WO2018083122A1

WO2018083122A1 - Human fcrn binding antibody for use in treatment of antibody mediated disease

Info

Publication number: WO2018083122A1
Application number: PCT/EP2017/077966
Authority: WO
Inventors: Peter Ulrichts; Torsten Dreier; Johannes Joseph Wilhelmus De Haard
Original assignee: Argenx Bvba
Priority date: 2016-11-01
Filing date: 2017-11-01
Publication date: 2018-05-11
Also published as: GB201618424D0

Abstract

A treatment regimen is disclosed for the treatment of human subjects afflicted with antibody mediated disease. The regimen comprises administering to the human subject an antibody molecule that binds to the human FcRn receptor with at least one of its CDRs and with a binding affinity at pH=6 of at least 10 nM. The regimen comprises administering a loading dose of at least 20 nM/Kg, and at least one maintenance dose of at least 10 nM/Kg.

Description

HUMAN FCRN BINDING ANTIBODY FOR USE IN TREATMENT OF

ANTIBODY MEDIATED DISEASE

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates generally to the fields of immunology and molecular biology. More specifically, the invention relates to a treatment regimen for IgG related diseases.

2. Description of the Related Art

Immunoglobulin G (IgG) constitutes the most prevalent immunoglobin class in the serum of man and other mammals. Despite fluctuations in rates of synthesis by B cells, IgGs are maintained at remarkably constant levels in the serum. If IgG homeostasis is disturbed, then pathology due to too high (hypergammaglobunemia) or too low

(hypogammaglobunemia) IgG levels can result. Studies indicate that the major

histocompatibility complex (MHC)-class I related receptor, FcRn, is involved in the homeostasis of serum IgGs (Ghetie et al., 1996; Junghans and Anderson, 1996; Israel et al., 1996). This receptor most likely acts as a salvage receptor, and this would be consistent with its known ability to transcytose IgGs in intact FcRn across the neonatal gut (Wallace and Rees, 1980; Rodewald and Kraehenbuhl, 1984) and yolk sac (Roberts et al., 1990; Israel et al., 1995) or placenta (Kristoffersen and Matre, 1996; Simister et al., 1996; Leach et al., 1996; Firan et al., 2001). More recent studies indicate that FcRn is also involved in the transport of IgGs across epithelial and endothelial cell barriers of diverse origin (Antohe et al., 2001; McCarthy et al, 2000; Spiekermann et al., 2002; Dickinson et al., 1999;

Kobayashi et al., 2002; Yoshida et al., 2004), and this has relevance to the delivery of IgG to different sites in the body. Thus, the use of protein engineering to modify the interaction site of an IgG with FcRn offers a way of modulating the serum persistence, distribution and transport of that antibody.

The interaction sites of FcRn on mouse IgGl (mlgGl ) and human IgGl (hlgGl) have been mapped using site-directed mutagenesis of recombinant Fc-hinge fragments, followed by analysis of these fragments both in vivo and in vitro (Kim et al., 1994b; Medesan et al., 1996; Medesan et al., 1997; Kim et al., 1999). From these studies, 1253 (EU numbering (Edelman et al, 1969)), H310, H435 and to a lesser extent, H436 (Y436 in hlgGl) play a central role in this interaction. These amino acids are located at the CH2-CH3 domain interface (Deisenhofer, 1981), and the mapping of the functional site to these residues is consistent with the crystallographic structure of rat FcRn complexed with rat Fc (Burmeister et al, 1994b; Martin et al, 2001).

The FcRn interaction site encompasses three spatially close loops comprised of sequences that are distal in the primary amino acid sequence. The central role of Fc histidines in building this site accounts for the marked pH-dependence (binding at pH 6.0, release at pH 7.2-7.4) of the Fc-FcRn interaction (Rodewald and Kraehenbuhl, 1984; Raghavan et al, 1995; Popov et al, 1996), as the pKa of one of the imidazole protons lies in this pH range. This pH dependence is essential for the release of FcRn bound IgG molecules when they come to the cell surface following intracellular recycling or transcytosis (Ghetie and Ward, 2000; Ober et al, 2004a). 1253, H310, H435 and to a lesser degree, H436, are highly conserved in IgGs of both human and rodent IgGs (Kabat et al, 1991). This, taken together with the isolation of a human homo log of FcRn (Story et al, 1994), indicates that the molecular mechanisms involved in IgG homeostasis and transport are common to both mouse and man and this has implications for the modulation of pharmacokinetics, distribution and delivery of IgGs to different body sites.

