understanding clinical trials - quoetr … clinical trials. ... • natural history of ms and number...

Understanding Cl in ical Tr ia ls

Introduction and Objectives

Introduction

• This module provides learners with an educational resource that focuses on

understanding both why and how clinical trials are conducted as well as the

importance of clinical trial results

Objectives

• Understand basic concepts of the drug approval process

• Describe the types and variations of clinical trials

• Review outcome measures in multiple sclerosis (MS) clinical trials

• Understand statistical measures

• Understand challenges involved in designing and conducting MS clinical trials

2.

Table of Contents

Chapter 1: Regulatory Pathway

• Food and Drug Administration (FDA) Approval in the United States

• Drug Approval in the European Union

Chapter 2: Types of Clinical Trials

• Overview

• Trial Objectives

Chapter 3: Trial Design and Terminology

• Clinical Trial Planning

• Randomization

• Blinding

• Crossover

• Inclusion/Exclusion Criteria, Baseline Data, and Adverse Effects

Chapter 4: Outcome Measures in MS Trials

• Annualized Relapse Rate (ARR)

• Expanded Disability Status Scale (EDSS)

• Multiple Sclerosis Functional Composite (MSFC)

• Magnetic Resonance Imaging (MRI)

• Quality of Life (QOL) Measures

Chapter 5: Statistical Measures Utilized in MS Clinical Trials

• Statistical Significance

• Statistical Power

• Relative Risk (RR)

• Odds Ratio (OR)

• Survival Analysis, Hazard Ratio (HR), and Confidence Interval

• Intention to Treat

• Mean Compared to Median

• Standard Deviation

Chapter 6: Factors Affecting the Validity and Relevance of

MS Clinical Trial Results

• Short Duration of Trials

• Duration of MS

• Randomization

• Seasonality

• Attrition and Lack of Adherence

• Natural History of MS and Number Needed to Treat (NNT)

• Bias

• Outcome Measures

Summary. 3

© iStock.com/adiekoetter

Chapter 1

Regulatory Pathway

• Drug discovery

• Preclinical testing

• Submit an investigational new

drug application (IND) to FDA

Drug Approval Process

Preclinical1

• Phase I studies

• Phase II studies

• Phase III studies

• Submit new drug application (NDA) to FDA

• Initial NDA review (FDA decides whether to

accept NDA for review)

• Application reviewed

• Receives approval or a complete

response letter

5

Clinical Review1,2 NDA1-3

Imagery provided by iStock

1. PhRMA. http://www.phrma.org/innovation/clinical-trials#sthash.xfkGHNsb.dpuf. Accessed [March 1, 2017].

2. US Food and Drug Administration. http://www.fda.gov/Drugs/ResourcesForYou/Consumers/ucm289601.htm. Accessed [March 1, 2017].

3. Randall B. The U.S. Drug Approval Process: A Primer. Washington, DC: Library of Congress, Congressional Research Service; 2001.

Therapeutic biologic application

(also known as biologic license

application [BLA])1,2

• Required by the FDA for therapeutic

biologic products

• Approval processes for BLAs and NDAs

are essentially the same

• Reviewed by the FDA’s Center for Drug

Evaluation and Research (CDER) and

Center for Biologics Evaluation and

Research (CBER)

Therapeutic Biologic Application

6

Traditional Drug

Treatments2

Biologic

Treatments2

Chemically synthesized Derived from living material

Smaller, well-defined structure Complex structure

NDA BLA

1. US Food and Drug Administration. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/ApprovalApplications/

TherapeuticBiologicApplications/default.htm. Accessed [March 1, 2017].

2. US Food and Drug Administration. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/ApprovalApplications/

TherapeuticBiologicApplications/ucm113522.htm. Accessed [March 1, 2017].

Therapeutic Biologics

Categories for therapeutic biologics1

• Monoclonal antibodies for in vivo use

• Cytokines, growth factors, enzymes, immunomodulators,

and thrombolytics

• Proteins intended for therapeutic use that are extracted

from animals or microorganisms, including recombinant

versions of these products (except clotting factors)

• Other non-vaccine therapeutic immunotherapies

7

1. US Food and Drug Administration. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/ApprovalApplications/TherapeuticBiologicApplications/ucm113522.htm.

Accessed [March 16, 2017].

2. National Multiple Sclerosis Society. http://www.nationalmssociety.org/NationalMSSociety/media/MSNationalFiles/Brochures/Brochure-The-MS-Disease-Modifying-Medications.pdf. Accessed [March 1, 2017].

