published 07/11/2001

For half a century, transplant surgeons have dreamed of tricking the immune system into accepting foreign organs without immunosuppression. Now government and industry have placed their chips on the table, launching a run of clinical trials designed to achieve tolerance in humans. This effort, unnoticed by the press and public, has the potential to transform organ transplantation and cure autoimmune disease. But it also could backfire disastrously.

At a special Pentagon briefing in August 1997, the U.S. Navy announced a “breakthrough” in organ transplantation. Under the cover of a revolutionary new antibody, two monkeys received kidney transplants -- without taking any immunosuppressive drugs -- and did not reject them, even after six months. Yet the immune systems otherwise remained normal, thus apparently achieving “tolerance,” the elusive but much-desired state of specific acceptance. “Navy medical researchers have discovered a way to teach the human immune system to accept even completely mismatched transplanted organs,” read the Defense Department press release. Yes, these were monkeys, not humans, but the immunological differences were considered minor. There seemed no reason it wouldn’t work in people.

The dramatic news catapulted the tolerance field from an obscure, utopian research backwater to a national priority. Within two years, a $144 million clinical trial consortium, the Immune Tolerance Network (ITN), was up and running, funded by the National Institutes of Health (NIH). “One could argue… that that [Navy] work was singlehandedly the reason why the tolerance network came into being,” says ITN director Jeff Bluestone. “It was a proof of principle that people were waiting for.”

Companies also eagerly entered the tolerance race. By 1999, six human trials of the same tolerance-inducing antibody were underway, all sponsored by Biogen Inc., the Cambridge, MA-based biotech giant that made the antibody. Biogen’s ambitious program reflected the faith it had in the drug, a humanized anti-CD40 ligand monoclonal (trade name Antova). A new era of transplantation, shedding forever the fearful side effects of immunosuppressant drugs -- infection, kidney toxicity, cancer -- seemed imminent.

Then, in November 1999, in a stunning reversal, Biogen called a halt to all trials, due to what a press release called “thrombo-embolic events” -- blood vessel clots. Details were never released, results never published and no deaths reported. But at least one patient (a kidney transplant recipient) died from a pulmonary embolism, a blood clot that migrated to the lung.

The Antova trials were a Waterloo for tolerance, and Biogen has no plans to resurrect the drug. But its failure only temporarily slowed tolerance’s momentum, and a new wave of human trials has now begun.

The ITN alone has already approved eight separate trials, and industry is independently conducting others. (See the tables that follow for details.) These trials, largely unpublicized, mark the coming-of-age of experimental tolerance therapy in humans. But they also represent a huge gamble: that the science of tolerance has matured enough to work, and work safely, in people. If tolerance fails again, and more patients get hurt, then the field may be faced with a new dark age. Is tolerance moving to the clinic too fast, too soon?

Although it has only emerged from obscurity in the last few years, tolerance is not a new idea. In 1944 English immunologist Sir Peter Medawar theorized that rejection takes place when the immune system recognizes that a transplanted graft is “foreign” and must be destroyed. But how this happened, and why the body didn’t attack itself, were complete mysteries. Then Medawar, in 1953, injected fetal mice with adult tissue cells of another mouse strain, and the recipient mice, after birth, proved tolerant to skin grafts from the donor mice. This landmark experiment -- the first to achieve tolerance in animals -- suggested that the body defines as acceptable “self” anything it encounters in fetal development, and unacceptable “foreign” anything it encounters later.

But how the immune system learns to tolerate self remained a mystery for decades. In 1957 Australian researcher F. McFarlane Burnet proposed the “clonal selection theory,” which postulated that clones of lymphocytes, with receptors on their surfaces that recognize self antigens, are selectively deleted during early development. (Medawar and Burnet shared the 1960 Nobel Prize in Medicine.) Gradually it became clear that T-cell tolerance is a learning process that takes place in the thymus, where immature T cells migrate.


