FRONTLINE presents Organ Farm
photo of thomas e. starzl
thomas e. starzl, m.d., phd home
four patients
animal welfare
the business
the regulators

A pioneer in the field of transplantation, Starzl performed the first successful liver allotransplant in 1967. In 1992 and 1993, while at the University of Pittsburgh Medical Center, he performed two unsuccessful baboon-to-human liver transplants. In this interview he describes his current research, which is focused on understanding how chimerism -- the coexistence of donor and recipient cells in a transplant patient -- can potentially help lead to organ acceptance, known as tolerance. (Interviewed Spring-Winter 2000.)

. . . Did you and your institute have a pioneer mindset? You've always been interested in . . . transplants.

A few months before we did that first case, which was in June, 1992, we studied a group of 30 patients who were the longest surviving liver recipients in the world, and also the longest surviving kidney recipients in the world. We wanted to know if those patients had donor leukocytes throughout their bodies and their skin, blood, lymph nodes, and in several of the cases, even in their hearts. So we did biopsies of the tissues in those 30 patients. And all of them had donor leukocytes disseminated throughout their bodies.

Using only that amount of information was a little like walking into an Alfred Hitchcock movie at the very end and figuring out the whole mystery that had preceded your arrival. For a variety of reasons, but mostly from experience, we were able to go right back to the time of transplantation, and derive a hypothesis about what had happened then that subsequently would lead to the presence of these cells 30 or more years later. . . .

The long and short of it is that the donor leukocytes that are distributed, first induce, and then exhaust an immune reaction by the recipient. And that is the basis for allograft acceptance. The kind of microorganisms that induce that kind of a response -- and it is the same as the response that allografts induce -- apply only to very specific situations. In infectious disease, it applies only to microorganisms that do not immediately kill the host tissues themselves. The allograft is being recognized by the recipient as a non-psychopathic microorganism -- in other words, as a rather benign infection.

. . . On the other hand, the xenograft is recognized by a different kind of highly toxic microorganisms . . . that actually kill the host cells. . . . There is a very direct linkage between that kind of microorganism and the xenograft, because both these microorganisms and the xenograft possess this alpha gel antigen, which has been construed to be the main barrier to xenotransplantation. So that brings us to the stone wall that everybody is facing: what to do about that target, which induces what is known as an innate immune reaction? It's quite different than the one that allografts or the more benign infectious regions induce.

When the graft is in place . . . what kind of migration of cells, both good and bad, occur that can both assist the transplant and may also raise problems?

That's a beautiful way of putting it, because the migratory cells constitute a two-edged sword. The purpose of that migration and all that follows is not the acceptance of allografts. The immune systems didn't evolve these protective mechanisms to frustrate transplant surgeons, or to help them, either. It was to deal with this particular kind of infection. . . .

Now you've got an immune reaction of a very special kind, and the question is, how is it turned off? One way it can be turned off is simply removal of the antigen. And removal of the antigen really involves removal not only of the graft from which the cells have flared, but also those cells that have peripheralized.

There is insufficient money from government sources to really finance this kind of research.  It has come to depend very largely on venture capital, and so then you have the additional pressure of fiscal expectations. The other mechanism of switching off this kind of immune reaction is to exhaust a clone or a subpopulation of the effector cells that kill grafts or bacteria. They are specifically directed against the antigen. When that occurs, then the recipient or the patient that's infected retains the ability to react against everything else, but has lost the ability to react against that one antigen or set of antigens that are, for example, present in the organ.

. . . I believe that this is the only way that you can actually reduce tolerance. There are a lot of theories about many mechanisms. But there is one seminal mechanism, and that's it. And that is, exhaustion, deletion, and that occurs by so-called cellular suicide. . . .

So what does this mean? It means that there is a double-edged sword here. Those are the cells that are causing a graft to be rejected. But the reaction that is mounted is absolutely necessary for the induction of tolerance to allografts. But that is the same guideline that has to be met, the same condition that has to be met, if one is going to succeed with xenotransplantation. . . .

I don't think that a strategy to get from A to B can ever be wisely put in place unless you know what the pathway is. In other words, if you can't understand tolerance, if you can't explain allograft acceptance, how in the world does anyone think that you can engineer a strategy that will allow xenografts -- a much more difficult problem -- to succeed? ...

