Basic Science and Clinical Trials: Interview with William Paul, Director of the Office of AIDS Research


On February 14 AIDS TREATMENT NEWS interviewed William E.
Paul, M.D., Director of the Office of AIDS Research at the
U.S. National Institutes of Health (NIH). The new Office of
AIDS Research, due to Congressional action in 1993, has
budgetary authority over all AIDS research at NIH -- the
world's largest AIDS research program by far.

Dr. Paul has long proposed that we need more basic research
in AIDS -- leading to public fears of de-emphasis on clinical
research intended to find treatments now. We wanted this
interview to address these concerns.

We especially wanted to clarify "the role of small, rapid,
data-intensive, exploratory human trials of potential
treatments for HIV disease," as we wrote in a letter to Dr.
Paul in preparation for this interview. We most wanted to
learn whether this kind of science-intensive human trial is
included within the basic-research emphasis at the Office of
AIDS Research.

Basic Research and Human Trials

ATN: People in the community have the image that "back to
basic science" means retreating into the laboratory and then
in five, ten, or 15 years coming out with something useful.
What can you say to reassure them?

Paul: That is a valid concern; people are right to raise that
point. You may remember the article in NATURE on May 12,
1994, by the late Dr. Bernard Fields, that said "back to
basics." Our position has always been that we do need to
build a knowledge base, but that cannot be done at the
expense of efforts to do therapeutic research now. We are
searching for a balance. In our opinion the balance shifted
too heavily away form "basic" research.

There is great concern throughout the HIV research community,
and certainly among patients and their advocates, that the
drug pipeline is not very full. People are worried, very
rightly, that after the protease inhibitors they do not see a
good picture. Our position is that the only way to fill that
pipeline is through this kind of scientific research.

People differ in what they mean by basic research. I was
struck by one of the proposals in your letter, in which you
described what I would call the clinical investigation of
people who are on research protocols using one or more agents
as therapies, coupled with very intensive laboratory study.
We know the very recent and successful example of using this
approach to gain important knowledge -- the two papers, one
by David Ho and colleagues, and the other by George Shaw's
group. This certainly qualifies as the kind of basic research
I am talking about.

We do not mean let's abandon everything we have done in HIV
and go back to square one. What we mean is that we need to
assure a balanced approach aimed at understanding the
mechanisms of this disease. Some of that will be done by
laboratory-based basic research on model systems, and we need
to support that. But [human] work like that in the Shaw and
Ho papers, and I think like that which you pointed out in
your letter, clearly falls under our mandate. We feel very
enthusiastic about that kind of work.

Our position is not an abandonment of therapeutic research
that benefits people who are sick now, but rather an effort
to build a better knowledge base, that will help us fill up
the therapeutic pipeline.

ATN: Will your Office of AIDS Research make sure that this
kind of work is done?

Paul: At OAR our responsibilities are not to choose the
individual trials; that is the responsibility of the
individual Institutes [of the U.S. National Institutes of
Health], such as the National Institute of Allergy and
Infectious Diseases (NIAID), and the National Cancer
Institute (NCI). Many people have been concerned about "the
extra layer of bureaucracy" that OAR introduced. We regard
ourselves as having a very valuable function in setting
overall goals, in conducting a clear evaluation and assuring
that precious funds for AIDS research are allocated wisely
and used effectively; but we do not intend to get into
micromanaging specific trials, which lies in the purview of
the Institutes. Instead, our job is to make sure that there
are programs in place that allow this kind of really terrific
work to go forward. We have been asking how we can facilitate
the kind of research you described, what mechanisms are
available?

There are two classes of mechanisms which I believe would be
suitable. Particularly with the upcoming recompetition of the
ACTG in 1996 [NIAID's AIDS Clinical Trials Group, which has
done most of the government's large studies of AIDS
treatments], there is going to be a sea change, I think, in
the recognition that much more must be done at the level of
these smaller, laboratory-intensive, pathogenesis trials.
[Note: The recompetition will involve a new selection of ACTG
sites, on the basis of competitive review; some sites may be
discontinued, and some new sites may be added. In addition, a
reorganization of the ACTG management took effect on January
1, 1995.] Dr. Chip Schooley is committed to that; and I
believe the Institute (NIAID) is committed to that as well.
So the reorganized ACTG is clearly one mechanism through
which that kind of trial could be supported.

