Albert Lasker
Clinical Medical Research Award

An Interview with Marc Feldmann
Interview by John Paul

Dr. William Paul, Lasker Jury member and an immunologist at the National Institutes of Health, spoke with Marc Feldmann before the Lasker Awards Luncheon. Feldmann comments on winning the 2003 Albert Lasker Medical Research Award and his work with Sir Ravinder Maini.

Date of interview: September 18, 2003

Paul: Good afternoon. I am William Paul, an immunologist at the National Institutes of Health, and a member of the Lasker Jury that chose this year's winners. I am delighted to have the opportunity to discuss the significance of Dr. Feldmann's work today. This work is a real landmark in advancing human health.

I begin by apologizing to all for not being present in person, but Hurricane Isabel has grounded everyone in D.C. I am here at my office at NIH. It is a real treat for me to be able to speak to my good friend and colleague whom I have admired for many years, Marc Feldmann. I hope we can have a really delightful discussion about the significance of his work and its importance for advancing human health.

Marc, would you be kind enough to just sort of lay out for us the basic principle behind the treatment that you and your colleague, Professor Maini, have introduced to treat rheumatoid arthritis?

Feldmann: What our work has shown is that blocking one pro-inflammatory molecule, tumor necrosis factor, usually abbreviated TNF, has much more dramatic effects on inflammation and on chronic diseases than would have been anticipated when the work started. We've been able to show over the last ten years that blocking TNF, which is done in patients with either antibodies or TNF receptor-fusion proteins (there are now three of these drugs on the market), not only has a profound effect on the inflammation in rheumatoid patients by reducing symptoms—it also it protects their joints from disease-driven damage to a remarkable degree. So our research focusing on the molecular mediators has enabled us to find that a single one, TNF, has a particular importance in orchestrating the disease process.

It has also been very gratifying that whereas our own clinical work has been restricted to rheumatoid arthritis, colleagues and other clinicians have picked up on the TNF blockade principle, which has now been tested in a whole plethora of diseases. Now there are approved treatments by blocking TNF for juvenile rheumatoid arthritis, Crohn's disease, psoriatic arthritis, ankylosing spondylitis, and about 20 more diseases responding in trials. Another important thing this work shows is that the biomedical research community now has good technology available for discovering which are the pathogenic molecules in a given disease. We now have good tools at our disposal to work out important therapeutic targets that have real biological and medical meaning.

Paul: The Lasker jury was motivated to award you and Tiny Maini this prize because the work was both so dramatic and represented the application of a new principle to treatment of patients based on the rational determination of a key factor in disease pathogenesis, and then using a monoclonal antibody specific for this factor for its treatment. Before we talk more about that, since this treatment has had such a big impact on the lives of patients, I wonder if you could say a little more about the impact on individuals who suffer from rheumatoid arthritis.

Feldmann: Yes, the patients I remember most clearly are the first half dozen patients that ever got this treatment. We were able to develop our idea efficiently because the work of Tony Cerami and his colleagues. Bruce Beutler and Kevin Tracey had identified TNF as of major importance in the sepsis syndrome, from which 300,000 Americans die every year. This prompted a number of pharmaceutical and biotech companies to develop a range of TNF inhibitors for clinical use in sepsis. So when our laboratory research came to the point that TNF appeared to be the important target, as TNF was orchestrating the recruitment of leucocytes and other inflammatory mediators, a number of therapeutic agents were potentially available and ready to use in patients, as they had already been used in clinical trials for sepsis. But getting our hands on them for testing our hypothesis was not easy. The first company we approached, after 6 months of negotiation, refused.

I vividly remember that there was initially a lot of concern about the safety of blocking TNF, based on the compelling data that TNF is a very important component of host defense, important in protection against bacterial infections. Some workers thought that blocking TNF was going to be very risky, and might cause a lot of harm to our patients.

When the first trial started, we had patients in a private room with our own research nurse sleeping at the foot of the bed. Thus if disaster struck, we would be rapidly alerted. But within a couple of hours, before patients even finished their three hour intravenous infusion of the drug that we were using, they were feeling better. At the time the drug was called cA2, which stood for chimeric A2. A2 was a monoclonal antibody developed in Jan Vilcek's laboratory, and cA2 is now better known as infliximab or Remicade®.

