TREATMENT AND PREVENTION: CLINICAL AND PREVENTIVE TRIALS JEFFREY L. PROBSTFIELD, MD, FACP, FACC, FAHA, FESC, FSCT DIRECTOR, CLINICAL TRIALS SERVICE UNIT.

TREATMENT AND PREVENTION: CLINICAL AND PREVENTIVE TRIALS

JEFFREY L. PROBSTFIELD, MD, FACP, FACC, FAHA, FESC, FSCT

DIRECTOR, CLINICAL TRIALS SERVICE UNIT

PROFESSOR OF MEDICINE (CARDIOLOGY)

UNIVERSITY OF WASHINGTON SCHOOL OF MEDICINE

AD. PPROFESSOR OF EPIDEMIOLOGY

UNIVERSITY OF WASHINIGTON SCHOOL OF PUBLIC HEALTH AND COMMUNITY MEDICINE

SHORT COURSE: CLINICAL TRIALS OVERVIEW

• OVERIEW OF TRIALS METHODOLOGY

• BASIC ISSUES: CLINICAL TRIALS DESIGN

• RECRUITMENT AND ADHERENCE

• SUBGROUP ANALYSES

• EQUIVALENCE AND NON-INFERIORITY TRIALS, AND SURROGATE OUTCOMES

CLINICAL TRIALS OVERVIEW

FUNDAMENTAL POINT: DEFINITION

A properly planned and executed clinical trial is a powerful experimental technique for assessing the effectiveness of an intervention.

“Ask a good question, get a clear answer!”

not

“Ask a good question, get a positive/negative answer!”

Study Question

In high risk people with IFG, IGT or early diabetes, does omega 3 PUFA supplementation reduce the risk of:

Primary• Cardiovascular death?

Secondary• MI, stroke or CV death?• Mortality?• Presumed arrhythmic death or cardiac arrest?

Years of Follow-up

Pro

po

rtio

n w

ith

eve

nts

0.0

0.0

50

.10

0.1

50

.20

0.2

50

.30

0 1 2 3 4 5 6 7

n-3 fatty acids

Placebo

# at Risk 1 2 3 4 5 6 7

n-3

P

6281 6161 6034 5882 5706 5503 3896 879

6255 6143 6017 5848 5685 5492 3893 837

Time to Adjudicated CV Death

Primary Outcome: CV Death

HR 0.98; 95% CI, 0.87-1.10;P=0.72

Years of Follow-up

Pro

po

rtio

n w

ith

eve

nts

0.0

0.0

50

.10

0.1

50

.20

0.2

50

.30

0 1 2 3 4 5 6 7

n-3 fatty acids

Placebo

# at Risk 1 2 3 4 5 6 7

n-3

P

6281 6161 6034 5882 5706 5503 3896 879

6255 6143 6017 5848 5685 5492 3893 837

Time to Adjudicated All Death

HR 0.98; 95% CI, 0.89-1.07

Death

Fatal arrhythmia

Years of Follow-up

Pro

po

rtio

n w

ith

eve

nts

0.0

0.0

50

.10

0.1

50

.20

0.2

50

.30

0 1 2 3 4 5 6 7

n-3 fatty acids

Placebo

# at Risk 1 2 3 4 5 6 7

n-3

P

6281 6161 6034 5882 5706 5503 3896 879

6255 6143 6017 5848 5685 5492 3893 837

Time to Adjudicated Arrhythmic Death

HR 1.10; 95% CI, 0.93-1.30

INTRODUCTION

Fredrickson - “The Indispensable Ordeal”

Phases - I-IV

Why Needed

Problems in Timing

Ethics

Protocol

References:

Friedman, Furberg & DeMets - Fundamentals of Clinical Trials

Peto, R et al - Br. J Cancer, 1976; 34:585-612 and 1977; 35:1-39



FUNDAMENTAL POINT: WHAT IS THE QUESTION?