. . . Tell us about the patient who was involved in your trial that involved transplanting a baboon liver into a human.

. . . The first patient was unusually interesting. He was a young man in the end stages of liver failure from the Hepatitis B virus, which is a very questionable indication for transplantation, or at least it was. But he also was infected with HIV. . . . When the whole prospect came up to him, he had a very enthusiastic reaction -- a most unusual one. . . . The reaction was that, whether it succeeded or failed, he wanted to do something in his life that was worthwhile and that would outlive him.

. . .

The combination of the multiple infections . . . HIV and the B virus was really unacceptable for allotransplantation. But bear in mind, this all took place in 1992, and Hepatitis B virus then was considered much more of a menace to post-transplant recovery than it is now. . . .

What was discovered when your team investigated that, specifically, in the case of this liver transplant patient?

. . . If you put in an organ that goes in with a virus, or if you have a bad-acting pathogen, a microorganism that happens to be around at the time you do a graft, and we can host the immune system by immunosuppression, you can very easily induce tolerance to both the graft and to the unwanted microorganism.

When you say "induced tolerance to the unwanted microorganism," what does that actually mean, in layman's language?

In layman's language, and also in scientific language, tolerance means that you induce a state of non-reactivity against a specific antigen. Whether that antigen is a bug, a germ, a microorganism, or whether it's a piece of tissue from some other person, or from an animal; if you can induce the human recipient not to react against that particular thing . . . while retaining the ability to react against other antigens that you have to confront . . . then you've induced tolerance. And if you have that kind of tolerance and you've put in a graft -- a liver graft, for example -- and if the tolerance is complete, you can come back to that same donor and take a heart and put it in, and it will be accepted without any immunosuppression, without any therapy.

So tolerance in a transplant context means that, if you gave me a . . . liver and I came to accept that to the point that no treatment was required any longer, you could come back and give me a piece of skin, or your kidney, or your heart, and it would be accepted. That would mean that tolerance has been induced against those antigens that are in donor tissues.

So that's the good side of the coin. Is there a bad side of the coin, in the sense that a transplant can also amplify the effect of the virus? Does that ever happen?

Oh, sure. That constitutes the historical great risk of transplantation.

. . . The movement of microorganisms to the lymphoid system of the host is really the fundamental means by which the host can get rid of the unwanted pathogens. But, of course, it's also the means by which it can get rid of the graft. But you have to have that unwanted immune reaction, or you can't induce tolerance.

In order to strip out a subset of cells that are specifically reactive against the thing you want to save -- the graft -- you have to have some kind of host versus graft reaction response. . . . That means you have to have a rejection. So if you give a lot of immunosuppression to prevent the destructive side from taking control, in a sense you're undermining the very thing you would like to have, and that is the tolerance. . . . Both getting rid of the graft and getting rid of a bug requires the immune reaction. . . .

So the body can never be blind to a graft. You wish it to see the graft as a sort of benign infection that doesn't need to be attacked and destroyed?

Exactly. . . . In order to have tolerance, you cannot blind the recipient. You have to go through the process of immune activation. That is the seminal mechanism of tolerance. . . . The recipient has to deal with the graft. And when I say deal with the graft, I mean the recipient has to deal with that spray of cells that goes all over to the recipient in lymphoid organs. And it's absolutely obligatory.

Is it possible to remove the harmful pathogens from the donor?

Well, if there are pathogens around, I don't think that you can remove them completely. . . . But what you can do is screen for them. With the xenotransplant situation, or a xenograft situation, where you would have highly controlled farms, then the chances of infecting somebody from the animal, in my view, would be less than one one-hundredth as great as the risk you take in doing human-to-human transplantation.

The human-to-human transplants are always done on a relatively urgent basis, and there is no time to do a lot of studies, to wait for the results of viral cultures and all that. So although efficient screening is done, it's not uncommon to have something get through and infect the recipient. That's always been the case from day one. But you can largely prevent that. With the xenografts, there is this PERV problem. . . . In a long follow-up of those humans that received pig islets over in Stockholm, years later, they found cells that had that the PERV virus.