What they noticed first was relief from fatigue. Their mood improved, and by the next day, the pain and stiffness had got less and they were feeling remarkably better. And this benefit continued for a period of weeks, so in this initial trial there was virtually instant evidence suggestive of clinical benefit. Within the first few hours, Tiny and I knew that something significant had happened. Of course for a patient getting a needle in your veins and fluid pumped in for three hours is a pretty dramatic treatment, so we were overjoyed when the blood tests of inflammatory markers in these patients demonstrated to us that this wasn't one of the world's giant placebo effects.

The C-reactive protein in serum, a useful marker of the degree of inflammation in RA patients, had dropped to normal levels within a week. So that and the clinical benefit was the initial evidence that our idea had worked, and within a couple of weeks, we had patients whose mobility had clearly improved, being able to go up and down stairs quite readily, whereas in the past they were really struggling up the stairs. And we still have those videos. These were certainly momentous times for us because they had shown that our novel or heretical idea that a single mediator cytokine coordinated a large part of inflammatory response was true. And not only that—the patients' clinical response was more striking than we had anticipated, for patients 10 or more years into a chronic disease. These were heady times for us and our key collaborator at Centocor, Jim Woody, an ex-colleague, who, after a stellar career rising to the top of U.S. Navy Research, had joined this biotech company as Chief Scientist.

Paul: You point out that rheumatoid arthritis and the other diseases for which this treatment can be used, while they manifest themselves in certain areas, like the joints, are really systematic diseases, and blocking inflammation has a dramatic effect on the quality of life of these patients. I know having a cousin, a young woman, with rheumatoid arthritis. She is beginning treatment with a TNF inhibitor which has already shown this remarkable effect. But I'm sure you must have thousands of such stories. It must be very gratifying to have had a role in a contribution that has such an impact on so many people in such a positive way.

I understand, Marc, that there's evidence now that even some of the bone destruction that's seen in this disease can be reversed by treatment with antibody or with TNF receptor. Is that correct?

Feldmann: This is probably correct. What's very well documented now is that therapy with all three of the TNF inhibitors that are now in regular use will block ongoing cartilage and bone destruction. But in only one clinical trial has it been noted that there appear to be signs of joint repair. Now that trial was quite large, well carried out, had independent observers who didn't know what they were looking at. So I think it's a reasonable hypothesis that there may be repair. The confirming clinical experiment has been done, and the results will be known soon. Overall, the existing results demonstrate that human diseases which involve tissue destruction are potentially more repairable than initially thought, which is important, as it will focus much more research on the topic of tissue repair.

Paul: Right. And I'd like you to say another word or two about the other applications of neutralizing tumor necrosis factor in other conditions. You of course mentioned them briefly a moment ago. But it's been really, for those of us who follow this, it's been really quite gratifying to see that this concept is really generalizable. And you might mention again those other diseases where TNF neutralization has had a big impact and what you see is the future for this therapy.

Feldmann: Yes, it has been gratifying that the principle of TNF blockade is generalizable, but there are some interesting puzzles. Probably it's totally expected that juvenile rheumatoid arthritis of the adult type would respond, although there are certain differences in its pathology from the adult form. There is a very interesting puzzle, which is in the contrasting response of Crohn's disease using two different anti-TNF drugs.

Crohn's disease responds very well to a monoclonal antibody against tumor necrosis factor—infliximab. And my understanding is that the trials with the other antibody that's been approved for rheumatoid arthritis, adalimumab, are also successful. But the TNF receptor fusion protein, etanercept, while very effective in rheumatoid arthritis, doesn't work in Crohn's disease trials. Whether the difference is due to the different dosage regime of these drugs at present is unknown. Infliximab is given by intravenous injection, and so at the onset of treatment a large amount is given, much of which gets through the inflamed vessels into the diseased tissues. But etanercept is given under the skin in low dose and thus relatively little will get through into the tissues.

Another interesting aspect of Crohn's disease is that the tissue destruction in Crohn's is also relieved by TNF blockade. The tissue destruction in Crohn's disease is just as unpleasant as it is in rheumatoid arthritis; it is often manifest as fistulas, which are communication between the gut and skin, a condition which is very distressing. A puzzle and surprise is that in both Crohn's disease and in rheumatoid arthritis, there are many patients that have marked relief of tissue damage, be it in their joints or in the closing of fistulas, despite a modest and incomplete anti-inflammatory response. That tells us there are many aspects of these chronic inflammatory diseases we don't yet understand, such as the linkage of inflammation and tissue damage.

