Progress in Neurobiology 95 (2011) 629–635

Contents lists available at SciVerse ScienceDirect

Progress in Neurobiology
journal homepage: www.elsevier.com/locate/pneurobio

The Parkinson Progression Marker Initiative (PPMI)

Kenneth Marek, Danna Jennings, Shirley Lasch, Andrew Siderowf, Caroline Tanner, Tanya Simuni,
Chris Coffey, Karl Kieburtz, Emily Flagg, Sohini Chowdhury, Werner Poewe, Brit Mollenhauer,
Paracelsus-Elena Klinik, Todd Sherer, Mark Frasier, Claire Meunier, Alice Rudolph, Cindy Casaceli,
John Seibyl, Susan Mendick, Norbert Schuff, Ying Zhang, Arthur Toga, Karen Crawford, Alison Ansbach,
Pasquale De Blasio, Michele Piovella, John Trojanowski, Les Shaw, Andrew Singleton, Keith Hawkins,
Jamie Eberling, Deborah Brooks, David Russell, Laura Leary, Stewart Factor, Barbara Sommerfeld,
Penelope Hogarth, Emily Pighetti, Karen Williams, David Standaert, Stephanie Guthrie, Robert Hauser,
Holly Delgado, Joseph Jankovic, Christine Hunter, Matthew Stern, Baochan Tran, Jim Leverenz,
Marne Baca, Sam Frank, Cathi-Ann Thomas, Irene Richard, Cheryl Deeley, Linda Rees, Fabienne Sprenger,
Elisabeth Lang, Holly Shill, Sanja Obradov, Hubert Fernandez, Adrienna Winters, Daniela Berg,
Katharina Gauss, Douglas Galasko, Deborah Fontaine, Zoltan Mari, Melissa Gerstenhaber, David Brooks,
Sophie Malloy, Paolo Barone, Katia Longo, Tom Comery, Bernard Ravina, Igor Grachev, Kim Gallagher,
Michelle Collins, Katherine L. Widnell, Suzanne Ostrowizki, Paulo Fontoura, Tony Ho,
Johan Luthman, Marcel van der Brug, Alastair D. Reith, Peggy Taylor
The Parkinson Progression Marker Initiative1

A R T I C L E I N F O A B S T R A C T

Article history: The Parkinson Progression Marker Initiative (PPMI) is a comprehensive observational, international, multi-
Available online 14 September 2011 center study designed to identify PD progression biomarkers both to improve understanding of disease
etiology and course and to provide crucial tools to enhance the likelihood of success of PD modifying
Keywords: therapeutic trials. The PPMI cohort will comprise 400 recently diagnosed PD and 200 healthy subjects
biomarker followed longitudinally for clinical, imaging and biospecimen biomarker assessment using standardized data
Dopamine transporter imaging acquisition protocols at twenty-one clinical sites. All study data will be integrated in the PPMI study database
Diffusion tensor imaging
and will be rapidly and publically available through the PPMI web site- www.ppmi-info.org. Biological
cerebrospinal fluid
alpha synuclein
samples including longitudinal collection of blood, cerebrospinal fluid (CSF) and urine will be available to
scientists by application to an independent PPMI biospecimen review committee also through the PPMI web
site. PPMI will rely on a partnership of government, PD foundations, industry and academics working
cooperatively. This approach is crucial to enhance the potential for success of this ambitious strategy to
develop PD progression biomarkers that will accelerate research in disease modifying therapeutics.
ß 2011 Elsevier Ltd. All rights reserved.

Contents

1. PPMI study design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 630

1.1. Study cohort and subject selection criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 630
1.2. Subject recruitment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 631
1.3. Site selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 631
1.4. Study assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 631
1.5. Scheme for incorporating additional PPMI assessments and biospecimen analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 632
1.6. Standardization of data acquisition/training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 632
1.7. Database/statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 632
1.8. PPMI website . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 633
2. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 633
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 634

1
Please see Appendix A for authors list with affiliations.

0301-0082/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.pneurobio.2011.09.005
630 K. Marek et al. / Progress in Neurobiology 95 (2011) 629–635

