Minimizing risk by optimizing clinical trial design AGAH Workshop 19 April 2008

1

Minimizing risk by optimizing clinical trial designAGAH Workshop 19 April 2008

Workshop

Minimizing risk by

optimizing clinical trial design

and performance

Wolfgang Timmer, Sybille Baumann

2


Overview

Some general considerations about “risk“

General strategies of minimizing risk during the different phases of clinical development

Identification of the factors of risk – The new EMEA Guideline

What is missing in the EMEA Guideline?

Other proposals

Further considerations on design features of clinical studies and the clinical environment

Personal view on particular issues

Discussion at any time during this talk

3


Multiplicity of risk categories in clinical drug development

Project development risk

Competitive risk

Device technology development risk

Market adoption risk

Management risk

Liquidity risk

Health-related risk of clinical investigational subjects(which is referred to in the following)

4


The quality of risks may vary during clinical development

Early clinical development, Phase I (strict sense)First-in-man trials in healthy subjects

Risk mainly related to unanticipated serious toxicity

Special trial designs in studies with healthy subjectsRisk may be related to special design features such as DDI studies, supratherapeutic doses in TQT trials, provocation models, etc.

Clinical efficacy studies, PoC, Phase IIRisk due to lack of efficacyRisk related to special features of the underlying or concomitant

disease

Late-stage clinical development Phase IIIdto.Rare side effects may become apparent in large-scale studies

5


There are different standards of risk awareness

Identification of high riskcompounds and adequate action

Thorough consideration of the demands of aparticular study or novel mechanism of action which might

not be reflected by existing guidelines or by a published case report

Assurance of the standard quality of a clinical study or development programme:Adherence to GCP, ethical and legal requirements and applicable regulatory guidelines.

Application of “state-of-the-art“ study designs and implementation of quality standards that arecommonly accepted in the medical community.

Is the top of thepyramide still missing?

III

II

I

6


Strategies of minimizing risk depending on risk quality (I)

Early clinical development, Phase I (strict sense)First-in-man trials in healthy subjects

Improve the predictive value of preclinical models

Identify high-risk drugs and apply special safety procedures

Do not conduct the trial, if the drug is not safe !

7


Strategies of minimizing risk depending on risk quality (II)

Special trial designs in studies with healthy subjects

Don‘t conduct sophisticated studies too early during a clinical development programme

Some special studies may be needed to assess the risk of later trials in patients

If a very high supratherapeutic dose is suggested by the FDA for a TQT trial, dare to enter a scientific discussion on the suitable dose

Don‘t perform trials that are not essential for dossier submission at an early stage

8


Strategies of minimizing risk depending on risk quality (III)

Clinical efficacy studies, PoC, Phase II

Prospectively define clear ‘Go / No Go‘ Criteria, and don‘t proceed with project development, if the criteria are not met.

Use an adequate active comparator in trials with out-patients

Make use of drop out data and concomitant medication use to collect full information on clinical efficacy

(which may also offer the chance to optimize the sample size)

Carefully consider an additional risk possibly related to interaction with the underlying disease, and perform additional

methodological or safety trials, if applicable

9


Strategies of minimizing risk depending on risk quality (IV)

Late-stage clinical development Phase III

Base large-scale trials on results of adequate dose-range-finding study

Don‘t skip Phase II (!)

Give due consideration to rare side effects by on-line safety monitoring and adequate pharmacovigilance procedures

10

Minimising risk by optimising clinical trial designAGAH Workshop 19 April 2008

Risk-benefit assessment

Which kind of “benefit“ are we talking about ?

a) benefit for the individual b) possible later benefit for the investigational subject ? medical community ?

all studies in healthy subjects few studies in patients all studies in patients

oncology trials orphan indications diseases for which no satisfactory treatment is available

11


Reflection on risk management in clinical trials

A risk greater than minimal is not acceptable in a healthy volunteer trial. However, the implicit risk of a clinical trial will never be zero. Risk has to be minimized as far as possible. This can be achieved by optimizing clinical trial design. But there are more strategies to reduce risk, e.g., perform appropriate

preclinical or clinical studies to support the trial, or first perform other investigations that may make clinical testing unnecessary (identify the “critical path“ in project management).

