The minipig as an alternative non-rodent model for immunogenicity testing using the TNFα blockers adalimumab and infliximab

Abstract

Immunogenicity is a major issue of concern for monoclonal antibodies used in human diseases and is by default mainly determined in non-human primates (NHP), as target molecules are considered most similar in NHP compared to human. In this manuscript the predictive value of immunogenicity testing in minipigs for human safety is evaluated, as the immune system of the pig is functionally similar to that in other mammalian species. Adalimumab and infliximab (both monoclonal antibodies blocking TNFα) were used as model substances. Female Göttingen minipigs (4/group) were treated every other week with low (0.1 mg/kg), mid (1.0 mg/kg), or high dose (5 mg/kg) adalimumab or 5 mg/kg infliximab subcutaneous (SC) over a period of 8 weeks. After first and last dosing, pharmacokinetic analysis was performed. Anti-drug antibodies (ADAs) were measured on several time points. Furthermore, hematology, clinical chemistry, body weight, clinical signs, and histopathology of several organs were evaluated. No signs of toxicity of the treatments were observed in the limited organs and tissues collected. Eleven out of 12 minipigs treated with adalimumab elicited a detectable ADA response. Induction of ADA was correlated with decreased plasma levels of adalimumab. Infliximab clearance was comparable after first and last dose. Therefore, the presence of ADA directed to infliximab was considered highly unlikely. It was concluded that the minipig and NHP showed comparable suitability for immunogenicity prediction in humans. More studies with other biopharmaceutical products are needed to strengthen the status of the minipig as an alternative model for immunotoxicity testing including immunogenicity.

• Adalimumab
• alternative model
• anti-drug antibodies
• immunogenicity
• infliximab
• minipig
• pharmacokinetics

Introduction

Over the past decades, interest in monoclonal antibodies as a therapy for several disorders has rapidly increased. Many monoclonal antibodies have been brought to market and there is an increasing number of monoclonal antibodies and other therapeutic proteins under development in several pharmaceutical companies (Chapman et al., 2007, 2009; Descotes, 2009; Brinks et al., 2011). One of the issues of concern for the safety of biopharmaceuticals is their potential immunogenicity. The characterization of the immunogenicity and immunotoxicity of biopharmaceuticals remains an important aspect of safety and efficacy testing in pre-clinical studies (Ryffel, 1997; Bugelski & Treacy, 2004).

Although the mechanisms underlying immunogenicity of biopharmaceuticals are still under investigation, it is clear that its occurrence is influenced by several factors. Incidence and disposition of an immunogenic response is dependent on intrinsic (protein-specific) factors that include among others, the protein structure, presence of T-cell epitopes, drug formulations, impurities, aggregation, and factors such as glycosylation and the covalent attachment of polyethylene glycol to the therapeutic protein (PEGylation). Furthermore, immunogenicity is influenced by extrinsic factors such as patient’s genetics, immune competence, and co-medication, as well as the route, dose, and frequency of administration (Kromminga & Schellekens, 2005). Unwanted immunogenicity induced by biologics comprises humoral and cellular immune responses with varying consequences for human efficacy (e.g. alterations in pharmacokinetics or pharmacodynamic parameters) and human safety (e.g. immune complex formation or general immune system effects such as serum sickness like disease, anaphylactic and infusion reactions) (Brinks et al., 2011).

Until now prediction of immunogenicity in humans has proven to be difficult. In silico modeling may help to identify T-cell epitopes, but the major limitation is the high level of false positive results (Baker et al., 2010). Several in vitro models exist for prediction of immunogenicity, but their predictive value is questionable (Bhogal, 2010). The described in vitro immunogenicity tests focus on T-cell-dependent immune responses, as it is challenging to mimic the B-cell-dependent humoral response in vitro. Therefore, the T-cell independent immunogenicity is ignored in these assays. In in vivo animal studies it is generally expected that the immune response will be more pronounced than in humans as most biologically derived pharmaceu-ticals are human or humanized proteins (Ponce et al., 2009; Brinks et al., 2011). Mainly due to species differences in protein structure and, as a result, the perceived foreignness of the drug construct, immunogenicity might be more pronounced in animal models. Therefore, animal studies are in general considered to have a low predictive value for immunogenicity of biopharmaceuticals in humans (Wierda et al., 2001; Bugelski & Treacy, 2004; Brinks et al., 2011).

