Frank Bretz Global Head – Statistical Methodology, Novartis

46
The Role of Statistical Methodology in Clinical Research – Shaping and Influencing Decision Making Frank Bretz Global Head – Statistical Methodology, Novartis Adjunct Professor – Hannover Medical School, Germany Joint work with Holger Dette & Björn Bornkamp; Willi Maurer & Martin Posch 44 e Journées de Statistique – 21 au 25 mai 2012, Bruxelles

description

The Role of Statistical Methodology in Clinical Research – Shaping and Influencing Decision Making. Frank Bretz Global Head – Statistical Methodology, Novartis Adjunct Professor – Hannover Medical School, Germany Joint work with Holger Dette & Björn Bornkamp; Willi Maurer & Martin Posch - PowerPoint PPT Presentation

Transcript of Frank Bretz Global Head – Statistical Methodology, Novartis

Page 1: Frank  Bretz Global Head – Statistical Methodology, Novartis

The Role of Statistical Methodology in Clinical Research – Shaping and Influencing Decision Making

Frank Bretz

Global Head – Statistical Methodology, NovartisAdjunct Professor – Hannover Medical School, Germany

Joint work with Holger Dette & Björn Bornkamp; Willi Maurer & Martin Posch

44e Journées de Statistique – 21 au 25 mai 2012, Bruxelles

Page 2: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20112

Drug development ...

... is the entire process of bringing a new drug to the market

... costs between USD 500 million to 2 billion to bring a new drug to market, depending on the therapy

... is performed at various stages taking 12-15 years, where out of 10’000 compounds only 1 makes it to the market • drug discovery [10’000 compounds]• pre-clinical research on animals [250]• clinical trials on humans [10]• market authorization [1]

Page 3: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20113

Drug development process

Page 4: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20114

Four clinical development phasesPhase Number of

subjects per study

Length per study

Study population

Aim

IFirst in human

6 – 20 Weeks – months

Healthy Volunteers

Pharmacokinetics & -dynamics; single & multiple ascending dose studies; bioavailability

IIFirst in patients

50 – 200 Months Patients (narrow population)

Proof-of-concept; dose and regimen finding; exploratory studies

IIISubmission

200 – 10’000

Years Patients (broad population)

Confirmatory, pivotal studies

IVPost marketing

1’000 – 1’000’000

Decades Market New label claims & extensions; publication studies; health economics; pharmacovigilance

Page 5: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20115

Why do we need statisticians in the pharmaceutical industry?

Remember, one way of defining Statistics is ...

... and drug development is

a series of decisions under huge uncertainty !

The science of quantifying uncertainty,Dealing with uncertainty,

And making decisions in the face of uncertainty.

Page 6: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20116

Strategic Role of Statisticians Decision making in drug development

• Integrated synthesized thinking, bringing together key information, internal and external to the drug, to influence program and study design

Optimal clinical study design• Specify probabilistic decision rules and provide operating characteristics

to illustrate performance as parameters change

Exploratory Data Analysis• Take a strong supporting role in exploring and interpreting the data

Submission planning and preparation• Be integrally involved in the submission strategy, building the plans,

interpreting and exploring accumulating data to provide input to a robust and well-thought through dossier

Page 7: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20117

Examples

Page 8: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20118

Four clinical development phasesPhase Number of

subjects per study

Length per study

Study population

Aim

IFirst in human

6 – 20 Weeks – months

Healthy Volunteers

Pharmacokinetics & -dynamics; single & multiple ascending dose studies; bioavailability

IIFirst in patients

50 – 200 Months Patients (narrow population)

Proof-of-concept; dose and regimen finding; exploratory studies

IIISubmission

200 – 10’000

Years Patients (broad population)

Confirmatory, pivotal studies

IVPost marketing

1’000 – 1’000’000

Years Market New label claims & extensions; publication studies; health economics; pharmacovigilance

1 – Ph II dose finding study

2 – Ph III confirmatory study

Page 9: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 20119

Example 1

Adaptive Dose Finding

Page 10: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201110

Notation and framework

Page 11: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201111

Notation and framework

Page 12: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201112

Optimal design for MED estimation

Page 13: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201113

Optimal design for MED estimation

Page 14: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201114

Adaptive Design for MED estimation

Page 15: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201115

Priors for parameters

Page 16: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201116

Procedure: 1) Before Trial Start

Page 17: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201117

Procedure: 2a) At Interim

Page 18: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201118

Procedure: 2b) At Interim

Page 19: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201119

Procedure: 3) At Trial End

Page 20: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201120

Example 2

Multiple testing problems

Page 21: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201121

Scope of multiplicity in clincial trials

Wealth of information assessed per patient• Background / medical history (including prognostic factors)• Outcome measures assessed repeatedly in time: efficacy, safety, QoL, ...• Concomitant factors: Concomitant medication and diseases, compliance, ...

Additional information and objectives, which further complicate the multiplicity problem• Multiple doses or modes of administration of a new treatment• Subgroup analyses looking for differential effects in various populations• Combined non-inferiority and superiority testing• Interim analyses and adaptive designs• ...