So there's a lot of research still to do to understand the mechanism of action of anti-TNF therapy. Studying the mechanism of action of drugs is very important, as it provides the opportunity to make improvements in the future. Other diseases successfully treated with TNF blockade now include ankylosing spondylitis, psoriasis of skin and psoriasis of joints. So it has been, as you said, a generalizable phenomenon.

Paul: Marc, would you perhaps say a few words about how it was you and Tiny Maini started on this approach, and what led you to the conclusion that this idea was one that had real merit and that was worth the enormous amount of time and energy and the years that you put into it?

Feldmann: Well, as you know, I started my research career in Australia at the Walter & Eliza Hall Institute in Melbourne as a basic immunologist, studying in vitro immune responses in mice under the direction of Erwin Diener and Gus Nossal, and being very interested in the signals passing between immune cells and the active molecules involved. These molecules at the time were totally uncharacterized, and I wasted much time trying to characterize them. But now we know them as the pro-inflammatory cytokines. WEHI, as it was known, and still is, was a great place to learn science with eminent leaders such as Jacques Miller, Dan Metcalf and of course Gus Nossal. My interest in mediators, initiated in Australia, continued in London. There I was working at the Imperial Cancer Research Fund Tumor Immunology unit, run by Avrion Mitchison.

My interests continued in immune regulation, with the eventual goal being to generate immunity against human cancer. Most of the unit was studying viral-induced cancers. In these animal models, there were obvious immune responses. Whether the immune responses were to the virus or to the cancer was arguable. But in human systems in the early 1980s, there were no detectable immune responses to cancer. At this time I became aware that there were a number of studies that had described up-regulation of MHC molecules in local sites of auto-immunity. This was described in juvenile onset diabetes, in rheumatoid arthritis by Klareskog, Wigzell and Janossy, and in thyroiditis. The nearest scientists to me in London doing this type of work were Ricardo Pujol-Borrell and Franco Bottazzo studying Graves' disease, a thyroiditis.

Since the function of MHC class II is antigen presentation, in tissues with upregulation of MHC class II, antigen presentation was probably augmented. And since antigen presentation is an aspect of immune regulation that I had studied for many years, it occurred to me that perhaps the earliest important event in the induction of an auto-immune disease may be upregulation of antigen presentation. So in 1982, while on holiday in Spain, it dawned on me that the genesis of auto-immune diseases may be explained in a fairly simple way. For a local autoimmune disease, like thyroiditis, if there was a local activation of the immune system—by a virus/bacterium etc.—near the organ site, this could lead to the liberation of molecules from the immune system that would upregulate MHC class II and augment the antigen-presenting function within the local thyroid tissue.

It had been well documented that autoantigen reactive T cells can be detected in normal mice or people, and so with the upregulation of antigen-presenting function, and perhaps a bit of tissue damage releasing antigen in the vicinity, these autoantigen reactive T cells could be activated, and consequently induce the disease. This concept was published as a hypothesis in the Lancet in 1983.

To test this notion, I became very interested in directly studying autoimmune tissue. We were able to demonstrate that several aspects of this idea was correct; thus there was upregulated antigen-presenting function in the thyroid tissue. Also autoantigen reactive T cells, which were stimulated by these thyroid epithelial cells, could be cloned out by my colleague Marco Londei.

To take this concept one stage further, to ask what molecules may be initiating the disease process, it became apparent that what was needed was access to an autoimmune disease site at the height of the disease. In thyroid disease that's not possible; the tissue is taken out after the disease has been treated. But in rheumatoid arthritis, physicians stick needles into joints to give them pain killers and steroids and take diagnostic biopsies. And so that was the disease where I thought I could delve deeper into the disease pathogenesis, and try to identify molecular mediators which might be therapeutic targets.

An eminent immuno-rheumatologist, Nathan Zvaifler, had been on sabbatical in London in the late 70s. He'd come to work with Mitchison, but ended up working with me. And so I contacted him to ask him who was the best person in London to discuss rheumatoid arthritis research samples with. He recommended Tiny Maini, so I called him. We met and there was an immediate positive chemistry; we could communicate very well. He understood immunology, had studied molecular mediators from activated lymphocytes, had published a paper in Nature on this, and so there was a common background and shared interest. So we started working together.

Paul: Marc, you've given us a good picture of the scientific issues and the ways you went about attacking them. But those of us who don't do the hard work of development probably don't have a picture of the complexity of moving the principle that neutralizing TNF can block the release of other cytokines to the actual development of a drug. I know you must have encountered all sorts of problems. It would be interesting to get a feeling of where the big problems were and how you overcame them.