Nevertheless, although animal models may have a low predictive value for the human situation, the results of immunogenicity studies may have important utility (Büttel et al., 2011). In particular, immunogenicity data from animal studies can be useful for interpreting drug pharmacokinetics (PK) and pharmacodynamics (PD), and can help to elucidate the mechanisms underlying antibody responses against therapeutic proteins (Sauerborn et al., 2010). Immunogenicity testing is focused on PK measurement and characterization of anti-drug antibodies (ADAs). Development and optimization of immunoassays for detection of these ADAs is essential, as well as assays to determine whether the ADAs are neutralizing antibodies (anti-idiotype antibodies that result in reduction of drug efficacy). PK measurements using a capture assay with the drug’s target (tumor necrosis factor [TNF]-α in the case of adalimumab and infliximab) will in principle address both the issue of neutralization as well as altered clearance rate due to immune complex formation, as only drug that is still able to bind TNFα will be detected in this assay.

The safety issues concerning immunogenicity of biopharmaceuticals are mainly studied in non-human primates (NHP) as, based on their close relation to man, they are often considered to be the best predictive animal species for human safety (Chapman et al., 2010) and because non-clinical safety testing of biologically-derived pharmaceuticals is preferably studied in animal species in which the drug is also biologically active. However, the use of NHP in pre-clinical safety issues raises several questions. The relevance of the cynomolgus or rhesus monkey is often far less optimal than too often claimed (Descotes and Gouraud, 2008). When designing a safety study, it is crucial to select the most relevant animal species, not only based on species cross-reactivity but also on knowledge of target affinity, functional potency, and pharmacological relevance (Chapman et al., 2007). Moreover, for ethical reasons it is expected that in the near future monkey studies will be limited or restricted to animal experiments for which there is no known alternative test species to predict human safety. Furthermore, as the protein-based therapies are more and more humanized, the desire for chimpanzee studies will increase. As the chimpanzee is an endangered species and no longer used in drug development, the demand for alternative relevant animal species for medical research is rising.

Currently, the minipig is considered a useful alternative non-rodent species for safety evaluation of (bio)pharmaceuticals (Bode et al., 2010; Forster et al., 2010a, b). Minipigs have been shown to be suitable and are recognized and accepted by regulatory authorities worldwide (van der Laan et al., 2010). Although the role for minipigs in safety testing of pharmaceu-ticals, food additives, and agrochemicals is increasing, there is still a significant gap in knowledge concerning the utility of the minipig for biologics/biotechnology products (Bode et al., 2010). The potential use of minipigs for immunotoxicity (including immunogenicity) testing and immunopharmacology testing of biopharmaceuticals in order to further evaluate their predictive value for human safety still needs more attention. The immune system of the pig has anatomical and organizational specificities but is functionally similar to other mammalian species (Bode et al., 2010). In an earlier study, we focused on the possibilities of immunotoxicity testing in minipigs (Penninks and van Mierlo, 2012). Upon immunosuppressive treatment of minipigs, immune pathology and several immune function tests (as requested by ICH S8 note for guidance for immunotoxicity testing of human pharmaceutics; European Medicines Agency, 2005) were successfully implemented (Penninks & Van Mierlo, 2012). In order to gain a better vision of the potential utility of the minipig and strengthen its position as a candidate species for safety studies for human biologics, the research described here was extended to immunogenicity testing.

The aim of the current study was to evaluate the feasibility of the minipig as an alternative species for immunogenicity testing of biopharmaceuticals and evaluate whether minipigs have an equal or better predictive value for human immunogenicity than NHP. For this purpose, a study was performed wherein female minipigs were treated every other week for 8 weeks with the marketed products adalimumab (Humira®, Abbott Biotechnology Deutschland, Wiesbaden, Germany) and infliximab (Remicade®, Centocor, Leiden, the Netherlands). These biopharmaceutical products are active in humans as TNFα blockers and used in chronic inflammation disorders like rheumatoid arthritis and Crohn’s disease. Based on in vitro TNFα binding studies TNFα, it was shown that adalimumab inhibits pig TNFα (Food and Drug Administration, 2002), whereas infliximab does not (European Medicines Agency, 2002). As these two biologics are directed against the same target in vivo but do not share biological activity in the pig, the immunogenic potential of these biologics in minipigs would yield valuable information on the utility of the minipig in immunogenicity testing for human safety. The PK of adalimumab and infliximab was followed in this study following the first and last immunization. The formation and characterization of ADA was followed in time, and both were compared to the results found in the open literature for other species, including NHP.

Materials and methods

Minipigs

Female Göttingen minipigs (ca. 3–3.5 months-old) were provided by Ellegaard Göttingen Minipigs (Dalmose, Denmark). The welfare of the animals was maintained in accordance with the general principles governing the use of animals in experiments of the European Communities (Directive 86/609/EEC) and Dutch legislation. This included approval of the study by the Ethical Review Committee of TNO.

