Anapharm-Designing Safe and Efficient Phase 1 Studies

© 2010 PharmaNet Development Group, Inc. All rights reserved.

Designing Safe and Efficient Phase I Studies to Expedite Clinical Development

Mario Tanguay, B.Pharm, Ph.D.Vice President, Scientific & Regulatory Affairs, Anapharm

Guest Professor, Faculty of Pharmacy, University of Montreal

April 12, 2010

The views expressed herein are solely those of the author and do not necessarily reflect the official policy, position or opinions of PharmaNet Development Group, Inc. and its affiliates

Experience You Can Trust

© 2010 PharmaNet Development Group, Inc. All rights reserved.Page 2

Agenda

Study Design Issues

Attempts To Make Phase I Programs More Efficient

IntroductionAre you ready to go to Phase IObjectives of Phase I studies

Recent trends will be presented based on some FIH studies conducted at Anapharm over the last 2 ½ years (n=27)Determination of maximum recommended starting doseStudy population (age, inclusion of females)Sample size considerationsDose escalation scheme



Are You Ready to Move to Phase IThere is often a lot of pressure on the development teams to obtain data in humans (and proof of concept) as quickly as possible

For that reason, Phase I formulation development is often neglected (or simply delayed to a later date)

The challenge: ~60% of NCE in development are poorly soluble1 (poor solubility = dissolution problems, poor and/or erratic absorption …)

The risk: clinical testing may be initiated with a less than optimal formulation for which reliability, reproducibility and scalability are not fully understood

1) H. Dubin, Drug Del. Technol., 2006



Using a Non Optimal Formulation in FIH – Case Example

Less than proportional increase in AUC due to solubility issue

No real conclusion can be drawn about safety data and MTD



Formulations Used in Recent Studies Done at Anapharm

Powder in a bottle used in almost 50% of studies with oral administration



Are You Ready to Move to Phase ISo poorly soluble and/or poorly permeable drugs should require more attention before moving quickly to Phase I

Biopharmaceutical properties must be taken in considerationBCS classification

Neglecting those aspects can make the results from the FIH study useless

No clear demonstration of MTD if a plateau in bioavailability is reachedMay not reach systemic exposure required for the desired pharmacological effect

Some possible advantages of a good Phase I formulation strategyEnhance the bioavailability of the compoundReduce the effect of floodDecrease PK variabilityEnsure scalability to manufacture batch size

Page 6



Objectives of Phase I TrialsDetermination of safety and tolerability

To identify suitable dose or dose range for further studies

Pharmacokinetics of the drug

There is also a trend for adding various objectives in order to obtain more information earlier in drug development process:

Effect of foodDrug-drug interactionsAbsolute bioavailabilityEffect on biomarkersProof of concept in either healthy or patient populationsEtc.

Page 7



Determination of Maximum Recommended Starting DoseFDA proposed a methodology (based on NOAEL) to be used to determine the maximum recommended starting dose (MRSD) in FIH

However, other factors should be considered, especially in the case of high risk compounds (EMEA Guidance, FIH trials), such as:

Drugs affecting immune systemDrugs with novel mode actionDrugs with steep dose-response relationshipWhen there is little knowledge about the nature of the targetWhen available animal models are of questionable relevance

In that context, calculation of MABEL (the minimal anticipated biological effect level) may be warranted

Special attention to PK/PD data

Page 8



Determination of Maximum Recommended Starting Dose

Recent trends observed for studies conducted at Anapharm

NOAEL: safety factor of 10 was applied about 40% of the time

A more conservative safety factor was otherwise used (up to 300)



Study PopulationTrend to involve patients earlier in the clinical development, but healthy volunteers are still generally enrolled in single ascending dose (SAD) phase, unless unsafe or unethical

Phase I studies may include cohorts of target populations, but more often in the multiple ascending dose (MAD) phase, when effects on clinical endpoints or biomarkers may be expected

Recent examples we have seen during MAD:Overweight/obese subjects (diabetes drug)Elderly subjects (drugs used for cognitive disorders)HCV or HIV patients (antivirals)Type II diabetes subjects (hypoglycemia agent)Asthmatic subjects (bronchodilator)

Page 10



Study Population

Age limits - Recent trends observed for studies conducted at Anapharm

Uper age limit typically based on safety considerations

Elderly subjects may be considered if this is the target population



Study Population

Should Females Be Enrolled in FIH Trials

Regulatory positionMost regulatory agencies have encouraged the enrolment of female subjects earlier in clinical development

Safety considerationsDespite the use of appropriate method of contraception, it may not always be safe to include women of childbearing potentialPotential issue for long T ½ drugs that remain in the body following the period of confinement within the Phase I unit

Trends in recent studies that we have conductedOnly 11% of the studies enrolled males onlyFemales of childbearing potential were enrolled in 63% of studies that included females



How Many Subjects Are NeededNo clear requirements from regulatory agencies

Sample size rationale is most of the time not provided in protocols

General consensus is that a minimum of 6 subjects should receive the active drug at each dose level and some subjects should receive a placebo