The second mechanism is conventional grant support for
research. Most often [the data-intensive trials] will require
a very sophisticated laboratory in order to be carried out.
The people with those facilities are in a very strong
position to obtain conventional grant support for that kind
of work. We argue that both strategies [ACTG support, and
conventional grant support] should be applied; these seem
like the two most flexible approaches to be certain that this
kind of clinical work can be done, well and thoughtfully, and
that it has been peer reviewed. One could imagine an infinite
number of such studies; obviously there must be a way to
choose thoughtfully and well those most likely to provide
really useful information. Whoever wants to propose such a
program needs to be able to defend why that experiment should
be done. There must be a scientific justification that
suggests that this is a wise way to use these funds. There
will always be more proposals than we have money to support.

The new, recompeted ACTG will have a much greater emphasis on
this laboratory-intensive clinical trial. It will have an
internal mechanism for more rapid consideration of proposed
trials. The advantage of ACTG support over regular grants is
that it is somewhat faster. But the disadvantage is that it
is largely limited to people already in the ACTG system. We
do not want to close out others; we need opportunities for
people from outside to do this work.

These are the two approaches which I think are most likely to
give people the flexibility to carry out innovative work that
will have a message. Everyone was excited by the Shaw and Ho
papers that made clear certain things about the dynamics of
the infection that we had not understood. These very incisive
pieces of work were not exceedingly expensive to do, although
obviously the investigators put in a lot of effort. These
represent thoughtful experiments, and even more, the
recognition of what the result meant. Often that recognition
is as important as anything else.

ATN: Although we have heard concern about the amount of
enthusiasm for that work, in that it leaves aside the whole
immunological area.

Paul: Is their report the end? No, there is much more work to
be done; and many of us who are immunologically inclined feel
that we still do not understand in detail the T-cell
dynamics. The viral dynamics are very well specified in those
papers. The T-cell dynamics are more complex, and there is
more work that needs to be done.

This kind of work is what we would call a clinical
investigation. You carry out an ethically sound experiment on
a human. It is ethically sound in that you use something that
potentially can do the individual some good. The treatment
perturbs them, and the perturbation gives you the opportunity
to watch what the immune system or viral system does, in
response to a known perturbation. Because we do not have the
animal models we need, this work in people is essential. I am
a great advocate of that.

ATN: That is exactly the point I wanted this interview to
make.

Paul: We agree entirely on that point.


Funding Scientist-Initiated Work

ATN: What about the big problem of funding? Fewer and fewer
investigator-initiated grants [which fund work proposed by
the scientists who will do the work, usually in academic
centers, instead of contracts proposed by government
scientists] can be funded because of lack of money.

Paul: That is an issue I addressed directly recently in
SCIENCE ["Reexamining AIDS Research Priorities," SCIENCE,
February 3, 1995, pages 633-636]. One thing the OAR clearly
can do is to recognize where a problem exists and take steps
to change it.

Let me give you some numbers. You can calculate what
proportion of a budget is used to support what I call
investigator-initiated research. In discussions with the
Division of Research Grants at NIH, we have agreed on the
following definition. Traditional grant applications (of
which the "R-01" is most well known, but there are also
several other similar types) are regarded as investigator-
initiated if they are not submitted in response to a "Request
for Applications" (RFA). We took the amount of money that was
spent in all of fiscal year 1994 supporting that category of
grant, and divided it by the total amount of money to support
grants and contracts, to obtain the proportion of funding
spent on unsolicited investigator-initiated grants.

For non-AIDS grants at NIH, that proportion is about 54
percent. For AIDS, it is less than 25 percent. So right now
there is a disconnect. The key now is to make an effort to
change that ratio. We can do a lot by changing that ratio.

Obviously to change the ratio you have to alter something. As
a first approximation, the kinds of things we would like to
see altered are, for example, the heavy use of RFAs and
contracts in the Institutes. Some are justified; but they
need to be looked at again, because programs go on, and
sometimes they outlive their usefulness. Sometimes they are
theoretically useful -- but compared to what? There is very
little done that does not serve a useful purpose; but when
funds are limited, useful purpose has to be graded more
toughly. You need to compare it to what else could be done
with these resources. We regard this kind of analysis as a
very important role for OAR in the system.

Animal Models

ATN: A number of people think that SIV (simian
immunodeficiency virus) in macaque monkeys is a very good
model for studying HIV disease. What do you think?

Paul: I agree with that view. There is little doubt that in
broad outline, the macaque disease resembles the human
disease. The pathogenesis seems to follow very similarly: you
have the initial burst of viremia, followed by a rapid
immunological control and low virus titers; but eventually
the animals get disease. There is the same kind of trapping
of virus in the germinal centers on follicular dendritic
cells. In broad outline the diseases look very similar. Much
can be learned from the monkeys about pathogenesis and
immunopathogenesis. I pointed out in my SCIENCE article that
we believe this area has real promise. We would like to see
much more of such work done.

