Interview with Dr. Robert Gallo

This is the second oral history interview with Dr. Robert Gallo of the National Cancer Institute about the history of AIDS at the National Institutes of Health. The date is 4 November 1994. The interviewers are Dr. Victoria A. Harden, Director, NIH Historical Office, and Dennis Rodrigues, program analyst, NIH Historical Office.

Harden: Dr. Gallo, when we ended the first interview, we had set the stage for the discussion of AIDS. We had talked about when [Dr. James] Jim Curran [of the Centers for Disease Control and Prevention] came to the NIH [National Institutes of Health] and was prodding you to go into AIDS research. Much of your early work has been detailed in many different places–in your book and in a variety of other publications–so what we would like to do in this interview is to have a few points amplified, not to attempt to recount all the facts.

One of the key questions that has come up over and over again is how, when a new disease appears, can it be demonstrated that a particular agent is the cause of it? Chronologically, the French isolated their virus, LAV, in 1983, but they did not demonstrate conclusively that there was a causal link between their virus and AIDS. You waited until May 1984, and then published four papers in Science to do this. In fact, you wrote to [Dr.] Jean-Claude Chermann noting that you wanted to wait to publish in order to obtain a certain number of papers to establish the etiology. Why did it take four papers to establish it and what particular points were you trying to make with those papers?

Gallo: That is a good question. First, it did not take four papers; that was just the number that Science accepted. It is a large number obviously. We wanted to get the maximum amount of data published in as rapid a period of time in the most visible journal that we could. In fact, we sent a fifth paper with more antibody testing data in it, at almost exactly the same time, to The Lancet. It was a paper by [Dr. Bijan] Safai, a clinical collaborator, myself, and my colleagues, in which there was 100 percent accuracy in blind testing of patients with AIDS for antibodies, the tell-tale sign of the infection. Let me just say that there is nothing magical about having four, five, or six papers. There was a lot more data from many other collaborators, including the CDC, that we did not include in those five papers, but that we had in hand and were ready to write up in subsequent papers.

The question you asked is interesting. Jon Cohen of Science asked me the same thing following his long interviews with and questioning of Peter Duesberg. That article will come out in Science within a week or two. So I have had a chance to think about that question again. Jon said to me, “Obviously things pointed to the cause steadily thereafter, but how did you know that soon?" The answer is something like this. Somebody like Duesberg focuses on the .01 percent uncertainty, or the 0.1 percent uncertainty, but most scientific answers that you obtain related to a human disease–and very often in many aspects of science–are never 100 percent certain. If, let us say, you were 99–I think I was maybe 99.9–percent sure that the virus was the cause of AIDS, you would have to act.

The alternatives were not good. I felt as confident as I had ever felt about anything at the time, and the reasons were as follows:

First of all, this was a new kind of virus. That was already established in the paper from the Pasteur Institute in 1983. We could certainly confirm, and extend our knowledge, by greater characterization of the virus, that this kind of virus was new. It certainly was not HTLV-I. It was not HTLV-II, and, if it was related to them, it was distantly related.

Second, we demonstrated that it the virus targeted the CD4+ T cell, and we learned from clinical people that AIDS apparently was a disease chiefly of CD4+ cells. This fit.

Third, it fit with our hypothesis that a variant of a certain kind of retrovirus might produce immune deficiency. This was based on the feline leukemia virus model in early work from Glasgow by the Jarretts [Drs. William and Oswald], and work by [Dr. Myron] Essex and others in Boston that modification of feline leukemia virus could produce an immune deficiency. We were thinking that this was a retrovirus that might either be distantly related, or more closely related, to the HTLVs, and there was precedent for modifications of such viruses being able to cause immune deficiency. Along those lines, we knew by 1984, from the epidemiologists, that the disease was transmitted by blood, by sex, and from mother to child. We knew that retroviruses were, in general, often transmitted that way in animals, and certainly we knew that occurred for HTLVs.

Fourth, we obtained a very large number of short-term isolates of the virus–let us call them detections–or what I sometimes would call a true isolate, where we could transmit the virus, as well, or keep it going for a while, in association with AIDS. The French group's paper was on a patient with lymph gland enlargement. That patient did not have AIDS. He did not develop AIDS until many years later.

What we were trying to do was establish the linkage of the virus to AIDS itself. So, after the first paper, the second paper, which I co-authored, was on establishing frequent isolation, or short-term culture, of the virus in a great number of people with AIDS or the symptoms before AIDS, which at that time was called ARC (AIDS-related complex). We published on 48 such detections or isolates, and that gave us a lot of confidence that we were finding it in many people with AIDS. Not 100 percent of the people, but that is easy to explain with the technology at the time. Yet, in healthy donors we could not isolate the virus, and a tremendous number of attempts to do so were made in a variety of blood cells from a variety of normal donors. I cannot remember if we included donors with other diseases, but the association with AIDS was dramatic.