In studies to identify the FcRn interaction site on Fc, mutations of Fc fragments (comprising the Fc and hinge region) have been made that reduce the serum half-lives of the corresponding Fc fragments (Medesan et al, 1997; Kim et al, 1994a; Kim et al, 1999). In addition, Fc fragments or IgGs with increased affinity for binding to FcRn have been engineered (Ghetie et al, 1997; Shields et al, 2001; Hinton et al, 2004) and these molecules have increased serum persistence in mice (Ghetie et al, 1997) or cynomologous monkeys (Hinton et al, 2004).

Immunoglobulin Fc domains are also of great interest for purposes of studying the mechanisms of antibody interactions with further molecules of the immune system. These include, depending on the class of antibody, interactions with complement, and binding to specific receptors on other cells, including macrophages, neutrophils and mast cells. More detailed knowledge of the biology of Fc regions is important in understanding various molecular processes of the immune system, such as phagocytosis, antibody-dependent cell- mediated cytotoxicity and allergic reactions.

The production of a longer-lived Fc fragment or antibody having increased binding to Fc receptors is attractive, since such a fragment or antibody can be used, for example, to tag therapeutic reagents. This allows fewer doses of the agent to be used in therapy and possibly even allows lower doses of the agent to be used through its increased persistence in the bloodstream. Additionally, such molecules would be useful, in and of themselves, for therapy against pathogenic agents, cancer and autoimmune diseases. Such antibodies would also be predicted to be more efficiently transported across the placenta during the third trimester of pregnancy when FcRn is active in the maternal fetal transport of IgGs (Simister, 2003). As such, protective antibodies (e.g., anti-pathogen) could be delivered to the developing fetus.

In addition, there are multiple situations in which increased clearance of IgGs from the circulation would be desirable, e.g., in autoimmune diseases such as systemic lupus erythematosus where circulating autoreactive antibodies cause pathology, and in situations where toxins or drugs are to be cleared rapidly from the body using an antibody as a clearing agent. Increased clearance of an antibody should be achievable by using a molecule, such as an engineered antibody, that binds to FcRn with high affinity and does not dissociate rapidly at near neutral pH (unlike naturally-occurring antibodies). Such antibodies would not be released from cells, but would instead be predicted to remain bound to FcRn and block binding of other, lower affinity IgGs. As a result, FcRn function would be blocked and endogenous or therapeutic IgGs would be directed into the lysosomal pathway for

degradation (Ober et al., 2004b). The targeting of such 'blocking' antibodies to FcRn might also be useful in the prevention of transport of pathogenic (e.g., autoreactive) antibodies from mother to fetus during pregnancy. For the treatment of IgG mediated disease, in particular chronic IgG mediated disease, it is desirable to provide a treatment regimen that does not unduly compromise the patient's immune system. In addition, it is important to provide a treatment regimen that does not unduly burden the patient, so that compliance is more readily achieved. STJMMARY OF THE INVENTION

Thus, in accordance with the present invention, there is provided a method of treating a human subject having IgG mediated disease, said method comprising administering to the subject an antibody molecule having specific affinity for hFcRn at pH 6 of at least 10 nM, according to a regimen comprising administering to the subject a loading dose of at least 20 nMole/Kg and at least one maintenance dose of at least 10 nMole/Kg.

In particular, the present invention provides a method of treating a human subject suffering from antibody mediated disease, said method comprising administering to the subject an antibody molecule that binds to hFcRn with at least one of its CDRs and having a specific binding affinity for hFcRn at pH 6 of at least 10 nM, according to a regimen comprising administering to the subject a loading dose of at least 20 nMole/kg; and at least one maintenance dose of at least 10 nMole/kg.

In an embodiment the binding affinity of the compound i.e. the antibody molecule to hFcRn is pH dependent. In an alternate embodiment the binding affinity of the compound i.e. the antibody molecule to hFcRn is pH independent.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1: Single dose ARGX-113 dose-escalation study in cynomolgus monkeys;

pharmacodynamic effect on tracer antibody. Figure 2: Single dose ARGX-113 dose-escalation study in cynomolgus monkeys;

pharmacodynamic effect on endogenous IgG levels.

Figure 3: ARGX-113 PK profile of cynomolgus monkey dose-escalation study.

Figure 4: Pharmacodynamic effect on endogenous IgG levels in cynomolgus monkeys of single doses of ARGX-113 at 10, 30, 50 or lOOmg/kg. Figure 5: ARGX-113 PK profile following a single dose of ARGX-113 in cynomolgus monkeys at doses of 10, 30, 50 or lOOmg/kg.

Figure 6: Pharmacodynamic effect on endogenous IgG levels in C57BL/6 mice of single doses of ARGX-113 at 2, 20, or lOOmg/kg. Figure 7: ARGX-113 PK profile following a single dose of ARGX-113 in C57BL/6 mice at doses of 2, 20 or lOOmg/kg.