Reliable and well-validated biomarkers for Parkinson Disease underway. The overall study design and key study features are
(PD) progression would dramatically accelerate research into detailed below.
both PD etiology and therapeutics. During the past two decades
much progress has been made in identifying and assessing PD 1. PPMI study design
biomarkers, but as yet no fully validated biomarker for PD is
currently available (Maetzler et al., 2009; Marek et al., 2008; The PPMI cohort will comprise 400 recently diagnosed PD
Schapira, 2004; Olanow et al., 2008; Gerlach et al., 2008; Shi and 200 healthy subjects followed longitudinally and compre-
et al., 2009; Scherzer et al., 2007; Ravina et al., 2009). Given the hensively for biomarker assessment using standardized data
recent advances in molecular genetics, neurobiology, and acquisition protocols at twenty-one clinical sites with expertise
imaging technology and the recognition that the lack of PD in subject recruitment and biomarker assessment. The PPMI
biomarkers has created a roadblock for further studies of disease steering committee directs the study through the clinical,
modifying therapies, a major initiative to develop PD biomarkers imaging, genetics, bioanalytic, biorepository, statistics, and
is both timely and necessary. Therefore, the Parkinson Progres- bioinformatics cores (Fig. 1). The steering committee includes
sion Marker Initiative (PPMI), a collaborative effort of PD PD biomarker experts, study core leaders, and MJFF and industry
researchers with expertise in biomarker development, PD scientists. All study data will be integrated in the PPMI study
clinical study design and implementation, and bioinformatics, database and will be rapidly and publically available through
statistics, and data management, was developed to identify and the PPMI web site – http://www.ppmi-info.org/. Biological
validate PD progression markers (The Lancet, 2010). PPMI is a samples including longitudinal collection of blood, cerebrospinal
public–private partnership, largely sponsored by the Michael J fluid (CSF) and urine will be stored at the PPMI biorepository
Fox Foundation with substantial industry partnership (GE and will be available to scientists by application to an
Healthcare, Pfizer, Roche, Genentech, Merck, Abbott, Biogen, independent PPMI biospecimen review committee through the
Covance, Glaxo-Smith-Kline). PPMI web site. In addition the PPMI website will serve as a
PPMI is a five-year observational, international, multi-center source of ongoing information about standardized study
study designed to identify PD progression biomarkers both to procedures, study progress, and study results.
improve understanding of disease etiology and course and to
provide crucial tools to enhance the likelihood of success of PD 1.1. Study cohort and subject selection criteria
disease modifying therapeutic trials. The specific goals of PPMI are
to: PD subjects will be recruited at disease threshold. They will be
required to have an asymmetric resting tremor or asymmetric
1. Establish standardized protocols for acquisition, transfer and bradykinesia or two of bradykinesia, resting tremor and rigidity
analysis of clinical, imaging and biospecimen data that can be with diagnosis within two years and to be untreated for PD. All
used by the PD research community. subjects will undergo dopamine transporter (DAT) imaging and
2. Investigate existing and identify novel clinical, imaging and DAT deficit will be required for PD subject eligibility. Healthy
biospecimen PD progression markers that individually or in subjects must have no significant neurologic dysfunction, no first
combination will rapidly demonstrate interval change in PD degree family member with PD and Montreal Cognitive Assess-
patients in comparison to healthy controls or in sub-sets of PD ment MoCA > 26. This first use of an imaging biomarker for
patients defined by baseline assessments, progression mile- eligibility in a PD biomarker study offers three major advantages.
stones and/or rate of clinical, imaging or biospecimen change. First, use of DAT imaging will enhance the accuracy of diagnosis
3. Optimize bioassays and conduct preliminary verification studies given increasing evidence that subjects enrolled in PD clinical trials
on promising biological markers using stored biospecimen. with scans without DAT deficit are unlikely to have PD (Parkinson
Study Group, 2004, 2007; Whone et al., 2003). Second PD subjects
The PPMI study was planned during a two-year period in a can be enrolled in the study earlier in their disease course with only
series of workshops with input from academic PD experts, Michael a single asymmetric sign. There is growing consensus that
J Fox Foundation scientific staff, and government and industry enrolling PD subjects at the earliest detectable stage of disease
partners. The study was launched in June 2010 and enrollment is would likely enhance the potential to both identify a progression

Fig. 1. PPMI study organization and governance.

