Preclinical Designs/Efficacy Models Supporting Proof of Concept … · 2019-08-28 · Preclinical Designs/Efficacy Models Supporting Proof of Concept Studies Luana Pesco Koplowitz,

2014 ACCP Annual Meeting

Preclinical Designs/Efficacy Models Supporting Proof

of Concept Studies

Luana Pesco Koplowitz, MD, PhD, FCP, FFPM

DUCK FLATS Pharma, LLC

Key Considerations for Efficient Study Designs in Early Stage Clinical Development and Proof of Concept Studies


The author is an employee of DUCK FLATS Pharma, LLC.

Disclosures:

2


Learning Objectives:

1. Identify challenges in the transition from animal

models to clinical development.

2. Understand how drug type (chemical, biologic) and

immunological differences between animals and

humans impact drug development programs.

3. Evaluate how previous successes and failures can

be utilized to improve future drug development

programs.

3


Overview

Most drugs that enter clinical development fail

due to:

• Lack of efficacy – 56%1

(Phase 2 to Submission, 2011-2012)

• Safety - 28%1

(Phase 2 to Submission, 2011-2012)

4


OverviewFailure rates (provided for areas where a failure reason

is given) are particularly high in the areas of oncology

and central nervous system (CNS).

Therapeutic Area Failure Rate1,2

Oncology 29.5%

Central Nervous System (CNS) 14%*

Infectious Disease 9.5%

Musculoskeletal 8.5%

Cardiovascular 8.5%

Other 30%

* Rates provided for areas where a failure reason is reported. Almost half of CNS

failures excluded due to non-disclosure of reason.5


Considerations and Challenges

• Animal models versus human response

o Toxicity

o Innate and adaptive immunity

o Predictability

• Drug Development

o New chemical entity versus biologic versus

approved compound

• Estimation of First-in-Human (FIH) dose

6



• Adequate study designs

• Demonstration of efficacy versus placebo or

marketed drug

o Due to variability of placebo effects, some treatments

may fail to prove drug superiority over placebo even

though their total effects are of clinical relevance3

7



Animal Models and Selection

• Species-specific toxicity

o Liver metabolism in animals very different from humans

o Immunogenicity may prevent proper study in animals

• Assessment of efficacy, PK, and safety in the same

model may not be possible

• Animal models

o Rodent versus non-rodent models

o Model is not well-validated

8



Chemical Entities versus Biologics(Small versus Large Molecule)

• Target and species specificity

• More complex CMC processing needed to optimize

effectiveness of biologics4

• Oral vs. parenteral route (examples: IV, IM, SC, IT)o Impacts drug absorption and bioavailability: Distribution

of biologics via SC administration into tissues is generally

slower compared to IV or oral administration5,6

9



Chemical entities versus Biologics (continued)

• Differences in metabolism/catabolism7

o Metabolites may be active in chemical entities

o Biologics may be catabolized to amino acids or peptides

• Toxicity from biologics typically due to exaggerated

pharmacology8

• Potential for formation of anti-drug antibodies (ADAs)

for biologics

o Can result in decreased drug efficacy9,10

10



Innate and adaptive immunity differences between

animals and humans such as:11,12

• Leukocyte subsets - monocytes, mast cells, CD4+ cells

• Cytokines and cytokine receptors

• Ig isotypes in allergic responseo Both IgE and IgG1 in mice

o IgE only in humans

• Significant bronchus-associated lymphoid tissue in

mice

11


Preclinical Testing

• In vitro experiments

• In vivo experiments

• Wide-range of study drug doses

• Preliminary efficacy, toxicity, and pharmacokinetic,

and metabolism data

• Basis for decision whether to further develop a drug

candidate

12


Estimation of FIH Dose

No observed adverse effect level (NOAEL)

• Highest dose tested in animal species that does not

produce a significant increase in adverse events

versus a control group; toxicology studies

Pharmacologically Active Dose (PAD)

• Starting exposure expected to have pharmacological

activity

• Position on exposure-response curve justified by

risk-benefit considerations

13


Estimation of FIH Dose

Minimum Anticipated Biological Effect Level (MABEL)

• Estimate exposure associated with minimal

pharmacological activity

• Usually apply additional safety factor to give a

No-Effect Level

Highest Non-Severely Toxic Dose (HNSTD)

• Applications in cancer patients

14


Proof of Concept Studies

• Generally Phase 1B/2A

• Provide preliminary evidence of efficacy and safetyo Translation into a beneficial therapeutic effect

• Utilize placebo or a comparator treatment

• Small population to limit potential drug exposure risks

• Biomarkers/surrogate markers to bridge gap between

animal/human and determine efficacy endpoint

• Inform the decision whether to proceed into full drug

development

15


Phase 2 Studies – Potential Issues

• Inadequate design

o Number of studies

o Sample size – must consider patient heterogeneity

• Choice of endpoints

o Provide limited or misleading information on drug efficacy

• Improper execution

o Good Clinical Practice (GCP) failures – can lead to poor

quality data

16


Implementation After Phase 2

Poor implementation of programs at this stage leads

to Phase 3 study failures.