Figure 8: Pharmacodynamic effect on endogenous total IgG and IgG subclass levels in healthy volunteers following a single dose of ARGX-113 at 0.2, 2, 10, 25 or 50 mg/kg. Figure 9: Dose/PD relationship after single infusion of ARGX-113 in healthy volunteers at day 3, 6, 14 and 28 post infusion.

Figure 10: Pharmacokinetic profile of ARGX-113 in healthy volunteers following single infusion of ARGX-113 in healthy volunteers at doses of 0.2, 2, 10, 25 and 50 mg/kg.

Figure 11: Pharmacodynamic effect on endogenous total IgG and IgG subclass levels in healthy volunteers following multiple infusions of ARGX-113 in healthy volunteers (either 6 infusions at lOmg/kg once every 4 days (q4d) or 4 infusions at lOmg/kg once every 7 days (q7d).

Figure 12: Pharmacokinetic profile of ARGX-113 in healthy volunteers following multiple infusions of ARGX-113 in healthy volunteers (either 6 infusions at lOmg/kg once every 4 days (q4d) or 4 infusions at lOmg/kg once every 7 days (q7d).

DETAILED DESCRIPTION OF THE INVENTION

The following is a detailed description of the invention. Definitions The term "IgG mediated disease" as used herein means a disease characterized by the presence in a subject's serum of pathological IgG molecules, such as autoimmune and alloimmune IgG antibodies. The term "IgG mediated disease" thus includes autoimmune and alloimmune diseases.

The term "hFcRn" as used herein refers to the human neonatal receptor for Fc. The term "antibody molecule having a specific binding affinity for hFcRn" as used herein refers to any molecule having a specific affinity for hFcRn that is greater, at pH 6, than the affinity of the wild type ("WT") Fc domain of a human IgG antibody molecule. Examples include full anti-hFcRn antibodies, for example as disclosed in US Patent No. 7,662,928 to Balthasar; WO 2006/118772 A2 to The Jackson Laboratory; WO 2009/131702 A9 to Dyax; WO 2014/019727A1 to UCB; WO 2015/167293 A2 to Hanall Biopharma; WO 2016/123521 to Momenta; and US 2016/0264668 to UCB; the disclosures of all of which are incorporated herein by reference. The term "specific binding affinity to hFcRn" as used herein refers to the KD value of the binding molecule, expressed in nM, as measured with Biacore. As explained in more detail herein below, the binding may be pH dependent or may be pH independent. For its mechanism of action the compound i.e. the antibody molecule must have greater affinity at endosomal pH, i.e., a pH of about 6, than WT hFc. The term "administering" as used herein refers to any method for introducing the molecule into the body of the subject, including but not limited to intravenous infusion; intramuscular injection; subcutaneous injection (including injection by a wearable device); transdermal administration; transmucosal administration, for example sublingual

administration or by means of an inhaler; oral administration or in the form of suppositories, and the like. The type of administration available to the practitioner depends in large measure upon the nature of the molecule, its molecular weight, its resistance to stomach acids and digestive enzymes; and the like.

The term "loading" dose as used herein refers to a first dose of the compound administered to the subject at the start of the treatment regimen. In general the loading dose is intended to achieve a significant degree of saturation of hFcRn receptors in the body of the subject.

The term "maintenance dose" as used herein relates to a dose of the compound administered to the subject at a point in time after administration of the loading dose, in general at a point in time when the level of the antibody molecule in the subject's serum shows a downward trend.

The term "reference IgG serum level" as used herein refers to the level of IgG in the subject's serum prior to commencement of the treatment regimen of the invention. The reference IgG serum level serves as a benchmark in determining the effectiveness of the compound in lowering the overall IgG serum level of the subject. The term "s.c. penetration enhancer" as used herein refers to a penetration aid administered in conjunction with the compound in the case of subcutaneous (s.c.)

administration. In general, s.c. administration is preferred over intravenous (i.v.)

administration for ease of use, lower risk of infections, and lower burden on the patient. However, the amount of compound i.e. antibody molecule that can be successfully administered s.c. is limited, which is particularly a problem when the compound has a high molecular weight, as is the case for full anti-FcRn antibodies and, to a lesser extent, mutated Fc domains. Penetration enhancers generally increase the space between cells in the s.c. administration area, thereby providing more space for the compound. In this manner, penetration enhancers increase the amount of compound that can be administered s.c, and may make the difference between allowing s.c. administration in lieu of i.v. administration.