. . . Tell me exactly what happened to the patient who received the baboon liver. How long did he live, and what eventually did he die of?

He lived for 70 days, and eventually, I believe, he died of over-immunosuppression. And he also died of inadequate function of his baboon liver.

The baboon is often used as a model for our own immune system in animal experiments, since it is genetically very closely related to us. Did it surprise you that you had to go down such a rocky road?

It was somewhat surprising. We had done a great deal of laboratory research with a small animal model, which was said to be a very good analogy in terms of genetic diversity to the baboon-human relationship. The model we had used was a hamster-rat, which has a philogenetic evolutionary distance of about 40 million years. And that certainly would be about the same as the sub-human primates, like baboons and humans. In every other respect, it looked as if that would be an appropriate conclusion. . . .

How did the baboon liver perform in the patient?

Well, in some respects it performed adequately. It cleared the jaundice that the patient had, and it did a number of other functions properly. But there were two problems. First of all, maintaining the function that we had required excessive immunosuppression -- more than would be compatible with life for more than two or three months. And then secondly, despite that, some of the liver functions were inadequate. And these included the production of albumin, which is one of the key synthetic functions of the liver.

So we were faced with a dilemma of beating this course . . . the recipient . . . to the point of submission or death in order to maintain what was sub-optimal function of the allograft. This was reflected in the pathology. We had seven or eight biopsies at 70 days of patient life. They showed perfect control of the classical cellular rejection that you see with allografts, and perfect control of the vascular lesions that are very often associated with xenotransplantation due to antibodies.

But there was decay in the appearance of the liver cells themselves, with a lot of fatty infiltration. And all this time, throughout the entire course, there was laboratory evidence of continuous complement activation. So we concluded that this could have been an idiosyncratic experience. We did another case, had the same general experience, and we concluded that we could not succeed with xenotransplantation as a service. By that, I mean that we could not give the patient a reasonable, if any, chance of being restored to a meaningful life and role in society.

So that was the result of our analysis of the events in these two cases. And so although we had permission to do two more, and we now thought you could probably succeed -- if you did 20 cases, you could maybe get one non-survival -- but it wasn't our right to do that. So we cancelled the other cases, and resolved that we would never do another one unless we could find a way to deal with this complement activation and obvious innate immune response. . . .

Everything you've discovered suggests that xenotransplantation is a lot more complicated than some of its proponents will give credit for.

You can conclude that, and it's not unreasonable. But my view is that we could not hope to succeed at a service level, which is what we have to do, with any existing tools of immunosuppression. And furthermore, the key to this would never be controlled by alteration of the recipient's immune system -- that's something that would have to be done with the graft. We further concluded that, if you couldn't succeed with concordant transplantation . . . then there was really no point at all in looking at discordant transplantation, for example, pig to human, or sheep to human. And that remained my viewpoint until it became obvious that there was a prospect of humanizing pigs to the extent that they could be simulates; they could be recognizers if they were allografts. But that, in itself, is a large order.

Were you surprised that the baboon, who is closely related to us, didn't perform better in the transplant?

Yes, I was, I was. I was surprised the first time, and I was crestfallen the second time to realize that the same thing had happened twice in a row. . . .

Do you think there's a lot of pressure within the scientific community at the moment to push xenotransplant?

There are multiple pressures to develop xenotransplantation. From a strictly altruistic point of view, the pressures do derive from the fact that many people die because there are no available human organs. And that has put a ceiling on the worldwide use of organ transplantation for organ-defined diseases. That is a terrible quandary, because organ transplantation has, for every one of the organs, changed the philosophy by which that special branch of medicine, whatever it is, is practiced. Kidney, liver, heart -- you can replace the engine. That's an electrifying concept. So, in order to pursue and develop that ideology, you've got to break down the ceiling. And I like to think that that is the principal thing that motivates this movement.

Of course, there are more crass considerations . . . because there is insufficient money from government sources to really finance this kind of research. It has come to depend very largely on venture capital, and so then you have the additional pressure of fiscal expectations. Somebody would benefit greatly from the development of this technology. The market is construed to be a multi-billion dollar market. It almost certainly would be. So that's another consideration.

