkers of disease could have limited valuefor the selection of patients for trials of putative disease-modifying interventions Targeting α-syn aggregation mayhave beneficial effects in selected subsegments of PD (egα-syn mutationsmultiplications) but could be futile orharmful in others

Should distribution of pathologydefine progressionIf the presence of α-syn aggregates defines PD its distributionhas to define PD progression The autopsy study quantifyingLewy pathology from individuals with and without PD thatled to the Braak hypothesis was anchored on cases with pa-thology in the dorsal motor nucleus of the vagus nerve andaccepted a single Lewy neurite as positive for any regionWhen considered together these brains suggested a patternwith a beginning in the peripheral nervous system specificallythe Meissner plexa in the gut and in the olfactory bulb and anend in the neocortex28 Although never proved it has beenassumed that there must be a sequential spread over timebetween these distribution patterns with support from animalstudies showing self-propagation of α-syn from neuron toneuron through permissive templating in a prion-likefashion2930 Stage 3 of the 6 hypothesized Braak stages of

α-syn progression represents the intersection between theconstructs of prodrome and clinically apparent diseaseaccording to current motor-centric diagnostic criteria

The many exceptions to the hypothesis of stereotypical pro-gression of pathology31 and the fact that symptomatic deficitsalign with neuronal degeneration rather than with α-syn ag-gregation32 render problematic the efforts to calculate diseaseprogression clinically (eg phenoconversion from prodrome todisease)33 or with longitudinal changes of CSF total α-synSimilarly clinical (eg rate of change in Movement DisordersSocietyndashsponsored Unified Parkinsonrsquos Disease Rating Scalescores) and imaging (eg dopamine transporter striatal bind-ing ratio signal loss) markers of progression have not detectedmeaningful effects on disease progression in trials conducted todateMoreover whatmight define progression on onemeasuredoes not on another Indeed there is poor concordance ofprogression between the Unified Parkinsonrsquos Disease RatingScale score and dopamine transporter striatal binding ratios34

Key roadblockWhile convergent dopamine- clinical- and α-synndashbased bio-markers may have utility in some populations biomarkers ofdivergence (present in some absent in most) will be neededto address the etiologic heterogeneity

Would testing treatments that failed in earlyPD be successful in prodromal PDBecause earlier treatment is better in many medical con-ditions the ldquotoo laterdquo argument has often been cited asexplaining past failures in neurodegenerative diseases If this iscorrect then retesting the interventions of prior trials inprodromal populations may yield the first success However ifPD subsumes many biological subtypes what is the re-lationship of prodromal PD to these different subtypes

In the absence of molecularbiological identification of pro-dromal disease we run the risk of extending the negative trialsto prodromal PD as the AD field has already demonstratedwith antiaggregation interventions in prodromal AD (mildcognitive impairment) This may be magnified by the fact thatthe most common prodromal PD phenotype REM sleep be-havior disorder can also evolve into dementia with Lewy bodiesand multiple system atrophy35 It is possible that earlier will notbe better if we target a large clinically defined syndrome ratherthan smaller but more homogeneous biological constructs

Key roadblocksMolecular criteria for subtyping are needed for both pro-dromal PD and manifesting PD

Tying biomarkers todisease mechanismsLike the examples of breast cancer and CF genotype andmolecular subtyping will need to replace phenotypic



subtyping which does not predict pathobiology pathology orpathophysiology67 Greater granularity in the clinical char-acterization of patients based on more sophisticated semio-logic neuroimaging or wearable technologyndashbased activitysensors should be of considerable utility in providing quan-titative and objective endpoint measures However likeldquoputting the cart before the horserdquo these approaches areunlikely to succeed alone in guiding the development ofbiomarkers or selecting patient subtypes for disease-modifying treatments These ideas lead to 2 formulations

c Revision fitting data to purpose Genetic risk profiles havebeen undertaken with the purpose of enhancing riskprediction for a disease considered polygenic Under thisparadigm the use of multiple common risk alleles canapproach the predictive accuracy of a single gene mutationin a mendelian disease36 For instance rare risk variants inthe glucocerebrosidase (GBA) and leucine-rich repeatkinase 2 (LRRK2) genes can serve ldquoto distinguish patientswith Parkinsonrsquos disease from healthy controlsrdquo37 Geneticbiomarkers are aimed at enhancing predictive or manifest-ing (PDnot-PD) diagnostic accuracy in at risk or clinicallydefined cohorts respectively as well as predicting risk ofprogression to key phenotypic outcomes such as de-mentia38 Clinical phenotypes act as the independentvariable genetic and biological measures the dependentvariable8 Research on extremes of clinical phenotype (egyoung-onset PD) or genotype (eg GBA-PD) inform thelarger whole by providing linked pieces to a disease puzzlethat can be brought into full view through their study1

c Reconstruction fitting purpose to data The exponentialincrease in significant hits from genome-wide associationstudies (to date 90 genetic risk loci for PD) may not reflectcontributors to a single phenomenon but expressions ofmultiple and mostly independent biological mechanismswhich may even be conditional on other unmeasuredfactors An examination of these loci as independentvariables might reveal a range of phenotypes potentiallyincluding but not restricted to the PD spectrumUnder sucha framework a given GBA mutation explains GBA-PD butmay not meaningfully contribute to the whole of (sporadic)PD According to this approach geneticbiological meas-ures act as the independent variable clinical phenotype thedependent variable There is not a single disease puzzle butseveral puzzles each explaining related but molecularlyseparate diseases albeit converging in the selectivevulnerability of specific neuronal populations most notablynigrostriatal neurons

GBA-PDSimilar to CFTR in CF different GBA mutations lead to dif-ferent properties of the encoded glucocerebrosidase (GCase)protein While some mutations may result in retention of theprotein in the endoplasmic reticulum other mutations canaffect the interaction with saposin C (GCase activator) in thelysosome or lead to a complete lack of protein (null muta-tions)39 Therefore therapeutic strategies such as chaperones

may be beneficial for only specific mutations while otherstrategies such as gene therapy may conceivably be beneficialfor all mutation carriers This is further complicated by the factthat different mutation types (eg mild pN370S or severepL444P) have differential effects on the progression of thedisease39ndash41 This differential progression will require strati-fication by mutation type even in GBA-specific clinical trialsto avoid a differential rate of progression in 1 arm of the trialdue merely to different frequencies of specific mutation typesin each group

Understanding the mechanism for each type of GBA-PD islikely to allow the identification of GBA-specific biomarkersand the development of mechanism-specific drugs Becausea subgroup of patients with sporadic PD have low GCase de-spite the absence of a GBA mutation42 recruitment for theongoing phase II GBA-specific trial (ClinicalTrialsgov identi-fier NCT02906020) includes a subset of patients with sporadic(non-GBA) PD with low GCase If the outcome were to benegative or futile it would support avoiding further develop-ments in a wider population than biologically appropriate

RevisionAlthough overextending the scope of molecular therapiescompromises their signal-to-noise ratio studies encompass-ing gt1 GBA mutation (or sporadic PD with low GCase) maydeserve consideration In reconstruction subtyping formutation-specific GBA-PD will reduce the population avail-able for recruitment but correspondingly increase the odds ofsuccess (by targeting smaller but better defined subsets)