A particular class of s.c. penetration enhancers is formed by the hyaluronidase enzymes, in particular human hyaluronidase enzymes. Specific examples include Hydrase, Amphadase, Wydase, and Hylenex, the last one being available from Halozyme

Therapeutics, 11388 Sorrento Valley Road, San Diego, CA 92121, USA.

The term "body mass index" or "BMI" of an individual is defined as the weight of the individual (in Kg) divided by the square of the height of the individual (in meters). The SI unit of BMI accordingly is Kg/m². An individual is considered overweight if his or her BMI exceeds 25 Kg/m². The term "ARGX 113" as used herein refers to a 54 kDa variant human IgG Fc region molecule comprising the amino acids Y, T, E, K, F and Y at EU positions 252, 254, 256, 433, 434, and 436, respectively, as described in WO2015/100299, the contents of which are incorporated herein in their entirety. It binds to hFcRn with binding affinity at pH=6 of about 8 nM. Its binding affinity to hFcRn is pH dependent.

Scientific Background of the Treatment Regimen

IgG antibodies have a significantly longer half-life than other antibody isotypes, such as IgA or IgM. The reason is that IgG antibodies comprise an Fc domain, which binds to the hFcRn receptors that are present in the cells of many tissues in the human body. IgG molecules, together with other serum proteins, are taken up by the cells in pinocytosis. In general, the proteins taken up by a cell are degraded in the lysosome of the cell. IgG molecules with an intact Fc domain are rescued from degradation through binding to FcRn in an acidified endosome at pH=6. IgG bound to FcRn is recycled to the surface of the cell, and released back into the serum at the physiological pH= about 7.4. This release is possible because the binding of the Fc domain of an IgG molecule to hFcRn is pH dependent. The binding is much stronger at the acidic pH of the endosome than it is at the neutral pH of the serum.

The antibody molecules used in the treatment regimen of the present invention interfere with this mechanism by binding to the hFcRn receptor, and by doing so more strongly than the Fc domain of an IgG molecule. It will be appreciated that the relevant comparison is the binding affinities at pH=6. The compound used in the therapy binds more strongly to the hFcRn receptor at pH=6. This prevents an IgG molecule from binding to the hFcRn receptor in the acidified endosome. Not being bound to the hFcRn receptor, the IgG molecule is degraded by the cell, just like other serum proteins. In other words, the compounds i.e. the antibody molecules used in the treatment regimen of the present invention prevent IgG molecules present in the serum of the subject from binding to FcRn receptors present in the subject's cells. As a result, the IgG molecules are not rescued from degradation, resulting in a significantly shorter half-life of the IgG molecules. The efficacy of the compounds i.e. the antibody molecules in reducing serum IgG levels are generally demonstrated in one, or both, of the following animal models.

In the mouse model, transgenic mice are used that do not express the mouse FcRn receptor (mFcRn-/-), and instead express the human FcRn receptor (hFcRn+/+). The animals receive a large dose of human IgG, for example 500 mg/Kg. One cohort of the animals receives a dose of the hFcRn binding compound, while a control cohort receives just the vehicle. The cohort receiving the hFcRn binding compound clears the hlgG more rapidly than the control cohort. Although typically a dose-response relationship is observed, these results cannot be used for establishing a desired dose for human treatment, because it is unknown how the expression rate of hFcRn in transgenic mice relates to the expression rate of hFcRn in humans. In the cynomolgus monkey model, cynomolgus monkeys receive the compound, and the serum IgG level of the animals is monitored. The FcRn receptor of cyno is highly homologous to hFcRn, so that the compound binds to cynoFcRn. Although a dose-response correlation can be established in these experiments, the results cannot be used for establishing a desired dose for human treatment, because it is unknown how the expression rate of cynoFcRn relates to the expression rate of hFcRn in humans.

Many hFcRn binding compounds are designed to exhibit pH independent binding to hFcRn. Although such compounds are effective in blocking hlgG binding to hFcRn, and thus in reducing the half- life of hlgG, the half- life of the compound is affected by the fact that the compound is not released from hFcRn, because its binding to hFcRn is pH independent. This shortens the half-life of such compounds.

Many hFcRn binding compounds are designed to exhibit high affinity to hFcRn, for example in the picomolar range. Although such compounds are effective in blocking hlgG binding to hFcRn, and thus in reducing the half- life of hlgG, the half- life of the compound is affected by the fact that the compound is not released from hFcRn, because its binding to hFcRn is too strong.

The treatment regimen of the present invention can be conducted with any compound that binds more strongly to hFcRn than does WT hFc. As such, any compound having affinity for hFcRn at pH=6 of 10 nM or less is suitable. For the reasons explained above, however, preferred for use in the treatment regimen of the present invention are compounds having affinity for hFcRn at pH=6 in the range of from 10 to 1 nM.