Selected Tolerance Trials

Protocol & Investigator	Reagent & Company	Indication	ITN?*	Status
T cell depletion Kevan Herold (Columbia Univ.)	hOKT3 gamma1 (ala-ala) (Johnson & Johnson)	New-onset diabetes	Yes	Phase I/II ongoing. Phase II to begin in October 2001
T cell depletion Marcus Clark (Univ. of Chicago)	hOKT3 gamma1 (ala-ala) (Johnson & Johnson)	Psoriatic arthritis	Yes	Phase II to begin in late 2001
Co-stimulatory blockade (anti-B71 and B72 antibodies) E. William St. Clair (Duke Univ.)	WAY-175143 (American Home Products)	Rheumatoid arthritis	Yes	Phase I pending
T cell depletion in kidney transplantation Stuart Knechtle (Univ. of Wisconsin)	Campath 1H [See story] (Millennium Pharmaceuticals)	Kidney transplantation	Yes	Phase I pending
Co-stimulatory blockade with CTLA4-Ig Samia Khoury (Brigham & Women’s Hospital)	CTLA4-Ig (Repligen)	Early multiple sclerosis	Yes	Phase I pending
Co-stimulatory blockade with anti-CD40 ligand Lloyd Kasper (Dartmouth College)	IDEC-131 (IDEC Pharmaceuticals)	Multiple sclerosis	Yes	Phase I/II

* Immune Tolerance Network


Depending on when and where it happens, the T cell receptor/MHC (major histocompatibility complex) link produces two opposite results. (MHC molecules present antigens to the immune system.) In the lymph nodes, if the T cell receptor matches the MHC/antigen complex, the T cell activates, multiplies and attacks. But when it happens in the thymus, where T cells are exposed to self, these T cells die. Self antigens then can circulate unharmed in perpetuity. Today’s scientists seek to mimic this process and induce tolerance.

Until the 1990s, tolerance approaches were primitive and impractical. Doctors tried feeding antigen to patients by mouth, or irradiated patient lymph nodes. But three discoveries dramatically changed the tolerance landscape and paved the way for today’s clinical trials. First, in the late 1980s it became clear that, for many T cells, the MHC/T-cell receptor interaction wasn’t enough to generate immunity -- a “second signal” was needed. This “co-stimulation” concept galvanized the field. “This second signal really totally changed the landscape, in thinking of how one might go from a… sledgehammer approach to a more specific approach to tolerance induction,” says Bluestone.

Another breakthrough was the gradual acceptance that much tolerance takes place not only in the thymus (“central tolerance”), but also in the lymph nodes (“peripheral tolerance”). Finally, scientists realized that tolerance is a process happening constantly throughout our lives, not just during fetal development. “It’s not just a default pathway of ignorance, but an active pathway,” says Bluestone. “So it’s a pathway we can tap into.”


On the heels of these advances, tolerance seems poised for success in the clinic. But some transplant surgeons say it’s no longer necessary, due to advances in immunosuppressive drugs. Since the early 1980s, when cyclosporine revolutionized transplantation, organ rejection rates have plummeted. More than 90 percent of patients now keep their new kidneys more than a year. (The rate is 84 percent for hearts and 80 percent for livers.) Since 1997 the FDA has approved four new transplantation drugs, and short-term “acute” rejection continues to drop. So, the skeptics ask, why bother with tolerance when we can achieve great immunosuppression with drugs?

Immunosuppression “is very successful,” acknowledges Bluestone, “but it’s very successful at a price.” Long-term “chronic” rejection rates have barely budged over the last two decades. Patients can expect to keep their new kidneys and livers about 10 years, fewer for lungs and pancreases. Given the long waiting lists for all organs, graft failure is a catastrophe for patients.

And immunosuppressive drugs all have nasty side effects: kidney toxicity for cyclosporine and FK506; bone disease, cataracts and diabetes risk for steroids; and dramatically increased risk of infections and non-Hodgkin’s lymphoma for almost all drugs. So, even with today’s improved drugs, “there are compelling reasons to do something at the time of the transplant that doesn’t result in the expense and the complications [of] long-term immunosuppression,” says University of Michigan transplant surgeon Jeffrey Punch. And for autoimmune disease, Bluestone points out, powerful and indiscriminate immunosuppressant drugs are hardly acceptable. “It's somewhat myopic to think that immunosuppression has solved most of our problems,” he says.

