EP3314267A1

EP3314267A1 - Method of detecting or monitoring an igg4-related disease by detecting igg4 kappa/lambda hybrid antibodies

Info

Publication number: EP3314267A1
Application number: EP16736564.2A
Authority: EP
Inventors: Gregg Wallis; Stephen Harding
Original assignee: Binding Site Group Ltd
Current assignee: Binding Site Group Ltd
Priority date: 2015-06-29
Filing date: 2016-06-28
Publication date: 2018-05-02
Also published as: WO2017001841A1; JP7020924B2; GB201511364D0; JP2018519525A

Abstract

Methods and assay kits for detecting or monitoring an Ig G4-related disease are provided, using antibodies or binding agents such as antibodies specific for immunoglobulins, the immunoglobulins comprising an Ig G class and a heavy chain subclass, further characterised by being bound to a kappa or lambda light chain class or multiple light chain classes. The total amount of such an immunoglobulin may be detected, or the ratio between immunoglobulins of the same Ig G class and heavy chain subclass but bound to either a kappa or lambda light chain class or both a kappa and a lambda light chain class may be detected.

Description

METHOD OF DETECTING OR MONITORING AN IGG4-RELATED DISEASE BY DETECTING

IGG4 KAPPA/LAMBDA HYBRID ANTIBODIES

The invention relates to assay kits and methods for detecting or monitoring IgG4-related disease, using antibodies or binding agents such as antibodies specific for immunoglobulins, the immunoglobulins comprising an IgG class and a heavy chain subclass, further characterised by being bound to a kappa or lambda light chain class or multiple light chain classes. Compositions and methods of using the assay, for example in the detection of an IgG4-related disease, are also provided.

Antibody molecules (also known as immunoglobulins) have a twofold symmetry and are composed of two heavy chains and two light chains, each containing variable and constant domains. The variable domains of the heavy and light chains combine to form an antigen- binding site, so that both chains contribute to the antigen-binding specificity of the antibody molecule. The basic tetrameric structure of antibodies comprises two heavy chains covalently linked by a disulphide bond. Each heavy chain is in turn attached to a light chain, again via disulphide bond. This produces a substantially "Y"-shaped molecule.

Heavy chains are the larger of the two types of chain found in antibodies, with typical molecular weight of 50,000-77,000 D, compared with the smaller light chain (25,000 D).

There are five main classes of heavy chain which are γ, α, μ, δ and ε which are the constituents heavy chains for: IgG, IgA, IgM, IgD and IgE respectively. IgG is the major immunoglobulin of normal human serum, accounting for 70-75% of the total immunoglobulin pool. This is the major antibody of secondary immune responses. It forms a single tetramer of two heavy chains plus two light chains.

IgM accounts for approximately 10% of the immunoglobulin pool. The molecules, together with J-chains, form a pentamer of five of the basic 4-chain structures. The individual heavy chains have a molecular weight of approximately 65,000 and the whole molecule has a molecular weight of about 970,000. IgM is largely confined to the intravascular pool and is the dominant antibody produced in the primary immune response.

IgA represents 15-20% of human serum immunoglobulin pool. More than 80% of IgA occurs as a monomer. However, some of the IgA (secretory IgA) exists as a dimeric form. IgD is present on the plasma membranes of mature B-lymphocytes and accounts for less than 1% of the total immunoglobulin in the plasma.

IgE is the least common serum immunoglobulin isotype, and IgE is most commonly found bound to Fc receptors on surface of mast cells and basophils.

In addition to the five main classes, there are four subclasses for IgG (IgGl, IgG2, IgG3 and IgG4). Additionally there are two subclasses for IgA (IgAl and IgA2). In healthy adult serum, IgGl, IgG2, IgG3 and IgG4 account for 60-70%, 14-20%, 4-8% and 2-6% of the total IgG pool respectively. These percentages may be altered in certain disease types.

IgG4 is therefore the least abundant IgG subclass in human plasma but has a 100-fold normal concentration range of 0.01-2. lg/1, and a mean range of 0.35-0.51 mg/ml. IgG4 constant regions show 95% amino acid sequence similarity with other IgG subclasses. IgG4 has a half- life of approximately 21 days and is recycled by a neonatal fragment crystallisable receptor (FcRn) mediated process.

There are two types of antibody light chain: Lambda (λ) and Kappa (κ). There are approximately twice as many κ as λ molecules produced in humans, but this is quite different in some mammals. Each chain contains approximately 220 amino acids in a single polypeptide chain that is folded into one constant and one variable domain. Plasma cells produce one of the five heavy chain types together with either κ or λ molecules. There is normally approximately 40% excess free light chain production over heavy chain synthesis. Where the light chain molecules are not bound to heavy chain molecules, they are known as "free light chain molecules". The κ light chains are usually found as monomers. The λ light chains tend to form dimers.

The classic antibody paradigm is that a single mature plasma cell produces one type of immunoglobulin heavy chain γ, α, μ, δ and ε in humans) and one type of immunoglobulin light chain (κ or λ). These are combined within the cell to produce a tetrameric molecule composed of two identical heavy chains and two identical light chains. IgG4 molecules are dynamic and can exchange half molecules to become hybrid bi-specific (monovalent) antibodies. This process may also be known as "Fab arm exchange" and was first characterised in a study which focused on the anti-inflammatory properties of IgG4 in potential immunotherapy (van der Neut Kolfschoten et al 2007).

IgG4 does not activate the classical and alternative complement pathways and has only a weak affinity for macrophage and phagocyte Fc receptors. Fab arm exchange effectively renders the antibody monovalent and thus, for the proportion of IgG4 that has undergone Fab arm exchange, limits the ability of IgG4 to cross link antigens. IgG4 molecules that have undergone Fab arm exchange (hybrid IgG4 molecules) may have two different kappa light chains one bound to each heavy chain, two different lambda light chains one bound to each heavy chain or one kappa light chain and one lambda light chain bound to their respective heavy chains. The latter is referred to as the IgG4K/ mixed hybrid molecule. Therefore, upregulation of IgG4 could potentially suppress an immune response. IgG4 production is normally associated with prolonged exposure to antigens and the increase of T helper 2 (Th2) cytokines which mediate allergic responses and IgE production. Unlike IgE, IgG4 does not exhibit high affinity binding to mast cells and basophils. IgG4 has been reported to interact with antibodies of the IgG and IgE classes through their Fc domains; analogous to rheumatoid factor interactions.

Fab arm exchange involves the swapping of a heavy-light chain pair (half antibody) between different IgG4 molecules, possibly reforming new disulphides in the hinge region but without disruption of the heavy-light chain disulphide bond. Hence, an asymmetrical or hybrid immunoglobulin can be produced as described above (also see Figure 1). A serine residue at position 228 in the core hinge of IgG4 that allows the formation of intra-chain rather than inter-chain disulphides, and an arginine at position 409 within the CH₃ domain have been implicated in the control of this mechanism. Alternative amino acids at both of these positions, as found in other IgG subclasses, and in low abundance in IgG4 isoallotype molecules, abolishes or reduces the exchange process.

With exception of IgG4, all antibody subclasses have unique Gm allotypes. IgG4 has isoallotypes at position 309, whereby leucine is changed to a valine (L309/V309), and at position 409, whereby arginine is changed to a lysine (R409/K409). The arginine residue R409, in the R409 IgG4 isoallotype, enables the IgG4 Fab arm exchange, whereas the lysine residue K409, in the K409IgG4 isoallotype results in it resembling IgGl , IgG2 and IgG3 which do not undergo Fab arm exchange.

Half molecule exchange may play a physiological role as these naturally produced bi- specific molecules cannot cross link antigens or elicit lymphoid responses, which may dampen the inflammatory response. The Fab arm exchange effectively renders the IgG4 antibody monovalent and thus limiting the ability to cross link antigens in vitro. The κ light chains usage bias in IgG4 is much higher than for the IgG class per se, and is reported to be approximately 5: 1 respectively (Young et al. 2014).