 K. Marek et al. / Progress in Neurobiology 95 (2011) 629–635 631

biomarker and provide a better population for eventual disease for the PPMI study. These clinical sites have undergone a two stage
modifying drug trials. Third, enrolling PD subjects earlier would assessment by the PPMI steering committee beginning with a site
likely prolong the duration of biomarker evaluation prior to questionnaire followed by an in person visit by a steering
initiating symptomatic treatment. This strategy would allow more committee member to assess site resources and capabilities for
complete assessment and comparison of PD biomarkers prior to PD and healthy subject recruitment and retention and to conduct
symptomatic medications, while still allowing assessment during all study clinical assessments, for collection of biospecimens and
the transition period when symptomatic medications are needed, for acquisition of positron emission tomography (PET), single
and after longitudinal follow-up while on symptomatic medications. photon emission computerized tomography (SPECT) and MRI. All
PD subjects enrolled in PPMI should not require symptomatic PD sites have undergone and will undergo ongoing training to ensure
medications for at least six months (subjects will be treated and standardization of data acquisition and biospecimen collection.
retained in the study if treatment is required before six months). PD PPMI has included sites in both the US and Europe to enhance the
subjects will be allowed to participate in clinical trials after 12 generalizability of the study, to broaden the recruitment pool, and
months participation in PPMI to encourage retention in PPMI. to encourage wide reaching collaboration among the PD research
community.
1.2. Subject recruitment
1.4. Study assessments
Given the acknowledged difficulty in identifying early-untreat-
ed PD study subjects and healthy subjects compounded by the Both the PD and healthy subjects in PPMI will undergo a
comprehensive longitudinal PPMI assessment requiring substan- comprehensive longitudinal schedule of clinical and imaging
tial subject time and commitment, it is anticipated that study assessments and biosampling (Table 1). Evaluations will occur at
recruitment will be a major challenge for PPMI. Within the PPMI screening/baseline and at 3 month intervals during the first year of
study we have piloted several novel recruitment strategies. participation and then every 6 months thereafter. In general the PD
Working closely with the Michael J Fox Foundation we have and healthy subjects will be assessed identically. PPMI will provide
developed a template of recruitment material and an ongoing both extensive motor and non-motor longitudinal clinical data.
series of focused local events at each site to introduce the study to Subjects will be assessed with the MDS-UPDRS (Colosimo et al.,
both referral physicians and potential study candidates. PPMI will 2010) and with additional clinical tests evaluating cognition,
also be publicized more globally with an international publicity depression, anxiety, autonomic function, sleep and olfaction. All PD
campaign to highlight the need for participation by both PD subjects will undergo longitudinal DAT imaging to monitor the
subjects and healthy subjects. An important strategy has been to change in DAT density during the study. A subset of subjects will
engage referral physicians, a key source of early-untreated PD undergo longitudinal MRI DTI building on recent data suggesting
subjects, to assist with recruitment by fully educating them about that DTI measures may distinguish early PD from healthy subjects
the critical need for biomarkers. We anticipate that sites can (Peran et al., 2010; Vaillancourt et al., 2009). Based on extensive
recruit one early-untreated PD subject each month and one discussion within the PPMI imaging core DTI MRI will be restricted
healthy subject every two months. We therefore expect recruit- to those sites with a uniform camera and image acquisition system
ment to be completed within 24 months. Subject retention to best acquire poolable longitudinal data.
strategies are also a key to study success given that the goal of Longitudinal collection of biospecimens including blood, CSF
PPMI is to identify biomarkers of disease progression. We plan to and urine is a key component of PPMI. For example, CSF samples
provide information to subjects throughout the study through will be collected every six months the first year and every twelve
websites and newsletters and other outreach. We plan to establish months thereafter. Samples will be collected according to
a PPMI community (within the confines of confidentiality) for standardized procedures and will be stored in the PPMI
interested participants and will continue to work closely with the biorepositories in the US (Coriell, Camden, NJ) and Europe (BioRep,
Michael J Fox Foundation and other public relations resources to Milan, Italy). The focus on CSF collection in PPMI is both novel and
promote study retention. ambitious. Recent data from both the Alzheimer Disease Neuro-
imaging Initiative (ADNI) and from single site studies of PD
1.3. Site selection subjects demonstrating that CSF biomarkers such as alpha
synuclein, tau, and ß-amyloid can potentially provide useful and
Twenty-one clinical study sites (sixteen US and five European) novel insights for PD provide the scientific rationale for CSF
with expertise and experience in PD research have been selected collection (Montine et al., 2010; Shaw et al., 2009; Siderowf et al.,

Table 1

Clinical assessments Imaging assessments

Motor MDS-UPDRS Dopamine transporter imaging DaTSCAN

NeuroBehavior Geriatric depression scale (GDS) MRI Volumetric, DTI, resting state
State – trait anxiety inventory (STAI)
Questionnaire for impulsive-compulsive Biospecimen collection
disorders (QUIP)
Cognitive testing MoCA total score
Hopkins verbal learning test – revised Blood Alpha synuclein, DJ1, amyloid,
Benton judgment of line orientation CSF tau, urate
Semantic fluency Urine
Letter number sequencing Genetics DNA, RNA
Symbol digit modalities test
Autonomic testing SCOPA-AUT total autonomic score
SCOPA-AUT sub scores
Sleep disorders Epworth sleepiness scale total score
REM sleep disorder
Olfactory testing UPSIT
632 K. Marek et al. / Progress in Neurobiology 95 (2011) 629–635

Table 2
Scheme for incorporating imaging and biospecimen biomarkers.

Modality Target Ligand

A. Imaging markers
TIER 1 – in use in PPMI in all or sub-set of sites
SPECT Dopamine transporter DaTSCAN
MRI DTI/resting state
TIER II – supportive data and clear rationale for biomarker for PD, but limited data in disease progression
PET VMAT2 DTZB, AV133
PET Glucose metabolism FDG
Ultrasound Substantia nigra
PET Amyloid AV45, BAY94, PIB
SPECT Cardiac sympathetic function MIBG
TIER III – limited data in PD but rationale for biomarker for PD
PET/SPECT PBR – inflammation PK11195, PBR06, PBR111
PET/SPECT Striatal function – MGluR5, CB1, A2a, GlyT1 FPEB, MK9640, FCPyPB
PET/SPECT DASB, MZIENT, INER, MPPF SERT, NET, 5HT1a, 5HT2a
Optical coherence tomography (OCT) Retinal morphology
TIER IV – strong rationale for biomarker for PD, no data for PD
PET/SPECT Alpha-synuclein
PET/SPECT Tau
B. Biospecimen biomarkers
TIER I – plans for use in PPMI. Markers for which there is some evidence for a disease association, CSF alpha synuclein, DJ1, amyloid, tau, blood urate
preliminary data around the detection of the marker in a biochemical assay exist
TIER II – putative markers with weak data correlating to PD, standardized assays exist and are EGF, cytokines, inflammatory markers, glutamine/glutamate
straightforward to study in PD subjects
TIER III – minimal data available, relationship to PD hypotheses and mechanisms of disease exist Proteomics, glutathione, 8-OHdG