LRRK2-PDCertain LRRK2 mutations (eg pG2019S) are associatedwith increased LRRK2 kinase activity Preliminary data sug-gest that LRRK2 kinase activity may also be increased in somePD brains without a LRRK2 mutation43 Thus a disease-modifying strategy that aims to inhibit kinase activity is cur-rently being tested Cancer has seen successes in applyinga similar strategy with kinase inhibitors (34 kinase inhibitorswere approved by the Food and Drug Administration forcancer by 2017)44 These successes have in part resultedfrom applying the concept of kinase dependency in that (1)kinase inhibitors will be successful only in diseases withupregulated kinase activity and (2) optimum use of kinaseinhibitors as therapies can be achieved only if coupled withbiomarkers that identify kinase-dependent disease45 Learningfrom these successes in cancer researchers know that the useof LRRK2 kinase inhibitors in PD requires development ofdiagnostic biomarkers to identify patients with PD withupregulated LRRK2 kinase activity in the CNS (LRRK2kinasendashdependent PD) Therefore treatment with LRRK2kinase inhibitors could potentially be extended to otherpatients with PD with enhanced LRRK2 kinase activity(analogous to patients with PD without GBA mutations butwith evidence of reduced GCase activity) Biomarkers tomeasure LRRK2 kinase inhibition in the CNS by these drugswill also be required Collectively these biomarkers will be



necessary for clinical trials to test the hypothesis that LRRK2kinase inhibition slows disease progression in patients withLRRK2 kinasendashdependent PD

Revision without reconstructionIf LRRK2 kinase activity is a pathogenic driver in mostLRRK2-PD and some nonndashLRRK2-PD opportunities fortherapies may materialize in gt1 genotype and perhaps insporadic kinase-dependent PD subtypes

InflammationMarkers of immune activation have been observed in mostsubtypes of PD These include peripheral inflammatorycytokines such as IL-6 tumor necrosis factor IL-1β IL-2 IL-10 and C-reactive protein46 A more proinflammatory cyto-kine profile at diagnosis is associated with faster subsequentmotor progression and more cognitive decline47 However itis unclear whether these markers represent downstreammediators (overlapping areas in the Venn diagram of figure 1)or upstream etiopathogenic events

RevisionIf causal or contributing adjunctive anti-inflammatory inter-ventions may lessen disease progression across most diseasesubtypes Inflammation or immune activation might be par-ticularly critical at an earlier time than putative interventionscan be enacted and influenced by host factors such as themicrobiome Alterations in the composition of gut microbiotahave been linked to inflammatory metabolic oncologic andpsychiatric disorders48 The evidence of aggregated α-syn inthe intestine early in PD potentially even in normal individ-uals makes the gut-brain connection particularly compellingIn addition different microbiomes coexist in other body sitessuch as the nose mouth or the skin and they too can po-tentially influence the triggering of PD The microbiome ishighly dynamic influenced by host genome variability regionof residence age sex diet medications and PD itself (egsecondary to constipation)49 Because it is still unclear howthe microbiome might be modulated to affect disease ex-pression measuring intestinal microbiota in observational andinterventional studies is imperative to better define its roleacross PD subtypes

ReconstructionAlthough epidemiologic evidence suggests that nonsteroidalanti-inflammatory drugs do not influence PD50 it remainspossible that subgroups with different patterns of immuneactivation could be targeted with selective anti-inflammatoryor immunomodulatory strategies

Challenges to clinical trials of disease-modifying drugs Matching the whatto the whoWhile many of our explanations for past negative outcomeshave emphasized the ldquohowrdquo (inadequate clinical trials) and

ldquowhenrdquo (patients too advanced) if the ldquowhatrdquo (right drugright target) and ldquowhordquo (right patients) are perfectly alignedthe larger anticipated effect size may allow imperfections inthe how (right clinical trial design and execution) and when(right patient staging)51

WhoPotential therapies known to affect a presumed pathogenicpathway will require a demonstration that the relevantmechanism (biomarker of pathogenesis) is active and path-ogenic in the intended recipients The selection of studyparticipants should be based on mechanism rather thanphenotype As with the experiences in CF and breast cancerthe first successes may materialize in small populations (likelylt5 of those with PD) The obvious first population to targetwill likely be individuals carrying the relatively common riskfactor genes LRRK2 and GBA

StratificationDifferent treatments are best suited to different subgroupsFuture studies should ideally bank biosamples (blood andCSF) and cells (eg peripheral blood mononuclear cells in-duced pluripotent stem cells) to assess for biomarkers of re-sponse and failure after study participation Study consentforms should secure participantsrsquo permission for samples anddata to be used in future analyses after their participation inthe study is completed This would admittedly representa challenge because neuronal cells are more difficult to obtainthan tumor samples (although adult somatic cells could bereprogrammed into neurons)

GeneralizabilityGBA- and LRRK2-targeted therapy trials may not generalizeto other forms of PD or perhaps to all expressions of GBAand LRRK2 mutations in manifesting individuals given thatdifferent GBA and LRRK2 mutations may result in differentpathogenetic mechanisms as learned from the experiencein CF

WhatAs noted above the what should not be considered in absenceof the who Of the recent or ongoing phase IIIII trials(inosine isradipine deferiprone and passive α-syn immuni-zation) only one paired the what with the who most likely tobenefit (inosine for patients with PD with low serum uratelevels in the Study of Urate Elevation in Parkinsonrsquos Disease[SURE-PD3]) Regardless of how promising a treatment mayhave seemed in preclinical studies it may be possible totranslate that promise to a patient population without etio-logic enrichment for the targeted pathogenic mechanism andin a manner that demonstrates target engagement

From single to multiple therapiesCombined or cocktail treatments will potentially be morebeneficial by targeting gt1 dominant pathogenic mechanism inany given disease subtype51 However it is likely that a singleintervention will need to be proven effective in a small patient



subgroup before the targeted pool can be increased throughthe addition of a second intervention (as discussed for CF) Inaddition a combined approach may be best positioned toaddress the ldquotriggers facilitators and aggravatorsrdquo52 for eachsubtype (eg LRRK2-PD) according to any changes in therelative dominance or contribution of a pathogenic mecha-nism over time or if several cellular mechanisms act in parallelrather than in series for that subtype To add further com-plexity defining a therapeutic window for each agent mayrequire considering a linear relationship between duration oftreatment and side effects but a potentially nonlinear re-lationship between duration of treatment and efficacy Finallymore collaboration to expedite and derisk drug developmentby industry will also enhance the ability to study combinationtherapies Precompetitive collaboration (ldquoco-opetitionrdquo) isa business model whereby competitor companies work to-gether in a competitively safe space to solve problems co-operatively to the benefit of all53

HowThe traditional approach to clinical trials tends to be costlyand time-consuming going stepwise from preclinical de-velopment to clinical phases 1 through 3 An alternativenimbler and more efficient approach is that of learning andconfirming clinical trials54 The interest in innovative trialdesigns has given rise to adaptive designs which allow thereview of accumulating information during an ongoing clinicaltrial to potentially modify trial characteristics Challenges tothe application of these concepts to PD in contrast to selectedother diseases such as breast cancer include the lower in-cidence slow progression lack of reliable rapid readouts ofchanges in underlying disease status and availability of potent

symptomatic treatments that mask important changes Tar-geting patients whose symptomatic management has beenoptimized before study entry may mitigate the last issue andaddress concerns that any disease-modifying effects may inpart represent (or be masked by) symptomatic benefits