. . . But I would say that need is the great incentive, at least as far as I'm concerned. An enormous number of people die in this country. I've forgotten what the estimated number is. But whatever the number is, it's only a fraction of those who never even can be considered for transplantation, because they fall outside the fairly narrow qualifications that have been set up to restrict the candidacy.

But it does those patients no help to raise up expectations that don't really exist, does it?

But I think the expectations do exist. And I think that there is a substantial chance that they will be fulfilled, by understanding . . . that the seminal mechanism is allograft acceptance. The seminal mechanism derives from the fact that the human immune system is recognizing human organs as if they were non-psychopathic microorganisms, to which tolerance -- i.e., graft acceptance -- can be induced. Whereas, the xenografts -- even the baboon xenograft, as close as it is -- was recognized as a less favorable microorganism. In the case of the pig organ, it is recognized as a frankly psychopathic microorganism that must be destroyed, no matter what the cost. . . .

That does raise another very important question. If you are creating transgenic animals that are humanized and you're transferring the humanized pig organs into baboons, you're testing the system against an assay, which is essentially a xenotransplantation assay. So I believe that has to be looked at very carefully, and some means need to be developed to test humanized pig organs in humans under certain circumstances that are ethical and proper. . . .

Do you believe that we're at the stage now where there is nowhere else really to go except to try human clinical trials involving pig organs?

. . . Well, one of the possibilities at the moment is that, with some kinds of end-stage organ disease, you can buy time by using xenografts as a bridge. Probably the favorite one of all is the liver, where you have full . . . failure, and you can test transgenic organs for that purpose. But that technology was very thoroughly tested about 40 years ago. . . .

You can get some information, but it isn't going to tell you what things are going to be like one week or one month down the line. But what could be done in those instances -- and I think this would be ethical -- where a person has stipulated in a living will that they want their organs or their body used for research, but this proves to be not possible, because they're infected, which is a contraindication for a donation -- or, more commonly, they have a malignancy that you would easily transplant, and therefore you can't use those organs -- but with the family's permission in the heart-beating category of donors, you could sew in a kidney or any of the other organs, and, with everybody's agreement, observe those organs for six hours. You could observe in a very clean and refined way, with all the functions and, at the end, with complete pathology and complete examination. You would learn more from one such experience than you would from all of these other surrogate models, because you would have transplanted an organ into a human with an intact circulation and done a real study of whether that organ is on its way down the drain, whether it's got a lot of complement in it, and so on.

That's never been done. And the institutional review boards, I think, would have no problem with that. . . . They would try to help. A guy named Paul Tarasaki, who has done massive epidemiologic studies after kidney transplantation, has shown that within six to twelve hours after a kidney transplantation, with the appropriate data, you can predict very accurately what the outcome of that organ is going to be five years down the line. . . . I think the use of the humanized pig-to-baboon surrogate model for a human clinical transplantation actually could lead us astray and give us false answers.

What about cloning? What opportunities might cloning provide in the future?

Well, I think it's a big opportunity, and maybe it offers a supreme possibility.

What does it actually hold out the prospect of doing?

Cloning . . . has been done by nuclear transfer. At the time of the nuclear transfer, one has access to the whole genome. You can, at that time, take out or put in genes, and then you would have a cloned pig that could be reproduced. And then you could carry out breeding experiments with your product, and replace or knock in multiple human genes. That is what I think may allow enough alteration in pig organs to make them comparable to allografts. And in my judgement, that's what you have to have. If you're going to use xenotransplantation, you have to be able to offer the patient something that is comparable or close to being comparable to allografts.

But even the way it is now, there's a sliding scale of expectations, even with allotransplantation. . . . So if you're putting in a cadaveral organ, it isn't as good as if you had a live, related match donor. And if you put in a xenograft, it may be that it is not quite as good as a cadaver human organ. But there may be such substantial advantages that you would outweigh that set of disadvantages. For example, you could get an organ when you exactly needed it. You could have teams working on the donor and recipient who were fresh and coming from a good night's sleep. You could use your donor tissues to induce tolerance or some degree of allograft acceptance before the allograft, the definitive organ, even arrives.

