Brian Strom discusses drug safety and the limitations of pre-marketing clinical trials in accurately detecting adverse drug reactions (ADRs). Post-marketing studies using pharmacoepidemiological methods are needed to further evaluate drug safety in large populations over long time periods. Pharmacoepidemiology applies epidemiological study designs like cohort studies and case-control studies to examine drug use and ADRs in real-world settings. However, drug safety research is challenging given issues like confounding factors, measurement of drug exposure, and potential conflicts of interest. Strom argues for increased focus on common ADRs rather than just rare events to better understand drug safety.
Value Proposition canvas- Customer needs and pains
Studying drug induced-disease
1. Brian L. Strom, M.D., M.P.H.
Chair and Professor, Department of Biostatistics and
Epidemiology
Director, Center for Clinical Epidemiology
and Biostatistics
George S. Pepper Professor of Public Health and
Preventive Medicine
Professor of Biostatistics and Epidemiology,
Medicine, and Pharmacology
Vice Dean for Institutional Affairs
University of Pennsylvania School of Medicine
Senior Advisor to the Provost for Global Health
Initiatives
University of Pennsylvania
2. What Are Your Drugs Really Doing To Your
Patients? Epidemiological Approaches
To Studying Drug-induced Disease
• Introduction
• Current System
–Premarketing
–Postmarketing/
Pharmacoepidemiology
• Pharmacoepidemiology and
Dermatology
CCEB
3. What Are Your Drugs Really Doing To Your
Patients? Epidemiological Approaches
To Studying Drug-induced Disease
• Introduction
• Current System
–Premarketing
–Postmarketing/
Pharmacoepidemiology
• Pharmacoepidemiology and
Dermatology
CCEB
4. Conflict of Interest Disclosure
•
•
•
•
•
•
Funding from the National Institutes of Health; Agency for Healthcare Research and Quality (including CERT funding, DEcIDE [Developing
Evidence to Inform Decisions about Effectiveness] funding, and patient safety funding; Pfizer Pharmaceuticals; and Takeda Pharmaceuticals
North America
Grants from Alza Corporation, Andrew W. Mellon Foundation, Asia Foundation, Bayer Corporation, Berlex Laboratories, the Burroughs
Wellcome Company, Charles A. Dana Foundation, Ciba-Geigy Corporation, Health Information Designs, Inc., Hoechst-Roussel
Pharmaceuticals, Hoffman-La Roche, Inc., Integrated Therapeutics, Inc., a subsidiary of Schering-Plough Corporation, International Clinical
Epidemiology Network, Inc., International Formula Council, John Wiley & Sons, Ltd., Joint Commission on Prescription Drug Use, Marion
Merrell Dow, Inc., McNeil Consumer Products, McNeil Pharmaceuticals, Mead Johnson Pharmaceuticals, Merck and Company, Institute of
Medicine of the National Academy of Sciences, Novartis Pharmaceuticals Corp., Pfizer Pharmaceuticals, PharMark Corp, A.H. Robins
Company, Rockefeller Foundation, Rowell Laboratories, Sandoz Pharmaceuticals, Schering Corporation, Smith Kline and French
Laboratories, Sterling Winthrop Inc., Syntex, Inc., Takeda Pharmaceuticals North America, the Upjohn Company, US Agency for International
Development, US Pharmacopeia, US Veterans Administration, Wyeth-Ayerst Research
Pharmacoepidemiology training program support has been provided by NIH and from Alza Corporation, Amgen, Inc., Aventis
Pharmaceuticals, Inc., Bayer Corporation, Berlex Laboratories, Inc., Ciba-Geigy Corporation, Genentech, Inc., Hoechst-Marion-Roussel, Inc.,
Integrated Therapeutics Group, Inc., Johnson and Johnson, Merck and Company, Inc., McNeil Consumer Product Company, McNeil
Consumer Healthcare, Novartis Pharmaceuticals Corporation, Pfizer, Inc. , SmithKline Beecham Pharmaceuticals, Whitehall-Robins
Healthcare, and Wyeth-Ayerst Research
US FDA Special Government Employee for serving on FDA advisory committees, and was a member of the FDA Drug Safety and Risk
Management Advisory Committee
Consultant to: Abbott Laboratories, Aetna, Alza Corporation, Astellas Pharma Europe BV, Astra-Merck, AstraZeneca LP, Aventis
Pharmaceuticals, Bayer Corporation, Berlex Laboratories, Blue Cross and Blue Shield, Biogen Idec, Bracco Diagnostics, Inc., Bristol-Myers
Squibb Company, Centocor, Inc., Cephalon, Inc., Churchill Communications, Ciba-Geigy, Inc., Connaught Laboratories, CV Therapeutics,
Cygnus Corporation, Inc., Daiichi Pharmaceuticals UK, Ltd., Dupont-Merck, Eli Lilly and Company, Ethicon, Forest Research Institute.