Fifth, or whatever number I am up to, we developed a blood test–I will not go through the technology that led to that, perhaps we have discussed that already–but we established a very sensitive and specific blood test in which we brought the Western blot into clinical medicine for the first time. We based it on an ELISA and a Western blot and we based it on the ability to mass produce the virus in continuous culture. We knew, although obviously, in retrospect, the public health people were not aware of this, that antibodies meant active replicating virus. In many other viral diseases if antibodies are present, you may simply have been exposed to a virus, but in animal retrovirology and HTLV, we knew that the presence of antibody for any length of time definitely meant infection with a retrovirus. By the time you mount an immune response, you have integrated the viral gene, so you are infected, and when you keep antibodies at a moderately high level, you are replicating virus. So we knew that the presence of]antibodies was equivalent to meaning that there was active replicating virus. I repeat that some of the public health people did not know this. I remember one of the first difficulties I got into in debating the cause of AIDS was when an epidemiologist on the West Coast said, “We do not have any idea what is the meaning of the antibodies.” I responded, “You may not, but we do.” It was important and it meant active infection.

What then did that antibody test tell us? It told us that 90 percent–at the time 88.9, 89 percent, then in the Safai study, we already had the data of 100 percent from blind studies–that if you had AIDS, you had antibody, and you were infected with this virus, whereas if you looked at the healthy heterosexual population antibody was very rare. I think in the total sera we had–remember we only published part of the data in Science because of authorship problems–people forget the reality that so many people want and deserve first authorship and we had to divide the papers–but all together we were holding in our hands, I suspect, somewhere upward of 1,000 to 2,000 tests of healthy heterosexuals in which antibody was rare. I think it was only once or twice that we found antibody. I do not want to make this an official record of the exact number of test results that we had, but it was large.

Now, in the midst of this research, we looked at what were known as the four “H's” at the time–people who were having a high incidence of AIDS–homosexuals, hemophiliacs, heroin addicts, and Haitians. In collaboration with a wide variety of groups, we had shown and we had the data in front of us–part of which was published in the Science papers, part of which was published later–that all these groups had a very significant percentage of people positive for this new virus, in contrast to people not in the risk groups at the time, where it was rare. This was perfectly compatible.

Sixth, we had data we did not publish, from collaboration with the CDC. I cannot remember exactly when we got this data, but it was early. The data was that blood transfused from persons who had AIDS–I think the numbers were small, but it was like six out of six–were all positive. People without AIDS who received blood were negative. But what was more interesting, the people who received blood who were positive–who had AIDS–had one or more donors who were positive and who had AIDS, whereas if you looked at the blood donors in general it was unusual to find a positive. This was rather dramatic.

Finally, we had data that in a number of countries, where we looked at some sera that were older, before AIDS was present, and with the antibody negativity but in the same risk groups and the same people with the disease AIDS, we had positive serology. We were preparing papers on this for publication either at that time or soon thereafter. I cannot give you a precise chronology, but I remember [Dr.] Jörg Shüpbach, who was a key postdoctoral fellow with me from Zurich, Switzerland, had a great deal to do with the introduction of the Western blot into clinical medicine while here. He had obtained serum from Switzerland and there were many people developing AIDS in Switzerland. That paper was subsequently published in the New England Journal of Medicine. We were looking at a number of countries and collaborating with many people and things consistently fit.

I could not conceive of any other data that we would need in order to say that the virus was the cause of AIDS. What other data were there? To try putting it into an animal and seeing if the animal got AIDS? Something that many people do not appreciate is that viruses are often species specific. In fact, they usually are. Sometimes you get the right result. Take gibbon ape leukemia virus. You can put it in a young gibbon and it produces leukemia, and so everything seems to fit fine; but if you are a human, you cannot put the virus in a human so you go into another species. If gibbon ape leukemia virus is put into another monkey, it does not infect at all or, if it does, it does not produce leukemia. On the contrary, Herpes saimiri in its native species produces nothing. But if you put it into some other monkey–I cannot remember the type–you can produce a lymphoma. Doing that kind of experiment–putting a virus into another species–is not necessarily meaningful and many people do not understand that about viruses. I did not see what else could be done. What else was there to do to establish etiology until you saw more people dying? We felt that the blood test was, in fact, an emergency, and to say that this was the etiology of AIDS was the right thing, and that it was urgent to do so.

Harden: From a philosophical point of view, then, you were saying that if a person did not have the AIDS virus, he or she did not have AIDS?

Gallo: That is correct.

Harden: That was your bottom line?

Gallo: Absolutely.

Harden: You did not believe that you had to demonstrate that this virus would cause AIDS in some other species. You could not use humans obviously.

Gallo: Exactly. My position has been unchanged from the beginning to now. I think it has been misinterpreted here and there in news articles, but my position has not altered.

Harden: This brings me to the next question, which has to do with the critics of AIDS, beginning with Dr. Peter Duesberg, and recently there has been Robert Root Bernstein who has written a book called Rethinking AIDS. Also [Dr.] Luc Montagnier has suggested that there might be cofactors in the development of AIDS. Here we may be moving from etiology to pathogenesis. Then there are all the quacks and all sorts of....

Gallo: I am glad you said it.

Harden: I did say it, but what I am trying to get at is that there are a range of people, from the fringe to the mainstream, who are questioning the virus as the cause of AIDS.

Now, I believe it was in 1988, that you, Dr. [William] Blattner and Dr. [Howard] Temin published a position paper in Science in which this was all hashed out. It seemed to be a straightforward matter, but the debate is still going on six years later.

Gallo: Yes.

Harden: What is happening, and what impact do you think that all this controversy has on AIDS prevention?

continued on Page 02

Back To Top | Photography Credits