HK1254871B - Improved bacterial endotoxin test for the determination of endotoxins - Google Patents

Description

Improved bacterial endotoxin test for the determination of endotoxins

Technical Field

Here we report a bacterial endotoxin test (BET) sample preparation method that overcomes the low endotoxin recovery (LER) effect due to endotoxin shielding.

Background Art

Protein therapeutics (such as monoclonal antibodies) are typically produced using genetically transformed eukaryotic and prokaryotic cells (such as bacteria). Bacterial production is performed using fast-growing bacteria such as Escherichia coli. However, during the growth and cultivation of recombinant proteins, highly toxic lipopolysaccharides (LPS) are secreted into the culture medium. These components are called bacterial endotoxins (abbreviated as endotoxins). Gram-negative bacteria have LPS as an important component of their cell wall. Gram-negative bacterial cells contain approximately 3.5 x 10 ⁵ LPS molecules, occupying a total area of approximately 4.9 μm ² (Rietschel, 1994, FASEB J. 8: 217-225). In the case of E. coli, this means that LPS occupies about three-quarters of the total bacterial cell surface. Approximately 10,000 CFU (colony forming units) of Gram-negative bacterial species corresponds to 1 endotoxin unit (EU) (Rietschel, 1994, FASEB J. 8: 217-225). EU refers to endotoxin/LPS; depending on the LPS used, 1 EU ≈ 100 pg LPS. However, even when products are not produced by recombinant means, most reagents used are contaminated with endotoxins because their production is rarely carried out under aseptic or even sterile conditions. Therefore, LPS is a ubiquitous potential contaminant if aseptic and/or sterile conditions cannot be maintained during drug production. Of all known bacterial compounds, endotoxins are among the most toxic naturally occurring compounds to mammals. LPS, present in the cell walls of Gram-negative bacteria, is known to cause profound immune activation, including fever upon entry into the human bloodstream. Upon entry into the cardiovascular and lymphatic systems, respectively, it can cause unconsciousness effects at extremely low concentrations (in the picogram range). Unfortunately, bacterial endotoxins are heat-stable, and their toxicity is unrelated to the presence of bacterial cells. It is also generally known that all protein therapeutics, regardless of their production method, are contaminated with low trace amounts of bacterial endotoxins (so-called "naturally occurring endotoxins" (NOEs) that are expected or believed to be present. Therefore, endotoxin contamination remains an ongoing challenge for the production of drugs such as therapeutic monoclonal antibodies. This is clearly outlined in the "Guidance for Industry, Pyrogen and Endotoxin Testing" issued by the Food and Drug Administration (FDA) in June 2012.

To ensure that injectable protein therapeutics (such as monoclonal antibodies) are safe for human use, endotoxin testing is necessary. Endotoxin testing is typically performed using the pharmacopoeial methods in the United States Pharmacopoeia <85>, the European Pharmacopoeia 2.6.14, or the Japanese Pharmacopoeia 4.01, using gel-clot, chromogenic, or turbidimetric Limulus amoebocyte lysate (LAL) technique (also known as the LAL assay or LAL test). The pharmacopoeial name for the LAL assay is the bacterial endotoxin test (BET). BET is used to detect the presence of unsafe levels of endotoxins, particularly Gram-negative bacterial endotoxins, in a given sample or substance.

The LAL assay is typically performed with a diluted test sample along with a positive control, which is a sample with a known amount of spiked control standard endotoxin (CSE). CSE is a commercially available, defined form of endotoxin (e.g., provided by Lonza, Associates of Cape Cod, Inc. (ACC), or Charles River Laboratories International, Inc.). According to the pharmacopeia LAL assay specifications, CSE is spiked into the diluted sample at a non-interfering concentration (NIC) to achieve an acceptable recovery of 50-200%. This method fails to recognize that the composition of the pharmaceutical formulation sample matrix and storage conditions potentially affect the LAL reactivity of endotoxins present in the undiluted product sample. When the LAL assay is performed after spiking endotoxins such as CSE into the undiluted product sample, low endotoxin recoveries (<50%) are observed for certain biological products. This low endotoxin recovery is particularly observed if the product formulation contains amphiphilic compounds such as detergents. Detergents are added to the product to solubilize the therapeutic protein. This endotoxin shielding results in a significant reduction in endotoxin detection, especially in cases where LPS contamination is low. If endotoxin recovery after spiking cannot be increased by sample dilution, this phenomenon is referred to as "endotoxin shielding."

Various sample pretreatments are known for use in the LAL assay to overcome inhibition and/or enhancement of the assay. However, these sample pretreatments do not currently produce satisfactory results. Consequently, there is still a risk of endotoxin contamination occurring during pharmaceutical manufacturing that cannot be detected by the LAL assay due to endotoxin shielding. According to current knowledge, there are two different types of endotoxin shielding:

1) Endotoxin shielding caused by endotoxin-binding proteins present in the sample ("protein shielding", Petsch, Anal. Biochem. 259, 1998, 42-47). For example, it is known that protein-endotoxin aggregates are formed with, for example, human lipoprotein Apo A1, lysozyme, ribonuclease A, or human IgG, which reduce the LAL reactivity of endotoxin (Emancipator, 1992; Petsch, Anal. Biochem. 259, 1998, 42-47).

2) Endotoxin shielding caused by certain formulation ingredients or buffer components often present in drug products. For example, the endotoxin shielding caused specifically by the combination of polysorbate plus citrate or phosphate is called "Low Endotoxin Recovery" or LER (Chen, J. and Williams, K.L., PDA Letter 10, 2013, 14-16, Williams, American Pharmaceutical Review, October 28, 2013: Endotoxin test Concerns of Biologics). Endotoxin shielding can also be caused by any other buffer component and non-ionic detergents, or a combination thereof.

Due to the LER effect, potential endotoxin contamination occurring during manufacturing remains underestimated or undetected when using conventional LAL assays. The LER effect is an ongoing challenge for pharmaceutical products (Hughes, BioPharm. Asia March/April 2015, 14-25).

Therefore, the technical problem of the present invention is to provide means and methods for overcoming the LER effect.

Summary of the Invention

This technical problem has been overcome by the method of the present invention as described below.

Here we report an improved LAL assay for quantifying endotoxins. This improved LAL assay is particularly useful in cases where the endotoxin assay is shielded by an amphiphilic matrix (Low-Endotoxin-Recovery; LER).

In particular, in the context of the present invention, it was surprisingly found that the LER effect can be successfully overcome by the sequence of adding magnesium ions (e.g. in the form of _MgCl2 ) to the sample; diluting the sample; and dialyzing the sample with a pH value of 5.7-8.0. Or in other words, the sample preparation method reported herein is suitable for overcoming the LER effect in LAL assays.

More specifically, within the context of the present invention, a sample preparation method for antibody-containing samples, such as therapeutic monoclonal antibody samples, has been discovered. The sample preparation method of the present invention has the advantage that, when subjected to the LAL assay, it surprisingly and unexpectedly eliminates the LER effect. More specifically, the present invention relates to a method for preparing an antibody-containing sample for BET, preferably for the LAL assay, wherein the method comprises the following steps in the following order:

(a) adding magnesium ions to the sample, preferably in the form of _MgCl2 ,

(b) diluting the sample, and

(c) dialyzing the sample having a pH of 5.7-9.0, preferably 5.7-8.0, against an endotoxin-free aqueous solution.