• Current failure rate for Phase 3 studies is about 45%13

• Majority of failures due to efficacy14

• Overall, failure rate at the pivotal development stage

(Phase 3) could be reduced by more adequate Phase

2 designs and dose-ranging studies

17


Efficacy and Safety Models

The following areas of drug development will be

discussed to provide examples of how animal

models translated to human studies:

• Cardiovascular

o Arrhythmia

• Central Nervous System (CNS)

o Inflammatory Disease - Multiple Sclerosis

o Pain Management

• Oncology

18


Cardiovascular (Arrhythmia)

Anti-arrhythmic drugs generally affect cardiac

arrhythmias by modulating conduction velocity, or

effective refractory period, or both.

Drug classification based on electropharmacologic

and electrophysiological properties.

Different animal models needed based on arrhythmia

type (supraventricular and ventricular) and drug

action; choice of model important to determine best

clinical results.

19



Supraventricular tachycardia15,16

Animal models closely resemble clinical features of patients

• In vitro: Microelectrode studies on isolated rabbit heart

preparations provided insight into arrhythmia

mechanisms on a cellular level

o However, this model does not mimic AV-nodal re-entrant

tachycardia in humans

• In vitro and in vivo: Canine models used for atrial flutter [right atrial crush, acetylcholine infusion and rapid pacing

(isolated blood perfused heart), aconitine or delphinine

application to right atrium (anesthetized dogs)]

20



Ventricular arrhythmias15,16

More difficult to align animal models to human patients

• In vivo and in vitro: Porcine models for ventricular

fibrillation (pacing electrode through right cephalic vein to

right ventricle, induction by 60-hz alternating current in

isolated right ventricle)

• In vivo: Mongrel dog models for sudden cardiac death (electrical stimulation performed after induction of anterior

myocardial infarction by occlusion/perfusion on left anterior

coronary artery)

21



Case Study: Sotalol(Approved and marketed in the US and Europe)

• Class II (beta-adrenoreceptor antagonist) properties

and

• Class III (Potassium-channel blocker) properties

o Prolongs the plateau phase of the cardiac action

potential in the isolated myocyte, as well as in isolated

tissue preparations of ventricular or atrial muscle

22



Case Study: Sotalol

• Previous studies usually performed in anaesthetized

animalso Can produce bradycardia and prolong QT interval in

most species

• Dogs typically used in safety pharmacology studies but

differences in PK and drug metabolism compared to

humans

• Non-human primates may be better model to determine

the dose- and plasma-response effects of sotalol

23



Case Study: Sotalol

Findings of study conducted in non-anaesthetized

cynomolgus monkeys:17

• Decreased heart rate

• Prolonged RR, PR, QT and corrected QT intervals

• Little or no effects on the QRS complex, arterial

pressure, or body temperature

24



Case Study: Sotalol

The cardiovascular effects in monkeys occurred in a

similar pattern and to a comparable degree as those

reported in human studies.

Study demonstrated the validity of utilizing conscious

telemetry-instrumented non-human primates for the

cardiovascular safety pharmacology assessment of

drugs prior to FIH testing.17

25


CNS (Inflammatory Disease)

Relative to the human response, mice are highly

resilient to inflammatory challenge.

For example, the lethal dose of endotoxin is 5–25

mg/kg for most strains of mice, whereas in humans,

a dose that is 1 million-fold less (30 ng/kg) has been

reported to cause shock.18

26



Murine Model versus Human Model18

• In humans, acute inflammatory stresses from different

etiologies result in highly similar genomic responses

regardless of genetic make-up

• Poor correlation in individual murine model

responses to acute inflammatory stresses compared

to human and other mice models

27



Multiple Sclerosis

• Chronic inflammatory disease of central nervous system

• Appears to have a large autoimmune component

• Areas of inflammation and demyelination19

o CD4+ T-helper cells reactive to myelin

Produce proinflammatory cytokines [ex: interferon-

gamma (IFN-), osteopontin (OPN) and tumor necrosis

factor-a (TNF-a)]

Known to have a leading role in MS

28



Multiple Sclerosis Case Study 1: IFN- Studies11

Experimental autoimmune encephalomyelitis (EAE)

mimics the demyelination seen in central and peripheral

nerves in MS; widely-used model for this disease.

