Athersys, Inc.

R&D / Investor / Analyst Day 
Intercontinental Hotel
Intercontinental Hotel

April 8th, 2009
 Forward Looking Statements

The statements and discussions contained in this presentation that are not historical facts constitute
forward‐looking statements, which can be identified by the use of forward‐looking words such as
“believes,” “expects,” “may,” “intends,” “anticipates,” “plans,” “estimates” and analogous or similar
expressions intended to identify forward‐looking statements. These forward‐looking statements and
estimates as to future performance, estimates as to future valuations and other statements contained
herein regarding matters that are not historical facts, are only predictions, and that actual events or
results may differ materially. We cannot assure or guarantee you that any future results described in
this presentation will be achieved, and actual results could vary materially from those reflected in such
forward‐looking statements.

Information contained in this presentation has been compiled from sources believed to be credible and
reliable. However, we cannot guarantee such credibility and reliability. The forecasts and projections of
events contained herein are based upon subjective valuations, analyses and personal opinions.

This presentation shall not constitute an offer to sell or the solicitation of an offer to buy any securities.
Such an offer or solicitation, if made, will only be made pursuant to an offering memorandum and
definitive subscription documents.

2
Today’s Objectives

(Re‐) Introduce you to our exciting cell‐therapy 
product platform
product platform
Update you on our pharmaceutical program status 
and strategy
Illustrate why we are well positioned to create 
substantial value
– Strong science foundation
– Strong collaboration network
– Smart & efficient development approach
S & ffi i d l h
– Portfolio of opportunities
– Cash / investor support
Cash / investor support

3
Today’s Agenda

High Level Strategic Overview
Value creation strategy, product development philosophy & criteria
Value creation strategy, product development philosophy & criteria

Overview of key programs & core technologies
MultiStem
™ Product profile, manufacturing, distinctiveness relative to other cell types, MOA’s, IP summary
™ Cardiovascular disease
™ Stroke and other neurological indications
™ Immunological conditions
Pharmaceutical programs
™ 5HT2c agonist program for obesity
g p g y
™ H3 antagonist / inverse agonist program for conditions affecting cognition, attention, wakefulness

Financial Update

Q & A
4
Senior Management Team

Gil Van Bokkelen, Ph.D.
™ Chairman & CEO
Chairman & CEO

William (B.J.) Lehmann, J.D.
™ President & COO
John Harrington, Ph.D.
™ Chief Scientific Officer

Robert Deans Ph D
Robert Deans, Ph.D.
™ Senior Vice President, Regenerative Medicine
Laura Campbell
™ Vice President, Finance
Deborah Ladenheim, Ph.D.
™ Vice President, Regulatory Affairs
g y

5
Company Highlights

Emerging portfolio of “best‐in‐class” product candidates and technologies

Multiple clinical trials initiated with MultiStem, a distinctive biologic
Highly standardized “Off‐the‐shelf” cell therapy product, produced at scale
Ad i i t d ith t ti
Administered without tissue matching or immune suppression
t hi i i
Multiple disease indications in development – multiple mechanisms of benefit
Frost & Sullivan 2008 Product Innovation of the Year Award

Attractive small molecule preclinical pipeline
ll l l l l l
Focused on CNS/metabolic related indications, including obesity, cognition and others, in 
two  program areas: 5HT2c agonists and H3 antagonists
Recent suspension of development of ATHX‐105 (5HT2c agonist) for obesity 

Public company with strong cash position
NASDAQ: ATHX
NASDAQ: ATHX 

6
Our Business / Value Creation Strategy

Efficiently develop a portfolio of programs with best‐in‐class potential
Defining existing fundamental challenges in areas of activity
Committed to maintaining lean operational infrastructure / modest core burn

Apply a “Fast Follower” strategy in multiple areas
Proven strategy that offers substantial potential to reduce risk and development 
costs, can lead to superior value creation over time  
Leverage prior knowledge, validation, development efforts of others to produce a 
better safer and/or more convenient product
better, safer and/or more convenient product
Multiple potential advantages to being “best but not first” (but also leveraging 
“early mover” opportunity in areas where it makes sense)

Portfolio based approach enables development & partnering flexibility
Substantial value creation is predicated  on successful product development
Given attrition in drug development, most “one shot wonders” do not survive
With our capabilities and approach we have multiple shots on goal and can 
evaluate, pursue multiple partnering opportunities as we advance
7
Partnerships and Collaborations

Our technologies have broad potential application

C ff i l df i h h ll b i
Cost effectively expand footprint  through collaborations
Focused collaborations with academic and clinical experts that possess deep 
knowledge and understanding of specific disease areas 
Highly motivated KOL’s that seek to evaluate MultiStem in validated disease 
models to assess potential therapeutic relevance, deepen knowledge of biology
Jointly design studies that are implemented by the investigator with input, 
y g p y g p ,
oversight by ATHX (typically we provide cells and brainpower, little to no 
funding, maintain rights to IP)

Partnerships with companies
Growing awareness of opportunity in cell therapy, regenerative medicine
Partnering will be a key value driver for us, and is a major strategic priority
Partnering will be a key value driver for us, and is a major strategic priority

8
 Multi‐Institution Collaborative Center of Excellence 
in Cell Therapy

Founding Members:

Deep research / Infrastructure,

Clinical network
translational capability support, processing
9
 Selected Collaborating Clinical Research Centers

Note: not an endorsement 10
 Selected Collaborating Research Institutions 

Note: not an endorsement 11
Corporate Partners & Licensees  
Growing Interest in Stem Cells & Regenerative Medicine

November 2008 – Pfizer announces launch of Regenerative Medicine 
Centers
– $100 million program to develop therapies, focused in Cambridge UK (brain / 
sensory) and Cambridge MA (heart disease / diabetes)
November 2008 –
November 2008 Genzyme and Osiris announce partnership to 
Genzyme and Osiris announce partnership to
commercialize Prochymal and Chondrogen (MSC)
– $130 million upfront/committed, $1.25 billion in potential milestones
– Osiris to commercialize in U.S.; Genzyme in RoW
July 2008 – GSK announces collaboration with Harvard Stem Cell 
Institute
– $25 million research partnership
June 2008 – Pfizer announces investment in EyeCyte
– Treatment of eye diseases (e.g., diabetes
Treatment of eye diseases (e.g., diabetes‐retinopathy)
retinopathy) with adult stem cell (EPC)
with adult stem cell (EPC)

13
MultiStem®: Biologic 
Product Platform

14
 MultiStem®: A drug‐like cell therapy with 
potential application in multiple disease areas
i l li i i l i l di

15
Historical Limitations to Stem Cell Therapy

Requirement for close Donor – Recipient tissue matching
Necessary to avoid transplant rejection, reduce incidence of Graft vs. Host Disease 
Use of immunosuppressive drugs pose additional risks, complications

Lack of ability to scale production of cells

Historically => one donor for each recipient – logistically difficult and very costly
Biological limitations of most cells prevent large scale / consistent production
i l i l li i i f ll l l / i d i

Mechanistic focus has been primarily cell / tissue replacement

Most cell types can produce limited repertoire  of more differentiated cells
Goal has been to replace lost or damaged cells (e.g. HSC transplantation)
Safety
Ectopic tissue, tumor / teratoma formation, GVHD
16
What is MultiStem?

MultiStem is a proprietary cell therapy product that is produced 
is a proprietary cell therapy product that is produced
by isolating a patented class of early progenitor stem cells 
(Multipotent Adult Progenitor Cells) obtained from bone marrow 
or other tissue and organ systems
or other tissue and organ systems.

17
MultiStem®:  Best‐in‐Class Potential Among Cell Therapies

“Off the shelf” administration
No tissue matching needed
No tissue matching needed
Non‐immunogenic ‐ No immunosuppression required

Well defined, FDA‐approved manufacturing process in place (with Lonza)
Banked product, highly characterized
Large scale production / yield (100k’s to millions of doses possible from a single donor –
represents a substantial advantage over alternative stem cell platforms)
Well validated & consistent safety profile based on extensive preclinical testing

Multiple potential mechanisms of therapeutic benefit
A drug like therapy that is dynamically responsive
Therapeutic effect primarily factor mediated: anti‐inflammatory / immunomodulatory,    
cytoprotective, trophic & growth factors, angiogenic / vasculogenic
Di
Direct cell / tissue replacement plays a minor role
ll / i l l i l

Leading IP position for pluripotent, multifunctional non‐embryonic stem cells
18
 Multiple Potential Mechanisms of Benefit

SHIFTING BALANCE IN REPAIR PROCESSES

IMMUNOMODULATION
Certain Secreted Proteins
INFLAMMATION
REDUCTION
Cytokines / Chemokines
CYTOPROTECTION
Growth Factors

Other Proteins / ASCULOGENESIS
ANGIO‐ / V

MultiStem
MultiStem® cells
cells
express  multiple 
CELLULAR REGENERATION / 
/
therapeutically‐relevant  REPLACEMENT
proteins at significant levels 
& can dynamically regulate 
other cell types
th ll t
Recent Investigational New Drug Applications