Adaptive designAdaptive design actually means adaptive by design Toproperly define the simulations and to preserve the integrityand validity of the trial properly designed simulations andadaptation rules must be specified in advance (planned ad-aptation) (figure 2) Adaptive enrichment designs representa variation that fulfills the desire to target therapies to patientswho can benefit the most from treatment In such designsa trial would initially consider a broad population (eg PDwith and without GBA mutations but low GCase) The firststudy period a discovery phase reveals participant groupsmost likely to benefit from treatment (eg only patients withmutation-specific GBA-PD) Participants from this subgroupare then randomized to treatment groups in a validationphase Hence the study power is increased despite a re-duction in sample size by focusing only on subgroups mostlikely to show benefit The adaptive design sample size re-estimation allows adjustment of the sample size based ona review of the interim data An internal pilot is a sample sizere-estimation used to reassess nuisance parameters midstudyWith moderate to large sample sizes internal pilot designs canbe used to make appropriate modifications with minimal in-flation of the type I error rate Thus there is little reason(statistically) not to apply this design for most clinical trialsHowever such trial designs are logistically complicated Strictattention must be paid to maintaining the blinding and

Figure 2 Planned adaptation allows learning and confirming in adaptive clinical trials

Only planned adaptations can beguaranteed to avoid unknown biasesresulting from the adaptation (fromKairalla et al60) GS = Group sequen-tial SSR = Sample-size re-estimation



minimizing time gaps during which adaptation decisions aremade to ensure the efficiency and integrity of adaptive studies

Master protocolsMaster protocols are protocols ldquodesigned with multiple sub-studies which may have different objectives and involve co-ordinated efforts to evaluate one or more investigational drugsin one or more disease subtypes within the overall trialstructurerdquo according to the Food and Drug AdministrationGuidance for Industry definition Master protocols bring ef-ficiency by creating 1 optimized trial infrastructure applying itperpetually to study multiple therapies using the followingbasket trials (test single drug or drug combination in multiplepopulations or settings) umbrella trials (test multiple in-vestigational drugs in a single population) and platform trials(test multiple therapies in a single disease in a perpetualmanner with therapies allowed to enter or leave the platformaccording to a decision algorithm) (figure 3)

Master protocols use a trial network with shared in-frastructure to streamline trial logistics to improve dataquality and to facilitate data collection and sharing anda common protocol that incorporates innovative statisticalapproaches to study design and data analysis enablinga broader set of objectives to be met more efficiently thanwould be possible in independent trials The scientific effi-ciencies of this type of adaptive design include a shared pla-cebo group sample size savings the study of multiple drugsand multiple markers and the flexibility to addremoveagents Successful examples in which a platform trial has beenimplemented include the Investigation of Serial Studies toPredict Your Therapeutic Response through Imaging andMolecular Analysis 2 (I-SPY 2) to screen drugs in neo-adjuvant breast cancer55 and Dominantly Inherited AlzheimerNetwork Trials Unit (DIAN-TU) in at-risk or early-onsetgenetic AD56 The first Platform Trial Initiative for amyo-trophic lateral sclerosis was recently established at the HealeyCenter for ALS at Massachusetts General with 5 promisingdrugs entering evaluation as of September 2019 (massgeneralorgneurologyalsresearchplatform-trial) Public-private

partnership efforts similar to those of the European Pre-vention of Alzheimerrsquos Dementia project57 will be requiredfrom the PD community to facilitate the needed application ofthese study designs to PD Efforts from the Critical PathInstitute and the Alzheimerrsquos Association will also be essentialin cementing the needed collaboration among industry aca-demia and regulatory agencies in developing effective com-bination therapies for PD AD and other neurodegenerativedisorders58 Challenges in the execution of master protocolsinclude ensuring the willingness of all sponsors to share dataand matching multiple dosing regimens (frequencies formsand routes) within the trial infrastructure

Special considerationsThere is much to be learned from existing cohorts includingdeep clinical genetic and -omics phenotyping with emphasison genetic forms of PD (revision) As argued previously largepopulation-based aging cohorts agnostic to phenotype willhave the power to find biological signals of pathogenicmechanisms for which putative therapies already exist andcould be repurposed Proof-of-concept adaptive clinical trialsof putative disease-modifying drugs can be embedded withinbiomarker-development cohorts (reconstruction)

Financial riskhypothesis certaintyWhile the mechanism adopted by the CF research communitywas instrumental in derisking pharmaceutical industry drugdiscovery and early-phase clinical trials particularly to managefailures due to safety drugs can fail on efficacy Thus it isimperative to have high confidence in the therapeutic hy-pothesis being tested in a clinical trial and the likelihood ofadequate target engagement No derisking effort can preventcandidate drugs from failing on efficacy because patients werenot selected on the basis of the mechanism of action that thepotential drug is purported to have The complex interplaybetween the financial risks and the certainty of the therapeutichypothesis has been elegantly addressed for rare diseases59 Ifthe fundamentals about drug development for disease modi-fication were to move from a large indication (eg sporadicPD) for which a high risk tolerance is mitigated by the po-tential of future astronomical returns even with very lowprobability of success (revision) to a small indication (ega molecular PD subtype reconstruction) the developmentprocess would tolerate less risk and the therapeutic hypoth-esis would have to be definitive rather than tentative A rare-disease model would also need to rely on orphan designationfor cost and revenue advantages and exploit the potential ofdrug repurposing

Animal models in drug developmentAnimal models cannot be expected to recapitulate the com-plexity of a human disease that exists in many forms andtypically presents late in life Although some fields (eg CF)have abandoned the reliance of therapeutic development onanimal models they could serve to understand specificmechanisms suggested by biospecimens and neuroimaging

Figure 3 Platform trial example

It allowsmultiple drugs examined from the outset which can be removed orcombined (eg A + C) It also permits new drugs (D) to be examined at a latertime point in the same trial



studies in humans (revision) The need for research applica-tions to demonstrate efficacy first in animal models beforeprospective trials can proceed in humans will need to bereconsidered (reconstruction)

ConclusionsAs recently articulated51 more critical than the tools ofwarfare (clinical trial designs endpoint sensitivity treat-ments themselves) may be the reconfiguration of warfareitself The whowhat we are fighting for has not evolvedfrom the days of blunt targeting of early PD in which so-phisticated weapons were deployed on a population with-out evidence of vulnerability to the effect of those weaponsSome aspects of our approach can benefit from a revisionstrategy an immediate evolutionary adaptation or short-cut Many areas require reconstruction which needs not

take longer but may be more revolutionary than evolu-tionary and will certainly be disruptive The tension be-tween revision and reconstruction may not be solved in theshort term (table 2)

