Abstract: Nearly all therapeutic proteins induce antibodies in patients. However this immunogenicity has been neglected in the use of these products, even though the antibodies may have severe consequences. During the last few years, progress has been made in understanding why patients do not tolerate these protein therapeutic products and also how to manage the problem of immunogenicity.
About Intolerant Patients and Tolerant Mice
Therapeutic proteins such as growth factors, cytokines, and monoclonal antibodies are increasingly being used and have revolutionized the treatment of many life-threatening and debilitating chronic diseases. Therapeutic proteins do not have the intrinsic toxicity like small molecule drugs and their side effects are mainly caused by an exaggerated pharmacodynamic effect (Clarke, 2010).
The main limitation of their use is immunogenicity. Nearly all therapeutic proteins induce antibodies (Schellekens, 2002a; 2002b). Sometimes the immunogenic response leads to failure of a drug’s development during clinical trials (Moreland et al., 2000; Rau et al., 2003). For the therapeutic proteins that have reached the market, their immunogenicity may interfere with efficacy and may lead to serious and sometimes life-threatening complications (Casadevall et al., 2002; Mayer and Young, 2006).
Treatment with therapeutic proteins is expensive and the costs may exceed 100,000 U.S. dollars per year. Hundreds of millions of dollars of the already strained health systems are wasted in the use of therapeutic proteins in patients who are not responding to the treatment because of antibodies developed against them (Hartung et al., 2005; de Vries et al., 2007; Sorensen, 2008).
The scientific, clinical, and economical implications of immunogenicity are becoming increasingly clear. In this article I will try to review what we know about the causes and the consequences of antibody induction. And also what clinicians can do to manage the problem in daily practice.
Causes of Immunogenicity of Therapeutic Proteins
The proteins used before the invention of recombinant DNA technology were mainly of animal origin such as bovine or porcine insulin. The human natural proteins such as growth hormone extracted from cadaver pituitary glands and donor derived clotting factors were used mainly in children with an innate deficiency and they lack immune tolerance to these products. The induction of antibodies by these products was considered to be the inevitable result of exposing the patients to non-self antigen.
The advent of the recombinant DNA technology enabled the production of copies of human proteins for which patients were immune tolerant and therefore immunogenicity was considered unlikely. However, nearly all these products also proved to induce antibodies, sometimes even in the majority of patients.
This experience has repeated itself during the different generations of monoclonal antibodies. The first monoclonal antibodies produced by the technology developed by Kohler and Milstein were of murine origin and proved to be highly immunogenic in patients. The murine sequences of the antibody constant regions were exchanged with the human equivalents, resulting in chimeric antibodies. Further humanization was achieved by grafting antigen binding parts, the complementarity-determining regions (CDR) of murine antibodies in a human IgG backbone. The immunogenicity of chimeric and humanized antibodies is less than that of murine monoclonal antibodies, but can still be substantial. Fully human monoclonal antibodies produced by phage display technology or transgenic mice were predicted to be non-immunogenic in patients (Yang et al., 2001). However, this proved to be incorrect.
Even completely human proteins induce antibodies, and because these antibodies cross-react with endogenous homologues they are by definition the result of breaking tolerance. Human proteins are foreign proteins for experimental animals and are therefore not used to study breaking of tolerance. Over the past 8 years we have studied the induction of antibodies by interferons in transgenic mice that are immune-tolerant to the expressed different proteins. By modifying the proteins, we try to correlate protein characteristics to breaking tolerance (Hermeling et al., 2005; van Beers et al., 2010).
High order aggregation has been identified in our models as the cause of breaking B cell tolerance (Rose et al., 1997). Also in clinical studies a strong correlation has been found between immunogenicity and the presence of aggregations (Ryff, 1997; Rosenberg, 2006). We reasoned that aggregates may form the repeated epitope structure which Zinkernagel proposed as being capable of directly activating B cells (Bachman and Zinkernagel , 1997).
Our models also enable us to study the characteristics of the immune response to human therapeutic proteins. It strongly resembles the T cell independent response described for polysaccharide vaccines with restricted affinity maturation, lack of induction of memory, and lack of effect of adjuvants (Gonzalez-Fernandez et al., 2008; Sauerborn et al., 2010).