Both compounds having pH dependent binding to hFcRn and compounds having pH independent binding to hFcRn are suitable for the treatment regimen of the present invention. For the reasons explained above, however, preferred for use in the treatment regimen of the present invention are compounds having pH dependent binding to hFcRn, for example compounds having a pH dependence ratio (defined as (affinity for hFcRn at pH=6)/(affinity for hFcRn at pH=7.4)) of less than 0.1, preferably less than 0.05, more preferably less than 0.02.

WO 2015/167293 to Hanall Biopharma discloses anti-hFcRn antibodies exhibiting pH independent binding to hFcRn. This binding is reported to be non-competitive with hFc. The reference theorizes that non-competitive binding permits a lower dosing level. We believe this is incorrect. According to our understanding, the required dosing is determined by the molecular weight of the compound and the desired degree of saturation of hFcRn receptors, as explained in more detail below. We conducted clinical trials with the ABDEG™ (ARGX 113) molecule in healthy human volunteers. The ABDEG™ molecule consists of the constant domain of hlgGl with variant Fc domains comprising comprise amino acids Y, T, E, K, F, and Y at positions 252, 254, 254, 433, 434 and 436 (EU numbering), respectively. The molecular weight of the molecule is 54 kDa. In the trial the ABDEG™ molecule was tested at single ascending doses of 0.2; 2; 10;

25 and 50 mg/Kg, respectively. Details are provided in Example 1. IgG clearance was observed in a dose dependent manner for doses 0.2; 2 and 10 mg, and at identical rates for doses 10, 25 and 50 mg/Kg. This indicates that a dose of 10 mg/Kg is sufficient to obtain saturation of the available hFcRn receptors. In other words, the amount required for obtaining saturation of the available hFcRn receptors is in the range of from 2 mg/Kg to 10 mg/Kg (37 to 185 nMole/Kg), most likely near the upper end of the range. The binding stoichiometry ABDEG™/hFcRn is believed to be 1 :2.

The treatment regimen of the present invention is based upon the insight that the rate of IgG clearance is determined by the degree of saturation of the hFcRn receptors by the hFcRn binding compound, independent from the binding affinity of the compound (provided that the threshold binding activity of 10 nM is met). Accordingly, the treatment regimen comprises administering a loading dose of at least 20 nMole/Kg, preferably at least 50, 80, 100, 120 nMole/Kg, more preferably at least 150 nMole/Kg.

Since saturation is obtained at 185 nMole/Kg, it is preferred that the loading dose not exceed 300 nMole/Kg.

The loading dose is followed by at least one maintenance dose. The at least one maintenance dose may be equal to the loading dose, or less than the loading dose. In most cases the at least one maintenance does is less than the loading dose. The at least one maintenance dose is in the range of 10 nMole/Kg to 300 nMole/Kg, preferably from 15, 20, 30, 40, 50, 100, 120, 150, 180, 200 nMole/Kg to 50, 80, 100, 120, 150, 200 nMole/Kg. -l ilt will be understood that the body weight of a subject is not a perfect proxy for the number of hFcRn receptors present in the individual's body. The bodies of overweight and obese subjects are characterized by a larger number of fat cells, which do not contain hFcRn receptors but do contribute to the body's weight. Dosing overweight or obese subjects based on body weight would result in overdosing such individuals. It is desirable to use a corrected weight instead. The corrected weight of a subject having a BMI > 25 is calculated as the weight of a person having the same height as the subject and having a BMI of 25.

By way of example, a subject of height 1.80 m and weight of 120 Kg has a BMI of 120/(1.80)² = 37. The corrected weight of the subject is 25 * (1.80)² = 81 Kg. If the desired dose is 100 nMole/Kg, preferably this subject is administered a dose of 8,100 nMole (based on the subject's corrected weight) instead of 12,000 nMole (based on the subject's actual weight).

The interval between the loading dose and the (first) maintenance dose is determined by factors such as the desired rate of IgG clearance from the subject's serum and the half-life of the hFcRn binding compound. We have found that the half- life of ARGX 113 ABDEG™ in a human subject is 3-4 days, which is considerably longer than in cyno (about 1.5 days). hFcRn binding compounds that bind hFcRn very strongly and/or in a pH independent manner can be expected to have a shorter half-life than ABDEG™. This can be compensated by selecting shorter intervals between subsequent maintenance doses. Another consideration is the amount of compound to be dosed in each maintenance dose. It may be desirable to decrease this amount in order to permit s.c. administration instead of i.v. administration, even at the expense of a shorter interval between maintenance doses.