Revision (shortcut)Opportunities exist for harnessing existent and accruinggenetic cohorts most meaningfully in GBA-specific clinicaltrials for GBA-PD subtypes stratified by mutation type andin kinase-dependent PD subtypes selected through bio-markers of upregulated kinase activity for targeting withkinase inhibitors If LRRK2 kinase activity is a pathogenicdriver in nonndashLRRK2 kinase-dependent PD treatmentsuccess may materialize in gt1 kinase-dependent PDgenotypesubtype In addition ongoing and future clinicaltrials in genetic and even nongenetic cohorts can gatherdata to generate markers of response and failure to refinethe ldquowhordquo in subsequent efforts

Table 2 Point-counterpoint Revision vs reconstruction

Revision Reconstruction

About the lessons from other diseases (eg cystic fibrosis and breast cancer)

Many treatments were developed without subtyping (egbisphosphonates for osteoporosis antibiotics for peptic ulcers)

Other diseases achieved first successes once divided into biomarker-driven subgroups (eg calcium Helicobacter pylori)

About clinical subtypes

Data-driven clinical subtyping can be used to explain differences inclinical course and may eventually predict biological subtyping

Clinical data-driven subtypes are not stable or reproducible betweenstudies45 or are related to unique pathologies

About pathology as axis of disease

α-Syn aggregation is so common that it must be a key to pathogenesisThe reason for poor correlation with cell loss is that cells are dead sono α-syn is left Other species of α-syn (ie nonaggregated oligomers)may be of greater importance

α-Syn aggregation is neither necessary nor sufficient for cell loss to occuror for clinical parkinsonism α-Syn aggregation has never been proven tolead to degeneration or to correlate with symptoms

Whereas tau pathology might bemore important than β-amyloid in ADthere is no support for a second pathology in PD thus the parallelsare unjustified The role of copathologies is uncertain and could besecondary to α-syn (eg seeding)

Pure α-syn aggregation has been shown to be the exception not the rule(80 copathologies in PD see also table 1) β-Amyloid and tau co-occurwith α-syn in 810 pathology-proven cases of PD

About pathology as target for therapy

If the wrong species of α-syn is targeted a negative trial would notmeanthat α-syn is irrelevant to PD α-Syn is important for PD even ifaggregation may not be the appropriate target

All α-syn species are present across all PD subtypes a single speciescannot be pathogenic in all α-Syn antiaggregation drugs might be ofbenefit in SNCA-PD but futile in other subtypes

About clinical trials of disease-modifying interventions

Until trials prove adequate brain penetration and target engagementhypotheses are not invalidated Failure of antiamyloid drugs in ADdoes not predict failure of targeting α-syn in PD

Refinements in trial design and execution suggest that negative trials area problem with the hypotheses Despite target engagement theantiamyloid approach has yielded 33 null trials

If the cohort is too selective recruitment becomes impossible and willfail to detect even moderate benefit Negative tests of any pathwaybiomarker may not exclude a possible contributing role of thispathway and potential benefit would be missed for those lacking thebiomarker of interest

Targeted recruitment to those with biomarkers of diseasetargetengagement will predict large effect sizes and small sample sizes largeearlyprodromal cohorts without biomarkerswill be unlikely to respond tomechanistic treatments

Mechanism of action for drugs is often unknown target engagement isunmeasurable andCSF changes are not equal to brain changes In theabsence of highly predictive biomarkers it is reasonable to targetbroadly common mechanisms including the toxic effects of α-syn orinflammation

It is imperative to be certain about the therapeutic hypothesis being testedand the likelihood of adequate target engagement If PD is many diseasesand α-syn and inflammation are common to most they are unlikely to bepathogenic in all

Abbreviations AD = Alzheimer disease α-syn = α-synuclein PD = Parkinson disease SNCA = synuclein gene



Reconstruction (long-cut)While drug combinations are a future expectation provingsuccess of individual drugs will require ascertainment ofmolecular PD subtypes regardless of progress in clinical trialdesigns The ongoing phenotype-based biomarker de-velopment programs will need to evolve into phenotype-agnostic biology-driven biomarkers able to distinguish caus-ative mechanisms from nonspecific late or commonconvergent mechanisms inappropriate or inadequate astherapeutic targets The therapeutic paradigm for neuro-protection will have to emulate the lessons from other fields ofmedicine tailoring biology-specificmultimechanism strate-gies to well-defined PD subtypes Although there will be valuein post hoc responder analyses of future large clinical trials themore biologically enriched the targeted groups for the treat-ment of interest are the greater the odds of therapeuticsuccess

Study fundingMajor support for the Revision vs Reconstruction meetingleading to this manuscript was provided by The KrembilFoundation the Ronald Kimel Foundation Trust the Par-kinsonrsquos Foundation the International Parkinson andMovement Disorder Society and the American ParkinsonDisease Association Additional support was provided by theInternational Parkinson and Movement Disorders Societythe Parkinson Foundation and the following pharmaceuticalcompanies Sunovion UCB Valeo Pharma Acadia Pharma-ceuXcals Inc Medtronic of Canada Ltd ProtoKineticsTheravance Biopharma US Inc and Neurocrine BioscienceThe Revision vs Reconstruction meeting was endorsed by theMichael J Fox Foundation and the American ParkinsonDisease Association CH Williams-Gray holds a ResearchCouncils UKUK Research Innovation Research InnovationFellowship awarded by the Medical Research Council(MRR0074461) RA Barker is supported by theWellcomeTrustMedical Research Council Stem Cell Institute(203151Z16Z)

DisclosureAJ Espay has received grant support from the NIH and theMichael J Fox Foundation personal compensation as a con-sultantscientific advisory board member for Abbvie AdamasAcadia Acorda Neuroderm Impax Sunovion LundbeckOsmotica Pharmaceutical and USWorldMeds publishingroyalties from Lippincott Williams amp Wilkins CambridgeUniversity Press and Springer and honoraria fromUSWorldMeds Lundbeck Acadia Sunovion the AmericanAcademy of Neurology and the Movement Disorders Soci-ety LV Kalia holds a Canadian Institutes of Health Research(CIHR) Clinician-Scientist Award receives research supportfrom CIHR Michael J Fox Foundation for Parkinsonrsquos Re-search Natural Sciences and Engineering Research Council ofCanada Ontario Brain Institute Parkinson Canada andToronto General amp Western Hospital Foundation holdscontracts with ApoPharma and received educational grants