Feldmann: An initial problem in our research was how to culture synovium. Many previous rheumatology studies had passaged the fibroblast-like cells of synovium and discarded the recruited cells of the immune system. As immunologists we didn't think it made sense to discard 90 percent of the cell mixture, so we cultured the whole complex mixture of cells—T lymphocytes, macrophages, endothelium, as well as fibroblasts.

A major general issue in research is how much to trust in vitro analysis of complicated systems. This is something that many people have argued, that in vitro models are very accident-prone. You've heard the expression, "In Vivo Veritas," which I think it has a lot of merit. In TNF studies, this is important, as TNF can be made in vitro in 30 minutes or so, and so the potential for artifacts is high. The rapid synthesis of TNF magnifies the research problem, because it turns out that probably one of the important reasons that TNF is involved in so many diseases is that it is the body's major "fire alarm." It's the most rapidly produced cytokine after any stress, and so it "alerts" leukocytes to the site of damage to remedy the problem.

Because of these potential problems, we didn't really cross the threshold of being certain that the in vitro analysis of cytokine interactions was correct until the animal models of arthritis also responded very well to anti-TNF therapy. And while of course the relationship of animal models of disease to their human counterparts is not clearly defined, this positive result really gave both us and everybody else we talked to considerably more confidence, as we could treat collagen-induced arthritis, and subsequently others treated every other model of arthritis by blocking TNF after the disease onset.

The capacity to block animal models of arthritis after they've started was a key result. In the pharmaceutical industry, often there's a lot of work being performed—not on treating established disease models, but preventing them from developing. Preventing disease is much more likely to work than treating an established disease.

So I think these were key research events—that we developed a new model of synovial culture, and we could show that the conclusions from this model in vitro were also borne out in an animal model of the disease, after disease onset.

And while at the time the collagen-induced model of the disease arthritis was rather poorly validated as a model of rheumatoid arthritis, subsequently it has become a very well-validated model for anti-cytokine therapy.

In parallel with our work, George Kollias in Athens had made transgenic mice which overexpressed TNF, and these developed a destructive arthritis resembling rheumatoid. This work independently supported the concepts we had been developing from human studies.

Paul: So Marc, may I now ask you a different kind of question? So you and Tiny have sort of led the way in the development of cytokine neutralization, or TNF neutralization, to really have an enormous impact on major human diseases. But of course there are still issues, and I'm sure you'd agree that rheumatoid arthritis is still a problem—that there are some that don't respond, and as we talked earlier...people who have serious bone lesions may or may not repair them under the treatment. So where do you see the further application of this general approach? Do you think there are still major opportunities that can be used to develop other but similarly principled ideas to attack other inflammatory diseases?

Feldmann: Yes, one of the interesting things about getting a major prize such as this one is that it makes you think in more general terms. And one of the points that we've become aware of is that in general terms in chronic inflammatory diseases, pro-inflammatory cytokine blockade is usually effective. In contrast, attempts to block the adaptive immune system directly in autoimmune diseases have been remarkably unsuccessful, unlike the situation in transplantation.

Just to amplify on this point a little bit: anti-TNF is obviously very successful in many diseases. Blockade of interleukin-1 is also approved. There's also good preliminary data for blocking interleukin-6 receptor, or interleukin-15 in rheumatoid arthritis. So there are a whole set of molecular targets in inflammation whose blockade is effective, and at this stage it's impossible to really know which one is better than another, due to the lack of available data. But in contrast, many attempts to induce immunological tolerance, for example with peptides, or induce oral tolerance, or deplete lymphocytes with antibodies have really made relatively little progress.

So it may well be that blocking the "innate" immune system, which is the less flexible "hardwired" part of the immune response, is easier to accomplish than the blockade of the more flexible adaptive immune system. We probably don't yet understand enough about the adaptive, flexible immune response to obtain the clinical results desired.

We've become aware that human diseases are really very complex, and to expect that blocking a single protein is going to a cure a complex multifactorial disease is probably impossible. But what we have shown, and other have subsequently confirmed, is that it is possible to obtain measurable and clinically meaningful effects by blocking a single molecule. And so the important question becomes how to get closer to what the patients want, which is a cure. We believe that we already have made a first step on that particular direction. And it's what is now essentially standard practice for rheumatoid arthritis, which is to use the TNF-inhibitors in conjunction with the anti-rheumatic drug methotrexate.

