Cellular Immunology 295 (2015) 118–126

Contents lists available at ScienceDirect

Cellular Immunology journal homepage: www.elsevier.com/locate/ycimm

Review

Therapeutic outcomes, assessments, risk factors and mitigation efforts of immunogenicity of therapeutic protein products Liusong Yin a, Xiaoying Chen b, Paolo Vicini c, Bonita Rup a, Timothy P. Hickling a,⇑ a

Pharmacokinetics, Dynamics and Metabolism-New Biological Entities, Pfizer, Andover, MA, United States Pharmacokinetics, Dynamics and Metabolism-New Biological Entities, Pfizer, Cambridge, MA, United States c Pharmacokinetics, Dynamics and Metabolism-New Biological Entities, Pfizer, San Diego, CA, United States b

a r t i c l e

i n f o

Article history: Received 20 January 2015 Revised 6 March 2015 Accepted 9 March 2015 Available online 14 March 2015 Keywords: Therapeutic protein Immunogenicity Risk assessment Efficacy Safety Prediction

a b s t r a c t Therapeutic protein products (TPPs) are of considerable value in the treatment of a variety of diseases, including cancer, hemophilia, and autoimmune diseases. The success of TPP mainly results from prolonged half-life, increased target specificity and decreased intrinsic toxicity compared with small molecule drugs. However, unwanted immune responses against TPP, such as generation of anti-drug antibody, can impact both drug efficacy and patient safety, which has led to requirements for increased monitoring in regulatory studies and clinical practice, termination of drug development, or even withdrawal of marketed products. We present an overview of current knowledge on immunogenicity of TPP and its impact on efficacy and safety. We also discuss methods for measurement and prediction of immunogenicity and review both product-related and patient-related risk factors that affect its development, and efforts that may be taken to mitigate it. Lastly, we discuss gaps in knowledge and technology and what is needed to fill these. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction Since the approval of the first recombinant therapeutic protein product (TPP), recombinant human insulin, in 1982, more than 200 TPPs have entered the marketplace with an estimated annual revenue of over 100 billion dollars [1–3]. Examples of TPP include monoclonal antibodies (mAbs), Fc fusion proteins, anticoagulants, blood factors, hormones, cytokines, growth factors and engineered protein scaffolds derived from non-human, humanized or human origins [1]. TPPs have been widely used to treat cancer, rheumatoid arthritis (RA), multiple sclerosis (MS), inflammatory bowel disease (IBD), hemophilia, and anemia (Table 1). The successes of TPP are

Abbreviations: TPP, therapeutic protein product; ADA, anti-drug antibody; mAb, monoclonal antibody; RA, rheumatoid arthritis; MS, multiple sclerosis; IBD, inflammatory bowel disease; NAb, neutralizing ADA; non-NAb, non-neutralizing ADA; PK, pharmacokinetics; PRCA, pure red cell aplasia; MHC II, major histocompatibility complex class II molecules; ELISA, enzyme-linked immunosorbent assay; SPR, surface plasmon resonance; ECLA, electrochemiluminescence assay; RIA, radioimmunoassay; PIA, pH-shift anti-idiotype antigen-binding test; HMSA, homogenous mobility shift assay; DC, dendritic cell; PBMC, peripheral blood mononuclear cell. ⇑ Corresponding author at: PDM-NBE, Pfizer, 1 Burtt Road, Andover, MA 01810, United States. Tel.: +1 978 247 2909. E-mail address: Timothy.Hickling@Pfizer.com (T.P. Hickling). http://dx.doi.org/10.1016/j.cellimm.2015.03.002 0008-8749/Ó 2015 Elsevier Inc. All rights reserved.

related to their increased specificity, slower clearance from the body (and hence longer duration of effect) and reduced intrinsic toxicity. These provide an advantage over small molecule drugs, which can be associated with off-target effects and harmful metabolites. The versatility of TPP and the growing resources that pharmaceutical companies have put into large molecule drug development are expected to lead to the continued expansion of the TPP portion of the drug marketplace, as evidenced by the 54 new approvals of TPPs in the United States and European Union between 2010 and 2014 [3]. However, when TPPs are administered to patients, unwanted immune responses, such as generation of anti-drug antibody (ADA), have impacted drug efficacy and caused patient safety problems, although in some cases little or no impact of ADA on efficacy and safety was observed [4–9]. Here, we present an overview of immunogenicity of TPP and its impact on drug efficacy and patient safety. We will also review experimental assays to measure ADA, and efforts to assess or predict immunogenicity risk, as well as product- and patient-related risk factors contributing to immunogenicity and efforts that may be prospectively taken to mitigate immunogenicity. We contend that, to reduce the occurrence and impact of immunogenicity, significant gaps in knowledge about its mechanisms and technologies to conduct robust assessments must be filled using intellectual input from the broader immunology science community.

L. Yin et al. / Cellular Immunology 295 (2015) 118–126

119

Table 1 Examples of TPP, their primary indications and proposed mechanisms. Primary indication

Examplea

Category

Proposed mechanism

Cancer

Alemtuzumab Bevacizumab Etanercept Adalimumab Natalizumab Interferon beta 1a Infliximab Vedolizumab Factor VIIa Epoetin alfa

mAb mAb Fusion protein mAb mAb Cytokine mAb mAb Blood factor Hormone

Treat Treat Treat Treat Treat Treat Treat Treat Treat Treat

RA MS IBD Hemophilia Anemia a

B cell chronic lymphocytic leukemia by targeting CD52 metastatic colorectal cancer by targeting vascular endothelial growth factor RA by targeting TNF-alpha RA by targeting TNF-alpha MS by targeting cell adhesion molecule a4-integrin MS by balancing pro- and anti-inflammatory signals Crohn’s disease and ulcerative colitis by targeting TNF-alpha Crohn’s disease and ulcerative colitis by antagonizing integrin receptor hemophilia by inducing coagulation anemia by stimulating erythropoiesis

The full list of approved TPP is discussed in Ref. [3].

2. ADA impact on drug efficacy and patient safety Formation of ADA against TPP has been widely observed in clinical practice, such as in treatment of Crohn’s disease and RA patients with anti-TNF adalimumab [10,11], hemophilia A (Factor VIII deficiency) with recombinant Factor VIII [12] and MS patients receiving interferon beta [13], although the incidence rate of ADA varies considerably among studies, even using the same drug [14–17]. The production of ADA against TPP has been linked to reduced clinical drug efficacy (Fig. 1). ADAs can be classified into two groups: neutralizing ADA (NAb) or non-neutralizing ADA

(non-NAb) depending on whether they inhibit the TPP pharmacological activity [18]. There are two possible mechanisms through which NAb and non-NAb could contribute to reduced drug efficacy. First, NAb directly blocks the binding of TPP to its targeting molecule, therefore reducing its therapeutic efficacy [19,20]. Second, NAb and non-NAb could contribute to increased clearance affecting the pharmacokinetics (PK) of TPP therefore compromising drug efficacy, although they could also increase the exposure of TPP in the case of a small protein such as an Fc conjugate [21]. For TNFantagonist TPPs used to treat RA or IBD, a high incidence of ADA is often associated with impaired or absent response to treatment

Fig. 1. Overview of risk factors that contribute to immunogenicity, therapeutic outcomes that result from immunogenicity and mitigation efforts to reduce immunogenicity. Upper left: Risk factors that contribute to immunogenicity include product-related and patient-related factors. Central: immunogenicity could be measured by experimental approaches or conceivably predicted by mathematical models and in vitro/in vivo assays. Upper right: therapeutic outcomes affected by immunogenicity include both drug efficacy and patient safety. Bottom: mitigation efforts to reduce immunogenicity are recommended following a risk-based approach. Image credit: structure of an IgG2 antibody created from PDB 1IGT (Wikimedia Commons, public domain).

120

L. Yin et al. / Cellular Immunology 295 (2015) 118–126

[17,22–24], although in some cases ADA levels are not linked with impaired clinical response, presumably as long as sufficient active drug is present [25,26]. The influence of ADA formation on therapeutic outcomes has been extensively and systematically reviewed elsewhere, with a strong focus on the treatment of inflammatory diseases using anti-TNF mAb [8,25,27,28]. ADA has also been associated with increased frequency of clinical adverse effects, such as infusion reactions, or even in some cases with life-threatening autoimmune syndrome, due to crossreactivity with endogenous proteins (Fig. 1). The most notorious example is the development of ADA targeting both administered epoetin alfa and endogenous erythropoietin, which resulted in severe pure red cell aplasia (PRCA) in more than 100 patients receiving epoetin alfa and ultimately led to the withdrawal of marketing authorization for subcutaneous administration of the drug product associated with increased incidence of PRCA [29–33]. Several Factor VIII products were also withdrawn due to increased development of NAb in hemophilia A patients [34–36]. Other examples of ADA-associated adverse effects include infusion reactions in Crohn’s disease and RA patients receiving anti-TNF treatments [10,11], severe anaphylaxis with recombinant interferon beta [37], and anaphylaxis after treatment with Factor IX [38]. Various types of ADA have been observed during TPP treatment, mostly IgG, but also IgE, IgM and IgA. Sustained production of affinitymatured IgG is responsible for most of the adverse effects, including PRCA by NAb against erythropoietin [29], thrombocytopenia by NAb against thrombopoietin [39], anaphylactic reactions [40,41], and activation of the complement system [42]. TPP-specific IgE have also been demonstrated to mediate anaphylactic reactions [43–46]. Low affinity and early stage transient IgM, capable of activating the complement system [47,48], has been reported with anti-TNF mAb treatment [46].