Therefore, according to the present invention, a sample comprising an antibody (e.g., a sample of a therapeutic monoclonal antibody) is processed by implementing steps (a) to (c) of the sample preparation method of the present invention. These steps and combinations thereof surprisingly result in the provision of a sample in which the LER effect does not occur when the sample is subjected to the LAL assay. Or in other words, after performing steps (a) to (c) of the sample preparation method provided herein, the sample comprising the antibody is reactive to factor C in the LAL enzymatic cascade. Therefore, advantageously, the sample preparation method of the present invention is performed before bacterial endotoxins are determined by the LAL assay. Therefore, the present invention also relates to a method for determining (i.e., detecting and/or quantifying) endotoxins in a sample. In particular, the endotoxin assay method provided herein allows for determining (i.e., detecting and/or quantifying) endotoxins in a sample comprising an antibody (e.g., a therapeutic monoclonal antibody). Specifically, the present invention relates to a method for determining bacterial endotoxins in a sample comprising an antibody (preferably showing the LER effect), wherein the method comprises the following steps in the following order:

(a) adding magnesium ions, preferably in the form of _MgCl2 , to the sample (i.e. the sample containing the antibody),

(b) diluting the sample,

(c) dialyzing the sample having a pH of 5.7-8.0 against an endotoxin-free aqueous solution,

(d) Determination of bacterial endotoxins in the sample by using the LAL assay.

Preferably, in the sample preparation method or endotoxin determination method of the present invention, a 1.5-5 ml transparent glass crimp-top flat-bottom container is used. Most preferably, the container is a screw-top glass bottle (1.5 ml or 4 ml) from Macherey-Nagel GmbH.

Endotoxin contamination is a high risk in the production of pharmaceuticals such as monoclonal antibodies. In the prior art, endotoxin testing, in particular for therapeutic antibodies, is performed by using the conventional LAL assay test. However, as demonstrated in the accompanying examples, the LAL assay test is unable to detect/underestimate endotoxin contamination in antibody preparations that exhibit the LER effect. Undetected/underestimated endotoxins are an extreme safety risk for any drug sample, in particular for drugs administered intramuscularly or intravenously. However, despite its enormous practical importance, the physicochemical mechanism of the LER effect is unknown. Therefore, the prior art fails to provide a method for the correct determination of endotoxins in therapeutic products that exhibit the LER effect.

In the context of the present invention, a robust physico-chemical setup has been discovered that can eliminate the LER effect and result in satisfactory recoveries from CSE-spiked samples. In particular, as demonstrated in the illustrative examples, the methods reported herein allow for the recovery of CSE spiked into a given sample at defined concentrations (0.5 or 5.0 EU/ml). Importantly, the methods provided herein result in recoveries ranging between 50% and 200%, thus meeting FDA requirements. Thus, the present invention advantageously provides methods capable of removing endotoxin shielding and overcoming the LER effect. More specifically, in the context of the present invention, it was surprisingly discovered that a specific combination and sequence of steps (a) to (c) (i.e., (a) adding magnesium ions to the sample to be tested; (b) diluting the sample to be tested; and (c) dialyzing the sample to be tested (wherein the sample has a pH of 5.7-8.0)) eliminates the LER effect in samples to be tested for endotoxins. Alternatively, in other words, performing steps (a) to (c) can result in de-endotoxin shielding in the sample, thereby making it possible to detect endotoxins using the LAL assay test. The examples demonstrate that the methods provided herein overcome the LER effect, for example in formulated rituximab. In contrast, the same protocol for (which does not comprise an antibody but does comprise epoetin-β) was not able to obtain satisfactory results. This suggests that the methods provided herein are particularly useful for eliminating the LER effect in antibody formulations, preferably in formulations with monoclonal antibodies, citrate buffer, and polysorbate 80.

Therefore, the sample preparation method provided herein and the endotoxin determination method provided herein beneficially eliminate the LER effect. Therefore, these methods improve the detection of endotoxins in drugs. This leads to the production of drugs with fewer adverse reactions. Therefore, the method provided herein will improve the health status of consumers and can save the lives of critically ill patients.

In the methods provided herein, the antibody contained in the sample can be an antibody produced in bacteria or eukaryotic cells and/or purified from bacteria or eukaryotic cells. For example, antibodies can be produced and purified from Chinese hamster ovary (CHO) cells. In one aspect of the invention, the sample (i.e., the sample comprising the antibody) is a dissolved solid sample. In another aspect of the invention, the sample (i.e., the sample comprising the antibody) is a liquid sample. In the sample preparation method and endotoxin assay method provided herein, it can be considered that the antibody (i.e., the antibody contained in the sample) is a therapeutic antibody. Preferably, the antibody (i.e., the antibody contained in the sample) is a monoclonal antibody. However, in the methods provided herein, the antibody (i.e., the antibody contained in the sample) can also be a polyclonal antibody. In the text, multispecific antibodies (e.g., bispecific antibodies) or antibody fragments are also included in the term "antibody", as long as they show desired biological activity. The antibody can be human, humanized or camelized.

The methods provided herein advantageously render LER-prone pharmaceutical preparation samples reactive toward Factor C in the LAL enzymatic cascade. The LER effect has been reported in biological products formulated with amphiphilic compounds such as non-ionic detergents, particularly when combined with citrate or phosphate as buffers. The examples demonstrate that the methods provided herein reliably eliminate the LER effect in such therapeutic preparations. Thus, it is contemplated that in the sample preparation methods and endotoxin assays provided herein, the therapeutic antibody (i.e., the therapeutic antibody contained in the sample) is formulated with at least one detergent, preferably a polysorbate.

However, it is contemplated that the therapeutic antibodies are formulated with polysorbates, which do not contain the structural motifs of the lipid A cavity in the C-reactive protein of the LAL cascade. More specifically, straight-chain fatty acids such as lauric acid can mimic the fatty acids in the lipid A molecule of the LAL cascade, as this molecule also contains fatty acids with 12 carbon atoms and no double bonds (i.e., C:D is 12:0). Such straight-chain fatty acids can negatively interfere with the LAL cascade. Therefore, in the methods provided herein, it is contemplated that the therapeutic antibodies are not formulated with detergents containing straight-chain fatty acids such as lauric acid. Polysorbate 20 contains lauric acid. Therefore, it is contemplated that in the methods provided herein, samples (particularly samples of therapeutic antibodies) are not formulated with polysorbate 20. Phosphate buffers, particularly sodium phosphate buffers, can also interfere with the LAL cascade. Therefore, these buffers are less useful for the sample preparation methods provided herein. Therefore, the present invention relates to sample preparation methods or endotoxin assay methods provided herein, wherein the (therapeutic) antibody contained in the sample is not diluted with phosphate buffer. In one aspect of the invention, the sample does not contain phosphate buffer above 0.1 mM and does not contain polysorbate 20 at a concentration higher than 1/100 of its critical micelle concentration (CMC). In a preferred aspect of the invention, the sample does not contain phosphate buffer and polysorbate 20, or contains phosphate buffer and/or polysorbate 20 in an amount below the detection limit using a standard detection method.