MS disease-modifying products

considered by FDA to be biologics2

Avonex® (interferon beta-1a)

Betaseron® (interferon beta-1b)

Copaxone® (glatiramer acetate)

Extavia® (interferon beta-1b)

Glatopa® (glatiramer acetate)

Lemtrada® (alemtuzumab)

Plegridy® (peginterferon beta-1a)

Rebif® (interferon beta-1a)

Tysabri® (natalizumab)

Zinbryta® (daclizumab)

Drug approval process

• Similar to drug approval process in the United States

o Begins with a clinical trial process followed by approval to market

and sell the drug

• Two drug approval procedures

1. Centralized1

• Commission approval of a new drug allows a pharmaceutical company

to market its drug in all Member States without having to obtain separate

approval in each Member State

2. Decentralized1,2

• If a product does not meet the requirements for consideration under the

centralized process, the application may be submitted under the

decentralized process

• Allows for simultaneous consideration by Member States that have not

yet approved the product

Drug Approval in the European Union

8

1. European Medicines Agency. http://www.ema.europa.eu/ema/index.jsp?curl=pages/about_us/general/general_content_000109.jsp&mid=WC0b01ac0580028a47&jsenabled=true. Accessed [March 1, 2017].

2. European Medicines Agency. http://www.ema.europa.eu/ema/index.jsp?curl=pages/about_us/general/general_content_000310.jsp&mid=WC0b01ac058074758f. Accessed [March 1, 2017].

3. European Union. http://europa.eu/european-union/about-eu/countries/member-countries_en. Accessed [March 1, 2017].

Member States of

the European Union3

Austria Germany Norway

Belgium Greece Poland

Bulgaria Hungary Portugal

Croatia Ireland Romania

Cyprus Italy Slovakia

Czech Republic Latvia Slovenia

Denmark Lithuania Spain

Estonia Luxemburg Sweden

Finland Malta United Kingdom

France Netherlands

The centralized process is mandatory for all:

• Human medicines for the treatment of HIV/AIDS, cancer,

diabetes, neurodegenerative diseases, autoimmune and

other immune dysfunctions, and viral diseases

• Veterinary medicines for use as growth or yield enhancers

• Medicines derived from biotechnology processes, such

as genetic engineering

• Advanced-therapy medicines, such as gene therapy, somatic

cell-therapy, or tissue-engineered medicines

• Officially designated “orphan medicines” (medicines used for

rare human diseases)

Centralized Process

9

HIV, human immunodeficiency virus.

European Medicines Agency. http://www.ema.europa.eu/ema/index.jsp?curl=pages/about_us/general/general_content_000109.jsp&mid=WC0b01ac0580028a47&jsenabled=true. Accessed [March 1, 2017].

© iStock.com/Oria

Chapter 2

Types of

Clinical Trials

© iStock.com/RTimages

Clinical trials1,2

• “Investigational” trials or experiments conducted in humans during which researchers will test the effects of

a treatment or prevention strategy on pre-specified outcomes

• Determines a treatment’s safety and/or efficacy based on pre-specified outcomes

• Categorized as:

Clinical Trials

11

Retrospective3

• Investigators look at medical records to

determine what happened to 2 or more

groups in the past

Interventional2

• Participants take a drug or undergo therapy

• Investigators record results and events

that occur

Observational2

• Investigators assess health outcomes

and record these observations in the

absence of investigator-induced

intervention

Prospective3

• Investigators follow patients into the

future for various periods

1. Appel LJ. Clin J Am Soc Nephrol. 2006;1(6):1360-1367.

2. ClinicalTrials.gov. https://clinicaltrials.gov/ct2/about-studies/learn#WhatIs. Accessed [March 9, 2017].

3. Boston University School of Public Health. http://sphweb.bumc.bu.edu/otlt/MPH-Modules/EP/EP713_AnalyticOverview/EP713_AnalyticOverview3.html. Accessed [March 9, 2017]. Imagery supplied by iStock

Types of Clinical Trials

12

US Food and Drug Administration. https://www.fda.gov/ForPatients/ClinicalTrials/Types/default.htm. Accessed [March 15, 2017].

Treatment trialsTest experimental treatments, new combinations of drugs, new devices,

or new approaches to surgery or radiation therapy

Prevention trialsSeek better ways to prevent disease in people who have never had the

disease or to prevent a disease from reoccurring

Diagnostic trialsConducted to find better tests or procedures for diagnosing a particular

disease or condition

Screening trials Test the best ways to detect certain diseases or health conditions

Quality of life trials

(or supportive care trials)

Explore ways to improve comfort and quality of life for individuals with

a chronic illness

Genetic studiesImprove prediction of disorders by identifying and understanding the

relationship between genes and illnesses

Epidemiology studies Identify patterns, causes, and control of disorders in groups of people

