The Unknown Unknown of Biorepositories: Overcoming Obstacles

1
The Unknown Unknown of BiorepositoriesOvercoming
Obstacles

• Benjamin M. Greenberg, MD, MHS
• Director, Transverse Myelitis and Neuromyelitis
  Optica Program
• University of Texas Southwestern Medical Center

2
Potential Roles of a Biorepository

• Understand biology of a disease
• Understand response to therapy
• Create diagnostic technologies
• Identify prognostic biomarkers
• Identify new conditions

3
Multiple Sclerosis Throughout History

• 1868 Jean-Martin Charcot describes in detail,
  clinical multiple sclerosis and relates it to
  white matter lesions.

4
(No Transcript)
5
Eugene Devic, 1858-1930
What devic described was a monophasic illness of
bilateral ON and myelitis in quick succession. He
saw one case, but his fellow Fernand Gault
(1873-1936), reviewed sixteen cases, some had
relapses and published in 1894
6
Neuromyelitis Optica Revisited

• In 1999, Wingerchuk, et. al., reviewed 71 cases
  from the Mayo clinic and defined NMO as a
  relapsing disorder with longitudinally extensive
  transverse myelitis (LETM) and normal brain MRI
• That turns out to be inaccurate

7
How Should We Define Diseases?

• Clinically
• Non-specific lumping of various syndromic
  phenotypes into one heterogenous category
• Pathologically
• Ignores potential phenotypic variation
• Multiple biologic pathways result in same
  pathology
• Molecularly
• Limited markers available
• Incomplete understanding of biology

8
The NMO BattlefieldBiology and Semantics

• Is NMO a spectrum of Multiple Sclerosis?
• Patients dont seem to respond to MS therapy
• 40 died within 5 years of respiratory failure
• Preferential effect on spinal cord and optic
  nerves
• When brain is involved, unique patterns can be
  seen
• Can have unusual CSF patterns compared to MS
• Yet, some patients can have clinical and
  radiographic patterns that initially look like MS

9
Incomplete NMO Definitions

• Optic neuritis and partial spinal lesions, were
  usually not NMO, but rather MS even with normal
  brain MRI.
• If you used ON, TM and normal brain scan would be
  wrong 50 of the time. So more specific criteria
  (LETM) was important for distinguishing MS from
  NMO

10
NMO IgG ArrivesAnnounced at 2004 AAN

• Lennon et al lancet 2004 364 2106-2112
• Antibody binds to pia and virchow robin spaces,
  distribution suggested of antigen in perivascular
  region.
• Claudia Lucchinetti identified pathology around
  blood vessels

11
The Alchemy of Biorepositories Turning Samples
into Gold

• A samples value is based on several factors
• When was the sample obtained
• How was the sample obtained
• How was the sample processed
• How was the sample stored
• What prospective data is subsequently gathered
  about the patient.

12
Example Patient
13
The Ideal Biorepository
14
Reality
15
There are Two Kinds of Biorepositories
16
The Classic Approach to Disease Specific
Biorepositories

• Cancer
• Patient suspected of cancer based on symptoms or
  tests.
• Tissue acquired and banked
• Genetic Conditions
• Patient diagnosed and DNA acquired
• Rare Non-Cancer/Non-Genetic Diseases
• The repository is research lab specific
• Genetics research labs acquire DNA
• Infectious Disease labs draw serum
• Nutrition labs draw whole blood

17
The Donald Rumsfield Analysis of Biorepositories
18
Reports that say that something hasnt happened
are always interesting to me, because as we know,
there are known knowns, there are things we know
we know. We also know there are known unknowns,
that is to say, we know there are some things we
do not know. But there are also unknown unknowns
the ones we dont know we dont know.

• -Donald Rumsfeld, Frmr. Secretary of Defense
• February 12, 2002

19
Translation We Dont Know The Questions That Are
Going to Be Asked Tomorrow

• Everything cant be designed around hypothesis
  driven research

20
Remodeling Repositories

• Most diseases are the result of the following
  formula
• Disease Genetics Environment Timing

21
Examples of Discoveries Since 2000

• siRNA
• miRNA
• Granulebacter bethesdensis in patients with CGD
• Metapneumovirus in children with hospitalizations
  for respiratory infections.
• Th17 immune cells
• 3 articles in 2003
• Over 700 in 2009

22
The Broken Repository Model

• The silo approach has failed
• Every institution has their own program.
• Hypothesis driven research is the only way to get
  funding
• Academia is not inherently collaborative
• Intellectual Property issues create obstacles.

23
Increased Challenges For Rare Disease Repositories

• Too few numbers
• Too few points of entry into study
• Too little infrastructure for longitudinal
  expenses.
• Solution Combining repositories with registries
  and partnering up with common conditions

24
Finding Common Ground

• Much can be learned by studying outliers.
• Comparing related, albeit distinct diseases
  allows for new discovery
• Neuromyelitis Optica versus Multiple Sclerosis
  versus Myasthenia Gravis
• The notion of controls
• Allows for collaborations between common and
  rare disease groups.

25
Collaborative Solutions
26
Status

• 10 enrolling sites
• Over 1700 cases
• 1393 MS
• 110 NMO
• 141 TM
• 500 controls

Overrepresentation of the rare diseases!!!!
27
Broad Sample Type Acquisition

• Whole Blood
• PaxGene Tubes
• Serum
• Plasma
• Cells

28
Removing The Silo Approach to Research

• The ACP repository model not only collects broad
  amounts of samples and matched clinical data
• Samples provided to researchers without
  IP/publication restrictions
• Samples provided to researchers on the condition
  that data generated in research is returned to
  the repository.

29
Third Party Biorepositories Advance Discovery
30
Third Party Biorepositories Advance Discovery
31
Conclusions

• The Unknown Unknowns require broad sample and
  data set acquisitions
• Rare diseases need to link with common ones
• Third party repository stewards are required to
  breakdown the barriers of complex research
• Public policy
• every visit is a research visit