As demonstrated in the examples using the methods of the present invention, the LER effect can be overcome in prepared rituximab samples as well as rituximab placebo samples. The only difference between the rituximab placebo sample and the rituximab sample is the absence of the antibody. In addition to this difference, the rituximab placebo sample contains all other components of the rituximab formulation, such as detergents and buffers. This shows that the methods provided herein do not rely on formulations containing specific monoclonal antibodies, but can be used, for example, to eliminate the LER effect in each formulation that exhibits the LER effect. Such formulations include formulations containing polysorbate 80 and chelating buffers (such as sodium citrate). Such formulations are typically used for antibodies, particularly monoclonal antibodies. Therefore, it is expected that the above method can be used to overcome the LER effect in each monoclonal antibody formulation. Rituximab was formulated with a mixture of polysorbate 80 and sodium citrate buffer (i.e., 25 mM sodium citrate buffer, pH 6.5; 700 mg/l polysorbate 80 and 154 mM NaCl). It is envisaged that samples comprising antibodies in the context of the present invention have this formulation.

The examples demonstrate that the LER effect can be overcome by using the methods provided herein in exemplary samples of therapeutic antibodies formulated with polysorbate 80 and citrate buffer. Therefore, in the sample preparation methods or endotoxin assay methods provided herein, the sample (i.e., a sample comprising an antibody) is preferably formulated with polysorbate 80. Therefore, in the methods provided herein, it is envisioned that the sample (i.e., a sample comprising an antibody) comprises polysorbate 80. Preferably, the sample comprises 500-1000 mg/l polysorbate 80, more preferably about 700 mg/l polysorbate 80. It is further envisioned that in the methods provided herein, the sample (i.e., a sample comprising an antibody) is formulated with a chelating buffer (e.g., a citrate buffer). The citrate buffer can be 5-50 mM citrate buffer, pH 6.0-7.0; preferably 25 mM citrate buffer, pH 6.5. Preferably, the citrate buffer is a sodium citrate buffer. For example, in the methods provided herein, the sample (i.e., the sample comprising the antibody) can comprise 5-50 mM sodium citrate, preferably 25 mM sodium citrate. Most preferably, the sample comprises polysorbate 80 and a sodium citrate buffer. For example, the sample can comprise approximately 700 mg/l polysorbate 80 and 5-50 mM, preferably approximately 25 mM sodium citrate buffer. Most preferably, in the methods provided herein, the sample is an antibody sample prepared with approximately 25 mM sodium citrate buffer and approximately 700 mg/l polysorbate 80 and having a pH value of approximately 6.5.

In the sample preparation methods or endotoxin assays provided herein, the antibody (i.e., the antibody contained in the sample) is preferably an anti-CD20 antibody. More preferably, the antibody is the anti-CD20 antibody rituximab. The amino acid sequences of the heavy and light chains of rituximab are shown herein as SEQ ID NOs: 1 and 2, respectively. A person skilled in the art will readily understand how to obtain an encoding nucleic acid sequence from a given amino acid sequence. Therefore, using SEQ ID NOs: 1 and 2, the encoding nucleic acid sequence of rituximab can be readily obtained. Rituximab is commercially available, for example, as RIWA®, ...

In step (a) of the sample preparation methods or endotoxin assay methods provided herein, magnesium ions (Mg ²⁺ ), for example, in the form of MgCl ₂ , are added to the sample (i.e., a sample containing an antibody). As used herein, the term "magnesium chloride" or "MgCl ₂ " refers to a chemical compound having the formula MgCl ₂ and its various hydrates, MgCl ₂ (H ₂ O) _x (i.e., MgCl ₂ ·xH ₂ O). For example, in step (a) of the methods provided herein, MgCl ₂ hexahydrate (i.e., MgCl ₂ ·6H ₂ O) can be added to the sample. Illustrative examples demonstrate that the addition of magnesium ions to a final concentration of 10-100 mM Mg ²⁺ in step (a) significantly reduces the LER effect. Furthermore, the examples show that a Mg ²⁺ concentration twice that of the sample buffer results in the best endotoxin recovery. For example, when using rituximab as a sample, the best endotoxin recovery was achieved when the magnesium salt MgCl ₂ was added in step (a) such that the final Mg ²⁺ concentration was twice that of the sodium citrate concentration (i.e., 50 mM ^{Mg 2+} ). Therefore, it is contemplated that magnesium ions in the form of a salt (e.g., MgCl ₂ ) are added in step (a) of the methods provided herein such that the final Mg ²⁺ concentration is twice that of the buffer (e.g., sodium citrate buffer). For example, in the methods provided herein, magnesium ions are preferably added to the sample such that the final Mg ²⁺ concentration [in step (a)] is 10-100 mM Mg ²⁺ , more preferably 25-75 mM Mg ²⁺ , even more preferably 40-75 mM Mg ²⁺ , and most preferably about 50 mM Mg ²⁺ (i.e., 45-55 mM Mg ²⁺ ). Alternatively, if the sample already contains magnesium ions, the amount of Mg ²⁺ added is adjusted so that the resulting final Mg ²⁺ concentration [in step (a)] is preferably 10-100 mM, more preferably 25-75 mM, even more preferably 40-75 mM, and most preferably about 50 mM Mg ²⁺ (i.e., 45-55 mM Mg ²⁺ ). After step (a), i.e., in step (b), the sample is diluted. However, in step (a), the term "magnesium ions are added to a concentration of..." or grammatical variations thereof and the term "magnesium ions are added to a final concentration of..." or grammatical variations thereof refer to the final Mg ²⁺ concentration in step (a). For example, adding MgCl ₂ to a (final) concentration of 45-55 mM MgCl ₂ in step (a) means that, after the addition of MgCl ₂ in step (a), the concentration of MgCl ₂ is 45-55 mM. Thus, if, for example, the sample is diluted in a ratio of 1:10 (sample:buffer/water) in step (b), the concentration of magnesium ions and likewise of _MgCl2 is 4.5-5.5 mM.

The examples demonstrate that an incubation step after the addition of magnesium ions further improves the recovery in the LAL assay. Thus, in the methods provided herein, after the addition of magnesium ions, the sample is preferably incubated for 30 minutes to 6 hours, more preferably 1 to 4 hours, and most preferably about 1 hour. In a preferred aspect of step (a) of the methods provided herein, after the addition of magnesium ions, the sample is incubated at room temperature for about 1 hour. The sample may be shaken before and after the incubation step [e.g., in a Heidolph MultiReax shaker at high speed (2,037 rpm)]. For example, the sample may be shaken for 30 seconds to 10 minutes, preferably 1 minute, before and after the incubation step.

