Addex Pharmaceuticals

Investor PresentationSeptember 2010

Disclaimer

These materials do not constitute or form part, or all, of any offer or invitation to sell or issue, neither in the United States of America nor elsewhere, or any solicitation of any offer to purchase or subscribe for, any securities, nor shall part, or all, of these materials or their distribution form the basis of, or be relied on in connection with, any contract or investment decision in relation to any securities.

These materials contain forward-looking statements based on the currently held beliefs and assumptions of the management of Addex Pharmaceuticals Ltd, which are expressed in good faith and, in their opinion, reasonable. Forward-looking statements involve known and unknown risks, uncertainties and other factors, which may cause the actual results, financial condition, performance, or achievements of Addex Pharmaceuticals Ltd, or industry results, to differ materially from the results, financial condition, performance or achievements expressed or implied by such forward-looking statements. Given these risks, uncertainties and other factors, recipients of this document are cautioned not to place undue reliance on these forward-looking statements. Addex Pharmaceuticals Ltd disclaims any obligation to update these forward-looking statements to reflect future events or developments.

These materials are strictly confidential and must not be disclosed or distributed to third parties.

3

•Goal: allosteric modulators for human health

•Focus: CNS, metabolic disorders, inflammation

•Proprietary allosteric modulator discovery platform

•15 discovery/development programs

• Pharma validation– Partners: Johnson & Johnson and Merck & Co., Inc.

– Investors: SR-One (GSK) and Roche Venture Fund

•138 staff / founded 2002 in Geneva, Switzerland

The Company

4

Financials• Cash for operations through end of 2011

CHF56.7 (US$54/€42) million in cash as of June 30

• Market cap (6 Sep): CHF52m (€40m / US$51m)

• SIX Swiss Exchange: ADXN (ISIN:CH0029850754)

• 5,871,242 shares outstanding as of June 30, 2010

• Five analysts covering:

Piper Jaffray Sam Fazeli & Michael AitkenheadJefferies Peter Welford & Philippa GardnerHelvea Olav Zilian Bank Vontobel Andrew C. Weiss  & Silvia SchanzBank am Bellevue Bob Pooler      

Partner Phase IIPreclinical Phase I MilestoneLead

OptimizationHit-to-Lead

Assay Development &

Screening

Molecule / Mechanism

PIPELINE

Merck & Co., Inc.

Ortho-McNeil-Janssen

ADX68692FSHR NAM

ADX63365mGluR5 PAM

ADX71943GABA-B PAM

ADX71149mGluR2 PAM

Start Ph IIa1Q11

Start Ph II 4Q10

Start Ph IIa 1Q11

Ortho-McNeil-JanssenStart Ph IIa

1Q11

ADX48621mGluR5 NAM

NAM = negative allosteric modulator (an inhibitor) ‡ and undisclosed additional indications PAM = positive allosteric modulator (an activator) *

Ortho-McNeil-Janssen Pharmaceuticals, Inc., a Johnson & Johnson subsidiary

Parkinson’s Disease Levodopa Induced Dyskinesia (PD-LID)

Dystonia

Schizophrenia

Anxiety

Osteoarthritic Pain

Schizophrenia ‡

Endometriosis / Benign Prostatic Hyperplasia

funded & developed by OMJPI*

funded & developed by OMJPI*

funded & developed by Merck

Start Ph I2011

Partner Phase IIPreclinical Phase I MilestoneLead

OptimizationHit-to-Lead

Assay Development &

Screening

Molecule / Mechanism

DISCOVERY PROGRAMS

NAM = negative allosteric modulator (an inhibitor) ‡ and undisclosed additional indications PAM = positive allosteric modulator (an activator)

Inflammation

CNS

Metabolic Disorders

Merck & Co., Inc.