In an embodiment the interval between the loading dose and the (first) maintenance dose is from 4 days to 20 days, preferably from 5 to 10 days. In an embodiment the interval between individual maintenance doses is from 5 days to 4 weeks, preferably from 1 week to 2 weeks.

The duration of the treatment regimen is determined in function of the disease being treated. Certain types of IgG mediated diseases are characterized by occasional flare-ups. An example is myasthenia gravis (MG). Treatment may be started when a flare-up is first diagnosed, and continued for a period long enough to allow the subject's body to regain a certain equilibrium, for example 3 weeks, 4 weeks, 6 weeks, 8 weeks, 3 months or 4 months. Effectiveness of the treatment may be monitored by monitoring a biomarker, such as total IgG titer, the titer of a specific autoimmune antibody, or one of the symptoms of the disease being treated, for example the subject's muscle strength in the case of MG.

Other types of IgG mediated diseases may require the treatment regimen to be continued over a prolonged period of time, for example 6 months or more, 9 months or more, 12 months or more, 18 months or more, 2 years or more, or 3 years or more. It is useful to define a target overall serum IgG titer to be achieved by the treatment, for example less than 80%, less than 70%, less than 60%, less than 50%, less than 40% or less than 30% of the reference IgG serum level. The IgG serum level may be monitored during the treatment regimen, and the maintenance dose level may be adjusted, and/or the maintenance dosing interval may be adjusted, if the serum IgG level starts deviating from the target.

In an embodiment the compound is administered s.c. In an embodiment s.c.

administration is performed in conjunction with administration of a s.c. penetration enhancer, such as a hyaluronidase enzyme, in particular a human hyaluronidase enzyme. Specific examples include Hydrase, Amphadase, Wydase, and Hylenex, the last one being available from Halozyme Therapeutics, 11388 Sorrento Valley Road, San Diego, CA 92121, USA.

In an embodiment s.c. injection is done by means of a wearable injection device. Such devices offer several advantages in terms of patient comfort and ease of compliance.

Examples include SmartDose Electronic Injector from West Pharmaceutical Services Inc, Exton, PA, USA; Enable Injector from Enable Injections Inc., Cincinnati, OH, USA; and Precision-Therapy™ of Unilife Corporation, King of Prussia, PA, USA.

EXAMPLES

Example 1 : ARGX-113 dose-escalation PK/PD in cynomolgus monkeys

A dose-escalation study of ARGX-113 was performed in cynomolgus monkeys to determine the onset of the pharmacodynamics effect as well as the saturating dose. To this end, cynomolgus monkeys were administered 1 mg/kg of an anti-murine CD70 hlgGl tracer antibody (FR70-hIgGl) by i.v. bolus injection. Animals were infused 48 hours later with various doses of ARGX-113 (0.2 mg/kg, 2 mg/kg, 20mg/kg or 200mg/kg) or vehicle (PBS). Infusion was performed within 3 hours and animals were administered a volume of 36.36 ml/kg. Each test group consisted of 2 animals. Blood samples (3χ150μ1) were taken 5 minutes prior to dosing ("pre-dose") and 5 min, 2h, 6h, 24h, 48h, 72h, 5 days, 7 days, 10 days and 14 days after the end of the infusion. Both tracer IgG (Figure 1) and endogenous IgG levels (Figure 2) were determined by ELISA and plotted relative to pre-dose levels. Dosing animals with 0.2 mg/kg ARGX-113 does not significantly influence the rate of tracer clearance nor does it affect endogenous IgG levels. A clear pharmacodynamic effect is seen starting at 2mg/kg and levels out at doses starting from 20mg/kg. A single administration of ARGX-113 reduces cynomolgus monkey IgGs by maximally 55% within 3-4 days. Next, the pharmacokinetic profile of ARGX-113 was determined via ELISA. ARGX-

113 has a short half- life as has been calculated from its PK profile displayed in Figure 3 (estimated half-life -1.5 days). At saturating levels, non-FcRn binding ARGX-113 will be cleared efficiently due to its own mode of action. In addition, the molecular size of ARGX- 113 (54 kDa) is close to the cut-off for renal clearance (~60kDa). In a follow-up experiment, the FcRn saturating dose in cynomolgus monkeys was further explored. In this study, ARGX-113 was infused intravenously over 2 hours at doses of 10, 30, 50 and 100 mg/kg. A vehicle-treated control group was also included in this study. Each dose cohort consisted of 4 animals. Blood samples were taken on test day 1 (prior to the start of the infusion and at the end of the 2h infusion), and on test days 3, 5, 7, 9, 12, 15, 18 and 21. Endogenous IgG levels were determined by ELISA and are shown in Figure 4. A clear pharmacodynamic effect was observed for all ARGX-113-treated animals. Very similar profiles were observed for the 30, 50 and 100 mg/kg dose groups whilst the 10 mg/kg group displayed an intermediate response. In this experiment a maximal depletion of endogenous cynomolgus IgG of -60% was observed. An ELISA assay was used to determine the serum concentration of ARGX-113