from Allergan and honoraria from Pfizer Shire and the NIHZ Gan-Or has served as an advisor for Lysosomal Thera-peutics Inc Idorsia Prevail Therapeutics Inception Sciences(now Ventus) andDenali received grants from theMichael JFox Foundation the Canadian Consortium on Neuro-degeneration in Aging the Canadian Glycomics Network theCanada First Research Excellence Fund awarded to McGillUniversity for the Healthy Brains for Healthy Lives programand the Fonds drsquoacceleration des collaborations en santegranted by MESICQDM and is supported by Fonds derecherche du Quebec-Sante Chercheurs-boursiers award withParkinsonrsquos Quebec and by the Young Investigator Award byParkinsonrsquos Canada CH Williams-Gray has received grantsfrom the Michael J Fox Foundation the Rosetrees Trust theEvelyn Trust Addenbrookersquos Charitable Trust ParkinsonrsquosUK the Academy of Medical Sciences UK the StevenageBiosciences Catalyst and the Patrick Berhoud Trust and issupported by the National Institute for Health Research(NIHR) Cambridge Biomedical Research Centre grant146281 (the views expressed are those of the author[s] andnot necessarily those of the NIHR or the Department ofHealth and Social Care) she has received honoraria fromLundbeck and consultancy fees from Modus Outcomes PLBedard has received grant support (to his institution) fromBristol-Myers Squibb Sanofi AstraZeneca GenentechRoche Servier GlaxoSmithKline Novartis SignalChemPTCTherapeutics Nektar Merck Seattle Genetics MersanaImmunomedics and Lilly SM Rowe has received grantsupport from the NIH the Cystic Fibrosis Foundation VertexPharmaceuticals Inc Bayer Forest Research Institute Astra-Zeneca N30Nivalis Novartis GalapagosAbbVie Proteo-stasis Eloxx Celtaxsys PTC Therapeutics and VertexPharmaceuticals Inc and personal honoraria from BayerNovartis Renovion and Vertex Pharmaceuticals Inc FMorgante received honoraria from UCB Pharma BIALChiesi Medtronic Zambon Chiesi Abbvie and Merz shereceives royalties from Springer for the book Disorders ofMovement A Fasano received honoraria for consultanciesfrom Abbvie Abbott BrainLab Boston Scientific Chiesifarmaceutici Ipsen Medtronic Sunovion and UCB hono-raria for participation in advisory boards from Abbvie BostonScientific and Ipsen and research grants from Abbvie BostonScientific Cummings Foundation Dystonia Medical Re-search Foundation Canada Michael J Fox FoundationMedtronic University of Toronto and Weston FoundationB Stecher is a patient and Parkinson advocate He writes ontmrweditioncom He has received honoraria from ZambonBiogen Abbvie Roche the Buck Institute McGill and theTanenbaum Open Science Institute MA Kauffman is anemployee of the CONICET He has received grant supportfrom Ministry of Science and Technology of Argentina andMinistry of Health of Buenos Aires He has received honorariapayments for educational activities from Janssen Pharma-ceuticals and Bago Pharmaceuticals MJ Farrer has receivedresearch grant support from the Canadian Federal Govern-ment British Columbia (Leadership Fund-Life Labs GenomeBC amp the Province) and the Weston Foundation He has



intellectual property rights including US patent WO 2005004794 A2 amp JP 2011-30292 International Publication NoWO 2006045392 A2 Method of Treating NeurodegenerativeDiseases International Publication No WO 2006068492 A1US Publication No US-2008-0009454-A1 and Norwegianpatent 323175 ldquoIdentification of LRRK2 in parkinsonismincluding mutation c6055GgtA (pG2019S) and mousemodelsrdquo He has received royalties from Mayo FoundationAthena Diagnostics amp H Lundbeck AS CS Coffey hasreceived grant funding from the Michael J Fox FoundationNational Heart Lung and Blood Institute and National In-stitute of Neurological Disorders and Stroke (NINDS) andconsulting funds from Michael J Fox Foundation MASchwarzschild is an employee of Massachusetts GeneralHospital and a member of the faculty of Harvard MedicalSchool and is supported by research funding (from the NIHMichael J Fox Foundation the Parkinson Foundation TargetALS The Maximilian E amp Marion O Hoffman FoundationBiotieAccorda and the US Department of Defense) clinicalpractice compensation an endowment and university fundsHe serves on the scientific advisory boards of the Michael JFox Foundation CBD Solutions and Cure Parkinsonrsquos TrustHe has served as a consultant for New szlig Innovation He hasreceived royalties from Massachusetts General Hospital forlicensing of adenosine A2A receptor knockout mice T Shereris the chief executive officer of the Michael J Fox Foundationfor Parkinsonrsquos Research RB Postuma reports grants andpersonal fees from Fonds de la Recherche en Sante theCIHR Parkinson Canada the Weston‐Garfield Foundationthe Michael J Fox Foundation and the Webster Foundationas well as personal fees from Takeda RocheProthena TevaNeurosciences Novartis Canada Biogen Boehringer Ingel-heim Theranexus GE HealthCare Jazz PharmaceuticalsAbbVie Janssen and Otsuko AP Strafella has receivedfunding from Canada Research Chair Program CIHR Na-tional Parkinson Foundation Parkinson Disease FoundationParkinson Canada Ontario Gambling Association TouretteSyndrome Association Brain Canada and Weston Brain In-stitute as well as honoraria fromGEHealthcare Canada LTDAB Singleton has nothing to disclose RA Barker has servedas an advisor to UCB Cellino Sana Biotherapeutics Blue-Rock Therapeutics Oxford Biomedica LCT Novo Nordiskand Fujifilm Cellular Dynamics International He has receivedgrant support from the Medical Research Council WellcomeTrust Parkinsonrsquos UK Cure Parkinsonrsquos Trust RosetreesTrust EU Evelyn Trust John Black Charitable Trust andCHDI and is supported by the NIHR Cambridge BiomedicalResearch Centre (the views expressed are those of the author[s] and not necessarily those of the NIHR or the Departmentof Health and Social Care) He has received royalties fromWiley and Springer-Nature K Kieburtz has received supportas consultant from Clintrex LLC RocheGenentechNovartis and Blackfynn grant support from the NIH(NINDS National Center for Advancing Translational Sci-ences) and Michael J Fox Foundation and support fromownership of Clintrex LLC Hoover Brown LLC and SafeTherapeutics LLC CW Olanow has served as a consultant to