Experience of Anapharm over the last two yearsIn 67% of studies, each cohort included 6 patients on active treatment and 2 subjects on placeboLowest number of subjects (on active treatment) per cohort was 3, while highest was 12Number of cohorts (1 cohort per dose tested) varied from 4 to 12, with a mean of 7 cohorts

Page 13



How Many Subjects Are NeededStatistical considerations

Minimum AE Incidence Needed to Observe at least one AE Occurrence (70%, 80% and 90% Probability) in a Treated

Subjects Cohort

00.10.20.30.40.50.60.70.80.9

1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Number of subjects on active treatment

AE

inci

denc

e

70%80%90%



What Should be the Dose Escalation ApproachCase-by-case approach based on safety profile of the drug and potential risk

Frequently used approaches:Classic approach using “doubling scheme”

N, 2N, 4N, 8N, 16N, etc.Fibonacci scheme

N, 2N, 3N, 5N, 8N, etc.Hybrid approach

N, 2N, 4N, 6N, 10N, etc.

Should be well justified to prevent questions from regulatory agencies or IRBs

Page 15



What Should be the Dose Escalation Approach

Experience of Anapharm over the last 2 ½ years

Lower dose rangesDoses increased in at least 2-fold icrement manner for 90% of the studies

Higher dose rangesMore conservative (e.g. Fibonacci) approach was used 63% of the time

Examples of dose escalation schemesMore conservative approach

100, 200, 300, 450, 600, 810, 100 mgMRSD was based on NOAEL with safety factor of 10

More aggressive approach1, 3, 10, 30, 100, 300, 600 mg

MRSD was based on MABEL (80 times < NOAEL/10)Interim blinded PK evaluation done before moving to the next cohort in about half the FIH studies

Strongly recommended



Dosing Precautions Within A CohortCase-by-case approach based on safety profile of the drug, starting dose and potential risk

Staggered dosing clearly recommended for high-risk compounds

Approach often used: Dosing 2 subjects (e.g., 1 active and 1 placebo) at least 24 hours before remaining subjects of the cohort

This strategy should ideally be used for all cohorts but is sometimes used only for first cohort(s) (with proper justification)

Dosing may also be spread over the same day, e.g., first two subjects dosed 4 hours apart, on the same day

Again, the dosing strategy within a cohort should always be well justified in protocol

Page 17



Dosing Precautions Within A Cohort

Page 18


Staggered dosing was widely used for biologics and for injectable products



Should We Use Parallel or Cross-Over DesignParallel design is generally considered as a safer approach to SAD Phase I studies

Unknown PK in human (unknown T ½) Unknown PD properties in human (any possibility of prolonged or irreversible toxicitySafety population is larger (more subjects are exposed to the drug as compared to a cross-over approach)

Cross-over designLimit the number of subjects that need to be recruitedAllow a better assessment of PK linearityBut increased risk of losing subjects, thereby limiting safety evaluation

Page 19



Should We Use Parallel or Cross-Over DesignMixed approach sometimes used

Example: 2 or 3 different cohorts of subjects who will receive more than one dose

Page 20

Parallel cohorts with cross-over dosing

From: Zhou Y. Choice of designs and doses for early phase trials. Fundam Clin Pharmacol 18 (2004) 373-8



Should We Use Parallel or Cross-Over Design




How Can We Make a Phase I Program More EfficientNew reality:

Increased pressure for reducing cost of clinical developmentNeed to obtain a proof-of-concept earlier in the drug development processPressure for obtaining more information from a single trial

More demands for “Integrated Phase I protocols”, such as:Single Ascending Dose (SAD) and Multiple Ascending Dose (MAD) within the same protocolFood effect armDrug-drug interaction armCohorts of patients (in addition to healthy volunteers)

Page 22



How Can We Make a Phase I Program More Efficient

Recent Trends Observed for Studies Conducted at Anapharm

44% of the FIH studies were integrated SAD/MAD studies

48% of the FIH studies included a PD evaluation

47% of the FIH studies performed with oral formulations included a food effect evaluation (usually in a cross-over)

Other additional objectives more occasionally seenEffect in CYP2D6 poor metabolizersDrug-drug interaction evaluation



Example of an “Integrated” FIH Protocol (SAD/MAD)A Phase I, Single-Centre, Randomized, Double-Blind, Placebo-Controlled Study of YY-123 in Healthy Subjects and Patients

Page 24

Phase A: Single ascending doses (SAD): (Cohorts 1 to 5) 8 subjects (6 active/2 placebo) per cohort/dose5 cohorts plannedFood effect studies in 1 cohort in a cross-over fashion

Phase B: Multiple doses for 14 day (MAD): (Cohorts 7 to 11)8 subjects (6 active/2 placebo) per cohort/dose4 cohorts of healthy volunteers1 cohort of patients with targeted indication

Phase B of the study planned to start after the 4th SAD cohort

1-2 weeks between each cohort



S = Screening; E = Enrollment, R = Randomization; D = Study Discharge

Cohorts in SAD (n = 6): 4 active drug : 2 placebo, in MAD (n=12): 8 active drug : 4 placebo