GlaxoSmithKline, Hoechst-Roussel Pharmaceuticals, Inc., Hoffman LaRoche, IBEX Technologies Corporation, IMS Health, Inflexxion, Inc.,
Inveresk Research North Carolina, Inc., IOM/National Academies of Science, Janssen Pharmaceuticals, McNeil Consumer Products
Company, Javelin Pharmaceuticals, Luitpold Pharmaceuticals, Mediwound, Mikalix and Company, Novartis, Omnicare, Inc., Orchid
Bioscience, Inc., Oscient Pharmaceutical Corp., Pfizer, Inc., PharMark Corporation, Quintiles Strategic Research and Safety/The Lewin
Group, Inc, Rhone Poulenc Rorer Pharmaceuticals, Inc., Roche Laboratories, Inc., RW Johnson Pharmaceutical Research Institute, SanofiAventis, Sanofi Pasteur, Inc., Schering-Plough Research Institute, Science, Toxicology, and Technology Consultants, Searle, Shire
Pharmaceuticals, Syntex,USA, Inc., Takeda Pharmaceuticals North America, TAP Pharmaceuticals, Teva Neuroscience, Inc., Value Health
Sciences, Warner Lambert, Wyeth Consumer Healthcare Division, and numerous law firms
Former Member of the Board of Directors of Medco Health Solutions, Inc.
•
CCEB No support was provided for this talk
5. • “A desire to take
medications is, perhaps,
the greatest feature which
distinguishes man from
other animals.”
Sir William Osler, 1891
CCEB
6. • “If the whole materia medica,
as now used, could be sunk to
the bottom of the sea, it would
be all the better for mankind ,
and all the worse for the
fishes.”
Oliver Wendell Holmes
Medical Essays, “Comments and Counter”
Currents in Medical Science
CCEB
7. What Are Your Drugs Really Doing To Your
Patients? Epidemiological Approaches
To Studying Drug-induced Disease
• Introduction
• Current System
–Premarketing
–Postmarketing/
Pharmacoepidemiology
• Pharmacoepidemiology and
Dermatology
CCEB
8. Phases of Drug Development
PC 1
2
3
4
Drug Approval
PC: Preclinical studies
1: Dose escalation in normals
2: Dose ranging, first time in patients
3: Pivotal trials for registration
4: Post-marketing, not always required
CCEB
9. Limitations of
Pre-marketing Trials-1
• Carefully selected subjects may
not reflect real-life patients in
whom drug will be used
• Study subjects may receive
better care than real-life patients
• Short duration of treatment
• No info on comparative
effectiveness
CCEB
10. Limitations of
Pre-marketing Trials-2
development costs lead to need
for immediate huge sales (“blockbuster
drugs”), and aggressive marketing
practices
• Yet, development programs with 3000
patients cannot reliably detect adverse
events with an incidence of < 1 per
1000, even if severe
CCEB
12. Resulting Opportunities
• 51% of drugs have label changes
due to major safety issues
discovered after marketing
• 20% of drugs get new “black box”
warnings after marketing
• 4% of drugs are ultimately
withdrawn for safety reasons
CCEB
13. Other Issues in Current System
• No incentive for sponsor to
complete promised postmarketing safety studies
• DTC ads lead to over-use of
the drug by patients for whom
use of the drug is not
compelling
CCEB
14. Net
Effect
• Public misunderstands “safety”:
post-marketing discovery of a drug
ADR means someone “messed up”
• Increasing concern about the
safety of our drugs
• Over-reaction leads to increased
pre-marketing requirements with
delayed access and drugs dropped
from development
CCEB
16. What Are Your Drugs Really Doing To Your
Patients? Epidemiological Approaches
To Studying Drug-induced Disease
• Introduction
• Current System
–Premarketing
–Postmarketing/
Pharmacoepidemiology
• Pharmacoepidemiology and
Pediatrics