In the step (b) of the sample preparation method or endotoxin determination method provided herein, the sample (i.e., the sample comprising the antibody) is diluted. The sample can be diluted with endotoxin-free water. The embodiments confirm that if the sample has a pH value of 5.7-8.0 during dialysis, a good recovery rate can be obtained. If the sample has a pH value of 6.0-8.0 during dialysis, an even better recovery rate is obtained. If the sample has a pH value of 6.5-7.5 during dialysis, the best recovery rate is obtained. Therefore, one aspect of the present invention relates to the method provided herein, wherein in step (b), the sample is diluted with endotoxin-free water, and wherein after dilution and before dialysis, the pH value of the sample is adjusted to 5.7-8.0, more preferably 6.0-8.0, and most preferably 6.5-7.5. Therefore, in one aspect of the present invention, in the step (b) of the method provided herein, the pH value of the sample is adjusted to pH 5.7-8.0, more preferably to pH 6.0-8.0. Most preferably, in the step (b) of the method provided herein, the pH value of the sample is adjusted to pH 6.5-7.5. For example, the pH value of the sample can be adjusted to pH 5.7, pH 5.8, pH 5.9, pH 6.0, pH 6.1, pH 6.2, pH 6.3, pH 6.4, pH 6.5, pH 6.6, pH 6.7, pH 6.8, pH 6.9 or pH 7.0. However, it is preferred herein that the pH value of the sample is adjusted in step (b) by diluting the sample with a 10-50 mM buffer solution (such as Tris/HCl buffer) of pH 6.0-9.0, more preferably with a 10-50 mM buffer solution (such as Tris/HCl buffer) of pH 6.0-8.0. Therefore, it is envisioned that in a preferred aspect of the method provided herein, in step (b), by diluting the sample with 10-50mM Tris/HCl buffer, pH 6.0-9.0 to adjust the pH value of the sample. More preferably, by diluting the sample with 10-50mMTris/HCl buffer, pH 6.0-8.0 to adjust the pH value of the sample. Most preferably, by diluting the sample (in step (b)) with 50mM Tris/HCl, pH～7.0 to adjust the pH value of the sample. Therefore, in the method provided herein, during the dialysis of step (c), the sample has a pH value of 5.7-8.0, preferably 6.0-8.0, more preferably 6.5-7.5.

As described above, the sample may contain a detergent, such as polysorbate 80. The illustrative examples demonstrate that, in particular, diluting a sample containing a detergent (e.g., polysorbate 80) can make endotoxin molecules accessible in the LAL assay. Without being bound by theory, it is believed that diluting a sample containing a detergent to near the CMC concentration can reduce micellar compartmentalization of the sample and, therefore, reduce the LER effect.

The examples show that a dilution of 1:5 to 1:20 significantly affects the recovery rate in the LAL assay. Therefore, the present invention relates to a sample preparation method and an endotoxin assay method provided herein, wherein in step (b), the sample is diluted at a ratio of 1:5 to 1:20 (sample: buffer/water), preferably 1:10 (sample: buffer/water). In the methods provided herein, the antibody is preferably formulated with about 25mM sodium citrate buffer and about 700mg/l polysorbate 80. Therefore, in the methods provided herein, the sample can be diluted in step (b) so that the concentration of the buffer is reduced to 5-1.25mM, preferably to 2.5mM. In addition, the sample can be diluted in step (b) so that the concentration of the detergent is reduced to 140-35mg/l, preferably to 70mg/l. In an embodiment, the sample is an antibody preparation having an antibody concentration of about 10mg/ml. These samples are diluted in step (b) of the method of the present invention to obtain an antibody concentration of 2-0.5mg/ml. Therefore, in the methods provided herein, the sample can be diluted in step (b) so that the antibody concentration is reduced to 2-0.5 mg/ml, preferably to 1 mg/ml. In the methods provided herein, an undiluted control can also be prepared. In addition to the undiluted control not being diluted (in step (b)), the undiluted control is treated in the same manner as the test sample. Herein, "buffer/water" means "buffer or water".

In step (c) of the sample preparation methods and endotoxin assay methods provided herein, the sample is dialyzed against an endotoxin-free aqueous solution. The endotoxin-free aqueous solution can be endotoxin-free water. However, the endotoxin-free aqueous solution can also be an endotoxin-free aqueous solution containing magnesium ions (e.g., added in the form of a salt, MgCl ₂ ). Therefore, one aspect of the present invention relates to the methods provided herein, wherein in step (c), the endotoxin-free aqueous solution contains magnesium ions, e.g., 2.5-10 mM MgCl ₂ .

Before starting dialysis, the sample can be shaken [e.g., in a Heidolph Multi Reax shaker, at room temperature, at high speed (2037 rpm)], for example, for 30 seconds to 10 minutes, preferably for 1 minute. Preferably, the dialysis in step (c) lasts for 1-48 hours, more preferably for 4-24 hours, and most preferably for about 24 hours. Therefore, preferably, the dialysis lasts for about 24 hours in step (c). The dialysis can be carried out at 15-30°C, preferably at room temperature (i.e., 21±2°C). After the dialysis, the sample can be shaken [e.g., in a Heidolph Multi Reax shaker, at room temperature, at high speed (2037 rpm), for example, for 20 minutes to 1 hour, preferably for (at least) 20 minutes.

Dialysis can be performed using a rotary dialyzer (Spin Dialyzer, e.g., Harvard rotary dialyzer, catalog number Nb.74-0314)) or a fast rotary dialyzer (Fast Spin Dialyzer, e.g., Harvard Fast Spin Dialyzer, catalog number Nb.74-0412)). The preferred rotational frequency of the agitator is high (particularly when using a fast rotary dialyzer), which means that the frequency of the agitator is 50 to 300 rpm, preferably 200 to 300 rpm. The agitator preferably has a length of 20-60 mm and a diameter (i.e., cross-sectional dimensions) of 5-25 mm. More preferably, the agitator has a length of about 40 mm and a diameter of about 14 mm. The agitator is most preferably a heat-sterilized (e.g., 4 hours at 250°C) magnetic agitator having a length of about 40 mm and a diameter of about 14 mm. Such agitators are available from OMNILAB. In fact, dialysis is usually performed using a high-frequency agitator because it is beneficial for diffusion through the dialysis membrane. The dialysis of the patient's blood is carried out under a high frequency stirrer. The dialysis of the patient's blood is carried out under a high frequency stirrer. The dialysis of the patient's blood is carried out under a high frequency stirrer. The dialysis of the patient's blood is carried out under a high frequency stirrer. Therefore, the dialysis of the patient's blood using a high frequency stirrer is a standard dialysis step. The container for dialysis preferably has a capacity of 500-5000 ml, more preferably 1000-3000 ml, most preferably 1500-2500 ml. The container can have a diameter of 120 mm and a height of 240 mm. For example, the container for dialysis can be a 2000 ml tall beaker (e.g., available from German OMNILAB, P/N: 5013163). It is preferred to use a fast rotating dialyzer because it has a double-area dialysis membrane and is therefore considered to be suitable for more effective and faster dialysis.

It is envisaged that for the dialysis in step (c), a membrane with a molecular weight cut-off of 100 to 16 kDa, preferably 500 to 10 kDa, and most preferably 10 kDa is used. For the dialysis in step (c), a cellulose ester or cellulose acetate membrane may be used. Preferably, for the dialysis in step (c), a cellulose acetate membrane is used. Most preferably, a cellulose acetate membrane with a molecular weight cut-off of 10 kDa is used during dialysis.

Therefore, in the step (c) of method provided herein, preferably by using the cellulose acetate membrane with molecular weight cut-off 10kDa to carry out dialysis for about 24 hours.Before dialysis, dialysis membrane can be washed, preferably in endotoxin-free water, washing.Especially, dialysis membrane can be vibrated in endotoxin-free water (for example, using shaking table or equivalent of German SG 20IDL GmbH, 50 to 300rpm, preferably 100rpm).For example, dialysis membrane can be washed by vibrating dialysis membrane in endotoxin-free water for 10 minutes to 3 hours, preferably 1 hour.After this washing step, dialysis membrane is preferably transferred to fresh endotoxin-free water, and again by vibrating dialysis membrane for 10 minutes to 3 hours, preferably 1 hour, washing.

