Innovations in Drug Discovery - Novartis Institutes for BioMedical Research

Pharmaceutical Innovation – a personal perspective:how we got here and now where do we go?

Sanjeev Thohan, PhDMarch 28, 2013

Sthohan | March 2013 | Pharma Innovation | Confidential2

Overview…

Historical perspectiveTarget prosecution and safetyLeverageToday and beyond…

Cartoonstock.com


4000 years of medicine


Serendipity to deliberate drug discovery….

In the past, drugs were at times discovered in a haphazard, or even accidental way, with pharmaceutical companies encouraging employees to take vacations in exotic locations and bring back dirt, fungus and other organic material, with tests then done on animals to see what the outcome would be… Gavin MacBeath, Harvard University, (2003).

Multidisciplinary efforts intersecting at a common ground of “target-driven-drugs” that are a consequence of deliberate research and development…


Historical perspective – influential pharmaceuticals

Medicine year ImportanceMorphine 1827 Commercialized by a pharmacy (Merck), pain management (Germany)

Aspirin 1897 Synthetic salicylic acid was commercialized (Germany)

Ether 1842 General anaesthetic, transformed surgery (US)Arsphenamine 1910 Syphillis Treatment (Hoechst, Germany)Insulin 1922 1st hormone therapy, transformed diabetes management

Penicillin 1929 Transformed the treatment of microbial diseases

Chlorpromazine and Haladol

1950 & 1958

Transformed management of psyschosis. (France) (Belgium)

Estrogen+ Progestin 1961 Birth Control Pills, deep social impact (USA)

Digoxin 1962 Changed treatment of heart failure and hypertension (Germany) (France)

Furosemide 1993 Loop diuretic, effective treatment of hypertension

Atorvastatin 1996 Cholesterol lowering medicine (USA)HAART 1996-7 Transforming effect on AIDS patients

L-Dopa (Sweden); Hydrocortisone; Viagra (1996, USA); Ritalin


0.120.23

0.30

0.570.45

0.76

0.56

1.07

0.62

1.37

0.70

1.65

0.79

1.96

0.0

0.5

1.0

1.5

2.0

2.5

Nu

mb

er

of

Ye

ars

In

cre

as

ed

Lo

ng

ev

ity

1988 1990 1992 1994 1996 1998 2000

Increase in Longevity Due toNew Drug Launches

Total Increase in Longevity

New Medicines Increase Longevity

40% of Increase in Life Expectancy

Data source: Lichtenberg. National Bureau of Economic Research Working Paper No. 9754 (Cambridge, MA: NBER, June 2003).

11

“They can mean an extra three months or five months or a year- another Christmas with the family, another season to plant a garden, another passage in the life of a child.” —Donna St. George on new targeted cancer therapies, The Washington Post, 2004


Pharmaceutical Lifecyle


Biotech: larger share of innovative approved drugs

Source: USFDA

Innovativeness of FDA approved drugs (2001 – 2007)


Drug Approvals 2012

39 new drugs last year, marking a 15‑year high. Most approvals since 1997 (↑33% over the last 20 years).

20 of 2012 approvals were first‑in‑class agents

Mullard, NatRev DD 2013

So how do we get there?

It’s a team effort……


Discovery processes: Target identification strategies

Gene expression profiling Focused proteomics, e.g. activity-based protein profiling Pathway analysis – pathway databases, e.g. GeneGo Metacore & Ariadne Phenotype analysis – phenomic database Functional screening (siRNAs, shRNAs) Genetic association Scientific Literature


Properties of an ideal drug target

Target is disease-modifying and/or has a proven function in the pathophysiology of a disease.

Modulation of the target is less important under physiological conditions or in other diseases.

If the druggability is not obvious (e.g. as for kinases) a 3D-structure for the target protein or a close homolog should be available for a druggability assessment.

Target has a favorable ‘assayability ’ enabling high throughput screening. Target expression is not uniformly distributed throughout the body. A target/disease-specific biomarker exists to monitor therapeutic efficacy. Favorable prediction of potential side effects according to phenotype data

(e.g. in k.o. mice or genetic mutation databases). Target has a favorable IP situation (no competitors on target, freedom to

operate).