Even a study which bears a minimal risk may not be acceptable, if there is neither a benefit for the individual subject, nor a chance for a later benefit for other patients.

Sponsors and investigators should implement strategies to early identify and re-assess risk during an ongoing clinical trial.

If the risk assessment changes during a clinical trial, adequate measures should be taken.

12


GUIDELINE ON STRATEGIES TO IDENTIFY AND MITIGATE RISKS FOR FIRST-IN-HUMAN CLINICAL TRIALS WITH INVESTIGATIONAL MEDICINAL PRODUCTS

COMMITTEE FOR MEDICINAL PRODUCTS FOR HUMAN USE(CHMP)

London, 19 July 2007Doc. Ref.EMEA/CHMPSWP/28367/07

DRAFT AGREED BY CHMP EXPERT GROUP 6 March 2007

ADOPTION BY CHMP FOR RELEASE FOR CONSULTATION 22 March 2007

END OF CONSULTATION (DEADLINE FOR COMMENTS) 23 May 2007

AGREED BY CHMP EXPERT GROUP 4 July 2007

ADOPTION BY CHMP 19 July 2007

DATE FOR COMING INTO EFFECT 1 September 2007

KEYWORDS First-in-human, Phase I clinical trials, identification of risk, non-clinicalrequirements, animal models, MABEL, risk mitigation strategies

13


Scope of the Guideline

applies to all new chemical and biological investigational drugs except gene and cell therapy medicinal products

“should be read in conjunction with the published EU guidelines“

outlines factors of risk

covers non-clinical issues for consideration

covers design issues for first-in-man trials

14


Identification of the factors of risk

“For some novel medicinal products, the non-clinical safety programme might not be sufficiently predictive of serious adverse reactions in man, and the non-clinical testing and the design of the first-in-human study requires special consideration.

…

Concerns may be derived from particular knowledge or lack thereof regarding

(1) the mode of action,

(2) the nature of the target, and/or

(3) the relevance of animal models“

15



(1) The mode of action

Novelty and extent of knowledge of the supposed MOA, including …

– Nature and intensity (extent, amplification, duration, reversibility) of the effect on the target,

– Effects on non-targets and subsequent mechanisms,– Type of the dose-response as measured in experimental systems:

linear vs. non-linear (e.g., plateauing, over-proportional increase,U-shaped, bell-shaped).

The following MOAs require special attention:

– Target is connected with multiple signalling pathways (pleiotropic effects),– There is a biological cascade or cytokine release which may lead to an

amplification of the effect (e.g., in the immune system or blood coagulation system).

16

Identification of the factors of risk – (1) The mode of action

Three data points are required to describe a sigmoidal dose-response curve:

There is a risk of misinterpreting the data, if the dose-response curve is non-sigmoidal.

More titration steps are needed to detect an atypical dose-response curve.


Effect

Log Dose

Effect

Log Dose

17


Identification of the factors of risk – (1) The mode of action

If an effect is triggered by a biological cascade or cytokine release, already the effect of a low dose may be amplified, and there may not be any safe dose at all.

Effect

Log Dose

Toxic effects CD3 or CD28 (super-) agonists might serve as an example.

Desired effect size

18



(2) The nature of the target

Consider the extent of knowledge on the target, including …

– Structure,

– Tissue distribution (including expression on cells of the immune system),

– Disease specificity,

– Regulation and level of expression,

– Biological function (subsequent mechanisms should also be considered),

– Polymorphisms of the target and their impact on pharmacological effects.

19



(3) Relevance of animal species and models

The available animal species should be compared to humans, regarding …

– Structural homology of the target,

– Target distribution,

– Signal transduction pathways,

– Nature of the pharmacological effects,

– Metabolism and pharmacokinetic aspects.

