N117 Midterm

Clinical Trial Requirements

-prospective experiment on humans

-should compare to a control

-must have an “intervention”

-goal to examine safety and/or efficacy of therapeutic intervention

-have to fulfill requirements of US FDA to gain approval to market

Drug (FDA def)

any article intended for use in the diagnosis, cure, mitigation, treatment or prevention of disease and intended to affect the structure or any function of the body of man or animals

Clinical Trial Requirements

-prospective experiment on humans

-should compare to a control

-must have an “intervention”

-goal to examine safety and/or efficacy of therapeutic intervention

-have to fulfill requirements of US FDA to gain approval to market

Drug (FDA def)

any article intended for use in the diagnosis, cure, mitigation, treatment or prevention of disease and intended to affect the structure or any function of the body of man or animals

Drug examples

dietary supplements, antioxidants in natural substances…

Purpose of Clinical Trial

answer a scientific question and/or test a hypothesis.

Protocol

written document describing the means by which the hypothesis will be tested.

Primary objective

goal of the study

Primary outcome

measures used by the protocol to support or reject the null hypothesis

What must the hypothesis include?

-population studied

-primary outcome of interest

-intervention studied

-period of observation

-dose(s)

Gold Standard

-randomized

-double-blind

-placebo-controlled

Type I Error

concluding there is an effect when one does not exist (saying drug works when it does not)

Type II Error

concluding there is no effect when one does exist (saying drug does not work when it does)

Efficacy trials

determine whether an intervention produces the expected result under ideal circumstances

Effectiveness trials

measure the degree of beneficial effect under “real world” clinical settings

Failed Drug

an investigational product that must be halted from further development

Failed trial

a study of an intervention that fails to answer the proposed scientific question (good bc it tells us that this drug doesn’t work (narrows down the drug))

Ambroise Parè

first accidental clinical trial

soldiers with wounds

standard treatment was boiling water

test treatment was turpentine/egg yolks/oil of roses

soldiers with test treatment felt little pain compared to those with standard treatment

James Lind

conducted one of the first controlled clinical trials, testing citrus fruit on sailors with scurvy, which demonstrated the effectiveness of vitamin C in preventing the disease.

two arms of trial:

-oranges and lemons

-cider/vinegar/seawater/nutmeg…

Flint’s Rheumatism Study

First Placebo Controlled Clinical Trial

too much importance was being attached to use of medicines, people are just going to get better anyway

Medical Research Council Patulin Trial

first double blind controlled trial

suffering from colds

patulin vs placebo

no protective effect of patulin

Medical Research Council Streptomycin trial

first randomized controlled trial

treatment: streptomycin & bed rest vs placebo (bed rest alone)

randomized using random sampling numbers and sealed envelopes

streptomycin was beneficial in treating TB but patients developed resistance to drug/time

Food and Drug Act

response to deaths of several children due to contaminated smallpox vaccines.

only required drug to meet standards of strength and purity, no requirement for submitting any info to FDA before marketing

US vs Johnson

“Johnson’s Treatment for Cancer”

found that new law did not prohibit false claims, only false and misleading claims regarding ingredients/identity of drug

led to Sherley Amendment

Sherley Amendment

prohibited labeling medicines with false therapeutic claims

Elixir Sulfanilamide Disaster

first wonder drug to treat strep

due to high demand, they made it into liquid form, not realizing that the solvent used was a poison.

led to more than 100 deaths in US

Food Drug and Cosmetics Act

set of laws giving authority to FDA to oversee the safety of food, drugs and cosmetics.

prohibited false therapeutic claims & allowed FDA to block/delay marketing of new drugs but no true requirement for “approval”

Thalidomide

anti nausea medication during pregnancy

babies born with defects & miscarriages

drug was never approved in US bc of cosmetic act

Kefauver-Harris Amendment

a drug must be both SAFE and EFFECTIVE

sponsors are required to file a notice for exemption for use of an investigational new drug (IND)

Secondary Questions

more exploratory objectives

subgroup analyses

different from primary

Primary Question

-most important goal of the trial

-will determine if trial is negative/positive

-will determine trial sample size

-varies per phase of development

-limited in number

-pre-specified

Study Population

-need to be able to report what people were studied and how they were selected

-helps others interpret results, replicate findings, build on findings

How is successful recruitment defined?