2010; El-Agnaf et al., 2006; Mollenhauer and Trenkwalder, 2009). PPMI biologic working group or those unaffiliated with PPMI in the
In addition the recent availability of non-traumatic spinal needles scientific community will be reviewed by the PPMI BRC. In general,
reducing discomfort and shortening the time of collection of CSF PPMI biospecimen will be used to verify or help to validate
provide the practical justification for requiring longitudinal CSF in biomarkers rather than for biomarker discovery. The use of PPMI
PPMI (Peskind et al., 2009). samples to study the same analytes already under investigation in
It is anticipated based on previous clinical trials that most PD PPMI with different assay platforms or methods will be discour-
subjects enrolled in PPMI will require symptomatic treatment aged in favor of proposals that focus on new, albeit validated
within six-twelve months of enrollment (Parkinson Study Group, biomarker assays. All assessment, biosampling, and biomarker
2007; Olanow et al., 2009). The timing of the PPMI study data will be rapidly made public on the PPMI website.
assessments will be adjusted as needed to ensure that subjects
are fully assessed just prior to initiation of symptomatic 1.6. Standardization of data acquisition/training
medications. The goal of PPMI is to identify biomarkers that track
progression both before and after initiation of commonly used PPMI has developed and implemented standardized procedures
symptomatic PD medications. for acquisition of all study data. The goal is that these procedures
will set standards for future multi-center PD studies. The PPMI
1.5. Scheme for incorporating additional PPMI assessments and study cores have established study manuals detailing the
biospecimen analyses collection of data. All sites undergo extensive training in the
MDS-UPDRS, neuropsych and neurobehavioral testing, electronic
PPMI is designed to be a flexible and adaptive biomarker data capture, collection and handling of biospecimen, and
identification consortium. Biomarker development is an iterative acquisition of imaging outcomes before enrolling study subjects.
process and as novel biomarkers are identified by the research This includes web based and in person training. Retraining will be
community they can be tested and further validated in PPMI with implemented when quality control issues occur and/or staff and site
its robust clinical, biospecimen, imaging and genetic data. Working resources change during the study. Comprehensive quality control
groups within PPMI will continue to assess novel clinical, imaging of data provides rapid feedback to sites to ensure that standard
and biospecimen biomarkers throughout the study. The PPMI procedures remain fully implemented and to optimize data quality.
Biospecimen Review Advisory Committee (BRC) formed by experts For example, biospecimen collection parameters (such as volume
within and independent of PPMI will evaluate proposals for collected, aliquoting, time to freezing) are monitored and quality
additional clinical and imaging studies utilizing existing and/or control of specimens received at the biorepository (such as number
adding additional assessments and biosampling. of samples received, visual inspection of samples, hemoglobin in
The imaging working group has identified several potential CSF samples) are assessed and reported to sites as necessary.
imaging biomarkers for consideration for inclusion in PPMI (Table Imaging acquisition has also been standardized at each site and
2). The inclusion of these imaging markers in PPMI is dependent on customized study phantoms are acquired to enable ongoing
the available data supporting their use as PD progression markers, assessment of camera performance (Seibyl et al., 2010). PPMI will
the ease of incorporating these markers in a multi-center study and continue to harmonize with the common data elements establish
the subject burden associated with implementing these imaging by the NIH taskforce (http://www.commondataelements.ninds.-
assessments. Similarly the biologic working group has established nih.gov/PD) in selecting tools for data acquisition.
a scheme for identifying CSF, blood and urine biomarkers for
inclusion in PPMI as additional data is available. The biologic 1.7. Database/statistical analysis
working group will be charged with identifying biomarkers with
adequate assay systems that could be PD progression markers. PPMI is fully committed to rapidly making study data
Proposals to assess PPMI biospecimen generated from either the publically available to the PD research community through a
 K. Marek et al. / Progress in Neurobiology 95 (2011) 629–635 633