Tolerance remains a very alluring goal. “The idea of retraining the immune system, everyone calls it the Holy Grail of transplantation,” says Punch. “It’s what everyone searches for. Because besides a sure Nobel Prize… it would change the face of transplantation.”


By far the trendiest approach to tolerance today is co-stimulatory blockade. Block the “second signal” when presenting antigen to the T cell, the theory goes, and you block T cell activation just for that antigen. The immune system remains otherwise intact to fight other invaders. At the time the idea emerged, in the early '90s, such specificity “was very exciting conceptually,” recalls Bluestone. “You could allow the T cell to get the first signal, and then block the second signal and as a consequence of that shut down only those T cells that got signal One.” Give some donor antigen just before transplant surgery to the patient -- while blocking the second signal -- and you get tolerance to the new organ. The theory was irresistible in its elegance.

But what molecules delivered the second signal? The first important co-stimulatory molecule found was CD28, which is still considered a promising target for blockade, although major problems have surfaced. Later, in the wake of the 1997 Pentagon press conference, the darling molecule became CD40 ligand.

Discovered by German researchers in the early '90s, CD40 ligand, expressed on helper T cells, appeared vital for immune system activation. When an antigen-presenting cell such as a dendritic cell displaying foreign antigen (on MHC) encounters the helper T cell, the required second signal is sent via CD40 ligand. That drives the dendritic cell into a superactivated state, priming killer T cells to hunt down the foreign antigen. Block the second signal by blocking CD40 ligand, the theory went, and you render the dendritic cell impotent. The result: tolerance.


Selected Tolerance Trials

Protocol & Investigator	Reagent & Company	Indication	ITN?*	Status
T cell depletion in islet transplantation James Shapiro (Univ. of Alberta)	“Edmonton Protocol” using daclizumab (Zenapax) (Roche)	Autoimmune diabetes	Yes	Ongoing multi-center Phase I/II trial
Mixed chimerism in kidney transplantation Megan Sykes (Massachusetts General)	None [See story; collaboration with BioTransplant]	Kidney transplantation and myeloma	Yes	Phase II in progress
Mixed chimerism in kidney transplantation David Sachs (Massachusetts General)	None [See story; collaboration with BioTransplant]	Kidney transplantation	Yes	Phase II to begin in late 2001
Co-stimulatory blockade using anti-CD40 ligand (humanized Mab) (IDEC Pharmaceuticals)	IDEC-131	Psoriasis	No	Phase II
Co-stimulatory blockade using anti-CD40 ligand (humanized Mab) (IDEC Pharmaceuticals)	IDEC-131	Immune thrombocytopenic purpura (ITP)	No	Phase II

* Immune Tolerance Network

Antibodies to CD40 ligand were soon tried in animals, with spectacular results. The Pentagon's 1997 stunner was only a prelude. In June 1999, Navy transplant surgeon Allan Kirk announced that nine monkeys briefly treated with Biogen’s version of anti-CD40 ligand all accepted their kidney transplants, for periods then pushing a year-and-a-half, with no immunosuppression -- and no adverse effects. These extraordinary results generated a wave of enthusiasm from powerful government scientists like Tony Fauci, director of the National Institute of Allergy and Infectious Diseases, and prominent NIH immunologist Polly Matzinger, originator of the “danger model” of immunity (which holds that the immune system responds not to foreign antigens but to “danger” signals caused by injury). “If the promise of signal Two blockers holds up, the days of signal One blockers are numbered, and short,” Matzinger wrote enthusiastically two years ago in Nature Medicine (Nat. Med. 5: 616-617 [June 1999]). “By mimicking the body’s own tolerance-inducing mechanisms, we will be able to induce tolerance to transplants at will and then withdraw the drugs to allow the recipient to step back from the abyss of lifelong immunosuppression.”