Each clinical trial must have a primary question. The primary question, as well as any secondary or subsidiary questions, should be carefully selected, clearly defined and stated in advance.


WHAT IS THE QUESTION?

Primary Question Intervention

Secondary Question(s) Response Variable

Adverse Effects Surrogate Outcomes

Ancillary Questions, Substudies

Natural History

Large, Simple Clinical Trials


FUNDAMENTAL POINT: STUDY POPULATION The study population should be defined in advance, starting unambiguous inclusion (eligibility) criteria. The impact that these criteria will have on the study design, ability to generalize and participant recruitment must be considered.


STUDY POPULATION Definition of Study Population: The study population is the subset of the Population with the condition or characteristics of interest defined by the eligibility criteria:

GeneralizabilityRecruitment

KEY INCLUSION/EXCLUSION CRITERIA (HOPE)

Inclusion CriteriaPatients (age 55) at high risk for cardiovascular

events because of:• any evidence of vascular disease (CHD, stroke,

PVD)

• diabetes + one other coronary risk factor

Exclusion CriteriaHeart failure or low EF

On ACE-I or Vitamin E

KEY HOPE BASELINE CHARACTERISTICS/DRUGS

Mean age 65.9

Female 26.7%

Any CAD 80.6%

MI 52.8%

PVD+AABP 43.4%

Stroke + TIA 10.8%

Any Diabetes 38.3%

Hypertension 46.5%

High Cholesterol 65.8%

ASA/antiplatelet 76.3%

Lipid Lowering 28.9%

Beta-Blockers 39.5%

Diuretics 15.1%

CCB 47.0%

Any anti-ischemicagents

74.4%


FUNDAMENTAL POINT: BASIC STUDY DESIGN

Sound scientific clinical investigation almost always demands that a control group be used against which the new intervention can be compared. Randomization is the preferred way of assigning participants to control and intervention groups.


BASIC STUDY DESIGN

• Randomized control studies• Nonrandomized concurrent control studies• Historical controls/databases• Cross-over designs• Withdrawal designs• Factorial designs• Hybrid designs• Studies of equivalency• Large simple clinical trials

CROSSOVER DESIGNS

• Each person serves as own control.

• Two period crossover.– 1/2 receive A first and 1/2 B.– Order of interventions excluded.

• Multiple Period crossover.– Order and sequence effects eliminated by design.

• Insertion of “Washout Period”.

FACTORIAL DESIGN

• Motive: to ask two or more questions in same trial in an efficient manner.

• Types:

– 2x2

– 2x2x2 etc.

– Incomplete/Partial

Ramipril + Rosiglitazone

DREAM: 2 x 2 factorial design-main effects analysis

DREAM Trial Investigators. Diabetologia. 2004;47:1519-27.

N = 5269 with IFG and/or IGT

Ramipril

Rosiglitazone Placebo

Ramipril + Placebo

PlaceboRosiglitazone +

PlaceboPlacebo +Placebo

FACTORIAL DESIGN

Indications• Interaction - small chance• Interaction - important• Interaction - must estimate

Contraindications• Design too complex - adherence• Power of simple comparison greater• Interaction - large change


FUNDAMENTAL POINT: THE RANDOMIZATION PROCESS

Randomization tends to produce study groups comparable with respect to known and unknown risk factors, removes investigator bias in the allocation of participants, and guarantees that statistical tests will have a valid significance levels.


THE RANDOMIZATION PROCESS

“Your most powerful ally - preserve at all costs”

• Fixed AllocationsSimpleBlockedStratified

• Adaptive RandomizationBaseline AdaptiveResponse Adaptive


FUNDAMENTAL POINT: BLINDNESS

A clinical trial, ideally, should have a double-blind design to avoid potential problems of bias during data collection and assessment . In studies where such a design is impossible, a single- blind approach and other measures to reduce potential bias are favored.