Dialysis can be performed in a 1 ml cell, for example, in a 1 ml rotary dialyzer (Harvard) equipped with a membrane (e.g., cellulose acetate membrane) with a molecular weight cut-off range of 500 Da to 10 kDa (e.g., a molecular weight cut-off of 10 kDa). During the dialysis, the water is preferably changed, more preferably twice. For example, the water can be changed after 2 and 20 hours of dialysis or after 18 and 22 hours of dialysis. Preferably, the water is changed after 2 and 4 hours of dialysis.

Preferably, in the methods provided herein, after dialysis in step (c), the sample is shaken [e.g., in a Heidolph Multi Reax shaker at room temperature, high speed (2037 rpm). Preferably, after dialysis for 10 minutes to 1 hour, more preferably 20 minutes, the sample (i.e., the sample comprising the antibody) is shaken. In addition to or in lieu of shaking, the sample can be sonicated after dialysis. Thus, one aspect of the invention relates to the methods provided herein, wherein in step (c), the sample is sonicated after dialysis.

Combining the sample preparation methods provided herein with the LAL assay test advantageously meets FDA requirements for quantitative and reproducible detection of defined amounts of CSE spiked into a sample. Preferably, the LAL assay test of the methods provided herein is the LAL assay test described below.

The examples show that the addition of Mg ²⁺ (i.e., magnesium ions) has the further advantage that it retains endotoxins in the inner compartment of the dialysis chamber. Thus, the dialysis in step (c) only results in the removal of the buffer (e.g., sodium citrate buffer) and not the endotoxins. The dilution step can reduce the concentration of the detergent (e.g., polysorbate 80), thereby eliminating the inhibition of the LAL cascade by the detergent. The examples demonstrate that the LER effect is particularly reproducibly overcome if the steps of the method according to the invention are carried out in the following order: (1) addition of Mg ²⁺ ; (2) dilution; and (3) dialysis. Thus, the combination of steps (1), (2) and (3), or the combination of steps (a), (b) and (c) in the claims, reproducibly overcomes the LER effect. Preferred amounts of Mg ²⁺ added, preferred dilutions and preferred dialysis parameters are described in detail above and below.

In a preferred aspect, the present invention relates to a method provided herein for preparing a sample containing antibodies for use in a LAL assay, wherein the method comprises the following steps in the following order:

(a) adding magnesium ions to the sample, for example in the form of _MgCl2 , to a final concentration of 10-100 mM, preferably 40-75 mM, most preferably 45-55 mM,

(b) diluting the sample with 10-50 mM Tris/HCl buffer, pH 6.0-8.0, preferably with 50 mM Tris/HCl buffer, pH ~ 7.0, at a ratio of 1:5 (sample: buffer) to 1:20 (sample: buffer), preferably 1:10 (sample: buffer),

(c) The sample is dialyzed against endotoxin-free water at a pH of 5.7-8.0 (preferably 6.5-7.5) for 1-48 hours, preferably 4-24 hours, and most preferably 24 hours. Preferably, a cellulose acetate membrane with a molecular weight cut-off of 10 kDa is used, and the water is changed after 2 and 4 hours. Most preferably, a fast rotary dialyzer is used with an agitator frequency of 50 to 300 rpm, preferably 200 rpm.

As described above, the antibody is preferably a monoclonal antibody. More preferably, the antibody is rituximab. Most preferably, in the methods provided herein, the sample is an antibody sample prepared with about 25 mM sodium citrate buffer (7.35 mg/ml) and about 700 mg/l polysorbate 80 and having a pH of about 6.

The term "about" and the symbol "~" are used interchangeably herein to indicate that the particular value provided can vary to some extent. For example, "about" or "~" (e.g., in the case of about/~25 mM sodium citrate buffer) means that the given value includes a variation within a range of ±10%, preferably ±5%, and most preferably ±2%.

As indicated, it is contemplated that in the present invention, the sample preparation methods provided herein are combined with the LAL assay.The LAL assay has the advantage of detecting low concentrations of endotoxin.

The validated lower limit of endotoxin detection in the kinetic chromogenic LAL technique is 0.005 EU/mL, as given by the CSE standard curve. The LAL reagents for these techniques contain a complete enzymatic amplification cascade of a serine protease purified from the Limulus crab.

In a recently developed assay (Hyglos GmbH, Germany), the manufacturer's stated lower limit of endotoxin (CSE) detection is 0.05 EU/mL (Hyglos advertisement: Grallert et al. in: Nature Methods, Oct. 2011; p://www.hyglos.de/fileadmin/media/Application_note_EndoLISA_Nature_Methods_October_2011.pdf). This assay uses only a recombinant form of the initial enzyme of the Limulus cascade (i.e., Factor C). Unlike the certified LAL assay, the assay also includes an initial endotoxin adsorption step by pre-coating microtiter plates with phage-encoded proteins, which has not yet been demonstrated to bind the broad spectrum of bacterial endotoxins known to be detectable by the LAL method.

In particular, in a preferred aspect, the present invention relates to a method for determining bacterial endotoxins in a sample containing a polypeptide, wherein the method comprises the following steps in the following order:

(a) adding magnesium ions, preferably in the form of _MgCl2 , to the sample to a final concentration of 10-100 mM, preferably 40-75 mM, most preferably 45-55 mM,

(b) diluting the sample with 10-50 mM Tris/HCl buffer, pH 6.0-8.0, preferably with 50 mM Tris/HCl buffer, pH ~ 7.0, at a ratio of 1:5 (sample: buffer) to 1:20 (sample: buffer), preferably 1:10 (sample: buffer),

(c) The sample is dialyzed against endotoxin-free water at a pH of 5.7-8.0 (preferably 6.5-7.5) for 1-48 hours, preferably 4-24 hours, and most preferably 24 hours. Preferably, a cellulose acetate membrane with a molecular weight cut-off of 10 kDa is used, and the water is changed after 2 and 4 hours. Most preferably, a fast rotary dialyzer is used with an agitator frequency of 50 to 300 rpm, preferably 200 rpm.

(d) Determination of bacterial endotoxins in the sample by using the LAL assay.

As described above, the antibody is preferably a monoclonal antibody. More preferably, the antibody is rituximab. Most preferably, in the methods provided herein, the sample is an antibody sample formulated with about 25 mM sodium citrate buffer and about 700 mg/l polysorbate 80 and having a pH of about 6.5.

As noted in the Examples, a "LER positive control" (also referred to as a "positive LER control") can be used in the LAL assay of step (d) of the endotoxin assay method provided herein. The "LER positive control" is an indicator to show that the sample to be tested (i.e., a sample containing antibodies) will show a LER effect if steps (a) to (c) of the method described herein are not performed. Or in other words, the "LER positive control" is used as a positive control in the LAL assay to show that a known amount of endotoxin (in the sample to be tested) cannot be recovered using only the LAL assay (i.e., without performing steps (a) to (c) of the method provided herein). In the context of the present invention, it was surprisingly and unexpectedly discovered that only a positive LER effect can be obtained by shaking the sample for 45 minutes to 2 hours, preferably about 60 minutes to 2 hours, and most preferably about 60 minutes after CSE is added to the sample. Thus, in the context of the present invention, a "LER positive control" is prepared by spiking an aliquot of a sample to be tested for endotoxin (e.g., an aliquot of a sample containing an antibody) with a known amount of endotoxin and shaking the spiked sample for 45 minutes to 2 hours, preferably about 60 minutes to 2 hours, and most preferably about 60 minutes. Thus, the present invention relates to an endotoxin assay method as provided herein, further comprising generating a LER positive control by spiking an aliquot of a sample with a known amount of endotoxin and shaking the spiked sample aliquot for ≥60 minutes (more preferably 60 minutes to 2 hours).