(Figure 5). In line with previous experiment, the calculated serum half-life elimination t_½ of ARGX-113 ranged between 21.3 to 34.5 hours in this study. Example 2: ARGX-113 dose-escalation PK/PD study in C57BL/6 mice

Murine FcRn is able to bind human IgGs and IgG Fc-fragments and therefore we explored the IgG-depleting potency of ARGX-113 in mice. To this end, C57BL/6 mice were injected with ARGX-113 via the tail vein. Doses of 2, 20 and 100 mg/kg were injected and each dose group consisted of 4 animals. Serum samples were taken at Test Day -3 (pre-dose), Test Day 1 (pre-dose and lh after injection) and on Test Days 2, 3, 5 ,8 and 15. In analogy with the experiments in cynomolgus monkeys, endogenous IgGs and ARGX-113 serum levels were determined via ELISA.

A similar maximal decrease in IgGs was seen in C57BL/6 mice compared to the experiments in cynomolgus monkeys (-55-60%) but this maximal effect was reached much earlier (at 2 days after injection vs. 4-5 post infusion for cynomolgus monkeys) (Figure 6). Pharmacokinetic analysis also revealed big inter-species differences. Whilst ARGX-113 had a serum half-life of about 1 ,5 days in cynomolgus monkeys, a much faster clearance was seen in mice (no reliable half-life calculation possible; Figure 7).

Example 3: ARGX-113 dose-escalation PK/PD study in healthy volunteers

Next, a dose-escalation study in healthy volunteers was initiated to evaluate the pharmacokinetic and pharmacodynamic effects of ARGX-113 in humans.

Healthy volunteers were dosed with 0.2, 2, 10, 25 or 50 mg/kg via a 2-hour intravenous infusion. A total of four subjects were infused per dose group. Serum samples were taken at following Test Days: Test Day 1 (pre-infusion, directly after infusion, 4 hours after start of infusion and 8 hours after infusion), 2, 3, 4, 5, 7, 15, 22 and 29. Endogenous total IgGs were determined via an ELISA whilst IgG subtypes were determined via a

Luminex assay. A clear dose-dependency of the pharmacodynamic effect was observed in the healthy volunteers (Figure 8). No pharmacodynamics effect was observed at 0.2 mg/kg. As of 2 mg/kg, IgG levels started to decline in some subjects following ARGX-113 infusion. This reduction in IgG levels was significantly increased at doses of 10 mg/kg and higher. A maximal decrease of 60% was observed and the nadir of this PD effect was situated between day 7 and day 15 of the study after which IgG levels returned to baseline in these subjects. For ARGX-113-treated individuals in the 10, 25 and 50 mg/kg dose groups, serum IgG concentrations returned to 55%-75% of baseline concentrations at the last day of the study (Day 29).

A saturation of the PD effect was observed at a dose of 10 mg/kg as no significant difference was observed comparing healthy subjects dosed with 10, 25 or 50 mg/kg ARGX- 113 in terms of depth of nadir, time to nadir, or kinetics of Ig increase following nadir for all IgG subtypes (Figure 9). This contrasts with the findings in other species where saturation was only achieved at higher doses. In addition, the pharmacodynamic effect was much more prolonged in humans compared to similar experiments in cynomolgus monkeys. The pharmacokinetic profile of ARGX-113 in healthy volunteers is shown in Figure

10. Again, large interspecies differences are observed. Whilst the serum half- life in cynomolgus monkeys on average was ~1,5 days, the serum half- life in humans varied from 86 to 99 hours, dependent on the dose group.

Next, the effect of multiple doses of ARGX-113 was evaluated. Five healthy volunteers were infused with lOmg/kg ARGX-113 either at 6 occasions once every 4 days and six volunteers were infused with lOmg/kg ARGX-113 at 4 occasions once every seven days. Pharmacodynamic effects are displayed in Figure 11. Compared to single infusion, IgG concentrations decreased significantly more in some subjects following repetitive infusions. No significant differences in PD were seen when comparing both multiple dose treatment regimen as both depth of the PD effect and time to reach nadir are very similar. At nadir, IgG levels are decreased by 70-85% from baseline dependent on IgG subclasses and subjects. In some subjects dosed every 4 days maximum PD effect was reached before the 5^th infusion (prior to Day 17) for IgGl and IgG3 and before the last infusion (Day 21) for IgG2 and IgG4. Additional dosing did not further decrease the IgG levels. In subjects dosed every week, maximum PD effect was reached around Day 20 for all IgG subtypes. Starting from 1 week post last infusion, a slow, steady increase in serum IgG concentrations was observed for all subjects in both treatment groups. Terminal half-life in the multiple dose was 87.5 h in the group treated every four days and 95.3 hours in the group treated every week, which was comparable with the half- life calculated in the single dose groups (Figure 12). From these experiments we conclude that effective therapy in humans is possible with lower doses than would be predicted based on comparable animal data. We further conclude that in a treatment regimen for human subjects the dosing intervals may be longer than would be predicted based on comparable animal data.