Alzheimer’s / Depression

Parkinson’s Disease ‡

DepressionPost Traumatic Stress Disorder

Sleep Disorders

Type II Diabetes

Type II Diabetes

Rheumatoid Arthritis, Psoriasis,

Alzheimer’s, Multiple Sclerosis

Psoriasis, Osteoarthritis

Gout, Type II Diabetes

funded by MerckmGluR4 PAM

GIPR PAM

TNFR1 NAM (CD120a)

GLP1 PAM

Orexin 2R NAM

mGluR7 NAM

mGluR2 NAM

A2A PAM

IL1R1 NAM (CD121a)

7

Partnered Programs

8

Agreement• With Ortho-McNeil-Janssen Pharmaceuticals, Inc. (OMJPI)

– Utilize Addex platform to identify & develop orally available mGluR2 PAM for schizophrenia, anxiety and undisclosed indications• OMJPI funds collaboration on discovery & lead optimization• OMJPI funds & performs preclinical and clinical development

– Addex sits on oversight committees

ADX71149

Terms

• €3 million upfront

• Research funding to Addex during discovery collaboration (2005-2007)

• €112 in potential milestones upon completion of clinical and regulatory milestones

• Low double-digit royalties

9

Progress to Date

• Collaboration generated clinical candidates

– €4.2 million in R&D funding between 2005-2007

• Comprehensive Ph I program started in June 2009

– €1 million milestone paid upon initiation of Phase I

– More than 5 Phase I trials in healthy volunteers• SAD, MAD

• Food & gender

• Ketamine challenge (schizophrenia model)

• Anxiety challenge

• Phase II program scheduled to start in 1Q11

– schizophrenia

– anxiety

– potentially other indicationsNote

mGluR2 activation is clinically validated in anxiety & schizophrenia

ADX71149

* Nature Medicine 2007: http://bit.ly/bbnsyQ

10

ADX63365

Agreement

• Merck & Co., Inc. licensed Addex mGluR5 PAMs in 2008

• mGluR5 PAM expected to be highly differentiated– Merck already had demonstrated mGluR5 PAM have efficacy in

animal models of schizophrenia

– Preclinical data show efficacy for cognitive deficit (& psychosis)

– Despite efficacy of marketed drugs many schizophrenia patients remain unable to learn new skills to support themselves

– FDA has recognized cognitive deficit as an unmet medical need in schizophrenia

• Merck is responsible for development of ADX63365 & backups

Terms

• $22 million upfront

• $680 million in milestones

• Undisclosed royalties

11

Progress

• Deal signed (Dec 07)

• 1st Preclinical milestone (Feb 08)– $250,000

• 2nd preclinical milestone (Jul 09)– $500,000

– Orally available mGluR4 PAM showed efficacy in model of PD

• Collaboration extended (Dec 09)– Merck commits $1.8 million in research funding

– Going forward all costs transferred to Merck

Agreement

• Discover and develop metabotropic glutamate receptor 4 (mGluR4) positive allosteric modulators (PAM) with Merck & Co., Inc.

• The deal includes mGluR4 PAM leads already discovered by Addex

• Merck is responsible for preclinical and clinical development

• Addex will sit on oversight committees Terms

• $3 million upfront

• $167.5 million in potential milestones

• Annual tech access fee $250,000

• Research funding 2009/2010

• Undisclosed royalties

• Option to co-promote in EU

mGluR4 PAM

12

Platform Revenues

Partner Product Indication(s)Status

at signingStatus April

30, 2010Upfront

Cash

Milestones or R&D funding

received

Total Potential

Milestones

Ortho-McNeil-Janssen

Pharmaceuticals, Inc.mGluR2 PAM

ADX71149 Anxiety &

schizophreniaHit-to-Lead

(Dec 2004)Phase I €3 million €5.2 million €112 million

Merck & Co., Inc. mGluR4 PAMParkinson’s

diseaseHit-to-Lead

(Dec 2007)

preclinical efficacy in PD model

$3 million $2.5 million $167.5 million

Merck & Co., Inc.mGluR5 PAM

ADX63365 schizophrenia

Clinical Candidate

(Jan 2008)

not disclosed

$22 million n.a. $680 million

Proprietary Platform Revenues

• Addex has received partnering revenue every year since 2004

• Cash inflows generated to date: CHF43 million

• All three partnerships are fully funded by our partners

• Potential for up to about $1 billion in milestones plus royalties

* and undisclosed indications

13

Wholly Owned Programs

14

ADX48621 Overview• Phase I completed

– Three studies: SAD, MAD, gender & food effects– 110 patients treated to date, including older volunteers– Safety & tolerability justify further clinical study– Ph II to start in 4Q10 for PD-LID and 1Q11 for dystonia