3. Factors that influence immunogenicity Immunogenicity against TPP can be generated in both T celldependent and T cell-independent pathways [49]. In the T celldependent pathway, T cells are activated through the recognition of TPP-derived antigenic peptides presented by major histocompatibility complex class II molecules (MHC II) in antigen-presenting cells. The activated T cells then stimulate B cells to generate TPP-specific ADA. ADAs generated from the T cell-dependent pathway are usually class-switched and affinity matured IgG [9]. In the T cell-independent pathway, B cells can be activated by multivalent ligands through B cell receptor cross-linking [50]. It has been speculated that aggregates of TPP that display repeating paracrystalline epitopes might cross-link B cell receptors and activate B cells in a T cell-independent pathway [51]. ADAs generated in T cell-independent pathway would most likely be IgM with some low affinity IgG [51–55]. The presence of additional danger signals provided from the TPP or impurities in TPP formulations [56] or the disease environment could potentially lead to increased immune responses against TPP. TPP of non-human origin could induce immunogenicity through recognition of non-self-antigens, while TPP of human origin could primarily induce immunogenicity due to the breakage of immune tolerance [52,57,58]. The detailed mechanisms leading to immunogenicity (i.e., ADA formation against TPP) have yet to be fully established and characterized, but several patient-related and product-related risk factors have been proposed (Fig. 1). The presence or absence of these factors when a particular TPP is administered in the context of a specific patient population and treatment regimen is used to assess risk for immunogenicity and to guide development of risk mitigation strategies. Patient-related factors that are thought to affect immunogenicity include genetic background, immunologic

status, prior sensitization, and route, dose and frequency of administration. Genetic factors may modulate the immune response and predispose patients carrying certain MHC alleles to the development of ADA to specific TPP [59]. Polymorphisms in cytokine genes may also influence the level of immunogenicity, as demonstrated by the association between IL10 genotype and ADA formation against adalimumab and Factor VIII [60,61]. Immune competency status of patients affects the propensity to develop unwanted immune responses against TPP. For example, it was found that immune-competent cancer patients generated NAb at a much higher percentage compared with immunecompromised cancer patients receiving a granulocyte-macrophage colony-stimulating factor product [62]. Thus, patients with a highly active immune system may increase the risk of ADA development, due to their lower activation threshold for antibody production. In contrast, co-medication with immunomodulators suppressing the immune system, such as methotrexate, could substantially lower the risk of immunogenicity [10,63]. Prior sensitization to homologous protein may result in presence of preexisting antibodies, although a strong association between pre-existing antibodies and subsequent immunogenicity was only found in some cases [64,65]. Lastly, route, dose and frequency of administration can affect immunogenicity. In general, intravenous administration of TPP poses the lowest risk of immunogenicity compared with intramuscular, intradermal, subcutaneous or inhalational treatments [66–68]. Some dosing regimens could reduce immunogenicity by inducing tolerance [10,69,70]. Product-related factors that affect immunogenicity include product origin, characteristics and formulation, all of which have been reviewed extensively by Singh and colleagues [9,57]. Immunogenicity of TPP originating from non-human species, such as mice and bacteria, is expected and often more pronounced than that of TPP derived from humanized or fully human-origins. This is thought to be attributed to the non-human sequences or structures, even though some fully human TPP can also notably elicit a high rate of immunogenicity, probably due to contributions from the other risk factors discussed, e.g., formulation or impurities [9,57]. For TPP derived from a human endogenous counterpart such as cytokines or enzyme replacement therapies, the low endogenous levels of these proteins expressed in specialized tissues may not result in the induction of complete central tolerance. Evidence for incomplete tolerance exists for Factor VIII, where antiFactor VIII antibodies have been observed in otherwise healthy individuals as a result of tissue damage or trauma [71–73]. Development of immunogenicity against these TPP is likely due to a patient specific breakage of peripheral tolerance, central tolerance or due to antigenic differences between endogenous and replacement factor TPPs. For example, peripheral tolerance may be broken when treating MS patients with high doses of interferon beta over a long period of time [74,75]. Novel structural formats and molecular designs, and primary sequences that contain B cell and T cell epitopes, could be recognized as non-human and induce immune responses [76–80]. Product properties that affect immunogenicity include molecular structures, post-translational modifications, impurities, immunomodulatory properties and aggregation [7,9,57]. Among these, aggregation of TPP carries particular concern, as it occurs under some conditions with many TPP and several studies have identified associations between aggregates and increased propensity for immunogenicity [51,81–86]. Compared with non-aggregated forms, higher immunogenicity in mice has been observed for aggregates generated under different stress conditions for various TPPs, such as human interferon alpha2b [87–89], human mAbs [90–92], human epoetin alfa [83], human Factor VIII [93,94], human interferon beta [95], and murine growth hormone [96]. The detailed mechanisms of aggregate-induced immunogenicity are still unknown, but it has been observed

121

L. Yin et al. / Cellular Immunology 295 (2015) 118–126

that aggregation enhances antigen uptake, increases the total amount of peptides associated with MHC II molecules, and increases danger signals to maturate dendritic cells and activate T cells [52,86,91,97,98]. Therefore, aggregates could contribute to increased immunogenicity through enhancing antigen processing and presentation in the T cell-dependent pathway. In addition, as mentioned previously, aggregates of TPP that display repeating para-crystalline epitopes might cross-link B cell receptors and activate B cells in a T cell-independent pathway [51]. Debates still exist on what quantities of aggregates and which properties of aggregates (size, solubility, stability, whether native-like or nonnativelike) determine immunogenicity. Formulation may also affect immunogenicity through the generation of co-aggregates between excipients and drug substances [99], or through leachables from drug containers which can chemically modify the TPP, introducing new epitopes, or have intrinsic adjuvant activity [83]. More extensive mechanistic studies of immunogenicity and tolerance are being pursued with some higher risk TPPs such as Factor VIII [58,100,101], however more data are needed for the majority of TPP. In summary, multiple risk factors have been associated with immunogenicity, but the detailed mechanism and causal linkages between each risk factor and immunogenicity induction onset have yet to be established, due to the limited amount of data from mechanistic studies and lack of multi-factorial analysis. Also, the lack of standard immunogenicity assessment methods complicates the comparison of outcomes across different products. Addressing these outstanding questions would be crucial for the field to advance. 4. Immunogenicity measurement TPP immunogenicity has been primarily assessed by monitoring the presence and magnitude (titer) of ADA responses and in vitro neutralizing ability of ADA following TPP administration. Assay strategies are driven by product-specific, indication-specific, risk assessment- and performance-based objectives [102,103]. Therefore, multiple assay formats, technology platforms and sample preparation protocols have been applied to measure ADA responses including enzyme-linked immunosorbent assay (ELISA), surface plasmon resonance (SPR), radioimmunoassay (RIA), electrochemiluminescence assay (ECLA), pH-shift anti-idiotype antigen-binding test (PIA), and homogenous mobility shift assay (HMSA) (Fig. 1). ELISA is the most commonly used assay technology platform due to its ease of administration. In the direct ELISA format, ADA from patients’ sera is captured by the coated TPP and detected by a labeled anti-Ig reagent. This format is unsuitable for most mAb and Fc fusion proteins, therefore these can be measured in the bridge ELISA format, in which ADA bound to the coated TPP is detected by a labeled version of the TPP [104].