3 MultiStem IND’s authorized by FDA within ~12 months (12/07 – 12/08)

IND Protocol Title
IND Protocol Title Key elements

Phase I, Multicenter, Dose‐Escalation Trial Evaluating Maximum‐ IV‐delivered product

Tolerated Dose of Single and Repeated Administration of  Phase I: open label – SAD, 
Allogeneic MultiStem® in Patients with Acute Leukemia, Chronic  MAD
Myeloid Leukemia, or Myelodysplasia
l id k i l d l i Adjunctive to BM/HSC 
dj i /
transplant
Phase I, Multicenter, Dose‐Escalation Trial Evaluating the Safety of  Catheter‐delivered product
Allogeneic AMI MultiStem® in Patients with Acute Myocardial  Phase I: open label (w/ 
Infarction registry)
Phase I, Multicenter, Dose‐Escalation Trial Evaluating the Safety of  IV‐delivered product
Allogeneic MultiStem® in Patients with Ischemic Stroke Phase I: blinded, placebo 
controlled

Working within Master File framework enables efficient advancement of

multiple
lti l opportunities
t iti

20
Athersys Stem Cell Patent Portfolio

Freedom to operate
27
27 patent families
f ili
– 11 granted patents, approximately 110 applications in global prosecution → platform 
coverage 2019‐2028
– Covers cell compositions, methods of making them, methods of using them
– New IP generated internally, and through ongoing outside collaborations

Proprietary methods techniques: trade secrets

Proprietary methods, techniques: trade secrets
Active management, with view to competitive efforts
Make technology available to collaborators on a controlled basis for 
Make technology available to collaborators on a controlled basis for
targeted research, certain grant funded projects

21
Overview of MultiStem® Production Process

Lot Release & Product Characterization 
Testing
Sterility
Potency
Purity and Viability
Stable Cytogenetics

Absence of tumorigenic potential in vivo

22
MultiStem: Additional Safety Studies 

▲ GLP Toxicology and Clinical Pathology (2 week, 4 week)
ƒ Studies indicate no evidence of acute toxicity or abnormal clinical pathology 
▲ Genetic Stability and Tumorigenicity Testing
ƒ Karyotypic stability
ƒ MCB / clinical product tested in standard Nude mouse tumor models (both i.v. / s.c.)
▲ Long Term GLP Histopathology Analysis (one year for stroke)
ƒ Extensive histopathology analysis of animals receiving clinical grade MultiStem 
indicates no evidence of tumorigenicity or ectopic tissue after one year
ƒ No other abnormalities or other adverse events noted
▲ Immune Sensitization Analysis
ƒ Single
Single or repeat administration (5x) of MultiStem does not cause immune 
or repeat administration (5x) of MultiStem does not cause immune
sensitization or abnormal clinical pathology
▲ Gene Expression, Protein Expression and SNP Array Analysis
ƒ No evidence of variability between working cell banks and production runs after 
y g p
significant expansion of clinical grade cellular product

23
Advantaged Expansion Profile for MultiStem

Cell Expansion over Time
Human Cells Isolated From Same Donor
Doublings

Expansion profile 
p p
enables significant
manufacturing 
advantage, using 
Master / Working 
/ g
Cell Bank 
approach 

Days

24
Master Cell Bank Creation

Isolation / expansion research grade product
– Routine internal production of large scale research banks of human, rat and 
p g ,
porcine cells
– Consistent and validated manufacturing process 
™ More than 50 donor marrows processed with ~75‐80% success rate
– Continuing process development & optimization with Lonza, but already have 
a commercially viable process
Clinical grade product
‒ Two MCB established under GMP, fully characterized to enable production 
runs for clinical use (provides ample basis for producing material for multiple 
clinical studies, programs)
‒ “Clinical grade” donors: meet stringent physical health / screening in 
“ li i l d ”d i h i lh l h/ i i
accordance with Good Tissue Practices (GTP), with history of prior marrow 
donations
‒ Extensive characterization / testing (e.g., sterility, safety, viability, purity
Extensive characterization / testing (e.g., sterility, safety, viability, purity 
stability)

25
 Cell Expansion from Master Cell Bank (MCB)

Cell Expansion from MCB
Averages from long‐term expansions of clinical product

D bli *
Doublings*
Stability
50 (e.g., growth, 
40 cytogenetics) 
30 well beyond 
ll b d
targeted cell
20 MultiStem expansion 
range expansion range  
10 2
for clinical 
1
0 product
1 4 7 10 13 16 19 22 25 28 31 34 37 40 Days
*  Additional doublings from MCB (itself expanded ~18 doublings from donor isolation) 

1 Cell expansion range: production run direct from MCB
2 Cell expansion range: production run from Working Cell Bank 
established from MCB
t bli h d f MCB

Confidential 26
Consistent & Well Validated Production Process

Consistent product
Production Runs (n=8) from MCB • Consistent morphology, marker profile
Aggregate Avg Doublings • Absence of hematopoietic contaminants
• Secreted protein expression
S t d t i i
1,000‐fold expansion potential
15 Stable product
10 • Stable cytogenetic profile
• Consistent potency, viability
5
Ongoing product shelf‐life studies
0 • 2‐year + real‐time data (viability, cell 
count, sterility, flow)
Day 3 Day 6 Day 9 • 5‐year stability targeted

27
 Consistent Expression of Relevant Factors

Protein Concentration, Measured by ELISA
Average Concentration / Standardized Dose

Growth Factor A Chemokine B Interleukin C

5 180
Concentration [pg/ml] 

4.5 14
160
Staandardized to Dosee

4 12
140
3.5
120 10
3
2.5 100 8
2 80
6
1.5 60
4
Avg. C

1 40
0.5 20 2
0 0 0
X7032    x7057    x7792     x7793 X7032    x7057    x7792     x7793    X7032    x7057    x7792     x7793   
Production lot

Expression of proteins relevant to potential therapeutic benefit / 
product identity → provides foundation for potency assay 
development

28
 M l iS
MultiStem®
®

A Distinctive Product  

29
The Distinctive Profile of MultiStem

MultiStem differentiated from mesenchymal
differentiated from mesenchymal stem cells (MSC), 
stem cells (MSC)
e.g.,
– Substantial expansion / production advantage
– Distinctive phenotypic and functional profile / characteristics
– Some distinctive mechanisms of action, though some potentially common 
mechanisms as well
MultiStem differentiated relative to other stem cell types, e.g., 
– Immuno‐privileged, enabling allogeneic and potential off‐the‐shelf usage 
– Robust biological activity, multiple potential mechanisms of benefit
R b t bi l i l ti it lti l t ti l h i fb fit
– Manufacturability of cell product

30
Distinctive Transcription Profiles for MultiStem and MSC

ATH113
ATH114
ATH102
ATH104
ATH107
MultiStem
ATH103
ATH106
ATH105
ATH117
ATH116
ATH115
ATH110 MSC
ATH111
ATH112
0.125 0.100 0.075 0.050 0.025 0

• Blinded analysis of gene expression demonstrates distinctive transcription profiles for 
MultiStem and MSC
• X‐axis represents the correlation value.
X i t th l ti l

31
Distinctive Protein Expression

MultiStem  MSC

P
Protein 1
i 1

MultiStem  MSC

Protein 2

32
 33
0.12

I
0.038
Heatmap of Differentially Expressed Genes: MSC, MultiStem

Parameters for above heatmap: MultiStem Signal > 500 and MSC < 100; MSC >500 and MultiStem <100
I
0.015

Correlation > 0.99 within MultiStem and MSC samples, ~0.85 between MultiStem and MSC

I II
0 0.009

go_component: nucleus
n
Sampling of 48 genes showing substantial differences between MultiStem and MSC

go_component: microfibril
m
go_component: in
ntegrin complex
go_component:eextracellular
go_component: ly
ysosome
go_component: in
ntegral to membrane
go_component: baseme
b nt membrane
go_component: uncha
u racterized
go_component:ccytoplasm
go_component:eextracellular matrix
go_component: uncha
u racterized
go_component:GGolgi apparatus
go_component: membrane
m
go_component: uncha
u racterized
go_component:ccytoplasm
go_component:eextracellular matrix
go_component:eextracellular
go_component:eextracellular matrix
go_component:ccytoplasm
go_component: uncha
u racterized
go_component:eextracellular matrix
go_component:eextracellular matrix
go_component: in
ntegral to membrane
go_component:ccytoplasmicvesicle
go_component:eextracellular
go_component: nucleus
n
go_component: uncha
u racterized
go_component:ccytoplasm
go_component:ccytoplasmicvesicle
go_component:eextracellular matrix
go_component: nucleus
n

~48,000 transcripts surveyed 
go_component:ccytoplasm
go_component:riibosome
go_component: nucleus
n
go_component:riibosome
go_component: nucleus
n
go_component: nucleus
n
go_component: ly
ysosome
go_component: uncha
u racterized
go_component: nucleus
n
go_component:pperipheral plasma membrane
go_component: uncha
u racterized
go_component: nucleus
n
go_component: nucleus
n