Each study group was housed in one pen (indoors) under conventional conditions. The experimental room was ventilated with ≈10 air changes per hour and was maintained at a temperature of 20–24 °C and a relative humidity of at least 40% and not exceeding 70%. Lighting was artificial with a sequence of 12 h light and 12 h dark. The minipigs were fed a commercial diet (SMP (E) SQC; SDS Special Diets Services, Whitham, UK) and received a measured amount of food twice each day. The minipigs were provided ad libitum access to tap-water suitable for human consumption (quality guidelines according to Dutch legislation based on EC Council Directive 98/83/EC). The study was started after acclimatization to the laboratory conditions for 6 days.

Experimental design

Animals (4/group) were injected subcutaneously (SC) 5 times in the neck region with either adalimumab (Humira®, Abbott Biotechnology Deutschland, Wiesbaden, Germany) at three dose levels, viz. a low, mid, and high dose of, respectively, 0.1, 1, or 5 mg/kg body weight (BW), or infliximab (Remicade®, Centocor, Leiden, the Netherlands) at one dose level (5 mg/kg BW) with a 2-week interval between injections (i.e. on study days 0, 14, 28, 42, and 56). Test substance formulations were prepared at a fixed concentration of 0.5, 5, or 25 mg adalimumab/ml or 25 mg infliximab/ml, respectively, such that a fixed dose of 200 µl/kg BW could be administered. Dose volumes were adjusted for the latest recorded body weight for each individual animal to maintain a constant dose-level in terms of BW. On each treatment day, a fresh batch of test substances was prepared. After the final SC injection, a 4-week recovery period was initiated.

Test parameters

Clinical observations

Cage-side observations to detect signs of ill health and reaction to treatment as well as moribund or dead animals were conducted at least once daily throughout the study, up to the day of necropsy. Special attention was given to the injection sites. The body weight of each animal was recorded starting at Day −4, then at Day 0, and once weekly thereafter until the end of the study (last body weight measurement on day of euthanization).

Blood sampling

Blood samples of all animals were collected via the neck vein. Blood for serum preparation for ADA analysis was sampled in coagulation tubes (BD Vacutainer, SST™II Advance, Franklin Lakes, NJ) on Days 0, 14, 28, 42, 56, and on necropsy days 84/85 (at the necropsy days, n = 2 for each group). Blood for plasma preparation for PK analysis was sampled with K₂-EDTA as anti-coagulant pre-dose and at 2, 4, 6, 8, 10, 12, and 14 days after the first and last dose. Blood samples for hematology analysis were taken with K₂-EDTA as anti-coagulant at Days 0, 4, 14, 56, 60, and 70. When blood sampling (for ADA analysis and hematology) and study substance administration were scheduled on the same day, blood was sampled prior to administration of the study substances, except for Day 28, when blood sampling was performed 2–3 h after dosing.

Hematology

Hematologic parameters such as hemoglobin, packed cell volume, red blood cell count, reticulocytes, thrombocyte count, and white blood cell count (total and differential) were all assayed using an Advia 120 hematology analyzer (Bayer Healthcare, Dublin, Ireland). Prothrombin time was analyzed by Normotest (Technoclone, Vienna, Austria). The following parameters were also calculated using Provantis (Version 6.5; Instem, Staffordshire, UK): mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC).

Necropsy and pathology

On necropsy day 84 or 85 (n = 2 per group on each necropsy day) the animals were sedated by Dex Domitor (Dexmedetomidine HCl, 0.5 mg/ml) 0.5 ml/10 kg BW intramuscular (IM) +Dormicum (midazolam, 5 mg/ml), 0.5 ml/10 kg body weight i.m. Thereafter the animals were anesthetized with Na-pentobarbital 60 mg/ml (ca. 1 ml/kg BW) intracardially until deep anesthesia. The animals were then sacrificed by exsanguination from the axillary artery and immediately examined for external changes and (grossly) for pathological changes.

The following organs were collected for microscopic exami-nation: liver, skin (injection site and skin lesions), gastrointestinal tract, spleen, thymus, peripheral lymph nodes (draining and non-draining), kidneys, and lungs. Organ weights (paired organs together) were determined of the following organs; liver, spleen, thymus, kidney, and lung. The relative organ weights (g/kg) were calculated on the basis of the terminal body weight of the animals. The collected organs were preserved in a neutral aqueous phosphate-buffered 4% solution of formaldehyde. The tissues were embedded in paraffin wax, sectioned at 5 µm and stained with hematoxylin and eosin. Histopathological examination was performed by light microscopy on the collected organs and tissues of all animals.