The IgG heavy chain specific class, IgG4, can therefore exist as three different forms determined by the light chains bound to the heavy chains. The first form exists as an IgG4 molecule bound to two κ light chains, one attached to each of the two heavy chains (IgG4K). The second form exists as an IgG4 molecule bound to two λ light chains, one attached to each of the two heavy chains (IgG ). Finally, the other isoform exists as an IgG4 molecule bound to two light chains, one being a κ light chain and the other being a λ light chain (IgGK/λ), as a so-called IgGK/λ mixed hybrid molecule.

The exact proportion of IgGK/λ mixed hybrid molecules to IgGK and IgG molecules that have not undergone fab arm exchange varies widely between healthy subjects, but is expected to represent a significant proportion of total IgG4. IgG4K/ mixed hybrid molecules are reported to be represent between 20-30% of the total IgG4 immunoglobulins; similar values are reported in both Young et al. 2014 (21-33%) and Silva et al. 2015 (28%).

Increased IgG4 antibody production is associated with a number of proliferative, autoimmune and chronic inflammatory diseases which feature infiltration of target organs by IgG4- expressing cells. These diseases are known collectively as IgG4-related diseases. Chronic exposure to an antigen also results in the increase of serum IgG4. It has been proposed, in Karagiannis et al. 2013, that the increase in IgG4 levels diverts the humoral immune response away from an IgE-dominated response. However, it is unknown if IgG4 is pathogenic or mediates the host's response in IgG4-related diseases. The epidemiology of IgG4-related diseases is relatively unknown, although the incidence in the Japanese population is reported to be 0.28- 1.08 per 100,000 individuals. IgG4-related diseases comprise a collection of disorders which share specific pathological, serological and clinical features. The common features may comprise tumour- like swelling of organs, a lymphoplasmacytic infiltrate enriched in IgG4-positive plasma cells and CD4+T lymphocytes, obliterative phlebitis and a degree of fibrosis which has a characteristic "storiform" pattern typified by a cartwheel appearance of arranged fibroblasts and inflammatory cells. Modest tissue eosinophilia is also common. Subjects often present with the development of a mass in an organ or diffuse enlargement of an organ. Multiple organs are affected in 60 to 90% of subjects.

Elevated serum concentrations of IgG and IgG4 are found in 60 to 70% of subjects with IgG4-related disease. The upper limit of normal for IgG4 for an adult human is approximately 2. lg/1. The increased severity of IgG4-related disease correlates with an increased amount of serum IgG4. The sensitivity of IgG4 measurement in IgG4-related diseases ranges from approximately 50-100%, although the sensitivity of IgG4 measurement in many of these diseases has not been determined. Typically, subjects are responsive to glucocorticoids, therefore glucocorticoid responsiveness has been considered one diagnostic criteria for these disorders. IgG4-levels decrease in response to glucocorticoid treatment.

The common organs affected in IgG4-related disease in order of frequency are: pancreas, salivary gland, kidney, lacrimal gland, aorta, bile duct, lung, paravertebra, orbita, retroperitoneum and artery (Inoue et al. 2015).

Usually, IgG4-related disorders occur more commonly in middle-aged and older men. However, IgG4-related disorders of organs of the head and neck are seen equally in both males and females. The major disorders contained within IgG4-related disease are Type 1 IgG4-related autoimmune pancreatitis, salivary gland disease (Mikilukz's disease), orbital disease and retroperitoneal fibrosis. IgG4-related disease also appears to increase the risk of malignancy, particularly lymphomas and gastric cancers.

The pathogenesis of IgG4-related disorders remains largely unknown. Elevations in serum and tissue IgG4 are not specific to IgG4-related diseases; this may also occur in other disorders such as multicentric Castleman's disease, allergic disorders and Churg-Strauss syndrome. An allergic response will often include increased serum levels of IgE, and increased levels of Th2 cytokines, IL-10 and TGF-beta in affected tissues. Additionally, subjects with IgG4-related disease have an increased prevalence of allergic rhinitis and bronchial asthma. Up to 40% of subjects with an IgG4-related disease have a peripheral eosinophilia.

IgG4-related disease is currently diagnosed upon a radiological finding or histopathological examination of a tissue sample. The tissue sample is extracted using an invasive core needle biopsy. The tissue sample is observed for signs of lymphoplasmacytic tissue infiltration by IgG4-positive plasma cells and lymphocytes, accompanied by fibrosis which has storiform features and obliterative phlebitis. Tissue IgG4-positive cell counts and the ratio of IgG4- negative to IgG4-positive cells are also considered. However, cell counts vary widely between tissues and subjects. PET, CT and MRI investigation is also recommended to identify areas of fibrotic tissue. Serum total IgG4 levels may be measured but are not regarded for diagnosis alone, as they are considered neither sufficiently sensitive or specific for IgG4-related disease.

The initial observations regarding IgG4-related disease were made in patients with autoimmune pancreatitis, which can be mistaken for pancreatic cancer. Elevated IgG4 may be seen in pancreatic cancer, although to a lesser degree than in autoimmune pancreatitis. Steroid treatment for type 1 autoimmune pancreatitis results in a decrease in IgG4, therefore differentiating the disease from pancreatic cancer. Therefore, measuring IgG4 may be beneficial in ensuring pancreatic cancer is not autoimmune pancreatitis and vice versa. An increase in IgG4 may also be observed in IgG4-related sclerosing cholangitis but not primary sclerosing cholangitis and/or cholangiocarcinoma, which may be beneficial in diagnosis as biopsies are seldom deep enough to define the histological features of IgG4-related sclerosing cholangitis.

IgG4 has also been shown to be present in IL-10-driven Th2 immune response in some inflammatory diseases. Th2-mediated inflammation is also usually indicative of tumour growth. IgG4 has been shown to infiltrate tumour cells and accumulate around tumours in melanoma (Karagiannis et al 2014). IgG4 has also been shown to inhibit the antitumor function of IgGl, possibly due to IgG4 binding to tumour antigens in the absence of Fc receptor binding and effectively dampening the anti-tumour immune function of IgGl by blocking further non IgG4 antibody binding to the tumour antigens. It has been suggested that IgG4 may also inhibit the anti-tumour function of IgGl via antagonism of Fc receptors (Karagiannis JCI 2013). This study is focused on the ratio of IgG4:total IgG as indicative of a switch to production of the IgG4 class in melanoma. The study also highlights the need to avoid stimulating IgG4 immunoregulation in designing future therapeutics. IgG4 is therefore regarded as having immunosuppressive effects and low cytotoxic potential.

The Applicants have previously developed a sensitive assay that can detect the free κ light chains and separately, the free λ light chains. This method uses a polyclonal antibody directed towards either the free κ or the free λ light chains. The possibility of raising such antibodies was also discussed as one of a number of different possible antigens, in W097/17372. This document discloses methods of tolerising an animal to allow it to produce desired antibodies that are more specific than prior art techniques could produce. The free light chain assay (Freelite™) uses the antibodies to bind to free λ or free κ light chains. The concentration of the free light chains is determined by nephelometry or turbidimetry. This form of assay is highly sensitive. Indication of clonal production of light chains, such as in light chain multiple myeloma, can be provided by identifying an abnormal ratio of free kappa to free lambda light chains in the serum.

Measuring antibody ratios assists in the diagnosis and monitoring of diseases. Furthermore, if the disease is treated, the technique allows the progression of the disease to be monitored. If the disease is successfully being treated, then the concentrations of the free light chains, which have a relatively short life span within the blood will change and move more towards the normal concentrations observed for normal sera.