•
go_component:eextracellular
go_component:ccytoplasm
go_component: nucleus
n
go_component: uncha
u racterized

MSC 1
MSC 2
MSC 3
MultiStem1
MultiStem2
MultiStem3
MultiStem4
 Multiple Potential Mechanisms of Benefit

SHIFTING BALANCE IN REPAIR PROCESSES

IMMUNOMODULATION
Certain Secreted Proteins
INFLAMMATION
REDUCTION
Cytokines / Chemokines
CYTOPROTECTION
Growth Factors

Other Proteins / ASCULOGENESIS
ANGIO‐ / V

MultiStem
MultiStem® cells
cells
express  multiple 
CELLULAR REGENERATION / 
/
therapeutically‐relevant  REPLACEMENT
proteins at significant levels 
& can dynamically regulate 
other cell types
th ll t
Selected Research on Mechanism: Immunomodulation 

Broad potential relevance across multiple disease areas
MultiStem are immunoprivileged
are immunoprivileged
– Do not elicit T‐cell response in vitro (e.g., MLR)
– Immunosuppression not required with MultiStem administered allogeneically or 
xenogeneically
i ll (human → rodent) for benefit in acute MI or stroke models
(h → d t) f b fit i t MI t k d l
– MultiStem serial administration safe → no evidence of allo‐antibody / T‐cell 
sensitization response
MultiStem actively modulates immune response
actively modulates immune response
– Dose dependent downregulation of allo T‐cell response in vitro (e.g., MLR)
– Reduction in immune response in vivo (e.g., GvHD models, AMI models)
– Immune response modulation appears to operate locally at site of injury –
I d l i l ll i fi j not a 
global immunosuppressive effect (e.g., ovalbumin studies, homing) 
Multiple pathways involved
– Anti‐inflammatory cytokines
– Other unique mechanisms
35
 MultiStem® Immuno‐Privileged In Vitro

Mixed Lymphocyte Reaction MultiStem does not elicit In Vitro T-

Cell Response in MLR Studies
Donor 1 Cells
(Rare alloreactive T-
Donor 2 Cells
cells in red)

Mixture

Allogeneic T-
cell controls

Recognition of allogeneic cells

causes T-cell activation and
proliferation
T-cells don’t react
Proliferation measurable by Self to self
increase DNA synthesis to MultiStem (MAPC)

36
 MultiStem® Immunosuppress Allogeneic T Cells

MultiStem (like MSC) Exhibits  Dose Dependent Suppression of Allogeneic T Cell 
Immunosuppressive Effects On MLR  Response in MLR (Lewis rat)
(human)
180,000

160,000

140,000 None 0.03x10^5

120,000 0.06x10^5 0.125x10^5

ymidine counts
100,000
0.25x10^5 0.5x10^5

80,000
1 10^5
1x10^5 2 10^5
2x10^5

3H-thy
60,000

40,000

20,000

0
R (Lewis )+ S (DA) R (Lewis ) No responder or s tim ulator

MAPC (MultiStem)   Dose Dependent 
Suppresses Immune  Effect
Response

37
MultiStem + Activated T‐Cells Result in Gene Expression Changes 

MultiStem response when exposed to Activated PBMC’s Activated PBMC response when exposed to MultiStem
MultiStem+Ab, MultiStem+Ab + activated PBMC Activated PBMC, activated PBMC + MultiStem+Ab

MultiStem response to PBMC activation agent (control) Activated PBMC’s  (change relative to unactivated PBMC’s)

MultiStem, MultiStem + T‐cell activator (“MultiStem+Ab”) PBMC, PBMC plus T‐cell stimulator (“activated PBMC”) 38
Dynamic Regulation of Activated T Cells 

3 chemokines

13 chemokines 3 cytokines
Activated T Cells
5 cytokines
5 cytokines >10 receptors

>25 secreted factors >20 intracellular proteins

>10 secreted factors
10 t df t 3 intracellular proteins
3 intracellular proteins

>40 receptors

3 receptors
MultiStem
>40 intracellular factors
> 5 fold difference in expression
> 5‐fold difference in expression
>10 intracellular factors
39
Factor‐driven Modulation of T Cell Response by MultiStem

Evaluation of Contact Independent / Dependent Inhibition 
in T‐Cell Proliferation Assay 
3H‐Thymidine Incorporation
120000

100000
dine uptake

80000

60000
Thymid

40000

20000

40
MultiStem Inhibits Leukocyte Extravasation

MultiStem associated with down‐regulation of key factors limiting cell surface expression 
of specific receptors, and inhibiting extravasation
f p f p , g into areas of inflammation
f f

Leukocyte extravasation:
• Contributes to inflammation 
and tissue damage (e.g. in 
regions of ischemia)
• Occurs mainly in post‐
capillary venules (minimized 
hemodynamic shear forces)
• Includes several steps, 
including chemo‐attraction, 
rolling adhesion, tight 
adhesion, (endothelial) 
transmigration 
• Process halted whenever any 
Process halted whenever any
of these steps is suppressed

41
 MultiStem Reduces Adhesion Receptor Expression on 
Activated Lymphocytes
FACS profiles of activated lymphocytes co‐cultured for 72 hours with MultiStem®
PBMC PBMC + CD3/28
PBMC PBMC + CD3/28
+ MultiStem®
+ MultiStem + MultiStem®
+ MultiStem

3% 2% 23% 2%
CD4
4

3% 3% 11% 1%
CD8

CD15s

Y-axes (identified immune cell markers), x-axis (certain adhesion receptor)

42
In Vivo Studies Confirm Favorable Immunological Profile

pp
Immunosuppression not q
required with MultiStem used 
allogeneically or xenogeneically (human → rodent) for 
benefit in acute MI or stroke models
MultiStem serial administration safe → no evidence of allo‐
antibody / T‐cell sensitization response
MultiStem appears to modulate immune / inflammatory
MultiStem appears to modulate immune / inflammatory 
response regionally, not globally
– No interference with systemic immune response, as evaluated in ovalbumin 
y p ,
antigen challenge studies

– MultiStem homes / accumulates in sites of injury in preclinical animal 
models 

43
MultiStem Non‐Interference with Systemic Immune Response 

Ova-Antibody Measurement
Designs 3

25
2.5

ut
min Elisa Read-ou
• Healthy buffalo rats immunized IP  2
with ovalbumin (OVA)
1.5
• Antibody study
1 PBS
S
– Multiple MultiStem
Multiple MultiStem injections and 
injections and

Ovalbum
MultiStem
evaluation points over time 0.5
Control
• T‐cell study 0

– Single IV MultiStem injections 1/100 1/200 1/400 1/800 1/1600 1/3200 1/6400 1/12800

Serum Dilution

Results 250000
T-Cell Responses

PBS
• OVA‐Antibodies: no difference  3H-Thymidine (CPM) 200000 MultiStem
between MultiStem treated and
between MultiStem‐treated and  Naïve
150000
PBS control groups
• T‐cell response:  no difference  100000

between systemically‐treated  50000
M ltiSt
MultiStem and PBS groups
d PBS
0
Day 3 Day 4
Day of Measurement
44
Pathophysiology of Neuronal Retraction Following Injury

Immediate

2 days

14 days

28 days

Dex‐TR = Neurons
ED‐1 = Macrophages
CSP= Scar (site of injury)

Rat spinal cord crush model

45
Inhibition of Pathways of Inflammation & Neuronal Damage

MultiStem modulates multiple mechanisms that can directly contribute to reducing inflammation 
and neurological damage, and displays distinct differences relative to other cell types.
d l i ld d di l di ti t diff l ti t th ll t

Figure modified from Jeyakumar et al., Nat Rev Neurosci 6:713-725 (2005) 46
Clinical Development

47
Focused Product Development Approach

o Chronic ischemia / CHF
o Peripheral vascular disease
MultiStem® Treatment 
Treatment o Traumatic brain injury & related
Traumatic brain injury & related
Acute/Ischemic 
Injury o Other Neurological Indications
o Other ischemic injury (e.g., kidney)
o Acute Myocardial Infarction 
(Ph 1 ongoing)
o Ischemic Stroke (IND authorized)

o Inflammatory Bowel Disease
Immune System  o Transplantation
Modulation
o Diabetes (type 1)

o HSC / Bone Marrow Transplant  Support /  o Multiple Sclerosis
GVHD (Ph 1 ongoing) o Other autoimmune disorders
o Other Neurological Indications
Next generation 
opportunities
Other themes, e.g., protein deficiencies, 
bone growth
48
MultiStem for Acute 
Myocardial Infarction
d l f

49
MultiStem®:  Acute Myocardial Infarction

AMI remains a major area of need for improved therapies

865,000 heart attacks annually in the U.S.
156,000 deaths
Significant incidence of progression to CHF
Significant incidence of progression to CHF

Local (catheter) delivery of MultiStem following heart attack

Reduces inflammation‐related damage and promotes revascularization
d i fl i l dd d l i i
Also exploring administration via i.v. 

