Scribd
Upload
214 views14 pages

Success Probability Estimation With Applications To Clinical Trials 1st Edition Premium Ebook Download

The book 'Success Probability Estimation with Applications to Clinical Trials' provides an introduction to statistical techniques for estimating success probabilities in clinical trials, focusing on planning and analyzing trial outcomes. It covers basic statistical tools, reproducibility probability estimation, and sample size estimation, along with applications to widely used statistical tests. The content is divided into two main parts: one for general understanding applicable to both statisticians and clinicians, and another for detailed statistical analysis aimed at statisticians.

Uploaded by

ghuynh.ghoangroongn.am
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
214 views14 pages

Success Probability Estimation With Applications To Clinical Trials 1st Edition Premium Ebook Download

The book 'Success Probability Estimation with Applications to Clinical Trials' provides an introduction to statistical techniques for estimating success probabilities in clinical trials, focusing on planning and analyzing trial outcomes. It covers basic statistical tools, reproducibility probability estimation, and sample size estimation, along with applications to widely used statistical tests. The content is divided into two main parts: one for general understanding applicable to both statisticians and clinicians, and another for detailed statistical analysis aimed at statisticians.

Uploaded by

ghuynh.ghoangroongn.am
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

Success Probability Estimation with Applications to Clinical

Trials, 1st Edition

Visit the link below to download the full version of this book:

https://medipdf.com/product/success-probability-estimation-with-applications-to-
clinical-trials-1st-edition/

Click Download Now


CONTENTS IN BRIEF

PART I SUCCESS PROBABILITY ESTIMATION IN PLANNING


AND ANALYZING CLINICAL TRIALS

1 Basic statistical tools 3

2 Reproducibility probability estimation 25

3 Sample size estimation 53

4 Robustness and corrections in sample size estimation 91

PART II SUCCESS PROBABILITY ESTIMATION FOR SOME


WIDELY USED STATISTICAL TESTS

5 General parametric SP estimation 107

6 SP estimation for Student's / statistical tests 115

7 SP estimation for Gaussian distributed test statistics 125

8 SP estimation for chi-square statistical tests 135

9 General nonparametric SP estimation - with applications to the


Wilcoxontest 145

vii
CONTENTS

Preface xv

Acknowledgments xvii

Acronyms xix

Introduction: Clinical trials, success rates, and success probability xxi


1.1 Overview of clinical trials xxii
1.2 Success rates of clinical trials xxiv
1.3 Success probability xxv
1.4 Starting from practice xxvii
1.4.1 Situation I: reproducibility problems xxvii
1.4.2 Situation II: sample size problems xxviii

PART I SUCCESS PROBABILITY ESTIMATION IN PLANNING


AND ANALYZING CLINICAL TRIALS

1 Basic statistical tools 3


1.1 Pointwi se estimation 4
IX
CONTENTS

1.2 Confidence interval estimation, conservative estimation 6


1.3 The statistical hypotheses, the statistical test and the type I error
for one-tailed tests 10
1.4 The power function and the type II error 11
1.5 The p-value 15
1.6 The success probability and its estimation 17
1.7 Basic statistical tools for two-tailed tests 19
1.7.1 Two-sided hypotheses and two-tailed statistical test 20
1.7.2 Two-tailed power function, type II and III errors and SP 21
1.7.3 Two-tailed p-value 22
1.8 Other statistical hypotheses and tests 23

Reproducibility probability estimation 25


2.1 Pointwise RP estimation 26
2.2 RP-testing 29
2.3 The RP estimate and the p-value 32
2.4 Statistical lower bounds for the RP 35
2.5 The 7-stability criterion for statistical significance 37
2.5.1 7-stability and type I errors 39
2.5.2 7-stability and type II errors 39
2.6 Other stability criteria for statistical significance 40
2.7 Comparing stability criteria 43
2.8 Regulatory agencies and the single study 44
2.9 The RP for two-tailed tests 47
2.9.1 RP estimation for two-tailed tests 47
2.9.2 Two-tailed RP-testing and relationship with the p-value 48
2.9.3 Two-tailed stability criteria 48
2.10 Discussing Situation I in Section 1.4.1 49

Sample size estimation 53


3.1 The classical paradigm of sample size determination 54
3.2 SP estimation for adapting the sample size 57
3.3 Launching the trial in practice 59
3.3.1 Equivalence of launching criteria 60
3.3.2 Numerical comparison of launching criteria 61
3.4 Practical aspects of SSE 62
3.4.1 Different SSE strategies 63
3.4.2 The variability in SSE 65
CONTENTS XI

