Go to main content
Type 1
Type 2

Inside Drug Development

by brendan milliner and lisa rotenstein

In diaTribe, we sometimes talk about drugs that are “not yet approved”, “at the FDA” or “in the regulatory process”. Some recent approvals of diabetes drugs in the US have stirred some controversy, and we think it's important for our readers to understand the basics of how drugs are discovered and approved. In this issue of Learning Curve, we'll take you through the development and regulatory process for drugs. As an example, we'll be following a recently approved diabetes drug called Cycloset, which acts on the brain to help improve blood sugar (see this issue's NewNowNext for more detail) - this drug, made by VeroScience, actually spent TEN years at FDA before it was approved. This is a timely topic, because there are a number of drugs currently before the FDA in the US and other regulatory agencies globally - just to name a few, liraglutide, exenatide once-weekly, saxagliptin, alogliptin, and Afresa inhaled insulin are all at the FDA now, waiting for a decision.


Many older drugs were discovered by accident, when somebody realized incidentally that a substance had a therapeutic effect. Today, most diabetes drug development follows a more linear track. First, researchers identify a “target” - something in the body that if acted on will have a positive effect. Next, the pharmaceutical industry works to develop a drug - usually a compound that is made by chemists, but sometimes a “biologic” that is made by living cells - that will act on this target.

Before this new drug can be evaluated in people, it must first be tested for safety and effectiveness in test tubes and in animals. These non-human studies are termed preclinical studies. The kinds of cells or animals used for these trials varies depending on the type of compound used, how it will be administered (as an injection, a pill, etc.), and which part of the body it is designed to affect. Many trials of diabetes medications are done in mice, which can be genetically manipulated to show symptoms of the disease being studied and are easy to care for, and primates, which are biologically much more similar to humans.

The types of preclinical trials conducted with animals can vary quite a bit as well, but the primary goal is to establish that the drug is both effective enough to warrant human study and safe enough that human volunteers won't be putting themselves in serious danger. Studies in animals also test to make sure the compound won't cause cancer or mutations, or damage an embryo or fetus. By giving animals such as mice very high concentrations of a potential drug, researchers can find the limits of safety so that they don't accidentally give a dangerous dose to a human subject. Mouse models are also often used to simulate human diseases.

clinical trials

Human clinical trials for a particular drug are initiated in volunteers after all laboratory and animal work has been completed and the data have been presented to the regulatory bodies for assessment. Clinical trials are designed to assess the safety, efficacy, and action of a drug in human subjects. Each clinical trial phase asks a different question about the drug.

phase 1 trials

A phase 1 trial is the first stage of testing of a drug in people. There are usually two of these trials and they generally involve small groups of healthy people without diabetes (usually 20-80 individuals). They explore what dose of a drug should be given by looking at the safety of different doses. Most of the work is done in a hospital or clinic while health professionals assess the drug’s tolerability and mechanism of action in patients (including the speed of action and the drug’s interactions with other substances in the body). The first studies start with a single dose of drug; subsequent studies are called multiple dose studies. Volunteers may dose for a period of hours to a couple weeks, depending on the drug. Typically a company can establish a range of doses that are perceived to be sufficiently safe and tolerable for further testing.

For example, a phase 1 Cycloset trial was started in April of 1997 and ended in June of 1997. It studied 120 healthy subjects aged 18-45 who were placed randomly in two groups. Patients either received a single dose of Cycloset or a dose of a drug similar to Cycloset (named Parlodel) daily and were studied throughout the two-month period. Phase 1 trials are usually held only in one location. This phase 1 Cycloset trial was held, for example, only in Quebec.

phase 2 trials

Phase 2 trials typically have larger groups of participants (100-300 individuals), and these people usually have the disease the drug is intended to treat. They look to see how well the drug works for its specified purpose, and data are still collected about its safety. This is the first time the company sees whether the drug works as planned, but also the first time when a drug can fail based on a lack of efficacy. This type of trial examines several potential doses of the drug within the range identified in phase 1 to determine at which dose levels the drug is efficacious. These trials are still usually held in a limited number of locations. For Cycloset, phase 2 trials determined that the best dose is around 1.6 mg per day, with a maximum of 4.8 mg per day. The studies showed that patients should take the drug in the morning and before meals for a maximum effect, and that they can increase the number of tablets gradually if they don't see an effect at first-up to the maximum of 4.8 mg.

phase 3 trials

This phase of testing is usually much larger than the other two phases-typically, several thousand patients will be enrolled. Phase 3 trials help to provide large-scale efficacy and side effect data so that researchers can understand how best to use the drug. They help to make sure that the drug works in a range of different patients, and doesn't cause rare side effects that might have been missed in smaller trials. Typically there are at least two phase 3 trials with similar endpoints, so that researchers can prove that the effects and safety findings weren't just a function of chance, and that the effects can be repeated. These trials are large, expensive, and are the last hurdle before sending the drug's information to the FDA for approval. Often as these studies are being conducted, additional phase 1 studies looking at drug interactions (with commonly used medications) or in special populations (those over age 65 or those with kidney disease and/or cardiovascular disease) are conducted as well for an even greater understanding of the effects of the drug across all relevant groups of potential users. For example, many people with type 2 diabetes are over 65, take several other medications for their cholesterol or blood pressure, and may have some decrease in kidney function - it is important to know how these patients will do with a new drug for diabetes.