Where animal models are perceived to be of questionable relevance,this should be considered as adding to the risk.

20


Relevance of animal species and models– An algorithm on how to deal with the uncertainty

A relevant model or surrogate exists, it is well-known and the method is validated. Use that appropriate model or surrogate.

A relevant model or surrogate exists, but it is not yet known. Search for appropriate model or surrogate.

A particular model or surrogate is of questionable relevance. Perform in-depth evaluation of its relevance, Search for other model or surrogate, Integrate information from in-vivo, ex-vivo and in-vitro studies.

A relevant model or surrogate does not exist. Use of homologous proteins or transgenic animals expressing the human target may be the only choice. Be aware of the risk and take appropriate measures.

A model which is actually not relevant is regarded as relevant by mistake. This is the most dangerous case which must be avoided!

(I)

(IIa)

(IIb)

(IIc)

(III)

21


Unanticipated Serious Toxicity – The TGN 1412 Experience

The CD28-specific mAb TGN1412 rapidly caused a life-threatening “cytokine-storm“ in all six healthy volunteers who received this superagonist . Preclinical testing failed to predict toxicity in man. Further studies were conducted by the NIBSC to develop improved tests for emerging immunomodulatory biologicals. Novel in-vitro procedures have now been reported, in which TGN1412, immobilized in various ways, is presented to human white blood cells in a manner that stimulates the striking release of cytokines and lymphocyte proliferation that actually occurred in vivo in humans. These novel procedures would have predicted the toxicity of TGN1412, but at

the time when the Phase I study was released, these procedures were not known

The Journal of Immunology, 2007, 179: 3325-3331“Cytokine Storm“ in the Phase I Trial of Monoclonal Antibody TGN1412: Better Understanding the Causes to Improve Preclinical Testing of ImmunotherapeuticsStebbings R, Findlay L, Edwards C, Eastwood D, Bird C, North D, Mistry Y, Dilger P, Liefooghe E, Cludts I,Fox B, Tarrant G, Robinson J, Meager T, Dolman C, Thorpe S, Bristow A, Wadhwa M, Thorpe R, Poole C

22



“TGN1412 is here shown to have the capabilityto evoke cytokine release and proliferation ofhuman CD4+ lymphocytes only whenpresented to human PBMC using the effectivemethods identified in these studies.

The methods are: immobilisation by dryingonto plates, binding to endothelial cells andcapture by immobilised anti Fc-antibody.“

Robin Thorpe showed this slide on the occasion of the Conference “Bioanalysis in Clinical Trials“ 2008

23



…

“In contrast to man, Cynomolgus macaques given TGN1412 at any of the doses tested did not experience any gross adverse reaction.

Cynomolgus macaque lymphocytes do not undergo proliferation when stimulated with immobilised TGN1412, unless IL-2 or immobilised anti-monkey CD3 is added to cultures(i.e., it acts like a conventional anti-CD28 mAB).

TGN1412 is superagonistic for human PBMC, but not for Macaque PBMC.“

Robin Thorpe showed this slide on the occasion of the Conference “Bioanalysis in Clinical Trials“ 2008

24



(some of the) lessons learned:

Due to the novelty of the mechanism of action, there was insufficient knowledge about the validity of the preclinical testing procedures.

The preclinical tests that were actually performed prior to human experimentation were not able to predict the toxicity in humans.

“Concerns should have been derived“ from TGN1412‘s mode of action, the nature of the target, and the insufficient knowledge about the relevance of animal species and models.

25


Section 4.4 of the EMEA Guideline: Clinical Aspects

“… Key aspects of the trial should be designed to mitigate … risk factors:

Study population Trial sites First dose Route and rate of aministration Number of subjects per dose increment Interval between dosing of subjects within the same cohort Dose escalation increments Transition to next dose cohort Stopping rules Allocations of responsibilities for decisions with respect to dosing …“

26


Subject for discussion:

What should have been put more clearly in the EMEA Guideline?