• Preclinical studies: IFN- protective in EAE as

neutralizing Abs exacerbate disease, potentially by

blocking induction/activation of suppressor activity

• Clinical trials: Treatment with IFN- was found to

exacerbate disease

29



Multiple Sclerosis Case Study 2: Integrin Studies11,20

• Preclinical (mice) studies: Blocking VLA-4 (41

integrin)-VCAM-1 interaction may help in MS

• Clinical trials: Demonstrated successfully in human

trials

These examples highlight how caution is required when

extrapolating results from mouse studies to the clinic, but

suggest that mouse models can successfully predict some

therapies for human disease.

30


CNS (Pain Management)

• Currently, more than 1.5 billion people worldwide

suffer from chronic pain of varying degrees21

• Neuropathic pain, which is the most difficult to

medically treat, affects approximately 3% to 4.5% of

the global population21

• Demand for better management and therapies for

both chronic and acute pain expected to rise with

increased age of the population

31



N-type calcium channels have been recognized as

crucial targets in controlling pain because of their

key role in transmitting pain through the spinal

nerves to the brain.

32



Case Study: Ziconotide(Approved and marketed in the US and Europe)

• Synthetic equivalent of a naturally occurring conopeptide

found in the piscivorous marine snail, Conus magus

• Non-narcotic analgesic drug; acts by blocking N-type

calcium channels

• Available as an intrathecal injection to treat chronic pain

33



Case Study: Ziconotide

• Binds to N-type voltage-sensitive calcium channels

(NVSCCs) [Cav2.2] located on the primary nociceptive

(A- and C) afferent nerves in the superficial layers of

the dorsal horn in the spinal cord22

• Produces potent antinociceptive effects reducing

release of pronociceptive neurotransmitters in dorsal

horn in the spinal cord23

o Inhibits pain signal transmission

34




Non-clinical studies conducted in several different animal

models:22

• Toxicology (acute and subchronic) in rats, dogs, and

monkeys

• Reproductive toxicity in rats and rabbits

• In vivo/in vitro mutagenic and carcinogenic evaluations

• Immunogenicity in mice, rats, and guinea pigs

• Intraspinal granuloma formation in dogs

35




Indication 1 - Chronic/Neuropathic Pain22,23

• Animal models - Potent antinociceptive profile

• Clinical studies

o Severe pain (due to cancer or AIDS): Moderate to

complete pain relief

o Severe chronic pain (mostly neuropathic): Significant

pain relief; decreased consumption of opioids

36




Indication 2: Post-Operative Pain22,23

• Rat incisional model: Demonstration of more potent

and longer activity than morphine

• Clinical study: Significant pain relief at both low and

high doses; decreased consumption of morphine

37


Oncology

The average rate of successful translation from

animal models to clinical cancer trials is less

than 8%.24

38


Oncology

Translation failures due to:

• Complexity of human carcinogenesis process,

physiology, and progression

• Genetic, molecular and physiological limitations

• Animal Models

o While there is some uniformity, best-practice standards do

not exist; experiments do not always include the

appropriate species

• Negative data/findings often unpublished for oncology

(and other drug development areas)

39


Oncology

Case Study: TGN1412

TGN1412 was described as an immunomodulatory

humanized agonistic anti-CD28 monoclonal antibody

developed for the treatment of certain cancers with

immunological etiology.

40


Oncology

Case Study: TGN1412 Preclinical Studies25

• In vitro evaluation in human and non-human cells (rodent and primate)

• In vivo studies in cynomolgus and rhesus monkeys (CD28 receptors affinity for TGN1412 similar to humans)

• Repeat-dose administration in cynomolgus and rhesus

monkeys

• Toxicology studies in relevant species using rat

anti-CD28 antibody jj316 or TGN1112 (Ig variant of

TGN1412)

41


Oncology

Case Study: TGN1412 Dose Calculation25

• Safe dose calculated from non-human primate model

was considered of suitable relevance for calculation of

FIH dose

o TGN1412 showed specificity toward CD28 receptor

expressed on human and non-human primate T cells

• Prior tests in cynomolgus and rhesus monkeys had

demonstrated a dose of 50 mg/kg (administered for

four consecutive weeks) to be safe

42


Oncology

Case Study: TGN1412 Dose Calculation25

• Based on the repeat-dose toxicity studies in cynomolgus

monkeys, NOAEL was considered to be 50 mg/kg per

week for not less than four consecutive weeks.