Preferably, in the method for determining bacterial endotoxins provided herein, a "LER positive control" is prepared by spiking CSE to a final concentration of 5.0 EU/ml into an aliquot of the sample to be tested. Thereafter, the spiked aliquot is shaken [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature] for ≥ 60 minutes, most preferably 60 minutes. After shaking, the aliquot spiked with endotoxin is preferably diluted to the same extent as the sample to be tested in step (b) of the method provided herein. Preferably, the spiked aliquot is diluted with endotoxin-free water. After dilution, the aliquot spiked with oscillation is preferably performed [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature] for example for 1 minute.

Therefore, the "LER positive control" is preferably prepared in the following order with the following procedure:

- CSE is spiked into aliquots of the sample to be tested to a final concentration of 5.0 EU/ml. Preferably, a "LER positive control" is prepared in a 1.5-5 ml clear glass crimp-top flat-bottom container, more preferably in a screw-top glass vial (1.5 ml or 4 ml) from Macherey-Nagel GmbH.

- Shake the spiked aliquot for ≥ 60 minutes (more preferably 60 minutes to 2 hours), most preferably 60 minutes. Preferably, the spiked aliquot is shaken at high speed (2037 rpm) at room temperature (i.e., 21 ± 2°C). Most preferably, the spiked aliquot is shaken at high speed (2037 rpm) at room temperature in a Heidolph MultiReax shaker.

- Dilute the spiked aliquot with endotoxin-free water. Dilute the spiked aliquot to the same extent as the test sample in step (b) of the methods provided herein (i.e., if the test sample was diluted at a 1:10 ratio in step (b), then also dilute the spiked aliquot at a 1:10 ratio).

- Vortex the spiked aliquot (eg 1 minute).

As described above, during the preparation of the "LER positive control", a dilution is performed. However, it is envisaged that in the context of the present invention, the "LER positive control" is not processed as described in steps (a) to (c) of the method provided herein, except for the dilution. However, the "LER positive control" is used in step (d) of the method for determining bacterial endotoxins to show that the sample to be tested (i.e., the sample containing the antibody) will exhibit the LER effect if steps (a) to (c) of the method described herein are not performed. The "LER positive control" can be prepared in the time period of any one of steps (a) to (c) (e.g., during the dialysis time) for use when performing the LAL assay test.

In order to identify that a given substance (e.g., a sample or buffer of a therapeutic antibody) exhibits the LER effect, the endotoxin content can be monitored over time, for example, in an endotoxin retention time study. An endotoxin retention time study requires the spiking of an undiluted sample with endotoxin and the storage of the endotoxin-spiked sample for a period of time. For example, the sample can be stored for up to several months. Preferably, in a retention time study, the endotoxin-spiked sample is stored for several days (e.g., 7 to up to 28 days) and the LAL assay is performed at the specified time points. A recovery rate of less than 50% of the amount of endotoxin spiked in indicates that the sample exhibits the LER effect.

As mentioned above, the LAL assay is routinely performed with a diluted test sample and a diluted positive control (PPC), which is a sample with a known amount of CSE spiked in. Thus, in the LAL assay performed in step (d) of the endotoxin assay method provided herein, it is contemplated that each sample is measured in duplicate, each time with and without spiked control standard endotoxin (PPC). Thus, using each given sample, it is easy to test whether the sample preparation method provided herein or the endotoxin assay method provided herein has the following beneficial effect: that is, the endotoxin present in the sample (or at least 50-200% of the endotoxin present as required by the FDA) can be detected by using the LAL assay. Thus, it is contemplated that the LAL assay in step (d) of the endotoxin assay method provided herein includes testing the positive control (PPC) together with the test sample (i.e., the sample containing the antibody to be tested). The positive control is identical to the test sample, except that the PPC is spiked with a known amount of CSE. Or in other words, steps (a) to (c) of the methods provided herein must be performed on the test sample and the PPC in the same manner. Therefore, the PPC is prepared before step (a) of the methods provided herein.

In the context of the present invention, it was surprisingly and unexpectedly found that a positive LER effect could only be achieved if the sample was shaken for 45 minutes to 2 hours, preferably for about 60 minutes to 2 hours, and most preferably for about 60 minutes after CSE was incorporated into the sample. Thus, in the context of the present invention, the PPC was shaken [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm)] for 45 minutes to 2 hours, preferably for about 60 minutes to 2 hours, and most preferably for about 60 minutes after incorporation. More preferably, the PPC was shaken [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm)] for about 60 minutes at room temperature after incorporation.

Therefore, a preferred aspect of the present invention relates to a method provided herein for determining bacterial endotoxins in a sample containing antibodies, wherein the method comprises the following steps in the following order:

(a0) PPC was prepared by the following steps

- spiking a first aliquot of the sample containing the antibody with a known amount of endotoxin,

- shaking the endotoxin-spiked aliquot for 60 minutes to 2 hours (preferably about 60 minutes at room temperature),

(a) adding magnesium ions to a second aliquot of the sample to be tested and to the PPC,

(b) diluting a second aliquot of the sample to be tested and the PPC,

(c) dialyzing a second aliquot of the test sample and the PPC having a pH of 5.7-8.0 (preferably 5.8-7.0) against an endotoxin-free aqueous solution, wherein the test sample and the PPC have a pH of 5.7-9.0, and

(d) Assaying a second aliquot of the test sample and the PPC for bacterial endotoxins using the LAL assay.

In one aspect of the invention, endotoxin is spiked into PPC to give a final endotoxin concentration of 5.0 EU/ml.

All aspects and definitions disclosed in conjunction with the method for determining bacterial endotoxins provided herein may be applied, with appropriate changes, to the method using PPC. Thus, a preferred aspect of the present invention relates to the method provided herein for determining bacterial endotoxins in a sample containing antibodies, wherein the method comprises the following steps in the following order:

(a0) PPC was prepared by the following steps

- spiking a first aliquot of the sample containing the antibody with a known amount of endotoxin, and

- shaking the endotoxin-spiked aliquot for ≥ 60 minutes (preferably 60 minutes at room temperature),

(a) adding magnesium ions (preferably in the form of _MgCl2 ) to a second aliquot of the sample to a final concentration of 10-100 mM, preferably 40-75 mM, most preferably 45-55 mM,

(b) diluting a second aliquot of the sample with 10-50 mM Tris/HCl buffer pH 6.0-8.0, preferably with 50 mM Tris/HCl buffer pH ~7.0, in a ratio of 1:5 (sample:buffer) to 1:20 (sample:buffer), preferably 1:10 (sample:buffer),

(c) The sample is dialyzed against endotoxin-free water at a pH of 5.7-8.0 (preferably 6.5-7.5) for 1-48 hours, preferably 4-24 hours, and most preferably 24 hours. Preferably, a cellulose acetate membrane with a molecular weight cut-off of 10 kDa is used, and the water is changed after 2 and 4 hours. More preferably, a fast rotating dialyzer is used with a high agitator frequency.

(d) Determination of bacterial endotoxins in the sample by using the LAL assay.

In addition, water controls can be applied to the endotoxin assay methods provided herein. Preferably, at least two water controls are used; one of which is composed of endotoxin-free water and the other is endotoxin-free water spiked with a known amount of endotoxin (e.g., to obtain a final concentration of 5.0 EU/ml of CSE). The water controls are treated in the same manner as the test samples.