• Differentiated– Only product shown to reduce dystonia in MPTP model– Chemical series unrelated to other mGluR5 NAM

Unique metabolic profile

– NCE patents valid through 2025 in most territories – Unrelated series of backup molecules in clinical candidate

selection

15

•ADX48621 effects in HIC model suggestIt should be tested further as a potential drug for PD It has potential to be a dopamine sparing agent

ADX48621 efficacy in HIC model• Haloperidol induced catalepsy (HIC) is a preclinical model of PD

ADX48621 dose-dependently reversed HIC in 3 independent experimentsMTEP mGluR5 antagonist is well documented to work in the HIC model

0

10

20

30

40

50

60

70

80

90

Vehicle ADX48621, 1mg/kg

ADX48621, 3mg/kg

ADX48621, 10mg/kg

ADX48621, 30mg/kg

MTEP, 30mg/kg

Late

ncy

(sec)

1st experiment2nd experiment3rd experiment

***

*********

****

***

**

***

+ 1 mg/kg haloperidol

**p<0.01, ***p<0.001 versus vehicle group

16

MPTP model of PD-LID

What is the MPTP Model?

• MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine)

– Neurotoxin destroys dopaminergic neurons

– Symptoms similar to PD, treatable with levodopa

• MPTP monkeys develop levodopa induced dyskinesia (LID), including both chorea (trembling) and dystonia (cramping) seen in humans

• MPTP PD-LID model is predictive of efficacy in humans

– KOLs consider it among the most translational models in the CNS space

– AFQ056 (mGluR5/Novartis) showed efficacy in MPTP PD-LID model

– AFQ056 showed efficacy in Phase IIa PD-LID trial

17

• ADX48621 or vehicle administered 30 min prior to levodopa

• Behavioral assessment began upon levodopa administration – trained observers performed video review

– dyskinesia & PD disability scoring (10 min every 30 min for 2hrs)

lower scores (left axis) indicate fewer symptoms/disabilitydyskinesia symptoms are side effects from levodopadisability is a measure of Parkinson’s disease severity

ADX48621 MPTP Study Design

18

ADX48621 reduced dyskinesia without compromising levodopa efficacy

0

5

10

15

L-DOPA (100%)

vehicle

ADX48621 (mg/kg)

3 10 30

*

dysk

ines

ia (

0-2

hr)

ADX48621 dose-dependently reduced dyskinesia

(chorea + dystonia)

Dyskinesia

ADX48621 efficacy in MPTP model

ADX48621 does not reduce levodopa efficacy

Disability

+ levodopa

19

ADX48621 is the first drug-candidate known to have efficacy on dystonia

0

5

10

15

L-DOPA (100%)

vehicle

ADX48621 (mg/kg)

3 10 30

*

dys

ton

ia (

0-2

hr)

Dystonia

Dystonia = sustained muscle contractions (e.g. cramps)

0

5

10

15

L-DOPA (100%)

vehicle

ADX48621 (mg/kg)

3 10 30

**

chor

ea (

0-2

hr)

Chorea

Chorea = involuntary

movements (e.g. trembling)

ADX48621 efficacy in MPTP model

20

ADX48621Commercial Potential

• Levodopa induced dyskinesia (LID) and dystonia are unmet medical needs– Most Parkinson’s disease (PD) patients develop LID– There are no drugs specifically approved for PD-LID – PD represents a growing unmet medical need – Dystonia is poorly addressed by existing drugs– Dystonia market could be significant

• Occurrs in multiple conditions • Occurrs as a side effect of drugs

– PD-LID/dystonia is a faster path to market than PD

• Addex can retain co-promotion rights– PD, PD-LID & dystonia are treated by specialists– geographic split possible– mGluR4 PAM & mGluR2 NAM represent potential additions to future

neurodegenerative disease franchise

21

ADX71943

• ADX71943 is a gamma-aminobutyric acid subtype B (GABAB)receptor PAM

–Clinically/commercially validated mechanismgeneric GABAB agonist, baclofen, is marketed

other orthosteric GABAB agonists are clinically validated

–ADX71943 is differentiatedADX71943 is the only allosteric modulator of GABAB in development

Demonstrated analgesic effects in three preclinical pain models Potential for chronic pain (e.g. osteoarthritis) and other indications

• Phase I to start in 2011

• ADX71943 is available for partnering

22

Phase I

Dose-dependent reduction in nocifensive behaviour.