web portal. PPMI has developed a comprehensive yet flexible, user The PPMI leadership is committed to rapid data sharing of the
friendly, secure and easily accessible database. The database has PPMI data. Individuals requesting data access will simply need to
been developed and will be maintained by the Laboratory of submit brief identifying information and agree that data analyses
NeuroImaging (LONI) at UCLA. The bioinformatics core will would be provided to the PPMI database upon completion. The PPMI
receive diverse datasets from several sources including the data committee will be informed of any request for data access and
clinical, imaging, bioanalytic and genetics cores and the bior- would be expected to respond with approval within twenty-four
epository. Clinical core data are transferred to the PPMI database hours. The database will provide for data access at several levels of
through nightly transfers including the multi-site clinical expertise with simple data queries possible. The goal would be to
assessments and reconciled biospecimen collection, and inven- stimulate data analyses and data mining and to incorporate some
tory and QC data from the biorepository. Similarly raw imaging of these additional data analyses into the PPMI database.
data and PPMI analyzed imaging data and PPMI biospecimen All requests for biospecimen access would also flow through the
analyses will be transferred from the imaging core and integrated PPMI website. The inventory of biosamples available on the PPMI
into the PPMI database. website will be updated though nightly transfer of these data from
The database will allow a coordinated assessment of clinical, the biorepository. There is a two-step request process for
imaging and biospecimen data. While the specific analysis plans biosample access including a brief request outlining the research
and sample size estimates to assess biospecimen and imaging question, the samples requested and the associated PPMI data
progression biomarkers remain for the most part exploratory, it requested followed by a more detailed proposal. The application
will be possible to examine the data from subject assessments in will be reviewed by the biospecimen review committee (BRC), an
PPMI to compare healthy controls to PD subjects and/or subsets of independent group of biomarker experts. Successful applicants
PD subjects defined by clinical, imaging, genetic, and/or biochemi- will be required to provide their data for entry into the PPMI
cal characteristics. The first set of analyses will compare baseline database to further enhance the PPMI data set.
characteristics among PD subjects and healthy controls. The
second set will use backwards selection to build a model for 2. Discussion
examining short-term change during the first six months for each
progression endpoint of interest. The third set of analyses will While PPMI seeks to discover and validate PD biomarkers for
examine whether the short-term changes in the progression all phases of PD, the most critical need is to identify one or more
endpoints are predictive of changes in long-term endpoints, such biomarkers to monitor disease progression. During the past two
as the MDS-UPDRS score. A ten-fold cross validation procedure decades numerous studies have investigated potential disease
will be used to test the predictive validity of each model. If modifying therapies (Parkinson Study Group, 1993, 2004, 2007;
successful, the final model will provide a subset of one or more Olanow et al., 2006, 2009; Vornov et al., 2006; Schapira et al.,
short-term progression endpoints that are predictive of the 2010). Much has been learned from these studies, but all have
change in one or more of the long-term endpoints. This would failed to clearly demonstrate slowing of disease. While there are
suggest that these short-term progression endpoints are valid many challenges to developing a disease-modifying drug
biomarkers for future studies of interventions in PD patient including limited animal models and uncertain clinical designs,
populations. Finally, each of the first three sets of analyses will be the lack of validated biomarkers remains a major barrier to
repeated comparing subsets of PD subjects. If successful, the final success. Current disease modifying drug development is limited
model from these subset comparisons will determine whether to large long duration very costly studies. Even when the study
some of the short-term progression endpoints are more predictive subjects are enrolled early in disease, substantial neurodegen-
of long-term change in the MDS-UPDRS score for only some eration has already occurred (Braak and Del Tredici, 2008; Eckert
subsets of PD subjects. et al., 2007; Fearnley and Lees, 1990; Langston, 2006; Marek and
While formal sample size estimation is difficult for these Jennings, 2009). Furthermore, the test drug may have symptom-
potential biomarkers we can examine the ability of the proposed atic benefit and/or study subjects are typically treated with
sample size to detect meaningful effects of interest for the available symptomatic medications during the study further
preliminary comparisons of baseline characteristics and univariate confounding the study outcomes (Parkinson Study Group, 1993,
assessments of progression markers across the groups of interest. 2004; Ravina et al., 2005; Shults et al., 2002). Identifying PD
For all calculations, we assume a two-sided alpha level of 0.05 and biomarkers enabling both accurate early detection and objective
a target power of 80%. With 600 total subjects (400 PD patients and monitoring of PD would transform disease modifying drug
200 healthy controls), the PPMI study is adequately powered for a development. The primary goal of PPMI is to identify markers
detectable difference in prevalence of 13% (for a dichotomous that would provide more rapid and sensitive demonstration of
endpoint) and a detectable standardized mean difference of 0.24 change. This might allow a biomarker signal to be assessed in a
(for a continuous endpoint). This suggests that the PPMI study is relatively small sample size short duration Phase 2 study rather
adequately powered to detect effects that would generally be of the large Phase 3 studies that have been unsuccessful. The biomarker
clinical interest. Given that PPMI is designed to identify biomarkers driven Phase 2 studies might provide a rationale strategy to identify
of change that would be useful in clinical studies of disease promising drugs to move to the costly Phase 3. These studies would
modifying drugs, it is reasonable to focus first on those biomarkers also enable more rational drug dosage determination in Phase 2
that demonstrate detectable change during a six-month interval. studies, another major impediment to disease modifying drug
The analysis plan and sample size estimate will continue to be development. Finally, PD biomarkers would also enable more
refined as additional biomarker data becomes available. accurate early detection of study subjects both ensuring that the
study population actually has early PD and enabling study subjects
1.8. PPMI website to be assessed for a longer period before requiring potentially
confounding symptomatic drugs.
The PPMI website, http://www.ppmi-info.org/ is an essential PPMI directly addresses yet another potential flaw in current
component of the study design. The website will provide access to disease modifying trials – that medications tested may be of
both PPMI data and biospecimen and to PPMI protocols, standard benefit to specific sub-groups of PD patients and given our inability
operating procedures, recruitment and retention materials and to define these subgroups effectively the potential drug effect may
study presentations and publications. be washed out and overlooked. For example, biomarkers defining
634 K. Marek et al. / Progress in Neurobiology 95 (2011) 629–635