Benefits

• Provide the strongest evidence supporting cause-effect relationships

• Form the basis for clinical practice and public health policy

• Minimize or eliminate bias and confounding factors

Challenges

• Expensive and time-consuming

• Ethical concerns

• Difficult to recruit patients into study

• Logistically complex

• Many research questions cannot be answered

due to study design restrictions

Benefits and Challenges of Clinical Trials

13

Appel LJ. Clin J Am Soc Nephrol. 2006;1(6):1360-1367. © iStock.com/Terminator3D

Placebo1,2

• Substance with no known treatment value that is given to a study participant

• Made to appear, smell, and/or taste similar to the actual treatment

Placebo-controlled trial• Determines whether a drug of interest is better than a placebo3

• Blinding so that study participants and investigators do not know

who is receiving the active treatment and who is receiving the placebo2

Superiority trial3

• Tests whether a new treatment is superior to the existing treatment,

which may be the current standard of care

• May include placebo-controlled efficacy trials, because the new treatment

is being tested for superiority to placebo

Placebo-Controlled and Superiority Trials

14

1. ClinicalTrials.gov. http://clinicaltrials.gov/ct2/info/glossary#inclusion. Accessed [March 14, 2017].

2. Gordis LM. Assessing the efficacy of preventive and therapeutic measures: randomized trials. In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009:chap 7.

3. Appel LJ. Clin J Am Soc Nephrol. 2006;1(6):1360-1367. Imagery supplied by iStock

Non-inferiority Trials

Non-inferiority trial1,2

• Shows that an experimental treatment is not worse than an active control by

more than the equivalence margin

• If non-inferiority is established, a treatment may be preferred based on other

considerations such as easier to use, fewer side effects, less costs, etc2,3

15

1. Snapinn SM. Curr Control Trials Cardiovasc Med. 2000;1(1):19-21.


3. Greene GJ. J Trauma Stress. 2008;21(5):433-439.

Dose Comparison or Dose-Ranging Trial

Method

• Used to determine a reasonable initial dose

• The initial dose is the lowest dose tested that has a

response that is statistically greater than the response

after a placebo test

• Placebo periods may follow the active drug period in

trials designed with more than one active drug period

16

Sheiner LB, et al. Clin Pharmacol Ther. 1989;46(1):63-77. © iStock.com/Jezperklauzen

Method1

• Tests the effects of using 2 or more drugs in combination to determine whether the drugs may work

together and be more effective, or more toxic, than using either drug as monotherapy

Three variations of a combination trial2

1. Combination therapy

• When 2 or more drugs (or a drug plus a immunomodulator or antibody) are administered

together to a patient who had not received any part of the combination previously

2. Add-on therapy

• When 1 drug is administered for some period and then a second drug

is added to the regimen

3. Sequential therapy

• When the first drug that is administered is discontinued before initiation of

therapy with a second agent

Combination Trial

17

1. Miles D, et al. Oncologist. 2002;7(suppl 6):13-19.

2. Powers JH. Clin Infect Dis. 2004;39(suppl 4):S228-S235. © iStock.com/ozgurdonmaz

Method1

• Simultaneously measure exposure and disease status in a given population

• Cannot be used to determine the following

o Whether the exposure preceded or followed the disease

o Cause-and-effect relationship

Advantages2

• Results can be highly generalizable, if the sample is representative of the population of interest

• Inexpensive

• Can be completed quickly

Disadvantages1

• Can only identify patterns or trends in disease occurrence, but cannot differentiate between an

association and a causal effect

Cross-sectional or Prevalence Study

18

1. Gordis LM. Case-control studies and other study designs. In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009:chap 10.

2. Aschengrau A, Seage GR III. Overview of epidemiologic study designs. In: Essentials of Epidemiology in Public Health. 2nd ed. Sudbury, MA: Jones and Bartlett Publishers; 2008:135-162.

Method1,2

• Observational study that does not involve an intervention or an attempt to alter the disease course

• Compares patients who have a disease or outcome of interest (cases) with patients who do not have the disease or outcome (controls)

• Retrospective analysis to compare how frequently a specific exposure

to a risk factor is present in each group

Advantages1

• Good for studying rare conditions or diseases

• Less time is needed to conduct the study

• Allows researchers to simultaneously look at multiple risk factors

• Useful as initial study to establish an association

Disadvantages

• Difficult to find a suitable control group1

• Greater chance of bias than experimental studies2

• Difficult to establish temporal relationship between exposure and disease2

Case-Control or Retrospective Study

19

1. Gordis LM. Case-control studies and other study designs. In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009:chap 10.

2. Aschengrau A, Seage GR III. Overview of epidemiologic study designs. In: Essentials of Epidemiology in Public Health. Sudbury, MA: Jones and Bartlett Publishers; 2008:135-162.