Five months after Matzinger made this glowing prediction, Biogen shocked the field by suspending its trials. Almost two years later, important questions remain unanswered. Did the drug work at all? How many blood clots, strokes and pulmonary emboli occurred? And were these side effects caused by actions specific to Antova, Biogen’s antibody; by impurities in the antibody prep; or because of the mechanism of action of anti-CD40 ligand itself? This last question is crucial, since co-stimulatory blockade might still work, and because at least one other version of anti-CD40 ligand (presumably targeting a different epitope) is in the clinic. “We don’t know the answer to that really,” says University of Wisconsin transplant surgeon Stuart Knechtle, who participated in Biogen’s kidney transplantation trial. “The company was supposed to solve that problem.”

Biogen isn’t talking -- company officials declined comment in response to questions from Signals -- so it’s hard to know what caused the blood clots. Biogen sponsored Antova trials for six different conditions: kidney transplantation, multiple sclerosis, lupus, hemophilia, islet transplantation in diabetes and immune thrombocytopenic purpura (ITP), a rare autoimmune disease in which the body attacks its own blood. All were suspended on Nov. 2, 1999.

In kidney transplantation, which had succeeded so brilliantly in monkeys, Antova was a disaster. During May's joint annual meeting of the American Society of Transplantation and the American Society of Transplant Surgeons, in a five-minute poster session attended only by about 20 people, the Navy's Kirk described how almost all patients rejected their kidneys (one of them within five days of transplant surgery). Worse, three of the first four patients had a blood clotting episode. One of them, Kirk mentioned, died of a pulmonary embolism, a clot that migrated to the lung. Although such clots can occur in any major surgery, the mortality rate from conventional kidney transplant surgery at one month is only one percent, according to data from the United Network for Organ Sharing (UNOS).

The trial, conducted at three different sites, continued after the death, with Kirk actually increasing the Antova dose while administering anti-coagulant drugs. (He also declined to comment.) The six Biogen trials only were suspended after patients suffered blood clotting events in trials conducted elsewhere (and in different indications). “The worst side effects were in other trials,” says Knechtle, who performed four of the seven kidney transplants. Whether other deaths occurred remains unknown.


In moving forward with human trials, Biogen may have ignored warning signs about its antibody. Earlier studies in Germany (by Richard Kroczek of the Robert Koch Institute and Gert Muller-Berghaus of the Max Planck Institute) showed that activated platelets express CD40 ligand, which acts to initiate an inflammatory cascade in blood vessel walls. Researchers at Massachusetts General Hospital, trying anti-CD40 ligand in monkeys, observed clotting complications similar to what were later seen in humans taking Antova. Following surgery, “if an antibody fixes to those platelets, it sets up an inflammatory process there that leads to complete thrombosis,” explains A. Benedict Cosimi, chief of transplantation at Mass General. “We had observed it before… that’s why we didn’t participate in the [Biogen-sponsored] clinical trials.”

It’s Cosimi’s impression that Biogen and its collaborators knew this. “The clinical trials nevertheless went ahead, and they got into some problems with it, which wasn’t too surprising to us,” he says. Cosimi, who thinks the drug is safe if given with anticoagulants, speculates that inconsistency in the animal thrombosis results led the Antova group to discount the side effects, until it was too late.

Despite the side effects, there are some hints that Antova may have been effective, at least for certain diseases. For lupus, Antova treatment led to a decrease in the frequency of anti-DNA antibody-secreting B cells in patients. (Such antibodies are thought to cause the disease’s symptoms.) “It seemed that it was working,” says trial investigator Richard Furie, chief of rheumatology at North Shore University Hospital in New York. For ITP, increased platelet counts for some patients were reported, suggesting some efficacy. And in hemophilia, one of four patients treated experienced a dramatic reduction in inhibitory antibodies to Factor VIII clotting factor. (The idea was to ‘tolerize’ hemophilic patients otherwise unable to use clotting factor to stop their bleeds.) “The results we obtained suggested we were on the right track,” wrote trial investigator Bruce Ewenstein, of Brigham and Women’s Hospital in Boston, in response to questions from Signals.