The Applicants have also produced antibodies and assays (known as Hevylite™) that would be able to distinguish between, for example, IgG λ and IgG κ, as disclosed in EP 1842071 and EP2306202. They produced antibodies which are specific for intact immunoglobulins and which had specificity for both a heavy chain class and a light chain type. They have also produced assays that allow the rapid quantitative measurement of, e.g. IgG and IgGK ratios to allow the rapid identification and/or follow the progression of diseases associated with production of a specific heavy chain class, or even heavy chain subclass, in conjunction with a bound λ or κ chain, as disclosed in WO2011021041. By determining the composition of immunoglobulins using an antibody specific to a heavy chain class at the same time as a light chain type or by using a first antibody against a heavy chain class and a second antibody to determine the light chain type bound to the heavy chain, the inventors produced a sensitive assay for specific immunoglobulin related diseases. The assays developed allow more sensitive progression of the diseases than, for example, by SPE. There is currently no serological assay to differentiate between mixed hybrid and single type light chain IgG4 molecules in serum. Heavy chain-light chain specific assays have been produced by the Applicant as Hevylite™. Hevylite™ uses a capture antibody which is heavy chain-light chain specific and is able to differentiate between immunoglobulins with a particular heavy chain class bound to either λ or κ light chains. However, Hevylite™ is unable to differentiate on its own between IgG4 mixed hybrid molecules from IgG4 molecules with only λ or κ light chains as it requires a highly specific anti-IgG4 capture antibody. If Hevylite™ was used to measure the amount of IgG4 molecules bound to κ light chains in a sample, it would give a positive result for every IgG4 molecule that is bound to a K chain. This would include the single type κ light chain IgG4 and the mixed hybrid IgG4, which also contains a λ light chain and is therefore an entirely different class of molecule.

Therefore, Hevylite™ is not able to discriminate on its own between the hybrid IgG4 and the single type light chain IgG4, leading to a misrepresentative measurement of the increase of the single type light chain IgG4. Hevylite™ is also concerned with the detection of expansion of monoclonal heavy-chain-light-chain specific antibodies in malignant plasma cell diseases. IgG4 only represents a very small proportion of total IgG (-4%) and it is therefore highly rare that a malignant plasma cell disease will result in an expansion of IgG4 molecules. Therefore, Hevylite™ when used on its own would not be suitable for use in diagnosing or monitoring an IgG4-related disease.

The restriction of using Hevylite™ to detect IgG4i<A. hybrids has been reported by the Applicants (Young et al 2014). The majority of previous work on IgG4 half molecule exchange has been made possible using purified monoclonal or antigen- specific IgG4 molecules spiked into human or animal plasma. The present Applicants realised that it was possible to use a different analytical-extraction procedure in order to capture, purify and identify IgG4 hybrids (IgG4K/ ) and IgG4 single type light chain (IgG4K and IgG4 ). The extraction process produced very pure samples of polyclonal IgG4 without any other plasma proteins present. Further fractionation produced pure samples of IgG4K, IgG4 and IgG4K/ . The Applicants then used a combination of ELISA, immunoprecipitation and Hevylite™ to estimate the percentage of ¾Θ4κ/λ in these purified samples. Table 1 shows the results of using a total IgG assay or Hevylite™ with either an anti-κ or an anti-λ capture antibody to measure each fraction. There is a discrepancy between the amounts of IgG4 captured with each of the Hevylite™ assays when compared to the total amount of IgG in the sample. It is clear that Hevylite™ is unable to distinguish an IgG4i<A. mixed hybrid molecule and therefore both Hevylite™ assays yielded a positive result upon detecting the hybrid, causing those molecules to be counted twice. Both the Hevylite™ assay with an anti-κ capture antibody and the Hevylite™ assay with an anti-λ capture antibody showed detection of the IgG4K/ mixed hybrid and neither equated to the total amount, thus demonstrating an impact on the assay calibration. Furthermore, the Applicants comment that the effect of IgG4i<A. hybrid molecules on assay calibration is unknown.

The Applicant has realised that there are advantages in having an assay which is able to detect immunoglobulins in a sample from a subject which does not require initial purification steps. This reduces the amount of time taken to achieve a result and is less labour intensive. It also removes the need to possess specialist equipment and techniques such as affinity columns and HPLC, and is therefore ideal for use in the clinic.

The Applicants had shown that Hevylite™ IgGK and IgG antibodies react with not only IgGK and IgG respectively, but also that both IgGK and IgG Hevylite™ antibodies react with IgG4 IgG4K/ mixed hybrid molecules. This indicated that the IgG hevylite antibodies were not suitable to detect IgG4K/ mixed hybrid molecules. The percentage of IgG4K/ hybrids was only characterised upon SDS-PAGE analysis after these discrepancies were discovered with the use of Hevylite™. These results demonstrated that a considerable portion (approximately 30%) of IgG4K/ hybrids are present in normal healthy human serum, as a result of Fab arm exchange in vivo.

Although the presence and measurement of IgG4K/ mixed hybrids is known in the field, there has been no evidence that their measurement is useful. It has been reported that individuals with IgG4-related disease do not have a substantially increased frequency of the K409 variant of IgG4 which compromises Fab arm exchange (Ahmad et al. 2014). This suggests that mixed hybrid molecule dynamics are unaffected in IgG4-related disease, and would not, therefore, be selected in the diagnosis or monitoring of an IgG4-related disease. Another study has reported that therapeutic antibodies engage in Fab arm exchange with endogenous human IgG4 in vivo, but does not comment on the clinical relevance of this (Labrijn et al. 2009), nor does it refer specifically to the ¾Θ4κ/λ hybrids. Furthermore, the development of an assay which is able to determine the levels of ¾Θ4κ/λ mixed hybrids is unlikely to be able to account for the levels of IgG4X/ or IgG4K/K hybrids, which may be desirable. Therefore, there has been no motivation to produce a simpler, efficient assay to determine the exact levels of the IgG4i<A. mixed hybrids.

An MSD (Meso Scale Discovery) immunoassay has been developed which uses highly specific anti-IL-6 IgG4 S228P mutated antibodies to detect only those IgG4 molecules possessing the S228P mutation which inhibits Fab arm exchange, which are raised against IL-6 (Silva et al. 2015). This relies on antigen specificity and cannot be used to measure the total number of IgG4K/ hybrids in a sample. Furthermore, there is no suggestion by Silva et al. 2015 that measuring the number of IgG4i<A. hybrids, or even estimating Fab arm exchange is clinically relevant.

There are a number of existing patent applications which cover isolation or amino acid modifications of IgG4 antibodies. US2004092719 discloses a method for isolating IgG4 antibodies which lack inter-heavy chain disulfide bonds in antibody preparation, whilst EP1810979 discloses stabilised human IgG4 antibodies for immunotherapy. US2013323236 discloses an IgG4 antibody which contains specific heavy chain amino acid substitutions to confer preferred properties in immunotherapy. WO2013124451, WO2013124450 and EP2626372 disclose bispecific heavy chain modifications for immunotherapy. US2011293607 also discloses IgG4 antibodies having amino acid modifications, but in the constant region. None of the prior art concerns the isolation or quantitative measurement of the IgG4K/ hybrids.

The Applicants hypothesised that the relationship between IgG4 levels and total IgG in tumour survival in melanoma such as that reported by Karagiannis et al. 2014, or in the presence or severity of IgG4-related disease, may be affected by the proportion of IgG4KA, hybrid antibodies in the total IgG4 antibody pool. The Applicants identified that the ratio of IgG4K/ hybrids to total IgG4 levels are perturbed in certain IgG4-related disorders, such as the skin blistering disorder pemphigus or autoimmune pancreatitis. As IgG4 antibodies are known to antagonise the complement fixing action of IgGl antibodies, thereby dampening down inflammation, the Applicants speculate that the proportion of ¾Θ4κ/λ hybrids may alter this effect. IgG4 antibodies also appear to interfere with antibody mediated cell killing via Fc receptor binding (Karagiannis et al. 2014).