The speaker‘s personal opinion is the following:

Clear recommendations concerning specific design features of first-in-human studies are missing, e.g., guidelines regarding group size, staggering of subgroups, dosing intervals, … etc.

The Guidance should dare to mention the term “high risk medicinal products“.

Some mechanisms of actions or substance classes should be named which are per definitionem at high risk.

When is a “frontrunner“ (“Pilotproband“) mandatory?

The Guideline should clearly name some proceedings which are forbidden.

A discussion on the interpretation of the NOAEL / MID may be helpful:Not only the dose, but also the quality of the observed events is important(e.g., target organ toxicity vs. unspecific side effects)

27


Further published EU Guidelines:

Non-clinical aspects:

Non-Clinical Safety Studies For The Conduct Of Human Clinical Trials For Pharmaceuticals (ICH M3), CPMP/ICH/286/95,

Preclinical Safety Evaluation of Biotechnology-derived Pharmaceuticals (ICH S6), CPMP/ICH/302/95,

The Non-clinical Evaluation Of The Potential For Delayed Ventricular Repolarisation(QT Interval Prolongation) By Human Pharmaceuticals (ICH S7B), CPMP/ICH/423/02,

Safety Pharmacology Studies For Human Pharmaceuticals (ICHS7A), CPMP/ICH/539/00, Toxicokinetics: The Assessment Of Systemic Exposure In Toxicity Studies (ICH S3A),

CPMP/ICH/384/95, Position Paper On The Non-clinical Safety Studies To Support Clinical Trials With A

Single Microdose (CPMP/SWP/2599/02).

Clinical aspects:

Guideline For Good Clinical Practice (ICH E6), CPMP/ICH/135/95, General Considerations For Clinical Trials (ICH E8), CPMP/ICH/291/95, EUDRALEX – Volume 10: Clinical Trials. In particular: Chapter I: Application and

Application Form, and Chapter II: Monitoring and Pharmacovigilance.

28


ICH Guideline E8, “General Considerations for Clinical Trials“

This is a very general guideline which is not intended to give special advice on how to deal with high-risk compounds.

However, that guideline may be helpful to optimize clinical trials designin light of the fact that it describes accepted principles in the conduct of clinical trials and overall development strategy.

The guideline integrates several aspects and gives many references to important ICH guidelines and topics.

It does also mention some special circumstances which are safety-relevant,e.g.: - appropriate timing of particular studies,

- how to deal with cases of pregnancy during a clinical trial,- criteria of subject selection which are subject to the clinical phase.

29


Design features of ‘First-in-Man‘ Trials – Prerequisites

Consider all preclinical data (pharmacology, toxicology, pharmacokinetics)

Make sure all required preclinical tests have been conducted(Note that additional preclinical investigations may be required prior to human experimentation, if certain critical issues were identified)

Properly identify the initial dose / Select adequate dose steps

Consider possible class effects

Identify possible “high-risk“ substances, and, if necessary, take additional measures to minimize risk and take care of the subjects‘ safety:

Consider the novelty of the MOA, the nature of the target, and the extent of knowledge about the relevance and validity of the preclinical testing procedures.

Is the biological effect reversible?

Could a particular adverse effect be adequately monitored?

In case of an intoxication, would a causal treatment be available?

30


Estimating the maximum safe starting doseFDA Guidance for Industry

Determine NOAELs [mg/kg] in toxicity studies in appropriate species,

Convert each animal NOAEL to Human Equivalent Dose (HED) based on body surface area,

Select lowest HED, or HED from most appropriate species,

Choose safety factor (normally “10“),

Divide HED by that factor,

Maximum Recommended Starting Dose (MRSD)

Consider lowering the MRSD based on Pharmacologically Active Dose (PAD) (converted to HED, if it is from an in vivo study)

31


When should an increased safety factor (> 10) be applied?

Steep dose-response curve

Severe toxicities

Nonmonitorable toxicity

Unexplained mortality in animal studies

Toxicities without advance warning

Irreversible toxicity

Variable bioavailability

Non-linear pharmacokinetics

Inadequate dose-response data

Novel targets

Animal models with limited relevance

Anything else?