But the sad kidney outcome augurs poorly for the tolerance field, because it was so shockingly different from the spectacular results in monkeys. Kirk is now trying to find out why. “There’s some [redundant] pathway that we have just not figured out,” he told his poster session audience in May. “Some co-stimulatory pathway in humans that compensates for blocking CD40 ligand.”


The lesson: Never underestimate the complexity of the human immune system. In fact, new co-stimulatory molecules appear in the scientific literature almost monthly. That might be good -- but it’s probably bad. “The exciting thing might be that one of these molecules will be… even better than the CD28 or the CD40,” says Jonathan Bromberg, professor of surgery at Mount Sinai School of Medicine. “[But] the… pessimistic view is that the immune system is going to have so much redundancy and so many alternative ways of activating, that we’re not going to be able to come up with a very simple, single, elegant way to hit the immune system and get tolerance. The answer right now is: Nobody knows.”

Although some still hope to identify a “master switch” co-stimulatory molecule that trumps the others, Bromberg doesn’t think that’s likely. “We’re going to find more and more co-stimulatory receptors and ligands, and discover that the immune system is a lot more clever than we are,” he says.


Selected Tolerance Trials

Protocol & Investigator	Reagent & Company	Indication	ITN?*	Status
Co-stimulatory blockade using anti-B71 antibody (primatized) (IDEC Pharmaceuticals)	IDEC-114	Psoriasis	No	Phase II
“Toleragen” therapy targeting B cells using DNA construct (La Jolla Pharmaceutical)	LJP 394	Lupus	No	Phase III
Anti-CD2 monoclonal antibody (humanized) (MedImmune; BioTransplant)	MEDI-507	Psoriasis	No	Phase II
Co-stimulatory blockade, CTLA4-Ig (Bristol-Myers Squibb)	BMS-224818	Solid organ transplantation	No	Phase II
CTLA4-Ig (Bristol-Myers Squibb)	BMS-188667	Arthritis	No	Phase II
Selective, anergy-inducing MHC-derived peptide vaccines (Corixa)	AnervaX and AnergiX	Rheumatoid arthritis and multiple sclerosis	No	Phase I

* Immune Tolerance Network

In fact, Kirk recently reported (in Philosophical Transactions of the Royal Society of London, Biological Sciences, 2001 May; 356 (409):696) that some of his original monkeys rejected their kidneys once Antova cleared from the circulation, suggesting that Biogen’s drug never induced tolerance to begin with. “It wasn’t tolerance,” contends Bromberg. “It was just a different form of immunosuppression.”

The failure to translate animal results to humans haunts the tolerance field. “Water, water everywhere but not a drop to drink,” in the words of Harvard transplant immunologist Hugh Auchincloss, speaking at the May transplantation conference. “We know of the 57 varieties of tolerance induction that work in small animals, and yet there is essentially no clinical [protocol] that’s ready for widespread clinical applications.” The human immune system seems to possess a full level of complexity beyond that of primates. This is good for grant-hungry academic researchers, but bad for drug companies.

A good example is CTLA4-Ig. After the discovery of CD28, the first co-stimulatory molecule, immunologists sought a way to block it and induce tolerance. They seized on an already-cloned molecule, CTLA4, which bound to the same ligand as CD28, but more tightly. Bristol-Myers Squibb Co. made a soluble form of the molecule by fusing it to certain immunoglobulin domains -- hence CTLA4-“Ig” -- and tried it out in animals. “In many animal models, including the early work in the '90s that we did, you could show tolerance,” says ITN director Bluestone.