Diagram of a Case-Control Study

Study individuals

Selected cases Selected controls

Exposed Not exposed Exposed Not exposed

Method1

• Follows prospectively over time one or more populations (or cohorts)

• Determines which patient characteristics or risk factors are associated with the development of a disease or outcome

Advantages2,3

• Easier and less expensive than a randomized controlled trial

• Best way to ascertain both the incidence and natural history of a disorder

• Time sequence between putative cause and outcome is usually clear

• Other outcomes, besides the pre-determined outcomes, are discovered

• May be useful in the study of rare exposures

Disadvantages1,2

• Cohorts can be difficult to identify due to confounding variables

• No randomization, which means an imbalance in patient characteristics could exist

• Blinding or masking is difficult

• Outcome of interest could take time to occur

Cohort, Prospective, or Longitudinal Study

20

1. Gordis LM. Cohort studies. In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009:chap 9.

2. Himmelfarb Health Sciences Library. https://himmelfarb.gwu.edu/tutorials/studydesign101/cohorts.html. Accessed [March 30, 2017].

3. Grimes D, Shulz KF. Lancet. 2002;359(9303):341-345.

Diagram of a Prospective Cohort Study

Year: 202520152006

Diseased

Non-

diseased

Diseased

Non-

diseased

Risk

factor(s)

exposed

Risk

factor(s)

non-exposed

Population

at risk

Diseased

Non-

diseased

Nested Case-Control Study

Method

• Hybrids of retrospective

and prospective analyses

• Participants are often selected from

ongoing cohort studies

• Compares the course of individuals

who develop the disease (case) with

those who do not develop the disease

but are at risk of developing the

disease (control)

21

Diagram of a Nested Case-Control Study

Gordis LM. Cohort studies. In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009:chap 10.

Do not develop

disease

(controls)

Compare case and control

groups during follow-up

Develop

disease

(cases)

Select population and

collect baseline data

Method

• Uses statistical methods to combine results of individual studies that are focused on related research

Advantages

• Makes it possible to see the results of several studies at one time

• Can see treatment effects in a more diverse population with a larger number of people

• Higher statistical power than individual studies

Disadvantages or weaknesses

• Difficulty in selecting which studies are included or excluded in a meta-analysis

• Not all studies may have subject characteristics and data suitable for inclusion and analysis

• Requires advanced statistical techniques

Meta-analysis

22

Himmelfarb Health Sciences Library. http://himmelfarb.gwu.edu/tutorials/studydesign101/metaanalyses.html. Accessed [March 10, 2017].

Chapter 3

Trial Design and Terminology

© iStock.com/Maridav

Goals of well-designed clinical trials

• Differences observed between different

intervention groups may be attributed to the

treatment under investigation1

• Generalizable results1

o Results are applicable to real people outside

of the study

• Fair to the participants2

• Ethical to the participants2

Clinical Trial Planning

24

1. Jüni P, et al. BMJ. 2001;323(7303):42-46.

2. National Institutes of Health Clinical Center. https://clinicalcenter.nih.gov/recruit/ethics.html. Accessed [March 15, 2017]. © iStock.com/goir

Why are participants randomized?

• To avoid bias on the part of investigators1

o Bias consists of any effect or interference that can affect a clinical trial’s results2

• To facilitate blinding of study investigators and participants1

Goals of randomization1,3

• To produce comparable groups in terms of general participant characteristics

o Characteristics may include age, gender, duration of disease, and disability status

• To conclude that study results are valid and cannot be attributed to assignment arm bias

Randomization

25

Randomization occurs when trial participants are

randomly assigned to a particular treatment arm of a trial,

rather than someone deciding which arm they should be in1


2. Taber’s Online. https://www.tabers.com/tabersonline/view/Tabers-Dictionary/730664/0/bias?q=bias&ti=0. Accessed [March 15, 2017].


© iStock.com/zentilia

© iStock.com/Kai Wong

Blinding prevents1

• Response bias from patientso The tendency for patients to report symptoms differently, based on the treatment they

are getting

• Behavior changes in patients that will affect the trialo Attrition

o Secretly adding in medication or other therapies if the patient thinks they are

receiving placebo

• Bias on behalf of the people collecting and reporting the data

How is blinding done?1,2

• By randomization

• By making all the interventions received by each arm look identical and be

dosed identically

Blinding

26

Blinding means that investigators and/or participants do not

know which treatment arm a participant is in1

1. Appel LJ. Clin J Am Soc Nephrol. 2006;1(6):1360-1367

2. Gordis LM. Assessing the efficacy of preventive and therapeutic measures: randomized trials. In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009: chap 7.

Single-blinded1,2

• Either the investigator or the participant is unaware of which medication the

participant is taking or which intervention they have been exposed to

Double-blinded1,3

• Neither the investigator nor the participant knows which drugs are being taken

• Participants are randomized to one of two or more study groups

• All medications and procedures that are given are designed to look identical

Double-dummy4

• Each group of participants receives one of the active interventions and a placebo

(or dummy) that looks and tastes the same as the other intervention

• Used to compare interventions that are administered by different routes or on

different schedules

Types of Blinding

27

Double-blind trials

are considered the

most objective type of

study and are known

as the “gold standard”

of drug trials5




4. Friedman LM, et al. Protecting the double-blind design. In: Fundamentals of Clinical Trials. 4th ed. New York, NY: Springer; 2010:123-141.

