An overview of the drug development process.

Pharmaceutical medicine uses all the scientific and clinical knowledge acquired by physicians in medical school and postgraduate training--combined with additional regulatory and business skills--to provide a challenging and rewarding career

Unlike clinical medicine, pharmaceutical medicine is part of an industry with huge up front investments for rewards that may or may not come years later. To develop new drugs takes a very, very long time--2 to 12 years from discovery to market, on average--and the cost is extremely high. It costs about $1.8 billion to take a new compound to market and success is quite limited.

Only one in 10,000 compounds ever reach the market. Of those only one in three ever recaptures its development costs. High risk indeed!

Drug development is a scientific endeavor that is highly regulated because of legitimate public health concerns.

Drug development phases

There are three major phases in drug development:

1. Pre-clinical research and development

2. Clinical research and development

3. After the compound is on the market, a possible "post-marketing" phase

[GRAPHIC OMITTED]

The pre-clinical phase represents bench (in vitro) and then animal testing, including kinetics, toxicity and carcinogenicity. In the U.S., an investigational new drug application (IND) is submitted to the Food and Drug Administration seeking permission to begin the heavily regulated process of clinical testing in human subjects.

The clinical research (IND) phase--representing the time from beginning of human trials to the new drug application (NDA) submission that seeks permission to market the drug--is by far the longest portion of the drug development cycle and can last from 2 to 10 years.

Phase I trials, sometimes called, "first in human" trials, are generally conducted on relatively small groups (typically 10 to 30) of healthy volunteers (except for oncology drugs or other potentially toxic compounds) in specialized units resembling small hospitals with 20 to 50 monitored beds.

The "inpatient" portion of Phase I trials usually lasts from a day or two to a week (though follow up can last up to about a month), and are designed to assess the safety of a compound and study its pharmacokinetics (Pk--what the body does to the drug) and pharmacodynamics (Pd--what the drug does to the body).

In some cases, human metabolism can differ markedly from animals so that a drug with a half life of a few hours in dogs may turn out to have a half life of several days in humans, or a compound with no animal toxicity may cause elevation in liver functions or a prolongation of QT interval in humans.

A rough idea of the maximum safe or tolerated dose, as well as a general side effect profile is obtained during Phase I trials. Many compounds never make it past Phase I, as they are found to have unacceptable side effects.

Assuming a compound is shown to be safe for healthy subjects and survives Phase I, then development proceeds to a series of Phase II trials. These trials typically enroll anywhere from about 20 or 30 patients up to a few hundred at most.

These patients usually have a relatively "pure" form of the disease for which the drug is intended. In other words, they suffer from as little other intercurrent disease as possible, and the list of concomitant medications they can be taking is usually restricted.

For example, patients with newly diagnosed, but untreated, diabetes, with no evidence of end organ damage, would be used to test a new antidiabetic agent.

Phase II trials tend to last only a few weeks to, at most, a few months. Initial Phase II trials (sometimes called, IIa) are pilot trials to determine dose range. They tend to be conducted at specialized centers, like university medical centers, by specialized investigators, such as medical school faculty.

[GRAPHIC OMITTED]

[GRAPHIC OMITTED]

Subsequent Phase II trials (often called, IIb) are aimed at elucidating dose response relationships, safety and, for the first time, efficacy, of the compound treating the disease or condition for which it is intended.

Drug-drug interactions are also studied carefully during Phase II as well as Pk and Pd in diseased patients, which can sometimes differ markedly from what was observed in healthy volunteers.

Phase II can encompass anywhere from a few to 20 or more clinical trials, and the "development plug" can be pulled--and frequently is--after any of them. Once again, assuming the drug shows sufficient evidence of efficacy and no major safety concerns--whether purely from drug effect, or from drug-drug interactions--a go/no go decision will be made to proceed to Phase III.

[GRAPHIC OMITTED]

[GRAPHIC OMITTED]

Phase III is where the "rubber meets the road." At least two pivotal Phase III trials demonstrating efficacy and safety in large numbers of patients, including special populations with all forms of the disease or condition to be treated, who may be on multiple other medications, are required for regulatory approval in the U.S.

Few drugs have been approved with data from less than two pivotal trials, and, if so, generally require post-marketing commitments to ensure that safety and efficacy is validated after marketing.

These trials are randomized, usually placebo-controlled (unless it would be unethical to use a placebo), and often involve an active comparator. They are conducted by less specialized investigators in countries all over the world. Thousands of patients may be enrolled and trials can cost a sponsor $50 million to $100 million each.

