|
|
Course
4: Analysis of Microarray Gene Expression Data with Applications in
Pharmacogenomics (PM Session)
Instructor: by
Dr. Mei-Ling Ting Lee (meilinglee@cph.osu.edu),
Ohio
State University
Abstract: The course will begin with a brief introduction to
the usefulness of microarrays, the pros and cons of different types of
microarray platforms and data types. We will discuss the inherent variability in
microarray data and the need for normalization. Using case studies, I’ll
illustrate statistical methods which can be used in analyzing microarray data,
including experimental design, ANOVA, Bayesian methods, multiple testing
procedures, permutation tests, nonparametric tests, and power and sample size
considerations. Unsupervised clustering methods and supervised machine learning
methods will be discussed. Applications to pharmacogenomics will also be
discussed.
Textbook:
Lee,
Mei-Ling T. (2004). Aalysis of Microarray
Gene Expression Data, Kluwer Academic Publishers (Now merged with Springer),
Boston
.
About
the Instructor:
Dr.
Mei-Ling Ting Lee is Distinguished Professor of Biostatistics and Computational
Biology at the
College
of
Public Health
,
Ohio
State
University
. Dr. Lee is a biostatistician with a wide range of research interests in
statistical modeling, methods and applications, including survival and
time-to-event studies, latent disease progression, and nonparametric methods for
clustered data. Her areas of medical application include cancer, occupational
risk, the environment, epidemiology, microbiology, pharmacokinetics, genomics
and proteomics. She was among the first to demonstrate the importance of
replication in microarray studies and the need for assessing sample size and
power for these kinds of studies. Dr. Lee is the founding editor and
editor-in-chief of the international journal Lifetime Data Analysis, the
only international statistical journal that is specialized in modeling
time-to-event data.
Course
5: Meta-analysis (Whole Day Session)
Instructor:
Dr. Michael A Stoto,
Georgetown
University
and
Harvard
School
of Public Health
Abstract:
Concerned with the effective use of existing clinical studies to inform decision
making and health care policy, this short course introduces the basic methods of
systematic review of the medical literature, including meta-analysis.
The principles and methods of systematic reviews, as well as statistical
approaches to meta-analysis for clinical trials and observational studies, will
be introduced and their application illustrated in the context of actual
clinical examples. The use of
meta-analysis to explore data and identify sources of variation among studies is
emphasized, as is the use of meta-analysis to assess drug safety.
About
the Instructor:
Dr.
Michael A. Stoto is a Professor of Health Systems Administration and Population
Health at
Georgetown
University
. An epidemiologist, statistician,
and health policy analyst, Dr. Stoto’s research includes methodological topics
in epidemiology, statistics, and demography, research synthesis/meta-analysis,
community health assessment, risk analysis and communication, drug and vaccine
safety, and performance measurement. He
also works on substantive topics in public health practice, especially with
regard to preparedness; the evaluation of public health interventions, and
infectious disease policy, and ethical issues in research and public health
practice. Dr. Stoto has worked with
the District of Columbia Department of Health to evaluate its hospital emergency
room syndromic surveillance system, and has published extensively in related
areas. He is currently leading the
evaluation team for the DC Healthcare Facilities Emergency Care Partnership
Program.
Dr. Stoto is also an Adjunct Professor of Biostatistics at
the Harvard School of Public Health, and director of the evaluation core of the
CDC-funded
Center
for Public Health Preparedness. He
previously served on the faculty of Harvard’s John F. Kennedy School of
Government, the George Washington University School of Public Health and Health
Services, the Georgetown Public Policy Institute, and the
RAND
Graduate
School
. Before coming to
Georgetown
on a full-time basis in August 2006, Dr. Stoto was a Senior Statistician at the
RAND Corporation and the Associate Director for Public Health in the Center for
Domestic and International Health Security.
From 1987 to 1998 he was a professional staff member at the Institute of
Medicine (IOM), where served as director of the Board on Health Promotion and
Disease Prevention and led numerous projects in public health practice.
Dr. Stoto received an AB from
Princeton
University
and a PhD in Statistics from
Harvard
University
, and is a
Fellow of the American Statistical Association.
Course
6: The Essence of Active-controlled Noninferiority/Equivalence Trials (Whole Day
Session)
Instructor:
Dr. Irving K. Hwang from
Irving
Consulting Group (ICG) /
University
of
Medicine
and Dentistry of
New Jersey
(UMDNJ)
Abstract:
The double-blind, placebo-controlled trials have been the gold standard
for new drug development for many decades. It provides a well-accomplished means
to confirm the efficacy of a new test drug by showing its superiority to
placebo. However, clinical trials with placebo as the “control” sometimes
posed “ethical” dilemma. As more effective drugs become available, the
objectives of clinical investigation of new drugs amend. Oftentimes, it seeks
noninferiority/equivalence of the new drug to an existing effective standard
drug in active-controlled trials.