One approach to purify and analyse IgG4 mixed hybrid antibodies has been recently reported (Yang et al. 2015). Purified human IgG was mixed with IgG4 monoclonal antibodies and the amount of half molecule exchange that occurred was detected by the separation of IgG4 mixed hybrid antibodies by mixed mode chromotography. IgG4 mixed hybrid antibodies which underwent half molecule exchange with the IgG4 monoclonal antibodies were then quantitatively measured using UV absorption or protein fluorescence. This method involves several purification steps of both the IgG sample from a human subject and the production of monoclonal IgG4 antibodies. This is a lengthy, labour intensive approach to measuring IgG4 mixed hybrids and would not be able to be carried out without the use of specialist equipment, such as a mass spectrometer. Furthermore, Yang teaches that the use of immunoassays to measure IgG4 mixed hybrids is currently limited to immobilising an antigen for one of the hybrid Fab arms on the solid phase and using an anti-idiotype antibody against the second Fab arm, or an anti-kappa or anti-lambda antibody for the other arm. Yang highlights that this may lead to multiple cross-reactivity between the administered antibody and endogenous IgG4, and concludes that the use of immunoassays to measure IgG4 mixed hybrid antibodies are not accurate.

The Applicants realised it was possible to develop a highly specific quantitative serological assay to detect the presence of IgG4i<A. hybrids from an unpurified sample directly from a subject. All the previous attempts in the prior art to analyse the percentage of IgG4i<A. have used lengthy protocols involving purified fractions of IgG4 and subsequent SDS-PAGE or immunoprecipitation analysis. Using their knowledge of heavy light chain specific antibodies, the Applicants have developed an immunoassay which identifies the ratio of IgG4K/ hybrids to total IgG4 in order to diagnose or monitor an IgG4-related disease such as autoimmune pancreatitis or pemphigus.

A first aspect of the invention provides a method of detecting or monitoring an IgG4-related disease comprising detecting in a sample the ratio between the relative amounts of antibodies having: (i) an IgG heavy chain class; and

(ii) an IgG4 heavy chain class bound to both a κ light chain and a λ light chain; and optionally,

(iii) immunoglobulins having the same heavy chain class but bound to either κ light chains only or λ light chains only;

the method comprising the quantitative detection of said immunoglobulins comprising:

(i) binding said immunoglobulins to a binding agent specific for the immunoglobulin; the binding agent immobilised on a substrate or conjugated with a reporter molecule,

(ii) using a labelled detecting agent which detects immunoglobulin bound to said binding agent or conjugated with a reporter molecule.

The method preferably quantitatively measures the amounts of the immunoglobulins in the sample.

Immunoglobulins bound to binding agent, for example immobilised binding agent, may be washed to remove unbound immunoglobulins.

Step (iii), as described above, may be used to determine the total amount of an IgG4 heavy chain class which is bound to either kappa light chains or lambda light chains. The total amount of IgG4 is determined using step (i). The total amount of an IgG4 heavy chain class which is bound to both a kappa and a lambda light chain is also determined using step (ii) and is subtracted from the total amount of IgG4 from step (i). The remaining fraction indicates the total amount of an IgG4 heavy chain class which is bound to either kappa light chains or lambda light chains.

The inventors have found that it is possible to use single antibodies in combination to discriminate between different heavy chain class/single light chain type antibodies, and heavy chain class/hybrid light chain type antibodies. Hence, the method of the invention may be determined using:

(i) at least one binding agent specific for the IgG heavy chain class; and a) a binding agent specific for the IgG heavy chain class bound to κ light chains in combination with a binding agent specific for λ light chains; and/or b) a binding agent specific for the IgG heavy chain class bound to λ light chains in combination with a binding agent specific for κ light chains; and/or c) a binding agent specific for the IgG heavy chain class bound to κ light chains in combination with a binding agent specific for the IgG heavy chain class bound to λ light chains.

An IgG4 - specific binding agent may be bound to the IgG4, which is also bound to an IgG4K- specific binding agent. The amount of the binding may be configured to a calibration curve obtained for predetermined concentrations of IgG4i<A. mixed hybrid.

Figure 2 shows two worked examples of the combination of an anti-IgG4K capture antibody and an anti-λ detecting antibody, or an anti-IgG4 capture antibody and an anti-κ detecting antibody to detect the IgG4i<A. mixed hybrid.

The binding agent specific for the immunoglobulin to be detected is preferably an antibody or fragment thereof, or an aptamer. The fragments of antibody used in all aspects of the invention may be Fab or F(ab')2 fragments. Aptamers are short single stranded DNA or RNA molecules with high affinity and specificity, and may be referred to as "nucleic acid antibodies".

The heavy chain class to be detected may be selected from IgGl, IgG2, IgG3 and IgG4, most preferably IgG4. A combination of all classes may also be detected as total IgG.

The sample may optionally be enriched for the IgG heavy chain subclass before use in the detection method by affinity purification or adsorption.

The method of the invention may also be used using one or more of the following methods wherein the binding of the binding agents to the antibodies in the sample is determined by using an ELISA (Enzyme Linked Immunosorbent Assay), flow cytometry or fluorescently labelled beads such as Luminex™ beads. These assay methods may be singleplex (measures one analyte of interest) or multiplex (measures multiple analytes of interest). Alternatively, a microarray assay may be produced using the specific binding agents. The total amount of a particular IgG heavy chain class, or the total amount of IgG detected as described above is preferably measured as part of a multiplex ELISA assay, or as a singleplex assay to be carried out alongside the detection steps of either a), b) and/or c) as described above.

Preferably the ratio of IgG -K/ mixed hybrid to total IgG4, or IgG4-K/ mixed hybrid to total IgG, is determined immunologically, most preferably via ELISA. ELISA-type assays per se are well known in the art. They use specific binding agents such as antibodies to detect blood groups, cell surface markers, drugs and toxins. In the case of the current invention, this type of assay has been used for the method of the invention.

ELISA uses antibodies or other binding agents such as aptamers, or fragments of antibodies to detect specific antigens. One or more of the antibodies, aptamers or fragments of antibodies used in the assay may be labelled with an enzyme capable of converting a substrate into a detectable analyte. Such enzymes include horseradish peroxidase, alkaline phosphatase and other enzymes known in the art. Alternatively, other detectable tags or labels may be used. These include radioisotopes, a wide range of coloured and fluorescent labels known in the art, including fluorescein, Alexa fluor, Oregon Green, BODIPY, rhodamine red, Cascade Blue, Marina Blue, Pacific Blue, Cascade Yellow, gold; and conjugates such as biotin (available from, for example, Invitrogen Ltd, United Kingdom). Dye sols, metallic sols or coloured latex may also be used. One or more of these labels may be used in the ELISA assays according to the various inventions described herein, or alternatively in the other assays, labelled antibodies or kits described herein.

The construction of ELISA-type assays is itself well known in the art. For example, a "binding antibody" specific for the antigen is immobilised on a substrate. In this case, the antigen is an antibody comprising an IgG heavy chain, or an IgG heavy chain of a particular subclass, attached to either a λ light chain or a κ light chain, or both a λ light chain or a κ light chain. The "binding antibody" may be immobilised onto the substrate by methods which are well known in the art. Antigens in the sample are bound by the "binding antibody" which binds the antigen to the substrate via the "binding antibody". Unbound antibodies may be washed away.

In ELISA assays the presence of bound antibodies may be determined by using a labelled "detecting antibody" specific to a different part of the antigen of interest than the binding antibody.

Flow cytometry may be used to detect the binding of the antibodies of interest and measure the ratios. This technique is well known in the art for, e.g. cell sorting. However, it can also be used to detect labelled particles, such as beads, and to measure their size. Numerous text books describe flow cytometry, such as Practical Flow Cytometry, 3rd Ed. (1994), H. Shapiro, Alan R. Liss, New York, and Flow Cytometry, First Principles (2nd Ed.) 2001, A.L. Given, Wiley Liss.