SPR can also be used to measure ADA against mAb and Fc fusion proteins and is more sensitive for detection of lower affinity ADA than ELISA. Because the bridge and SPR formats do not use Ig-specific detectors, other molecules that bind to the TPP can also score positively in these assay formats. In RIA, free ADA is captured by a Sepharose-bound reagent such as protein A or antibody, and then detected by 125I-labeled TPP [104]. Most assay formats predominantly detect free ADA, but not TPP-bound ADA. TPP in the sera can interfere with ADA detection and potentially causes either underestimation of ADA levels or false negative results. More complex assay formats and sample preparation strategies may be needed if TPP interference is likely to be an issue (i.e., for monitoring ADA vs TPP with longer half-lives). ECLA platforms using a bridge format generally have increased tolerance against TPP interference because higher concentrations of TPP can be used to capture the ADA on higher density ECLA platform surfaces; longer incubation times can also increase the chance of detecting free ADA. PIA is capable of detecting both free and bound ADA, because ADA bound with TPP is first released by lowering pH and subsequently detected by an antigen-binding test, as in ELISA, ECLA or RIA [105]. HMSA can detect ADA, by measuring the binding to fluorochrome-labeled TPP [106]. PIA and HMSA have been only recently developed, with only a few studies to date using these assays to assess immunogenicity; they therefore require further investigation. The advantages and limitations of each commonly used assay for ADA detection are listed in Table 2. The detection of NAb is directly assessed by its ability to neutralize the effect of TPP, as in the case of detecting NAb against Factor VIII based on the ability to inhibit its clotting function [107] and NAb against anti-TNF therapies in a cell-based bio-assay [108]. However, there are systematic differences between assay performance when directly comparing multiple assays, which can lead to discrepancies in estimating the actual ADA incidence and thus in interpreting the impact of ADA on clinical outcomes [109– 111]. For example, in a study directly comparing drug concentration and ADA incidence rate following infliximab treatment in Crohn’s disease patients measured by ELISA, RIA, HMSA and reported gene assay (RGA), it was found that infliximab concentrations are comparable, but incidence rate of anti-infliximab antibodies varies by 4-fold [111]. Another similar study also reported 3-fold difference in anti-infliximab antibodies incidence rate measured by ELISA, RIA and HMSA in Crohn’s disease patients receiving infliximab treatment [109]. Therefore, caution should be used when comparing relative immunogenicity of various TPPs measured by different assay formats, as these comparisons could be misleading. Besides those assay-intrinsic factors, potential interactions between ADA and biological matrix components such as lipids, rheumatoid factors, and heterophilic antibodies could also lead to under- or over-reporting of ADA level [112].

Table 2 Summary of current major assays for immunogenicity measurement. Assay

ADA detection

Advantages

Limitations

ELISA

Captured by coated TPP Detected by labeled anti-Ig or TPP Captured by coated TPP Detected by resonant oscillation Captured by Sepharose-bound reagent Detected by 125I-labeled TPP Dissociated from TPP using acid treatment Captured and detected by antigen-binding test Captured by TPP Detected by ruthenium-labeled reagent Dissociated from TPP using acid treatment Captured by fluorescence-labeled TPP Detected by mobility shift

Ease of administration

False positives TPP interference False positives TPP interference TPP interference Use of radioactivity Limited study

SPR RIA PIA ECLA HMSA

High sensitivity High sensitivity Less TPP interference High sensitivity Less TPP interference Less TPP interference No radioactivity High sensitivity Able to detect all Ig isotypes

Limited study Limited study Multiple labeling Low throughput

122

L. Yin et al. / Cellular Immunology 295 (2015) 118–126

5. Immunogenicity assessment and prediction tools While the immunogenicity risk profile of a TPP is ultimately characterized by measurement of ADA levels in patients and correlation with therapeutic outcomes, immunogenicity prediction tools serve to assess immunogenicity risk preclinically and conceivably offer the advantage of fast, and high throughput assessment, especially in the initial phase of drug development such as at TPP candidate design and selection stages. In silico prediction algorithms and mathematical models complement experimental data for assessing immunogenicity potential. In silico CD4+ T cell epitope prediction algorithms based on peptide binding affinity to MHC II or sequence analysis of naturally processed peptides [113–115], and (currently less reliable) B cell epitope prediction algorithms [116,117], are the common tools used. Gaps still exist between algorithm captured characteristics and in vivo processes leading to generation of immuno-dominant epitopes. For example, HLA-DM has been demonstrated to play a key role in antigen presentation and CD4+ T cell epitope selection by promoting the presentation of highly stable MHC II-peptide complexes [114,118,119]. However, to date no epitope prediction algorithms have directly incorporated the effect of HLA-DM [115]. Algorithm development efforts would benefit from studies directed at identifying sequence motifs of peptides presented following MHC II-binding and HLA-DM editing. Recently, mechanistic and multiscale mathematical models have been proposed to simulate ADA formation over time [120–122]. These models provide estimation and forward simulations of TPP PK data and ADA formation, and could conceivably capture well-characterized immunological phenomena such as memory response and antibody affinity maturation. The recent discipline of mathematical modeling of ADA formation following mAb exposure has been comprehensively reviewed elsewhere [123]. Experimental tools to predict immunogenicity risk include HLA binding assay, dendritic cell (DC) activation and antigen presentation assay, T cell stimulation assay and peripheral blood mononuclear cell (PBMC) stimulation assay, as well as animal models [115]. HLA binding and DC antigen presentation assays evaluate the presence and generation of potential T epitopes derived from the TPP, while T cell and PBMC stimulation assays test the ability of TPP to activate the immune cells in vitro and ex vivo in terms of cell proliferation, surface marker and cytokine expression. In vivo animal models also can provide a useful approach to evaluate immunogenicity, with the predictive value depending on the similarity of processes underlying immunogenicity across species [124,125]. For TPP with sequences conserved across species, animal models may have higher prediction values, but for TPP of human origin, animal models often over-estimate immunogenicity [125]. Each of these various approaches has strengths and limitations in assessing and predicting immunogenicity. In silico epitope prediction algorithms provide a high throughput assessment of TPP and an opportunity in early drug development to modify a TPP to reduce its immunogenic potential [115]. A previous study demonstrated a correlation between the in silico evaluation of T cell epitopes from a recombinant Fc fusion protein and its observed immunogenicity rate when administered to human subjects in a clinical trial [126]. The limitations of these in silico algorithms lie in the limited information captured. For example, most CD4+ T cell epitope prediction algorithms are based on binding affinity and stability of peptides to MHC II molecules [113–115], but it is known that other factors such as protease cleavage sites [127,128], T cell precursor frequency [129,130] and T cell competition [131,132] all have certain roles in defining T cell epitopes, and the affinity of MHC II–peptide complex to TCR also influences

the immunological outcome of an epitope [133–135]. Algorithms taking these factors into account will provide more prediction power. Recent mathematical models of ADA production following TPP administration provide a comprehensive evaluation of immunogenicity in physiological context covering antigen processing and presentation, immune cell activation and ADA disposition, and a platform to generate new hypotheses on the underlying mechanisms of immunogenicity [121–123]. The corresponding limitations are that multiple parameters especially on a TPP-specific basis constraining the model are not well defined and therefore impair the predictability. With further experimental validation, these mathematical models could potentially support a platform for prospective immunogenicity assessment. In vitro assays, such as HLA-binding, DC activation, T cell and PBMC stimulation are straightforward and could provide a preview of the immunogenicity of a TPP pre-clinically [115]. However, these assays are often labor intensive, are impractical to run with a large number of TPP candidates, and the interpretation of results can sometimes be complex. For example, these assays are often performed with cells derived from a naïve population and therefore the precursor frequency of antigen-specific cells is quite low. In that case, it is hard to define a true positive response, due to low signal-to-noise ratio. Future considerations for improving these assays include using statistically derived criteria to distinguish responders from non-responders, evaluation of samples from subjects with disease status relevant to the intended therapeutic indication, and selection of biomarkers to better define responsiveness. Finally, in vivo animal models provide an assessment of immunogenicity in the context of functional innate and adaptive immune systems and have been employed to study the underlying mechanisms of immunogenicity [125]. Species differences and low throughput remain the major obstacles to using animal models to assess immunogenicity [115,125]. Animal models that may be more fitfor-purpose for assessing human immunogenicity risk include mice transgenic for HLA, the TPP, and transplanted with human immune cells or lymphoid tissues for assessing relative immunogenicity, neo-epitopes and breaking of tolerance [115,125]. 6. Mitigation of unwanted immunogenicity As discussed, immunogenicity of TPP contributed by multiple risk factors affects drug efficacy and patient safety. Both the United States Food and Drug Administration and the European Medicines Agency have outlined guidelines and recommendations on risk-based approaches to assess and mitigate immunogenicity against TPP, although currently not all of these are part of common clinical practices [136,137]. A better understanding of immunogenicity risk factors and underlying immune system mechanisms is required to develop safer and more efficacious medicines. For each patient-related and product-related risk factor, a corresponding mitigation strategy could be identified (Fig. 1). For example, to address risk related to the immunological status of patients, co-medication with immunomodulators could be taken into consideration to modulate the immune response. A screening of prior sensitization to the same or related TPP may help the overall risk–benefit assessment. An appropriate route, dosing regime and frequency of administration could be selected based on the expected immunogenicity risk. For product-related factors, the design strategy of candidate TPP should consider reducing their foreignness through humanization and modification of the primary sequence to remove or mask potential T cell and B cell epitopes, or even to introduce T regulatory cell epitopes, as identified by in silico epitope prediction algorithms, HLA binding assays, DC antigen presentation assays, T cell or PBMC stimulation assays. Handling, formulation and container closure system should avoid