MultiStem demonstrated safe and effective in multiple pre‐clinical models

MultiStem demonstrated safe and effective in multiple pre‐clinical models

IND approved, clinical trial initiated with co‐development partner (Angiotech)

50
Innovative Approach to Treating Heart Conditions

Administration of “off‐the‐shelf” cellular biologic to improve outcomes and function
– St d di d
Standardized product (administered without matching, immuno‐suppression agents)
d t ( d i i t d ith t t hi i i t)
– Production of trophic factors likely driver of benefit
• Stimulate revascularization and improved circulation 
• Intervene in inflammatory processes
• Override processes of cell / tissue decline, contribute to endogenous tissue regeneration

Efficient delivery to injury site with micro‐infusion catheter
– Administration of cell product into perivascular region
Administration of cell product into perivascular region
– Relative ease of use, comparable to standard angioplasty
– Also exploring IV‐delivery as complementary approach (if multiple treatment believed to be beneficial)  
or alternative

51
Extensive Cardiovascular Pre‐clinical Development Efforts

Cell Safety / Biodistribution
– Absence of acute infusional toxicity (e.g., mortality, lung embolism, fever, granuloma) 
Absence of acute infusional toxicity (e g mortality lung embolism fever granuloma) – rats, syngeneic & 
rats syngeneic &
allogeneic; human cells in NOD/SCID mice; human & pig cells in nude mouse studies
– Absence of tumor or ectopic tissue formation in all models – nude mice i.v. & s.c.; NOD/SCID

Pre‐clinical proof‐of‐concept: allogeneic utility and dose
– Rodents – permanent ischemia (van’t Hof et al, Cytotherapy, 2007)
– Pigs – permanent ischemia, direct injection (Zeng et al, Circulation, 2007)
– Pigs – transient ischemia, catheter‐delivery, exploring administration timing, alternative catheters
– Drug sensitivity study
Drug sensitivity study

Delivery device performance and safety
– Biocompatibility studies show high viability and efficient delivery
– Catheter deliver options tested in pig model for optimized selection
Catheter deliver options tested in pig model for optimized selection
– Stent and trauma studies showed safe delivery in models reflecting human disease

• GLP study
– Pigs –
g transient ischemia, catheter‐delivery
, y

52
Rat LAD Ligation Model – Potential Mechanisms

Rat Ischemia Models ‐ direct injection in infarct zone

Neutrophil count in infarcted hearts Reduction in neutrophil tissue elastase

MultiStem
p=.005 PBS MultiStem
Day 3

V
Vessel
lDDensity
it

o. VWF positive vessels/mm^2

* p<.05
40
Day 14
Day 14 30 *
20
10
0
No

PBS MultiStem

See Penn et al Cytotherapy (2007) 53

Transarterial Catheter Delivery Approach

Mercator MedSystems, 510(k) approved MicroSyringe Infusion Catheter

• Site‐specific delivery into perivascular space and adventitia
– Retain greater number of cells at/near injury site (reduce wash‐away of cells into bloodstream)
– Relative ease‐of‐use
• Good cell viability, efficient ease of use

54
Biocompatibility of MultiStem with Catheter Delivery

MultiStem following catheter 
passage (micrograph)

Viable cell recovery,
Million cells
300
250
200
150
100
50
0
Cells infused Viable cells post‐infusion

55
Improvements in Functional Performance Observed in GLP Study

™ Long‐term safety study in AMI pigs
™ Delivery of MultiStem with the transarterial catheter 2 days after 
transient ischemia
i i h i

Left Ventricular Ejection Fraction Wall Motion Score (Echo)

*
*

56
Phase I Clinical Protocol Summary ‐ AMI

Phase I Study, open label, dose escalation
STEMI, LVEF between 30‐45%
Administration of MultiStem in coronary artery (via transarterial catheter) 
delivered on day 2‐5 after Acute MI
‐ Three dose groups (6 patients each) plus 10
Three dose groups (6 patients each) plus 10‐patient
patient registry cohort
registry cohort
Multiple sites, largely regional

Objectives
Primary endpoints: safety (drug or procedure related arrhythmias, acute 
toxicity, hospitalization, death, mechanical complication)
Secondary endpoints: functionality measures (e.g. LVEF)

Strategy
Provide safety foundation and information to enable design of meaningful 
Ph
Phase II exploratory study (e.g., dose levels, delivery timing)
II l d ( d l l d li i i )

57
Delivery of MultiStem in AMI patient

5 sec 30 sec 60 sec

Delivery, retention of cells

in area of ischemic damage
Vessel patency
Rapid, efficient procedure
Well tolerated

58
Update from Ongoing Phase I Study

Seven clinical sites screening, enrolling patients

Leading cardiovascular treatment centers, clinical researchers 
participating in trial 
– Cleveland Clinic, Henry Ford, University of Michigan, Columbia University, 
, y , y g , y,
Care Group
– Select regional cardiovascular treatment centers also participating

First patient cohort enrolled, treatment administered (lowest dose), and 
First patient cohort enrolled treatment administered (lowest dose) and
registry patients on study
8th site initiating and potential to add other clinical sites if necessary
Target enrollment completion by end of year and announcement of 
top‐line results shortly thereafter

59
Cardiovascular Area Strategy

Establish, and then leverage, initial proof‐of‐concept
– Standardized / scalable product manufacturing, pre‐clinical data, delivery, mechanism, 
basic safety in humans
– Establish initial human safety in Acute MI setting: allogeneic MultiStem, catheter‐
delivered in critical setting
– Leverage data from other (non‐cardiovascular) trials as necessary (e.g., IV‐delivery)

Establish efficacy proof‐of‐concept in specific indications through focused 
exploratory development
– Studies with discreet endpoints / readouts over short term → showing of desired 
biological activity and benefit
– Leverage BB
Leverage BB‐13554
13554 to step
to step‐out
out into other indications, where appropriate
into other indications, where appropriate

Share risks and investment – cardiovascular development can be difficult
– Some state grant funding
– Partnership (e.g., Angiotech)
60
MultiStem for Ischemic Stroke 
and Other CNS Indications

61
MultiStem®:  Ischemic Stroke

Substantial unmet need in the treatment of Ischemic Stroke

Approximately 780,000 strokes annually in U.S. in 2008, and ~85%+ ischemic strokes 
(Note: This number expected to increase over time as risk to “boomers” grows)
Substantial functional loss and rehabilitation and follow‐up care costs 
Limited treatment options, rtPA must be administered within 3 hrs of stroke
f

IV delivery of MultiStem 48 – 60 hours following Ischemic Stroke

Broad potential treatment window
Benefit trophic‐factor mediated: reduce inflammation, stimulate revascularization, 
override processes of cell / tissue decline & contribute to tissue regeneration

MultiStem demonstrated safe and effective in pre‐clinical models

IND authorized December 2008

62
Animal Models of Cerebral Ischemia

MCA Occlusion

MCA Ligation

63
Motor and Neurological Tests Overview

EBST (Elevated Body Swing Test)
– Gross Motor Test
– 50%= Normal Behavior
Bederson Test
– Battery of 4 Neurological Tests
– Each Test can be scored from 0 (Normal Behavior) to 3 (Absolute Neurological 
Deficit)
– All 4 scores are added then divided by 4 to give an average score
– Tests:
1) IIpsilateral
il t l Circling (Gross Test)
Ci li (G T t)
2) Hind Limb Replacement (Gross Test) 
3) Beam Walking (Fine Test)
4) Bilateral Forepaw Grasp (Fine Test)
Bilateral Forepaw Grasp (Fine Test)

64
 Pre‐Clinical Experimental Approach
Experimental approach
1. Immunosuppression (+/‐) with allo‐/ 
xenogeneic cells, intracranial delivery
2. Route of administration: viability of IV‐
delivery
3. Delivery window: 1‐7 days post‐stroke
4. Dose escalation
Key results
Key results
• Immunosuppression not required for safe improvement to 
neurological function
• Significant functional improvement (locomoter, neurological) 
statistically over control
• Comparable improvement in locomotor or neurological 
function observed among animals receiving cells at 1, 2 or 7 
days
• Dose response observed with IV‐infused cells, as measured by 
neurological improvement
• Engrafted cells display neuronal markers in neonatal model
• No abnormal tissues or abnormal pathology observed in 
animals kept on study for 1 year post cell transplantation
65
Allogeneic / Xenogeneic Cells Show Significant Improvement +/‐ CsA

3
ore 
ologic sco

2
ean neuro

1
****
Me

* *** *** * ****

0
Baseline Post‐ 14 28 42 56
stroke Days post‐transplant:

IC Rat MultiStem + Vehicle
IC Rat MPCs + Vehicle IC Rat MultiStem + CsA
IC Rat MPCs + CsA
IC Human MultiStem
IC Human MPCs 
IC Human MultiStem
MPCs + Vehicle
Vehicle
+ Vehicle
Vehicle IC Human MultiStem+
IC Human MPCs 
IC Human MultiStem
MPCs + CsA
CsA CsA
CsA
Control (Irradiated MPCs + CsA)
Control (Irradiated MultiStem + CsA)