S, E-R 40 mg or Placebo - D

Cohort MD1


Cohort MD2


Cohort MD3

S, E-R Oral 20 mg or Placebo

Cohort 1 Period 1

IV 20 mg or Placebo - D

Period 2

S, E-R 40 mg or Placebo

Cohort 2

S, E-R 80 mg or Placebo

Cohort 3 Period 1


Cohort 4

SAD

MAD


Cohort 5

80 mg or Placebo - D

Period 2

- D

Typical Time Course of a SAD-MAD Combo with IV arm and ISCV assessment arm

Page 25



Pros and Cons of “Integrated” FIH ProtocolsPros:

Cost and time savingOnly one protocol to submit to regulatory agency and IRBOnly one study planningMaximize amount of information obtained from a single clinical trial

Cons (or challenges):Less experience from regulatory agencies and IRBsMore complex studies from a logistic standpointNot as much time for review process and decision making before MADNeed for flexible protocol and well defined safety measurements with clear stopping criteria according to the potential risksIncreased risk of amendments during the course of the studyConclusions drawn from non optimally designed study

Page 26




Intensive QT Evaluation in Phase I Trial (SAD/MAD)

Early detection of a cardiac safety (QT) issue can potentially save enormous time and resources spent on development of an unsafe drug

ECG may be recorded and analyzed in Phase I with the same quality as in a thorough QT (TQT) study.

ECGs at multiple time points

Unique opportunity to study the effect on QT interval at doses higher than those that may eventually be studied in a formal TQT study

If QT prolongation is observed, this may prompt the conduct of a TQT study earlier in the drug development process, or cessation of further development




Example from Literature of Intensive QT Evaluation in FIH

Ref: Near-Thorough QT Study as Part of a First-In-Man Study.Malik et al, Journal of Clinical Pharmacology, 2008;48:1146-1157

SAD with 6 cohorts of 8 volunteers (6 active / 2 placebo)

In addition to standard screening and safety ECGs, each subject underwent 3 continuous 12-lead ECG recordings (13-hour duration each) on Day -1, Day 1 and Day 2

Automatic QT-interval measurements made at 63 time points (28 at baseline and 35 on treatment), with points most frequent between 2 and 3 hours (around expected Tmax)

Data points were synchronized to allow calculation of time-matched differences from Day -1 baseline

Placebo subjects of all cohorts were pooled




Intensive QT Evaluation in Phase I Trial (SAD/MAD)

We however must consider the following:No positive control (to establish assay sensitivity)Insufficient power to detect QTc prolongation of 5 ms, but data may allow PK-QTc modellingAbsence of QT prolongation in Phase I would generally not preclude the need for a TQT studyData may be supportive for drugs usually not requiring a TQT study (e.g., large molecules and cytotoxic drugs)



Questions on FIH Studies Frequently Asked by IRB

Justification of the time allowed between cohorts

Justification of the time allowed between sentinels and main cohort

Justification for the non-use of sentinel subjects

Justification of the provision for starting the MAD before SAD cohorts are completed

Dose escalation criteria (should be included in the protocol)

Page 30



Questions Frequently asked by Health Canada (TPD or BGT)

Clinical reviewClarification/justification of the upper age limitClarification regarding the selection of the starting doseClarification regarding selection of doses, dose escalation scheme and stopping criteriaRequest for allowing more time between each cohortRequest for allowing more time between dosing of each subject on the same dayClarification of the stopping rules

Page 31



Phase I and Special PK Studies Often Performed

First time in human, single-and multiple dose

Pharmacokinetics

BA/BE studies

Dose proportionality studies

Drug-drug interactions

Mass balance/metabolism

Food effective study

Page 32

Effect of age

Effect of gender

Renal impairment study

Hepatic impairment

PK/PD studies

Effect on biomarkers

Proof of concept (included in FIH if possible

Many Guidance documents are available from the different regulatory agencies to assist sponsors and CROs in the design and analysis of BA/BE and clinical pharmacology studies



ConclusionSponsors must assure that they have a drug product really suitable for Phase I trials

Safety must always prevail when designing Phase I program

There is a trend for more complex “integrated” FIH study protocols

There is room for creativity in designing Phase I studies; but approach used must assure safety of study participants and must be scientifically sound

The Phase I study represents a unique opportunity to gather information about the study drug within a large range of doses

Growing interest for intensive QT evaluation in FIH studies

Page 33



Acknowledgements■ Jean-François Gagné, M.Sc.

Manager, Protocol Writing Sector, Anapharm

■ Stéphane Lamouche, Ph.DAss. Director, Drug Development & Regulatory Affairs, Anapharm

■ Richard Larouche, B.Pharm, MDDirector, Medical Affairs, Anapharm

■ Annie Ouimet, M.Sc.Senior Regulatory Affairs Associate, Anapharm

■ Eric Shink, Ph.DSenior Biostatistician, Anapharm

■ Fethi Trabelsi, Ph.DDirector, Scientific & Regulatory Affairs, Anapharm

Page 34

Anapharm-Designing Safe and Efficient Phase 1 Studies

Documents