In addition to the two successful pivotal Phase III trials needed before an NDA can be filed, numerous additional special trials are usually demanded by regulatory agencies throughout the course of the IND clinical development period encompassing Phases I through III. A few examples of special trials would be to evaluate:

Special populations

* Renal insufficiency

* Hepatic insufficiency

* Elderly vs. young

* Lactating women

Interactions

* Food or liquids

* Drugs used in same indication

* Drugs interfering with metabolism or protein binding

* Drugs or substances modifying pharmacodynamic response (e.g., alcohol, sedatives)

* Drugs or substances which prolong cardiac repolarization, i.e., QT interval (currently an FDA "hot button" after withdrawal of Seldane[R] and Propulsid[R] for safety concerns)

Special conditions

* Effects on driving automobiles or operating machinery

* Effects on performing activities requiring alertness or concentration

* Effects on psychometric or psychological testing

* Effects of abrupt drug withdrawal

Special toxicities

* Ocular

* Ototoxicity 

* Rhabdomyolysis 

* Allergy/Anaphylaxis

* Hormonal (e.g., prolactin)

* Cardiovascular (QT prolongation)

Addiction potential

If the pivotal trials prove efficacy (usually by meeting or exceeding a predefined statistical "p-value" for a primary efficacy endpoint) and safety, and none of the special trials requested by regulatory agencies uncovers any serious problems, then all data--pre-clinical and clinical--is compiled into an NDA for submission to regulatory agencies.

The NDA includes an integrated summary of efficacy (ISE) and of safety (ISS). It is not unusual for an NDA to run several hundred thousand pages and be delivered to the FDA for regulatory review in one or more large trucks.

When evaluating NDAs, regulatory agencies look at:

* Validity of pivotal studies

* Replicability of pivotal studies (consistency across studies)

* Generalizability across populations (demographic groups, concomitant medications, intercurrent diseases, geographic regions, and even cultural groups)

* Establishment of supportable dosage and dose regimen(s)

* Clinical relevance of efficacy results

* Clinical seriousness of safety profile (in context of seriousness of condition being treated)

* Overall usefulness of drug (risk/benefit ratio)

In the U.S., the FDA does not actually approve the drug itself for sale. It approves the labeling--the package insert. United States law requires truth in labeling, and the FDA assures that a drug claimed to be safe and effective for treatment of a specified disease or condition has, in fact, been proven to be so. All prescription drugs must have labeling, and without proof of the truth of its label, a drug may not be sold in the United States.

[GRAPHIC OMITTED]

[GRAPHIC OMITTED]

The FDA takes, on average, about a year to review a typical, non-expedited NDA, give or take a few months. It may approve the proposed labeling, approve modified labeling, send the sponsor back to conduct additional special, or even pivotal trials, or may refuse approval outright (though, usually it will warn the sponsor if that is likely, giving them an opportunity to withdraw the NDA).

Sometimes the FDA will give conditional approval but require additional post-marketing trials to answer specific additional efficacy or safety questions.

In addition to mandated conditional regulatory approval or post-marketing surveillance trials, other reasons sponsors may conduct post-marketing trials include:

* Comparing with competitors (prove non-inferiority or superiority)

* Widening population (pediatric)

* Changing formulation or dose regimen (antihypertensive-diuretic combination, or new extended release formulation of already marketed compounds)

* Applying a label extension (such as expanding indication, e.g., Paxil[R] for obsessive compulsive disorder, Serafem[R] for premenstrual dysphoric disorder, Neurontin[R] for diabetic neuropathy)

Even when an NDA is approved unconditionally, regulatory scrutiny of a drug does not end. In most countries, yearly safety reports must be filed with the applicable regulatory agencies as long as a drug remains on the market, and these agencies independently monitor drug safety.

If safety concerns arise, the FDA may demand withdrawal of a drug from the market at any time (terfenadine {Seldane[R]}, cisapride {Propulsid[R]}, and cervistatin {Baycol[R]}, for example).

The author wishes to thank regulatory expert, Raymond Huml, MD, legal expert, Judith Beach, and communications advisor, Jay Johnson, for their valuable assistance in preparing this manuscript, and to PhRMA, Pharmaceutical Research and Manufacturers of America, for permission to use the "Stages of Drug Development" graphic.

By Ross Tonkens, MD

Ross Tonkens, MD, is global scientific head of the cardiovascular therapeutics division of Quintiles, an international pharmaceutical research organization.