Statistically significant from the dose of 3 mg/kg.

Phase II

Dose-dependent reduction in nocifensive behaviour.

Statistically significant from between the doses of 3 and 10 mg/kg.

*p<0.05, **p<0.01, ***p<0.001 vs. ADX71943 vehicle ++p<0.01, +++:p<0.001 vs baclofen vehicle

0

20

40

60

80

100

ADX71943vehicle

0.3 1 3 10 30 100 Baclofen,6 mg/kg

Noc

ifen

sive

beh

avio

ur (

s)

.

1st experiment2nd experiment3rd experiment

PHASE I

* *

+++

*****

*

***

ADX71943 (mg/kg)

***

Baclofen,vehicle

ADX71943 significantly reduces nocifensive behaviour in the formalin test in mice after single oral administration

0

20

40

60

80

100

120

140

160

180

200

vehicle 0.3 1 3 10 30 100 Baclofen,6 mg/kg

Noc

ifen

sive

beh

avio

ur (

s)

.

1st experiment

2nd experiment

3rd experiment

PHASE II

***

***

++

***

***

**

ADX71943 (mg/kg)

Baclofen,vehicle

ADX71943

23

ADX71943 significantly attenuates CFA-induced mechanical hyperalgesia after single oral administration

• Single oral of administration ADX71943 caused a dose-dependent increase of the withdrawal threshold in CFA-inflamed rats.

• The effect was statistically significant from the dose of 10 mg/kg.

non-CFA CFA 1 hr 2 hr0

5

10

15

20

25

***

**

Time post-dose (hr)

With

dra

wal

Th

resh

old

Vehicle1 mg/kg ADX-719433 mg/kg ADX-7194310 mg/kg ADX-7194330 mg/kg ADX-7194330 mg/kg Naproxen

24

ADX71943 significantly reduces nocifensive behaviour in the writhing test in mice after single oral administration

• ADX71943 caused a dose-dependent reduction of acetic acid-induced writhing in mice.

• The effect was statistically significant from the dose of 3 mg/kg.

*p<0.05, **p<0.01, ***p<0.001 vs. JQ-02 vehicle; +++p<0.001 vs. 1%CMC vehicle

Experiment 1

Experiment 2

Experiment 3

Experiment 4

ADX71943 (mg/kg)

0

2

4

6

8

10

12

14

16

18

20

22

Baclofenvehicle

ADX71943vehicle

0.3 1 3 10 30 100 3 mg/kg

Nu

mb

er

of

wri

the

s

+++

*********

***

*

** ***

*

Baclofen

25

ADX71943 demonstrates antihyperalgesic effects in the MIA model of OA in rats

0

50

100

150

200

250

300

350

Pre-MIA Post-MIA Day 1 Day 8

With

draw

al th

resh

old

(g)

Max

imum

resp

onse

bew

teen

1 a

nd 2

hr

Vehicle 1 mg/kg ADX71943 3 mg/kg ADX7194310 mg/kg ADX71943 30 mg/kg ADX71943 Celecoxib (30 mg/kg)

* ***

***

*****

Pre-treatment Treatment

Day post-MIA -1 14 14 21

###p<0.001 vs. Pre-MIA baseline; paired t-test, n=10 animals per group. *p<0.05, **p<0.01, ***p<0.001 vs. vehicle;one-way ANOVA with Dunn's multiple comparison n=10 per group.