VeroScience conducted a phase 3 trial comparing the adverse events in patients receiving Cycloset with those receiving a placebo, and continued to look at whether Cycloset effectively lowered blood glucose and A1c. The phase 3 trial included 3,000 people with type 2 diabetes over two years. It allowed VeroScience to be able to label its drug and report observed side effects to the FDA-nausea, fatigue, vomiting, etc. Other serious adverse effects weren't seen with Cycloset use in this trial.

submission and approval

Upon completion of phase 3 trials, the company must submit a New Drug Application (or NDA) to the FDA. The NDA contains information from the entire development program of the compound (preclinical and clinical), and also includes information on plans for manufacturing and marketing the drug to consumers. The company then waits for the FDA to reach a decision on whether the drug is marketable. After the FDA has accepted a filing, it sets a “PDUFA date”, which is the time by which the FDA must reach a decision. The review process may take a year or longer (as is currently the case, especially in diabetes, although Cycloset's ten years at the FDA was extreme), and additional studies are sometimes required to further convince the FDA of the safety or efficacy of a candidate drug.

Typically, the FDA will convene a panel of experts to advise it based on a more detailed analysis of the data (this panel is called an Advisory Committee). See diaTribe 14's Dialogue for an interview with Rebecca Killion, the inspiring patient representative on this board. During a one-day meeting, the Committee is asked to debate specific questions put forward by the FDA and then give its final opinion by a vote. The FDA is not required to follow the recommendations of the Advisory Committee, but experience would suggest that it usually does. See NewNowNext in diaTribe issue 15 for our coverage of the Advisory Committees for two diabetes drugs that haven't yet been approved, saxagliptin and liraglutide.

If a drug is approved by the FDA, the sponsoring company may begin to sell the drug and advertise to consumers and health care professionals. In many cases, the company may be asked to conduct additional large-scale trials of the drug once it is already on the market-these are known as post-marketing trials. Cycloset was just approved by the FDA in May 2009, and its sponsoring company VeroScience was able to begin marketing at this year's ADA meeting. Notably, regulatory processes must take place in every single country or region in which a company wants to sell a drug.

The approval process for diabetes drugs has become more complicated since the FDA released new guidance intended to reduce the risk of cardiovascular (CV) side effects from diabetes drugs. This new guidance is in addition to the previous FDA requirements. To rule out CV effects, a drug must be given to a large number of people over a long period of time (often many years), and this type of trial can be expensive and difficult for drug companies to complete before the drug is approved and making money. There has been a great deal of controversy surrounding this issue in the diabetes community, because of fears that the new restrictions will make companies less willing to develop new diabetes drugs that could be helpful for patients. Click here to see the letter sent to companies by the FDA. Particularly following new data given at this year’s ADA, we think the requirement for pre-approval screening is not necessary and may delay innovative new treatments for people with diabetes.

how can i enroll in a clinical trial?

Enrolling in a clinical trial is often a good way to explore new therapies and gain access to cutting-edge medications that haven't yet been approved by the FDA. It is also often a great way to learn more about your disease. Everyone that participates in a clinical trial will go through an informed consent process, and will hear about all the known risks and benefits and everything that will be involved in their participation. They can then choose whether or not to participate. Most clinical trials are blinded and randomized, which means you won't know whether you are on the test medication, a comparator drug, or a placebo “dummy pill.” It is important to know that you and the doctor won't have any say in the treatment that each individual participant receives--treatment groups are randomly assigned, and neither you nor your doctor will know. This helps with reducing bias in trials, but is important to know before you participate. At the same time, clinical trials are just that-trials-and so not all questions about these drugs have been answered.

One way to get involved with a trial is to ask your doctor-many doctors are part of trials or lead trials themselves. You can also go to several websites to see if you'd qualify for any trials or some of the drugs being tested might meet your needs:

how to find out about or report drug side effects

The FDA plays an important role in monitoring the safety of drugs even after they get to market, and as a member of the public you have access to FDA safety warnings. Also, you can do your part in informing others by reporting side effects that you experience while on a medication. Both the FDA safety warnings and the reporting system can be accessed through MedWatch.

Whew! A long proess, but an important one! Outside the US, many countries have similar processes, all in the name of bringing safe, effective medicine to those who need it.