Measurement of effective adalimumab and infliximab concentration

Adalimumab and infliximab concentrations were measured as described in van Schouwenburg et al. (2010). In short, recombinant TNFα was indirectly coated to Maxisorp ELISA plates (Greiner Bio-One, Frickenhausen, Germany) via a monoclonal antibody against TNFα (anti-TNF-7, Sanquin Reagents, Amsterdam, the Netherlands). Bound adalimumab and infliximab were detected using biotinylated polyclonal rabbit anti-idiotype specific for adalimumab or infliximab, respectively (Sanquin Reagents). Results were extrapolated from a titration curve generated using pure stocks of adalimumab or infliximab. For both assays, the detection limit was ≈2 ng/ml.

Analysis of ADA by antigen binding test (ABT)

Antibodies against adalimumab were detected by an antigen-binding test as described in van Schouwenburg et al. (2010). Briefly, sera were incubated with Sepharose-immobilized Protein A (Pharmacia/GE Healthcare, Uppsala, Sweden). Specific anti-adalimumab was detected using [¹²⁵I]-labeled adalimumab F(ab)₂ prepared and labeled as described by Hart et al. (2011). Antibody levels were compared to a standard serum containing ADA and expressed in arbitrary units (AU). This assay has proven to be less susceptible to drug interference than a bridging ELISA, but will still mainly pick up antibodies in excess to drug (Hart et al., 2011).

Pharmacokinetic analysis

Pharmacokinetic parameters [C_max, T_max, T_1/2, total clearance, AUC_0–∞ and mean residence time (MRT)] were calculated by non-compartmental regression analysis using WinNonLin 6.2 (Phoenix, AZ). The area under the plasma concentration vs time curve was extrapolated to infinity by mixed-log-linear regression analysis of the final part of the curve. A concentration of 0 µg/ml was assigned to all plasma samples which were below the lower limit of quantification of the analysis method, as adalimumab and infliximab were assumed not to be present endogenously.

Statistical analysis

Statistical analyses were performed using Provantis (Version 6.5; Instem). A one-way analysis of variance (ANOVA) was performed for the pre-treatment body weight data, pre-treatment hematology data, and organ weights. When variances were not homogeneous or data were not normally distributed, the data were step-wise log- or rank-transformed prior to the ANOVA, or a non-parametric test was performed (Kruskal-Wallis). If the ANOVA yielded a significant (i.e. p < 0.05) effect, inter-group comparisons with the control group were made using a Dunnett’s multiple comparison test. If Kruskal-Wallis yielded a significant effect, inter-group comparisons with the control group were made using Dunn’s test. A Fisher’s exact probability test was performed for incidences of histopathological changes. For body weight and hematology data at other timepoints, an ANCOVA test was performed with pre-treatment values as a covariate. Incidences of histopathological changes were statistically analyzed by a Fisher’s exact probability test. Tests were performed as two-sided tests, with results taken as significant where the probability of the results was p < 0.05.

Results

Clinical observations and body weights

No treatment-related clinical signs were observed during the whole study period. The minipigs were not significantly affected in their growth (expressed as body weight) by adalimumab or infliximab treatments. During the study period of almost 3 months the body weights of the minipigs increased from ≈6–7 kg at study start to ≈12–13 kg at necropsy (data not shown).

Hematology

In general, no consistent and/or dose-related changes in hematologic parameters were observed between treatment groups, except for a higher percentage of monocytes in the high-dose adalimumab group (compared to the low-dose group) during the first 2 weeks of the study (Days 0–14; reaching statistical significance on Day 14 only).

Organ weights

Mean absolute and relative organ weights of the kidneys, thymus, spleen, liver, and lungs collected from the minipigs in the different treatment groups are presented in Table 1. Absolute and relative organ weights were comparable in the different treatment groups and no statistically significant differences between groups were seen.

Table 1. Relative organ weights after adalimumab or infliximab treatment.

Compound	Dose (mg/kg)	Terminal body weight (kg)	Kidney	Thymus	Spleen	Liver	Lungs
Adalimumab	0.1	12.45 ± 1.9	3.64 ± 0.56	0.70 ± 0.32	1.35 ± 0.25	17.4 ± 0.2	6.03 ± 0.67
Adalimumab	1	12.73 ± 2.5	3.55 ± 0.57	0.60 ± 0.22	1.48 ± 0.22	16.8 ± 0.8	5.71 ± 0.39
Adalimumab	5	12.28 ± 2.1	3.97 ± 0.13	0.69 ± 0.14	1.60 ± 0.32	17.4 ± 2.1	5.73 ± 0.47
Infliximab	5	11.85 ± 1.8	4.14 ± 0.55	0.68 ± 0.34	1.37 ± 0.21	17.7 ± 1.2	5.65 ± 0.43

Organ weights are expressed in g/kg of the terminal body weight. Values shown are group means ± SD.

