ACTIVITY 5: NON-INFERIORITY (NI) TRIALS

NON-INFERIORITY TRIALS

• RCT are utilized to _

• determine one of 3 different outcomes between comparative drugs

RCT are utilized to determine one of 3 different outcomes between comparative drugs:

• superiority 

• equivalency 

• non-inferiority 

Superiority

Determine that a drug is superior to a comparator, comparator is often placebo.

Equivalency

Determine whether the new drug is therapeutically similar to the control (determine bioequivalence between two drugs)

Non-Inferiority

• Seek to show that any difference between two treatments

  • Shows the test drug to be statistically and clinically not inferior to the reference drug based on efficacy.

Non-inferiority trial or NI trial is a

variation to the controlled clinical trial design.

Non-inferiority trial or NI trial seeks to

• show that any difference between two treatments is small enough to conclude the test drug has "an effect not too much smaller than the active control" or reference drug.

• establish that the experimental treatment is not clinically worse than the standard treatment comparison by more than a small, predetermine margin.

The concept on NI trials is to

compare two possible treatment options and determine if the experimental treatment is not clinical worse than the standard treatment.

If the hypothesis is proven to be true, then patients can

select the less expensive forms of treatment without sacrificing their chances at getting better since both treatments offers comparable clinical efficacy. 

More NI trials are getting published over the years since 1999, as evidenced by

more FDA registrations using NI trials as references (US FDA).

Non-inferiority (NI) trials have hypotheses that are_

inverse of the ordinary hypotheses in a controlled clinical trial

Superiority

null hypothesis

there is no statistically significant difference between treatment groups

Superiority

alternative

there is a statistically significant difference between the treatment groups

NI

null hypothesis 

the test drug fails to exhibit non-inferiority to the reference drug within a certain margin

NI

alternative hypothesis 

states that the test drug is noninferior to the reference drug (not significantly worse) within a certain margin (i.e., the NI margin)

SUPERIORITY TRIALS

objective

To determine if one treatment is superior over another

SUPERIORITY TRIALS

null hypothesis

There is no difference between the two treatments

SUPERIORITY TRIALS

alternative hypothesis

One treatment is superior over the other

SUPERIORITY TRIALS

limit

p-value depends on the statistical probability

EQUIVALENCE TRIALS

objective 

To determine if the experimental and standard treatment are equal within a predetermined range

EQUIVALENCE TRIALS

limit 

-◿ to +◿ 

from negative to positive differences

NON-INFERIORITY TRIALS

objective 

To determine if the experimental treatment is not clinically worse than the standard by a small preset margin

NON-INFERIORITY TRIALS

null hypothesis 

The experimental is inferior to the standard treatment within the preset margin

NON-INFERIORITY TRIALS

alternative hypothesis

The experimental is no worse than the standard treatment within the preset margin

NON-INFERIORITY TRIALS

limit 

0 to +◿ 

from zero to positive differences

ADVANTAGES OF NI TRIALS

• Useful when placebo control is inappropriate (ethical nature of the study)

• Not limited to pharmaceutical therapy

• Can conduct risk-benefit analyses for therapies

• Appropriate in comparing doses of similar drug agents or formulations of similar treatments

• Provides a peek at the non-inferiority, equivalence, and superiority of a particular treatment

DISADVANTAGES OF NI TRIALS

• Requires specific design and analytical parameters for the results to be useful and meaningful

• Appropriate for use where standard treatment does not show a consistent efficacy vs placebo (unestablished efficacy of standard)

• Often requires a large sample size compared to a superiority trial

NI MARGIN

• A pre-specified value used to determine if the test drug’s treatment effect is not worse than the reference drug by more than this specific degree

• The extent to which the test drug’s therapeutic effect can be less than the reference drug, but still be considered not worse

METHOD TO DETERMINE NI MARGIN

Confidence Interval (CI) for the reference drug treatment effect

Confidence Interval (CI) for the reference drug treatment effect

(reference drug – placebo = treatment effect) → Metaanalysis

The CI around the_ is used to establish the NI margin. 

mean reference drug treatment effect

The CI around the mean reference drug treatment effect is used to establish the NI margin. 

• The lower bound of that CI represents the smallest expected reference drug treatment effect

 if the CI around the mean reference drug treatment effect CI 6–15 mmHg,

• What can be used to set the NI margin?

6 mmHg change in DBP could be considered the smallest clinically significant change.

NI margin is set using

historical placebo-controlled studies or controlled trials to determine the actual treatment effect of the reference drug. 