genetic risk and/or change in biochemical or imaging outcomes Study Cores:

may identify PD subgroups that are rapid or slow progressors and Clinical Coordination Core: Alice Rudolph, PhD, Cindy Casaceli,
at whom specific therapies might be directed and assessed. In Clinical Trials Coordination Center, University of Rochester, NY
these cases medications that might be focused on specific disease Imaging Core: John Seibyl, MD, Principal Investigator, Susan
mechanisms may slow disease for PD subgroups and may Mendick, MPH, Institute for Neurodegenerative Disorders, New
ultimately provide clues for more generalizable PD therapeutics. Haven, CT; Norbert Schuff, PhD, University of California, San Francisco
In addition, PPMI would allow us to combine biomarkers to better Statistics Core: Ying Zhang, University of Iowa, Iowa City, IA
define subgroups to more effectively assess potential disease Bioinformatics Core: Arthur Toga, PhD, Principal Investigator,
modifying drugs. Given the multiple genetic etiologies for PD Karen Crawford, Laboratory of Neuroimaging (LONI), University of
already identified, the marked variability in the loss of dopami- California, Los Angeles
nergic markers measured by imaging at motor symptom onset and BioRepository: Alison Ansbach, MS, Principal Investigator,
the well-known heterogeneity of clinical symptoms in PD onset Coriell Institute for Medical Research, Camden, NJ; Pasquale De
and clinical progression, it is clear that many biomarkers with a Blasio, Michele Piovella, BioRep, Milan, Italy
focus ranging from clinical symptoms to PD pathobiology to Bioanalytics Core: John Trojanowski, MD, PhD, Principal
molecular genetic mechanisms will be necessary to fully map PD Investigator, Les Shaw, PhD, Principal Investigator, University of
progression (Jankovic, 1990; Klein and Schlossmacher, 2007; Beyer Pennsylvania, Philadelphia, PA
et al., 2007). Finally, the longitudinal design of PPMI will further Genetics Core: Andrew Singleton, PhD, Principal Investigator,
allow us to assess the temporal pattern of biomarker change. National Institute on Aging, NIH, Bethesda, MD
Recent data from the ADNI study has developed a working Neuropsychological and Cognitive Assessments: Keith Hawkins,
hypothesis for the timing of AD biomarkers that has informed PsyD, Yale University, New Haven, CT
clinical AD studies (Jack et al., 2010). Likely PD biomarkers also Michael J Fox Foundation: Jamie Eberling, PhD, Deborah Brooks
have a temporal pattern that may be dependent on PD clinical and/
or pathological stage. Of particular interest in PPMI is the Site Investigators and Coordinators:
opportunity to compare progression biomarkers prior to symp- David Russell, MD, PhD, Laura Leary, BS, Institute for
tomatic treatment, at the time symptomatic treatment is required Neurodegenerative Disorders, New Haven, CT; Stewart Factor,
and following long-term symptomatic treatment. Finally the DO, Barbara Sommerfeld, RN, MSN, Emory University of Medicine,
longitudinal study design will enable PPMI to flexibly assess Atlanta, GA; Penelope Hogarth, MD, Emily Pighetti, Oregon Health
biomarkers ‘on the fly’ and discard putative markers that are not and Science University, Portland, OR; Karen Williams, Northwest-
useful while adding other promising markers that may be newly ern University, Chicago, IL; David Standaert, MD, PhD, Stephanie
available. The accessibility of PPMI data and biospecimen will Guthrie, University of Alabama at Birmingham; Robert Hauser,
enable the PD research community to both learn from and MD, Holly Delgado, RN, University of South Florida, Tampa, FL;
contribute to PPMI accelerating this adaptive approach. Joseph Jankovic, MD, Christine Hunter, RN, CCRC, Baylor College of