Types of Immune Response: Breaking Tolerance Versus Vaccine Type of Reaction
Most therapeutic proteins are copies of human proteins and they induce the production of antibodies by breaking B cell tolerance. Induction of antibodies by human proteins such as the interferons and the epoetins (synthetic form of erythropoietin) in general only occurs after months of chronic treatment. The incidence varies widely between different products. It may be rare as in the case of epoetins alpha and it may occur in the majority of patients as with interferon beta-1b. These antibodies in general disappear rapidly after stopping treatment or may also disappear during the course of treatment.
Proteins of animal or microbial origin such as bovine adenosine deamidase (ADA) or microbial streptokinase induce a classical immune response comparable to the response to a vaccine. The great majority of patients will produce antibodies after a single or a few injections and the antibodies may persist for years.
Both types of immune response may occur with the same product in the same population. In hemophilia patients, immune tolerance may exist in a number of patients depending on their type of factor VIII gene defect. So in some patients the antibody response is based on breaking tolerance. For the non-tolerant patients, factor VIII is a foreign protein and the antibodies are the result of the classical activation of the immune system.
Factors Influencing the Immunogenicity of Therapeutic Proteins
The primary factor by which therapeutic proteins of non-human origin such as streptokinase or asparaginase induce antibodies is the level of non-self. The only identified primary cause by which human proteins induce antibodies is the presence of aggregations.
Many factors may alter the immune response such as genetic factors, although the role of HLA type is not clear (Schellekens, 2005). Patients treated with tumor necrosis factor (TNF) inhibitors showing an immune reaction to a specific product are more likely to also make antibodies to another product suggesting that the immune reaction has a genetic predisposition (Bartelds et al., 2009).
The type of disease being treated and the concomitant therapy also affect the immunogenicity. Cancer patients are less likely to produce antibodies than patients with other diseases. Apparently, these patients are immune depressed by their disease and/or anticancer treatment.
The route of administration may influence the immunogenicity of therapeutic proteins. Subcutaneous treatment is associated with the highest risk and intravenous administration with the lowest risk.
The Clinical Consequences of Antibodies
Studies of the clinical consequences of antibodies induced by therapeutic proteins have been hampered by the lack of standardization of the assays, which has led to conflicting results about both the incidence of immunogenicity and its biological effects (Schellekens et al., 1997; Schellekens, 2008). There have been observations that patients who produce antibodies initially do better, as an epiphenomenon of an activated immune system or because the low affinity antibodies during the initial antibody response may extend the half-life of the therapeutic protein. Carefully designed follow-up studies are needed to fully explore the effects of antibodies.
In diseases like multiple sclerosis, the unpredictable course and the relative modest clinical effect of the therapeutic proteins were major hurdles for showing unambiguously that loss of antibodies’ efficacy occurred. Because the sometimes long periods between start of treatment and appearance of antibodies, the clinical consequences of immunogenicity only became clear after the products have been in clinical use for a number of years.
The biological effect of the antibodies is dependent on their type, level, and pharmacokinetics. In many cases the immunogenicity is restricted to triggering a transient appearance of a low level of binding antibodies with no clinical effect. However a persisting level of neutralizing antibodies inhibits efficacy. There are an increasing number of potential therapeutic proteins failing during the final clinical development stage because of their immunogenicity. Also there are now well documented cases of therapeutic drugs on the market of which the clinical efficacy is restricted by antibody formation, such as the interferons, TNF-inhibitors, and enzyme replacement therapy. Because antibody testing is rarely part of the clinical decision making and because therapeutic proteins are, in general, expensive, the healthcare systems are wasting hundreds of millions worldwide for ineffective treatments.
The extent of immunogenicity contributed to the adverse effects is dependent on the amount of a protein therapeutic product administered. With products administered in relatively low doses like the interferons, the first sign resulting from immunogenicity in patients is often the disappearance of the flue-like symptoms, a common side effect caused by the interferons. With some monoclonal antibodies given in relative high amounts, immune complexes can be formed as a consequence of immunogenicity, and the adverse effects may intensify. These immune complexes may lead to transfusion reactions, anaphylactoid symptoms, and serum sickness.
The most dramatic clinical consequence of immunogenicity occurs if the antibodies induced by the product cross-neutralize an endogenous factor with an important biological function. A good example is the reformulation of an epoetins alpha in 1998 which resulted in an upsurge of Pure Red Cell Aplasia (PRCA) cases in Europe, Australia, and Canada where the product was marketed. The formulation change resulted in an immunogenic response in about 250 subcutaneously treated patients with chronic renal failure with neutralizing antibodies cross-reacting with endogenous erythropoietin. The upsurge was stopped by the end of 2002 after the subcutaneous use of the product was contraindicated. However, the problem has reappeared in Thailand recently (Praditpornsilpa et al., 2009).