There is also another model which is not as close but should
not be fully excluded. There is a disease in cats, caused by
a related lentivirus called the feline immunodeficiency virus
(FIV). While it is further removed from HIV, it has
advantages because it is easier to work with cats than with
monkeys. This is another area where we should not miss the
opportunity presented.

Some AIDS research would be very difficult to do in humans --
for example, all of the attenuated virus vaccine work [which
involves injecting a live but weakened virus into uninfected
animals, as a preventive vaccine]. We would be exceedingly
reluctant to begin that work in humans. But there is very
little doubt that a robust vaccine response can be induced in
the macaques. At the very least we can learn the nature of
the immunity that really is protective. Even if we finally
reject this strategy for a human vaccine, we may still
benefit enormously from such studies in animals.

Specific Issues: Management Analysis; Hydroxyurea

ATN: One suggestion and critique of Federal AIDS research
says that NIH should use the services of professional
management consultants, who analyze large industrial
enterprises to find what is working and what is not, and how
to improve the system. The suggestion is to bring in these
experts and see what they can suggest for making procedures
work more smoothly than in the past.

Paul: I don't have any antipathy to systems management, etc.
But my own view is that this is a misreading of where we are
at, and what our task is. When industry embarks on a research
program, it is usually quite definite in what it wants to
achieve. A pharmaceutical company, for example, will have
identified a series of molecular targets, and then try to
build a drug that will react with a particular target. It
uses a very clear, precise, and also narrow but deep
approach; it wants to get a drug, and the question is what is
the best way to go about it.

We are faced with a different problem, because we don't
already have the knowledge to accomplish our goals. I do not
believe that the systems engineers have grappled with this
kind of problem. In general, American industrial research has
been very good in the development stage, but it is not known
for its discovery. We are dealing with discovery, with
research; I do not believe that there is enormous experience
in the planning of the acquisition of new knowledge -- in
contrast to the planning of the utilization of knowledge to
create some product or technology. I know there is a group
that presses this view very heavily; but I am not in
agreement with their views that this is fundamentally a
management problem, that the reason we have not solved AIDS
is that we have not had the right managers. We have not
solved AIDS because we have not made the right scientific
progress, because it is a problem, unfortunately, of real
complexity. I do not believe that what we are facing right
now is a crisis in management.

ATN: One example of a needed change is illustrated by the
problems we have had over the last year and a half in getting
any U.S. study done on hydroxyurea. Today there is tremendous
interest in doing a trial, but it has been a year and a half
since hydroxyurea came out publicly in Dr. Robert Gallo's
plenary talk at the Berlin conference. We have not had any
trial in the U.S. in that time, not even with a few people to
get a sense of where we are; and now people are starting to
use the drug without the benefit of those trials. What can be
done to prevent this kind of problem in the future?

Paul: I am aware that hydroxyurea has been widely used in
therapy for several other diseases, including some forms of
cancers, and most recently was shown to be effective for
sickle cell anemia. Dr. Gallo reported his findings in
SCIENCE in November of 1994, suggesting that hydroxyurea
appears to be a possible candidate for AIDS therapy. In fact,
I understand that the ACTG is currently reviewing the data
for the development of a treatment protocol within the ACTG.

The Mission of the Office of AIDS Research

ATN: People ask what is the justification for the new Office
of AIDS Research. I hear from your answers that OAR should
make sure the proper ways of decision making are in place,
should increase the proportion of funds going to
investigator-initiated research vs. government-requested
contracts, should deal with such management without
micromanaging by deciding on individual trials.

Paul: I would go beyond that in one respect. The OAR's
responsibility is to take into account that research on AIDS
is involved in enormously broad and diverse areas. We use a
planning process with much outside input -- with many
scientists from outside NIH, with scientists from within NIH,
and with community representatives.

There is AIDS research support at NIH in all the Institutes.
There must be an overall vision for what we are going to do.
There must be some process through which we decide, over
time, where we put our emphasis. It is not enough to simply
let the whole budget go up or go down in exactly the same
lockstep without any thought in it. That is the opposite of
what we want to achieve.

As long as the planning process occurs only within the
Institutes, what individual Institutes will be able to do is
simply to look after their own areas. They cannot influence
the balance of research across the areas. So one very
important role for OAR, in a process as diverse as AIDS
research, is to recognize where the opportunities lie, where
the greatest hopes are, and to try to put the resources
there.