by Jeffrey P. Cohn
FDA Consumer special report
The scene is a typical one. A patient,
perhaps you or I, goes to a doctor and gets a prescription. Then a
pharmacist fills the prescription, with instructions to take the drug in
the prescribed amount and manner over the following days, weeks or
months. This scene is repeated millions of times across this country
every day–some 2 billion prescriptions are filled every year in the
United States. In fact, the process is so commonplace that the pills,
tablets, capsules, and other medications that virtually every one of us
relies on to restore or maintain good health at some point in our lives
come to be taken for granted.
Yet these drugs–and the improved
quality of health they bring to the American people–are truly
“miracles of modern science.” In fact, the process for
discovering, developing and testing new drugs encompasses some of the
most exciting areas of scientific discovery today. The endeavor runs the
gamut from basic biomedical investigation of living cells and molecules
to applied research that yields new consumer products to improve health
The Cutting Edge
“We are on the cutting edge of the
biological sciences,” says Rhoda Gruen, Ph.D., special assistant to the
president of international research and development at Hoffmann-La
Roche, Inc., a pharmaceutical research and manufacturing firm
headquartered in Nutley, N.J. “We suck up new information like a
sponge. Everything we do is subject to change as new scientific
information becomes known.”
The research process is a complicated,
time-consuming, and costly one whose end result is never known at the
outset. Discovering a new drug has been likened to searching for the
proverbial needle in a haystack. Literally hundreds and sometimes
thousands of chemical compounds must be made and tested to find one that
can achieve the desirable result without too-serious side effects.
The complexity of the process can be
gauged, in part, by the diversity of scientific disciplines engaged in
finding new drugs. Traditional organic chemists, physiologists and
statisticians have been joined in recent years by new kinds of
specialists. Biochemists study the chemistry of life processes.
Molecular biologists study the molecules that make up living matter.
Toxicologists investigate chemicals’ potential for harm. Pharmacologists
look at how drugs work. And computer scientists apply the power of their
sophisticated machines to analyze and assess new chemicals. Each
provides a different way of looking for that needle.
Such a complicated process costs vast
amounts of time and money. FDA estimates that, on average, it takes
eight and a half years to study and test a new drug before the agency
can approve it for the general public. That includes early laboratory
and animal testing, as well as later clinical trials using human
Drug companies spend $359 million, on
average, to develop a new drug, according to a 1993 report by the
Congressional Office of Technology Assessment. A company such as
Hoffmann-La Roche, whose annual sales in the United States alone are
about $3 billion, spends about $1 billion a year on research worldwide.
Building on Good Science
There is no standard route by which the
thousands of drugs now sold in the United States were developed. ”Each
drug has its own way of being born,” says Clement Stone, a former
senior vice president for Merck and Co. Inc. research laboratories, West
Point, Pa. “Often we consciously search for a drug for a specific
use, but more often it is serendipity. What is required, though, is good
science building on good science.”
In some cases, a pharmaceutical company
decides to develop a new drug aimed at a specific disease or medical
condition. In others, company scientists may be free to pursue an
interesting or promising line of research. And, in yet others, new
findings from university, government or other laboratories may point the
way for drug companies to follow in their own research.
Indeed, the process typically combines
elements of all three avenues. New drug research starts by studying how
the body functions, both normally and abnormally, at its most basic
levels, Ronald Kuntzman, vice president for research and development at
Hoffmann-La Roche, says. The pertinent question is: “If I change it
[the body’s functioning], will I have a useful drug?” That, in
turn, leads to a concept of how a drug might be used to prevent, cure or
treat a disease or medical condition. Once the concept has been
developed, the researcher has a target to aim for, Kuntzman adds.
Gruen elaborates: “Disease
processes are complex and involve a sequence of events. If you want to
intervene in the disease process, you try to break it down into its
component parts. You then analyze those parts to find out what abnormal
events are occurring at the cellular and molecular levels. You would
then select a particular step as a target for drug development with the
aim of correcting the cellular or molecular dysfunction.”