3.4.3 Overall power and overall type I error 66


3.4.4 Evaluating SSE strategies 68
3.5 Frequentist conservative SSE 69
3.6 Optimal frequentist CSSE 72
3.6.1 Optimal conservativeness 73
3.6.2 The Calibrated Optimal Strategy (COS) 75
3.6.3 How COS works 77
3.7 Bayesian CSSE 77
3.8 A comparison of CSSE strategies 82
3.8.1 Study design and comparison tools 83
3.8.2 Results 83
3.9 Discussing Situations I and II in Section 1.4 85
3.10 Sample size estimation for the two-tailed setting 87

Robustness and corrections in sample size estimation 91


4.1 CSSE strategies with different effect sizes in phases II and III 92
4.2 Comparing CSSE strategies in different scenarios 93
4.2.1 Study design 94
4.2.2 Results 95
4.3 Corrections for CSSE strategies 96
4.3.1 Corrections and their equivalence 97
4.3.2 Corrected CSSE strategies 98
4.4 A comparison among corrected CSSE strategies 99
4.4.1 Study design 100
4.4.2 Results 100

PART II SUCCESS PROBABILITY ESTIMATION FOR SOME


WIDELY USED STATISTICAL TESTS

General parametric SP estimation 107


5.1 The parametric model 107
5.2 Power, SP and noncentrality parameter estimation 108
5.3 RP estimation and testing 109
5.4 Sample size estimation 110
5.5 Statistical tests included in the model 111

SP estimation for Student's t statistical tests 115


6.1 Test for two means - equal variances 116
xii CONTENTS

6.1.1 Power and RP estimation 116


6.1.2 Sample size estimation 117
6.1.3 Final comments 118
6.2 Test for two means - unequal variances 119
6.2.1 Power and RP estimation 121
6.2.2 Sample size estimation 121
6.2.3 Final comments 122
6.3 On Student's t RP estimates 122

7 SP estimation for Gaussian distributed test statistics 125


7.1 Test for two proportions 125
7.1.1 Power and RP estimation 126
7.1.2 S ample size estimation 127
7.1.3 Final comments 129
7.2 Test for survival: the log-rank test 129
7.2.1 Power and RP estimation 131
7.2.2 Sample size estimation 131
7.2.3 Final comments 133

8 SP estimation for chi-square statistical tests 135


8.1 Test for two multinomial distributions: 2 x C comparative trial 135
8.1.1 Power and RP estimation 136
8.1.2 Sample size estimation 138
8.1.3 Final comments 139
8.2 Test for S couples of binomial distributions: the Mantel-
Haenszel test 139
8.2.1 Power and RP estimation 141
8.2.2 Sample size estimation 143
8.2.3 Final comments 143
8.3 On chi-square RP estimates 143

9 General nonparametric SP estimation - with applications to the


Wilcoxontest 145
9.1 The nonparametric model 146
9.2 General nonparametric SP estimation 147
9.3 The Wilcoxon rank-sum test 148
9.3.1 Power, SP and their approximations 150
9.3.2 SP estimation 152
CONTENTS xiii

9.3.3 RP estimation and testing 154


9.3.4 Sample size estimation 155
9.3.5 Corrections in sample size estimation 160
9.3.6 Discrete distributions 161
Appendix A Tables of quantiles 163
Appendix B Tables of RP estimates for the one-tailed Z-test 171

References 181
Topic index 187
Author index 195
PREFACE

This book aims to provide a simple and understandable introduction to the statistical
techniques of success probability estimation.
Success probability estimation offers an original and practical perspective to the two
problems: 1) evaluating the statistical significance and the stability of an experiment
whose data analysis is based on a statistical test (e.g. a phase III trial), in order to
avoid, if possible, further confirmatory studies; and 2) planning experiments on the
basis of pilot data (e.g. phase III trials on the basis of phase II data), taking into
account the variability of pilot data.
It is worth noting that these two problems have a common mathematical core, that is,
the estimation of the true power of the test, namely the success probability. Usually,
the power of statistical tests is viewed as a mathematical function, and it is studied
analytically to compare these tests. Here, the new perspective is to estimate the true
power of the tests, in other words, to estimate the success probability of experiments
based on statistical tests.
The introduction, regarding clinical trials in general and reporting of some remark-
able numbers related to them, opens the book. Then, the book is divided into two
Parts: I) Success Probability Estimation in Planning and Analyzing Clinical Trials;
and II) Success Probability Estimation for Some Widely Used Statistical Tests.

xv
XVI PREFACE

The first part presents the concepts related to success probability estimation and their
usefulness in applied statistics, and in clinical trials in particular. Part I is devoted to
both statisticians and non-statisticians (for example, clinicians who are involved in
clinical trials).
The second part is mainly of interest to statisticians. An in depth analysis is provided
on the techniques for success probability estimation, with applications both to repro-
ducibility probability estimation and to conservative sample size estimation of some
widely used statistical tests.