• A dose of 0.1 mg/kg was decided to be administered to

healthy volunteers based on:

o FDA guidelines

o Minimal Anticipated Biological Effect Level (MABEL)

o Safety Criteria for the safe first dose

43


Oncology

Case Study: TGN1412 Phase 1 Clinical Study (2006)

Clinical study was conducted in six volunteers.

After the very first infusion of a dose:

• All six volunteers faced life-threatening conditions

involving multi-organ failure resulting from rapid

release of cytokines by activated T cells

This was despite the fact that multiple pre-clinical

studies had shown that this drug, (given at much

higher doses in animals), was safe.

44


Oncology

Case Study: TGN1412 What happened and Why?25

• Interpretation of preclinical (primate) studies

o Low-level cytokine release in primate studies should

have prompted more caution in clinical trial

• Insufficient in vitro human studies performed

o Important that human material is as close as possible to

the target tissue

• Choice of starting dose

o Prediction of risk and dose range from animal studies

not always reliable (even from non-human primates)

45


In Summary:Transition from animal models to human studies

• Choice of animal model(s)

o Best fit to potential human response

• Correct dose(s)/dosing range/biomarker(s)

o Minimize toxicity and efficacy issues

• Adequate Proof of Concept/Phase 2 study design

o Critical in lessening Phase 3 failures

• Recognition that certain diseases (such as cancer)

are harder to treat due to nature of disease

o Requires more vigilant interpretation of animal data

46


References:1. Arrowsmith J, Miller P. Trial Watch: Phase II and Phase III attrition rates 2011–2012. Nature

Reviews Drug Discovery 2013;12:569. doi:10.1038/nrd4090.

2. Graul AI, Rabasseda X, Tracy M, Cruces E, Dulsat C, Jago C, Lupone B, Navarro, D, Stringer

M. The Year’s New Drugs & Biologics, 2012. Drugs of Today 2013;49(1):33-68 and 49(2):117-

143. doi:10.1358/dot.2013.49.1.1933991 and doi:10.1358/dot.2013.49.2.1948333.

3. Meissner K, Kohls N, Colloca L. Introduction to placebo effects in medicine: mechanisms and

clinical implications. Phil Trans R Soc B 2011;366, doi:10.1098/rstb.2010.0414.

4. Morrow T. Defining the Difference: What Makes Biologics Unique. Biotechnol Healthc

[internet]. 2004 [cited 24 Jun 2014]; 1(4):24-26,28-29. Available from:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3564302/

5. Prueksaritanont T, Tang C. ADME of Biologics-What Have We Learned from Small Molecules?

The AAPS Journal 2012;14(3):410-419. doi:10.1208/s12248-012-9353-6.

6. Golan DE, Tashjian AH Jr, Armstrong EJ, Armstrong, AW. Principles of Pharmacology. 3rd ed.

Wolters Kluwer; 2011. Chapter 3, Pharmacokinetics. [cited 22 July 2014]. Available from:

https://www.inkling.com/read/principles-of-pharmacology-david-golan-3rd/chapter-3/absorption

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3564302/

https://www.inkling.com/read/principles-of-pharmacology-david-golan-3rd/chapter-3/absorption


References:7. Xu X, Vugmeyster Y. Challenges and Opportunities in Absorption, Distribution, Metabolism,

and Excretion Studies of Therapeutic Biologics. The AAPS Journal 2012;14(4):781-791.

doi:10.1208/s12248-012-9388-8.

8. Vugmeyster Y, Xu X, Theil FP, Khawli LA, Leach MW. Pharmacokinetics and toxicology of

therapeutic proteins: Advances and challenges. doi:10.4331/wjbc.v3.i4.73. 2012.

9. Vincent FB, Morand EF, Murphy K, Mackay F, Mariette X, Marcelli C. Antidrug Antibodies

(ADAb) to Tumour Necrosis Factor (TNF)-specific Neutralising Agents in Chronic Inflammatory

Diseases. Ann Rheum Dis [internet]. 2013 [cited 24 Jun 2014];72(2):165-178. Available from:

http://www.medscape.com/viewarticle/777685

10. Matucci A, Petroni G, Nencini F, Pratesi S, Maggi E, Vultaggio A. Anti-drug antibodies and

clinical implications. Clinical Dermatology [internet]. 2013 [cited 22 Jul 2014];2(2):77–80.

Available from: http://www.clinicaldermatology.eu/materiale_cic/697_1_2/6015_anti/article.htm

11. Mestas J, Hughes C. Of Mice and Not Men: Differences between Mouse and Human

Immunology. Immunol 2004;172:2731-2738. doi:10.4049/jimmunol.172.5.2731.