Fuji Lundbeck Newron Teva UCB and Zambon He ownsstock in Clintrex which provides consulting services forAstraZeneca AcordaCivitasBiotie Blackthorne BritanniaCynapsus Dart Denali EMD Serono Inhibikase IntecIpsen Jazz Kyowa‐Kirin Lundbeck Michael J Fox Foun-dation Monosol Neuraly Neurocrine Neuroderm OtsukaINC Pfizer Pharma Two B PhotoPharmics SanofiGenzyme Sarepta Serina Sunovion SynAgile TakedaTeva Ultragenyx US WorldMeds Vaccinex Voyager vTvand Weston Foundation He has served on the boards of theMichael J Fox Foundation the National Space BiomedicalResearch Institute UCB and Zambon A Lozano is a con-sultant for Medtronic Boston Scientific and Abbott JHKordower is a paid consultant to Clintrex Inc NsGeneincAxovant Inc Fuji-CDI Inc Inhibikase Inc Abbvie and theMichael J Fox Foundation JM Cedarbaum is an in-dependent consultant He is former employee of and share-holder in Biogen P Brundin has received commercial supportas a consultant from Axial Biotherapeutics CuraSen Fujifilm-Cellular Dynamics International IOS Press Partners LifeSciCapital LLC Lundbeck AS and Living Cell TechnologiesLTD He has received commercial support for grantsresearch from Lundbeck AS and Roche He has ownershipinterests in Acousort AB and Axial Biotherapeutics and is onthe steering committee of the Nilotinib in Parkinsonrsquos Diseasetrial DG Standaert is a member of the faculty of the Uni-versity of Alabama at Birmingham and is supported by en-dowment and university funds Dr Standaert is aninvestigator in studies funded by Abbvie Inc Avid Radio-pharmaceuticals the American Parkinson Disease Associa-tion the Michael J Fox Foundation for Parkinson ResearchAlabama Department of Commerce the Department of De-fense and NIH grants P01NS087997 P50NS108675R25NS079188 P2CHD086851 P30NS047466 andT32NS095775 He has a clinical practice and is compen-sated for these activities through the University of AlabamaHealth Services Foundation In addition since January 12019 he has served as a consultant for or received honorariafrom Axovant Sciences Inc Censa Pharmaceuticals AbbvieInc Grey Matter Technologies Theravance Inc the Ken-nedy Krieger Institute McGraw Hill Publishers Sanofi-Aventis RTI Consultants Cerevance Inc Yale Universityand Michigan State University AE Lang has served as anadvisor for Abbvie Allon Therapeutics Avanir Pharma-ceuticals Biogen Idec Boerhinger-Ingelheim CeregeneLilly Medtronic Merck Novartis NeuroPhage Pharma-ceuticals Teva and UCB received honoraria from Med-tronic Teva UCB AbbVie received grants from BrainCanada CIHR Edmond J Safra Philanthropic FoundationMichael J Fox Foundation the Ontario Brain InstituteNational Parkinson Foundation Parkinson Society CanadaTourette Syndrome Association and W Garfield WestonFoundation received publishing royalties from SaundersWiley-Blackwell Johns Hopkins Press and CambridgeUniversity Press and has served as an expert witness incases related to the welding industry Go to NeurologyorgN for full disclosures



Publication historyReceived by Neurology October 2 2019 Accepted in final formJanuary 16 2020

References1 Fujita KA Ostaszewski M Matsuoka Y et al Integrating pathways of Parkinsonrsquos

disease in a molecular interaction map Mol Neurobiol 20144988ndash1022 Simuni T Caspell-Garcia C Coffey C Lasch S Tanner C Marek K How stable are

Parkinsonrsquos disease subtypes in de novo patients analysis of the PPMI cohort Par-kinsonism Relat Disord 20162862ndash67

3 Greenland JC Williams-Gray CH Barker RA The clinical heterogeneity of Parkin-sonrsquos disease and its therapeutic implications Eur J Neurosci 201949328ndash338

4 Marras C Lang A Parkinsonrsquos disease subtypes lost in translation J Neurol Neu-rosurg Psychiatry 201384409ndash415

5 Mestre TA Eberly S Tanner C et al Reproducibility of data-driven Parkinsonrsquosdisease subtypes for clinical research Parkinsonism Relat Disord 201856102ndash106

6 De Pablo-Fernandez E Lees AJ Holton JL Warner TT Prognosis and neuropath-ologic correlation of clinical subtypes of Parkinson disease JAMA Neurol 201976470ndash479

7 Espay AJ Marras C Clinical Parkinson disease subtyping does not predict pathologyNat Rev Neurol 201915189ndash190

8 Espay AJ Schwarzschild MA Tanner CM et al Biomarker-driven phenotyping inParkinsonrsquos disease a translational missing link in disease-modifying clinical trialsMov Disord 201732319ndash324

9 Accurso FJ Rowe SM Clancy JP et al Effect of VX-770 in persons with cystic fibrosisand the G551D-CFTR mutation N Engl J Med 20103631991ndash2003

10 Ramsey BW Davies J McElvaney NG et al A CFTR potentiator in patients withcystic fibrosis and the G551D mutation N Engl J Med 20113651663ndash1672

11 Veronesi U Saccozzi R Del Vecchio M et al Comparing radical mastectomy withquadrantectomy axillary dissection and radiotherapy in patients with small cancers ofthe breast N Engl J Med 19813056ndash11

12 Veronesi U Cascinelli N Mariani L et al Twenty-year follow-up of a randomizedstudy comparing breast-conserving surgery with radical mastectomy for early breastcancer N Engl J Med 20023471227ndash1232

13 Early Breast Cancer TrialistsrsquoCollaborative Group Tamoxifen for early breast canceran overview of the randomised trials Lancet 19983511451ndash1467

14 PegramMD Pauletti G Slamon DJ HER-2neu as a predictive marker of response tobreast cancer therapy Breast Cancer Res Treat 19985265ndash77

15 Malhotra GK Zhao X Band H Band V Histological molecular and functionalsubtypes of breast cancers Cancer Biol Ther 201010955ndash960

16 Parkkinen L OrsquoSullivan SS Collins C et al Disentangling the relationship betweenLewy bodies and nigral neuronal loss in Parkinsonrsquos disease J Parkinsons Dis 20111277ndash286

17 Patterson L Rushton SP Attems J Thomas AJ Morris CM Degeneration of dopa-minergic circuitry influences depressive symptoms in Lewy body disorders BrainPathol 201929544ndash557

18 Kalia LV Lang AE Hazrati LN et al Clinical correlations with Lewy body pathologyin LRRK2-related Parkinson disease JAMA Neurol 201572100ndash105

19 Irwin DJ Grossman M Weintraub D et al Neuropathological and genetic correlatesof survival and dementia onset in synucleinopathies a retrospective analysis LancetNeurol 20171655ndash65

20 Buchman AS Yu L Wilson RS et al Progressive parkinsonism in older adults isrelated to the burden of mixed brain pathologies Neurology 201992e1821ndashe1830

Appendix Author contributions

Name Location Contribution

Alberto J EspayMD MSc

University of CincinnatiOH

Conception of researchproject literaturesearch writing of thefirst manuscript draft

Lorraine V KaliaMD PhD

University of TorontoOntario Canada

Review and critique ofthe manuscript

Ziv Gan-Or MDPhD

McGill UniversityMontreal QCCanada


Caroline HWilliams-GrayBMBCh MRCPPhD

University of CambridgeUK


Philippe LBedard MDFRCPC



Steven M RoweMD MSPH

University of Alabama atBirmingham


FrancescaMorgante MDPhD

St Georgersquos University ofLondon UK


Alfonso FasanoMD PhD



BenjaminStecher



Marcelo AKauffman MDPhD

UBA and UniversidadAustral-CONICET BuenosAires Argentina


Matthew JFarrer PhD

University of FloridaGainesville


Chris S CoffeyPhD

University of Iowa IowaCity IA


Michael ASchwarzschildMD PhD

Massachusetts GeneralHospital Boston


Todd ShererPhD

Michael J Fox Foundationfor Parkinsonrsquos ResearchNew York City NY


Ronald BPostuma MDMSc

McGill UniversityQuebec Canada


Antonio PStrafella MDPhD FRCPC



Andrew BSingleton PhD

National Institute onAging NIH Bethesda MD


Roger A BarkerMBBS MRCPPhD



Karl KieburtzMD MPH

University of RochesterNY


Appendix (continued)


C WarrenOlanow MDFRCPCFRCP(HON)