###

26

Mechanical allodynia in the MIA model of OA in rats

• The effects of ADX71943 and celecoxib in the von Frey test did not reach statistical significance

##

Vehicle 1 mg/kg ADX71943 3 mg/kg ADX71943

10 mg/kg ADX71943 30 mg/kg ADX71943 Celecoxib (30 mg/kg)

0.0

2.5

5.0

7.5

10.0

12.5

15.0

Pre-MIA Pre-drug Day 1 Day 8

Max

imu

m P

aw W

ith

dra

wal

Th

resh

old

(g

)

Days post-MIA -1 14 14 21

Pre-treatment Treatment

##p<0.01 vs. Pre-MIA; paired t-test, 9≤n≤10 animals per group

27

ADX68692

Roles of FSH/LH

Females

• FSH involved in folliculogenesis– maturation of follicles

– estrogen production

• LH triggers ovulation, progesterone

Males

• FSH supports spermatogenesis

• LH stimulates testosterone production

Status

• ADX68692 is a follicle stimulating hormone receptor (FSHR) NAM

• Orally available non-steroidal molecule with drug-like characteristics

• In late preclinical development

• ADX68692 is available for partneringPreclinical Data & Potential Indications

• Statistically significant reduction in testosterone & prostate weight Benign prostatic hyperplasia (BPH)

• Statistically significant reduction in estradiol Endometriosis

28

Control group

02468

1012141618

-14 -11 -8 -5 -2 2 5 8 11 14 17 20 23 26 29

Day of treatment

ADX68692: 20 mg/kg/day

02468

1012141618

-14 -11 -8 -5 -2 2 5 8 11 14 17 20 23 26 29

Day of treatment

In vivo target related efficacy in female rats: 4 week treatment

– Treatment with ADX68692 at pharmacological and multiple of pharmacological doses for 4 weeks was well tolerated

– Treatment for 4 weeks with ADX68692 reduced the number of female rats in the estrus/proestrus phase (ovulatory phase), and increased the number of female rats in the diestrus phase, indicating disruption of menstrual cycle.

– Eventually at the highest dose, animals were found in persistent diestrus

– Histopathological examination showed follicular atresia which is the consequence of the FSH function blockade (blockade of ovulation)

– Estradiol levels were lowered, even in the proestrus phase

SynchronisationTreatment start

To

tal n

um

ber

of

anim

als

in P

roes

tru

s/E

stru

s st

age

29

In vivo target related efficacy in female rats: 4 week treatment

Effect on Estrous cycle duration

0

2

4

6

8

10

12

14

16

18

20

Number of Estrous cycles during treatment duration Mean duration of Estrous cycle (days)

0 mg/kg/day 2x10 mg/kg/day 2x30 mg/kg/day 2x100 mg/kg/day 2x300 mg/kg/day

***

***

***

***

POC that ADX68692 disrupts the estrous cycle and increases its mean duration, eventually leading to complete blockade at high dose.

30

In vivo target related efficacy in male rats: 4 weeks treatment

Circulating testosterone levels were significantly lowered at the lowest and intermediate dose.Relative weight changes were observed for prostate (). These changes were dose dependent and became statistically significant at the highest dose

Group 1 (0 mg/kg/day) Group 2 (2 x 10 mg/kg/day)

Group 3 (2 x 30 mg/kg/day) Group 4 (2 x 100 mg/kg/day)

0.0

5.0

10.0

15.0

20.0

25.0

After 3-week treatment with ADX68692

Tes

tost

ero

ne

(nn

mo

l/L)

******

0.0

0.1

0.2

0.3

0.4

0.5

Prostate: relative weight (g)

*

31

GLP-1 PAM project

• To develop a small molecule GLP-1 PAM for the

treatment of Type II Diabetes– Orally available

– Should increase insulin secretion and decrease blood glucose

– No or reduced nausea/vomiting compared to competition

– Should preferably induce weight loss

– Could be used in combination with other therapies

32

Oral GLP1R PAM in db/db mouse model

• Leptin receptor–deficient db/db knockout mice are considered an established model – develop human Type II diabetes mellitus – hypertension and obesity – disrupted circadian blood pressure (BP) rhythm