L. Yin et al. / Cellular Immunology 295 (2015) 118–126

drug agitation, freeze–thaw cycles, and direct light exposure, to minimize aggregation, denaturation and degradation. Beside those patient- and product-related risk factors, consideration should also be given to post marketing monitoring, which may identify risks that might be missed during preapproval clinical testing due to the limited size and diversity of the population under study. 7. Outstanding questions and call for help Despite the fact that extensive studies have been conducted and a variety of risk factors have been identified, elucidation of the detailed mechanism and underlying causes of immunogenicity will require help from the broad immunology community. For TPP of non-human origins (i.e., streptokinase and some mouse-derived mAb against TNF), or for TPP as a replacement in patients deficient for endogenous proteins, immunogenicity is probably induced through recognition of non-self-antigens [138]. Even the fully human mAb adalimumab carries unique stretches of amino acids distinct from germ line responsible for specific TNF recognition and induction of NAb against adalimumab [139]. For TPP of human origin, immunogenicity is likely to be caused by breakage of immunological tolerance [140], although the mechanism is not well understood. For example, it is still unknown how recombinant human interferon beta breaks immunological tolerance in multiple sclerosis patients [75]. Tolerance might be broken through several mechanisms, including increased danger signals due to the impurities in TPP [56], bystander activation [141] and molecular mimicry. However, even though aggregation has been linked to breaking tolerance [140], the mechanistic details such as size of aggregates, T cell dependency, B cell epitopes, functional B cells, and antibody isotypes have yet to be characterized [140]. There are multiple steps involved in ADA formation, including DC activation, antigen presentation, T cell and B cell activation, and ADA production [121,122]. Experimental validation of each of these steps would improve quantitative approaches, such as mathematical modeling, to ultimately predict immunogenicity. A full understanding of mechanisms of immunogenicity would enable drug developers to assess, predict and mitigate unwanted immune responses, which will ultimately lead to the development of safer and more effective drugs. Significant discrepancies have been shown when measuring ADA formation with different assays which could lead to biased clinical outcome interpretation and treatments strategies [111,142,143] and could be caused by differences in assay performance as well as sampling strategies. As discussed above, each assay has its own advantages and limitations with factors such as sensitivity, specificity, drug interference and antibody isotypes, specifics of the sampled population, dosing and sampling time frames all contributing to discrepancies. Therefore, enough attention should be given to the nature of the analyte and assay limitations, and assays able to measure both TPP and ADA concentration robustly are highly desired. Together with ADA incidence rate, other ADA characterizations, such as relative binding affinity, antibody isotypes, cross-reactivity to endogenous proteins, and sample characteristics, such as patients’ genetic background, dosing and sampling regime may also be reported to better understand the underlying mechanisms of immunogenicity development. 8. Concluding remarks TPPs have been highly successful in treating a variety of human diseases, and are expected to continue to expand in the drug market, due to their low intrinsic toxicity and high specificity. However, immunogenicity of TPP has raised critical concerns both

123

for drug efficacy and patient safety. Multiple product-related and patient-related risk factors have been proposed to be associated with increased immunogenicity, and tremendous efforts have been devoted towards assessing and mitigating immunogenicity using a risk-based approach. Outstanding questions still remain on the detailed mechanisms of immunogenicity and the generation of robust assays for ADA assessment, each of which urgently needs intellectual input from the broader immunology community. Acknowledgment This work was supported by Pfizer Worldwide Research and Development Postdoctoral Fellowship. References [1] D.S. Dimitrov, Therapeutic proteins, Methods Mol. Biol. 899 (2012) 1–26. [2] G. Walsh, Biopharmaceutical benchmarks 2010, Nat. Biotechnol. 28 (2010) 917–924. [3] G. Walsh, Biopharmaceutical benchmarks 2014, Nat. Biotechnol. 32 (2014) 992–1000. [4] M.D. Barbosa, Immunogenicity of biotherapeutics in the context of developing biosimilars and biobetters, Drug Discovery Today 16 (2011) 345–353. [5] M.D. Barbosa, S. Kumar, H. Loughrey, S.K. Singh, Biosimilars and biobetters as tools for understanding and mitigating the immunogenicity of biotherapeutics, Drug Discovery Today 17 (2012) 1282–1288. [6] P. Chamberlain, Assessing immunogenicity of biosimilar therapeutic monoclonal antibodies: regulatory and bioanalytical considerations, Bioanalysis 5 (2013) 561–574. [7] A.C. Moss, V. Brinks, J.F. Carpenter, Review article: immunogenicity of antiTNF biologics in IBD – the role of patient, product and prescriber factors, Aliment. Pharmacol. Ther. 38 (2013) 1188–1197. [8] K.S. Nanda, A.S. Cheifetz, A.C. Moss, Impact of antibodies to infliximab on clinical outcomes and serum infliximab levels in patients with inflammatory bowel disease (IBD) a meta-analysis, Am. J. Gastroenterol. 108 (2013) 40–47. quiz 48. [9] P.A. van Schouwenburg, T. Rispens, G.J. Wolbink, Immunogenicity of anti-TNF biologic therapies for rheumatoid arthritis, Nat. Rev. Rheumatol. 9 (2013) 164–172. [10] F. Baert, M. Noman, S. Vermeire, G. Van Assche, G. D.H’aens, A. Carbonez, P. Rutgeerts, Influence of immunogenicity on the long-term efficacy of infliximab in Crohn’s disease, N. Engl. J. Med. 348 (2003) 601–608. [11] D. Pascual-Salcedo, C. Plasencia, S. Ramiro, L. Nuno, G. Bonilla, D. Nagore, A. Ruiz Del Agua, A. Martinez, L. Aarden, E. Martin-Mola, A. Balsa, Influence of immunogenicity on the efficacy of long-term treatment with infliximab in rheumatoid arthritis, Rheumatology 50 (2011) 1445–1452. [12] G.L. Bray, E.D. Gomperts, S. Courter, R. Gruppo, E.M. Gordon, M. MancoJohnson, A. Shapiro, E. Scheibel, G. White 3rd, M. Lee, A multicenter study of recombinant factor VIII (recombinate): safety, efficacy, and inhibitor risk in previously untreated patients with hemophilia A. The Recombinate Study Group, Blood 83 (1994) 2428–2435. [13] G. Antonelli, Neutralising antibodies against interferon beta in multiple sclerosis, Lancet 363 (2004) 168. author reply 168–169.. [14] M.E. Weinblatt, E.C. Keystone, D.E. Furst, L.W. Moreland, M.H. Weisman, C.A. Birbara, L.A. Teoh, S.A. Fischkoff, E.K. Chartash, Adalimumab, a fully human anti-tumor necrosis factor alpha monoclonal antibody, for the treatment of rheumatoid arthritis in patients taking concomitant methotrexate: the ARMADA trial, Arthritis Rheum. 48 (2003) 35–45. [15] M.H. Weisman, L.W. Moreland, D.E. Furst, M.E. Weinblatt, E.C. Keystone, H.E. Paulus, L.S. Teoh, R.B. Velagapudi, P.A. Noertersheuser, G.R. Granneman, S.A. Fischkoff, E.K. Chartash, Efficacy, pharmacokinetic, and safety assessment of adalimumab, a fully human anti-tumor necrosis factor-alpha monoclonal antibody, in adults with rheumatoid arthritis receiving concomitant methotrexate: a pilot study, Clin. Ther. 25 (2003) 1700–1721. [16] G.M. Bartelds, C.L. Krieckaert, M.T. Nurmohamed, P.A. van Schouwenburg, W.F. Lems, J.W. Twisk, B.A. Dijkmans, L. Aarden, G.J. Wolbink, Development of antidrug antibodies against adalimumab and association with disease activity and treatment failure during long-term follow-up, JAMA 305 (2011) 1460– 1468. [17] A. Jamnitski, G.M. Bartelds, M.T. Nurmohamed, P.A. van Schouwenburg, D. van Schaardenburg, S.O. Stapel, B.A. Dijkmans, L. Aarden, G.J. Wolbink, The presence or absence of antibodies to infliximab or adalimumab determines the outcome of switching to etanercept, Ann. Rheum. Dis. 70 (2011) 284–288. [18] G. Shankar, S. Arkin, L. Cocea, V. Devanarayan, S. Kirshner, A. Kromminga, V. Quarmby, S. Richards, C.K. Schneider, M. Subramanyam, S. Swanson, D. Verthelyi, S. Yim, Assessment and reporting of the clinical immunogenicity of therapeutic proteins and peptides-harmonized terminology and tactical recommendations, AAPS J. 16 (2014) 658–673. [19] A. Bertolotto, S. Malucchi, A. Sala, G. Orefice, P.B. Carrieri, M. Capobianco, E. Milano, F. Melis, M.T. Giordana, Differential effects of three interferon betas

124

[20]

[21]

[22]

[23]

[24]

[25] [26]

[27]

[28] [29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37] [38]

[39]

[40]

[41] [42]