BLINDNESS

• Types of TrialsUnblindedSingle-blindDouble-blindTriple-blind

• Special ProblemsMatching of

drugsCoding of drugsUnblinding of

trialsAssessment of

the blind


FUNDAMENTAL POINT: SAMPLE SIZE

Clinical trials should have sufficient statistical power to detect differences between groups considered to be of clinical interest. Therefore calculation of sample size with provision for adequate levels of significance and power is an essential part of planning


SAMPLE SIZE

Power Level of Significance AdherenceEvent Rate Effect Size Variability

Dichotomous Response VariablesContinuous Response VariablesRepeated Measures“Time to Failure”Equivalency of InterventionsCluster RandomizationMultiple Response Variables

CLINICAL TRIAL RESULTS: TYPE 1 AND 2 ERRORS

TYPE 1 ERROR-FALSE POSITIVE-

‘YOU THINK IT IS THERE, BUT IT ISN’T.”

TYPE 2 ERROR-FALSE NEGATIVE-

“YOU THINK IT ISN’T THERE, BUT IT IS.”

ACE INHIBITORS, NOT THE SAME? HOPE vs. QUIET

HOPE• Effects of ACE Inhibition

(Ramipril 10 mg/day), CVD Morbidity and Mortality in 9297 High-risk CVD Participants

• Primary Outcome: MI/Stroke/CV Death

QUIET• Effects of ACE Inhibition

(Quinapril 20 mg/day) in 1750 Participants with CAD

• Primary Outcome: (MI/CV

Death/VT/VF

QUIET = Quinapril Ischemic Events TrialQUIET = Quinapril Ischemic Events Trial

Lees R, et al. Lees R, et al. Am J Cardiol.Am J Cardiol. 1996;78:1011-1016. 1996;78:1011-1016.The HOPE Study Investigators. The HOPE Study Investigators. N Engl J MedN Engl J Med. 2000;342:145-153.. 2000;342:145-153.

0

0.05

0.1

0.15

0.2

0 500 1000 1500

Days of Follow-up

Kap

lan

-Mei

er R

ates

Ramipril Placebo

Primary Outcome - Ramipril vs Placebo

RR=078(0.70-0.86) P=0.000002

Dagenais GR, et al. Lancet. 2006;368:581-88.





FUNDAMENTAL POINT: BASELINE ASSESSMENT

Relevant baseline data should be measured in all study participants before the start of intervention.


BASELINE ASSESSMENT

• Use of Baseline Comparability at baseline Stratification and subgrouping Evaluation of change Natural history analysis• What Is True Baseline Measurement?• Balance and Imbalance

FAR TOO MUCH DATA IS COLLECTED


FUNDAMENTAL POINT: Recruitment of Study Participants

Successful recruitment depends on developing a careful plan with multiple strategies maintaining flexibility, establishing interim goals, and preparing to devote the necessary effort.


RECRUITMENT OF STUDY PARTICIPANTS

PLAN, PLAN, PLAN, PLAN

Strategies and SourcesConductMonitoringProblemsSolutionsReasons for Participation

http://depts.washington.edu/cardweb/fellowship/educational-resources.shtml

RECRUITMENT:13 NHLBI STUDIESRECRUITMENT:13 NHLBI STUDIESNumber of Participants

RecruitedRecruitment

Time

Person-Years inPlanned Recruitment

PeriodStudy Actual Actual/Planned Actual/Planned Actual/Planned (R)

AMIS 4,524 1.06 1.00 0.83BHAT 3,837 0.95 1.21 0.82CAPS 502 1.00 1.08 0.82CARDIA 5,182 1.02 1.00 0.86CDP 8,345 1.00 1.22 0.55CSSCD 3,241 1.01 1.13† 1.16HDFP 10,940 1.04 1.50 1.02LRC 3,843 1.08 1.54 0.34MILIS 985 0.82 2.71 0.35MRFIT 12,886 1.07 1.13† 0.81POSCH 838 0.84 1.58† 0.25SHEP Pilot 551 1.10 1.17 0.71TIMI-1* 316 0.93 0.96 0.98*TIM-1 was stopped on the recommendation of Data and Safety MonitoringCommittee; the treatment showed strong evidence of efficacy.† The ‘projected’ time was revised after additional clinics joined the study.