5. US Food and Drug Administration. https://www.fda.gov/AboutFDA/WhatWeDo/History/Overviews/ucm304485.htm#_edn39. Accessed [March 13, 2017].

Single doseSingle dose

Washout

Washout

Visit

Drug A Drug A

Drug B Drug B

RInformed

consentEligibility

1 2 3 4 5

Method1,2

• Allows each patient to act as their own control

• Ensures that patients in different arms of a study are similar enough

to compare results

• May be used to compare the effects of two drugs to one another

• Most appropriate for short-term studies

Two main issues regarding crossover trials1,2

1. Order effects

o When the order in which the treatments

are given impact the outcome

2. Carryover effects

o When beneficial effects and/or negative side

effects carry over from the first treatment

o May be minimized by adequate washout periods

Crossover Study

28

Patients participate in each arm of the

study, usually in random order

Diagram of a Crossover Study


2. Mills EJ, et al. Trials. 2009;10:27. doi: 10.1186/1745-6215-10-27.

Inclusion/exclusion criteria1

• Characteristics of a person or illness that researchers use to determine who is eligible

to participate

Baseline data1

• Patient information collected before randomization

Adverse effects2

• Any negative changes in health or side effects that occur in a person who participates in a clinical

trial while the patient is receiving a study drug (active drug or placebo) or within a certain

(specified) amount of time after the trial ends

Relevant Terms

29



Chapter 4

Outcome Measures

in MS Trials

Disadvantages

• Difficult to compare across studies, as different researchers adopt different methods and

definitions to confirm relapses1

Annualized Relapse Rate

31

Total number of relapses experienced by the group (the treatment or

placebo arm) divided by the person/years at risk1

The most common summary measure of relapses2

1. Inusah S, et al. Mult Scler. 2010;16(12):1414-1421.

2. Lavery AM, et al. Mult Scler Int. 2014;2014:262350. doi: 10.1155/2014/262350.

EDSS1

• Rating system that is used frequently for classifying and standardizing the physical

condition of people with MS in clinical trials

• It is possible to move back and forth on the scale2

• A composite scale based on multiple outcome measures, not on an individual measure1

Calculating the EDSS score3

• Based on neurological testing and examination of 8 functional systems

Advantages3

• Familiar system for researchers

• Researchers and physicians understand the significance of

changes in EDSS

Disadvantages1

• Places emphasis on the ability to walk, when other symptoms

could be impacting a person’s ability to function

• Insensitive to cognitive dysfunction

• Moderate inter-rater reliability

• Not a linear scale

Expanded Disability Status Scale

32

1. Krieger S. Issues in the design and interpretation of multiple sclerosis clinical trials. In: Giesser B, ed. Primer on Multiple Sclerosis. New York, NY: Oxford University Press; 2011:435-450.

2. Kurtzke JF. Neurology. 1983;33(11):1444-1452.

3. National Multiple Sclerosis Society. http://www.nationalmssociety.org/For-Professionals/Researchers/Resources-for-Researchers/Clinical-Study-Measures/Functional-Systems-Scores-%28FSS%29-and-

Expanded-Disab. Accessed [March 13, 2017].

EDSS of Disability Progression

EDSS remains

the gold

standard in

clinical trials for

measuring

physical

impairment

from MS1

A multidimensional clinical outcome measure consisting of 3 components

Advantages

• Dimensions change relatively independently

over time

• Measures are quantitative and standardized

• Easy to administer by a non-medical person

• Inter-rater and test-retest reliability are good

Disadvantages

• Many researchers are unfamiliar with

the MSFC

• Clinical meaning of a 1-point change on the

scale is not as clear as a change on the EDSS

• “Practice effect” can influence test results—

scores tend to improve naturally after

individuals take the tests on separate

occasions

Multiple Sclerosis Functional Composite

33

Fischer JS, et al. Mult Scler. 1999;5(4):244-250.