One of the binding antibodies, such as the antibody specific for the heavy chain class, is bound to a bead, such as a polystyrene or latex bead. The beads are mixed with the sample and the second detecting antibody, such as antibody specific for λ light chains. The detecting antibody is preferably labelled with a detectable label, which binds the antibody to be detected in the sample. This results in a labelled bead when antibody to be assayed is present.

Labelled beads may then be detected via flow cytometry. Different labels, such as different fluorescent labels may be used for, for example, the anti-λ and anti-κ antibodies. This allows the amount of each type of antibody bound to be determined simultaneously and allows the rapid identification of the κ/λ hybrid:single type λ or κ ratio for a given heavy chain class when carried out in combination with the identification of the total amount of a given heavy chain class.

Alternatively, or additionally, different sized beads may be used for different antibodies, for example for different class specific antibodies. Flow cytometry can distinguish between different sized beads and hence can rapidly determine the amount of each heavy chain class in a sample. An alternative method uses the antibodies bound to, for example, fluorescently labelled beads such as commercially available Luminex™ beads. Different beads are used with different antibodies. Different beads are labelled with different fluorophore mixtures, thus allowing the single type or hybrid type ratio for a particular heavy chain class or subclass to be determined by the fluorescent wavelength. Luminex™ beads are available from Luminex™ Corporation, Austin, Texas, United States of America.

The immunoglobulin of interest may also be detected using a homogenous time resolved fluorescence (HTRF) assay platform. Homogenous assays require a simple mix and read procedure without the necessity for multiple processing steps such as separation or washing. HTRF technology combines fluorescence resonance energy transfer (FRET) with time resolved measurement (TR) of fluorescence. FRET relies on the energy transfer between two fluorophores which come into close contact upon the interaction of one protein with another, both of which have either a donor fluorophore attached, or an acceptor fluorophore attached. For example, the donor fluorophore (which has a long duration of fluorescence) could be attached to an anti-IgG4K antibody. The protein of interest from the sample, the ¾Θ4κ/λ mixed hybrid antibody, would then bind to the donor labelled anti-IgG4K antibody. An anti-λ antibody could then be added, which may have an acceptor fluorophore attached (which has a short duration of fluorescence), which would also bind to the ¾Θ4κ/λ mixed hybrid antibody. When these two labelled proteins, the anti-IgG4K antibody and the anti-λ antibody come into close proximity, such as that upon binding to the IgG4i<A. mixed hybrid, the level of energy transfer between the donor and acceptor fluorophores can be detected as the emission of fluorescence. This information can be used to indicate the amount of a particular protein of interest in a sample.

The immunoglobulin of interest may also be detected using a lateral flow assay platform, preferably a lateral flow sandwich assay platform. A lateral flow sandwich assay uses coloured or fluorescent particles which are used to label binding agents, typically an antibody. The labelled antibody is used to detect the antigen of interest from a sample which passes through a series of capillary beds. The labelled antibody is usually immobilised to the surface of one of the capillary beds and forms an antigen-antibody complex upon encounter of the antigen of interest in the sample. As the flow of the sample moves through the series of capillary beds, the matrix immobilising the antibody to the capillary bed dissolves and the antigen-antibody complex is free to migrate further where it becomes bound to a second specific binding agent, typically an antibody. A coloured/fluorescent band may be observed as the accumulation of the antigen- antibody particles as an indicator of the presence or amount of the antigen in the sample.

Preferably, the sample is obtained from tissue or fluid, such as whole blood, plasma or serum from the blood of an animal, such as a mammal, preferably a human and additionally rhesus monkeys (Macaca mulatta), cynomolgus macaques (Macaca fascicularis), rabbits

(Oryctolagus cuniculus), guinea pig (Cavia porcellus), rat (Rattus norvegicus), horse (Equus caballus), sheep (Ovis aries), cow (Bos taurus), donkey (Equus asinus), mouse (Mus musculus), goat (Capra hircus), and pig (Sus scrofa) (Labrijn et al, 2011). Additionally, other mammals such as mice may exhibit light chain mixed hybrid antibodies of another class, such as IgG3. Additionally, it may be possible to identify such proteins in tissue, preferably tumour tissue, or urine, saliva, cerebrospinal fluid or lymph fluid. The tissue or fluid may be taken from a specific area within the body, such as the local tissue or fluid surrounding a tumour, or fluid within a skin blister. Preferably the sample is assayed in vitro.

The presence of human antibodies may be determined using anti-human antibodies, e.g. from sheep, horse, goat, donkey, rabbit, chicken, mouse or rat.

Measurement of the heavy chain-light chain specific pair or hybrid pair is capable of being automated. Furthermore, the technique is more sensitive and allows the quantitative determination of the amount of different immunoglobulins. It can be used both to aid diagnosis of a disease and also to monitor the response of the disease to treatment, such as glucocorticoid treatment. It may be used to detect the presence of a tumour in an IgG4-related disease. It may also be used in prognosis of future disease progression and the risk of survival of a subject. These values may be calculated as a percentage of risk. The method may also be used to monitor the response of an allergy to treatment. It may also be used to indicate whether a specific antigen has been removed from the environment of a subject.

IgG4-related disease may be known as: IgG4-related systemic disease, IgG4- syndrome, IgG4-associated disease, IgG4-related sclerosing disease, IgG4-related systemic sclerosing disease, IgG4-related autoimmune disease, IgG4-positive multiorgan lymphoproliferative syndrome, hyper- IgG4 disease, systemic IgG4-related plasmacytic syndrome, , IgG4-related multifocal systemic fibrosis, multifocal fibrosclerosis, or multifocal idiopathic fibrosclerosis. Preferably, the IgG4-related disease is selected from: Type 1 (IgG4-related) autoimmune pancreatitis, IgG4-related sclerosing cholangitis, Mikulicz's disease (or IgG4-related dacryoadenitis and sialadenitis), sclerosing sialadenitis (or Kiittner's tumor, IgG4-related submandibular gland disease), IgG4-related orbital inflammation or IgG4-related orbital inflammatory pseudotumor, chronic sclerosing dacryoadenitis (or lacrimal gland enlargement, IgG4-related dacryoadenitis), retroperitoneal fibrosis (or Ormond's disease) and related disorders (IgG4-related retroperitoneal fibrosis, IgG4-related mesenteritis), chronic sclerosing aortitis and periaortitis (or IgG4-related aortitis or periaortitis), multifocal fibrosclerosis, fibrous variant of Hashimoto's thyroiditis, Riedel's thyroiditis (or IgG4-related thyroid disease), IgG4-related interstitial pneumonitis and pulmonary inflammatory pseudotumors (or IgG4-related lung disease), IgG4-related kidney disease (including tubulointerstitial nephritis and membranous glomerulonephritis secondary to IgG4-RD), IgG4-related hypophysitis, IgG4-related pachymeningitis, IgG4-related midline destructive disease, IgG4-related lymphadenopathy, IgG4-related orbital myositis, IgG4-related skin diseases (pemphigus, cutaneous pseudolymphoma), IgG4-related hepatopathy, hepatic inflammatory pseudotumor, lymphoplasmacytic gastritis associated with autoimmune pancreatitis, sclerosing mastitis, inflammatory pseudotumors of the breast, prostatitis, constrictive pericarditis, inflammatory aortic aneurysm, Sjogren's syndrome, lymphoplasmacytic sclerosing pancreatitis, familial multifocal fibrosclerosis, Kuttner's tumour, eosinophilic angiocentric fibrosis, Rosai-Dorfman disease, mediastinal fibrosis, periarteritis, idiopathic hypocomplementemic tubulointerstitial nephritis with extensive tubulointerstitial deposits and IgG4-related nasopharyngeal disease. The common organs affected in IgG4-related disease in order of frequency are: pancreas, salivary gland, kidney, lacrimal gland, aorta, bile duct, lung, paravertebra, orbita, retroperitoneum and artery. Additionally, cancers such as melanomas, or allergies, may exhibit elevated IgG4 and may be considered IgG4-related diseases.