L. Yin et al. / Cellular Immunology 295 (2015) 118–126 on neutralising antibodies in patients with multiple sclerosis: a follow up study in an independent laboratory, J. Neurol. Neurosurg. Psychiatry 73 (2002) 148–153. A. Bertolotto, A. Sala, S. Malucchi, F. Marnetto, M. Caldano, A. Di Sapio, M. Capobianco, F. Gilli, Biological activity of interferon betas in patients with multiple sclerosis is affected by treatment regimen and neutralising antibodies, J. Neurol. Neurosurg. Psychiatry 75 (2004) 1294–1299. R. Ponce, L. Abad, L. Amaravadi, T. Gelzleichter, E. Gore, J. Green, S. Gupta, D. Herzyk, C. Hurst, I.A. Ivens, T. Kawabata, C. Maier, B. Mounho, B. Rup, G. Shankar, H. Smith, P. Thomas, D. Wierda, Immunogenicity of biologicallyderived therapeutics: assessment and interpretation of nonclinical safety studies, Regul. Toxicol. Pharmacol.: RTP 54 (2009) 164–182. K. Bendtzen, P. Geborek, M. Svenson, L. Larsson, M.C. Kapetanovic, T. Saxne, Individualized monitoring of drug bioavailability and immunogenicity in rheumatoid arthritis patients treated with the tumor necrosis factor alpha inhibitor infliximab, Arthritis Rheum. 54 (2006) 3782–3789. G.J. Wolbink, M. Vis, W. Lems, A.E. Voskuyl, E. de Groot, M.T. Nurmohamed, S. Stapel, P.P. Tak, L. Aarden, B. Dijkmans, Development of antiinfliximab antibodies and relationship to clinical response in patients with rheumatoid arthritis, Arthritis Rheum. 54 (2006) 711–715. M. Svenson, P. Geborek, T. Saxne, K. Bendtzen, Monitoring patients treated with anti-TNF-alpha biopharmaceuticals: assessing serum infliximab and anti-infliximab antibodies, Rheumatology 46 (2007) 1828–1834. C. Krieckaert, T. Rispens, G. Wolbink, Immunogenicity of biological therapeutics: from assay to patient, Curr. Opin. Rheumatol. 24 (2012) 306–311. M. Benucci, G. Saviola, F. Meacci, M. Manfredi, M. Infantino, P. Campi, M. Severino, M. Iorno, P. Sarzi-Puttini, F. Atzeni, No correlations between the development of specific IgA and IgM antibodies against anti-TNF blocking agents, disease activity and adverse side reactions in patients with rheumatoid arthritis, Open Rheumatol. J. 7 (2013) 75–80. J.R. Maneiro, E. Salgado, J.J. Gomez-Reino, Immunogenicity of monoclonal antibodies against tumor necrosis factor used in chronic immune-mediated inflammatory conditions: systematic review and meta-analysis, JAMA Intern. Med. 173 (2013) 1416–1428. G. Solomon, Immunogenicity – implications for rheumatoid arthritis treatment, Bull. Hospital Joint Dis. 71 (2013) 200–203. N. Casadevall, J. Nataf, B. Viron, A. Kolta, J.J. Kiladjian, P. Martin-Dupont, P. Michaud, T. Papo, V. Ugo, I. Teyssandier, B. Varet, P. Mayeux, Pure red-cell aplasia and antierythropoietin antibodies in patients treated with recombinant erythropoietin, N. Engl. J. Med. 346 (2002) 469–475. C.L. Bennett, S. Luminari, A.R. Nissenson, M.S. Tallman, S.A. Klinge, N. McWilliams, J.M. McKoy, B. Kim, E.A. Lyons, S.M. Trifilio, D.W. Raisch, A.M. Evens, T.M. Kuzel, G.T. Schumock, S.M. Belknap, F. Locatelli, J. Rossert, N. Casadevall, Pure red-cell aplasia and epoetin therapy, N. Engl. J. Med. 351 (2004) 1403–1408. I.C. Macdougall, Antibody-mediated pure red cell aplasia (PRCA): epidemiology, immunogenicity and risks, Nephrol. Dial. Transplant. 20 (Suppl. 4) (2005) iv9–iv15. J. Yang, K.W. Joo, Y.S. Kim, C. Ahn, J.S. Han, S. Kim, J.S. Lee, Two cases of pure red-cell aplasia due to anti-erythropoietin antibodies, J. Nephrol. 18 (2005) 102–105. J.M. McKoy, R.E. Stonecash, D. Cournoyer, J. Rossert, A.R. Nissenson, D.W. Raisch, N. Casadevall, C.L. Bennett, Epoetin-associated pure red cell aplasia: past, present, and future considerations, Transfusion 48 (2008) 1754–1762. K. Peerlinck, J. Arnout, J.G. Gilles, J.M. Saint-Remy, J. Vermylen, A higher than expected incidence of factor VIII inhibitors in multitransfused haemophilia A patients treated with an intermediate purity pasteurized factor VIII concentrate, Thromb. Haemost. 69 (1993) 115–118. F.R. Rosendaal, H.K. Nieuwenhuis, H.M. van den Berg, H. Heijboer, E.P. Mauser-Bunschoten, J. van der Meer, C. Smit, P.F. Strengers, E. Briet, A sudden increase in factor VIII inhibitor development in multitransfused hemophilia A patients in The Netherlands. Dutch Hemophilia Study Group, Blood 81 (1993) 2180–2186. K. Peerlinck, J. Arnout, M. Di Giambattista, J.G. Gilles, R. Laub, M. Jacquemin, J.M. Saint-Remy, J. Vermylen, Factor VIII inhibitors in previously treated haemophilia A patients with a double virus-inactivated plasma derived factor VIII concentrate, Thromb. Haemost. 77 (1997) 80–86. T. Corona, C. Leon, L. Ostrosky-Zeichner, Severe anaphylaxis with recombinant interferon beta, Neurology 52 (1999) 425. I. Warrier, B.M. Ewenstein, M.A. Koerper, A. Shapiro, N. Key, D. DiMichele, R.T. Miller, J. Pasi, G.E. Rivard, S.S. Sommer, J. Katz, F. Bergmann, R. Ljung, P. Petrini, J.M. Lusher, Factor IX inhibitors and anaphylaxis in hemophilia B, J. Pediatr. Hematol. Oncol. 19 (1997) 23–27. J. Li, C. Yang, Y. Xia, A. Bertino, J. Glaspy, M. Roberts, D.J. Kuter, Thrombocytopenia caused by the development of antibodies to thrombopoietin, Blood 98 (2001) 3241–3248. G. Weber, J. Gross, A. Kromminga, H.H. Loew, K.U. Eckardt, Allergic skin and systemic reactions in a patient with pure red cell aplasia and antierythropoietin antibodies challenged with different epoetins, J. Am. Soc. Nephrol.: JASN 13 (2002) 2381–2383. F.D. Finkelman, Anaphylaxis lessons from mouse models, J. Allergy Clin. Immunol. 120 (2007) 506–515. quiz 516–507. A. Vultaggio, E. Maggi, A. Matucci, Immediate adverse reactions to biologicals: from pathogenic mechanisms to prophylactic management, Curr. Opin. Allergy Clin. Immunol. 11 (2011) 262–268.