Statistical test ANOVA: *p < 0.05 versus low-dose (0.1 mg/kg) adalimumab group.

Plasma adalimumab levels and anti-adalimumab antibody titers

Figure 1 presents the plasma concentration vs time curves of adalimumab for individual minipigs during the study. In minipigs treated with 0.1 mg/kg adalimumab (animals 1, 3, 5, and 7), plasma adalimumab levels remained low during the whole study period. These animals all had a measurable amount of antibodies against adalimumab from Day 14 onwards (Figure 2). Three out of four minipigs treated with 1 mg adalimumab/kg (animals 9, 13, and 15) had background levels of the drug in the plasma (Figure 1). These three animals also developed a clear antibody response towards adalimumab (Figure 2). One minipig treated with 1 mg adalimumab/kg (animal 11) had a high plasma level of adalimumab (Figure 1). This same animal showed very low levels of ADA as well (Figure 2).

Figure 1. Adalimumab levels in the plasma of individual minipigs during time. Minipigs were treated SC every other week for 8 weeks with (a) 0.1, (b) 1.0, or (c) 5.0 mg adalimumab/kg.

Figure 2. Anti-adalimumab antibody (AAA) levels in serum of individual minipigs. Minipigs were treated SC every other week for 8 weeks with (a) 0.1, (b) 1.0, or (c) 5.0 mg adalimumab/kg. Anti-adalimumab levels were followed during time.

In the group of four minipigs treated with five bi-weekly injections of 5 mg adalimumab/kg, three out of four (animals 17, 19, and 21) had very low adalimumab concentrations in the plasma. One of these (animal 17) showed a clear ADA response on Day 14 that reached a maximal value on Day 42 and remained constant for the rest of the study period. In two of these animals (animals 19 and 21), ADA rose slower and increased to maximal values on Days 56 or 84/85, respectively. In these two minipigs, adalimumab levels clearly decreased when ADA-levels were detected. The fourth animal in this group had a decreased plasma adalimumab level only at the end of the study period. This animal also showed a maximal ADA level at the end of the study (see Figure 2, animal 23, Days 84/85) that might be responsible for the drop in adalimumab levels at study end (Figure 1, animal 23).

Plasma infliximab levels and anti-infliximab antibodies

The plasma infliximab levels of the minipigs that were treated with five bi-weekly injections of 5 mg infliximab/kg remained at a constant high level during the whole study period (Figure 3). ADA against infliximab could not be measured reliably, as the assay for measurement of ADA against infliximab is prone to assay interference by infliximab in the sample. This has also been reported for clinical studies performed by the manufacturer of infliximab (Janssen Biotech, Inc., 2013).

Figure 3. Infliximab levels in individual minipigs. Minipigs were treated SC every other week for 8 weeks with 5.0 mg infliximab/kg. Infliximab and anti-infliximab levels were followed over time. Arrows indicate treatment days. No anti-infliximab antibodies were observed during the study.

Pharmacokinetic evaluation

For adalimumab (Figure 4, Table 2) peak concentration in the plasma (C_max) and the area under the curve from 0 to infinity (AUC_0–∞) after the first injection increased almost linearly with dose level of adalimumab. The half-life and clearance were not dependent on the dose as determined on the first day of dosing. The half-life of adalimumab in the plasma was ∼0.7 days for all dose groups after the first injection, with the exception of one animal in the mid-dose adalimumab group that showed a half-life of 8.2 days. After the last injection on Day 56, the adalimumab was cleared from the plasma very fast following injection, resulting in plasma concentrations that were below the limit of quantification (which was 0.002 mg/ml). Therefore, pharmacokinetic analysis could not be performed in most of the animals after the last adalimumab injection. Only in the two animals in which no anti-adalimumab antibodies could be detected, half-life was a bit decreased (animal 15) or slightly increased (animal 17).

Figure 4. Plasma concentration vs time curves for individual minipigs receiving intramuscular dose of (a) 0.1, (b) 1.0, or (c) 5.0 mg adalimumab/kg after the first (Day 0) and final (Day 56) dose sessions. Sessions are indicated by arrows above the figure. Results presented here were used for PK analysis.

Table 2. Pharmacokinetic parameters for adalimumab and infliximab following SC dosing to female Göttingen minipigs every other week for 8 weeks (n = 4 per group).