Page 22: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201122

Impact of multiplicity on Type I error rate

Probability to commit at least one Type I error when performing m independent hypotheses tests (= FWER, familywise error rate)

Page 23: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201123

Impact of multiplicity on treatment effect estimation

Distribution of the maximum of mean estimates from m independent treatment groups with mean 0 (normal distribution)

Page 24: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201124

Phase III development of a new diabetes drug Structured family of hypotheses with two levels of multiplicity

1. Clinical study with three treatment groups• placebo, low and high dose• compare each of the two active doses with placebo

2. Two hierarchically ordered endpoints• HbA1c (primary objective) and body weight (secondary objective)

Total of four structured hypotheses Hi

H1: comparison of low dose vs. placebo for HbA1c

H2: comparison of high dose vs. placebo for HbA1c

H3: comparison of low dose vs. placebo for body weight

H4: comparison of high dose vs. placebo for body weight

In clinical practice often even more levels of multiplicity

Page 25: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201125

How to construct decision strategies that reflect complex clinical constraints?

Page 26: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201126

Basic idea Hypotheses H1, ..., Hk

Initial allocation of the significance level α = α1 + ... + αk

P-values p1, ..., pk

α-propagation

If a hypothesis Hi can be rejected at level αi, i.e. pi ≤ αi, reallocate its level αi to other hypotheses (according to a prefixed rule) and repeat the testing with the updated significance levels.

Page 27: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201127

Bonferroni-Holm test (k = 2)

Page 28: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201128

Bonferroni-Holm test (k = 2)

Example with α = 0.05

Page 29: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201129

Bonferroni-Holm test (k = 2)

Example with α = 0.05

Page 30: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201130

Bonferroni-Holm test (k = 2)

Example with α = 0.05

Page 31: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201131

Bonferroni-Holm test (k = 2)

Example with α = 0.05

Page 32: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201132

Bonferroni-Holm test (k = 2)

Example with α = 0.05

Page 33: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201133

General definition

Page 34: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201134

Graphical test procedure

Page 35: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201135

Main result

Page 36: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201136

Example re-visited Two primary hypotheses H1 and H2

• Low and high dose compared with placebo for primary endpoint (HbA1c)

Two secondary hypotheses H3 and H4

• Low and high dose for secondary endpoint (body weight)

Proposed graph on next slide• reflects trial objectives, controls Type I error rate, and displays possible

decision paths• can be finetuned to reflect additional clinical considerations or treatment

effect assumptions

Page 37: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201137

Resulting test procedure

Page 38: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201138

Resulting test procedure

Page 39: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201139

Resulting test procedure

Page 40: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201140

Resulting test procedure

Page 41: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201141

Resulting test procedure

Page 42: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201142

Resulting test procedure

Page 43: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201143

Resulting test procedure

Page 44: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201144

Resulting test procedure

Page 45: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201145

Now and future In addition to building and driving innovation internally,

important to leverage strengths externally at the scientific interface between industry, academia, and regulatory agencies

At its best, cross-collaboration is greater than the sum of the individual contributions• Synergy on different perspectives and strengths

Provides opportunity to more deeply embed change throughout industry and to have greater acceptance by stakeholders

An exciting time to be a statistician !

Page 46: Frank  Bretz Global Head – Statistical Methodology, Novartis

| JDS | Frank Bretz | May 25, 201146

Selected References Bornkamp, B., Bretz, F., and Dette, H. (2011) Response-adaptive dose-finding under model uncertainty.

Annals of Applied Statistics (in press)

Bretz, F., Maurer, W., and Hommel, G. (2011) Test and power considerations for multiple endpoint analyses using sequentially rejective graphical procedures. Statistics in Medicine (in press)

Maurer, W., Glimm, E., and Bretz, F. (2011) Multiple and repeated testing of primary, co-primary and secondary hypotheses. Statistics in Biopharmaceutical Research (in press)

Dette, H., Kiss, C., Bevanda, M., and Bretz, F. (2010) Optimal designs for the Emax, log-linear and exponential models. Biometrika 97, 513-518.

Bretz, F., Dette, H., and Pinheiro, J. (2010) Practical considerations for optimal designs in clinical dose finding studies. Statistics in Medicine 29, 731-742.

Dragalin, V., Bornkamp, B., Bretz, F., Miller, F., Padmanabhan, S.K., Patel, N., Perevozskaya, I., Pinheiro, J., and Smith, J.R. (2010) A simulation study to compare new adaptive dose-ranging designs. Statistics in Biopharmaceutical Research  2(4), 487-512.

Bretz, F., Maurer, W., Brannath, W., and Posch, M. (2009) A graphical approach to sequentially rejective multiple test procedures. Statistics in Medicine 28(4), 586-604.

Dette, H., Bretz, F., Pepelyshev, A., and Pinheiro, J.C. (2008) Optimal designs for dose finding studies. Journal of the American Statistical Association 103(483), 1225-1237.

Bretz, F., Pinheiro, J.C., and Branson, M. (2005) Combining multiple comparisons and modeling techniques in dose-response studies. Biometrics, 61(3), 738-748.