As described above, in the sample preparation methods provided herein and the endotoxin assay methods provided herein, it is contemplated that, in step (a), the sample is incubated for 30 minutes to 6 hours, preferably 1-4 hours, and most preferably 1 hour. In addition, it is also contemplated that, after dialysis, the sample is shaken [e.g., in a Heidolph MultiReax shaker at high speed (2037 rpm) at room temperature] for example, 10 minutes to 1 hour, preferably 20 minutes. Thus, one aspect of the present invention relates to a method provided herein for preparing a sample comprising an antibody for use in a LAL assay, wherein the method comprises the following steps in the following order:

(a) adding magnesium ions (preferably in the form of _MgCl2 ) to the sample to a final concentration of 10-100 mM, preferably 40-75 mM, most preferably 45-55 mM; and incubating the sample for 30 minutes to 6 hours, preferably 1-4 hours, most preferably 1 hour,

(b) diluting the sample with 10-50 mM Tris/HCl buffer pH 6.0-8.0, preferably with 50 mM Tris/HCl buffer pH ~ 7.0, at a ratio of 1:5 (sample: buffer) to 1:20 (sample: buffer), preferably 1:10 (sample: buffer),

(c) The sample is dialyzed against endotoxin-free water at a pH of 5.7-8.0 (preferably 6.5-7.5) for 1-48 hours, preferably 4-24 hours, and most preferably 24 hours. (Preferably, a cellulose acetate membrane with a molecular weight cutoff of 10 kDa is used, and the water is changed after 2 and 4 hours. Most preferably, a fast-rotating dialyzer is used with a high agitator frequency.) After dialysis, the sample is shaken for 10 minutes to 1 hour, preferably 20 minutes.

Similarly, another aspect of the present invention relates to a method provided herein for determining bacterial endotoxins in a sample containing antibodies that exhibit the LER effect, wherein the method comprises the following steps in the following order:

(a) adding magnesium ions (preferably in the form of _MgCl2 ) to the sample to a final concentration of 10-100 mM, preferably 40-75 mM, most preferably 45-55 mM; and incubating the sample for 30 minutes to 6 hours, preferably 1-4 hours, most preferably 1 hour (wherein the sample may be shaken before and after incubation),

(b) diluting the sample with 10-50 mM Tris/HCl buffer pH 6.0-8.0, preferably with 50 mM Tris/HCl buffer pH ~ 7.0, at a ratio of 1:5 (sample: buffer) to 1:20 (sample: buffer), preferably 1:10 (sample: buffer),

(c) The sample is dialyzed against endotoxin-free water at a pH of 5.7-8.0 (preferably 6.5-7.5) for 1-48 hours, preferably 4-24 hours, and most preferably 24 hours. (Preferably, a cellulose acetate membrane with a molecular weight cutoff of 10 kDa is used, and the water is changed after 2 and 4 hours. Most preferably, a fast-rotating dialyzer is used with a high agitator frequency.) After dialysis, the sample is shaken for 10 minutes to 1 hour, preferably 20 minutes.

(d) Determination of bacterial endotoxins in the sample by using the LAL assay.

Furthermore, as described above, in the endotoxin assay method provided herein, it is contemplated that the PPC is prepared and shaken for 60 minutes to 2 hours after incorporation. Thus, the present invention relates to a method provided herein for determining bacterial endotoxins in a sample containing antibodies that exhibit the LER effect, wherein the method comprises the following steps in the following order:

(a0) PPC was prepared by the following steps

- spiking a first aliquot of the sample containing the antibody with a known amount of endotoxin (e.g. to a final concentration of 5.0 EU/ml), and

- shaking the endotoxin-spiked aliquot for 60 minutes to 2 hours (preferably 60 minutes at room temperature),

(a) adding magnesium ions (preferably in the form of _MgCl2 ) to a final concentration of 10-100 mM, preferably 40-75 mM, most preferably 45-55 mM to a second aliquot of the sample and to the PPC; (preferably incubating the sample and the PPC for 30 minutes to 6 hours, more preferably 1-4 hours, most preferably 1 hour; (wherein the sample may be shaken before and after incubation)),

(b) diluting the sample and PPC with 10-50 mM Tris/HCl buffer pH 6.0-8.0, preferably with 50 mM Tris/HCl buffer pH ~ 7.0, at a ratio of 1:5 (sample/PPC: buffer) to 1:20 (sample/PPC: buffer), preferably 1:10 (sample/PPC: buffer),

(c) The sample and PPC are dialyzed against endotoxin-free water at a pH of 5.7-8.0 (preferably 6.5-7.5) for 1-48 hours, preferably 4-24 hours, and most preferably 24 hours. (Preferably, a cellulose acetate membrane with a molecular weight cutoff of 10 kDa is used, and the water is changed after 2 and 4 hours. More preferably, a fast-rotating dialyzer is used with a high agitator frequency.) After dialysis, the sample and PPC are shaken for 10 minutes to 1 hour, preferably 20 minutes.

(d) Determination of bacterial endotoxins in the sample by using the LAL assay.

In addition, as described above, it is conceivable that in the endotoxin assay method provided herein, a "LER positive control" is prepared and used in step (d) to demonstrate the LER effect. Therefore, a preferred aspect of the present invention relates to a method provided herein for determining bacterial endotoxins in a sample containing antibodies that demonstrate the LER effect, wherein the method comprises the following steps in the following order:

(a0) PPC was prepared by the following steps

- spiking a first aliquot of the sample containing the antibody with a known amount of endotoxin (e.g. to a final concentration of 5.0 EU/ml), and

- shaking the endotoxin-spiked aliquot for 60 minutes to 2 hours (preferably 60 minutes at room temperature),

(a) adding magnesium ions (preferably in the form of _MgCl2 ) to a final concentration of 10-100 mM, preferably 40-75 mM, most preferably 45-55 mM to a second aliquot of the sample and to the PPC; (preferably incubating the sample and the PPC for 30 minutes to 6 hours, more preferably 1-4 hours, most preferably 1 hour; (wherein the sample may be shaken before and after incubation)),

(b) diluting the sample and PPC with 10-50 mM Tris/HCl buffer pH 6.0-8.0, preferably with 50 mM Tris/HCl buffer pH ~ 7.0, at a ratio of 1:5 (sample/PPC: buffer) to 1:20 (sample/PPC: buffer), preferably 1:10 (sample/PPC: buffer),

(c) The sample and PPC are dialyzed against endotoxin-free water at a pH of 5.7-8.0 (preferably 6.5-7.5) for 1-48 hours, preferably 4-24 hours, and most preferably 24 hours. (Preferably, a cellulose acetate membrane with a molecular weight cutoff of 10 kDa is used, and the water is changed after 2 and 4 hours. Most preferably, a fast-rotating dialyzer is used with a high agitator frequency.) After dialysis, the sample and PPC are shaken for 10 minutes to 1 hour, preferably 20 minutes.

(d) Determination of bacterial endotoxins in the sample and PPC by using the LAL assay, wherein a "LER positive control" is used in the LAL assay, which is prepared by the following steps:

- spike the third aliquot of the sample to be tested with CSE to a final concentration of 5.0 EU/ml;

- Shake the spiked aliquot for ≥ 60 minutes, most preferably 60 minutes;

- diluting the spiked aliquot with endotoxin-free water (diluting the spiked aliquot to the same extent as the sample to be tested in step (b) of the methods provided herein);

- Vortex the spiked aliquot (eg 1 minute).