CCEB
18. Options in Research Design
• Analytic Studies
– Experimental Study
– Prospective Cohort Study
– Retrospective Cohort Study
– Case-Control Study
• Descriptive Studies
CCEB
– Analyses of Secular Trends
– Case Series
– Case Reports
21. Pharmacoepidemiology:
Unique Setting
• A large population needs to be
studied
• Randomized clinical trials are
less likely to be productive
• Answers often must be
obtained very quickly
CCEB
24. Methodologic Issues of Special Concern
for Pharmacoepidemiology
• Measurement of exposure
• Confounding by indication/
channeling
CCEB
25. Pharmacoepidemiology:
Other Unique Characteristics
•
•
•
•
•
Some studies can be very expensive
Major role played by industry
- Premarketing studies
- Funding for postmarketing studies
- Contract Research Organizations (CROs)
Interplay of industry vs. regulators
Enormous public interest in drug safety
Rife with risk of conflict of interest
CCEB
26. Key Problem of “Historical”
Pharmacoepidemiology
• Adverse drug events are the most
common iatrogenic causes of patient
injuries
• Most are the result of an exaggerated but
otherwise usual pharmacological effect
of the drug
• Yet, historically these have been ignored
by pharmacoepidemiology, as they do
not represent a focus of commercial and
regulatory interest
CCEB
27.
28. Data Sources for Pharmacoepidemiology
Studies
• Spontaneous case reports of adverse
reactions
• Aggregate population-based data
sources
• Computerized collections of data from
organized medical care programs
• Data collected for pharmacoepi on an
ongoing basis
• Existing data collected as part of other
ad hoc studies
• Data collected de novo
CCEB
29. Spontaneous Reports of Adverse
Reactions: Advantages
• Incorporates all drugs
• Incorporates all prescribers
• Relatively inexpensive
CCEB
30. Spontaneous Reports of Adverse
Reactions: Disadvantages
• Under- or over-ascertainment
• Under-reporting
• External events can change
ascertainment or reporting
• No denominators
CCEB
31. Data Sources for Pharmacoepidemiology
Studies
• Spontaneous case reports of adverse
reactions
• Aggregate population-based data
sources
• Computerized collections of data from
organized medical care programs
• Data collected for pharmacoepi on an
ongoing basis
• Existing data collected as part of other
ad hoc studies
• Data collected de novo
CCEB
32. Computerized Collections of Billing Data:
Sources of Data
Provider: Pharmacy
Provider: Hospital
Provider: Physician
CCEB
Payor
Data
User
33. Data Sources for Pharmacoepidemiology
Studies
• Spontaneous case reports of adverse
reactions
• Aggregate population-based data
sources
• Computerized collections of data from
organized medical care programs
• Data collected for pharmacoepi on an
ongoing basis
• Existing data collected as part of other
ad hoc studies
• Data collected de novo
CCEB
34. Use of Pharmacoepi
to Study Drug Mechanisms
• Risk factors for druginduced disease
• Pharmacogenetics
• Molecular pharmacoepi
• Epidemiologic study of drug
interactions
CCEB
37. Evolution of Therapeutics
Empiric Choice of Therapy
Statistical Predictive Models of Patients
Likely to Benefit or Suffer Harm
Personalized Medicine
CCEB
38. Evolution of Therapeutics
Empiric Choice of Therapy
Statistical Predictive Models of Patients
Likely to Benefit or Suffer Harm
RiskMAPS
Personalized Medicine
CCEB
39. What Are Your Drugs Really Doing To Your
Patients? Epidemiological Approaches
To Studying Drug-induced Disease
• Introduction
• Current System
–Premarketing
–Postmarketing/
Pharmacoepidemiology