Therapeutic approaches: target modulation

Traditional Small Molecules

• Enzymes, receptors, transcription factors, ion channels, transport proteins, protein-protein interfaces

Biologics• Extracellular proteins, trans-

membrane receptors, cell surface receptors, substrates and metabolites

Nucleic Acids• RNAi

Novel – Good IP position Antibody drug conjugates Stapled or stabilized

proteins Nanotechnology Repurposed/repositioned

drugs Patent extension


Discovery Processes: Lead Optimization


Drug Failures in Ph II-III

Phase III failures 2007-2010Phase II failures 2008-2010

Nature Reviews Drug Discovery 10, 328-329 (May 2011) Nature Reviews Drug Discovery 10, 87 (February 2011)

Leading reasons for adverse events:• cardiovascular toxicity• hepatotoxicity


What we use in research models

What is

Population Responses

Intrinsic variability• Drug-target or metabolite target interaction• Type of target transduction• Access at the biophase• Delivery and input rate• Metabolism and pheno/genotype• Disease and homeostasis• Placebo response

Extrinsic Variability• Drug-drug interactions• Interactions with endogenous substances


Elements of early safety assessment

• In vitro Safety Pharmacology:– Human target based: associated with clinical adverse reactions

• GPCRs

• Nuclear hormone receptors

• Ion channels

• Transporters

• Kinases

• Proteases

• Other enzymes

• Phenotypic and Organ-toxicity

– Cell- & tissue-based

• Cardiomyocyte-based assays

• Hepatotoxicity

• Hematotoxicity

• Neurotoxicity

• Genotoxicity

Risk Assessment and Mitigation - Integration with ADME and PK data- Computer-assisted Drug Design (SAR)

- Clinical annotation - Bioinformatics and network prosecution


How to identify targets for safety profiling?Reverse translation

Therapeutic effect A

Common ADR

Therapeutic effect B

Relevance to EFPC

Cmax AC50

Define coverage

General application

Molecular target

EFPC: effective free plasma concentrationAC50: concentration necessary to achieve50% activity @ off-target, in vitro


O

Cl

O

O

Chlorotrianisene

Prenylamine

Domperidone

Terfenadine

Azatadine

Ritodrine

Reserpine

a

b

c

d

e

f

g

Side effect profile of synthetic estrogens

Requirements• Large-scale computational effort to predict the activity of marketed drugs against

adverse drug reaction (ADR) targets• In vitro profile of marketed drugs• Biomarker for the off-target related ADRs• Extrapolation to drug candidates to predict ADRs in silico

O

Cl

O

O

Chlorotrianisene

Prenylamine

Domperidone

Terfenadine

Azatadine

Ritodrine

Reserpine

a

b

c

d

e

f

g

Side effect profile of chlorotrianisene

Lounkine et al. Nature (2012)

Predicting targets based on ADRs


Prescription drug cocktails?There is a chance to enhance side effects by taking several promiscuous compounds

Do we take this into account?


Proof of Target – ADR link: 5HT2b agonism - VHD

Specifics of 5HT2b-related VHD

Long latency development, but irreversible*

Hard to detect/diagnose in the clinic

No signal in short regulatory animal studies

Needs special in vivo experimental design to confirm manifestation

30

Fen- Phen

Withdrawn

Dose dependent effect!

Common feature: 5HT2b agonism

Withdrawn

Clinical landscape of VHDRestrictions

Kvernmo, 2006

http://www.google.com/imgres?imgurl=http://spy-spider.com/skull_and_crossbones.png&imgrefurl=http://www.sodahead.com/fun/do-you-want-a-tatoo-if-so-where-and-what-is-it-going-to-be/question-833335/&h=341&w=344&sz=15&tbnid=bqicXfIYUSRQyM:&tbnh=119&tbnw=120&prev=/images?q=skull+and+crossbones+symbol&zoom=1&hl=en&usg=__ysOy7JGzRLWYYSIJH5sP5QLsAmI=&sa=X&ei=tRh0TKezD4L68Aa-u7yGCQ&ved=0CCUQ9QEwCA

http://www.google.com/imgres?imgurl=http://spy-spider.com/skull_and_crossbones.png&imgrefurl=http://www.sodahead.com/fun/do-you-want-a-tatoo-if-so-where-and-what-is-it-going-to-be/question-833335/&h=341&w=344&sz=15&tbnid=bqicXfIYUSRQyM:&tbnh=119&tbnw=120&prev=/images?q=skull+and+crossbones+symbol&zoom=1&hl=en&usg=__ysOy7JGzRLWYYSIJH5sP5QLsAmI=&sa=X&ei=tRh0TKezD4L68Aa-u7yGCQ&ved=0CCUQ9QEwCA