[43] R.T. Purcell, R.F. Lockey, Immunologic responses to therapeutic biologic agents, J. Investig. Allergol. Clin. Immunol. 18 (2008) 335–342. [44] D.H. Conrad, D.R. Gibb, J. Sturgill, Regulation of the IgE response, Biol. Rep. 2 (2010). [45] K. Stubenrauch, U. Wessels, H. Birnboeck, F. Ramirez, A. Jahreis, J. Schleypen, Subset analysis of patients experiencing clinical events of a potentially immunogenic nature in the pivotal clinical trials of tocilizumab for rheumatoid arthritis: evaluation of an antidrug antibody ELISA using clinical adverse event-driven immunogenicity testing, Clin. Ther. 32 (2010) 1597–1609. [46] A. Vultaggio, A. Matucci, F. Nencini, S. Pratesi, P. Parronchi, O. Rossi, S. Romagnani, E. Maggi, Anti-infliximab IgE and non-IgE antibodies and induction of infusion-related severe anaphylactic reactions, Allergy 65 (2010) 657–661. [47] G.Y. Minuk, J.H. Hoofnagle, V.J. McAuliffe, R.H. Purcell, IgM antibody responses to hepatitis B surface antigen in recipients of hepatitis B virus vaccine, J. Infect. Dis. 148 (1983) 330. [48] J. Richard, N. Prang, The formulation and immunogenicity of therapeutic proteins product quality as a key factor, IDrugs Invest. Drugs J. 13 (2010) 550–558. [49] A.S. De Groot, D.W. Scott, Immunogenicity of protein therapeutics, Trends Immunol. 28 (2007) 482–490. [50] Q. Vos, A. Lees, Z.Q. Wu, C.M. Snapper, J.J. Mond, B-cell activation by T-cellindependent type 2 antigens as an integral part of the humoral immune response to pathogenic microorganisms, Immunol. Rev. 176 (2000) 154–170. [51] S. Kumar, S.K. Singh, X. Wang, B. Rup, D. Gill, Coupling of aggregation and immunogenicity in biotherapeutics: T- and B-cell immune epitopes may contain aggregation-prone regions, Pharm. Res. 28 (2011) 949–961. [52] S. Sethu, K. Govindappa, M. Alhaidari, M. Pirmohamed, K. Park, J. Sathish, Immunogenicity to biologics: mechanisms, prediction and reduction, Arch. Immunol. Ther. Exp. 60 (2012) 331–344. [53] C.M. Snapper, T.M. McIntyre, R. Mandler, L.M. Pecanha, F.D. Finkelman, A. Lees, J.J. Mond, Induction of IgG3 secretion by interferon gamma: a model for T cell-independent class switching in response to T cell-independent type 2 antigens, J. Exp. Med. 175 (1992) 1367–1371. [54] R. Guinamard, M. Okigaki, J. Schlessinger, J.V. Ravetch, Absence of marginal zone B cells in Pyk-2-deficient mice defines their role in the humoral response, Nat. Immunol. 1 (2000) 31–36. [55] I. Puga, M. Cols, C.M. Barra, B. He, L. Cassis, M. Gentile, L. Comerma, A. Chorny, M. Shan, W. Xu, G. Magri, D.M. Knowles, W. Tam, A. Chiu, J.B. Bussel, S. Serrano, J.A. Lorente, B. Bellosillo, J. Lloreta, N. Juanpere, F. Alameda, T. Baro, C.D. de Heredia, N. Toran, A. Catala, M. Torrebadell, C. Fortuny, V. Cusi, C. Carreras, G.A. Diaz, J.M. Blander, C.M. Farber, G. Silvestri, C. CunninghamRundles, M. Calvillo, C. Dufour, L.D. Notarangelo, V. Lougaris, A. Plebani, J.L. Casanova, S.C. Ganal, A. Diefenbach, J.I. Arostegui, M. Juan, J. Yague, N. Mahlaoui, J. Donadieu, K. Chen, A. Cerutti, B cell-helper neutrophils stimulate the diversification and production of immunoglobulin in the marginal zone of the spleen, Nat. Immunol. 13 (2012) 170–180. [56] D. Verthelyi, V. Wang, Trace levels of innate immune response modulating impurities (IIRMIs) synergize to break tolerance to therapeutic proteins, PLoS ONE 5 (2010) e15252. [57] S.K. Singh, Impact of product-related factors on immunogenicity of biotherapeutics, J. Pharm. Sci. 100 (2011) 354–387. [58] S. Lacroix-Desmazes, A.M. Navarrete, S. Andre, J. Bayry, S.V. Kaveri, S. Dasgupta, Dynamics of factor VIII interactions determine its immunologic fate in hemophilia A, Blood 112 (2008) 240–249. [59] S. Hoffmann, S. Cepok, V. Grummel, K. Lehmann-Horn, J. Hackermuller, P.F. Stadler, H.P. Hartung, A. Berthele, F. Deisenhammer, R. Wassmuth, B. Hemmer, HLA-DRB1⁄0401 and HLA-DRB1⁄0408 are strongly associated with the development of antibodies against interferon-beta therapy in multiple sclerosis, Am. J. Hum. Genet. 83 (2008) 219–227. [60] G.M. Bartelds, C.A. Wijbrandts, M.T. Nurmohamed, G.J. Wolbink, N. de Vries, P.P. Tak, B.A. Dijkmans, J.B. Crusius, I.E. van der Horst-Bruinsma, Antiadalimumab antibodies in rheumatoid arthritis patients are associated with interleukin-10 gene polymorphisms, Arthritis Rheum. 60 (2009) 2541–2542. [61] J. Astermark, Prevention and prediction of inhibitor risk, Haemophilia 18 (Suppl. 4) (2012) 38–42. [62] P. Ragnhammar, H.J. Friesen, J.E. Frodin, A.K. Lefvert, M. Hassan, A. Osterborg, H. Mellstedt, Induction of anti-recombinant human granulocyte-macrophage colony-stimulating factor (Escherichia coli-derived) antibodies and clinical effects in nonimmunocompromised patients, Blood 84 (1994) 4078–4087. [63] P.A. van Schouwenburg, C.L. Krieckaert, T. Rispens, L. Aarden, G.J. Wolbink, D. Wouters, Long-term measurement of anti-adalimumab using pH-shift-antiidiotype antigen binding test shows predictive value and transient antibody formation, Ann. Rheum. Dis. 72 (2013) 1680–1686. [64] L. Xue, M. Fiscella, M. Rajadhyaksha, J. Goyal, C. Holland, B. Gorovits, A. Morimoto, Pre-existing biotherapeutic-reactive antibodies: survey results within the American Association of Pharmaceutical Scientists, AAPS J. 15 (2013) 852–855. [65] L. Xue, B. Rup, Evaluation of pre-existing antibody presence as a risk factor for posttreatment anti-drug antibody induction: analysis of human clinical study data for multiple biotherapeutics, AAPS J. 15 (2013) 893–896. [66] M.W. Konrad, A.L. Childs, T.C. Merigan, E.C. Borden, Assessment of the antigenic response in humans to a recombinant mutant interferon beta, J. Clin. Immunol. 7 (1987) 365–375.

L. Yin et al. / Cellular Immunology 295 (2015) 118–126 [67] C. Ross, K.M. Clemmesen, M. Svenson, P.S. Sorensen, N. Koch-Henriksen, G.L. Skovgaard, K. Bendtzen, Immunogenicity of interferon-beta in multiple sclerosis patients: influence of preparation, dosage, dose frequency, and route of administration, Danish Multiple Sclerosis Study Group, Ann. Neurol. 48 (2000) 706–712. [68] H. Schellekens, Factors influencing the immunogenicity of therapeutic proteins, Nephrol. Dial. Transplant. 20 (Suppl. 6) (2005) vi3–vi9. [69] R.N. Maini, F.C. Breedveld, J.R. Kalden, J.S. Smolen, D. Davis, J.D. Macfarlane, C. Antoni, B. Leeb, M.J. Elliott, J.N. Woody, T.F. Schaible, M. Feldmann, Therapeutic efficacy of multiple intravenous infusions of anti-tumor necrosis factor alpha monoclonal antibody combined with low-dose weekly methotrexate in rheumatoid arthritis, Arthritis Rheum. 41 (1998) 1552–1563. [70] J. Oldenburg, R. Schwaab, H.H. Brackmann, Induction of immune tolerance in haemophilia A inhibitor patients by the ‘Bonn Protocol’: predictive parameter for therapy duration and outcome, Vox Sang. 77 (Suppl. 1) (1999) 49–54. [71] M.T. Reding, H. Wu, M. Krampf, D.K. Okita, B.M. Diethelm-Okita, N.S. Key, B.M. Conti-Fine, CD4+ T cell response to factor VIII in hemophilia A, acquired hemophilia, and healthy subjects, Thromb. Haemost. 82 (1999) 509–515. [72] M.T. Reding, H. Wu, M. Krampf, D.K. Okita, B.M. Diethelm-Okita, B.A. Christie, N.S. Key, B.M. Conti-Fine, Sensitization of CD4+ T cells to coagulation factor VIII: response in congenital and acquired hemophilia patients and in healthy subjects, Thromb. Haemost. 84 (2000) 643–652. [73] M. Franchini, F. Capra, N. Nicolini, D. Veneri, F. Manzato, F. Baudo, G. Lippi, Drug-induced anti-factor VIII antibodies: a systematic review, Med. Sci. Monit. 13 (2007) RA55–RA61. [74] M.P. Baker, H.M. Reynolds, B. Lumicisi, C.J. Bryson, Immunogenicity of protein therapeutics: the key causes, consequences and challenges, Self/nonself 1 (2010) 314–322. [75] S.E. Grossberg, J. Oger, L.D. Grossberg, A. Gehchan, J.P. Klein, Frequency and magnitude of interferon beta neutralizing antibodies in the evaluation of interferon beta immunogenicity in patients with multiple sclerosis, J. Interferon Cytokine Res. 31 (2011) 337–344. [76] A.J. Chirino, A. Mire-Sluis, Characterizing biological products and assessing comparability following manufacturing changes, Nat. Biotechnol. 22 (2004) 1383–1391. [77] J. Carpenter, B. Cherney, A. Lubinecki, S. Ma, E. Marszal, A. Mire-Sluis, T. Nikolai, J. Novak, J. Ragheb, J. Simak, Meeting report on protein particles and immunogenicity of therapeutic proteins: filling in the gaps in risk evaluation and mitigation, Biologicals 38 (2010) 602–611. [78] V.P. Yeung, J. Chang, J. Miller, C. Barnett, M. Stickler, F.A. Harding, Elimination of an immunodominant CD4+ T cell epitope in human IFN-beta does not result in an in vivo response directed at the subdominant epitope, J. Immunol. 172 (2004) 6658–6665. [79] T.D. Jones, W.J. Phillips, B.J. Smith, C.A. Bamford, P.D. Nayee, T.P. Baglin, J.S. Gaston, M.P. Baker, Identification and removal of a promiscuous CD4+ T cell epitope from the C1 domain of factor VIII, J. Thromb. Haemostasis: JTH 3 (2005) 991–1000. [80] M. Kosmac, T. Avcin, N. Toplak, G. Simonini, R. Cimaz, V. Curin Serbec, Exploring the binding sites of anti-infliximab antibodies in pediatric patients with rheumatic diseases treated with infliximab, Pediatric Res. 69 (2011) 243–248. [81] M.K. Joubert, Q. Luo, Y. Nashed-Samuel, J. Wypych, L.O. Narhi, Classification and characterization of therapeutic antibody aggregates, J. Biol. Chem. 286 (2011) 25118–25133. [82] S. Kumar, M.A. Mitchell, B. Rup, S.K. Singh, Relationship between potential aggregation-prone regions and HLA-DR-binding T-cell immune epitopes: implications for rational design of novel and follow-on therapeutic antibodies, J. Pharm. Sci. 101 (2012) 2686–2701. [83] A. Seidl, O. Hainzl, M. Richter, R. Fischer, S. Bohm, B. Deutel, M. Hartinger, J. Windisch, N. Casadevall, G.M. London, I. Macdougall, Tungsten-induced denaturation and aggregation of epoetin alfa during primary packaging as a cause of immunogenicity, Pharm. Res. 29 (2012) 1454–1467. [84] K.D. Ratanji, J.P. Derrick, R.J. Dearman, I. Kimber, Immunogenicity of therapeutic proteins: influence of aggregation, J. Immunotoxicol. 11 (2014) 99–109. [85] J.G. Barnard, K. Babcock, J.F. Carpenter, Characterization and quantitation of aggregates and particles in interferon-beta products: potential links between product quality attributes and immunogenicity, J. Pharm. Sci. 102 (2013) 915–928. [86] V. Rombach-Riegraf, A.C. Karle, B. Wolf, L. Sorde, S. Koepke, S. Gottlieb, J. Krieg, M.C. Djidja, A. Baban, S. Spindeldreher, A.V. Koulov, A. Kiessling, Aggregation of human recombinant monoclonal antibodies influences the capacity of dendritic cells to stimulate adaptive T-cell responses in vitro, PLoS ONE 9 (2014) e86322. [87] S. Hermeling, L. Aranha, J.M. Damen, M. Slijper, H. Schellekens, D.J. Crommelin, W. Jiskoot, Structural characterization and immunogenicity in wild-type and immune tolerant mice of degraded recombinant human interferon alpha2b, Pharm. Res. 22 (2005) 1997–2006. [88] S. Hermeling, H. Schellekens, C. Maas, M.F. Gebbink, D.J. Crommelin, W. Jiskoot, Antibody response to aggregated human interferon alpha2b in wildtype and transgenic immune tolerant mice depends on type and level of aggregation, J. Pharm. Sci. 95 (2006) 1084–1096. [89] P. Human, H. Ilsley, C. Roberson, E. Grovender, B. Van Antwerp, E. Fogt, P. Zilla, Assessment of the immunogenicity of mechanically induced interferon aggregates in a transgenic mouse model, J. Pharmaceut. Sci. (2014).