66
IV Delivery Window: Equivalent Benefit across Window

Equivalent Functional Improvement Seen When 1 Million Xenogeneic
MultiStem Are Transplanted IV on Days 1, 2 or 7 Post Stroke

Cell Delivery on Day 1 Cell Delivery on Day 2 Cell Delivery on Day 7 Non‐Viable Cell Delivery 

on Day 7

67
Earlier IV‐delivery Results in Greater Cell Survival 

Day 1 cell delivery protects penumbra better than day 2 
and day 7 delivery

68
Single Dose of Human MultiStem Provides Robust, Durable Improvement 
in Rodent Model of Ischemic Stroke 

Bederson Composite Score of Neurological Function,

IV-delivery of MultiStem day 2 post-stroke
3

• Dose response –
Mean Neurologicall Score

2.5
Therapeutic benefit
2
proportional to dose
delivered
1.5 • Treatment timing –
Improvement
1 whether delivery at
day 1, 2 or 7
0.5

Day 42
Day 14

Day 28

Day 56
Stroke
Post-
Baseline

0.4
0 4 units 4 units
1 units 10 units
2 units 20 units
10 units non-viable cells (control)
69
Athersys in Leadership Position for Stroke Cell‐based Therapy

Stem cell Therapeutics as an Emerging Paradigm in Stroke
(STEPS) Conference

• “Bridging Basic and Clinical Science for Cellular and Neurogenic Factor Therapy in 
Treating Stroke,” Arlington VA, October 27‐ 28, 2007
‒ 50 participants/6 countries (Academia, Industry, FDA, NINDS)
‒ Athersys was an organizer / content leader
• Positioned to define the landscape for cell‐based therapy in stroke
‒ Major topics of discussion included: (1) mechanisms of actions of cellular 
therapy, (2) translation from animal models to human clinical situation, and (3) 
h (2) l i f i l d l h li i l i i d (3)
clinical trial design
‒ Publication, Bridging Basic and Clinical Science for Cellular and Neurogenic 
Factor Therapy in Treating Stroke Stroke 2009. 40:510
Factor Therapy in Treating Stroke. Stroke 2009 40:510‐515
515.
• Athersys positioned as an industry leader
‒ Increased FDA / thought leader exposure to Athersys approach and perspectives
‒ Guided position papers
Guided position papers
‒ New centers/investigators and advisors identified and brought on board 
70
Phase I Clinical Protocol Summary – Ischemic Stroke

Phase I study, double‐blind, placebo‐controlled dose escalation
– Ischemic stroke: DWI lesion, defined severity, arm weakness
Ischemic stroke: DWI lesion, defined severity, arm weakness
– Administration of single dose of MultiStem intravenously, 48 – 60 h post stroke
• Four dose tier groups including placebo (2:1)
• Modified continual reassessment methodology
– Multiple sites
Objectives
– Primary endpoints safety
y p y maximum tolerated dose based on composite of DLTs 
p
and AEs through first 14 days
– Secondary endpoints include functionality outcomes at 30, 90, 180 days, and 1 
year
Strategy
– Establish dose safety and tolerability of cell product (e.g., dose levels, delivery 
timing) to enable phase II proof‐of‐concept study

71
Additional Research Collaborations in CNS Area

Research strategy
– Work with independent research laboratories to establish preclinical proof 
p p p
of concept in selected diseases / conditions
– Develop better understanding of MultiStem’s modes of action in different 
settings
Research collaborations
– Neonatal hypoxic ischemia, Stroke, Medical College of Georgia
Neonatal hypoxic ischemia Stroke Medical College of Georgia
– Spinal cord & neurological injury, CWRU
– Amotrophic Lateral Sclerosis, U. of Wisconsin
Amotrophic Lateral Sclerosis, U. of Wisconsin
– Parkinson’s Disease, U. of Cincinnati
– Lysosomal disorders (Hurlers), U. of Minnesota
– Stroke, Traumatic Brain Injury, UT‐Houston
72
MultiStem®:  Potential in Other Neurological Injury Models

Profile for neurological injury and disease

MultiStem associated with substantial, durable functional improvement in multiple 
models (e.g., ischemic stroke, neonatal hypoxic ischemia, orphan disease indications)
Benefits seen using localized delivery or intravenous administration
In independent studies conducted in leading neuroscience labs, biological 
mechanisms of benefit observed that are not seen with other cell types
Multiple trophic factors expressed by MultiStem that appear to directly affect 
function and health of neurological tissue

Substantial unmet medical needs exist

Hypoxic injury – neonatal hypoxic ischemia is a leading cause of Cerebral Palsy
Progressive neurological conditions, especially those where chronic inflammation 
plays a contributory role
plays a contributory role

73
MAPC Provide Benefit in Hypoxic Ischemia Model 

Design
• Surgically induced hypoxic ischemic 
injury
• Administration of rat MAPC 7 days
Administration of rat MAPC 7 days 
after birth
– Syngeneic v. allogeneic (IC delivery) 
study
– Intracranial v intravenous
Intracranial v. intravenous 
(allogeneic) study
– Vehicle controls
• Behavioral tests (rotarod, EBST at 
d
days 7 and 14 post‐transplant)
7 d 14 l )
• Sacrifice at d14 post‐transplant, 
followed by tissue evaluation

See Yasuhara et al Cell Transplant (2006); Yasuhara et 

al J Cerebral Blood Flow Metabolism (2008)
74
MAPC Reduces Damage from Injury 

Intact Control

Syngeneic Allogeneic 75
Reduction in Cell Loss Relative to Vehicle

Cell Loss in Hippocampal CA3 Region, 
Measuring Injured Side Relative to Intact Side

100 *
d  to intacct side

*
80

60
%), injured

40

20
Ratio (%

0
Syngeneic Allogeneic
ll Vehicle
h l

76
IV‐delivered MAPC Reduce Cell Loss  

Intact Vehicle

MAPC, IV MAPC, IC

77
MultiStem for Transplantation 
MultiStem for Transplantation
Support in the Hematological 
Malignancies

78
MultiStem®:  Transplantation (GvHD)

Frequent, potentially life threatening consequence of HSC / BM transplants

Clear need for improved treatments beyond broad immunosuppression
Limited treatment options for complications (e.g., GvHD)
Other problems associated with conditioning regimen (e g GI function)
Other problems associated with conditioning regimen (e.g., GI function)

IV delivery of MultiStem in conjunction with HSC / BM transplant

Reduction of GvHD
d i f i
impact and promotion of tissue regeneration and engraftment
d i f i i d f
Potential for GvHD intervention

MultiStem demonstrated safe and effective in pre‐clinical models

MultiStem demonstrated safe and effective in pre‐clinical models

IND approved, clinical trial initiated

79
MultiStem Provides Survival Advantage in Rat Acute GvHD Model 

Design
Survival
• Rats sublethally irradiated and injected  100%

with bone marrow cells and T‐cells  90%

from different rat strain → creating 
g 80%

Graft vs. Host immune response 70%

• MultiStem administered I.V. at day 1, or  60%

at days 1 and 8  50%

40%
No treatment
30%
Treatment, day 1
Results 20%
Treatment days 1+8
10%
• MultiStem provides significant survival 
MultiStem provides significant survival 0%
benefit versus animals receiving no  0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
treatment Days
• Benefits observable for other GvHD Significant survival advantage in MultiStem treated animals
indicators (body weight activity
indicators (body weight, activity, 
Note: Published work from other labs did NOT demonstrate a durable
posture, fur texture, skin) survival advantage following administration of MSC’s in GVHD models

80
MultiStem Homing in Mouse GvHD Model

MultiStem Intravenous Infusion in Mouse GvHD Model
6 hours post‐i.v. 24 hours post‐i.v.

