At NYU, my colleagues and I were following the successful approval process of Remicade for Crohn’s disease and rheumatoid arthritis with personal satisfaction. The results of our work were helping to alleviate suffering from these diseases. Yet we didn’t pop bottles of expensive champagne. Our reaction was muted, because in 1998–99 it was still far from clear that Remicade would have a very significant impact on the diseases for which it had been licensed.
In both Crohn’s disease and in rheumatoid arthritis the treatment was approved only for short-term applications. Remicade did not cure the underlying disease and the symptoms returned when treatment was discontinued. There was only limited evidence suggesting a favorable therapeutic response when Remicade injections were repeated—perhaps three or four times—but there was not enough experience to know if long-term treatment was feasible. In principle, once a drug is approved, a physician may prescribe it essentially without restrictions, but in practice most physicians are reluctant to deviate from FDA recommendations, for liability reasons.
Questions remained: Would long-term administration of Remicade damage a person’s immune system? Would defenses against infectious agents (bacteria, viruses, fungi, and parasitic agents) be too severely compromised? There was a plethora of animal data indicating that blocking TNF action weakens the body’s defenses to infectious agents, such as Mycobacterium tuberculosis, the causative agent of tuberculosis. Also, would the body’s immune response against the chimeric antibody protein constituting Remicade render repeated administration impossible? It was well documented that the human body may produce an immune response to the administration of chimeric antibodies.
No one had answers to these questions and my own bias during this time was that, more likely than not, these issues would preclude the wider use of Remicade. If long-term administration wasn’t possible, the benefit of the treatment would amount to nothing more than short-term relief from the symptoms of two miserable diseases. Knowing that patients’ disease symptoms are likely to relapse would almost certainly discourage physicians from prescribing the drug.
There were other obstacles. The treatment was expensive—even short-term administration would cost several thousand dollars. Remicade had to be administered by slow intravenous infusion in a specialized medical facility, usually inside a hospital or a specially equipped outpatient clinic. Finally, as a monoclonal antibody directed against TNF, Remicade embodied a radically new treatment concept, and, as with most new therapies, medical practitioners were not likely to rush to embrace it. Not surprisingly, initial Remicade sales were slow.
But there was one piece of news suggesting that not everyone was bearish about the prospects of Remicade: in 1999 Johnson & Johnson acquired Centocor for $4.9 billion. This seemed to be a very generous price given that Centocor’s product sales were still modest.
In retrospect, Johnson & Johnson got a huge bargain. Subsequent clinical trials sponsored by Johnson & Johnson showed that Remicade was generally safe to administer to patients over a long term. In addition to Crohn’s disease (including pediatric Crohn’s disease) and rheumatoid arthritis, Remicade has also been approved—in the US and around the globe—for the use in ankylosing spondylitis (an inflammatory disease of the spine), ulceratitive colitis (another form of inflammatory bowel disease, including its pediatric form), juvenile idiopathic arthritis (a form of inflammatory arthritis occurring in children), psoriatic arthritis, and plaque psoriasis. It is estimated that through 2014 three million patients had been treated with Remicade worldwide.
The success of Remicade paved the way for the development and introduction of other drugs with a similar mechanism of action. Four other TNF-blocking agents are now approved for a variety of indications. (They are Amgen’s Enbrel, AbbVie’s Humira, UCB’s Cimzia, and Johnson & Johnson’s Simponi.) It is fair to say that anti-TNF therapy has revolutionized the management of many chronic inflammatory disorders.
Most drugs are approved for one disease or perhaps a couple of closely related diseases. Why is Remicade—along with the other anti-TNF agents—effective in so many diseases that at first glance seem to have very little in common? All of the diseases responsive to Remicade (and other TNF-blocking agents) are autoimmune chronic inflammatory disorders, in the pathogenesis of which TNF is now known to play a key role. In autoimmunity—dubbed horror autotoxicus by Paul Ehrlich, a pioneer of the study of autoimmunity—the immune system of the body attacks its own tissues and organs.
The damage inflicted by autoimmune diseases results from the action of antibodies and cells that constitute the immune system, the very same cells that are important in protecting bodies from the attack by pathogens. The result of such an attack is usually inflammation. In rheumatoid arthritis—a prime example of an autoimmune disease—joints become swollen, warm, and tender as a result of an inflammation of the soft tissue inside joints.