So I think combination therapy is going to be the way forward, and the difficulty is really going to be to plan combination therapy which is more effective, without compromising on safety, and without increasing the costs of therapy to absurd levels. This is because, at the moment, a major practical problem with anti-TNF therapy is that the cost is too high, so that the number of patients that could benefit is a lot more than the ones actually getting the treatment, and treatment is delayed. But this rationing is implicit in anything that costs about $15,000 a year just for the active drug.

Paul: Marc, you sounded somewhat pessimistic about the possibility of therapies that would actually cure these diseases in contrast to therapies that treat them. Do you think, in fact, there's a reasonable prospect that we could get either curative or preventive strategies for complex inflammatory disorders like rheumatoid arthritis, like Crohn's disease, like psoriatic arthritis?

Feldmann: I think we will eventually get to cures. Tiny and I were privileged to be able to discover one of the first rationally-based immunotherapies for chronic inflammatory diseases. Prior to anti-TNF therapy, the existing drugs were developed without much insight into the molecular nature of the diseases. The previous work could not encompass the molecular understanding of disease, as they were attempted before this molecular understanding could be generated. We have been very fortunate, starting our research at the same time that cytokines and other very important immunological molecules were being defined and cloned, and with much improved research technologies.

So we were able to base our research on the work of very many talented individuals, which had built up the knowledge of the field of molecular medicine, especially of cytokines, to the extent that we could then add our contribution to the pile. These include the pioneers of IL-1 research, such as Gery, Oppenheim and Dinarello; of TNF, Cerami, Beutler Gray and Goeddel' as well as the pioneers of autoimmunity, Burnet, McDevitt and so on. So my current belief is that the best approach towards a cure to rheumatoid arthritis would be to add other treatment modalities on top of anti-cytokine therapy, and for now it seems that anti-TNF is probably as good as any other in that field, and perhaps better.

But what to add on top of anti-TNF therapy? Very useful would be therapy that attacks the immune system, but without compromising the regulatory systems. It is likely that the first treatments with anti-CD4 antibodies didn't work because while they may have killed or blocked pathogenic T cells, they also deleted or blocked regulatory T cells. So I think that a more focused attack on the pathogenic aspects of the immune system by early targeting the class of pathogenic cells might be a way towards getting a cure. My colleague, Fionula Brennan, who was a pivotal worker in the synovial TNF regulation studies, is now focusing on defining pathogenic T cells in rheumatoid synovium, with encouraging progress.

Because the rate of progress in the past 20 years has been so fast, I can't be pessimistic about cures. I do think that in the next 20 or 30 years, we will have long-term remissions and cures for these diseases. There are many other approaches, and my colleagues, in particular Drs. Ewa Paleolog and Peter Taylor, and I are very interested in seeing what happens if both angiogenesis and TNF are blocked.

Paul: Great. Well, Marc, let me ask you an entirely different kind of question. We've heard about your early career in Australia and in the U.K. on more basic immunology, and then your recognition that you could turn your knowledge and your enthusiasm to human disease, and of your enormously useful collaboration with Tiny Maini. So what can the scientific community expect to hear from Marc Feldman in the future? Where are things that are exciting for you now?

Feldmann: Well, Tiny and I have worked together for 20 years, and in most of that time, I've had a sheltered, privileged life. Tiny's about seven years older, so he's been the one that's ended up doing nearly all the administration. But in England, there's still a retirement age at 65, so now he's turned 65, "retired," been knighted for his contribution to rheumatology—that really means he can still come to the lab and do whatever he wants, but now the admin has been turned over to me!

So I'm now head of the Kennedy Institute of Rheumatology, the Arthritis Research Campaign flagship research institute now at Imperial College London. My challenge in that role is "how do I build from a very good institution to an even better one?" So we've decided to do a number of things. One is to start looking at other rheumatic diseases, where the pathology is even more poorly understood than rheumatoid arthritis was when we started. The real challenge is osteoarthritis, where there's a complex, poorly understood series of events which ends up destroying the joints. At a late stage, it certainly involves cytokines and inflammation, but at the early cartilage specific stage it doesn't. By looking at that disease with fresh eyes and techniques, trying to work out how to diagnose it early, my colleagues Jerry Saklatvala and Hideaki Nagase hope to make an impact, and are making encouraging progress.

Personally I will focus my knowledge of how to translate laboratory research into medicine by helping others challenge other major arthritic diseases. But, another aspect that I am very interested in is exactly the one we've been discussing for the last ten minutes. It is how to convert significant medical benefit into remission or even a cure for rheumatoid arthritis. Towards that end, we've done a considerable amount of work on what therapies can be synergistic in animal models, and therefore have a number of candidates that we wish to test in the human situation.