125

[90] V. Filipe, W. Jiskoot, A.H. Basmeleh, A. Halim, H. Schellekens, V. Brinks, Immunogenicity of different stressed IgG monoclonal antibody formulations in immune tolerant transgenic mice, mAbs 4 (2012) 740–752. [91] M.K. Joubert, M. Hokom, C. Eakin, L. Zhou, M. Deshpande, M.P. Baker, T.J. Goletz, B.A. Kerwin, N. Chirmule, L.O. Narhi, V. Jawa, Highly aggregated antibody therapeutics can enhance the in vitro innate and late-stage T-cell immune responses, J. Biol. Chem. 287 (2012) 25266–25279. [92] A.J. Freitag, M. Shomali, S. Michalakis, M. Biel, M. Siedler, Z. Kaymakcalan, J.F. Carpenter, T.W. Randolph, G. Winter, J. Engert, Investigation of the immunogenicity of different types of aggregates of a murine monoclonal antibody in mice, Pharm. Res. (2014). [93] V.S. Purohit, C.R. Middaugh, S.V. Balasubramanian, Influence of aggregation on immunogenicity of recombinant human Factor VIII in hemophilia A mice, J. Pharm. Sci. 95 (2006) 358–371. [94] D.S. Pisal, M.P. Kosloski, C.R. Middaugh, R.B. Bankert, S.V. Balu-Iyer, Nativelike aggregates of factor VIII are immunogenic in von Willebrand factor deficient and hemophilia a mice, J. Pharm. Sci. 101 (2012) 2055–2065. [95] M.M. van Beers, M. Sauerborn, F. Gilli, V. Brinks, H. Schellekens, W. Jiskoot, Aggregated recombinant human interferon Beta induces antibodies but no memory in immune-tolerant transgenic mice, Pharm. Res. 27 (2010) 1812– 1824. [96] A.H. Fradkin, J.F. Carpenter, T.W. Randolph, Glass particles as an adjuvant: a model for adverse immunogenicity of therapeutic proteins, J. Pharm. Sci. 100 (2011) 4953–4964. [97] J.C. Jones, E.W. Settles, C.R. Brandt, S. Schultz-Cherry, Virus aggregating peptide enhances the cell-mediated response to influenza virus vaccine, Vaccine 29 (2011) 7696–7703. [98] M. Ahmadi, C.J. Bryson, E.A. Cloake, K. Welch, V. Filipe, S. Romeijn, A. Hawe, W. Jiskoot, M.P. Baker, M.H. Fogg, Small amounts of sub-visible aggregates enhance the immunogenic potential of monoclonal antibody therapeutics, Pharm. Res. (2014). [99] A. Braun, J. Alsenz, Development and use of enzyme-linked immunosorbent assays (ELISA) for the detection of protein aggregates in interferon-alpha (IFN-alpha) formulations, Pharm. Res. 14 (1997) 1394–1400. [100] J. Skupsky, A.H. Zhang, Y. Su, D.W. Scott, A role for thrombin in the initiation of the immune response to therapeutic factor VIII, Blood 114 (2009) 4741– 4748. [101] A.M. Fathallah, R.B. Bankert, S.V. Balu-Iyer, Immunogenicity of subcutaneously administered therapeutic proteins–a mechanistic perspective, AAPS J. 15 (2013) 897–900. [102] A.R. Mire-Sluis, Y.C. Barrett, V. Devanarayan, E. Koren, H. Liu, M. Maia, T. Parish, G. Scott, G. Shankar, E. Shores, S.J. Swanson, G. Taniguchi, D. Wierda, L.A. Zuckerman, Recommendations for the design and optimization of immunoassays used in the detection of host antibodies against biotechnology products, J. Immunol. Methods 289 (2004) 1–16. [103] S. Gupta, S.R. Indelicato, V. Jethwa, T. Kawabata, M. Kelley, A.R. Mire-Sluis, S.M. Richards, B. Rup, E. Shores, S.J. Swanson, E. Wakshull, Recommendations for the design, optimization, and qualification of cell-based assays used for the detection of neutralizing antibody responses elicited to biological therapeutics, J. Immunol. Methods 321 (2007) 1–18. [104] L. Aarden, S.R. Ruuls, G. Wolbink, Immunogenicity of anti-tumor necrosis factor antibodies-toward improved methods of anti-antibody measurement, Curr. Opin. Immunol. 20 (2008) 431–435. [105] P.A. van Schouwenburg, G.M. Bartelds, M.H. Hart, L. Aarden, G.J. Wolbink, D. Wouters, A novel method for the detection of antibodies to adalimumab in the presence of drug reveals ‘‘hidden’’ immunogenicity in rheumatoid arthritis patients, J. Immunol. Methods 362 (2010) 82–88. [106] S.L. Wang, L. Ohrmund, S. Hauenstein, J. Salbato, R. Reddy, P. Monk, S. Lockton, N. Ling, S. Singh, Development and validation of a homogeneous mobility shift assay for the measurement of infliximab and antibodies-toinfliximab levels in patient serum, J. Immunol. Methods 382 (2012) 177–188. [107] S. Lacroix-Desmazes, N. Misra, J. Bayry, C. Artaud, B. Drayton, S.V. Kaveri, M.D. Kazatchkine, Pathophysiology of inhibitors to factor VIII in patients with haemophilia A, Haemophilia 8 (2002) 273–279. [108] P.A. van Schouwenburg, L.A. van de Stadt, R.N. de Jong, E.E. van Buren, S. Kruithof, E. de Groot, M. Hart, S.M. van Ham, T. Rispens, L. Aarden, G.J. Wolbink, D. Wouters, Adalimumab elicits a restricted anti-idiotypic antibody response in autoimmune patients resulting in functional neutralisation, Ann. Rheum. Dis. 72 (2013) 104–109. [109] R. Khanna, B.G. Feagan, Current and future status of therapeutic drug monitoring in the treatment of IBD, Curr. Treat. Options Gastroenterol. 12 (2014) 76–89. [110] F.I. Scott, G.R. Lichtenstein, Therapeutic drug monitoring of anti-TNF therapy in inflammatory bowel disease, Current Treat. Options in Gastroenterol. 12 (2014) 59–75. [111] C. Steenholdt, K. Bendtzen, J. Brynskov, O.O. Thomsen, M.A. Ainsworth, Clinical implications of measuring drug and anti-drug antibodies by different assays when optimizing infliximab treatment failure in Crohn’s disease: post hoc analysis of a randomized controlled trial, Am. J. Gastroenterol. 109 (2014) 1055–1064. [112] B. Gorovits, J. McNally, C. Fiorotti, S. Leung, Protein-based matrix interferences in ligand-binding assays, Bioanalysis 6 (2014) 1131–1140. [113] T.J. Messitt, F. Terry, L. Moise, W. Martin, A.S. De Groot, A comparison of two methods for T cell epitope mapping: ‘‘cell free’’ in vitro versus immunoinformatics, Immun. Res. 7 (2011).