81
Gut Pathology in Rat Acute GVHD Model

Day 15 Pathology, Multi‐treatment Group

GVHD, MultiStem
GVHD MultiStem
Treated
GVHD, PBS Control 
Substantially less
Treated
gastro‐intestinal damage
in MultiStem treated
in MultiStem‐treated 
animals

82
Phase I Clinical Protocol Summary

Phase I study, open label, dose escalation
Patients (leukemia, myelodysplasia) undergoing PBSC / bone marrow transplantation
Administration of MultiStem intravenously
‐ Two treatment arms: Single dose co‐administered with transplant, multiple doses 
administered over first 30 days
administered over first 30 days
‐ Continual reassessment methodology

Objectives
Primary endpoints: safety: maximum tolerated dose based on composite of DLTs and 
AEs through 30 days
Secondary endpoints: incidence and severity of GVHD, survival, infection

Strategy
Provide safety foundation to allow for (a) prophylactic treatment and intervention for 
GVHD, and (b) single and multiple dose treatment approaches 

83
Update from Ongoing Phase I Study

Three clinical sites screening, enrolling patients
Leading bone marrow transplant clinical researchers at academic & 
clinical research institutions (University Hospitals, Oregon Health Sciences 
University, Ohio State University)
First dosing cohort (lowest dose) enrolled and treatment administered; 
CRM used to govern step‐ups in subsequent dosing level(s)
Enrollment slower than expected

Activities & options to accelerate progress
Improve enrollment dynamics and performance 
– Increase number of sites, potential protocol amendments to increase 
number of eligible patients

Considering possible implementation of GvHD treatment arm
– May support step out to treatment of other immunological conditions

84
Other Immunomodulation Opportunities 

Treating emergent or chronic autoimmune disease

Immunomodulatory activity of MultiStem for GVHD is mechanistically similar to 
biological conditions in many other indications 
Rapid clinical entry is possible (leveraging off of existing pre‐clinical and clinical data)
Manufacturing capability already in place

Multiple indications possible 
Wide range of autoimmune conditions with unmet medical need as potential 
therapeutic targets for MultiStem
– Other potential benefits to help address tissue damage
I.V. delivery
Potential to leverage IND BB‐13507 (Evaluation of MTD of Single and Repeated 
Administration of Allogeneic MultiStem in Patients with AL, CM and Myelodysplasia)

85
 Recent Publications 

Cardiovascular Van’t Hof W, Mal N, Huang Y, et al.  Direct delivery of syngeneic and allogeneic large‐scale expanded 

multipotent adult progenitor cells improves cardiac function after myocardial infarct.  Cytotherapy.  2007  
9(5): 477‐487.
Zeng L, Hu Q, Wang X, et al.  Bioenergetic and Functional Consequences of Bone Marrow‐Derived 
Multipotent Progenitor Cell Transplantation in Hearts with Postinfarction Left Ventricular Remodeling.   
Circulation.  2007 115: 1866‐1875.
Aranguren X, McCue J, Hendrickx B et al.  Multipotent adult progenitor cells sustain function of ischemic 
limbs in mice.  J. Clin. Investigation. 2008 118(2): 505‐514.
Wragg A, Mellad JA, et al.  VEGFR1/CXCR4‐positive progentior cells modulate inflammation and augment 
tissue perfusion by a SDF‐1‐dependent mechanism.  J Mol Med. 2008 86(11): 1221‐32.
Hematology / Kovacsovics M, Streeter P, Mauch K et al.  Clinical scale expanded adult pluripotent stem cells prevent
Immunology graft‐versus‐host disease.  Cellular Immunology.  2008.
YSerafini M, Dylla S, Weismann IL, et al. Hematopoetic reconstitution by multipotent adult progenitor 
cells: precursors to long‐term hematopoetic stem cells. J. Experimental Medicine. 2007 
CNS / Stroke
/ Mays R, Van't
y , Hof W, Deans R, et al. Development of adult pluripotent
, , p p p stem cell therapies for ischemic 
p
injury and disease. Expert Opin. Biol. Ther. 2007; 7(2):173‐184. 
Yasuhara T, Hara K, Maki M, et al. Intravenous grafts recapitulate the neurorestoration afforded by 
intracerebrally delivered multipotent adult progenitor cells in neonatal hypoxic‐ischemic rats. J Cerebral 
Blood Flow & Metab. 2008, 1‐7.
Yasuhara T, Matsukawa N, Yu G, et al. Behavioral and histological characterization of intrahippocampal
grafts of human bone marrow‐derived multipotent progenitor cells in neonatal rats with hypoxic‐
ischemic injury. Cell Transplant. 2006 15(3):231‐238. 
86
Thinking Ahead … Platform for Multiple Product Opportunities

Potential for addressing multiple indications
– Multiple possible mechanisms of action from MultiStem
– Broadly applicable mechanisms of action, e.g., immunomodulation
Product differentiation and segmentation possible
Product differentiation and segmentation possible
– Administration, dosing, formulation
– New product development; certain variants may perform better 
than others for certain indications
Broad technology platform and intellectual property creates 
potential for next generation products
potential for next generation products
– Enhanced biological activity targeted to certain conditions, 
indications
– Further advantaged production and administration characteristics
87
Pharmaceutical Programs

88
 Obesity

5HT2c Agonist Program
Obesity Market Opportunity

Clinical Landscape

Growing, global health epidemic contributing to heart disease, diabetes, cancer and stroke
Estimated 30% of Americans are clinically obese (BMI > 30); an additional ~30% are 
overweight (BMI > 25)
Economic cost in U S alone is estimated at $117 billion annually
Economic cost in U.S. alone is estimated at $117 billion annually
True blockbuster potential for safe and effective therapies

Therapeutic Landscape
Therapeutic Landscape

Increasing recognition of obesity as serious medical condition
No highly effective & safe drug therapies currently on market – few in clinical trials
Several targets are well known but have not been effectively exploited to date
Large potential market, patient variability (efficacy and tolerability) creates room for 
multiple players and MOA’s
Combination therapy is general focus to achieve maximum efficacy
Combination therapy is general focus to achieve maximum efficacy
Obesity Program – Overview of 5HT2c Agonists

5HT2c (serotonin) receptor agonists well known to suppress appetite & cause weight loss 

Mechanism extensively validated in humans (e.g. fenfluramine, 
dexfenfluramine recognized as highly effective weight loss agents, recent 
Lorcaserin data)…but…
N
Non‐selective agents (fenfluramine, dexfenfluramine) also activate the 5HT2b 
l i (f fl i d f fl i ) l i h 5HT2b
receptor in the heart and cause cardiovascular toxicity (valvular hypertrophy = 
valvular regurgitation/heart murmur)
‐ Mechanism
Mechanism not identified & understood until 2000 (agonist activity at 5HT2b)
not identified & understood until 2000 (agonist activity at 5HT2b)
‐ Effect also observed clinically with other agents (pergolide, cabergoline) that 
activate 5HT2b receptor, but not 5HT2c, 5HT2a (NEJM Jan, ‘07 – see Roth et al)
‐ Other agents (lisuride) that are have agonist activity at 5HT2c and 5HT2a do not 
g ( ) g y
cause valvulopathy
Selective 5HT2c agonists (i.e. that do not stimulate 5HT2b) believed to be safe –
recent Lorcaserin clinical experience provides important validation
Selectivity relative to 5HT2a important to limiting CNS related side effects
ATHX ‐ 5HT2c Agonist Program Summary

Program thesis and focus
Thesis: Compounds that display superior selectivity for target (5HT2c) relative to other 
serotonin receptors (5HT2b, 5HT2a) will be better tolerated, safer, and more effective
Focus: Development of a safer, better tolerated, and more effective 5HT2c agonist

High quality portfolio of therapeutic compounds established
High binding affinity, full agonist activity at target 5HT2c receptor
O t t di
Outstanding selectivity for other serotonergic
l ti it f th t i receptors, in particular 5HT2b and 2a
t i ti l 5HT2b d 2
Extensively screened to ensure selectivity against broad range of other receptor systems
ATHX‐105 was our lead (3 Phase I trials completed in 2008 demonstrating good safety 
and tolerability, outstanding regional absorption, high relative exposure levels)
ATHX‐105 placed on clinical hold due to compound specific non‐clinical tox finding (prior 
to commencement of planned Phase II study) – subsequently suspended 
Program currently focused on identification of clinical candidate with best‐in‐class 
potential building upon substantial experience base
ATHX‐105 – Demonstrated Reduction in Food Intake & Weight

Administration of ATHX‐105 to Obese Zucker Rats (QD)
Effect on Food Intake
Vehicle
Administration of 5HT2c 
300 antagonist reverses effect
Fen (10 mpk)
ATHX‐105 (2.5 mpk)
250
Cumulative Food Intake (g)

ATHX‐105 (5.0 mpk)
ATHX‐105 (10 mpk) *
200 * 6.0
ATHX 105 (20 mpk)
ATHX‐105 (20 mpk)
*

nsumption (g) 
ATHX‐105 (40 mpk)
150 * 5.0
**
100
*
* * 4.0
* *
50 * * 3.0
* ** *
}*

Food Con
0 2.0
0 2 4 6 8 10 12 14
Time After Initial Dose (days) 1.0
*

Effect on Body Weight 0.0

4 Hours After Dosing
0
Chaange in Body Weight (g)

0
‐10
* * Vehicle
‐20 * *
* Fenfluramine (10 mpk)
‐30 *
* * ATHX‐105 (10 mpk)
‐40
*
* *
* SB242084 (3mpk)
‐50 *
* SB242084 (3mpk) + 
‐60 * *
* ATHX‐105 (10 mpk)
‐70

‐80
2 4 6 8 10
Time After Initial Dose (days)
12 14 93
* p<0.05 compared to vehicle control (ANOVA, Dunnett's)
ATHX‐105 – Phase I Study Objective and Design

Key study objectives and parameters
Objective: Evaluate the safety, tolerability and pharmacokinetics of ATHX 105 after oral doses
Objective: Evaluate the safety, tolerability and pharmacokinetics of ATHX‐105 after oral doses
Study conducted in the U.K. with Charles River Labs
Basic design:
‐ Double blind, placebo controlled
‐ Two part study: SAD and MAD administration
‐ Males and females with target BMI 25 – 35 (overweight to modestly obese) within age range                
18 to 65 years
‐ Typically 8 subjects per cohort (6 subjects on ATHX‐105 and 2 on placebo)
‐ 107 total subjects evaluated in the study