12. Pedotti R, De Voss JJ, Steinman L, Galli SJ. Involvement of both ‘allergic’ and ‘autoimmune’

mechanisms in EAE, MS and other autoimmune diseases. TRENDS in Immunology

2003;24(9):479-484. doi:10.1016/S1471-4906(03)00233-3.

http://www.medscape.com/viewarticle/777685

http://www.clinicaldermatology.eu/materiale_cic/697_1_2/6015_anti/article.htm


References:13. Retzios A. Why Do So Many Phase 3 Clinical Trials Fail? Part 1: The Effect of Deficient Phase

2 Trials in Therapeutic Areas with High Failure Rates in Phase 3 Studies. Bay Clinical R&D

Services [internet]. 2010 [cited 24 Jun 2014];Ver 31. Available from: http://adrclinresearch.com

14. Hay M, Thomas DW, Craighead JL, Economides C, Rosenthal J. Clinical Development

Success Rates for Investigational Drugs. Nature Biotechnology 2014;32(1):40-51.

15. Bhatt LK, Nandakumar K, Bodhankar SL. Experimental animals models to induce cardiac

arrhythmias. Indian J Pharmacol [internet]. 2005 [cited 18 Jun 2014];37(6):348-357. Available

from: http://medind.nic.in/ibi/t05/i6/ibit05i6p348.pdf

16. Janse MJ, Opthof T, Kléber AG. Animal models of cardiac arrhythmias. Cardiovascular

Research [internet]. 1998 [cited 12 Jun 2014];39;165–177. Available from:

http://cardiovascres.oxfordjournals.org

17. Lynch JJ III, Wilson AW, Hernandez LE, Nelson RA, Marsh KC, Cox BF, Mittelstadt SW.

Dose–response effects of sotalol on cardiovascular function in conscious, freely moving

cynomolgus monkeys. British Journal of Pharmacology 2008;154:1439–1445.

doi:10.1038/bjp.2008.206.

18. Seoka J, Warren HS, Cuencac AG, Mindrinosa MN, Baker HV, Xua W, et al. Genomic

responses in mouse models poorly mimic human inflammatory diseases. PNAS [internet]

2013 [cited 21 May 2014];110(9):3507-3512. Available from:

www.pnas.org/cgi/doi/10.1073/pnas.1222878110

http://adrclinresearch.com/

http://cardiovascres.oxfordjournals.org/

http://www.pnas.org/cgi/doi/10.1073/pnas.1222878110


References:19. Fletcher JM, Lalor SJ, Sweeney CM, Tubridy N, Mills KHG. T cells in multiple sclerosis and

experimental autoimmune encephalomyelitis. Clinical and Experimental Immunology 2010;

162:1–11. doi:10.1111/j.1365-2249.2010.04143.x.

20. Kawamoto E, Nakahashi S, Okamoto T, Imai H, Shimaoka M. Anti-Integrin Therapy for

Multiple Sclerosis. Autoimmune Diseases 2012;6 pgs. doi:10.1155/2012/357101.

21. Vocus/PRWeb. Global Pain Management Market to Reach US$60 Billion by 2015, According

to a New Report by Global Industry Analysts, Inc. San Jose, CA 2011.

22. Smith HS, Deer TR. Safety and efficacy of intrathecal ziconotide in the management of severe

chronic pain. Therapeutics and Clinical Risk Management [internet]. 2009 [cited 17 Jul

2014];5:521–534. Available from: http://www.dovepress.com/articles.php?article_id=3332

23. McGivern JG, Ziconotide: a review of its pharmacology and use in the treatment of pain.

Neuropsychiatric Disease and Treatment [internet]. 2007 [cited 17 Jul 2014]:3(1) 69–85.

Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2654521/

24. Mak IWY, Evaniew N, Ghert M. Lost in translation: animal models and clinical trials in cancer

treatment. Am J Transl Res [internet]. 2014 [cited 21 May 2014];6(2):114-118. Available from:

www.ajtr.org

25. Attarwala H. TGN1412: From Discovery to Disaster. J Young Pharm 2010;2(3):332-336.

doi:10.4103/0975-1483.66810.

http://www.dovepress.com/articles.php?article_id=3332

http://www.ajtr.org/

Documents

Preclinical Designs/Efficacy Models Supporting Proof of Concept … · 2019-08-28 · Preclinical Designs/Efficacy Models Supporting Proof of Concept Studies Luana Pesco Koplowitz,