126

L. Yin et al. / Cellular Immunology 295 (2015) 118–126

[114] L. Yin, J.M. Calvo-Calle, O. Dominguez-Amorocho, L.J. Stern, HLA-DM constrains epitope selection in the human CD4 T cell response to vaccinia virus by favoring the presentation of peptides with longer HLA-DM-mediated half-lives, J. Immunol. 189 (2012) 3983–3994. [115] V. Jawa, L.P. Cousens, M. Awwad, E. Wakshull, H. Kropshofer, A.S. De Groot, Tcell dependent immunogenicity of protein therapeutics: preclinical assessment and mitigation, Clin. Immunol. 149 (2013) 534–555. [116] C. Lundegaard, O. Lund, C. Kesmir, S. Brunak, M. Nielsen, Modeling the adaptive immune system: predictions and simulations, Bioinformatics 23 (2007) 3265–3275. [117] V. Brinks, D. Weinbuch, M. Baker, Y. Dean, P. Stas, S. Kostense, B. Rup, W. Jiskoot, Preclinical models used for immunogenicity prediction of therapeutic proteins, Pharm. Res. 30 (2013) 1719–1728. [118] N.K. Nanda, A.J. Sant, DM determines the cryptic and immunodominant fate of T cell epitopes, J. Exp. Med. 192 (2000) 781–788. [119] A.J. Sant, F.A. Chaves, S.A. Jenks, K.A. Richards, P. Menges, J.M. Weaver, C.A. Lazarski, The relationship between immunodominance, DM editing, and the kinetic stability of MHC class II: peptide complexes, Immunol. Rev. 207 (2005) 261–278. [120] X. Chen, T. Hickling, E. Kraynov, B. Kuang, C. Parng, P. Vicini, A mathematical model of the effect of immunogenicity on therapeutic protein pharmacokinetics, AAPS J. 15 (2013) 1141–1154. [121] X. Chen, T.P. Hickling, P. Vicini, A mechanistic, multiscale mathematical model of immunogenicity for therapeutic proteins: part 2-model applications, CPT: Pharmacomet. Systems Pharmacol. 3 (2014) e134. [122] X. Chen, T.P. Hickling, P. Vicini, A mechanistic, multiscale mathematical model of immunogenicity for therapeutic proteins: part 1-theoretical model, CPT: Pharmacomet. Systems Pharmacol. 3 (2014) e133. [123] J.D. Gomez-Mantilla, I.F. Troconiz, Z. Parra-Guillen, M.J. Garrido, Review on modeling anti-antibody responses to monoclonal antibodies, J. Pharmacokinet Pharmacodyn. 41 (2014) 523–536. [124] L. Moise, C. Song, W.D. Martin, R. Tassone, A.S. De Groot, D.W. Scott, Effect of HLA DR epitope de-immunization of Factor VIII in vitro and in vivo, Clin. Immunol. 142 (2012) 320–331. [125] V. Brinks, W. Jiskoot, H. Schellekens, Immunogenicity of therapeutic proteins: the use of animal models, Pharm. Res. 28 (2011) 2379–2385. [126] E. Koren, A.S. De Groot, V. Jawa, K.D. Beck, T. Boone, D. Rivera, L. Li, D. Mytych, M. Koscec, D. Weeraratne, S. Swanson, W. Martin, Clinical validation of the ‘‘in silico’’ prediction of immunogenicity of a human recombinant therapeutic protein, Clin. Immunol. 124 (2007) 26–32. [127] B. Manoury, D. Mazzeo, L. Fugger, N. Viner, M. Ponsford, H. Streeter, G. Mazza, D.C. Wraith, C. Watts, Destructive processing by asparagine endopeptidase limits presentation of a dominant T cell epitope in MBP, Nat. Immunol. 3 (2002) 169–174. [128] T. Burster, A. Beck, E. Tolosa, V. Marin-Esteban, O. Rotzschke, K. Falk, A. Lautwein, M. Reich, J. Brandenburg, G. Schwarz, H. Wiendl, A. Melms, R. Lehmann, S. Stevanovic, H. Kalbacher, C. Driessen, Cathepsin G, and not the asparagine-specific endoprotease, controls the processing of myelin basic

[129]

[130]

[131]

[132] [133]

[134]

[135]

[136] [137] [138] [139]

[140]

[141] [142]

[143]

protein in lysosomes from human B lymphocytes, J. Immunol. 172 (2004) 5495–5503. G.Y. Liu, P.J. Fairchild, R.M. Smith, J.R. Prowle, D. Kioussis, D.C. Wraith, Low avidity recognition of self-antigen by T cells permits escape from central tolerance, Immunity 3 (1995) 407–415. C.J. Harrington, A. Paez, T. Hunkapiller, V. Mannikko, T. Brabb, M. Ahearn, C. Beeson, J. Goverman, Differential tolerance is induced in T cells recognizing distinct epitopes of myelin basic protein, Immunity 8 (1998) 571–580. R.M. Kedl, W.A. Rees, D.A. Hildeman, B. Schaefer, T. Mitchell, J. Kappler, P. Marrack, T cells compete for access to antigen-bearing antigen-presenting cells, J. Exp. Med. 192 (2000) 1105–1113. R.M. Kedl, J.W. Kappler, P. Marrack, Epitope dominance, competition and T cell affinity maturation, Curr. Opin. Immunol. 15 (2003) 120–127. P.A. Muraro, M. Vergelli, M. Kalbus, D.E. Banks, J.W. Nagle, L.R. Tranquill, G.T. Nepom, W.E. Biddison, H.F. McFarland, R. Martin, Immunodominance of a low-affinity major histocompatibility complex-binding myelin basic protein epitope (residues 111–129) in HLA-DR4 (B1⁄0401) subjects is associated with a restricted T cell receptor repertoire, J. Clin. Invest. 100 (1997) 339– 349. Y. Li, Y. Huang, J. Lue, J.A. Quandt, R. Martin, R.A. Mariuzza, Structure of a human autoimmune TCR bound to a myelin basic protein self-peptide and a multiple sclerosis-associated MHC class II molecule, EMBO J. 24 (2005) 2968– 2979. Y. Yin, Y. Li, M.C. Kerzic, R. Martin, R.A. Mariuzza, Structure of a TCR with high affinity for self-antigen reveals basis for escape from negative selection, EMBO J. 30 (2011) 1137–1148. FDA, Guidance for industry immunogenicity assessment for therapeutic protein products, (2014). EMA, Guideline on immunogenicity assessment of biotechnology-derived therapeutic proteins, (2007). H. Schellekens, Bioequivalence and the immunogenicity of biopharmaceuticals, Nat. Rev. Drug Discovery 1 (2002) 457–462. P.A. van Schouwenburg, S. Kruithof, C. Votsmeier, K. van Schie, M.H. Hart, R.N. de Jong, E.E. van Buren, M. van Ham, L. Aarden, G. Wolbink, D. Wouters, T. Rispens, Functional analysis of the anti-adalimumab response using patientderived monoclonal antibodies, J. Biol. Chem. 289 (2014) 34482–34488. M. Sauerborn, V. Brinks, W. Jiskoot, H. Schellekens, Immunological mechanism underlying the immune response to recombinant human protein therapeutics, Trends Pharmacol. Sci. 31 (2010) 53–59. K.B. Blyuss, L.B. Nicholson, The role of tunable activation thresholds in the dynamics of autoimmunity, J. Theor. Biol. 308 (2012) 45–55. M. Chaparro, I. Guerra, P. Munoz-Linares, J.P. Gisbert, Systematic review: antibodies and anti-TNF-alpha levels in inflammatory bowel disease, Aliment. Pharmacol. Ther. 35 (2012) 971–986. C. Steenholdt, M.A. Ainsworth, M. Tovey, T.W. Klausen, O.O. Thomsen, J. Brynskov, K. Bendtzen, Comparison of techniques for monitoring infliximab and antibodies against infliximab in Crohn’s disease, Ther. Drug Monit. 35 (2013) 530–538.