Phases of Clinical Trials

Clinical trial phases correspond to the different stages of testing that a drug goes through before approval. For people with brain tumors, the clinical trial phases are phase 0, phase 1, phase 2, or phase 3. The phases are sometimes referred to using the Roman numerals 0, I, II, III.

Phase 0 studies

Phase 0 clinical trials are usually done very early in the testing of a new drug. Phase 0 clinical trials test whether a drug gets into the brain or brain tumor cells and whether it “hits” the expected target. For example, if a new drug is developed to inhibit the growth protein EGFR, a phase 0 study can look at the activity of the EGFR protein in the brain tumor before and after treatment with the new drug. Phase 0 studies usually require surgery to obtain a sample of the tumor after treatment with the drug; however, depending on the specific drug and the expected target, testing of drug effects can sometimes be done on blood samples or other tissues. A common way of doing phase 0 studies is to have someone take a drug for a week before a planned surgery and then use some of the tumor removed during surgery to determine the effects of the drug on the tissue. One important aspect for patients to understand about Phase 0 studies is that they generally are not expected to benefit the individual entering the study. These studies are often done very early in the development of a drug before the optimal dose is determined, and treatment is often continued only for a short period of time. These studies do, however, potentially allow for a better understanding of how a drug works; this, in turn, aids in the planning of further trials. Phase 0 studies are generally not applicable for patients with inoperable tumors, such as diffuse intrinsic pontine gliomas, which some believe cannot be biopsied safely.

Phase 1 studies

Phase 1 clinical trials are done early in the testing of a new drug or new treatment combination to find what doses of a drug are safe to give. Phase 1 clinical trials often also test how long the drug stays in the body and how the body processes and gets rid of the drug. There are sometimes a number of extra tests required in phase 1 studies, such as blood tests, urine tests, EKGs of the heart, etc., that test for toxicity to different organs and to see how long the drug is staying in the body and how it is getting out of the body.

For traditional chemotherapy drugs, phase 1 studies often have the goal of defining what is called the Maximum Tolerated Dose, or MTD. The maximum tolerated dose is the dose just below where a high proportion of people have unacceptable side effects from a drug. The maximum tolerated dose is the highest dose that should be used for that drug because higher doses are generally too toxic. Sometimes, particularly for modern drugs that target specific molecules in cancer cells, phase 1 studies do not go all the way to the maximum tolerated dose, but stop at a dose where the drug is expected to completely inhibit the molecule it is targeting. This is called the Biologically Effective Dose.

People are enrolled in phase 1 studies as part of a group, which typically is called a cohort. Cohorts usually consist of three people but can vary in size from 1 person to many people. Each cohort is given a different dose of the drug being tested. The first cohort gets a very low dose. If the side effects are tolerable, the next cohort gets a higher dose. This continues until too many people in a cohort get intolerable side effects. These intolerable side effects are called Dose Limiting Toxicities (DLT). The highest dose whose cohort did not get intolerable side effects is the Maximum Tolerated Dose (MTD).

Phase 1 trials are a very important step in developing new cancer drugs, and everyone in a phase 1 trial is helping patients in the future. Traditionally Phase 1 studies have not focused on how well a drug works—just the safety and dosage. In most phase 1 studies, once you are in a cohort, the dose of the drug will stay the same for you. Other people who are in other cohorts may get a higher dose of the drug. Therefore, people in the earlier cohorts of a phase 1 study may get a dose of the drug that is below the dose that is thought to be needed to shrink the tumor. For this reason, as well as the fact that most drugs tested in Phase 1 will not go on to demonstrate significant efficacy in future studies, the expectation is that most people in a phase 1 trial are not likely to personally benefit from the treatment being tested. Therefore, phase 1 trials are often most appropriate after patients have failed standard treatments with proven benefit or when there are no proven treatment options available.

However, as drug companies have become more focused on drugs targeting specific proteins and molecular pathways in tumors, the role of Phase 1 testing has started to change. In the case of these drugs, the specific targets are sometimes known before the study. Thus, it is possible in these specific situations to hypothesize that patients whose tumors have that specific molecular target may be more likely to respond to that specific drug. So, in addition to defining the Maximum Tolerated Dose, some more recent Phase 1 studies also include initial measurements of tumor response or drug effects on tumors. This approach has been used successfully by scientists to demonstrate potential efficacy of promising drugs early in their development, which has accelerated the subsequent Phase 2 and 3 studies and FDA approval. It is possible that in the future, patients participating in Phase I trials of targeted drugs may have a higher chance of getting benefit from the treatment in the situation in which the patient and the physicians know that the patient’s tumor has that molecular target.

Phase 2 studies

Once the safe dose of a drug is known, phase 2 clinical trials are used to see if the drug is effective against a certain kind of tumor. Phase 2 trials may have many different designs. In some Phase 2 studies, called single arm studies, everyone gets the same drug and dose. In multi-arm or randomized studies, there are several different treatments being tested or compared. Patients may get a specific treatment based on particular criteria of the study, or patients may be randomly assigned to treatment arms. Randomly assigned, also called randomization, means that neither you nor your doctor can choose what treatment you get. Such randomization is necessary because otherwise conscious or subconscious biases can influence the results of a trial. For example, if doctors gave everyone with small tumors the new drug and everyone with big tumors the old drug, then the new drug might look better even though it might not be better than the old drug.