Combination therapy profiles

Look for caution signals and common pathways

Pimozide Risperidone Zolpidem Alprazolam Fluphenazine Haloperidol


Type 2 Diabetes mellitus treatment landscape

Multidrug regimen for diabetes can become more complex with atherogenic dyslipidemia, hypertension, and prothorombotic/proinflammatory states• Cholesterol management (simvastatin)• ACE inhibitors (enalapril)• Low dose aspirin

Monotherapy add add

Obese metformin sulfonylurea exenatide or Insulin orglitazone

Non-obese Sulfonylurea or metformin

exenatide or Insulin orglitazone

elderly Low dose secretagogue Switch to simple insulin regimen

---

Asian glitazone metformin Sulfonylurea orInsulin or Exenatide*(*not approved w glitazone)


Combination therapy profiles

sitagliptinmetformin pioglitazone simvastatin enalapril


Drug Innovation – “the new”

New tools to do things • reagents to explore biological phenomena or new types of drugs, such as aptamers,

chimeric proteins, peptidomimetics, multi-valent antibodies, etc.

New ways to measure things • techniques of scientific observation and measurement, including new visualization

methods, multiplexed assays, real-time biological kinetics measurements and others.

New ‘things’ themselves • devices, including the use of new materials with novel properties.

New ways to handle and extract insights from experimental observations • advances in bioinformatics, data integration, knowledge management, artificial

intelligence and others.

Leveraging the existing

New IP?....


Repositioning

"New technologies however enable the systematic evaluation of any drug or mechanism of action against any disease."

Aris Persidis of Biovista is in the process of filing for novel use patents for 12 drugs with potential in Parkinson's, Alzheimer's, epilepsy, depression and sleep disorders.

Clive Morris, head of the new opportunities division at AstraZeneca, existing medicines that have already been in clinical trials can skip the early phase of drug development and go straight into phase two (mid-stage) trials, which can save three to five years. AstraZeneca is now testing a failed diabetes and obesity drug for use as a glaucoma treatment.

AstraZeneca (October 2012) made 22 failed medicines available to academics through a partnership with the Medical Research Council and it also struck a £180m partnership with its bigger rival GlaxoSmithKline in May to boost the faltering development of new antibiotics, dubbed NewDrugs4BadBugs.

"Typically, repositioning is done by accident, or in a limited way,“- between 2007 and 2009, 30% of all newly marketed medicines

were either existing drugs or new formulations of old drugs. - - Aris Persidis (Biovista - 2012).


Repositioning

The safety advantage. • Existing drugs that are either approved or have been shown to be safe in late-stage trials, but have failed to meet

end points of their originally-targeted indications, can leverage their inherently reduced development risk into potentially new indications. Since safety accounts for approximately 30% of drug failures in clinical trials, this is a significant development advantage that repositioned drugs enjoy.

The money savings advantage. • Relaunching a repositioned drug averages $8.4 million, whereas to relaunch a new formulation of an existing drug

in its original indication costs an average $41.3 million. NCE/NME development averages more than $1.3 billion (160 million times less investment). Repositioning is in a completely different league of investment needed to create a new drug product in the market.

The market potential advantage.• Potential for market success depends on numerous factors, including market need, competition, differentiation, an

excellent product, IP barriers, payer acceptance, compliance and a successful market strategy. These factors apply for repositioned drugs in the same way as they do for NCE/NME drugs as well.

The return on investment potential. • Portfolio strategy: it is prudent to have a reasonable stable of repositioned drugs under development as a portfolio,

to allow for attrition due to potential lack of efficacy (but not safety), when any drug is tested in clinical trials.

The out-licensing potential. • Pharmaceutical companies are said to be exploring new models to out-license some of their clinical drug

candidates that may have been shelved for whatever reason. Benefits: met end points and have proven themselves to be safe. Repositioning grants a pharmaceutical company specific and novel business development possibilities for out-licensing.


Repositioning

Original Use Repurposed

Aspirin (Bayer) - 1897 Inflammation and pain antiplatelet drug for treating and preventing heart attacks and strokes

Ibuprofen (Boots) -1960 Anti-inflammatory Hangovers, rheumatoid arthritis, and Parkinson's disease prevention?

