Investigating the efficacy and safety of an agent or combination of agents through clinical trials involves multiple phases, beginning with smaller Phase I trials and culminating in larger, randomized Phase III trials. Phase IV trials may also be conducted to gather additional information.^1-3

Trials phases in the United States are defined as follows^2-4:

• Phase I: Initial trial of a drug in humans for dosing, safety, and early efficacy information (20-80 patients*).

• Phase II: Subsequent trial of a drug's safety and efficacy in a particular disease setting (100-300 patients*).

• Phase III: Larger trial comparing a drug with best available therapy to confirm efficacy and safety; often used for drug approval (100-3000 patients*).

• Phase IV: Trial conducted after US Food and Drug Administration (FDA) approval to gain additional information about the drug's risks and benefits. Phase IV studies may range in study size, from a few hundred to a few thousand patients.

*Patient numbers cited here apply to clinical trials in general.

When designing a clinical trial, investigators define a number of parameters in order to generate meaningful results. These parameters include⁵:

• Patient population to be studied

• Therapies to be investigated

• Endpoints

• How the trial will be conducted (eg, randomized vs non-randomized)

When determining the patient population to be studied, investigators should include patients who are likely to respond from the intervention being tested.⁵ The population should also be selected such that the results of the trial can be generalized to patients in clinical practice. The more diverse the patient population, the more generalizable the results may be to the wider patient population.⁶

In order to study a patient population of the appropriate disease state and level of diversity, investigators define criteria that determine if a patient is eligible for a trial.⁶

Inclusion and exclusion criteria can include patient characteristics (eg, age) as well as disease- and treatment-specific characteristics (eg, number and type of prior therapies).⁶

Efficacy and safety in clinical trials are measured by means of predetermined endpoints, or outcomes, that the trial is designed to evaluate. These may include clinical endpoints, as well as surrogate endpoints, which are expected to predict for a clinical outcome.⁵

The primary endpoint is the key measure from which clinical benefit is assessed. The primary endpoint selected by the investigators impacts the number of patients needed for the trial and must be determined before the trial is initiated.⁷ Secondary endpoints are other outcomes that provide additional and potentially valuable information about the investigational treatment(s). The trial protocol should pre-specify secondary endpoints to increase the likelihood that statistical analysis of those endpoints will be valid.⁵

Determining the primary endpoint requires consideration of several factors:

• What is the most clinically meaningful measure of benefit that could guide treatment decision-making in this disease and patient population?

• Can the trial be conducted in a reasonable timeframe?

  • Some endpoints require longer follow-up than others, lengthening the time required to complete the trial and obtain meaningful results.⁸

• Can a sufficient number of patients be recruited to complete the trial?

  • Some endpoints necessitate larger trials in order to demonstrate statistically significant differences between arms, potentially creating difficulties in recruitment.⁸

When the trial population, treatment, and endpoints have been identified and defined, the trial design is not yet complete. In Phase III and some Phase II trials, the patient population may be randomized and stratified. The gold standard in clinical research is a scientifically rigorous, randomized, and well-controlled trial.