In this
tutorial, the methods and practice of active-controlled trials will be
thoroughly covered. First, some critical definitions such as assay sensitivity
(AS), historical evidence of sensitivity-to-drug effects (HESDE), appropriate
trial conduct (ATC), and constancy assumption (CA) will be addressed. Next, the
design issues of superiority versus noninferiority/equivalence trials will be
discussed including the forms of null and alternative hypotheses, confidence
intervals, as well as sample size and power calculations. Key notions of
prespecification of a fixed margin and preservation of a fraction of the active
control effect for noninferiority trials will be specifically delineated. A
sample size comparison among these trial designs will be given and discussed. In
addition, switching objectives between superiority and noninferiority in
active-controlled trials will also be covered. Finally, the inherent
difficulties and some useful design alternatives to the noninferiority/equivalence
trials will be rendered.
The focus of this tutorial will be
primarily on concept, reasoning, and practices of well-controlled clinical
trials. Statistical theories and formulas will be provided, but kept to a
minimum. Issues of “why” and “how” in the design and conduct of
superiority versus noninferiority/equivalence trials will be extensively
addressed. Real-life examples for trials will be bestowed and tailored for
illustration and exercise purposes. Knowledge and comprehension of this tutorial
would ensure that when a particular confirmatory clinical trial (e.g., a
noninferiority active-controlled trial) is designed and conducted, its intended
objective(s) would be reached with scientific credibility as well as regulatory
approvability.
Keywords:
Placebo control; active control; substantial evidence; superiority;
noninferiority; equivalence; assay sensitivity; sensitivity-to-drug-effects;
constancy assumption; effects size; noninferiority margin; preservation of a
fraction of active control effect; switching objectives
(Note: A reprint of the book chapter
entitled, “Active-controlled noninferiority/equivalence trials: methods and
practice.” in Statistics in the
Pharmaceutical Industry, 3rd Ed. (Buncher and Tsay ed.), will be
furnished as a part of the tutorial material.)
About
the Instructor:
Dr.
Irving Hwang
is currently President, Irving Consulting Group (ICG) and Adjunct Professor,
University of Medicine & Dentistry of New Jersey (UMDNJ). He specialized in
high-level biostatistical consulting in global new drug development. He consults
on statistical methodologies in clinical trials including design and analysis of
exploratory, confirmatory, adaptive, and active-control trials. He provides
statistical trouble-shooting and resolution for client companies. He also
participates in the independent data monitoring committees (IDMCs).
Previously, Dr. Hwang was Sr. Vice President, Harvard
Clinical Research Institute; Vice President & Head, Global
Biometrics
, Hoechst Marion Roussel, Inc.; Sr. Director, Clinical Research Operations,
Hoechst Roussel Pharmaceuticals, Inc.; and Sr. Director, Clinical Biostatistics
& Research Data Systems, Merck. He was formerly PhRMA Deputy Topic Leader,
ICH E10 Expert Working Group; Member, PhRMA BSS Steering Committee; Program
Chair, ICSA Applied Statistics Symposium; and Co-Chair, PMA/FDA Workshop on
Clinical Trials Monitoring and Interim Analysis in the Pharmaceutical Industry.
He had taught graduate courses in the field of biostatistics in clinical trials
at both
Rutgers
and UMDNJ.
Dr. Hwang has over a quarter century of global drug
development experience with major pharmaceutical and biotech companies in design
and analysis of clinical trials for development of new drugs and vaccines. He
has expertise in many therapeutic areas of drug development (Phase I-IV) such as
cardiovascular-renal, metabolism-endocrinology, infectious disease, AIDS,
neuroscience, rheumatology-bone disease, cancer-oncology,
respiratory-allergy-immunology, dermatology, gastrointestinal disease,
ophthalmic, OTC, hepatology-vaccines, and clinical pharmacology. Notably, he is
an expert in design and analysis of landmark CV mortality/irreversible morbidity
megatrials (e.g., CONSENSUS, 4S, and AFCAPS/TexCAPS) as well as adequate and
well-controlled confirmatory trials. He
has hands-on experiences in many successful NDAs/BLAs and EU registrations (
MAAs
).
Dr. Hwang received his Ph.D. in Statistics from the
Wharton
School
,
University
of
Pennsylvania
. His research interests include PK/PD modeling,
survival analysis, longitudinal analysis, interim analysis/adaptive designs,
and confirmatory clinical trial methodology including design and analysis of
landmark megatrials and non-inferiority/equivalence trials. He has many
professional publications, presentations, and lectures in statistics and
clinical trial applications including DIA, ICSA, and SFDA tutorials.
|