But human tolerance trials of CTLA4-Ig produced inconsistent results. That was partly because doctors couldn’t get enough into patients to work, because it didn't bind tightly enough. But scientists uncovered a more profound obstacle: CTLA’s normal function in the human body, they discovered, is to turn off T cells. So by using CTLA4-Ig to block CD28 binding, they also blocked CTLA4 binding, sometimes making things worse for patients, not better. CTLA4-Ig, in a bizarre way, is an antagonist for itself. “You could… end up exacerbating an immune response in certain settings because you block CTLA4,” says Bluestone.

Two companies, Repligen Corp. and Bristol-Myers Squibb, have since made improved versions of CTLA4-Ig, enhancing its binding affinity. Both molecules are now in the clinic, but no one is under the illusion any more that CTLA4-Ig is a magic bullet for tolerance. Complexity, again, dashed that hope.


This complexity has caused many pharmaceutical and biotech companies to leave the tolerance field or avoid it altogether. Another deterrent: Tolerance therapy is, by definition, short-term, while immunosuppression is for life. Why develop an antibody that’s given only three times when you can make a pill that someone has to take every day for 20 years?

Bluestone encounters this mentality all the time. But, he says, it’s hardly universal. “There are enough small biotech and even larger companies that are doing [tolerance],” he says. “[They] realize that, if… someone else succeeds at doing it, and they’re not players, then they’ve lost both their immunosuppression market and they’re not in the tolerance market.” But only about 26,000 solid organ transplants a year take place in the U.S. -- not enough to make it worthwhile for many companies. IDEC Pharmaceuticals Corp., for example, isn’t testing IDEC-131, its version of anti-CD40 ligand, in transplantation. “For us, it does not make sense today to conduct transplantation trials with this particular antibody, or any other antibody,” says chief scientific officer Nabil Hanna. “Our marketing folks… believe it is not a large indication at this time.”

Autoimmunity, of course, is a different story. Rheumatoid arthritis, allergies, multiple sclerosis, asthma -- these are irresistibly lucrative markets. IDEC is trying its IDEC-131 in several autoimmune diseases. (The drug has not caused thrombosis, and IDEC contends that the side effect was peculiar to the Biogen version of the antibody.) But achieving tolerance is harder in autoimmunity. The typical organ transplant patient has a clean immune system that, in theory, can be primed in advance to accept donor cells (from living donors, anyway). “With an autoimmune patient, that patient is already all primed [to attack itself] with all sorts of memory activated T cells and B cells all over the body,” says Mt. Sinai's Bromberg. “[Tolerance] is a much more difficult task.”


Tolerance may first find success in organ transplantation, but proving it will be hard. There’s a major ethical dilemma: You can’t demonstrate tolerance without removing immunosuppression, but if you remove immunosuppression you risk the organ. “One of the problems in developing tolerance is, the results are excellent right now without it,” says the University of Michigan's Punch. “The standard is extremely high. If you do it and it works three-quarters of the time, that’s just no good. Because we can get 90 percent results with conventional immunosuppression, in kidneys, anyway.” How can doctors remove immunosuppression without exposing patients to life-threatening organ rejection?

One way is to go forward with tolerance plus immunosuppression, then remove immunosuppression in patients who demonstrate tolerance in a lab assay. The trouble is, a reliable tolerance assay does not yet exist. The ITN is pouring money into a variety of methods to test if patients are accepting or rejecting their organs: genetic profiling using arrays or real-time PCR; cytokine analysis; surrogate mouse models; and a tetramer assay originally developed for cancer vaccine trials.

No one knows how long it will take. “These are very hard assays to develop, in part because our current immunosuppressive drugs are so good that they cloud our ability to look at these things,” says Bluestone.

Meanwhile, tolerance trials are going forward -- without an assay. Besides co-stimulatory blockade, there are two major approaches: T-cell depletion and mixed chimerism. T-cell depletion seeks to eliminate the body’s current inventory of T cells and tolerize the new, immature T cells to the foreign graft before transplantation. “As [the] immune system matures and develops more memory and experience, it’s harder and harder to tolerize,” explains the University of Wisconsin's Knechtle. “So we’re converting the immune system back to a more immature state and [providing] a more receptive environment for tolerizing to the new antigen.”