The site and extent of inflammation varies with each type of autoimmune disorder. Certain cytokines play important roles in initiating and sustaining inflammation in autoimmune disorders, with TNF playing a key role in some, but not all, autoimmune conditions. Because of its important role in the inflammatory response accompanying many autoimmune disorders, curbing the action of TNF with Remicade or other TNF-blocking treatments can be effective in numerous conditions.
Yet treatment with Remicade and other TNF-blocking agents does not help all patients suffering from the diseases for which these treatments are recommended. For reasons that are still poorly understood, around 30–40 percent of patients with rheumatoid arthritis, Crohn’s disease, and the other conditions for which Remicade is approved do not respond adequately to these treatments. Unfortunately, it is still impossible to predict who will and who will not have a favorable response.
Moreover, Remicade and the other TNF-blocking agents can produce serious side effects in patients. Among the most common is an increased susceptibility to bacterial, viral, and fungal infections.
The treatments must be used with great caution in people who are infected with some pathogens even if they do not show overt signs of disease. For example, in people who have a dormant, symptomless infection with Mycobacterium tuberculosis, the administration of anti-TNF medications can lead to an activation of the infection resulting in overt tuberculosis. Appropriate precautions need to be taken with such patients. There is also a long-standing—though still unconfirmed—suspicion that in rare instances treatment with Remicade and other TNF-blocking agents might precipitate the occurrence of lymphomas and other malignancies, especially in children and adolescent patients.
And blocking TNF is not effective in all autoimmune diseases. In fact, in some autoimmune disorders, including multiple sclerosis and systemic lupus erythematosus (a common autoimmune disorder of women affecting the skin and multiple organs), Remicade and other anti-TNF agents are not only ineffective, they may make the disease worse. So these treatments can be used only in patients in whom the benefit from successful treatment outweighs the potential risks.
Another drawback is that anti-TNF medications, though bringing significant relief to patients who respond to the therapies, do not cure the underlying disease. If the treatment is stopped, patients relapse and the disease flares up anew. There are exceptions—I have met patients with Crohn’s disease who after responding to Remicade treatment had gone off therapy and yet their disease failed to relapse for reasons that are not understood.
Thus autoimmune diseases responsive to treatment with anti-TNF agents have joined the list of many other chronic diseases for which there exist treatments providing powerful relief from the symptoms of the disease without eliminating the root cause of the illness. Diseases that can be controlled by appropriate treatment, but not cured, include diabetes that can be managed by treatment with insulin and other drugs, and even AIDS that can be kept in check by the administration of antiviral drugs. A major form of diabetes, type 1 diabetes, though not known to be responsive to anti-TNF therapy, is an autoimmune disease. AIDS is caused by infection with a virus, HIV, that takes up permanent residence inside certain vital white blood cells—especially so-called helper T lymphocytes—gradually leading to their destruction.
That anti-TNF therapies don’t cure the underlying autoimmune disease is inherent in their mechanism of action. All of these agents act by blocking the activity of TNF, a key element in the chain of events causing inflammation. Inflammation, in turn, is the cause of the pathology in most autoimmune diseases, and upon disabling TNF and suppressing inflammation, disease symptoms cease. However, TNF is not the root cause of the disease—something else is—and if the administration of the anti-TNF agent is halted, TNF production and action generally resume, which is the reason why disease symptoms flare up anew.
Why is there no cure available for autoimmune diseases? The main reason is that we don’t fully understand the underlying causes of these disorders. Indications are that certain environmental factors, including infections with bacteria or viruses, along with genetic factors built into our own DNA, determine whether we develop autoimmunity. Much more work will be needed before these factors become better known. And even when we learn the underlying cause of these illnesses, it still may take a long time to develop a cure. For example, we do understand the cause of diabetes—it is either the failure of beta cells in the pancreas to produce enough insulin or the failure of other cells in the body to respond to its action—and yet, available medications, though effective, generally do not cure the disease. There is hope that, as a result of more research and progress in fields like stem cell research and gene therapy, the situation will change in the foreseeable future.