Galantamine (Sopharma) 1960’s - USSR

Plio paralysis/ anesthesia

Alzheimers

Zidovudine (GSK) 1964 Oncology HIV/ AIDS

Rogaine (Pfizer) - 1976 High blood pressure hair loss

Mifepristone (RU486) - 1980 Pregnancy termination Antipsychotic, major depression

Gemzar (Lilly) - 1980s Antiviral cancer drug

Cymbalta (Lilly) - 1990 Antidepressant fibromyalgia

Finasteride (Merck) -1993 Benign prostatic hyperplasia

Hair loss

Topiramate (JNJ) - 1996 Epilepsy Obesity

Viagra (Pfizer) -1996 Heart disease Erectile dysfunction Pulmonary Arterial Hypertension

Today and Beyond…..


Repositioning: explore alternative indications for existing drugs

Network modeling links repositioning objects in a network format.

The network biological relevance is measured by the purity of identified modules and topological parameters such as “betweenness centrality and closeness.”

The network can be used to predict novel repositioning opportunities.

Liu, et.al., DDT 2012

Bioinformatics-based approaches have the potential to offer systematic insights into the complex relationships among drugs, targets and diseases necessary for successful repositioning.


Biologics: 907 medicines and vaccines in development - 2013

338 cancer therapeutics that target several different types of solid tumors, leukemia and lymphoma. Monoclonal antibodies account for 170 of the 338 products in development.

176 candidates in development for an array of infectious diseases, including 134 vaccines.

71 medicines for autoimmune diseases, such as lupus, multiple sclerosis and rheumatoid arthritis.

58 treatments for cardiovascular diseases, such as congestive heart failure and stroke.

Other diseases include diabetes, digestive disorders, genetic disorders, neurologic and respiratory disorders

PhRMA Productivity report 2013


Protein-protein interactions

Today: high-quality interaction

networks of reasonable coverage are available for only a small number of interaction types and model organisms, thus limiting the evolutionary trajectories. Limiting our understanding.

Future opportunities: shift from static 2D to comparative spatiotemporal network. Higher density data with greater resolution means more opportunity.

Yama and Burk. Nat rev mol cell biol 2009


Tools of nanotechnology

Liposomes Nanoparticles Polymeric micelles Dendrimers Nanocantilever Carbon nanotubes Quantum dots

Misra, 2010, DDT


Stapled peptides

Can stapled peptides solve the “undruggable space?”

Historical perspective:• Not orally available in active form• Lack the ability to enter cells • Inactivated by proteases, filtered from the blood by the

kidneys within minutes.

A stabilized α-helix (pink) binds to a key portion of the HIV capsid (blue), as shown in an NMR structure. A hydrocarbon side-chain cross-link (yellow) helps this helix enter cells.Drahl, CEN. Volume 86 Issue 22 (2008)

http://www.aileronrx.com


New Technology: Histopathology imaging

Whole slide images of histology sections resolved into distinct patches (e.g., viable tumor, necrosis) so that each patch can be linked with the outcome.

Computed code, from the learned representation, is then utilized to classify patches from a curated library of images.

Evaluation over 1400 and 2500 samples of glioblastoma multiforme and clear cell kidney carcinoma indicates a performance of 84% and 81%, respectively.

Processes are computationally demandingSolutions may be in High Perfomance Computing – AI learning.

Color coding is black (tumor), pink (necrosis), and green (transition to necrosis).


Opportunities for innovation

FIPNet (fully integrated pharmaceutical network) model of drug development, in which the core capabilities of different stakeholders in the development process are leveraged. CRO, contract research organization

Katin, Clin Pharmacol Ther. 2010 March; 87(3): 356–361.


Sharing?

Securely profile every compound ever synthesized against every assay with multiple computational models in a collaborative manner, using standard technologies without structural disclosure.

There is a need for a universal platform for collaborative drug discovery and development that will allow researchers to collaborate, while retaining refined IP rights.

Bunin & Ekins, DDT 2011

“Integrated specialization”Operate within natural workflows or with minimal interruption to research individuals/teams.Securely and selectively collaborate with anyone or any organization with truly complementary and best in class capabilities… Bunn and Ekins DDT 2011


Questions and thoughts

Innovations in Drug Discovery - Novartis Institutes for BioMedical Research

Health & Medicine