	Adalimumab
Pharmacokinetic	Day 0			Day 56			Infliximab
parameter	0.1 mg/kg	1 mg/kg	5 mg/kg	0.1 mg/kg	1 mg/kg	5 mg/kg	Day 0	Day 56
Cmax (µg.ml)	1.2 ± 0.1	7.9 ± 1.6	42.5 ± 5.8	All plasma values < LLOQ	For n = 3,  all plasma values < LLOQ^a	For n = 3,  all plasma values < LLOQ^b	32.5 ± 7.9	57.4 ± 17.8
Tmax (days)	3.0 ± 1.0	7.5 ± 1.0	6.5 ± 1.0				2.5 ± 1.0	2.5 ± 1
Half-life (days)	0.7 ± 0.1	2.9 ± 3.6	0.7 ± 0.3				12.8 ± 4.1	19.6 ± 2.0
AUC(0–∞) (day*μg/ml)	8.5 ± 1.3	66.2 ± 25.9	357 ± 23				661 ± 129	1527 ± 472
Total clearance (L/kg*day)	0.012 ± 0.001	0.016 ± 0.007	0.014 ± 0.000				0.008 ± 0.002	0.004 ± 0.001
MRT (day)	4.9 ± 0.4	6.0 ± 1.1	6.0 ± 0.5				6.9 ± 0.2	11.8 ± 0.3

LLOQ, lower limit of quantification.

^aFor n = 3, all plasma values were < LOQ and no PK parameters could be calculated. For the remaining animal of this group, PK parameters were C_max = 18.8 µg/ml, T_max = 4 days, half-life = 6.2 days, AUC_(0-∞) = 295 day*μg/ml, total clearance = 0.003 L/kg*day, and MRT = 9.9 days.

^bFor n = 3, all plasma values were < LOQ and no PK parameters could be calculated. For the remaining animal of this group the PK parameters were C_max = 25.3 µg/ml, T_max = 2 days, half-life = 1.6 days, AUC_(0–∞) = 120 day*μg/ml, total clearance = 0.042 L/kg*day, and MRT = 4.1 days.

For infliximab (Figure 3, Table 2) the C_max after the fifth injection was higher than after the first injection, as a result of accumulation of the test substance in the plasma. Also, the half-life of infliximab in the plasma was slightly, but not significantly, increased after the fifth injection compared to the half-life after the first injection.

Pathology

Macroscopic changes

On the day the animals were euthanized, the animals were exami-ned for external changes and grossly for pathological changes. This macroscopic examination did not reveal treatment-related gross changes. One of the animals in the high-dose adalimumab group showed a small gall bladder with gallstones. While this is a known side-effect of adalimumab in humans, it cannot unequivocally be attributed to the treatment. Other observations were: a unilateral dark appearance of a superficial cervical lymph node in an animal treated with low dose adalimumab; a small spleen with irregular surface and few hemorrhages in an animal in the mid-dose adalimumab group; and agenesis of the gall bladder of one minipig treated with infliximab. These observations are considered to be incidental findings.

Microscopic changes

Although microscopic examination of the adalimumab-treated minipigs revealed several histopathological changes, no distinct treatment-related histopathological changes were noted. Some adalimumab-treated minipigs showed changes at the injection sites, such as mononuclear cell infiltrates; however, a relationship with the adalimumab dose was absent. The histopathological changes seen in other organs or tissues occurred in one or two animals only or the incidences were distributed about equally in all groups. Thus, a dose–response relationship for adalimumab was absent. In minipigs treated with infliximab, some histopathological changes were also observed. Most changes were observed in one animal only, like mononuclear cell infiltrate in the colon or liver and decreased cortex/medulla ratio in the thymus. In addition, three out of four infliximab-treated minipigs showed increased germinal center development.

Discussion

Despite the increasing role of minipigs in non-clinical safety testing of pharmaceuticals, food additives and agrochemicals, there is a significant gap in knowledge concerning the utility of the minipig for biologics and biotechnology products (Bode et al., 2010). Therefore, more research is needed to provide sufficient comparative experimental data for a rigorous evaluation of the value of the minipig for human drug safety testing and to evaluate whether minipigs may better reflect certain human drug-induced toxicities than traditionally used non-rodent toxicology models. In this study, the immunogenicity and pharmacokinetics of the clinically-used monoclonal antibodies (mAb) adalimumab (Humira®) and infliximab (Remicade®) was assessed in minipigs. This kind of informative studies will contribute to our understanding of the possible role of minipigs in safety evaluation, as immunogenicity data are of importance for interpretation of toxicological observations and pharmacokinetics in tests with biopharmaceuticals.