Despite the limitations, Remicade and other TNF-blocking agents can be highly beneficial. Even though these therapies do not eliminate the underlying disease, they do much more than alleviate the symptoms of disease.
Specifically, in inflammatory bowel disease (either Crohn’s disease or ulcerative colitis), anti-TNF agents can induce a clinical remission, promote the healing of lesions, prevent the appearance of new lesions, and eliminate the use of potentially harmful corticosteroids or the need for drastic surgery, thus often restoring the ability of patients to live normal, productive lives. Prior to the introduction of the anti-TNF agents, some patients with ulcerative colitis or Crohn’s disease required colostomy—the surgical removal of much of their large intestine (colon) and attachment of its remaining portion to the outside of the abdominal wall. Intestinal waste would then flow through the opening in the abdomen into a colostomy bag. Anti-TNF therapy has largely eliminated the need for this surgery. Another benefit of anti-TNF therapy is a reduction in the high risk of gastrointestinal cancers in Crohn’s disease and ulcerative colitis patients that correlates with the duration of the disease symptoms and severity of inflammation.
Similarly, in conditions whose major characteristic is joint inflammation (rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis), anti-TNF agents can arrest the progression of structural damage and improve physical function, thereby preventing the onset of a more disabling illness. These are no small accomplishments.
I’ve had the pleasure of meeting many patients treated with Remicade, who tell me how dramatically their lives have improved as a result of the medication.
And then there are letters and e-mails from patients or relatives of patients—many of them. Here are excerpts from three recent e-mails:
I’m nineteen years old now, but when I was around thirteen years old I was diagnosed with Crohn’s disease. When I was first diagnosed, and for about three years after that, my life was turning into a nightmare. The treatments I underwent at first, such as steroids, never worked. Many times they left me feeling more sick than before. I was in constant pain, and losing weight so rapidly that my mother had come to the point where she was not going to send me back to school. When my GI doctor told us about Remicade, it was our last hope, I was so sick. From the first twenty-four hours since I was hooked up to my IV and given Remicade, I was a completely new person. I cannot even begin to tell you what it feels like to be better after being sick for so long. It gives you a completely different outlook [toward] the world around you and your purpose in the world . . .
I was diagnosed with ulcerative colitis at age thirteen and before my treatment with Remicade my life was incredibly difficult dealing with this painful condition. I am now thirty-nine years old and this miracle drug continues to work after starting it in Nov. 2006. I haven’t had any problems whatsoever in these six years. In fact I have two healthy children and took the drug during my pregnancies . . . I just wanted you to know . . .
Five years ago our son, then seventeen years old, was diagnosed with Crohn’s disease. Today, after receiving Remicade for four years, he had a completely normal colonoscopy. While dealing with this has not been easy for him, he has no way of knowing what we—his parents—know, that Crohn’s disease used to mean pain and embarrassment, frequent hospitalizations, and regular surgeries. Our son has had a very normal life; even though he has had to take a handful of pills daily and receive intravenous medication every six to eight weeks, he has been able, along with his father, to raise thousands of dollars for Crohn’s research by riding his bike fifty miles every year. He graduates from college in two weeks, a healthy, vibrant, and creative young man . . . Our child has had a relatively pain-free, healthy life. That has made a world of difference . . .
These letters remind me of my own good fortune: to have had the opportunity to contribute to a scientific venture that has made a real difference in the lives of millions of people.
Remicade became a huge economic success for Centocor (which in 2011 changed its name to Janssen Biotech) and Johnson & Johnson. When Remicade was first approved in 1998, initial sales were relatively modest. By 2001, with Remicade on the market for its third full year, worldwide sales reached $850 million. I had thought that was a very respectable result and had not expected much further growth. Ten years later, in 2011, worldwide sales exceeded $7 billion. From the outset, Remicade sales ran ahead not only of my own expectations, but also of the pundits’ forecasts. In 2013, according to PMLiVE, a pharmaceutical industry newsletter, Remicade was the second-highest-selling drug in the world, with sales of $10.1 billion.
