Phase III studies are randomized, meaning that the treatment that is delivered to the patient is randomly assigned, among 2 or more treatment algorithms (or treatment ‘arms’). Often one treatment approach is considered “standard” and the other consists of the “standard” therapy plus additional treatment (i.e. a new drug). Other studies may compare “standard” treatment to a completely different treatment that was promising based upon Phase II studies. Usually 1-2 patients are randomized to receive an experimental treatment for every patient who is randomized to not receive the experimental treatment. There can be 2-4 (rarely more) different treatment ‘arms’- meaning 2-4 different treatment algorithms that a patient can be assigned to. For example, a study can have 2 randomizations (i.e. be randomized to receive or not receive a drug X, and be randomized to 2 different doses of drug Y), which would result in 4 ‘arms’- this is called a 2 x 2 randomization.
Phase III studies can be blinded – meaning that neither the patient nor the doctor know which of the two (or more) treatments the patient is receiving. It is possible that one of the treatments will use a placebo – an inert substance with no efficacy against cancer – and neither the patient, the doctor or study coordinators will know if a placebo is being given. The reason for using placebos is that the researchers want an unbiased assessment of how well the experimental treatment is working. By not knowing whether or not a placebo is given, neither the patient nor doctor will be biased one way or the other (consciously or subconsciously) in assessing treatment response or side effects.
The major difference between a randomized Phase II study and a Phase III study is that the Phase III study plans to enroll for more patients (>300-3000) versus a randomized Phase II study which may enroll dozens to hundred of patients. While randomized Phase II studies need to justify the number of patients they enroll, based upon the study’s goals, Phase III studies have stricter statistical requirements. For Phase III studies, the number of patients to be enrolled is determined beforehand on the basis of the number of patients needed to see a statistical difference, of a given magnitude, between treatment arms. For example, researchers decide beforehand what the anticipated difference might be (say 10% difference in survival), and the number of patients that need to be enrolled to see that difference. While smaller differences in treatment outcome (say 5% difference in survival) may also be considered clinically meaningful, to ascertain small differences in outcome between treatment arms may require an inordinate number of patients.
Next, learn about how mesothelioma clinical trials work.