• Pharmacoepidemiology and
Dermatology
CCEB
40. Pharmacoepidemiology and
Dermatology: Opportunities
• Skin reactions are among the most
common types of ADRs
• More toxic drugs are now being
used in dermatology
• That ongoing experience represents
an enormous opportunity to learn a
huge amount about the effects of
drugs on skin, and vice versa,
through the use of
CCEB
pharmacoepidemiology techniques
41. Pharmacoepidemiology and
Dermatology: Issues
• There are few trained pharmacoepidemiologists in the world
– Only 1 NIH training grant, with only
2 slots
– Multiple headhunter calls/week
– Under FDAAA, FDA doubling its
pharmacoepidemiology group
• There are many fewer trained
pharmacoepidemiologists in
CCEB dermatology
42. Selected Examples of Recent
Dermatopharmacoepidemiology Issues
•
•
•
•
Accutane: use and effects
Acne drugs: side effects
Immunosuppressives for psoriasis: ADRs
Stevens-Johnson Syndrome and TEN:
Drug-induced
• Wound healing: predictors and new
treatments
CCEB
Notes de l'éditeur
First, the obligatory coi disclosure. It is also a test of your distance vision!
I could read it all too you, but there would be no more time for the rest of the talk. However, [by category]
To quote Sir William Osler
Reading the newspaper nowadays makes one think of this quote from Sir Oliver Wendell Holmes.
While perhaps overly pessimistic, it has certainly fed the development of pharmacoepidemiology
•
•
•problems--significant piece is statistical
•
• Usually ONLY 500-3000 patients studied before marketing
• In the wake of the drug crises of the past year or so, phase 3 is being prolonged, and drug approval delayed
Inevitably leaves questions to be answered
These are successes, not failures--the issue is to detect them earlier.
• Includes clinical decisions
• Bridging field
• In many ways, it is also the application of clinical pharmacology to public health
PS and IV
I had the honor of, and suffered the turmoil from, being one of the first academics to try to use such techniques, in the early 1980s. At that point it was very controversial. Now it is considered second nature as a or the central tool of pharmacoepidemiologists.
• Risk factors for drug-induced disease
--Lesson for my trainees
--Interventions to prevent use in those at high risk
--Ex. sulfa study--is a prior sulfa AB allergy a RF for reactions to non-Ab sulfa exposures
• Pharmacogenetics
--Bench findings are leading to follow-up epi studies, eg nsaids and MI (Steve)
--In the future, genetics can be applied as risk factor in DUR
• Molecular pharmacoepi
-- using molecular measures as risk factor in epi studies,
-- eg WISE
• Epidemiologic study of drug interactions
--cisapride, terfenadine, ceruvastatin all led to epi studies
--epi studies of antipsychotics and sudden death/VT, based on QT prolongation
--beginning use of epi to examine interactions earlier, eg Worth using GRPD to study interactions between warfarin and antibiotics
• Common definitions
• In some ways, the therapeutic use of these terms is similar
Tailoring therapy by population averages
Tailoring therapy by individual
•
•
•problems--significant piece is statistical
•
Accutane: Studies of birth defects, using epi to change rxing, and then use of epi methods to evaluate
Acne drugs: does use of Ab cause drug resistance?
Immunosuppressives used for psoriasis: ?incidence of lymphoma and other serious ADRs?
SJS: Multi-center and multi-national SCAR case-control study
Wound healing: predictors and new treatments
Genetic mechanisms for these effects