• We orally administered to db/db mice– ADX91886 GLP1R PAM – Januvia sitagliptin DPP IV inhibitor– or vehicle

• 15 min later 2 g/kg glucose is given orally• Blood glucose + insulin levels were measured 10 ; 20 ; 30 ;

60 ; 90 min after glucose administration

33

Plasma glucose

15 30 45 60 75 90

0

10

20

30

40

50

C: Sitagliptin 10 mg/kg poB: ADX91886 (220 mg/kg po)

A: Vehicle po

B

*****

******

***

**

**

Time (min)

Pla

sma

glu

cose

(mM

)

Plasma insulin

15 30 45 60 75 90

0

5

10

15

20

25

30

C: Sitagliptin 10 mg/kg po

B: ADX91886 (220 mg/kg po)

A: Vehicle po

B

** ** **

***

***

*** ***

Time (min)

Pla

sma

insu

lin(n

g/m

l)

Oral GPL1R PAM compared to

sitagliptin in db/db mice

Glucose AUCB (0-90 min)

A B C0

5

10

15

20

25

30

A: Vehicle po

B: ADX91886 (220 mg/kg po)

C: Sitagliptin 10 mg/kg po

***

*

Glu

cose

AU

CB

(mM

.hr)

34

mGluR2 NAM

• Data from Addex and others show that mGluR2 inhibition can reverse cognitive deficit – in models of cognitive deficit– in physiologically relevant models of AD– mechanism may be complementary to marketed

drugs

• Published data suggest that mGluR2 inhibition may reduce generation of beta-amyloid*– mGluR2 NAM may also be disease modifying

*The Journal of Neuroscience, March 17, 2010; 30(11):3870-3875

35

Familiar object

Novel object

ADX92639 (mg/kg, p.o.)

120

sham β- Amyloid*

0

30

60

90

t1 t2 t1 t2 t1 t2 t1 t2

Veh 10 30 Donepezil

ADX92639 (mg/kg, p.o.)

Lin

e c

ros

se

s

Locomotor activity during the test

veh 10 30 Donepezil

18

0

3

6

9

12

15 *** *** ***

Exploration of novel vs familiar objects

veh veh veh veh 0

3

6

9

12

15

18

Exp

lora

tio

n t

ime

(se

c)

***

sham β-Amyloid*

• ADX92639 reverses cognitive impairment induced by intracebroventricular (icv) β-amyloid in

the rat NOR test after oral administration:

– Full and donepezil-like reversal of the memory deficit at 30 mg/kg

– No effect on locomotor activity observed during the test

ADX92639 is effective against icv β amyloid-induced deficit in NOR in the rat

*Single administration into the lateral ventricle of 8 μl solutionFinal concentration of amyloid = 2 mg/ml

(1 mg/kg, ip) (1 mg/kg, ip)

36

Allosteric ModulatorDiscovery & Optimization

37

Allosteric Modulation

Because they bind a different site on the receptor, allosteric modulators do not turn receptors on or off the way the body’s natural activators and most drugs do. Instead, they act more like a dimmer switch, offering control over the ease & intensity of (de-)activation while allowing the body to retain its natural control over initiating receptor activation via the active site (e.g. the on/off switch).

38

Allosteric Advantages

• Greater specificity than orthosteric molecules – e.g. mGluRs

• Can target receptors considered “intractable” or only tractable by biologics, peptides or proteins

– e.g. GLP-1

• Non-competitive mechanism–Un-exploited intellectual property–Less dose related toxicity

•Acts like a dimmer not “on/off” switch–Body maintains control of receptor activation cycle

Natural ligand

Time

PAM + natural ligand

NAM + natural ligand

Bio

log

ical

res

po

nse

Allostery preserves natural rhythm

Time

Natural ligand

Agonist

Antagonist

Bio

log

ical

res

po

nse

Orthosterics are steady state

39

Serendipity has yielded some marketed allosteric modulators. Most recent examples include:

• Sensipar/Mimpra cinacalcet (Amgen) – Positive allosteric modulator (PAM) of calcium sensing receptor– Approved 2004– treats secondary hyperparathyroidism

• Selzentry/Celsentri maraviroc (Pfizer) – Negative allosteric modulator (NAM) of CCR5– Approved 2007– Treats CCR5-tropic HIV-1

Allosteric Modulators 101

40

• Allosteric modulators are hard to find– Industrial tools for orthosteric drugs are not appropriate – Pharma has been focused on tools for finding orthosteric drugs– Why didn’t pharma industrialize allosteric drug discovery?