Billions of dollars are spent annually on researching and developing drugs, but it is estimated that just one out of approximately every five to ten thousand compounds studied in preclinical trials is ever going to be approved by the FDA and find its way to market. And of the drugs that are approved, just a very select few achieve blockbuster status, defined as reaching more than $1 billion in annual sales.
In 2013 the three highest-selling drugs in the world were all TNF inhibitors, a class of therapeutics whose development and introduction was trailblazed by Remicade. The number one best-selling drug was Humira—another TNF-blocking monoclonal antibody, manufactured and marketed by AbbVie, a company that in 2011 split off Abbott Laboratories—with 2013 worldwide sales stated by PMLiVE to have reached $11.1 billion. Approved by the FDA in 2000, about two years after Remicade, Humira is almost identical to Remicade in its efficacy and therapeutic spectrum. Whereas Remicade is a chimeric human-mouse antibody protein, with around 30 percent of its sequence stemming from the original mouse antibody, Humira is a fully human protein.
I am not aware of evidence that Remicade is less efficacious or that there are more complications observed with Remicade than with Humira. However, some patients and doctors may prefer Humira’s route of administration; it is applied by injection under the skin, usually self-administered by the patient, whereas Remicade must be given by slow intravenous infusion in a specialized facility.
The third-highest-selling drug in the world in 2013, according to PMLiVE’s “top pharma list,” was Amgen’s Enbrel, a synthetic antibody-like TNF inhibitor, with sales of $8.9 billion. Enbrel too is used for indications almost identical to those of Remicade. Enbrel was first approved for use in rheumatoid arthritis in November 1998, a few months after Remicade’s approval for Crohn’s disease. Another pharmaceutical newsletter, FierceBiotech, listed Enbrel as the second-highest-selling drug in the world in 2013, with Humira in first place and Remicade in third.
In 2014, Humira retained its first place on PMLiVE’s top pharma list, but Remicade was reduced to third place by a newcomer—Solvadi, the “one thousand dollar pill” against hepatitis C, manufactured by Gilead Science.
Remicade has turned out to be a boon for NYU and, yes, for its NYU-based inventors. According to the licensing agreement signed by NYU and Centocor in 1984, NYU is due a royalty based on quarterly worldwide sales of Remicade. NYU, in turn, pays a portion of the collected royalties to Jimmy Le and me. Royalty payments collected by NYU for Remicade sales are not publicly disclosed, but it is safe to assume that by now they will have exceeded $1 billion. The medical and commercial success of Remicade illustrates the value of university-based biomedical research for therapeutic advances and economic progress. (I hope policymakers in Washington will read these lines.)
Naturally, no party goes on forever. Most foreign Remicade patents expired in February 2015, with US patents due to run out in 2018. When patents covering conventional medications run out, the products may be replaced with generic drugs.
Monoclonal antibodies and other biologicals, usually large protein molecules produced by methods involving recombinant DNA technology, are more complex than conventional small-molecule drugs that can be defined by chemical formulas. Biologics such as Remicade are not only structurally more complicated, their characteristics are also more likely to be influenced by changes in the manufacturing process.
Regulatory agencies in the US and in Europe have developed guidelines for the approval process of generic biologic drugs, referred to as biosimilars. In order to be approved for marketing, biosimilars must be shown to be close in properties to the parent biological product as demonstrated through analytical, animal, and clinical studies. The main yardstick used for the approval of biosimilar biologic drugs is the demonstration that they are comparable to the parent drug in clinical efficacy and safety.
The first biosimilar monoclonal antibody recommended for approval by the European Medicines Agency in October 2013 was a biosimilar Remicade, developed under the name Inflectra by the Korean company Celltrion. Inflectra and another Remicade biosimilar are already distributed in some European countries and in Canada, where Remicade is no longer under patent protection. According to news reports Inflectra is priced at about 20 to 30 percent below the annual cost of $10,000–$20,000 per patient for branded Remicade sold by Johnson & Johnson or Merck & Co., Johnson & Johnson’s commercial partner in Europe. Additional fees that may be charged for the administration of the drug are likely to be the same for all forms of Remicade treatment.
I hope that introduction of a biosimilar Remicade will indeed make the treatment I helped to develop less costly and accessible to more patients who can benefit from its use.