32


The Minimal Anticipated Biological Effective Dose (MABEL)

The MABEL is the anticipated dose level leading to a minimal biologic effect in humans.

For “high-risk compounds“, the MABEL approach is recommended.

The following information should be considered (acc. to EMEA Guideline):- target binding and receptor occupancy studies in vitro in target cells from human and the relevant animal species,- concentration-response curves in vitro in target cells from human and the relevant

animal species, and dose/exposure-response in vivo in the relevant animal species,- PK/PD modelling, whereever possible.

A safety factor may be applied for the calculation of the first dose in human from MABEL.

The safety factor should take into account criteria of risk.

When the methods of calculation (NOAEL, MABEL) give different estimations of the first dose in man, the lowest value should be used.

33


Design features of ‘First-in-Man‘ Trials – Infrastructure

Sufficient and qualified staff must be available.

Technical equipment and remedies for the treatment of emergency situations must be available.

Physicians and study nurses must be trained in emergency procedures.

It is recommended that an anesthesiologist or a physician who has practical experience in the treatment of emergencies is available at the site.

The night following the day when study drug was administered, a physician should stay on the ward and be on duty next door to the volunteers.

A risk management plan must be available.

An emergency call system should be installed in the phase I unit.

The phase I unit should be located in reasonable proximity of a clinic with an intensive care unit.

Volunteers should never stay alone. Single rooms should be avoided.

34


Design features of ‘First-in-Man‘ Trials – InfrastructureEmergency equipment

35


Design features of ‘First-in-Man‘ Trials – InfrastructureEmergency call system

36


Insertion for discussion – clinical environment of FIM trials

Far-reaching recommendations on the clinical environment for first-in-man studies are given in this report:

Expert Scientific Group on Phase One Clinical Trials Final ReportAuthor: Gordon W. Duff (chairman) Publisher: http://www.tsoshop.co.uk

This particular Expert Scientific Group was established following the very serious adverse reactions that occurred in the first-in-man clinical trial of TGN1412 in March 2006.

The Expert Scientific Group has set out 22 recommendations that should increase the safety of volunteers in future clinical trials.

Published 6th December 2006Extent ca. 700p.ISBN 10 0117037222, 13 9780117037229Price £125.00

http://www.tsoshop.co.uk/

http://www.tsoshop.co.uk/bookstore.asp?FO=1159966&DI=507983&DISPLAY=PF

37


Insertion for discussion – clinical environment of FIM trials

Expert Scientific Group on Phase One Clinical Trials Final Report

Recommendation No. 20:

“First-in-man studies of higher risk medicines should always be conducted in an appropriate clinical environment supervised by staff with appropriate levels of training and expertise, with immediate access to facilities for the treatment and stabilisation in an acute emergency, and with pre-arranged contingency availability of ITU facilities in reasonable proximity.“

At present, that maximum recommendation is not common practice.

Note that that recommendation does only refer to higher risk medicines,not to first-in-man trials in general.

38


Design features of ‘First-in-Man‘ Trials – Study conduct (I)

8 subjects per dose level: 6 on active compound, 2 on placebo

If an investigational compound belongs to a novel substance class, it may be advisable, or even mandatory, to treat no more than 4 subjects on the same day.

For the first cohort of the first dose group, the period of time between two subsequent administrations should not be shorter than 20 minutes.When the drug is given p.o., an even longer distance may be appropriate taking into account the rate of absorption (tmax).

A full evaluation of the results of each dose group must be perfomed prior to proceeding to the next higher dose step:

evaluation of safety and tolerability based on individual subject data listings and summary listings, if appropriate, and

evaluation of pharmacodynamic data or surrogates which may provide hints on clinical efficacy, and

evaluation of the plasma levels of the drug and relevant metabolites(on-line pharmacokinetics).