The method or kit described herein may include co-assaying for IgG4-related disease inflammatory markers such as VEGF, IL-10, TGF-β or Th2 cytokines.

The presence of the IgG4 isoallotype K409 is usually tested for using a method such as PCR and post-PCR analysis. If the subject has a diploid K409 mutation, there will be no IgG4 mixed hybrid molecules to detect. If the subject has a haploid K409 mutation, the proportion of ¾Θ4κ/λ mixed hybrid molecules and IgG4 single type light chain molecules will be disrupted and reduced. Figure 6A shows clear outliers when detecting the ratio of IgG4 to IgG4K/ mixed hybrid molecules. Therefore, the identification of these outliers may be used to identify subjects with either a haploid or diploid K409 mutation.

The method may also be used to determine the amount or proportion of an IgG4 heavy chain class bound to both a κ light chain and a λ light chain antibody in synthetic antibody preparations, such as monoclonal antibodies produced as biologicals for immunotherapy, comprising detecting in a sample of the synthetic antibody preparation the ratio of two or more of the relative amounts of immunoglobulins having:

(i) an IgG4 heavy chain class bound to both a κ light chain and a λ light chain; and optionally,

(ii) immunoglobulins having the same heavy chain class but bound to either κ light chains only or λ light chains only; the method comprising the quantitative detection of said immunoglobulins by a binding assay comprising:

Polyclonal antibodies are preferably used, additionally monoclonal antibodies may also be used. Polyclonal antibodies allow an improved assay to be produced to monitor different immunoglobulins of, for example, the same class. Polyclonal antibodies allow a plurality of different antibodies to be raised against different epitopes for the specific IgG heavy chain- light chain combination. This allows for the slight variations between different immunoglobulins, but which nevertheless comprise the same IgG heavy chain-light chain combination. The polyclonal antibodies used in the various aspects of the invention may be capable of being produced by the method shown in WO 97/17372. This allows the production of highly specific polyclonal antibodies. The sample may be further characterised by measuring the amount of free λ or free κ light chains in the sample. The total amount of λ and κ free light chains may also be measured. A ratio may also be produced of λ to κ free light chains. This is preferably carried out using antibodies specific for free λ or free κ light chains, such as those sold under the trade mark Freelite™ and Combylite™ by The Binding Site Ltd, Birmingham, UK.

The resultant ratio value from the detection of total IgG or IgG heavy chain subclass specific versus IgG heavy chain subclass specific bound to both κ and λ light chains may be compared to a normal range or a reference value for each IgG4-related disease type. The normal range will typically have a median ratio of 0.328, a mean ratio of 0.358 with a standard deviation of 0.1272. The ratio value may also be used to further characterise an IgG4-related disease type according to a range specific for that disease.

A further aspect of the invention provides an assay kit, preferably an ELISA, comprising:

(i) a binding agent specific for the IgG4 heavy chain class bound to a specific light chain class; and

(ii) a binding agent specific for the opposite light chain class; and optionally

(iii) a binding agent specific for IgG4 and a binding agent specific for an IgG heavy chain class

A further aspect of the kit may comprise a predetermined amount of an IgG4i<A. mixed hybrid calibrator.

The opposite light chain class is considered to be kappa when the binding agent specific for the IgG4 heavy chain class is bound to a lambda light chain, and the opposite light chain is considered to be lambda when the binding agent specific for the IgG4 heavy chain class is bound to a kappa light chain.

A further aspect of the kit is for use in a method wherein the ratio of IgG4K/ mixed hybrid to total IgG4, or IgG4i<A. mixed hybrid to total IgG, is detected between two or more of the relative amounts of immunoglobulins having: (i) at least one binding agent specific for the IgG heavy chain class; and

a) a binding agent specific for the IgG heavy chain class bound to κ light chains in combination with a binding agent specific for λ light chains; and/or b) a binding agent specific for the IgG heavy chain class bound to λ light chains in combination with a binding agent specific for κ light chains; and/or c) a binding agent specific for the IgG heavy chain class bound to κ light chains in combination with a binding agent specific for the IgG heavy chain class bound to λ light chains.

A further aspect of the kit is for use in a method wherein the amount of IgG4-K/ mixed hybrid molecule is detected and compared to a calibrator, wherein the kit comprises :- a) a binding agent specific for κ light chains; and

b) a binding agent specific for λ light chains; and

c) a predetermined amount of IgG4X/K mixed hybrid calibrator agent.

The calibrator is used to calibrate the assay. Typically substantially pure hybrid IgGK/λ is diluted with a suitable buffer to form the calibrator agent.

The binding agent or detecting agent provided in the kit may be an antigen specific aptamer, an antibody or antigen- specific fragment thereof.

The antibodies, labels, etc. are preferably as described above.

Preferably the antibody specific for the IgG heavy chain class or the antibody specific for the IgG4 heavy chain class bound to a specific light chain class is immobilised to a substrate. The substrate may be a bead, but preferably is a microtitre plate well.

One or more of the antibodies preferably comprises a detectable label. Indirect labels, such as enzymes (e.g. alkaline phosphatase and horseradish peroxidase) can be used, as can radioactive labels such as ³⁵S. Each type of detecting antibody may be labelled with a different detectable label if used in combination, for example, in a multiplex assay. One or more controls (or calibrators), such as a known amount of a predetermined monoclonal protein, such as purified IgG, IgG4 or IgG4K/ , or a fragment thereof, may be provided in this and indeed other ELISA, flow cytometry, Luminex™ or other assays described herein. The calibrator is typically linked to an international standard concentration. It may for example, be at least 98% w/w pure mixed hybrid IgG4X/K substantially without the presence of IgG4 or IgG4K. The assay may report in standard units compared to these values. The fragments, when used will retain, e.g. antigenic determinants for detecting class and/or light chain type.

Flow cytometry kits with capture and Luminex beads are also provided. Preferably these kits comprise an antibody or fragments thereof, or additionally aptamers, specific for the IgG heavy chain class, and/or an IgG heavy chain subclass, and/or specific for the IgG4 heavy chain class bound to a specific light chain class, and each of the types of different antibodies, if used in combination, are attached to a different size of bead. Preferably the kit additionally comprises a labelled antibody for detecting the presence of antibodies from a sample bound, via the antibodies of the method, to the bead.

Competition assays, lateral flow immunochromatographic assays and HTRF platform assays are also provided.

The kits of the invention may additionally comprise antibodies specific for free λ or free κ light chains. Alternatively, anti-total free light chain (FLC) antibodies for measuring total FLC concentration may be provided or used to additionally measure total FLC, for example against a predetermined standard.

The kits may additionally comprise one or more of instructions for using the kit, substrate, a buffer, label, a preservative or a control.

The invention will now be described by way of example only, with reference to the following figures:

Figure 1 shows a schematic diagram of IgG4 half-molecule exchange. Table 1 shows immunoassay quantification of the purified polyclonal IgG4 Kappa, IgG4 Lambda and mixed light-chain IgG4 hybrids from normal human sera (taken from Young et al 2014). IgG4 κ/λ hybrids are positively detected by both Hevylite G assays.

Figure 3 shows a 5 point calibration curve using the current optimised assay parameters. The working range is 18.5-1500mg/L using a 1/5000 sample dilution. The calibrator is pure IgG4MM purified (quantified by Optilite G4 assay), the calibrator diluent is IgG4 depleted serum diluted 1/5000, the sample diluent is standard ELISA buffer T271, the conjugate is a- K-Perox diluted 1/3000. All steps are carried out at 30mins at RTP.

Figure 4 shows three samples corresponding to the lower middle and higher portions of the calibration curve which were tested. The percentage CV within sets of samples is shown. This demonstrates good inter-assay precision. Three regions of the calibration curve were targeted (25, 85, 580 mg/L IgG4MM) and there were 16 replicates per sample.

Figure 5A shows total IgG4 levels in 90 healthy controls (outliers removed) using EIA.