-efficiently achieving full enrollment

-getting a sample that is representative of the greater disease suffering population

Inclusion Criteria

necessary elements to be enrolled

-diagnosis

-age range

-scores on given test

Exclusion Criteria

cannot participate if:

-comorbidity (ex. liver disease)

-certain medications

Ideal setting of Population

study sample should look like a random subsample from the population of all diseased patients who would ultimately be judged suitable for the intervention

Randomization

process to produce study groups comparable with respect to known and unknown risk factors, remove investigator bias in allocation of participants, and guarantee that statistical tests will have valid significance

-prevent selection bias

-ensure comparability of the groups

Simple Randomization

coin toss/random number generator for each subject

Block randomization

randomize “blocks” of participants

-ensure equal numbers of participants/group (for every 20 we’ll make sure 10 get drug and 10 get placebo)

Stratified Randomization

-takes into account certain participant characteristics to ensure balance

-ex. treatment center in multi center study, age, race, etc.

What do non-randomized trials suffer from?

Significant heterogeneity

Unblinded

-investigator and participant know assignment

Unblinded Strengths

-straight forward

-cheaper

-may facilitate recruitment

Unblinded Weaknesses

-risk for participant dropout

-inherently biased

Single Blinding

subjects/but not investigators are blinded to treatment assignment

Single Blinding Strengths

-straight forward

-investigators may be more comfortable or better able to deal with adverse events

Single Blinding Weaknesses

-bias resulting from imbalanced use of concomitant therapies

-bias from desire for positive results

Double Blinding

-neither participants nor investigators know treatment assignment

Triple Blinding

-data monitoring committee is given blinded data, or

-participants and investigators are blind to treatment assignment; data collection or assessment performed by blinded third party

Control Groups

-placebo control

-dose comparison control

-no treatment

-active treatment control

-historical control

Non-Placebo Control Groups

-conditions in which treatment effect would be unquestioned (ex. survival)

-when unethical to not treat with standard of care therapy

-when scientific question centers around advancing current clinical practice

Outcome Measures

-primary outcome measure determines if the intervention is positive or negative

Primary Outcome Measure

-must be prespecified

-must be unbiased

-must be capable of being assessed identically in all participants

-in some trials, participation ends when outcome is achieved

-must be as complete as possible

Wiley Act

labeling (but could claim whatever you want)

Preclinical Phase of Investigation

-epidemiology including risk factors

-basic science: physiologic mechanisms

-animal experiments: toxicology

Clinical Phase of Investigation

-phase I: initial safety/dose finding

-phase II: preliminary efficacy/ further safety

-phase III: confirmatory efficacy/ effectiveness

Approval of indication based on total evidence to date

-evidence based medicine

-phase IV: post-marketing surveillance

Phase I

trying to find balance between safety and maximally tolerable dose.

because more efficacy

Phase II

dose has been determined

safety would be second thing to look at, after efficacy

Lecanemab Phase III

-double blind

-PET and cerebrospinal fluid

-patients randomly assigned in a 1:1 ratio to receive lecanemab or placebo

-primary endpoint: change from baseline

Approval Process

goal is to provide the context for weighing the drug’s risks and benefits

-expectation is drug maker will submit results from 2 well-designed clinical trials

Managing Risks for Approval Process

-FDA approved drug label, describing the drug’s benefits and risks, and how the risks can be detected and managed

-sometimes REM (risk management and mitigation strategy) is required

Accelerated Approval

approval of a new drug can sometimes be expedited using an accelerated approval process

-applies to promising therapies that treat a serious or life-threatening condition and provide therapeutic benefit over available therapies

-allows for approval of a drug that demonstrates an effect on a s”surrogate endpoint”

(drug makers have to do post-marketing clinical trials) and FDA can withdraw approval

Fast Track

process designed to facilitate the development and advance the review of drugs that treat serious conditions and fill an unmet medical need

Breakthrough Therapy

-expedites the development and review of drugs that are intended to treat a serious condition, and preliminary clinical evidence indicates that the drug may demonstrate substantial improvement over available therapy

Priority Review

means that FDA aims to take action on an application within six months, compared to ten months under standard review

-directs attention and resources to evaluate drugs that would significantly improve the treatment, diagnosis or prevention of serious conditions

FDA Advisory Committees

-advisory committee are subject specific

-usually around 15-25 members on the committee

Aducanumab

one trial was cancelled (going opposite way than intended), clinical benefits in second trial, unanimous against approval, FDA gave accelerated approval

Makena

drug for pregnant women (premature babies)

3+3 Method

early phase study, to find maximally tolerated dose.

start at lowest possible dose

treat 3 subjects, if none experience dose limiting toxicity increase dose and repeat

if 1 experience DLT, top study and assume previous level was dose

Early Phase Designs

-examine safety

-proof of concept (does drug do what we want it to do?)