39


Date Dose Group 11st subgroup

Dose Group 12nd subgroup

Dose Group 21st subgroup

Dose Group 22nd subgroup

01 Saturday

02 Sunday Admission (evening)

03 Monday Dosing of N=4

04 Tuesday Admission (evening) Information of subjects Information of subjects

05 Wednesday Dosing of N=4

06 Thursday Discharge Screening examination

07 Friday Start of bioanalytics

08 Saturday Discharge

09 Sunday

10 Monday Start of bioanalytics Screening examination

11 Tuesday

12 Wednesday

13 Thursday Results of Online PK available

14 Friday Dose Escalation Meeting (or TC)

15 Saturday

16 Sunday Admission (evening)

17 Monday Dosing of N=4

18 Tuesday Admission (evening)

19 Wednesday Dosing of N=4

20 Thursday Discharge

21 Friday Start of bioanalytics

22 Saturday Discharge

23 Sunday

24 Monday Start of bioanalytics

Flow Chart of an ascending-dose trial with staggered subgroups (excerpt)

40


Design features of ‘First-in-Man‘ Trials – Study conduct (II)

Allocation of responsibilities:

The results of each dose cohort are discussed in a conference by …

The Principal Investigator (“Hauptprüfer“, LKP acc. to § 4 AMG), Other Investigators who were present on the study day, Other experts of the trial site, e.g., the pharmacokineticist (if applicable), The sponsor‘s experts, e.g., the sponsor‘s responsible medical officer,

the clinical trial monitor, the sponsor‘s pharmacokineticist, etc.

The conference will come to a consensus conclusion which will be documented.Note: The ultimate medical responsibility for the subjects‘ well-being is up to the investigator.

Interim results will be forwarded to the BfArM / EC, if ... requested, ... any alarming findings became apparent which may alter the formerly

favourable risk-benefit assessment.

41


Design features of ‘First-in-Man‘ Trials – Study conduct (III)

Stopping criteria:

Stopping criteria should be pre-defined in the protocol.

It is important to differentiate the following:

Individual stopping criteria (usually not applicable to SD studies)

Stopping criteria referring to progression to the next dose cohort

Stopping criteria referring to termination of the trial, even within a dose cohort

General stopping criteria should always be given in the protocol (such as “more than 1/3 of subjects of a dose cohort experiencing drug-related SAEs …“).

Specific stopping criteria are particular events that are given in the protocol based on the knowledge of non-clinical data or class effects.

42


Design features of ‘First-in-Man‘ Trials – Study conduct (IV)

In special cases, it may be appropriate to dose only one (1) volunteer on aparticular study day [“Pilotproband“, “frontrunner“]:

… if the investigational drug is considered a “high-risk“ drug

… if a possible adverse effect, that could reasonably be expected by thedrug‘s mechanism of action, may only become apparent with a delay

… if any safety results of the ongoing study already give reasons for concern

… if an alarming adverse effect observed in an animal toxicology study might not necessarily be species-specific, and if that effect, in case it occurred in humans,… might not be reversible,… or, if a causal treatment was not available,… or, if the AE could possibly be life-threatening.

43


Design features of multiple-dose (MD) ‘First-in-Man‘ Trials (I)

The design of the MD study should be based on the results of the SD trial.

If certain critical issues have been identified in the SD trial, additional monitoring procedures should be included as a consequence.

The number of dose groups is below the number of dose groups in theSD trial.

The number of subjects per dose level is usually higher as compared to theSD trial: e.g., 12 subjects per dose level: 9 on active compound, 3 on placebo.

44


Design features of multiple-dose (MD) ‘First-in-Man‘ Trials (II)

The highest dose level of the MD trial should be below the MTD identified inthe SD trial, taking into account the higher exposure in steady state.

The period of time between two subsequent administrations can be reduced,if sufficient information has been obtained from the SD study.

However, consider the risk of sensitization after repeated administration which might not have become obvious during the SD trial.

As always, a full evaluation of the results of each dose group must be perfomed prior to proceeding to the next higher dose step.