Figure 5B shows total IgG4 κ/λ mixed hybrid levels in 90 healthy controls (outliers removed) using EIA.

Figure 5A shows the ratio of total IgG4 κ/λ mixed hybrid levels to total IgG4 levels in 90 healthy controls (outliers removed) using EIA.

Figure 6A shows a linear plot of total IgG4 and IgG4 κ/λ mixed hybrid with the inclusion of identified outliers.

Figure 6B shows a linear plot of total IgG4 and IgG4 κ/λ mixed hybrid without the inclusion of identified outliers.

Figure 7 shows the ratios of total IgG4 to total IgG4 κ/λ mixed hybrid using EIA from clinical samples from patients with known IgG4-related diseases; Pemphigus/Pemphigoid and Autoimmune pancreatitis type l(AIP). Enzyme immunoassay (EIA)

IgG4 κ/λ hybrid levels were measured using antigen capture EIA. EIA plates (Maxisorp™ flat-bottom, clear, 96-well plates; Nunc, Roskilde, Denmark) were coated overnight with a- IgG4 at 2^g/ml in phosphate-buffered saline (PBS) in a moist box at 22 °C. Following removal of the coat solution the plates were blocked with Stabilcoat™, (SurModics, Eden Prairie, MN, USA) for 30 min at 22°C. After removal of the block, the plates were dried under vacuum, before sealing in a foil pouch containing desiccant sacs. The diluted patient sample was added across all the strips in duplicate. Purified IgG4 κ/λ hybrid in IgG4 depleted serum (1/5000) was used as a calibrator and triply diluted from 300ng/ml to 3.7ng/ml. Serum samples were diluted at 1/5000 sample dilution in PBS containing 0.1% Tween™ (PBS-T) sample diluent and incubated for 30 min at 22°C. After washing (Bio-Teck instruments Inc, VT, USA) with PBS containing 0.1% Tween™-T bound IgG4 κ/λ hybrids were detected by 1/3000 dilution of anti-kappa-HRP (BindingSite, UK) in 10% StabliZyme-HRP (SurModics) in saline. After a further 30 min incubation and washing, the plates were developed using tetramethylbenzidine (TMB, SurModics) solution, and the absorbance was measured at 450 nm (Bio-Teck EL800 micro plate reader). The calibration range obtained is 18.5 to 1500mg/l.

IgG4 levels were measured using antigen capture EIA. EIA plates) were coated overnight with monoclonal anti-IgG4 clone HP6025 (Sigma Aldrich) at 5.0μg/ml in phosphate-buffered saline (PBS) in a moist box at 22 °C. Following removal of the coat solution the plates were blocked with Stabilcoat™, for 30 min at 22°C. After removal of the block, the plates were dried under vacuum, before sealing in a foil pouch containing desiccant sacs. The diluted patient sample was added across all the strips in duplicate. A turbimetric IgG4 calibrator fluid was diluted in sample diluent to produce a range from 339.33ng/ml to 1.40ng/ml. Serum samples were diluted at 1/20000 sample dilution in PBS containing 0.1% Tween™ (PBS-T) sample diluent and incubated for 30 min at 22°C. After washing with PBS containing 0.1% Tween™-T bound IgG4 was detected by 1/8000 dilution of anti-IgG-HRP (Binding Site, UK) in 10% StabliZyme-HRP in saline. After a further 30 min incubation and washing, the plates were developed using TMB solution, and the absorbance was measured at 450 nm (Bio-Teck EL800 micro plate reader). The calibration range obtained is equivalent to 27.93 to 6787 mg/1 Results

The total IgG4 and IgG4 κ/λ hybrid levels in 90 healthy controls (outliers removed) were determined using EIA (Figures 5A to 5C). Total IgG4 concentrations were between 17.25mg/L and 1585.7mg/L with a median of 267.44mg/L, a mean of 350.217mg/L and a standard deviation of 290.587(Figure 5A). The 95% reference range as determined by a non- parametric percentile method (CLSI C28-A3) was 28.9 to 1603 mg/L, which is within the expected published range for serum IgG4. The IgG4 κ/λ hybrid levels were between 16.309mg/L and 494.3mg/L with a median of 84.085mg/L , a mean of 108.284mg/L and a standard deviation of 82.7324 (Figure 5B). The 95% reference range as determined by a non- parametric percentile method (CLSI C28-A3) was 20.3 to 460 mg/L. The IgG4 κ/λ hybrid to total IgG4 ratio had a range between 0.163 and 0.9450 with a median of 0.328, a mean of 0.353 and a standard deviation of 0.127525 (Figure 5C). A linear plot of total IgG4 and IgG4 κ/λ hybrid with (Figure 6A) or without (Figure 6B) the inclusion of identified outliers gave an R² value of 0.95 and a slope between 0.27 and 0.28. The slope indicates that 27-28 % of total polyclonal IgG4 is composed of IgG4 κ/λ hybrid molecules. This agrees well with published values obtained by purification, fractionation and quantification of polyclonal IgG4 (Young et al, 2014). This analysis identified 3 samples with no IgG4 κ/λ hybrids. These are most likely from people with the K409 isoalloytypic variant of IgG4 that do not undergo Fab- arm exchange (Brusco et al, 1998).

The Total IgG4 and IgG4 κ/λ hybrid EIAs were also used with clinical samples from known IgG4-related diseases; Pemphigus/Pemphigoid and Autoimmune pancreatitis type l(AIP) (Figure 7). In 6 patient samples with Pemphigus/Pemphigoid total IgG4 was found to be between 141.3mg/L and 429.3mg/L with a median of 234.2mg/L, a mean of 257.9mg/L and a standard deviation of 118.7. The IgG4 κ/λ hybrid levels were found to be between 28.35mg/L and 111.2mg/L with a median of 57.33mg/L , a mean of 66.18mg/L and a standard deviation of 30.89. The IgG4 κ/λ hybrid to total IgG4 ratio had a range between 0.2000 and 0.3400 with a median of 0.2500, a mean of 0.2580 and a standard deviation of 0.05119. The ratio of IgG4 κ/λ hybrid to total IgG4 was significantly lower in the Pemphigus/Pemphigoid samples compared to healthy normal (P=0.0003, Mann-Whitney U test). Nine AIP samples were analysed using the IgG4 and IgG4 κ/λ hybrid EIA. The total IgG4 was found to be between 16138mg/L and 59478mg/L with a median of 26405mg/L, a mean of 35421mg/L and a standard deviation of 117514. The IgG4 κ/λ hybrid levels were found to be between 4100mg/L and 18711mg/L with a median of 7929mg/L, a mean of 9996mg/L and a standard deviation of 5359. The IgG4 κ/λ hybrid to total IgG4 ratio had a range between 0.2400 and 0.3100 with a median of 0.2800, a mean of 0.2778 and a standard deviation of 0.02635. The ratio of IgG4 κ/λ hybrid to total IgG4 was slightly lower in the AIP samples compared to healthy normals (P=0.0347, Mann-Whitney U test).

References

Young E, Lock E, Ward DG, Cook A, Harding S, Wallis GL. Estimation of polyclonal IgG4 hybrids in normal human serum. Immunology. 2014. Jul;142(3):406-13. doi:

10.1111/imm.12265. PubMed PMID: 24512211.

Karagiannis P, Gilbert AE, Josephs DH, Ali N, Dodev T, Saul L, Correa I,Roberts L, Beddowes E, Koers A, Hobbs C, Ferreira S, Geh JL, Healy C, Harries M, Acland KM, Blower PJ, Mitchell T, Fear DJ, Spicer JF, Lacy KE, Nestle FO, Karagiannis SN. IgG4 subclass antibodies impair antitumor immunity in melanoma. J Clin Invest. 2013

Apr; 123(4): 1457-74. PubMed PMID: 23454746;

Karagiannis P, Gilbert AE, Nestle FO, Karagiannis SN. IgG4 antibodies and cancer- associated inflammation: Insights into a novel mechanism of immune escape.