It is crucial to develop PD progression biomarkers to speed Medicine, Houston, TX; Matthew Stern, MD, Baochan Tran,
development of PD treatments. PPMI is an ambitious, but essential University of Pennsylvania, Philadelphia, PA; Jim Leverenz, MD,
approach for biomarker development and will rely on a partner- Marne Baca, University of Washington, Seattle, WA; Sam Frank,
ship of government, PD foundations, industry and academics MD, Cathi-Ann Thomas, RN, MS, Boston University, Boston, MA;
working cooperatively. As in the highly successful ADNI consor- Irene Richard, MD, Cheryl Deeley, MS, RNC, University of Rochester,
tium for Alzheimer disease, this collaborative and cooperative Rochester, NY; Linda Rees, The Parkinson’s Institute, Sunnyvale,
strategy is necessary to make progress in developing biomarkers CA; Fabienne Sprenger, Innsbruck Medical University, Innsbruck,
that would enhance the potential for success of PD modifying Austria; Elisabeth Lang, Paracelsus-Elena Klinik, Kassel, Germany;
therapeutics (Trojanowski et al., 2010; Aisen et al., 2010; Schmidt Holly Shill, MD, Sanja Obradov, BA, Banner Research Institute, Sun
et al., 2010; Weiner et al., 2010). City, AZ; Hubert Fernandez, MD, Adrienna Winters, BS, Cleveland
Clinic, Cleveland, OH; Daniela Berg, MD, Katharina Gauss,
Appendix A. The Parkinson Progression Marker Initiative University of Tuebingen, Germany; Douglas Galasko, MD, Deborah
(PPMI) Fontaine, RNCS, MS, University of California, San Diego; Zoltan
Mari, MD, Melissa Gerstenhaber, RNC, MSN, Johns Hopkins
Executive Steering Committee: University, Baltimore, MD; David Brooks, MD, Sophie Malloy,
Kenneth Marek, MD, Principal Investigator, Danna Jennings, MD, Imperial College London, UK; Paolo Barone, MD, PhD, Katia
MD, Shirley Lasch, Institute for Neurodegenerative Disorders, New Longo, MD, Universita Federico II, Naples, Italy
Haven, CT; Andrew Siderowf, MD, University of Pennsylvania, ISAB (Industry Scientific Advisory Board):
Philadelphia, PA; Caroline Tanner, MD, PhD (Site Investigator), The Tom Comery, PhD, Pfizer, Inc., Groton, CT; Bernard Ravina, MD,
Parkinson’s Institute, Sunnyvale, CA; Tanya Simuni, MD (Site MSCE, Biogen Idec, Cambridge, MA; Igor Grachev, MD, PhD, Kim
Investigator), Northwestern University, Chicago, IL; Chris Coffey, Gallagher, PhD, GE Healthcare, Princeton, NJ; Michelle Collins, PhD,
PhD, (Statistics Core, PI), University of Iowa, Iowa City, IA, Karl Abbott Laboratories, Abbott Park, IL; Katherine L. Widnell, MD,
Kieburtz, MD, MPH (Clinical Core, PI), Emily Flagg, (Clinical Core, PhD, Abbott Neuroscience Research & Development, Abbott Park,
Project Manager), Clinical Trials Coordination Center, University of IL; Suzanne Ostrowizki, MD, PhD, Paulo Fontoura, MD, PhD,
Rochester, NY; Sohini Chowdhury, Michael J. Fox Foundation, New F.Hoffmann La-Roche, Basel, Switzerland; Tony Ho, MD, Johan
York, NY Luthman, DDS, PhD, Merck & Co., North Wales, PA; Marcel van der
Brug, PhD, Genentech, Inc., South San Francisco, CA; Alastair D.
Steering Committee/Cores: Reith, PhD, GlaxoSmithKline, Stevenage, United Kingdom; Peggy
Werner Poewe, MD (Site Investigator), Innsbruck Medical Taylor, ScD, Covance, Dedham, MA
University, Innsbruck, Austria; Brit Mollenhauer, MD (Site
Investigator), Paracelsus-Elena Klinik, Kassel, Germany; Todd
References
Sherer, PhD, Mark Frasier, PhD, Claire Meunier, Michael J. Fox
Foundation, New York, NY Aisen, P.S., et al., 2010. Clinical core of the Alzheimer’s Disease Neuroimaging
Michael J. Fox Foundation, New York, NY Initiative: progress and plans. Alzheimers Dement. 6 (3), 239–246.
 K. Marek et al. / Progress in Neurobiology 95 (2011) 629–635 635