But there is a dilemma that some or many of these combinations may be pro-infective, and therefore risky, but others we think are not going to be pro-infective. And so that's an aspect that we definitely intend to investigate very actively in the next few years. Another aspect which I think is very important is how to prevent TNF blockade, which is very expensive, from being only accessible in the rich parts of the world. We are thus trying to work with various groups that may be able to develop more effective forms of TNF blockade, but at a lower cost to patients.

Because if you can reduce the price from say $15,000 a year to $5000 a year or less, the potential increased access will be very important, and it will also make it much easier to treat patients earlier in the course of disease. In rheumatoid arthritis, at the moment, there's an unfortunate paradox that once patients have been diagnosed with rheumatoid arthritis, they usually can't get TNF-blocking drugs, the most joint protective drugs, until they've failed one disease-modifying drug, and often more than one. So it means that for a period of two to five years, they're getting sub-optimal treatment for protecting their joints, and hence their joints are getting progressively damaged. If we had more flexibility on TNF blockade, it may be possible to change the paradigm and have TNF blockade as a first-line therapy, but only if the safety profile permits.

Paul: Winning this prize I'm sure is gratifying to you, and to all of us who've admired your work. Maybe we could close by my asking you to speak about what is the issue that most excites you, or what point you would like to make about careers in medicine and biomedical research?

Feldmann: Bill, you and I have had the amazingly good fortune to grow up in immunology and molecular medicine as the technology has progressed rapidly and dramatically, and so a whole plethora of interesting molecules have emerged from glimmerings of biological activity, to biochemical definition, to having their genes cloned and the proteins produced in kilograms. So we've lived through a revolution in molecular medicine. We know something that many younger people do not know—that there is a huge opportunity for creative ideas to be harnessed to do enormous medical good. The era of molecular medicine contributing to human health has actually barely begun, the vast majority has yet to come.

We now have all the pieces, or at least most of the pieces, of the "chess set" of the immune system and the rest of biology. The genome has been cloned, the proteome is being addressed, but we have yet to learn how these genes are regulated to producing appropriate proteins, and so I think the stage is really set for monumental changes in the way therapy can be delivered in the future. There are enormous opportunities for dedicated young scientists and physicians to make contributions on an even greater scale than ours. The dilemma, however, is that the career progression for academic physicians, both in Europe and the U.S., is becoming more and more difficult. And as the scientific knowledge increases, it's increasingly difficult for someone to become and remain a more than competent physician and a talented scientist at the same time.

This problem has really got to quite a difficult stage. So I think there is a very real need for young people to perhaps look at the example of how Tiny Maini and I have worked together, and ask whether, perhaps having a little team of people in a leadership role—two or perhaps even more scientists and physicians working together, whether that may be an effective paradigm for translating molecular medicine into clinical benefit in the future. That the complexity of both the laboratory science, the clinical science, and its application to patients is now such that a single individual may not be the optimal way to lead a group able to deliver this translational type of "laboratory to the bedside" medicine.

I've certainly found that having a colleague, a partner, a friend, Tiny Maini, with complementary/overlapping knowledge, to help going through this jungle of unknown and dangerous places in the application of scientific knowledge for medical benefit, to be extremely beneficial. And it may be an optimal approach for molecular medicine in the future. We need every scientifically trained physician available to be excited by the opportunities, and not discouraged by the endless training schemes and the risks. But also, we need the realization that these people probably need support from more active laboratory-based researchers if their potential as academic physicians is to be fully realized.

I'm a very strong supporter of the collaborative approach to scientific research. It's an ethos I learned in Australia, at WEHI, where I saw that leaders like Nossal, Miller and Metcalf were able to be highly productive and internationally competitive partly by leveraging resources through successful collaborations.

Life is too short to "re-invent the wheel." Collaborating with others can speed up the research progress for the group, and hence lead to greater success. Effective collaboration is just as important to science not yet ready for "translation" for clinical benefit.

Paul: Well, that's terrific Marc. Marc, I do want to thank you, and again to congratulate you and Tiny Maini on sharing this prize. It's always a great excitement to see important advances recognized and to know how individual decisions could have such a great impact on aiding human health. So please accept my congratulations. I know all of your colleagues and friends are simply delighted that you are going to be receiving this honor. Again, it's been a great pleasure to have this chat.

Feldmann: Thank you Bill. Thank you for these very kind words.