-later phase planning (determine doses, outcome selection, effect size)

Adaptive Dose Finding

-adaptive dose finding methods offer more efficient ways to learn about the dose-response relationship

Steps for Implementing a CRM

1. choose dose levels

2. choose a target toxicity level

3. choose a dose-response curve

4. obtain prior probabilities of toxicity at each dose level

5. assign a cohort of subjects to a dose level

6. update dose-toxicity curve based on occurrence of DLTs in cohort

7. next cohort of subjects are assigned the dose closest to target toxicity level based on updated curve

Parallel Design Strengths

-more straight-forward data analysis

-less issue with disease change over time

-less issue with dropouts

Parallel Design Weaknesses

-requires large number of subjects

-difficult to distinguish symptomatic from disease-modifying effects

Non-Randomized Controlled Trials

-comparison group receives no therapy

-but groups are not strictly comparable

Cross-Over Designs

-subjects are randomized to sequences of treatments (A then B or B then A)

-uses patient as his/her own control

-often a “wash-out” period is used to avoid a “carry over” effect

-can have a cross over design with more than 2 periods

-good for diseases that do not worsen over time

Crossover Designs Strengths

-decreased variability

-reduces necessary sample size

-useful in comparing 2 treatments or more

Crossover Designs Weaknesses

-difficult to eliminate carryover effects

-disease progression (or death) can bias results

-withdrawals make analysis challenging

-drug matching critical, since participants will see both therapies

Noninferiority/Equivalency Trials

-conducted in cases where withholding of a known effective therapy is unethical (syphilis/HIV)

-requires a positive control

-many possible endpoints

-must be adequately powered to detect a difference

Equivalency Trials

demonstration that the confidence interval of a new agent falls within the therapeutic measure of efficacy for an existing therapy

-absence of meaningful difference

Noninferiority Trial

one sided finding, demonstrating that the lower boundary of the confidence interval has been achieved

-not worse

Factorial Design

-attempts to evaluate two interventions compared to a control in a single experiment (simplest case)

-an important concept: interaction

-interaction: the effect of treatment A differs depending upon the presence or absence of intervention B and vice verse

Advantages of Factorial Designs

-if no interaction, can perform two experiments with less patients than performing two separate experiments

-can examine interactions

Disadvantages of Factorial Designs

-added complexity (interaction effects)

-potential for adverse effects due to “poly-pharmacy”

Withdrawal Studies

-removal or dose-reduction of patients from specific therapy

-useful in therapies known to be useful and those for which efficacy is questioned

-highly selected sample

Staggered Withdrawal

-everyone gets active treatment

-not always the best mechanism to demonstrate disease modification

-likely to result in patient dropout, due to loss of medication effects

N-of-1 Design

-special case of crossover/repeated measures

-a single subject undergoes several treatments over multiple time periods

-final outcome of the trial is a determination about the best treatment for that particular subject

Run-In Design

-addresses the regression to the mean problem

-assign all eligible patients a placebo to be taken for a “brief” period of time. patients who are “judged” compliant are enrolled into the study. this is often referred to as the “placebo run in” period

-can also use active drug to test for compliance

Futility Design

-screening tool to identify whether agents should be candidates for Phase III trials while minimizing costs/sample size

-main intent is to determine whether treatments seem unlikely to show a clinically meaningful effect (futile), and provide enough evidence to support a “go” decision to a larger, confirmatory trial (not futile)

Adaptive Design

-knowledge will accumulate as a trial progresses

-allow elements to be reviewed and used during the trial

-all changes should be planned for ahead of time

Cluster Randomization

-groups are randomized to treatment or control

-allows for mass intervention

Selenium Story Takeaways

-epidemiology as foundation for major intervention trials

-demonstrates the importance of confirmatory trials for subgroup effects in large trials

-selenium is not wonder drug

-prespecifying outcomes

-understanding that if you look hard enough, you will find something

Phase I

initial safety/dose finding

Phase II

preliminary efficacy/ further safety

Phase III

confirmatory efficacy/ effectiveness

Phase II Screening

-incidence of major adverse effects

-decide if worth studying in larger samples

-looking for any evidence of treatment efficacy

Public Health Objectives

-discover things that are true (does drug truly work)

-develop the science in order to provide public health benefit (therapy/prevention)

-want high prevalence of truly beneficial therapies

PPV

positive predictive values

-public health objectives

Sensitivity of Test

probability of positive in diseased

sensitivity = Pr[+l D]

probability that its positive given that you have the disease

Specificity of Test

probability of negative in healthy

specificity = Pr[-l d]

probability that its negative given that you dont have the disease

PPV of positive test

probability of disease when test is positive