Oncoimmunology. 2013 Jul l;2(7):e24889. Epub 2013 May 7. PubMed PMID: 24073371;

Silva JP, Vetterlein O, Jose J, Peters S, Kirby H. The S228P Mutation Prevents in Vivo and in Vitro IgG4 Fab-arm Exchange as Demonstrated using a Combination of Novel

Quantitative Immunoassays and Physiological Matrix Preparation. J Biol Chem. 2015 Feb 27;290(9):5462-9. doi: 10.1074/jbc.Ml 14.600973. Epub 2015 Jan 7. PubMed PMID:

25568323; PubMed Central PMCID: PMC4342462.

Ahmad M, Mahajan VS, Mattoo H, Stone JH, Pillai S. Individuals with IgG4-related disease do not have an increased frequency of the K409 variant of IgG4 that compromises Fab-arm exchange. J Rheumatol. 2014 Jan;41(l): 185-7. doi: 10.3899/jrheum. l31017. PubMed PMID: 24382929. Labrijn AF, Buijsse AO, van den Bremer ET, Verwilligen AY, Bleeker WK, Thorpe SJ, Killestein J, Polman CH, Aalberse RC, Schuurman J, van de Winkel JG, Parren PW.

Therapeutic IgG4 antibodies engage in Fab-arm exchange with endogenous human IgG4 in vivo. Nat Biotechnol. 2009 Aug;27(8):767-71. doi: 10.1038/nbt. l553. Epub 2009 Jul 20. PubMed PMID: 19620983. van der Neut Kolfschoten M, Schuurman J, Losen M, Bleeker WK, Martinez-Martinez P, Vermeulen E, den Bleker TH, Wiegman L, Vink T, Aarden LA, De Baets MH, van de Winkel JG, Aalberse RC, Parren PW. Anti-inflammatory activity of human IgG4 antibodies by dynamic Fab arm exchange. Science. 2007 Sep 14;317(5844): 1554-7. PubMed PMID: 17872445.

Inoue D, Yoshida K, Yoneda N, Ozaki K, Matsubara T, Nagai K, Okumura K, Toshima F, Toyama J, Minami T, Matsui O, Gabata T, Zen Y. IgG4-Related Disease: Dataset of 235 Consecutive Patients. Medicine. 2015 Apr 94(15):e680. doi:

10.1097/MD.0000000000000680. PubMed PMID: 25881845.

Xiaoyu Yang, Ying Zhang, Fengqiang Wang, Larry Wang, Daisy Richardson, Mohammed Shameem,Alexandre Ambrogelly. Analysis and Purification of IgG4 Bispecific Antibodies by a Mixed-Mode Chromatography. Anal. Biochem. Anal Biochem 2015 Jun 16. Epub 2015 Jun 16.

Claims

1. A method of detecting or monitoring an IgG4-related disease in a subject comprising detecting in a sample the ratio between two or more of the relative amounts of immunoglobulins having:

(i) an IgG heavy chain class; and

(iii) immunoglobulins having the same heavy chain class but bound to either κ light chains only or λ light chains only; the method comprising the quantitative detection of said immunoglobulins by a binding assay comprising:

(i) binding said immunoglobulins to a binding agent specific for the immunoglobulin immobilised on a substrate or conjugated with a reporter molecule,

2. A method of detecting an amount of IgG4 kappa/lambda hybrid molecules comprising detecting in a sample an amount of immunoglobulins having:

(i) an IgG heavy chain class; and/or

(ii) an IgG4 heavy chain class bound to both a κ light chain and a λ light chain; and/or

(iii) immunoglobulins having the same heavy chain class but bound to either κ light chains only or λ light chains; the method comprising the quantitative detection of said immunoglobulins by a binding assay comprising:

(i) binding said immunoglobulins to a binding agent specific for the immunoglobulin immobilised on a substrate or conjugated with a reporter molecule, using a labelled detecting agent which detects immunoglobulin bound to said binding agent or conjugated with a reporter molecule.

3. A method according to claim 1 or 2, wherein the ratio is determined using:

(i) at least one binding agent specific for the IgG heavy chain class; and

a) a binding agent specific for the IgG4 heavy chain class bound to κ light chains in combination with a binding agent specific for λ light chains; and/or b) a binding agent specific for the IgG4 heavy chain class bound to λ light chains in combination with a binding agent specific for κ light chains; and/or c) a binding agent specific for the IgG4 heavy chain class bound to κ light chains in combination with a binding agent specific for the IgG heavy chain class bound to λ light chains; or

(ii) a binding agent specific for IgG4 heavy chain class bound to κ light chains in combination with a binding agent for IgG4 heavy chain class bound to λ light chains.

4. A method according to claim 1, 2 or 3, wherein the IgG heavy chain class is IgG4.

5. A method according to claim 1 to 4, wherein the ratio is determined using enzyme- linked immunosorbent assays (ELISAs) comprising the antibodies.

6. A method according to claim 1, 2 or 3, wherein the binding is determined using flow cytometry or fluorescently labelled beads.

7. A method according to claim 1, wherein the sample is whole blood, plasma, serum, urine, cerebrospinal fluid, lymph fluid, tissue or tumour tissue, most typically whole blood, plasma or serum.

8. A method according to claim 1, 2 or 3, wherein the result is used to determine the subject's response to treatment of the IgG4-related disease.

9. A method according to claim 1, 2 or 3, wherein the result is used to determine the removal of an antigen from the subjects environment.

10. A method according to claim 1, 2 or 3, wherein the result is used to further characterise an IgG4-related disease or allergic response.

11. A method according to claim 1, 2 or 3, additionally comprising the step of determining the presence of the IgG4 isoallotype K409.

12. A method according to any preceding claim, additionally comprising the step of detecting the presence and or/amount of free κ light chains and/or free λ light chains in the sample.

13. A method according to claim 12, wherein a ratio is determined between the amount of free κ light chains and free λ light chains.

14. A method according to claims 12 and 13, wherein the free κ light chains and/or free λ light chains are detected by antibodies or binding agents specific for free κ light chains and/or free λ light chains.

15. A method of determining the amount or proportion of an IgG4 heavy chain class bound to both a κ light chain and a λ light chain antibody in synthetic antibody preparations comprising detecting in a sample of the synthetic antibody preparation the ratio between two or more of the relative amounts of immunoglobulins having:

(i) an IgG heavy chain class; and

(iii) immunoglobulins having the same heavy chain class but bound to either κ light chains only or λ light chains only; the method comprising the quantitative detection of said immunoglobulins by a binding assay comprising: (i) binding said immunoglobulins to a binding agent specific for the immunoglobulin immobilised on a substrate or conjugated with a reporter molecule,

16. An assay kit comprising:

17. An assay kit according to claim 16 comprising a predetermined amount of an IgG4i<A. hybrid calibrator.

18. A method to be carried out by the kit of claim 16 or 17, wherein a ratio is detected between two or more of the relative amounts of immunoglobulins having:

(i) at least one binding agent specific for the IgG heavy chain class; and

a) a binding agent specific for the IgG heavy chain class bound to κ light chains in combination with a binding agent specific for κ light chains; and/or b) a binding agent specific for the IgG heavy chain class bound to λ light chains in combination with a binding agent specific for λ light chains; and/or c) a binding agent specific for the IgG heavy chain class bound to κ light chains in combination with a binding agent specific for the IgG heavy chain class bound to λ light chains.

19. An assay kit comprising:

(i) a binding agent specific for κ light chains; and

(ii) a binding agent specific for λ light chains; and

(iii) a predetermined amount of IgG4X/K mixed hybrid calibrator agent.

20. A method or kit according to any preceding claim wherein the binding agent or detecting agent is an antigen specific aptamer, antibody or antigen- specific fragment thereof.

21. A method according to claim 1, or a kit according to claim 16, substantially as described herein with reference to the accompanying tables and figures.