FUNDAMENTAL POINT: DATA COLLECTION AND QUALITY CONTROL

During all phases of a study, sufficient effort should be spent to ensure that all key data critical to the interpretation of the trial are high quality.


DATA COLLECTION AND QUALITY CONTROL

Problems

DO WE CHECK ALL DATA---IF NOT, WHICH?

• Variability• Minimize Poor Quality

Protocol/Manual Training Reduce VariabilityForms Pretesting Data Entry

• Quality Control Monitoring-Source Documentation• Audits


FUNDAMENTAL POINT: ASSESSING AND REPORTING ADVERSE EFFECTS

Adequate attention needs to be given to the assessment, analysis and reporting of adverse effects to permit valid assessment of potential risk of interventions.


ASSESSING AND REPORTING ADVERSE EFFECTS

• Determinants of Adverse EffectsDefinitionAscertainmentFrequencyLength of Follow-upIndividual SusceptibilityAnalysis of Non-adherent Participants

• Reporting - FDA Format


FUNDAMENTAL POINT: ASSESSMENT OF HEALTH-RELATED QUALITY OF LIFE

Assessment of the effects of interventions of participants’ HRQL may be an important component of clinical trials, especially those that involve interventions directed at primary or secondary prevention of chronic disease.


ASSESSMENT OF HEALTH-RELATED QUALITY OF LIFE• Definitions• Primary HRQL Dimensions

Physical FunctioningPsychological FunctioningSocial FunctioningOverall Life Satisfaction/Well BeingPerceptions of Health Stages

•Additional DimensionsNeuropsychological FunctioningPersonal ProductivityIntimacy and Sexual FunctionsSleep DisturbancePainSymptoms


FUNDAMENTAL POINT: PARTICIPANT ADHERENCE

Many potential adherence problems can be prevented or minimized before participant enrollment. Once a participant is enrolled, taking measures to enhance and monitor participant adherence is essential.


PARTICIPANT ADHERENCE

“Prevention is the key issue”

Considerations before participant enrollmentMaintaining good participant adherenceSpecial InterventionsAdherence monitoringSpecial populations

EQUIPOSE AND THE ETHICS OF EQUIPOSE AND THE ETHICS OF CLINICAL RESEARCHCLINICAL RESEARCH

Benjamin Freedman, PhD

EQUIPOSE

“ - a state of genuine uncertainty on the part of the (all) clinical investigator(s) regarding the comparative therapeutic merits of each arm in a trial.”

NEJM 1987; 317:141-145

Key Point - Adherence correction term-sample size formula, a squared function.

2N = 2(z + z)2 (1 - 2)2(1-p)2

p = Reduction in Adherencek = Increase in Sample Size

SAMPLE SIZE ADJUSTMENT FOR SAMPLE SIZE ADJUSTMENT FOR REDUCED ADHERENCEREDUCED ADHERENCE

p k.01 1.02.05 1.11.10 1.23.20 1.56.30 2.04.50 4.00


FUNDAMENTAL POINT: SURVIVAL ANALYSIS

Survival analysis methods are important in trials where participants are entered over a period of time and have various lengths of follow-up. These methods permit the comparisons of the entire survival experience during the follow-up and may be used for the analysis of time to any dichotomous response variable such as a nonfatal event or an adverse effect.