Mount Sinai School ofMedicine New York CityNY


Andres LozanoMD PhD



Jeffrey HKordower PhD

Rush University MedicalCenter Chicago IL


Jesse MCedarbaum MD

Coeruleus ClinicalSciences Woodbridge CT


Patrik BrundinMD PhD

Van Andel InstituteGrand Rapids MI


David GStandaert MDPhD



Anthony E LangMD FRCPC


Critical revision of themanuscript overallsupervision



21 Visanji NP Lang AE Kovacs GG Beyond the synucleinopathies alpha synuclein asa driving force in neurodegenerative comorbidities Translational neurodegeneration2019828

22 Espay AJ Vizcarra JA Marsili L et al Revisiting protein aggregation as pathogenic insporadic Parkinson and Alzheimer diseases Neurology 201992329ndash337

23 De Franceschi G Fecchio C Sharon R et al Alpha-synuclein structural featuresinhibit harmful polyunsaturated fatty acid oxidation suggesting roles in neuro-protection J Biol Chem 20172926927ndash6937

24 Shen N Song G Yang H et al Identifying the pathological domain of alpha- synucleinas a therapeutic for Parkinsonrsquos disease Int J Mol Sci 201920E2338

25 Kishimoto Y Zhu W Hosoda W Sen JM Mattson MP Chronic mild gut in-flammation accelerates brain neuropathology and motor dysfunction in alpha-synuclein mutant mice Neuromolecular Med 201921239ndash249

26 Chen SG Stribinskis V Rane MJ et al Exposure to the functional bacterial amyloidprotein curli enhances alpha-synuclein aggregation in aged Fischer 344 rats andCaenorhabditis elegans Sci Rep 2016634477

27 Nicoll JAR Buckland GR Harrison CH et al Persistent neuropathological effects 14years following amyloid-beta immunization in Alzheimerrsquos disease Brain 20191422113ndash2126

28 Braak H Del Tredici K Rub U de Vos RA Jansen Steur EN Braak E Staging of brainpathology related to sporadic Parkinsonrsquos disease Neurobiol Aging 200324197ndash211

29 Visanji NP Brooks PL Hazrati LN Lang AE The prion hypothesis in Parkinsonrsquosdisease Braak to the future Acta Neuropathol Commun 201312

30 Luk KC Lee VM Modeling Lewy pathology propagation in Parkinsonrsquos diseaseParkinsonism Relat Disord 201420(suppl 1)S85ndashS87

31 Burke RE Dauer WT Vonsattel JP A critical evaluation of the Braak staging schemefor Parkinsonrsquos disease Ann Neurol 200864485ndash491

32 Nuber S Petrasch-Parwez EWinner B et al Neurodegeneration andmotor dysfunctionin a conditional model of Parkinsonrsquos disease J Neurosci 2008282471ndash2484

33 Berg D Postuma RB Adler CH et al MDS research criteria for prodromal Parkin-sonrsquos disease Mov Disord 2015301600ndash1611

34 Simuni T Siderowf A Lasch S et al Longitudinal change of clinical and biologicalmeasures in early Parkinsonrsquos disease Parkinsonrsquos progression markers initiativecohort Mov Disord 201833771ndash782

35 Postuma RB Iranzo A HuM et al Risk and predictors of dementia and parkinsonismin idiopathic REM sleep behaviour disorder a multicentre study Brain 2019142744ndash759

36 Khera AV Chaffin M Aragam KG et al Genome-wide polygenic scores for commondiseases identify individuals with risk equivalent to monogenic mutations Nat Genet2018501219ndash1224

37 Nalls MA McLean CY Rick J et al Diagnosis of Parkinsonrsquos disease on the basis ofclinical and genetic classification a population-based modelling study Lancet Neurol2015141002ndash1009

38 Collins LM Williams-Gray CH The genetic basis of cognitive impairment and de-mentia in Parkinsonrsquos disease Front Psychiatry 2016789

39 Gan-Or Z Liong C Alcalay RN GBA-associated Parkinsonrsquos disease and othersynucleinopathies Curr Neurol Neurosci Rep 20181844

40 Gan-Or Z Amshalom I Kilarski LL et al Differential effects of severe vs mild GBAmutations on Parkinson disease Neurology 201584880ndash887

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

42 Alcalay RN Levy OA Waters CC et al Glucocerebrosidase activity in Parkinsonrsquosdisease with and without GBA mutations Brain 20151382648ndash2658

43 Di Maio R Hoffman EK Rocha EM et al LRRK2 activation in idiopathic Parkinsonrsquosdisease Sci Transl Med 201810eaar5429

44 Ferguson FM Gray NS Kinase inhibitors the road ahead Nat Rev Drug Discov201817353ndash377

45 Sawyers CL Opportunities and challenges in the development of kinase inhibitortherapy for cancer Genes Dev 2003172998ndash3010

46 Qin XY Zhang SP Cao C Loh YP Cheng Y Aberrations in peripheral inflammatorycytokine levels in Parkinson disease a systematic review and meta-analysis JAMANeurol 2016731316ndash1324

47 Williams-Gray CH Wijeyekoon R Yarnall AJ et al Serum immune markers anddisease progression in an incident Parkinsonrsquos disease cohort (ICICLE-PD) MovDisord 201631995ndash1003

48 Falony G Joossens M Vieira-Silva S et al Population-level analysis of gut micro-biome variation Science 2016352560ndash564

49 Hill-Burns EM Debelius JW Morton JT et al Parkinsonrsquos disease and Parkinsonrsquosdisease medications have distinct signatures of the gut microbiome Mov Disord 201732739ndash749

50 Ren L Yi J Yang J Li P Cheng X Mao P Nonsteroidal anti-inflammatory drugs useand risk of Parkinson disease a dose-response meta-analysis Medicine 201897e12172

51 Lang AE Espay AJ Disease modification in Parkinsonrsquos disease current approacheschallenges and future considerations Mov Disord 201833660ndash677

52 Johnson ME Stecher B Labrie V Brundin L Brundin P Triggers facilitators andaggravators redefining Parkinsonrsquos disease pathogenesis Trends Neurosci 2019424ndash13

53 Stephenson D Perry D Bens C et al Charting a path toward combination therapy forAlzheimerrsquos disease Expert Rev Neurother 201515107ndash113

54 Cedarbaum JM Elephants Parkinsonrsquos disease and proof-of-concept clinical trialsMov Disord 201833697ndash700

55 Messmer MF Wilhelm EE Shoulson I I-SPY 2 breast cancer trial as a model forinnovation in Alzheimer disease therapies JAMA Neurol 2017741027ndash1028

56 Bateman RJ Benzinger TL Berry S et al The DIAN-TU Next Generation Alz-heimerrsquos Prevention Trial adaptive design and disease progression model AlzheimersDemen 2017138ndash19

57 Ritchie CW Molinuevo JL Truyen L Satlin A Van der Geyten S Lovestone SDevelopment of interventions for the secondary prevention of Alzheimerrsquos dementiathe European Prevention of Alzheimerrsquos Dementia (EPAD) project Lancet Psychi-atry 20163179ndash186

58 Perry D Sperling R Katz R et al Building a roadmap for developing combinationtherapies for Alzheimerrsquos disease Expert Rev Neurother 201515327ndash333

59 Minikel EV How pharmaceutical industry financial modelers think about your raredisease [online] Available at cureffiorg20190429financial-modeling-in-rare-disease Accessed June 22 2019