A Cholesterol Drug
Take cholesterol, a wax-like substance
found naturally in the body. Too much cholesterol, either naturally or
in the diet, can cause it to build up on the inside walls of blood
vessels. This can clog the arteries that deliver blood to the heart
muscle, blocking the flow of oxygen and nutrients, causing a heart
There have been few drugs that
effectively cut cholesterol levels without either toxic or unpleasant
side effects. This has limited their use. Others that were tested acted
too late in the process by which the body makes cholesterol to lower its
levels. What was needed, says Eve Slater, M.D., executive vice president
for worldwide regulatory affairs for Merck, was a drug that would act
earlier in the cholesterol-making process.
To find one, scientists at Merck and
elsewhere spent decades studying how the body makes and uses
cholesterol. Along the way, they identified more than 20 biochemical
reactions necessary for the body to make cholesterol, along with the
enzymes required at each step to turn one chemical into the next one in
The research problem, Slater says, was
to find the step where interference by a drug would effectively lower
cholesterol production. By the 1970s, scientists had found a
possibility. They had isolated a chemical, mevalonic acid, that was an
early link in the cholesterol chain and an enzyme called HMG-CoA
reductase that produced mevalonic acid.
What was needed, then, was a drug that
could either inhibit HMG-CoA reductase or prevent cells from correctly
using the enzyme.
Sometimes scientists are lucky and find
the right compound quickly. More often, Gruen says, hundreds or even
thousands must be tested. In a series of test tube experiments called
assays, compounds are added one at a time to enzymes, cell cultures, or
cellular substances grown in a laboratory. The goal is to find which
additions show some chemical effect. Some may not work well, but may
hint at ways of changing the compound’s chemical structure to improve
its performance. The latter process alone may require testing dozens or
hundreds of compounds.
A more high-tech approach is to use
computers to simulate an enzyme or other drug target and to design
chemical structures that might work against it. Enzymes work when they
attach to the correct site on a cell’s membrane. A computer can show
scientists what the receptor site looks like and how one might tailor a
compound to block an enzyme from attaching there.
Nevertheless, “computers give
chemists clues to which compounds to make, but they don’t give any final
answers,” says Kuntzman. “You still have to put any compound you
made based on a computer [simulation] into a biological system to see if
Yet a third approach involves testing
compounds made naturally by microscopic organisms. Candidates include
fungi, viruses and molds, such as those that led to penicillin and other
antibiotics. Scientists grow the microorganisms in what they call a
fermentation broth, one type of organism per broth. Sometimes 100,000 or
more broths are tested to see whether any compound made by a
microorganism has a desirable effect.
In the search for a new cholesterol
drug, scientists found a fungus that inhibited the HMG-CoA reductase
enzyme in a test tube. Chemists then had to identify which of the
fungus’ dozens of chemical byproducts was actually inhibiting the
enzyme. Once that was done, the chemical’s structure was analyzed and
improved on to enhance its effects.
To this point, the search for a new
drug has been confined to a laboratory test tube. Next, scientists have
to test those compounds that have shown at least some desired effects in
living animals. ”We have to find what the drug is doing on the down
side,” Kuntzman explains.
In animal testing, Kuntzman says, drug
companies make every effort to use as few animals as possible and to
ensure their humane and proper care. Two or more species are typically
tested, since a drug may affect one differently from another. Such tests
show whether a potential drug has toxic side effects and what its safety
is at different doses. The results “point the way for human testing
and, much later, product labeling,” Kuntzman says.
So far, research has aimed at
discovering what a drug does to the body. Now, it must also find out
what the body does to the drug. So, in animal testing, scientists
measure how much of a drug is absorbed into the blood, how it is broken
down chemically in the body, the toxicity of its breakdown products
(metabolites), and how quickly the drug and its metabolites are excreted
from the body. Sometimes such tests find a metabolite that is more
effective than the drug originally picked for development.
Of particular concern is how much of
the drug is absorbed into the blood. “If a drug’s active
ingredients don’t get into the blood,” Kuntzman says, ”it won’t
work.” Scientists may add other chemicals to the drug to help the body
absorb it or, on the other side, to prevent it from being broken down
and excreted too soon. Such changes in the drug’s structure mean even
Absorption rates can cause a host of
problems. For example, for a certain drug to be effective, 75 percent of
it may need to reach the bloodstream. But absorption rates can vary
among individuals from, say, 10 to 80 percent. So, the drug must be able
to produce the desired effects in those who absorb only 10 percent, but
not cause intolerable side effects in people who absorb 80 percent.