The goal of a phase 2 study is to see if a new drug or combination of drugs has some beneficial effect on the tumor or other preferred patient outcomes. However, even though some Phase 2 trials include more than 100 patients, they are generally not large enough to conclusively prove whether people live longer when they take the new drug or to prove absolutely that a new drug helps. Instead, phase 2 trials often look at endpoints other than how long people live that can indicate whether a drug is helpful. These alternate endpoints can include:

Response rate: the percent of tumors that shrink a certain amount (usually 25%) with a treatment.
Disease control rate: the percent of tumors that shrink or stay the same size with a treatment. Because of random variation between scans, tumors must grow by at least 20% to 25% to not be considered the same size. Thus, if a tumor grows 10% or shrinks 10% it is considered stable, i.e., the same size.
Progression Free Survival (PFS): the amount of time from the start of treatment until someone either dies or the tumor progresses, which means it grows more than a set amount. PFS is often described as a median, the amount of time until half of people die or have tumor progression. For example, if the median PFS is 6 months, then by 6 months after the start of treatment, half of the patients have either died or had their tumor grow and half of the patients are alive with tumors that are either the same size or smaller.
PFS3 or PFS6: the progression-free survival at 3 months or 6 months respectively, after the start of treatment. This is the percent of people who are alive with tumors that are the same size or smaller at the specified time point.

While it seems intuitive that treatments that shrink tumors or that delay tumor progression will make people live longer, there are occasional cases where this has turned out not to be true. For example, sometimes tumors can grow back faster after a drug stops working or a tumor may appear to shrink on an MRI but the cells are continuing to grow and invade new parts of the brain. Because the study is testing how effective a drug is, patients participating in a phase 2 trial have a higher chance of personally benefiting from taking the drug than in earlier Phase studies. However, trial participants may not benefit if the treatment turns out to be ineffective or if the treatment only works at certain doses or in certain people. Often as part of a phase 2 trial, tests are done on tumor specimens or on blood samples to try to identify which people are more likely to benefit from the drug. A test that can distinguish people who might benefit from a drug and those who will not benefit from a drug is called a biomarker.

Phase 3 studies

Phase 3 clinical trials are large trials designed to definitively prove whether or not a treatment really works. Phase 3 clinical trials are done to compare a new treatment to something else, often the treatment that is considered “standard of care” for that disease and situation. This standard treatment may be another older treatment or in some cases may not involve active treatment against the tumor. Phase 3 trials must be randomized; meaning the treatment you receive is decided randomly by a method determined by the study designers, not by you or your doctors. Random means no one can control what treatment you get, as if it is determined by the flip of a coin or the roll of a die. The group you are randomized into is called an arm of the trial. Randomized phase 3 trials are needed because the results of non-randomized trials, such as many phase 2 studies, can be misleading. Non-randomized trials compare their results to what one would expect to happen to the usual patient in that situation. However, non-randomized studies can give a falsely optimistic view of a new drug because:

People who volunteer for clinical trials are usually healthier and physically stronger than those who do not volunteer. The inclusion criteria for clinical trials weed out people who have other health problems or are expected to be very sick soon.
People who can travel to big hospitals for clinical trials tend to have slower growing cancers than people who cannot travel.
Over time doctors have gotten better at preventing and managing symptoms and side effects, so people may live longer.
People in clinical trials may see their doctor or be contacted by nurses more often than people not in clinical trials. Therefore, problems can be dealt with early before they become untreatable.
When doctors and patients think a treatment is working, they are more likely not to see evidence that it is not working.

Although not very common in oncology, when there is no known effective treatment for a particular situation, then it is considered ethical and appropriate to compare a new treatment to giving no anti-tumor treatment. Giving no anti-tumor treatment does not mean “doing nothing.” It may be called Best Supportive Care, which means treating all of the symptoms of the tumor but not giving anything to fight the tumor. Giving no anti-tumor treatment may also involve a placebo. A placebo is used so that neither the person in the study nor their doctors know whether the person is getting the new drug that is being tested. When people and their doctors do not know what treatment someone is receiving, they are said to be blinded. Placebos prevent doctors and patients from being subconsciously biased in interpreting MRIs and symptoms by knowing a person’s treatment arm. Placebos also prevent people from dropping out of the trial and entering another trial if they are randomly assigned to the arm with no tumor treatment. A placebo will be a pill or infusion that looks and feels identical to the active drug, but it does not have any drug in it. Some people call placebos “sugar pills” because they used to contain sugar instead of drugs.

Many people do not like the idea of getting a placebo or of getting randomized. Rarely is the outcome without treatment so certain or a treatment so good that randomized trials are not needed. Some people like to joke that randomized, placebo-controlled trials are not needed to show that parachutes are helpful if you jump out of an airplane. However, most cancer treatment situations are not that clear cut. Although clinical trials are done when it is believed a new treatment is a good thing, sometimes it turns out that the new treatment is worse than not treating the tumor as there are unexpected additional side effects.

Some phase 3 trials allow people to receive the experimental treatment from the other arm of the trial when the tumor grows. This is called crossing-over. For example, a trial may randomize people between receiving drug "A" or a placebo. When the tumors of the people receiving the placebo grow, those people may be able to cross-over to the other arm and receive drug "A". Not all trials allow crossing over. Whether crossing over is permitted or not depends on the purpose of the trial, the resources available for the trial, and regulatory requirements.