Such depleting antibodies are nothing new: Ortho Biotech Inc.’s OKT3 was the very first FDA-approved monoclonal antibody (for organ transplants, in 1986). But OKT3 has a serious side effect: cytokine release syndrome, which can lead to central nervous system damage and heart dysfunction. Safer, stripped-down versions of OKT3, which targets the CD3 antigen on T cells, are now ready for the clinic (see tables), but no one knows if they can induce tolerance.

Knechtle is trying Campath 1H, a powerful depleting antibody which Burroughs Wellcome abandoned in 1994 after disastrous clinical trial results in arthritis. (For details of Campath's history, see the Signals article "Campath's Path To Stardom.") In May 2001, the FDA approved Campath (now licensed by Millennium Pharmaceuticals Inc.) for patients with a rare leukemia. Knechtle has tried it, at a much lower dose, for kidney transplantation (along with the immunosuppressant drug rapamycin) with encouraging results, and is about to launch a new ITN-sponsored trial. “The burning question is, of course, is tolerance induced with this strategy?” Knechtle says.

The only way to tell would be to stop the rapamycin, and that can’t ethically happen without using a tolerance assay. When will one be ready? “I don’t know,” says Knechtle. “I don’t like to put dates on things that haven’t happened yet.”

In an even bolder experiment, two tolerance trials aren’t waiting for an assay. Doctors at Massachusetts General Hospital (and two other centers) will transplant kidneys after giving recipients bone marrow from the donor, creating a state of intermingled lymphocytes, or “mixed chimerism,” in patients. In theory, the donor lymphocytes will migrate to the thymus, along with antigen from the donor organ, and there induce tolerance to the new kidney. After three months on cyclosporine, patients who demonstrate engraftment of donor marrow will then go off immunosuppression. (The first trial will enroll myeloma patients, who benefit from the bone marrow infusion, but the second will only involve kidney transplants.) “If everything is stable, we stop all the drugs,” says Cosimi.

Cosimi’s group has already treated two patients, and both are doing well 32 and 8 months, respectively, post-transplant. “It’s courageous of the first few patients who do it,” acknowledges Cosimi, who believes that excellent results in animal models, and the high long-term graft failure rate using immunosuppressive drugs, justify the risk of stopping immunosuppression. “Obviously, for the patient to undergo this kind of therapy, they’re taking a higher risk early on,” he notes. “But for the long term probably the risk is much lower.”


Other tolerance advocates, to prove their therapies work, will also have to take patients off immunosuppressive drugs. But many transplant surgeons believe minimal immunosuppression is a more realistic goal than tolerance, and one that avoids putting patients’ organs at risk. They hope to wean transplant patients off the current three- or four-drug regimens down to a less toxic one or two, at lower doses. This camp has adopted terms like “near tolerance” and “almost tolerance” to describe the concept.

The terms bother pure tolerance advocates like Jeff Bluestone. “There’s no such thing as ‘almost tolerance,’” he says. “You can’t be a little pregnant.” Bluestone believes efforts to reduce immunosuppression should go forward, but separate from tolerance research. “There are millions of people that have a need for tolerance that would not benefit from these so-called ‘reduced immunosuppressive therapies.’”

So tolerance now goes forward in humans. Will it succeed? “You have no way to know,” says Ohio State University immunologist Charlie Orosz. “Some group could stumble on… the right [tolerance] strategy tomorrow, and within a year prove that this system is working very well, and virtually everyone will be using it in five years. Or it could be that nobody gives a damn for the next 20 years.” The potential rewards -- in terms of scientific prestige and economic return -- are driving the field’s star players forward. But the Antova debacle should give them pause. Biogen’s anti-CD40 ligand was as close to a sure bet as the field has ever had, and it failed miserably, leaving a patient dead in its wake. A few more such outcomes and tolerance might well end up looking back at 2001 as an illusory Golden Age.