• Upfront investment to build allosteric drug discovery platform was high

• Time to value creation was long & uncertain

• Addex is industrializing allosteric modulation discovery– Proprietary assays

• High throughput screening & optimization tools (384 well plates & robotics)

• Direct detection systems – proximal to target – continuous real-time observation– works for molecules (i.e. allosteric modulators) that do not activate target receptor

– Allostery biased library of over 70,000 compounds

The Addex raison d'être

41

Proprietary Screening AssaysG-Protein Coupled Receptors

• Phoenyx– a cAMP dynamic non stop assay

• FBBA (GLP1R, mGluR7)– Fluorescence-Based Binding Assay – Measures bi-molecular interactions

• Proxylite (GLP1R, GIP)– Proximal & dynamic assays for functional measurements of

all types of GPCRs

42

• APRA (TNFR1)–Accessory Protein Relocalization Assays

• ADX-tags series 1 (IL1R) –Proximal & dynamic assays for functional measurements –Measures activation-dependent association or dissociation of

binding partners

• ADX-tags series 2 (TNFR1, IL1R)–measures conformational changes that lead to activation signal–measures multimerization changes that lead to activation signal

Proprietary Screening Assaystype 1 single-pass transmembrane proteins

43

Allostery Biased Library in-silico analysis

Addex library and marketed drugs share the same physicochemical property space BUT Addex library occupies a unique structural space

Addex library and marketed drugs share the same physicochemical property space BUT Addex library occupies a unique structural space

Physicochemical Comparison Structural Comparison

Addex CompoundsMarketed Drugs

Addex CompoundsMarketed Drugs

44

Platform Validation

• Addex identified selective orally available small molecules for– Challenging GPCRs (mGluRs, GABA-B & A2A)– Peptide receptors (GLP1R, GIPR)– Cytokine receptors (TNFR1 & IL1R1)

• The proof is in the pudding – Merck & Co., Inc. licensed mGluR4 PAM & mGluR5 PAM– Johnson & Johnson licensed mGluR2 PAM

45

Summary

• 15 Programs for High Value Targets/Indications–Highly differentiated allosteric mechanism

–Low target related risk (i.e. mostly clinically validated targets)

• Allosteric Modulator Platform–Proprietary tools + tailored library

–Platform + multi-disciplinary approach are scalable

• 3 Validating Partnerships (MRK/J&J)

• Top Tier Investors

• Cash to end of 2011

46

Management & Boards

Vincent Mutel, Chief Executive Officer Tim Dyer, Chief Financial Officer Charlotte Keywood, Chief Medical Officer Sonia Poli, Head of Non-Clinical Development Laurent Galibert, Head of Inflammation & Metabolic Disorders

Board of Directors

André J. Mueller, Chairman

Vincent Mutel, Vice Chairman & CEO of Addex

Andrew Galazka, SVP Scientific Affairs, Merck-Serono

Ray Hill, former Head of EU Licensing, Merck & Co., Inc.

Vincent Lawton, former MD of Merck Sharp & Dohme U.K.

Beat E. Lüthi, CEO of CTC Analytics

Antoine Papiernik, Sofinnova Partners

Scientific Advisory Board

George F. Koob, Ph.D., Chairman

Bernhard Bettler, Ph.D.

Arthur Christopoulos, Ph.D.

Patrick M. Sexton, Ph.D.

Mark A. Geyer, Ph.D.

Barbara J. Mason, Ph.D.

Jean-Philippe Rocher, Head of Core Chemistry Robert Lütjens, Head of Core Biology Tatiana Carteret, Head of Human Resources

Chris Maggos, Investor Relations & Communications

Executive Management

47

www.addexpharma.com

allosteric modulators for human health