People, including some less experienced scientists, often believe that scientific breakthroughs proceed in a logical and straightforward way from the eureka moment, when the discovery is supposedly conceived, presumably in the bathtub or—in more modern times—in the shower, through the validation of the original idea and the confirmation of the significance of the discovery, straight to the Nobel Prize or whatever the appropriate accolades may be.
In reality, as pointed out by James Watson, Nobel Prize–winning codiscoverer of the structure of DNA, in his succinct book, The Double Helix, “science seldom proceeds in the straightforward logical manner imagined by outsiders.”
The A2 antibody that eventually became cA2 and Remicade is an object lesson in how science does not proceed in a “straightforward logical manner.” It was first conceived as a tool for the development of a diagnostic test, with some vague allusions to its possible worth in medical applications related to autoimmunity and malignant tumors. When, some years later, it was thought that it might be useful as a therapeutic agent, the intended disease target was sepsis. When the sepsis trial failed to show positive results, the trajectory changed once again as cA2 was found to provide relief to patients with severe rheumatoid arthritis. The ultimate outcome—a therapy with profound effectiveness in a variety of severe autoimmune disorders—is a far cry from the original idea.
Serendipity played a huge role in getting the project off the ground and in seeing it through to a successful conclusion. The project would not have materialized if I had not run into Michael Wall during the 1982 meeting at Haverford College or if I had not accepted his invitation to visit the newly opened Centocor laboratories. The project would not have happened if Jimmy Le—then freshly trained in monoclonal antibody technology—had not joined my group. Remicade would almost certainly not have become what it is today if Marc Feldmann had not been able to convince Centocor (with the help of Feldmann’s former colleague, Jim Woody) to provide him and Tiny Maini with cA2 for the first clinical trial in rheumatoid arthritis. I could go on.
Equally important in the successful outcome was the productive collaboration we were able to establish between my laboratory at NYU and the scientists and management at Centocor. Collaborations between academia and industry often falter because of their different cultures and because scientists in general with their strong egos have difficulty sharing credit. We got along and enjoyed our close interactions with the enthusiastic and dedicated Centocor scientists. Michael Wall and Hubert Schoemaker cared deeply about Centocor’s success in the world of business, but they were also interested in the science behind our joint project.
It is fortunate too that my lab recognized the importance of TNF at a time when only a small group of people even knew that such a thing existed. Personal contacts played a role. The fact that I was friends with Lloyd Old very likely influenced my decision to look for TNF in the materials we generated when producing IFN-gamma in my laboratory. In hindsight, there is a logical path from my experiments with interferon to cA2, yet in the moment, the path was not always clear. Yes, a similar drug may well have been developed elsewhere, as a result of other coincidences and serendipities, but these are the twists and turns that produced Remicade.
Or perhaps it can be all condensed to a sentence from E. B. White’s love letter to the city, Here is New York: “No one should come to New York to live unless he is willing to be lucky.” Amen.
1988-89
Monoclonal antibody specific for human TNF (named “A2”) is generated from a laboratory mouse at NYU School of Medicine
1990-91
Team of scientists at biotechnology company Centocor converts A2 antibody by genetic manipulation into a “chimeric,” predominantly human, monoclonal antibody (named “cA2”)
1991
Preclinical testing of cA2 antibody at Centocor and NYU School of Medicine in test tube experiments and laboratory animals shows cA2 is highly potent and selective, potentially suitable for human use
1991-92
Clinical trial of cA2 in patients with sepsis showed no significant benefit
1992
cA2 used successfully in rheumatoid arthritis patients by M. Feldmann, R. Maini, and colleagues
1993
cA2 used successfully in Crohn’s disease patients by S. van Deventer and colleagues
1998
cA2 (by then named Remicade/infliximab) approved by the Food and Drug Administration (FDA) for the treatment of Crohn’s disease
1999
FDA approval for rheumatoid arthritis
2004
FDA approval for ankylosing spondylitis
2005
FDA approval for psoriatic arthritis
FDA approval for ulcerative colitis
2006
FDA approval for severe plaque psoriasis
2013
Remicade/infliximab is the second-highest-grossing drug in the world
2014
Around three million patients treated worldwide with Remicade/infliximab