And there would be many things that could be done that would outweigh the ostensible disadvantages that you would be able to define in a purely scientific sense. So I think that those who are in the field, the ones that really know what the requirements are, would agree with me that it is going to be very tough, but that there's no room for opportunists. What has to be done is extremely straightforward, and I don't know if it's going to be possible. There is a possibility that no matter what is done, it will not prove to be possible.

. . . Do you think it will be ever safe, totally safe, to transplant pig organs into humans without knocking out the PERV element of the pig genome?

I think the PERV is a red herring, myself. Personally, I think that the scare about creating a new black plague or something like that is so remote that it's not a factor. The central problem is that we don't know how to get the discordant allograft, or even a concordant allograft, accepted by the human immune system. And if that objective were met, everything else would just dissolve.

The debates about PERV are interesting. Of course it needs to be studied. But the PERV is part of the pig genome, and you know, the pig and the human genome have picked up little bits and pieces of DNA from microorganisms in the course of evolution, so I don't think that . . .

Is it not risky to leave the PERV element inside the pig cells?

I don't think so.

If it were to activate, if it became a hot virus by whatever means, it would then possibly be cloaked in a cell with human genes?

I can't answer that question, because I'm really not qualified. . . . But you're asking me what my opinion is, and my opinion, as somebody who has seen tremendous cross section of experience in transplantation, is that control of the infectious disease problem with xenotransplantation would be vastly easier and more effective than has ever been accomplished with allotransplantation.

Now, maybe a million people have undergone allotransplantation. If there were going to be some kind of transmographication of weird viruses . . . that would have already occurred. If you had these clean pigs, I think that the possibility of infectious complications of that kind, or the emergence a of hot PERV virus, is very remote.

We do know that, in human-to-human transplants, we can amplify the effect of viruses, especially when the recipient has never seen that virus before. Is it not arguable that there will be viruses? Some of the retroviruses and endogenous retroviruses that can't be removed in the pig that the recipients of these xenoorgans will never have seen before? Do you accept that we could amplify the effect of the virus?

You're asking me if we could create infections that haven't been seen before. I'm sure you'll find people who will support that theoretical possibility. I think it's a small one, and with the xenografts, under the conditions that have been stipulated now by government agencies, that risk is very, very remote. And that risk is not in the same league as the risk that you might face just with allotransplantation.

Is the scientific community ready to expect the unexpected in xenografts?

The scientific community thrives on the unexpected, and certainly would be on the alert, for sure. In fact, part of the regulations that were fostered mainly by the FDA call for surveillance for exactly that. I think many precautions have already been mandated should something like that happen.

With 20-20 hindsight, looking back at the beginning of Industrial Revolution, we didn't know everything. We were handling new materials in ways that we didn't always understand. We're now at the beginning of the biotech revolution. Can the public be confident that we are handling things in a way that's so powerful that the chance of risk is absolutely minimalized?

There is a learning process in every new technology, and certainly there will be here. I'm sure there will be mistakes made. But as with the Industrial Revolution, eventually there will be corrective measures, probably much more rapidly now, because people are so ecologically aware.

Do you think there's a bit of a battle for the hearts and minds of the public around xenotransplant? Or will there be a battle?

Well, there has been, but I don't think that the sides are defined nearly so much as a split between two segments of the scientific community. It's more a split between the scientific community and the animal rights advocates. Xenotransplantation, for animal rights activists, is anathema. And I'm not unsympathetic to their point of view. But I think one has to weigh priorities, and I've done that painfully. Our house is full of animals, and they're part of our family. But everyone has to weigh their priorities, and I think when it comes to a choice between a child and an animal, the decision will be clear.

But there we get back to this point of service. We can't do xenotransplantation if the child, or whoever it is, is not served by what is offered. . . .So we have to get it right, in other words, and if that happens, I think all the camps will come together.

home · four patients · the risks · animal welfare · the business · the regulators
discussion · faqs · video · chronology · interviews
synopsis · tapes & transcripts · press · credits · carlton's organ farm
FRONTLINE · pbs online · wgbh

web site copyright 1995-2014 WGBH educational foundation