Single ascending dose study
Evaluated following doses: 2 mg, 6 mg, 20 mg, 50 mg, 100 mg, 150 mg
valuated following doses: mg, 6 mg, 0 mg, 50 mg, 00 mg, 50 mg
Fed‐fasted arm conducted at 50 mg dose level to evaluate effect of food on drug absorption

Multiple ascending dose study
Administration of ATHX‐105 or placebo for one week
Evaluated following doses: 25 mg, 50 mg and 75 mg QD, and 50 mg BID
94
ATHX‐105 – Summary of Phase I Results

Good safety and tolerability profile observed
Maximum tolerated dose = 100 mg
Generally well tolerated at dose levels below 100 mg
No severe adverse events occurred at any dose
No serious adverse events or discontinuations due to adverse events
Adverse events generally mild and transient (e.g. headaches, nausea, dizziness)
No clinically significant effects on heart rate, blood pressure or EKG parameters at 
any dose
No clinically significant effects on any hematology or clinical chemistry parameter at 
any dose
Other highlights

Well absorbed – good drug exposure observed
Food had no effect on total drug exposure
Maximum drug concentration observed to be dose proportionate
b b
Subsequent Phase I study demonstrated outstanding regional absorption
95
 ATHX‐105:  Superior Selectivity vs. Lorcaserin, Fenfluramine

Selectivity Ratios for 5HT2c vs. 5HT2b, 5HT2a
based on binding affinity (Ki) at each receptor

5HT2c / 5HT2b 5HT2c / 5HT2a

250 235 50
40
200 40

150 30

100 20

50 10 8
12 3
0.5
0 0
Fenfluramine*  Lorcaserin** ATHX‐105 Fenfluramine*  Lorcaserin** ATHX‐105
(Norfenfluramine) (Norfenfluramine)

ATHX‐105 displays substantially better  ATHX‐105 displays substantially better selectivity 
selectivity for target relative to 5HT2b for target relative to 5HT2a
((responsible for cardio AE’s)
p ) ((responsible for CNS AE’s –
p e.g. nausea, dizziness)
g , )
* From Fitzgerald et al (2000) Mol. Pharmacol. 57:75‐81
** From Thomsen et al (2008) J. Pharmacol. & Exp. Therapeutics, Online, 02/05/08.  Based on functional assays, Lorcaserin’s reported 2c/2b and 2c/2a 
selectivity is 105x and 19x, respectively (which is also inferior to ATHX‐105’s selectivity based on functional assays).
ATHX‐105 Well‐tolerated at High Plasma Drug 
Concentrations
Plasma Drug Concentrations at MTD
3

25
2.5
* Arena Poster Presentation
2
at 2009 NAASO meeting
1.5
** Athersys Phase I study
1

0.5

0
Lorcaserin* ATHX‐105**

™ Lorcaserin has similar potency at 5HT2c, but ATHX‐105 is approximately 5‐fold more 
selective at 5HT2a
™ At the MTD for each drug, ATHX
At the MTD for each drug, ATHX‐105
105 plasma drug levels were approximately 6
plasma drug levels were approximately 6‐fold
fold higher 
higher
than lorcaserin
™ Higher selectivity at 5HT2a appears to lead to better tolerability at high drug 
concentrations, allowing higher doses to be administered to achieve superior efficacy

97
 Relative Drug Levels of ATHX‐105

300

250
THX-105 (ng//ml)

200

Immediate Release
150 Distal Small Intestine Release
Colon Release
AT

100

50

0
0 5 10 15 20 25 30

Time Post-Release (hrs.)

98
Competitive Landscape ‐ Recent Results with Lorcaserin

Top line BLOOM date announced March 30, 2009
Weight loss statistically significant, but perhaps less than what was hoped for
g y g , p p p
Importantly ‐ No evidence of valvulopathy issues seen previously with fenfluramine
and dexfenfluramine
No apparent evidence of CNS effects that have plagued other drugs / MOA
No apparent evidence of CNS effects that have plagued other drugs / MOA’ss such as 
such as
Rimonabant and other CB‐1 antagonists (i.e. anxiety, depression, suicidal ideation)
Appears to be clear room for improvement in multiple dimensions, including efficacy, 
tolerability, convenience
tolerability, convenience
Fen‐Phen still appears to be most effective combination therapy developed to date 
Historical safety liabilities for this class now well understood, clinically validated
™ Selectivity for 5HT2c receptor vs. 5HT2b receptor essential for cardiovascular safety
Selectivity for 5HT2c receptor vs 5HT2b receptor essential for cardiovascular safety
™ Selectivity for 5HT2c receptor vs. 5HT2a receptor important for tolerability, maximizing 
exposure while minimizing CNS side effects
™ Weight loss effect is dose proportional, so exposure level and therapeutic index/window is 
g p p , p p /
key to achieving maximum therapeutic effect while achieving tolerability, convenience

99
 Combination Weight Loss Agents

Trial
Agent % Weight Loss Duration
(placebo adjusted) (weeks)
Fenfluramine/Phentermine (Fen‐Phen) 11‐12% 34 wk
(1992 Weintraub et al.)

Topirimate/Phentermine (Qnexa) 7.5% 28 wk

(Phase 3 Equate trial)

Bupropion/Naltrexone (Contrave) 4.2% 56 wk

(Phase 3 NB‐302 trial)

Bupropion/Zonisamide (Empatic) 3 4 7 5%
3.4‐7.5% 24 wk
24 wk
(Phase 2 trial)

100
 5HT2c Agonist Program Summary

5HT2c Receptor now extensively validated as an obesity target
In our view the “ideal compound” or combination for obesity is yet to be developed
Best‐in‐class selectivity achieved by ATHX for compounds targeting 5HT2c receptor vs. 
5HT2b, 2a 
™ Substantially higher drug exposure levels for ATHX‐105 relative to Lorcaserin
™ Extensive knowledge developed around MOA and SAR 
Building off our internal experience, historical validation and recent results, we are 
Building off our internal experience historical validation and recent results we are
focused on:
™ Continued development of potent, selective 5HT2c agonists with improved characteristics 
and best‐in‐class
and best in class profile (efficacy, safety, tolerability, convenience)
profile (efficacy safety tolerability convenience)
™ Improved potency, selectivity, half‐life & retention of other desirable characteristics
™ Elimination of compound specific liability seen with ATHX‐105 

Evaluating potential partnering opportunities
101
Cognition & Wakefulness – H3 Antagonist Program

Histamine H3 Receptor is a Highly Validate Target
At low doses antagonists or inverse agonists increase histamine levels, promote 
At l d t it i it i hi t i l l t
wakefulness, attention and cognitive performance
Multiple indications possible (i.e. recognized in the literature) including narcolepsy/EDS, 
chronic fatigue associated with Parkinson’s or other conditions, ADD/ADHD, 
Schizophrenia, Alzheimer’s and other dementias
At higher doses, potential to reduce appetite and food intake?

High quality portfolio of therapeutic compounds established at ATHX

High quality portfolio of therapeutic compounds established at ATHX
Multiple highly potent, selective chemical series developed
Initial safety and efficacy data demonstrated in animal models
M lti l
Multiple patent applications filed
t t li ti fil d
Initial lead compound highly effective but produced phospholipidosis in rats (also 
observed with other compounds described in literature)
Extensive and rigorous discovery effort has identified multiple lead compounds that do
Extensive and rigorous discovery effort has identified multiple lead compounds that do 
not produce phospholipidosis
Multiple promising potential clinical candidates under evaluation 102
Histamine H3 Receptor Biology

Esbenshade et al., Molecular Interventions (2006) 103

Early Lead (ATH‐90879) Profile
• Highly efficacious in animal models
– Minimum efficacy dose in sleep < 5 mg/kg
– Minimum efficacy dose in obesity < 30 mg/kg which produce 10% weight loss in 14 days
• Well tolerated at high doses
Well tolerated at high doses
– MTD > 200 mg/kg in rats
– No CNS clinical observations at any dose tested
• Excellent Pharmacokinetic Profile
– Long Half‐life (6 hrs. in rats, 4.5 hrs. in monkeys, >10 hours predicted for humans)
– Orally bioavailable (50% in rats, 35% in dogs)
– Excellent brain penetration (22‐fold higher brain exposure vs. plasma)
• Highly Selective
Highly Selective
– >1000‐fold vs. other histamine receptors
– No meaningful interaction at Herg channel, or any of the 70 other receptors, channels, and enzymes 
tested to date
– No cytochrome P‐450 interactions at relevant drug concentrations
No cytochrome P‐450 interactions at relevant drug concentrations
• No genotoxicity observed in Ames and Chromosome Aberration assays
• No organ toxicity or clinically meaningful blood changes observed in 14 day rat safety study; 
however, dose dependent phospholipidosis observed in the lungs
• Extensive effort undertaken  to identify a clinical candidate that does not produce 
phospholipidosis
104
ATH‐90879 – Positive Effect on Wakefulness