SURVIVAL ANALYSIS

•Estimation of the Survival Curveeg. Kaplan - Meier

•Comparison of Two Survival CurvesPoint-by-point comparisonComparison of median survival timeeg. Mantel - Haenszel - Gehan

•Generalizations•Covariate Adjustment

0

0.05

0.1

0.15

0.2

0 500 1000 1500

Days of Follow-up

Kap

lan

-Mei

er R

ates

Ramipril Placebo

Primary Outcome - Ramipril vs Placebo

RR=078(0.70-0.86) P=0.000002


FUNDAMENTAL POINT: MONITORING RESPONSE VARIABLES

During the trial, response variables need to be monitored for early dramatic benefits or potential harmful effects. Preferably, monitoring should be done by a person or group independent of the investigators. Although many techniques are available to assist in monitoring, none of them should be used as the sole basis for the decision to stop or continue the trial.


MONITORING RESPONSE VARIABLES

Data and Safety Monitoring Committee (Board)

•Repeated Testing for Significance•Decisions for Early Termination•Statistical Methods Used in Monitoring

Classical Sequential MethodsGroup Sequential MethodsFlexible Group Sequential Methods

“Spending the Alpha”Boundaries - Setting them A-priori

Asymmetrical

MONITORING BOUNDARIIESOVER TIMERamipril versus Placebo

1.51

2.322.75

4.54.31

4.76

-1

0

1

2

3

4

5

Z

O

ct. 9

5

Aug

. 96

Nov

. 97

Nov

. 98

Mar

. 99

Jan.

94

Nov

. 99

Ben

efit


FUNDAMENTAL POINT: ISSUES IN DATA ANALYSIS

Excluding randomized participants or observed outcomes from analysis and sub-grouping on the basis of outcome or response variables can lead to biased results of unknown magnitude or direction.


ISSUES IN DATA ANALYSIS

•Which Participants Should Be Analyzed?Intention to treatBy treatment analysis - Non-

adherence•Withdrawals•Ineligibility•Poor quality or missing data•Competing events •Covariate adjustment - Baseline, surrogates•Subgroup analysis•Not counting some outcomes•Comparison and multiple variables•“Cutpoints” categorical analysis•Meta-analysis of multiple studies


FUNDAMENTAL POINT: CLOSEOUT

The closeout of a clinical trial is usually a fairly complex process that requires careful planning, if it is to be accomplished in an orderly fashion.


Termination ProceduresPost Study Follow-upData Clean-up and VerificationStorage of Study MaterialDissemination of Results

CLOSEOUT

“You must account for every participant’s primary outcome status”

SYST - EURSYST - EURA WORST CASE ANALYSIS?A WORST CASE ANALYSIS?

LOST TO N STROKE CHD TOTAL FOLLOW-UP SYST-EUR 4695 124 (250) 131 255 237 Placebo 2297 77 73 150 116 Active 2398 47 58 105 121 SHEP 4736 262 (270) 245 507 10 Placebo 2371 159 141 300 5 Active 2365 103 104 207 5


FUNDAMENTAL POINT: REPORTING AND INTERPRETING OF RESULTS

The investigators have an obligation to review their study and its findings critically and to present sufficient information so that readers can properly evaluate the trial.

Potential Annual Global Impact Ramipril: Assume HOPE Results

• approximately 1 to 1.5 million deaths, myocardial infarction, stroke or revascularization procedures will be prevented globally every year

• Plus impact on CHF, diabetic complications and prevention of diabetes, which will prevent an additional 0.5 to 0.6 million such events/year

If 1/4 of eligible patients with vascular disease in developing countries and 1/2 in developed countries are given Ramipril:

Total benefit of about 2 million events prevented

TREATMENT AND PREVENTION: CLINICAL AND PREVENTIVE TRIALS JEFFREY L. PROBSTFIELD, MD, FACP, FACC, FAHA, FESC, FSCT DIRECTOR, CLINICAL TRIALS SERVICE UNIT.

Documents

clinical trials overview

simple clinical trials

clinical trials peto

death slide

study question

surrogate outcomes slide

vitamin e slide

noninferiority trials