In the clinic, it becomes more and more general practice to first determine (effective) drug levels, and measure anti-drug antibodies (ADAs) only when drug levels are low. In this way, drug interference in the ADA assay is circumvented. The assay used in this study to measure ADA has proven to be less susceptible to drug interference than a bridging ELISA, but will still mainly measure antibodies in excess to drug (Hart et al., 2011). Therefore, in the study described in this manuscript, the same strategy is used as in the clinic, i.e. we chose effective (free) drug levels as the primary determinant to monitor immunogenicity and the ADA tests were used for confirmation of the results.

Both adalimumab and infliximab were developed as TNFα blockers and approved for treatment of several (auto)immune disorders such as psoriasis, Crohn’s disease, and rheumatoid arthritis. Adalimumab is a recombinant human IgG₁ monoclonal antibody (Abbott Laboratories, 2011), whereas infliximab is a chimeric human–mouse monoclonal IgG₁ antibody produced by a recombinant cell line from a genetically engineered form of the mouse monoclonal antibody prototype (Paserchia, 1999). The choice for these biologics was made based on the availability of immunogenicity data in primates and humans.

Upon exposure to adalimumab and infliximab, a normal growth of the animals (in terms of body weights) was seen. In addition, no clinical signs or (histo)pathological changes were observed in the organs collected. Thus, it can be concluded that the dosages of adalimumab and infliximab used were not in the toxic range for the minipig, in respect to the limited organs and tissues collected. Comparably, no toxicological effects were observed in cynomolgus monkeys or mice treated weekly intravenously (IV) for 4 weeks with 32 mg/kg adalimumab (Food and Drug Administration, 2002). For infliximab, at doses up to 30 mg/kg, no overt signs of toxicity were observed after 1–5 IV administrations in chimpanzees (Food and Drug Administration, 1998). Even though high ADA titers were observed in adalimumab-treated minipigs, no clear signs of immune complex deposition were observed in the selected organs (although small depositions cannot be excluded with the staining performed for immunopathology).

PK data on half-life of the humanized antibody adalimumab are comparable in monkeys and in humans (14–21 days in monkeys (European Medicines Agency, 2004) vs 10–20 days in man [package insert, Humira]). In the minipig, the observed half-life of (functional) adalimumab was much lower. The reason for this difference in half-life between species is unclear but should be taken into account when interpreting toxicity and immunogenicity results. After five adalimumab injections, clearance is so fast in the minipig that half-life could not be determined. The most logical explanation for this quick clearance of functional drug-antibody after repeated injections is formation of ADA in minipigs. Also, in the human situation, it is recognized that production of ADA is associated with lower serum adalimumab levels, leading to reduced clinical responses (Bartelds et al., 2007).

Adalimumab elicited an ADA response in 11 out of 12 minipigs. Immunogenicity of adalimumab after IV injection was also tested in two mouse studies that showed the formation of ADA already after one injection (Abbott Laboratories, 2011). The results for assessments of immunogenicity of adalimumab in the minipig were also in line with those published by the USFDA in cynomolgus monkeys. Cynomolgus monkeys were considered to be the most relevant species based on in vitro TNFα binding studies with TNFα of several animal species compared to human TNFα (Food and Drug Administration, 2002). From the pre-clinical studies performed in cynomolgus monkeys, it is clear that dose levels up to 10 mg/kg induced ADA production in all monkeys. At higher dose levels, the percentages of ADA⁺ monkeys decreased. This might indicate that tolerance induction occurred at these higher dose levels (i.e. ≥15.5 mg/kg [Food and Drug Administration, 2002]). On the other hand, drug present in the serum might have interfered in the analysis of ADA as complexes can be formed between adalimumab and antibodies directed to adalimumab. In studies with healthy human volunteers, the incidence of ADA formation towards adalimumab varied from near-zero to 33% (European Medicines Agency, 2004). In pivotal trials in patients with rheumatoid arthritis (RA), ADA formation towards adalimumab was identified in ≈12% of RA patients not using concomitant methotrexate treatment (Food and Drug Administration, 2002; van de Putte et al., 2004; European Medicines Agency, 2006); studies using an assay format similar to that used in the present study report ADA in ≈20% of RA patients after 1 year of treatment (Bartelds et al., 2011).