Beyer, M.K., Larsen, J.P., Aarsland, D., 2007. Gray matter atrophy in Parkinson Parkinson Study Group, 2004. Levodopa and the progression of Parkinson disease.
disease with dementia and dementia with Lewy bodies. Neurology 69 (8), NEMJ 351, 18–28.
747–754. Parkinson Study Group, 2007. Mixed lineage kinase inhibitor CEP-1347 fails to
Braak, H., Del Tredici, K., 2008. Invited Article: nervous system pathology in delay disability in early Parkinson disease. Neurology 69 (15), 1480–1490.
sporadic Parkinson disease. Neurology 70 (20), 1916–1925. Peran, P., et al., 2010. Magnetic resonance imaging markers of Parkinson’s disease
Colosimo, C., et al., 2010. Task force report on scales to assess dyskinesia in nigrostriatal signature. Brain 133 (11), 3423–3433.
Parkinson’s disease: critique and recommendations. Mov. Disord. 25 (9), Peskind, E., et al., 2009. Safety of lumbar puncture procedures in patients with
1131–1142. Alzheimer’s disease. Curr. Alzheimer Res. 6 (3), 290–292.
Eckert, T., Tang, C., Eidelberg, D., 2007. Assessment of the progression of Parkinson’s Ravina, B., et al., 2005. The role of radiotracer imaging in Parkinson disease.
disease: a metabolic network approach. Lancet Neurol. 6 (10), 926–932. Neurology 64 (2), 208–215.
El-Agnaf, O.M., et al., 2006. Detection of oligomeric forms of alpha-synuclein Ravina, B., et al., 2009. A longitudinal program for biomarker development in
protein in human plasma as a potential biomarker for Parkinson’s disease. Parkinson’s disease: a feasibility study. Mov. Disord. 24 (14), 2081–2090.
FASEB J. 20 (3), 419–425. Schapira, A.H., 2004. Disease modification in Parkinson’s disease. Lancet Neurol. 3
Fearnley, J., Lees, A., 1990. Striatonigral degeneration: a clinico-pathological study. (6), 362–368.
Brain 113, 1823–1842. Schapira, A.H., et al., 2010. Rationale for delayed-start study of pramipexole in
Gerlach, M., et al., 2008. Early detection of Parkinson’s disease: unmet needs. Parkinson’s disease: the PROUD study. Mov. Disord. 25 (11), 1627–1632.
Neurodegener. Dis. 5 (3–4), 137–139. Scherzer, C.R., et al., 2007. Molecular markers of early Parkinson’s disease based on
Jack Jr., C.R., et al., 2010. Hypothetical model of dynamic biomarkers of the gene expression in blood. Proc. Natl. Acad. Sci. U.S.A. 104 (3), 955–960.
Alzheimer’s pathological cascade. Lancet Neurol. 9 (1), 119–128. Schmidt, M.E., et al., 2010. The Alzheimer’s Disease Neuroimaging Initiative:
Jankovic, J., 1990. Lower body (vascular) parkinsonism. Arch. Neurol. 47 (7), 728. perspectives of the Industry Scientific Advisory Board. Alzheimers Dement. 6
Klein, C., Schlossmacher, M.G., 2007. Parkinson disease, 10 years after its genetic (3), 286–290.
revolution: multiple clues to a complex disorder. Neurology 69 (22), Seibyl, J., Marek, K., Zubal, I.G., 2010. The role of the core imaging laboratory in
2093–2104. multicenter trials. Semin. Nucl. Med. 40 (5), 338–346.
Langston, J.W., 2006. The Parkinson’s complex: parkinsonism is just the tip of the Shaw, L.M., et al., 2009. Cerebrospinal fluid biomarker signature in Alzheimer’s
iceberg. Ann. Neurol. 59 (4), 591–596. Disease Neuroimaging Initiative subjects. Ann. Neurol. 65 (4), 403–413.
Maetzler, W., Liepelt, I., Berg, D., 2009. Progression of Parkinson’s disease in the Shi, M., Caudle, W.M., Zhang, J., 2009. Biomarker discovery in neurodegenerative
clinical phase: potential markers. Lancet Neurol. 8 (12), 1158–1171. diseases: a proteomic approach. Neurobiol. Dis. 35 (2), 157–164.
Marek, K., Jennings, D., 2009. Can we image premotor Parkinson disease? Neurolo- Shults, C.W., et al., 2002. Effects of coenzyme Q10 in early Parkinson disease:
gy 72 (Suppl. 7), S21–S26. evidence of slowing of the functional decline. Arch. Neurol. 59 (10), 1541–1550.
Marek, K., et al., 2008. Biomarkers for Parkinson’s disease: tools to assess Parkin- Siderowf, A., et al., 2010. CSF amyloid b1–42 predicts cognitive decline in Parkinson
son’s disease onset and progression. Ann. Neurol. 64 (Suppl. 2), S111–S121. disease. Neurology 75 (12), 1055–1061.
Mollenhauer, B., Trenkwalder, C., 2009. Neurochemical biomarkers in the differen- The Lancet, N., 2010. Biomarker promise for Parkinson’s disease. Lancet Neurol. 9
tial diagnosis of movement disorders. Mov. Disord. 24 (10), 1411–1426. (12), 1139.
Montine, T.J., et al., 2010. CSF Abeta(42) and tau in Parkinson’s disease with Trojanowski, J.Q., et al., 2010. Update on the biomarker core of the Alzheimer’s
cognitive impairment. Mov. Disord. 25 (15), 2682–2685. Disease Neuroimaging Initiative subjects. Alzheimers Dement. 6 (3), 230–238.
Olanow, C.W., et al., 2006. TCH346 as a neuroprotective drug in Parkinson’s disease: Vaillancourt, D.E., et al., 2009. High-resolution diffusion tensor imaging in the
a double-blind, randomised, controlled trial. Lancet Neurol. 5 (12), 1013–1020. substantia nigra of de novo Parkinson disease. Neurology 72 (16), 1378–1384.
Olanow, C.W., Kieburtz, K., Schapira, A.H., 2008. Why have we failed to achieve Vornov, J., et al., 2006. GPI 1485, a neuroimmunophilin ligand, fails to alter disease
neuroprotection in Parkinson’s disease? Ann. Neurol. 64 (Suppl. 2), S101–S110. progression in mild to moderate Parkinson’s disease. Mov. Disord..
Olanow, C.W., et al., 2009. A double-blind, delayed-start trial of rasagiline in Weiner, M.W., et al., 2010. The Alzheimer’s Disease Neuroimaging Initiative:
Parkinson’s disease. N. Engl. J. Med. 361 (13), 1268–1278. progress report and future plans. Alzheimers Dement. 6 (3), 202 e7–211 e7.
Parkinson Study Group, 1993. Effects of tocopherol and deprenyl on the progression Whone, A.L., et al., 2003. Slower progression of Parkinson’s disease with ropinirole
of disability in early Parkinson’s disease. N. Engl. J. Med. 328, 176–183. versus levodopa: the REAL-PET study. Ann. Neurol. 54 (1), 93–101.