1

2

3

Timed 25-foot walk

• Measure of leg function and

ambulation

9-hole peg test

• Measure of upper extremity

(arm and hand) function

Paced auditory serial addition test (PASAT)

• Measure of cognitive function that

specifically assesses auditory information-

processing speed and flexibility

Imagery supplied by iStock

Results of MRI scans1

• Provide a measure of disease activity in relapsing-remitting MS

• Show lesions indicative of breakdown of the blood-brain barrier and total

lesion burden

• MRI measures have been used from pilot studies to phase III studies

Advantages2

• More sensitive in identifying disease activity than measures of relapse

• Easily blinded

• Objective and less prone to reporting bias on the part of the patient or

bias on the part of the researchers

• Reliability and reproducibility are generally very good

Disadvantages1

• Disease activity as seen on MRI scans does not correlate exactly with

clinical signs or with disability

Magnetic Resonance Imaging

34

No disease-

modifying agent for

MS has MRI results

data in its indication

1. Daumer M, et al. Neurology. 2009;72(8):705-711.

2. Martinelli Boneschi F, et al. Mult Scler. 2004;10(4):341-347.

© iS

tock.c

om

/denis

ovd

Quality of life scales

• Are included in and required by the FDA for some trials

Health status questionnaire Short Form-36 (SF-36)

• Generic QOL scale covering 8 areas of health

• Used to compare QOL across different diseases

QOL scales for MS

• Multiple Sclerosis Quality of Life-54 (MSQOL-54)

o 54-item scale that incorporates elements of the SF-36

• Multiple Sclerosis Quality of Life Inventory (MSQLI)

o Covers some areas in more detail than the MSQOL-54

o Includes subscales that study particular aspects of MS

Quality of Life Measures

35

Herndon R. Outcome measures in multiple sclerosis. In: Giesser BS, ed. Primer on Multiple Sclerosis. New York, NY: Oxford University Press; 2011:149-155.

Statistical significance1,2

• The percent likelihood that a result obtained happened

by chance alone

• Expressed as a probability value (P value)

• Determined by the magnitude of the result and the

sample size

• Does not necessarily suggest clinical significance

• A common cutoff for P values is 0.05

• When there are multiple endpoints, this P value

may be divided among endpoints

• However, in some situations, endpoints are

deemed “exploratory” and cannot be used to

assess statistical significance

Statistical Significance and Statistical Power

37

Statistical power3

• Probability of a statistically significant result

• Increasing the probability of a statistically significant

result is accomplished by increasing the sample size

0.05

P value2

P value >0.05

Not statistically significant

P value ≤0.05

Statistically significant


2. Gordis LM. Randomized trials: some further issues. In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009: chap 8.

3. Kirkwood BR, Sterne JA. Calculation of required sample size. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:413-438.

Relative Risk or Risk Ratio

38

Relative Risk Meaning

1There is no difference in risk between the

2 groups

<1The event is less likely to occur in the

experimental group than in the control group

>1The event is more likely to occur in the

experimental group than in the control group

Gordis LM. Estimating risk: is there an association? In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009:chap 11.

Example

Relative Risk =

incidence of disease in exposed population

incidence of disease in non-exposed population

A cohort study reported that 280/1000 people

who chewed gum every day for at least

10 years developed chronic jaw pain, and

48/1000 people who didn’t chew gum

developed chronic jaw pain.

RR =

Incidence of chronic jaw pain in daily gum chewers

=

280

= 5.83Incidence of chronic jaw pain in non-gum chewers

48

This would translate to findings of “people

who chew gum every day for at least 10 years

are 5.83 times more likely to develop chronic

jaw pain than those who do not chew gum.”

© iStock.com/Hogie

Determine if using a computer keyboard for more than 6 hours per day contributes

to hand pain. A case-control study compared 100 people with hand pain to

100 people without hand pain. We find out that 60 of the people with hand pain use a

computer keyboard for more than 6 hours per day, while 20 people with hand pain

use a computer keyboard for less than 6 hours per day. Here is the OR calculation:

Example2

Odds Ratio

39

1. Gordis LM. Estimating risk: is there an association? In: Epidemiology. 4th ed. Philadelphia, PA: Saunders Elsevier; 2009:chap 11.

2. Kirkwood BR, Sterne JA. Comparing two proportions. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:148-164.

Odds Ratio1 =

Odds that an exposed person

develops disease

Odds that a non-exposed person

develops disease

Odds of hand pain in people using a

computer keyboard for more than

6 hours per day:

60%

=

60%

= 1.5(100%-60%) 40%

Odds of hand pain in people not using a

computer keyboard for more than

6 hours per day:

20%

=

20%

= 0.25(100%-20%) 80%

OR =

1.5= 6 =

The likelihood of hand pain in people using a computer

keyboard for more than 6 hours per day is 6 times higher

than in people not using a computer keyboard for more

than 6 hours per day0.25

Survival analysis1

• Statistical method in which individuals are followed until the occurrence of a predetermined event

• Described in 2 different methods

1. Life tables

• Used when researchers do not know the exact survival time of each individual study participant but do know the total number of study participants who have survived

at a succession of time points

2. Kaplan-Meier estimate of survival curve

• Curve is based on the exact time in the study that each individual study participant reaches the survival end point

• Primarily used to compare the survival patterns of different groups

Hazard ratio2

• Measure of how often a particular event happens in one group compared with how often it happens in another group, over time

• HR = 1 means there is no difference in survival between 2 groups

• HR >1 or HR <1 means that survival was better in one of the groups

95% Confidence interval3

• Range of numbers that represents the upper and lower 95% confidence limits for reported data

Survival Analysis, Hazard Ratio, and Confidence Interval

40

1. Kirkwood BR, Sterne JA. Survival analysis: displaying and comparing survival patterns. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:272-286.