Design % Total Response Time - Wakefulness

90

™ Adult male rats fitted with electrodes  80 * *
*
to record brain wave sleep patterns 70

™ Drugs administered i.p. immediately  60
prior to start of the animal’s sleep cycle 50

40
Results 30

20
™ Wakefulness significantly increased by     
ATH‐90879 and ≥ the effect of  10

modafinil (Provigil)
( g ) 0
Vehicle 1.25 5 20 Caffeine Modafinil
™ No observable hyperactivity caused by      mpk mpk mpk 300 150
mpk mpk
ATH‐90879 ATH-90879

105
H3 Antagonist Effect on Body Weight

Zucker Obese Rat Model

106
Safety Profile Considerations for H3 Antagonists

Potency and selectivity are key parameters
Selectivity against other Histaminergic receptors
Selectivity against other Histaminergic
Selectivity against broad range of other receptor systems

Phospholipidosis ‐ An underappreciated potential risk factor 
Phospholipidosis = phospholipid accumulation that may occur in certain organs that 
can result in long term toxicity and tissue damage (e.g. interstitial lung disease)
Some H3 antagonists / inverse agonists described in the literature thought to cause  
Some H3 antagonists / inverse agonists described in the literature thought to cause
phospholipidosis
Easy to miss in animal models due to typically short half‐life of compounds in rodent 
models, limited exposures

ATHX has established rigorous screening capabilities for phospholipidosis
Sensitive in vitro assays designed to detect phospholipidosis
y g p p p
Careful in vivo assessment
107
 Example H3 Leads
Histamine H3 Histamine H3
Compound (hu, Ki) (ms, Ki)
Phosphlipidosis

ATH‐91299 0 2 nM
0.2 nM 7 nM
7 nM ND
N.D.
ATH‐91301 0.15 nM 4 nM N.D.
ATH‐91305
ATH 91305 0 25 nM
0.25 nM 2 5 nM
2.5 nM ND
N.D.
ATH‐91308 0.24 nM 2.7 nM N.D.
ATH‐91310 0.15 nM 4.1 nM N.D.
ATH‐91311 0.17 nM 6.0 nM N.D.
ATH‐91313 0.2 nM 3.4 nM N.D.
ATH‐91335 0.1 nM 3.8 nM N.D.

* N.D., Not Detected 108

In Vitro Evaluation of Phospholipidosis

Acetominophen 10uM
(Negative Control)
Competitor’s 
ATH‐91313
Compound

Amiodarone 10uM
(Positive Control)

Clinical stage H3 antagonists developed by other 
companies test positive for phospholipidosis in vitro
companies test positive for phospholipidosis in vitro

109
ATH‐91263 is a Potent Inverse Agonist

120

100

80

GTPγS binding (% maxximal)
60

40

20

‐20
vehicle 100 nM R‐a‐ 10 uM  10 nM ATH‐
MH thioperamide 91263
‐40
IC50 (nM)
‐60
ATH‐
0.71
91263

110
 Pharmacokinetics of ATH‐91263 in Rats

Drug Concentrations Following Dose Proportionality

Oral Administration
2000
6000
1800
g/ml)

Plasma
1600 5000
Brain
entration (ng

max (ng/ml)
1400
4000
1200
1000 3000

800

Cm
2000
Conce

600
400 1000

200
0
0 0 50 100 150 200 250 300 350
0 5 10 15 20 25 30
Dose (mg/kg)
Time (hrs.)

ATH-91263
ATH 91263 is orally bioavailable,
bioavailable brain penetrant
penetrant, and
dose proportionate
111
ATH‐91263 Increases Cumulative Wakefulness

112
ATH‐91263 Increases Wakefulness

Oral administration at beginning of dark cycle, n=8 rats, cross-over design

ATH‐91263 Does Not Produce Hyperactivity
Properties of Histamine H3 Leads
Highly Potent and Selective
– 300 pM at human H3, 3 nM at mouse H3
– >1,000‐fold vs. other histamine receptors
– No meaningful interaction at Herg channel ( > 10 uM)
– No cytochrome P‐450 interactions (IC50 > 10 uM for each major CYP)
– No interactions in broad selectivity screen (IC50 > 10 uM for all proteins in Pan Labs panel)
E ll t Ph
Excellent Pharmacokinetic Profile
ki ti P fil
– Long Half‐life (> 6 hrs. in rats)
– Orally bioavailable 
– > 60% free drug concentration in rats and humans
– Good brain penetration (IC
db ( 50)
– No evidence of Phospholipidosis using in vitro assay
Well tolerated at high doses
– Acute MTD > 300 mg/kg in rats
Efficacious in animal models
– Efficacy at doses >5 mg/kg in rat sleep model
– Weight loss observed at doses > 100 mg/kg
Excellent safety and tolerability profile in 14 day rat safety and tolerability 
Excellent safety and tolerability profile in 14 day rat safety and tolerability
study
115
H3 Antagonist/Inverse Agonist Program Summary

Extensive preclinical and emerging clinical validation around target
Portfolio of extremely potent, highly selective compounds established at ATHX
yp , g y p
™ Rigorous assessment for phospholipidosis and other characteristics part of our evaluation
™ Extensive knowledge base developed internally regarding SAR, other parameters 
Unexpected issues seen with other programs (e g liver tox phospholipidosis)
Unexpected issues seen with other programs (e.g. liver tox, phospholipidosis)
™ Phospholipidosis particularly significant with long half‐life compounds, clinical indications 
that require extended use
Building off our internal experience, historical validation and recent results from other 
B ildi ff i t l i hi t i l lid ti d t lt f th
programs, we are focused on:
™ Continued development of potent, selective H3 antagonists and inverse agonists with        
best‐in‐class profile
™ Avoidance of liabilities seen with other compounds
™ Opportunity to develop first long half‐life compound free of phospholipidosis and suitable for 
once daily administration, application in clinical indications with extended use

Evaluating potential partnering opportunities
 RAGE‐VT Program Summary

Production of cell lines expressing well validated targets for dug development
RAGE (Random Activation of Gene Expression) technology developed at ATHX 
Extensively validated, patented w/ multiple scientific publications illustrating use in:
™ Gene discovery
™ Phenotypic screening and pathway analysis 
™ Drug development (creation of cell lines for HTS screening and optimization)
™ Protein expression (alternative approach to traditional cDNA)

Multiple partnerships established by ATHX for use in drug screening programs

Multiple partnerships established by ATHX for use in drug screening programs
Over past few years, partnerships/licensing agreements established by ATHX with 
multiple major pharma companies and other partners
™ Partners & licensees include BMS (renewed twice), Pfizer, J&JRI, Wyeth and others
Partners & licensees include BMS (renewed twice) Pfizer J&JRI Wyeth and others
™ Successful delivery and acceptance of ~24 targets (covering a range of therapeutic areas)
™ Most advanced partnered programs now in mid‐stage clinical development (Phase II)
™ Has also provided access to multiple targets for internal programs
p p g p g
To date ~$14 Million in revenue generated…with meaningful future revenue 
potential as programs continue to mature & advance
 Biopharma Summary

Multiple proprietary technologies, capabilities developed & validated at Athersys

Successfully applied for several years across multiple therapeutic areas for benefit of 
Successfully applied for several years across multiple therapeutic areas for benefit of
major pharma and emerging biopharma partners
Benefits include revenue generation & establishment of relationships with partners

Focused, cost effective approach implemented for internal programs
CNS focused
Building off of knowledge base developed by others, focused on addressing 
fundamental gaps to achieve best‐in‐class
Potential for synergy with MultiStem programs
Potential for synergy with MultiStem
Leveraging core internal capabilities and outsourced capabilities through CRO’s, CSO’s
Actively exploring partnerships

118
Financials

119
Summary Financial Data

Year ended
$ Thousands
December 31, 2008

Revenues $3,105

Operating expenses (22,197)

Interest income and other, net 1,100

Net loss ( ,
(17,992)
)

Net Cash Use in Operating Activities (15,711)

Cash and Investments 31,613

Debt 0

120
Financial Considerations

Favorable cash position
p
– Almost $32 million in cash and investments at end of 2008
– Operational focus and manageable burn rate constructed to allow 
approximately three years of operations with current cash (→ Q1, 
i l h f i ih h (→ Q1
2011)
– Exploring potential for collaborations across programs, and non‐
dil i (
dilutive (e.g., grant) opportunities
) ii

Substantial upside potential
– Portfolio
Portfolio of potential best‐in‐class opportunities, with nearer term 
of potential best in class opportunities with nearer term
impact possible with cell therapeutic programs
– Lower relative valuation, compared to other public stem cell / product 
platform companies
platform companies

121
 Athersys, Inc.
R&D / Investor / Analyst Day 
Intercontinental Hotel
Intercontinental Hotel

April 8th, 2009