Thus, the invention has been described by reference to certain embodiments discussed above. It will be recognized that these embodiments are susceptible to various modifications and alternative forms well known to those of skill in the art.

Many modifications in addition to those described above may be made to the structures and techniques described herein without departing from the spirit and scope of the invention. Accordingly, although specific embodiments have been described, these are examples only and are not limiting upon the scope of the invention.

Claims

1. A method of treating a human subject suffering from antibody mediated disease, said method comprising administering to the subject an antibody molecule that binds to hFcRn with at least one of its CDRs and having a specific binding affinity for hFcRn at pH=6 of at least 10 nM, according to a regimen comprising administering to the subject:

a. a loading dose of at least 20 nMole/kg; and

b. at least one maintenance dose of at least 10 nMole/kg.

The method of claim 1 wherein the maintenance dose is lower than the loading dose.

The method of claim 1 or 2 wherein the regimen comprises administering more than one maintenance dose.

The method of claim 3 wherein the maintenance doses are administered later in time than the loading dose.

The method of claim 4 wherein the first maintenance dose is administered between 4 days and 4 weeks after the loading dose.

The method of claim 5 wherein the first maintenance dose is administered between 6 days and 3 weeks after the loading dose.

The method of claim 6 wherein the first maintenance dose s administered between 6 days and 2 weeks after the loading dose.

The method of any one of claims 1 through 7 comprising administering more than one maintenance dose and wherein individual maintenance doses are administered from 4 days to 4 weeks apart.

9. The method of claim 8 wherein individual maintenance doses are administered from 6 days to three weeks apart.

10. The method of claim 9 wherein individual maintenance doses are administered from 6 days to 2 weeks apart.

11. The method of any one of the preceding claims wherein the antibody molecule is selected from a full antibody, a Fab, a scFv, or a VHH antibody.

12. The method of any one of the preceding claims wherein the binding affinity of the antibody molecule for hFcRn is substantially pH independent.

13. The method of any one of claims 1 through 11 wherein the binding affinity to hFcRn of the antibody molecule in nM at pH 6.0 is less than 20 % of its binding affinity to hFcRn at pH 7.4.

14. The method of any one of the preceding claims wherein the loading dose is from 50 to 300 nmole/kg, preferably from 100 to 250 nmole/kg, more preferably from 150 to 200 nmole/kg.

15. The method of any one of the preceding claims wherein a maintenance dose is from 30 to 250 nmole/kg, preferably from 50 to 200 nmole/kg, more preferably from 100 to 150 nmole/kg.

16. The method of any one of the preceding claims wherein the subject has, prior to

commencement of the treatment a reference IgG serum level, and 4 - 10 days after commencement of the treatment has a treatment IgG serum level less than 80% of the reference serum level.

17. The method of claim 16 wherein the treatment IgG serum level is less than 50% of the reference serum level.

18. The method of claim 16 or 17 wherein the treatment IgG serum level is reached within 7 days after commencement of the treatment and is maintained during the remainder of the treatment.

19. The method of claim 16 wherein the treatment IgG serum level remains less than 80% of the reference serum level for at least 10 days after administration if the last maintenance dose.

20. The method of any one of the preceding claims wherein at least one dose is

administered subcutaneously.

21. The method of any one of claims 8 through 20 wherein at least the maintenance doses are administered subcutaneously.

22. The method of any one of the preceding claims wherein all doses are administered subcutaneously.

23. The method of any one of claims 20 through 22 wherein the subcutaneous

administration is carried out in conjunction with administration of a subcutaneous penetration enhancer.

24. The method of claim 23 wherein the subcutaneous penetration enhancer comprises a hyaluronidase enzyme.

25. The method of any one of the preceding claims wherein the subject has a BMI > 25, and at least the loading dose is determined based on the subject's corrected weight.

26. The method of claim 25 wherein at least one maintenance dose is determined based on the subject's corrected weight.

27. The method of claim 25 or 26 wherein all doses are determined based on the subject's corrected weight.