2. Taber’s Online. http://www.tabers.com/tabersonline/view/Tabers-Dictionary/750370/0/hazard_ratio. Accessed [March 25, 2017].

3. Kirkwood BR, Sterne JA. Confidence interval for a mean. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:50-57.

Intention to Treat

Method1,2

• Preferred for the analysis of clinical trials

• Individuals are analyzed according to randomized assignment, independent of whether they remain on assigned therapy

• Provides an unbiased assessment of the efficacy of the intervention at the level of adherence that occurred in the trial

Figure is an example of modified intention to treat3

• “Modified” because the study only included patients who received at least 1 dose of study drug

41

Strategy for the analysis of randomized controlled trials that compares

all patients in the treatment arms to which they were originally randomly assigned1


2. Montori VM, Guyatt GH. CMAJ. 2001;165(10):1339-1341.

3. Cohen JA, et al. N Engl J Med. 2010;362(5):402-415.

431 were included in the modified

intention-to-treat and

safety populations

281 were excluded

6 were not treated 2 were not treated 4 were not treated



safety populations



safety populations

426 were assigned to receive

fingolimod, 1.25 mg


fingolimod, 0.5 mg


interferon beta-1a

1292 underwent randomization

1573 patients were assessed for eligibility

Mean1,2

• Mathematical average of a group of numbers

• Typically used for normal distributions

Mean Compared With Median

42

2 + 3 + 3 + 4 + 7 + 9 + 11 + 12 + 14 + 17 + 18 = 100/11 = 9.1

Adding the numbers up, you get a total of 100.

Divide this by 11 and you get a mean of 9.1 (median = 9)

2 + 3 + 3 + 4 + 7 + 9 + 11 + 12 + 14 + 17 + 86 = 168/11 = 15.3

Adding the numbers up, you get a total of 168. Divide this by 11

and you get a mean of 15.3 (median = 9)

2 3 3 4 7 9 11 12 14 17 18

To find the median, choose the middle number (the 6th one in

the list with 5 numbers on each side) and you get a median of 9

2 3 3 4 7 9 11 11 12 14 17 18

To find the median, choose the middle 2 numbers (the 6th and 7th

ones in the list with 5 numbers on each side), add them and

divide by 2, and you get a median of 10 (9 + 11 = 20 ÷ 2 = 10)

1. Kirkwood BR, Sterne JA. Displaying the data. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:15-30.

2. Kirkwood BR, Sterne JA. Means, standard deviations and standard errors. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:33-41.

Median1,2

• The middle number of a data set when the set

is sorted in numerical order

• Used for skewed distributions

Mean example (set of 11 numbers) Median example (odd-numbered data set)

Mean outlier example (set of 11 numbers) Median example (even-numbered data set)

Standard deviation1,2

• Shows how much variance is in the data or

how different the points are from the mean

• For a normal distribution, nearly all values

lie within 3 standard deviations of the mean

Figure of a normal curve

and standard deviations1-3

• Dark blue is +/- 1 standard deviation

from the mean, which is approximately

68% of the set for a normal distribution

• Medium and dark blue are +/- 2 standard deviations from the mean, which is approximately


• Light, medium, and dark blue are +/- 3 standard deviations from the mean, which is approximately


Standard Deviation

43

1. Kirkwood BR, Sterne JA. Means, standard deviations and standard errors. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:33-41.

2. Kirkwood BR, Sterne JA. The normal distribution. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:42-49.

3. Encyclopedia of Mathematics. http://www.encyclopediaofmath.org/index.php?title=Three-sigma_rule&oldid=17366. Revised February 7, 2011. Accessed [March 25, 2017].

Chapter 6

Factors Affecting the

Validity and Relevance of

MS Clinical Trial Results


2. Inusah S, et al. Mult Scler. 2010;16(12):1414-1421.


4. Salvi F, et al. BMC Neurol. 2010;10(105):1-6. doi:10.1186/1471-2377-10-105.

5. Mohr DC, et al. Mult Scler. 1999;5(3):192-197.

6. Kirkwood BR, Sterne JA. Calculation of required sample size. In: Essential Medical Statistics. 2nd ed. Malden, MA: Blackwell Science; 2003:413-428.