60 Kairalla JA Coffey CS Thomann MA Muller KE Adaptive trial designs a review ofbarriers and opportunities Trials 201213145

Disputes amp Debates Rapid online correspondence

The editors encourage comments on recent articles through Disputes amp Debates

Access an article at NeurologyorgN and click on ldquoMAKE COMMENTrdquo beneath the article header Responses will be postedas rapidly as possible

Before submitting a comment to Disputes amp Debates remember the following

Disputes amp Debates is restricted to comments about articles published in Neurology within the last 8 weeks Read previously posted comments redundant comments will not be posted Your submission must be 200 words or less and have a maximum of 5 references the first reference must be the article on

which you are commenting You can include a maximum of 5 authors (including yourself)



DOI 101212WNL0000000000009107202094481-494 Published Online before print February 26 2020Neurology Alberto J Espay Lorraine V Kalia Ziv Gan-Or et al

reconstructionDisease modification and biomarker development in Parkinson disease Revision or

This information is current as of February 26 2020

ServicesUpdated Information amp

httpnneurologyorgcontent9411481fullincluding high resolution figures can be found at

References httpnneurologyorgcontent9411481fullref-list-1

This article cites 59 articles 9 of which you can access for free at

Citations httpnneurologyorgcontent9411481fullotherarticles

This article has been cited by 1 HighWire-hosted articles

Subspecialty Collections

httpnneurologyorgcgicollectionparkinsons_disease_parkinsonismParkinsons diseaseParkinsonism

httpnneurologyorgcgicollectionall_clinical_trialsAll Clinical trials

httpnneurologyorgcgicollectionall_clinical_neurologyAll Clinical Neurologyfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

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

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

rights reserved Print ISSN 0028-3878 Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright copy 2020 American Academy of Neurology All

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

Phenotypic refinements aside Parkinson disease (PD)remains largely defined as it was clinically gt200 years ago andpathologically 100 years ago Technical advances in the ex-ploration and analysis of pathobiological processes have beenused to validate rather than to question this clinicopathologicconstruct As a result the whole has become the sum of manyand increasing parts To order the increasing complexity intoan etiopathologic whole bioinformatic tools integrate mo-lecular and genetic discoveries connecting synaptic dysfunc-tion failed vesicular trafficking mitochondrial dysfunctionimpaired proteostasis α-synuclein (α-syn) pathobiology andneuroinflammation as pieces in a single puzzle1 In the ther-apeutic realm promising molecular interventions have beenadministered to patients in the absence of relevant measuresof target engagement that is bioassays of relevant molecularpathways rescued by such interventions Clinical trials aimedat slowing or halting progressive PD disability are undertakenwith the hypothesis that a molecular modification will affectcommon neurodegeneration pathways in most patients withthe clinical diagnosis of PD Consequently disease modifi-cation has remained elusive during the 4 decades in which wehave been trying with hypotheses and disease constructslargely unaltered

The fundamental question is whether PD is one disorder withvarying clinical manifestations influenced by a number of ge-netic and environmental factors that need to be revisedthereby allowing current approaches to understanding patho-genesis and advancing disease modification therapies orreconstructed to frame PD as a syndrome in which pathologiccommonalities are subordinate to unique etiologic or patho-physiologic entities thereby forcing new approaches to ad-vancing our understanding of the disease and developingsuccessful disease modification Understanding the answer tothis questionwill have far-reaching implications in terms of trialdesign disease modeling biomarker development therapeutictargeting and so on A key question is whether the negativeclinical trials of putative disease-modifying interventions havebeen true failures (failure of the hypothesis negative outcomesdeserve to be trusted) or false failures (failure of the trialmethodology hypotheses deserve another chance)

The Krembil Knowledge Gaps in PD Symposium held inToronto from April 24 to 26 2019 tackled this fundamentalconceptual challenge if PD is not a single disorder canadvances in biomarkers and disease modification be revised toconcentrate on commonalities of pathogenic mechanisms in

large populations (assuming biologic convergence) or dothey need to be reconstructed for application to smallersubgroups of patients distinguished by well-defined molec-ular characteristics with some mechanistic understanding ofetiopathogenesis and a relevant bioassay to measure it Thisreport summarizes the deliberations at that meeting andacknowledges the unresolved tension between the revisionistand the reconstructionist approach to PD and its treatment

Key knowledge gaps in 2019 Revisionor reconstructionAt the core the key question is whether PD is one disease withmany facets each contributing to explaining the whole ormany diseases sharing some elements with one another butwith distinctive pathogenic elements As a field we have livedwith this many-diseasesone-disease tension by reasoningthat both can be correct (figure 1) The road ahead for bio-marker development and clinical trial efforts toward diseasemodification depends on whether PD is a multinodal diseaseprocess connected by α-syn aggregation or a collection ofindependent pathologic processes that are largely if not en-tirely independent The former is the more alluring modelmolecular genetic forms of parkinsonism are consideredpieces of the same puzzle driving the creation of key researchconstructs (table 1)

Knowledge gap 1 Does the heterogeneity of PDrepresent a disease spectrum with differentclinical subtypes and rates of progression ordistinct molecular entities each a separatedefinable diseaseThe former premise has been a dominant force Adopting theview that PD is a spectrum supports efforts at increasing ormerging ongoing cohorts to maximize scale and deployingimproved analytic processes to the accumulating data (re-vision) However the data-drivenndashgenerated subtypes thusfar created are not stable over time23 have not beenreproducible45 and do not predict unique patterns of ag-gregation of α-syn or other pathologies67 Moreover meas-urements of putative aggregated protein-based biomarkershave shown substantial signal overlap between patients andcontrols and between disease phenotypes and replication ofresults across cohorts is lacking8 Combining cohorts an-chored on clinical criteria and dependent on confirmation ofdopamine deficit is a strategy likely to run into the same

GlossaryAD = Alzheimer disease α-syn = α-synuclein CF = cystic fibrosis CFTR = CF transmembrane conductance regulatorDIAN-TU = Dominantly Inherited Alzheimer Network Trials Unit ER = estrogen receptor GCase = glucocerebrosidase HER2 =human epidermal growth factor receptor 2 I-SPY 2 = Investigation of Serial Studies to Predict Your Therapeutic Responsethrough Imaging and Molecular Analysis 2 IL = interleukin LRRK2 = leucine-rich repeat kinase 2 PD = Parkinson diseaseSNCA = synuclein SURE-PD3 = Study of Urate Elevation in Parkinsonrsquos Disease
































































































































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints

































































































































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints














































































































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints



































































































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints





















































































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints











































































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints





























































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints





















































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints










































































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints












































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints






































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints


































Ziv Gan-Or MDPhD


















BenjaminStecher






Matthew JFarrer PhD



Chris S CoffeyPhD






Todd ShererPhD















Karl KieburtzMD MPH








Andres LozanoMD PhD






Jesse MCedarbaum MD













































































Reprints



































































Reprints



















Reprints




Documents

Disease modiﬁcation and biomarker development in …A fundamental question in advancing Parkinson disease (PD) research is whether it represents one ... lecular and genetic discoveries,