Interestingly, in pilot experiments, the ADA response to adalimumab as detected in the minipigs could not be inhibited with human polyclonal IgG F(ab)₂. This would implicate that all minipigs that developed an ADA response to adalimumab in the present study developed these specifically directed to the idiotype part of adalimumab. This would be an interesting finding and is under further investigation. After IV treatment in cynomolgus monkeys both anti-idiotype antibodies and anti-human-framework IgG were formed (Loyet et al., 2009). In humans, the drug-specific antibodies detected are not directed to the constant domains of the antibody (Bartelds et al., 2010) that would be in line with our preliminary data in the minipig.

Due to assay interference, the presence or absence of anti-infliximab antibodies could not be conclusively determined in the present study. In humans, presence of anti-infliximab antibodies is associated with a higher clearance of infliximab (Wolbink et al., 2006). The most relevant consequence of antibody formation is either an increased clearance of immune complexes or neutralization of the antigen-binding site, both resulting in decreased pharmacological availability (Lobo et al., 2004). Therefore, the serum levels of functional therapeutic antibody and its rate of decay can be seen as the best indication of how much clinically relevant ADA is formed (Aarden et al., 2008). Based on the absence of increased infliximab clearance in the minipig in this study, it is plausible to conclude that the level of anti-infliximab antibodies is low in these animals, or at least that the amount of clinically relevant antibodies against infliximab in the minipig is low. This is a surprising finding as infliximab, like adalimumab, is a foreign protein in minipigs.

Species cross-reactivity of infliximab is limited to human and chimpanzee; no cross-reactivity with TNFα originating from rhesus macaques, cynomolgus macaques, baboon, pigtail macaques, cotton-top tamarin, marmoset, pig, rabbit, rat, or mouse was shown (European Medicines Agency, 2002). The limited information available from chimpanzee studies shows that no or a low-level ADA response was observed, whereas in cynomolgus monkeys no anti-infliximab antibodies were observed (Food and Drug Administration, 1998; Ponce et al., 2009). The lack of clinically relevant levels of anti-infliximab antibodies in the minipig might be related to the lack of biological activity of infliximab in the minipig. In mice in which infliximab does not exert biological activity (i.e., in the DBA/2, BALB/c, and CD1 strains), no-to-very low levels of ADA were measured after IV injection(s) with infliximab. In contrast, clear ADA responses were observed in Tg197 mice (that express human TNFα and, thus, infliximab exerts a biological activity in these hosts) after intraperitoneal (IP) infliximab treatment (Food and Drug Administration, 1998). Clearly, the mechanism behind these outcomes requires further investigation.

In humans, 10–61% of patients develop antibodies to infliximab after IV treatment (Food and Drug Administration, 1998; Paserchia, 1999; Baert et al., 2003; European Medicines Agency, 2004; Wolbink et al., 2006). The presence of anti-infliximab antibodies seems to be dependent on the dose level of infliximab used. Fifty-seven per cent of subjects who received monotherapy of 1 mg infliximab/kg proved positive for anti-infliximab antibodies vs only 25% and 10% in the 3 and 10 mg/kg groups, respectively (Food and Drug Administration, 2002). In a placebo controlled Phase 2 study in patients with RA, the elimination half-life increased with increasing dose level (Paserchia, 1999). This correlation might be caused by an excess of infliximab present in the circulation leading to only small immune complexes that do not result in increased clearance. On the other hand, it might also be a matter of tolerance induced by the large amount of infliximab injected.

Conclusion

In conclusion, immunogenicity results obtained in the minipig, using the test substances adalimumab and infliximab, are overall comparable with those obtained in NHP. The same holds true for immunogenicity results obtained in the minipig using the recombinant human protein anakinra (van Mierlo et al., 2013). For infliximab, no reliable ADA-measurements could be performed, based on the high levels of infliximab detected during the whole study. We therefore can only conclude that no clinically relevant anti-infliximab antibodies were formed in the minipig. The results are supportive for the future role minipigs may play in safety assessment of human biopharmaceuticals, including drug-induced immunogenicity, as immunogenicity plays a critical role in the interpretation of observed alterations in PK, PD, and toxicity end-points. More studies with other biopharmaceutical products are needed to strengthen the status of the minipig as an alternative model for (immuno)toxicity testing including immunogenicity.

Declaration of interest

Niels-Christian Ganderup is employed at Ellegaard Göttingen Minipigs A/S, breeder and provider of minipigs. Dr Theo Rispens received honoraria for lectures.

Abbreviations
ABT	=	antigen binding test
ADA	=	anti-drug antibody
AU	=	arbitrary units
MRT	=	mean residence time
NHP	=	non-human primate
PD	=	pharmacodynamics
PK	=	pharmacokinetics
TNFα	=	tumor necrosis factor-α

Acknowledgments

The authors would like to thank all colleagues who contributed to this study.