”If we can improve the absorption rate
we can reduce the variation in what real dosages people would be subject
to,” Kuntzman says. A more standard absorption rate for all
individuals, say around 75 to 80 percent, would mean that the dose could
be reduced and still have the desired effects.
The Wrong Road
By this time in the testing process,
many drugs that had seemed promising have fallen by the wayside. More
often than many scientists care to admit, researchers have to just give
up when a drug is poorly absorbed, is unsafe, or simply doesn’t work.
“In research you have to know when to cut your losses if you are
going down a wrong road,” says Clement Stone. And, he adds, there are
many more wrong roads than right ones.
Nevertheless, progress may yet be made.
Occasionally, Stone says, a stubborn scientist keeps looking and finds a
usable compound after others had given up. In other cases, compounds may
be put aside because they failed to work on one disease, only to be
taken off the shelf years later and found to work on another.
Such was the case with Retrovir (zidovudine,
also known as AZT), the first drug approved for treatment of AIDS. The
drug was first studied in 1964 as an anti-cancer drug, but it showed
little promise. It was not until the 1980s, when desperate searches
began for a way to treat victims of the AIDS virus, that scientists at
Burroughs Wellcome Co., of Research Triangle Park, N.C., took another
look at zidovudine. After it showed very positive results in human
testing, it was approved by FDA in March 1987.
Even so, ”a minuscule number of drugs
we test ever reach testing in man,” says Richard Salvador, Ph.D., a
Hoffmann-La Roche vice president and international director of
preclinical development. The Pharmaceutical Research and Manufacturers
of America organization estimates that only five in 5,000 compounds that
enter preclinical testing make it to human testing, and only one of
those five may be safe and effective enough to reach pharmacy shelves.
The role of FDA in the early stages of
drug research is small. The Federal Food, Drug, and Cosmetic Act
requires FDA to ensure that the new drugs developed by pharmaceutical
companies are safe and effective. It does not give the agency
responsibility to develop new drugs itself. So, FDA physicians,
scientists and other staff review test results submitted by drug
developers. The purpose: to determine whether the drug is safe enough to
test in humans and, if so–after all human testing is completed–to
decide whether the drug can be sold to the public and what its label
should say about directions for use, side effects, warnings, and the
FDA first becomes involved when a drug
company has completed its testing in animals and is ready to test a drug
on humans. (Actually, some animal testing continues after human tests
begin to learn whether long-term use of the drug may cause cancer or
birth defects. Also, more animal data may be needed if human tests turn
up unexpected effects. And new therapeutic uses may be found by
continued animal studies.)
Although FDA usually does not tell drug
companies what specific laboratory or animal tests to run, the agency
does have regulations and guidelines on the kinds of results FDA expects
to see in any request to conduct human testing.
And the drug companies listen to those
signals. Both Hoffmann-La Roche’s Kuntzman and Merck’s Stone say their
companies follow and sometimes exceed FDA’s guidelines. ”We want to
optimize our chances of taking a compound from animal to human
testing,” Stone says.
So drug research is a long, difficult
and costly road, certainly. But sometimes the hard work, the scientific
sleuthing, and the time and dollars spent pay off. Such was the case in
December 1992, when FDA approved in five months Taxol for treatment of
advanced cases of ovarian cancer. Taxol is an important second-stage
drug for ovarian cancer because, while most patients respond to
chemotherapy initially, the disease often recurs.
But to scientists like Kuntzman, drug
research goes even beyond preventing or curing disease or making money.
It is also a tool for finding out more about the human body and its
basic life processes.
“Research is an evolutionary
process,” Kuntzman says. ”You change studies and use experiments
to lead to other experiments. As you go along you may not even see the
connection between studies. In a sense, research has no end. The only
end would be when we understand everything there is to know about the
human body. I expect that we will never know enough about the body.”
Merck’s Slater agrees. ”We can make
progress,” she says, ”but we are unlikely to achieve perfection.” In
the end, that is what researching and developing new drugs is all
about–understanding and progress.
Jeffrey P. Cohn is a free-lance
writer in Washington, D.C.