Management of the Immunogenicity
Prevention at the level of the product is the best way to manage the problem of immunogenicity. However there are no tools that fully predict the immunogenicity of a therapeutic product other than prolonged clinical trials. There are claims of technologies to de-immunize proteins based on in silico algorithms or in vitro assays of the epitope make-up of a protein. There is hardly any clinical evidence showing the success of these approaches.
The presence of aggregation is the best predictor of the immunogenicity of therapeutic proteins. However, it is not known at what level and what type and size product aggregates are inducing antibodies. Therefore, optimal production, purification, and formulation are the best ways to prevent an immunogenic response.
Immunosuppressive treatment is used to prevent antibody induction with specific products but only for conditions for which immune suppression is also one of the treatment options.
There are several options for clinicians to manage antibody-positive patients, depending on the clinical consequences of the antibodies. In the case of low levels of antibodies with no signs of loss of efficacy or side effects, treatment can be maintained. However in many cases binding antibodies precede the appearance of neutralizing antibodies, and patients should be repeatedly tested for antibodies during treatment.
If patients have developed neutralizing antibodies and show loss of efficacy or other symptoms of immunogenicity, the clinician may decide to stop the treatment, because further treatment is not efficacious and may lead to the increase or prolongation of the side effects. In some cases the treatment can be changed immediately to a different therapeutic protein with a similar mode of action or a similar but less immunogenic product, after a wash out period (Malucchi et al., 2005; McDougall et al., 2009).
In cases alternative treatment is not available, induction of tolerance can be considered. This has been used successfully in the case of factor VIII deficiency and is also advocated for essential enzyme replacement therapy (Wang et al., 2009).
Immunosuppressive therapies have been used to shorten the immunogenic response in cases of anti-epoetin antibodies associated PRCA. However these treatments have never been evaluated in controlled trials (Bennett et al., 2005).
Bachmann MF, Zinkernagel RM. Neutralizing antiviral B cell responses. Annu Rev Immunol 15:235-270, 1997.
Bartelds GM, Wijbrandts CA, Nurmohamed MT, Stapel SO, Lems WF, Aarden L, Dijkmans BAC , Tak PP, Wolbink GJ. Anti-infliximab and anti-adalimumab antibodies in relation to response to adalimumab in infliximab switchers and anti-TNF naive patients: a cohort study. Ann Rheum Dis 69(5):817-21, 2010.
van Beers MMC , Sauerborn M, Gilli F, Brinks V, Schellekens H, Jiskoot W. Aggregated recombinant human interferon beta induces antibodies but no memory in immune tolerant transgenic mice. Pharm Res, epub ahead of print, May 25, 2010.
Bennett CL, Cournoyer D, Carson KR, Rossert J, Luminari S, Evens AM, Locatelli F, Belknap SM, McKoy JM, Lyons EA, Kim B, Sharma R, Costello S, Toffelmire EB, Wells GA, Messner HA, Yarnold PR, Trifilio SM, Raisch DW, Kuzel TM, et al. Long-term outcome of individuals with pure red cell aplasia and antierythropoietin antibodies in patients treated with recombinant epoetin: a follow-up report from the Research on Adverse Drug Events and Reports (RADAR) Project. Blood 106(10):3343-7, 2005.
Casadevall N, Nataf J, Viron B, Kolta A, Kiladjian JJ, Martin-Dupont P, Michaud P, Papo T, Ugo V, Teyssandier I, Varet B, Mayeux P. Pure red-cell aplasia and antierythropoietin antibodies in patients treated with recombinant erythropoietin. N Engl J Med 346(7):469-75, 2002.
Clarke JB. Mechanisms of adverse drug reactions to biologics. Handb Exp Pharmacol (196):453-74, 2010.
Hartung HP, Munschauer F 3rd, Schellekens H. Significance of neutralizing antibodies to interferon beta during treatment of multiple sclerosis: expert opinions based on the Proceedings of an International Consensus Conference. Eur J Neurol 12(8):588-601, 2005.
González-Fernández A, Faro J, Fernández C. Immune responses to polysaccharides: lessons from humans and mice. Vaccine 26:292-300, 2008.
Hermeling S, Aranha L, Damen JMA, Slijper M, Schellekens H, Crommelin DJA, Jiskoot W. Structural characterization and immunogenicity in wild-type and immune tolerant mice of degraded recombinant human interferon alpha2b. Pharm Res 22:1997-2006, 2005.