Short duration of trials1

• 1 to 2 years’ duration for disease-modifying

agents

• Makes it challenging to detect effectiveness

in prevention of long-term disability

Duration of MS2

• Frequency of relapse tends to diminish the

longer a person has had the diagnosis of MS

o Leads to difficulty interpreting ARR or

other measures of relapse

Randomization3

• Treatment arms not comparable in baseline

characteristics may affect study results

Factors Affecting Validity and Relevance of MS Trial Results

45

Seasonality4

• May impact MS symptoms, relapses,

and the appearance of new lesions

found in serial T2-weighted MRI

Attrition and lack of adherence5,6

• May affect statistical power

Factors Affecting Validity and Relevance of MS Trial Results (cont)

46

Natural history classifications

• 5 classifications of disease course

1. Clinically isolated syndrome (CIS)

2. Relapsing-remitting MS (RRMS)

3. Relapsing-progressive MS (RPMS)

4. Secondary progressive MS (SPMS)

5. Primary progressive MS (PPMS)

• Reliance on only these 5 classifications for

clinical trials limits the generalizability of the

results across the entire MS spectrum and

restricts FDA approval for other subtypes of MS


Placebo-controlled trials

• It is difficult to recruit and retain patients

to take placebo because many

approved therapies are available

• Ethical concerns

Common Categories of Bias

47

Krishna R, et al. J Clin Diagn Res. 2010;4:2320-2324.

Selection bias

• Occurs when the groups to

be compared are intended

to be similar but are

actually different

o Volunteer or referral bias

o Nonrespondent bias

o Occurs when those

who do not respond

to a survey differ in

important ways from

those who respond

or participate

o This bias can work in

either direction

Measurement bias

• Can occur when systematic

error is introduced when

data are collected

o Instrument bias

o Insensitive measure bias

o Expectation bias

o Recall or memory bias

o Attention bias

o Verification or workup bias

Intervention (exposure) bias

• Is generally associated with

research that compares

groups

o Contamination bias

o Co-intervention bias

o Timing bias

o Compliance bias

o Withdrawal bias

o Proficiency bias

o When interventions

or treatments are

not equally applied

to the subjects

Relapse reporting by patients1

• Difficulty distinguishing a true relapse from a

pseudoexacerbation

o Relapses are over-reported

• Ignores worsening or new symptoms

o Relapses are underreported

Non-standardized definition and

counting of relapse1

• Studies vary on the definition of relapse

• Interpretation of symptoms differ by examiner

Outcome Measures

48

1. Krieger S, et al. Issues in the design and interpretation of multiple sclerosis clinical trials. In: Giesser BS, ed. Primer on Multiple Sclerosis. 2nd ed. New York, NY: Oxford University Press; 2016:535-552.

2. Martinelli Boneschi F, et al. Mult Scler. 2004;10(4):341-347.

EDSS limitations1

• EDSS may not capture accruing disability in

functional areas besides ambulatory

disability

• Significant amount of inter-observer

variation at the lower end of the scale

MRI measures and clinical correlation2

• MRI measures cannot predict clinically

relevant relapse(s) or long-term disability

outcomes

Drug approval in the United States follows either an NDA process or a BLA process1

• Many pharmaceutical products used in MS require a BLA

Clinical trials determine a treatment’s safety and/or efficacy on pre-specified outcomes2-4

• Trials can be categorized as experimental or observational and prospective or retrospective

Well-designed clinical trials are planned to achieve valid results and eliminate bias5,6

• Trials are also designed to be fair and ethical to the participants7

There is no single ideal measure for MS outcomes

• ARR is a common summary measure of relapses8

• EDSS remains the gold standard in clinical trials for measuring physical impairment from MS9

Statistical significance is the percent likelihood that the result occurred by chance2

• Statistical significance does not necessarily mean clinical significance

In MS trials, evaluating efficacy can be challenging due to the nature of the disease2

• Clinical trial design ensures that results obtained are real and not due to unseen influences or problems in the research protocol

1. US Food and Drug Administration. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/ApprovalApplications/TherapeuticBiologicApplications/default.htm. Updated [March 14, 2017]. Accessed

[March 25, 2017].


3. ClinicalTrials.gov. https://clinicaltrials.gov/ct2/about-studies/learn#WhatIs. Accessed [March 9, 2017].

4. Boston University School of Public Health. http://sphweb.bumc.bu.edu/otlt/MPH-Modules/EP/EP713_AnalyticOverview/EP713_AnalyticOverview3.html. Accessed [March 9, 2017].

5. Jüni P, et al. BMJ. 2001;323(7303):42-46.


7. National Institutes of Health Clinical Center. https://clinicalcenter.nih.gov/recruit/ethics.html. Accessed [March 15, 2017].

8. Lavery AM, et al. Mult Scler Int. 2014;2014:262350. doi: 10.1155/2014/262350.


Summary

49

Adap

ted f

rom

iS

tock.c

om

/alm

agam

i

understanding clinical trials - quoetr … clinical trials. ... • natural history of ms and number...

Documents