Overall design of NI trials closely follows a__

superiority trial

ASSAY SENSITIVITY

Concluding that the test drug is efficacious can only be justified if the reference drug’s efficacy has been confirmed with high quality clinical trials against placebo

ASSAY SENSITIVITY

standard treatment to be used in the comparison must be __ 

a well-established treatment with at least one superiority trial to establish its effectiveness over a placebo treatment. 

In a noninferiority trial, it is key that the reference drug used is

superior to placebo. 

“historical evidence of sensitivity to drug effects (HESDE)

- used when past trials have been appropriately designed and conducted to confirm the superiority of the reference drug over placebo. 

Metaanalytic methods can be used to

develop more precise estimates of reference drug effect when several studies are available.

When can HESDE be used to select the NI margin? 

• Only after a determination is made that these past studies are similar in design and conduct compared to the NI trial

CONSTANCY ASSUMPTION

• The historical studies and the new NI study should be as identical as possible regarding important characteristics. 

• Randomization, blinding, inclusion/exclusion criteria, outcome measures, conduct of testing, etc. should be identical to previous studies

In establishing efficacy for the test drug using an NI trial design, __are used

assay sensitivity and constancy assumptions

NI margin must be set _

prior to trial testing and it must include justification of the margin, stating statistical and clinical validations

NI margin should include

• justification of the margin, stating statistical and clinical validations

• Investigators should provide a detailed description of how the NI margin was determined

NI MARGINS

• Should not be larger than the “smallest” value representing a clinically significant difference between treatments

• Commonly a value smaller than the difference in effect between the standard treatment and placebo

REPORTING RESULTS

• A combination of __ is advisable to create a balanced conclusion in consideration of biases produced

Intention-to-Treat (ITT) analysis and Per-Protocol (PP) analysis

PP analysis

Only the patients that completed the study are included in the analysis.

ITT analysis

Includes all patients that were randomized to treatment, regardless of whether they completed the study duration,

ITT analysis limitation

Increase the risk of falsely claiming noninferiority due to the potential of smaller observed treatment effects. 

The differences in the effects of treatment are presented in the form of

confidence intervals (CI). 

CI is compared to the NI margin to

determine if the experimental treatment is inferior, non-inferior, or superior to standard treatment

If CI is entirely below 0,

the experimental drug is SUPERIOR

if only the upper limit of CI is above 0 and within the NI margin,

the experimental drug is NON-INFERIOR

if the upper limit of CI is beyond the NI margin

the experimental drug is NOT NON-INFERIOR

if the entire CI is within the NI margin,

the experimental drug is NON-INFERIOR

if the CI is entirely above the line of non-inferiority,

the experimental drug is INFERIOR

ERRORS IN NI TRIALS

• use of improper or misleading terminology

• inappropriate NI margins

• confused error types

• incomplete analysis or results

• non-inferiority claims from a superiority trial 

INCOMPLETE ANALYSIS OF RESULTS

• Analysis using ITT and PP analyses to produce balance output

NON-INFERIORITY CLAIMS FROM A SUSPERIORITY TRIAL

• Data is removed from the NI margin or sample size did not consider NI margin

USE OF IMPROPER OR MISLEADING TERMINOLOGY

• In reporting results to confuse readers about the comparison between experimental and standard treatment

INAPPROPRIATE NI MARGINS

• Not set at the start of the study, unacceptable justification for both statistical and clinical relevance

CONFUSED ERROR TYPES

• Type I error

• An inferior treatment is accepted as non-inferior

  • Once a Non-inferiority is established, but the true effect was inconclusive

  • Assay sensitivity and constancy assumptions that cannot be completely verified

CONFUSED ERROR TYPES

Type II error

— A non-inferior treatment is rejected to be inferior

• Noninferiority is not demonstrated, but the test drug is noninferior to the reference

• There is an excessively large dropout rate due to safety and lack of return to the clinic

EVALUATING SOME OF THE IMPORTANT CHARACTERISTICS OF NON-INFERIORITY TRIALS

Formulate a null hypothesis.

Assess the noninferiority margin

Compare the intention-to-treat (ITT) and per-protocol (PP) analyses.

Translate CIs graphically

Summarize the clinical relevance of the results.

CONCLUSION

• Non-inferiority (NI) trials to show that any difference between two treatments is small enough to conclude the test drug has “an effect not too much smaller than the active control” or reference drug.

• Null hypothesis of the study is that the experimental and standard treatment are equal within the reset margin while the alternative hypothesis is that the experimental treatment is no worse than the standard treatment within the preset margin.

• Errors in NI trials include use of improper or misleading terminology, inappropriate NI margins, confused error types, incomplete analysis of results and non-inferiority claims from a superiority trial