45


Discussion: Combined Study Protocols (Food Effect + SD + MD)

PRO‘s CON‘sCombination offood effect and SD part: Early investigation of food effect may be Methodological problems:

useful to quantify a possible food Food effect cannot be investigated right at theinteraction which might affect subject safety. beginning acc. to a crossover design because

repeated administration would not be justifiedEarly identification of the recommended at that stage.clinical treatment regimen is useful to mimickthe clinical setting in safety studies.

Combination ofSD part and MD part: Sponsors may wish to combine the SD study The MD part contains a decision matrix and

and the MD study in one protocol to safe several dosing options.some time. This does not improve clarity and transparence.

Safety procedures and timing of measurements may no longer be appropriate in light of the resultsobtained in the SD study.

Any changes of the procedures in the MD part which are not reflected by the options given in the protocol would require submission of a protocol amendment.

The study as a whole (!) can only start when the tox data supporting the MD part are available.

46


Minimizing risk during clinical trials in healthy subjects(Phase I programme, extended definition) – some aspects

Make sure an adequate washout period has been observed since the subject‘s last trial participation. The use of a central volunteers data base is advisable.

Perform a drug screen and (in females) a pregnancy test at each admission to the study site.

Give due consideration to the use of adequate contraception methods.

Consider all available data and “state-of-the-art“ study design (= efficient knowledge management). Seek advice of competent authorities, if needed.

If supra-therapeutic doses are requested for special studies by regulatory authorities, up-titration may be appropriate.

If a different galenical formulation will be administered during the programme via a different mode of administration, e.g., to test bioavailability, perform an adequately designed pilot study to investigate safety and tolerability.

47


Minimizing risk during clinical efficacy trials – some aspects (I)

Consider possible differences in ADME between healthy subjects and (some) patients. Investigate the pharmacokinetics in patients with hepatic or renal impairment, if accumulation is to be expected. Adjust doses, if necessary.

Consider concomitant therapy. Conduct adequate DDI trials in healthy subjects.

A thorough dose-range-finding study should be conducted in clinical Phase II before a high number of patients will be exposed in large-scale trials.

The sample size should be appropriate for test and control groups. Bear in mind that the ratio ‘active drug : placebo‘ does not necessarily need to be 1:1.

48


Minimizing risk during clinical efficacy trials – some aspects (II)

Consider an active comparator to avoid putting the patients at risk due to lack of efficacy. Allow concomitant rescue medication use, if appropriate.

The sample size calculation of large-scale confirmatory trials (e.g., Ph.III) should be based on the results of exploratory trials (e.g., Ph.II).

The sample size of a first clinical PoC-trial should be high enough to detect a clear signal and generate further hypotheses, however, a PoC-trial should not be powered like a confirmatory trial based on a desired product profile.

49


Summary of some points

Different risk categories exist.

Among other factors, risk depends on the knowledge that is already available.

Risk may vary during the clinical development programme.

In first-in-human trials, risk is mainly related to unanticipated toxicity.

The basic procedure to minimize risk is the permanent assurance of the standard quality of a clinical study or development programme: adherence to ethical, legal and regulatory requirements, and “state-of-the art“ trial design.

But there may be special demands of a particular study or development programme which must be identified by consideration of the background story, i.e., by thinking, and thorough review of the available data.

Identifying risk is not enough - Adequate measures must be taken.

In some ways, guidelines should be more precise (personal opinion).

50


“Risk“ is derived from the early italian vocable ‘ris(i)co‘ which denotes the cliff which must be circumnavigated.

Regulatory authorities

Investigator or SponsorClinical investigational subjects

unexpected toxicity ?

may harm sponsors, investigators and regulators

may harm the clinical investigational subjects

H2O

unexpected toxicity which may be predictable when all hints are considered

http://www.risknet.de/fileadmin/template_risknet/images_content/cartoon_03.jpg

Minimizing risk by optimizing clinical trial design AGAH Workshop 19 April 2008

Documents

risk quality

clinical study

factors of risk

clinical development

talk minimizing risk

following minimizing

clinical efficacy studies

risk of later trials