Macdougall IC, Rossert J, Casadevall N, Stead RB, Duliege AM, Froissart M, Eckardt KU. A peptide-based erythropoietin-receptor agonist for pure red-cell aplasia. N Engl J Med 361(19):1848-55, 2009.
Malucchi S, Capobianco M, Gilli F, Marnetto F, Caldano M, Sala A, Bertolotto A. Fate of multiple sclerosis patients positive for neutralising antibodies towards interferon beta shifted to alternative treatments. Neurol Sci 26(Suppl 4):s213-4, 2005.
Mayer L, Young Y. Infusion reactions and their management. Gastroenterol Clin North Am 35(4):857-66, 2006.
Moreland LW, McCabe DP, Caldwell JR, Sack M, Weisman M, Henry G, Seely JE, Martin SW, Yee CL, Bendele AM, Frazier JL, Kohno T, Cosenza ME, Lyons SA, Dayer JM, Cohen AM, Edwards CK 3rd. Phase I/II trial of recombinant methionyl human tumor necrosis factor binding protein PEGylated dimer in patients with active refractory rheumatoid arthritis. J Rheumatol 27(3):601-9, 2000.
Praditpornsilpa K, Kupatawintu P, Mongkonsritagoon W, Supasyndh O, Jootar S, Intarakumthornchai T, Pongskul C, Prasithsirikul W, Achavanuntakul B, Ruangkarnchanasetr P, Laohavinij S, Eiam-Ong S. The association of anti-r-HuEpo-associated pure red cell aplasia with HLA-DRB1*09-DQB1*0309. Nephrol Dial Transplant 24(5):1545-9, 2009.
Rau R, Sander O, van Riel P, van de Putte L, Hasler F, Zaug M, Kneer J, vander Auwera P, Stevens RM; Rheumatology Group 791. Intravenous human recombinant tumor necrosis factor receptor p55-Fc IgG1 fusion protein Ro 45-2081 (lenercept): a double blind, placebo controlled dose-finding study in rheumatoid arthritis. J Rheumatol 30(4):680-90, 2003.
Rose AS, Braun A, Kwee L, Labow MA, Alsenz J. Protein aggregates seem to play a key role among the parameters influencing the antigenicity of interferon alpha (IFN-alpha) in normal and transgenic mice. Pharm Res 14:1472-8, 1997.
Rosenberg AS. Effects of protein aggregates: an immunologic perspective. AAPS J 8(3):E501-7, 2006.
Ryff JC. Clinical investigation of the immunogenicity of interferon-alpha 2a. J Interferon Cytokine Res 17(Suppl 1):S29-33, 1997.
Sauerborn M, Brinks V, Jiskoot W, Schellekens H. Immunological mechanism underlying the immune response to recombinant human protein therapeutics. Trends Pharmacol Sci 31:53-9, 2010.
Schellekens H. The immunogenicity of therapeutic proteins and the Fabry antibody standardization initiative. Clin Ther 30(Suppl B):S50-1, 2008.
Schellekens H. Factors influencing the immunogenicity of therapeutic proteins. Nephrol Dial Transplant 20(Suppl 6):vi3-9, 2005.
Schellekens H. Immunogenicity of therapeutic proteins: clinical implications and future prospects. Clin Ther 24(11):1720-40, 2002.
Schellekens H, Ryff JC, van der Meide PH. Assays for antibodies to human interferon-alpha: the need for standardization. J Interferon Cytokine Res 17(Suppl 1):S5-8, 1997.
Schellekens H. Bioequivalence and the immunogenicity of biopharmaceuticals. Nat Rev Drug Discov 1(6):457-62, 2002.
Sorensen PS. Review: Neutralizing antibodies against interferon-beta. Ther Adv Neurol Disord 1:125-41, 2008.
de Vries MK, Wolbink GJ, Stapel SO, de Vrieze H, van Denderen JC, Dijkmans BA, Aarden LA, van der Horst-Bruinsma IE. Decreased clinical response to infliximab in ankylosing spondylitis is correlated with anti-infliximab formation. Ann Rheum Dis 66(9):1252-4, 2007.
Wang J, Lozier J, Johnson G, Kirshner S, Verthelyi D, Pariser A, Shores E, Rosenberg A. Neutralizing antibodies to therapeutic enzymes: considerations fortesting, prevention and treatment. Nat Biotechnol 26(8):901-8, 2008.
[Discovery Medicine, 9(49):560-564, June 2010.]