D. D E M A R T I N I

Genoa, Italy
March, 2013
ACKNOWLEDGMENTS

I would like to thank the Department of Statistics of Stanford University, where I


visited several times, for the exceptional intellectual atmosphere I breathed there.
I would also like to thank the Department of Statistics and Quantitative Methods
(DiSMeQ) of the Université degli Studi di Milano-Bicocca for supporting this book.
With regards to single individuals, I wish to warmly thank Dr. Maurizio Rainisio
and Dr. Lucio De Capitani, who revised the manuscript and gave me some useful
suggestions.

D.D.M.

xvii
ACRONYMS

1SES One Standard Error Strategy


3QS Third Quartile Strategy
AN Asymptotic Normality
AP Average Power
BAS Bayesian Strategy
BAT Bayesian Truncated Strategy
COS Calibrated Optimal Strategy
CSSE Conservative Sample Size Estimation
dfs Degrees of Freedom
edf Empirical Distribution Function
iff If and Only if
MC Monte Carlo
MSE Mean Square Error
OP Overall Power
PWS PointWise Strategy
XX ACRONYMS

RP Reproducibility Probability
SP Success Probability
SS Single Study
SSE Sample Size Estimation
WRS Wilcoxon Rank-Sum
INTRODUCTION: CLINICAL TRIALS,
SUCCESS RATES, AND SUCCESS
PROBABILITY

This book considers experiments whose data are analyzed through statistical tests. A
significant outcome of a test is considered a success, whereas a non-significant one
is a failure.
Data are supposed to be collected with a certain amount of randomness, which im-
plies the adoption of statistical tests for data analysis. Consequently, also the out-
comes of the tests, i.e. success/failure, are affected by randomness. So, the probabil-
ity of a successful outcome in these experiments, i.e. the probability of a significant
outcome, is of great interest to researchers, sponsors of research and users of research
results.
Focus is placed on large experiments that have been preceded by pilot ones. A pilot
experiment is often performed in order to achieve data for deciding whether or not
to launch the successive, important study and, if this is the case, to adequately plan
the latter.
One of the contexts in which the framework above can be found is that of clinical
trials. Here, large experiments are phase III trials, and previous phase II studies

XXI
xxii INTRODUCTION

can be considered pilot studies in view of the subsequent phase III studies. A brief
introduction to clinical trials follows, together with some data on their success rates
and an introduction to their individual probability of success.
To conclude, in order to introduce applied problems related to success probability
estimation, and to motivate the latter, two practical situations often encountered in
clinical trials are presented, which can also be understood by those owning minimal
statistical skills.
The context of clinical trials is adopted throughout the book to present, explain,
and exemplify success probability estimation. Nevertheless, the fields of application
of success probability estimation are numerous, and one example is that of quality
control.

1.1 Overview of clinical trials

Clinical trials are implicit to drug development and are conducted to collect safety
and efficacy data for health intervention. Clinical drug development is structured
into four phases (see also the U.S. National Institute of Health - NIH; website:
clinicaltrials. gov) :

■ Phase I trials include introductive investigations to study the metabolism and


pharmacologie actions of drugs in humans, and the side effects associated with
increasing doses. They are also run to furnish early evidence of effectiveness.
■ Phase II trials are controlled studies conducted to evaluate the effectiveness of
the drug for a particular indication in patients with the specific disease under
study and to determine the common short-term side effects and risks.
■ Phase III trials are expanded controlled studies that are performed once prelim-
inary evidence suggesting effectiveness of the drug has been obtained. They
are also intended to gather additional information on the overall benefit-risk
relationship of the drug.
■ Phase IV trials are post-marketing studies that are run to obtain additional in-
formation including the drug's risks, benefits, and optimal use.

In recent years, on average (approximately) 2600 phase I, 3700 phase II, 2300 phase
III, and 1800 phase IV trials annually have been presented for approval under the
United States Food and Drug Administration (FDA) (source: clinicaltrials.gov).
These trials amount to about 60% of those run globally every year. Indeed, as a
rule of thumb, the total amount of trials run worldwide is divided as follows: 60%
under the United States Food and Drug Administration (FDA); 30% under the Eu-
ropean Medicines Agency (EMA); and the remaining 10% under other Agencies,
mainly the Japanese Pharmaceuticals and Medical Devices Agency (PMDA). It fol-
lows that an impressive number of trials are simultaneously in operation around the
world every year.

You might also like