Therefore, a preferred aspect of the present invention relates to a method as provided herein for determining bacterial endotoxins in a sample containing antibodies that exhibit the LER effect, wherein the method comprises the following steps in the following order:

(a0) PPC was prepared by the following steps

- spike a first aliquot of the sample containing the antibody with a known amount of endotoxin to a final concentration of 5.0 EU/ml, and

- Shake the endotoxin-spiked aliquot at room temperature for approximately 60 minutes,

(a) adding magnesium ions (preferably in the form of _MgCl2 ) to a final concentration of 45-55 mM to a second aliquot of the sample and the PPC, shaking the sample and PPC for 1 hour, incubating the sample and PPC for 1 hour, and shaking the sample and PPC again after the incubation,

(b) Dilute the sample and PPC at 1:10 (sample/PPC:buffer) using 50 mM Tris/HCl buffer, pH 7.0.

(c) dialyzing the sample and PPC against endotoxin-free water having a pH of 5.7-8.0 (preferably 6.5-7.5) for 24 hours using a cellulose acetate membrane with a molecular weight cut-off of 10 kDa; wherein the water is changed after 2 hours and 4 hours (preferably using a fast-spinning dialyzer with a high frequency of agitation), shaking the sample and PPC for 20 minutes, and

(d) Determination of bacterial endotoxins in the sample and PPC by using the LAL assay, wherein a "LER positive control" is used in the LAL assay, which is prepared by the following steps:

- spike the third aliquot of the sample to be tested with CSE to a final concentration of 5.0 EU/ml;

- Shake the spiked aliquot for 60 minutes,

- Dilute the spiked aliquot with endotoxin-free water at a ratio of 1:10,

- Vortex the spiked aliquot (eg 1 minute).

In addition, as described above, it is conceivable that a water control may be used in the LAL assay. For example, a water control consisting of endotoxin-free water may be used in the LAL assay of step (d) of the endotoxin assay method provided herein. Another water control may consist of endotoxin-free water spiked with endotoxin. After endotoxin spiked, the water is preferably shaken [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm)] for ≥ 60 minutes (e.g., 60 minutes at room temperature). In addition, in the LAL assay, standards are typically prepared according to the instructions of the kit used.

Steps (a0), (a), (b), (c) and (d) should be performed in the order of (a0) → (a) → (b) → (c) → (d). However, washing the dialysis membrane can be performed at any time, provided that the step is performed at the beginning of dialysis. Similarly, preparation of the LER positive control can be performed at any time, provided that the step is performed at the beginning of the LAL assay. In a preferred aspect of the present invention, the endotoxin assay method provided herein comprises the following steps.

Step (a00) : Sample preparation

Adjust the concentration of the sample to be tested to suit the PPC (e.g. 900 μl of antibody + 100 μl of endotoxin-free water)

Spike an aliquot of the test sample with endotoxin to generate PPC (e.g. 900 μl of antibody + 100 μl of CSE at 50 EU/ml = final concentration 5.0 EU/ml)

Prepare a water control (e.g. 1000 μl of endotoxin-free water)

Prepare another water control (e.g. 900 μl of endotoxin-free water + 100 μl of CSE at 50 EU/ml = final concentration 5.0 EU/ml)

Shake the sample at room temperature for about 60 minutes [e.g. in a Heidolph Multi Reax shaker, high speed (2037 rpm)],

Step (a01) : Washing the dialysis membrane

For example, use a 10 kDa cellulose acetate (CA) membrane and place it in a crystallizing dish with endotoxin-free water (e.g. 300 ml of distilled water from the manufacturer B. Braun, Melsungen)

Carefully shake for 1 hour (shaker SG20. IDL GmbH, Germany or equivalent, 50 to 300 rpm, preferably 100 rpm)

• transfer the membrane to a new crystallization dish with fresh endotoxin-free water (e.g. 300 ml distilled water from the manufacturer B. Braun, Melsungen),

Shake (shaker SG 20. IDL GmbH, Germany or equivalent, 50 to 300 rpm, preferably 100 rpm) for 1 hour

Step (a) : Add 25-100mM, preferably 50-100mM magnesium ions (Mg ²⁺ )

Add Mg ²⁺ (e.g. in the form of MgCl ₂ ) to the sample of step (a0) to a final concentration of 25-100 mM, preferably 50-100 mM (e.g. add 50 μl of 1 M MgCl ₂ stock solution to the sample of step (a0))

Shake [e.g. in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature] for 2-5 minutes, e.g. 1 minute

Incubate the sample at room temperature for 45 to 75 minutes, preferably 60 minutes

Shake [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature] for 1 minute

Step (b) : Dilution

Take one of the samples from step (a) and dilute it 1:10 with a buffer solution at pH 7.0 (e.g. 50 mM Tris/HCl buffer at pH 7.0) (e.g. 895 μl 50 mM Tris buffer + 105 μl sample)

Preferably prepare two diluted samples

·For example:

o 2x antibody, diluted 1:10 in Tris buffer (sample)

o 2x antibody, diluted 1:10 in Tris buffer, spiked with 5.0 EU/ml (PPC)

o 2x LAL water, diluted 1:10 in Tris buffer (background)

o 2x LAL water 5.0EU/ml, diluted 1:10 with Tris buffer (standard)

o LAL-water = please add

Step (c) : Dialysis

Shake [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature for 1 minute for all diluted samples]

Transfer it to a dialyzer (preferably a fast-spin dialyzer)

Place one dialyzer per beaker on a stirrer

Fill a beaker with endotoxin-free water (e.g. 200 ml of distilled water from the manufacturer B. Braun, Melsungen)

Dialyze for 24 hours at room temperature, exchanging the solution with endotoxin-free water after 2 and 4 hours.

The frequency of the stirrer is preferably 50 to 300 rpm, more preferably 200 rpm (especially if a fast-rotating dialyzer is used). The stirrer preferably has a length of 20-60 mm and a diameter of 5-25 mm. More preferably, the stirrer is a magnetic stirrer having a length of about 40 mm and a diameter of about 14 mm.

Step (d) : Oscillation

After dialysis, transfer the sample to a new container (e.g., a 1.5 ml screw-cap vial) and shake [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature] for 20-60 minutes

Step (d00) : Preparation of a "LER positive control" and an additional water control

1. (Step (d00) does not necessarily need to be performed after step (d). Step (d00) can be performed at any time, as long as the "LER positive control" and another water control are prepared at the beginning of the LAL assay.) Preferably, the "LER positive control" (i.e., "positive LER control") is prepared 1 hour before the end of dialysis [e.g., 900 μl of antibody + 100 μl of CSE at a concentration of 50 EU/ml = a final concentration of 5.0 EU/ml]

Prepare additional water controls, for example:

1. Antibody 900μl + 100μl LAL water

2. LAL water 1000 μl

900μl LAL water + 100μl CSE concentration 50EU/ml = final concentration 5.0EU/ml

Shake for 1 hour [e.g., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature]

Dilute the sample 1:10 with endotoxin-free water

Shake for 1 minute [e.g., in a Heidolph Multi Reax shaker at room temperature, high speed (2037 rpm)]

Step (e) : LAL assay

Prepare standards (i.e., those included in the LAL assay kit being used) according to the manufacturer's instructions and start the measurement as follows;

(1) Preparation of LAL reagent (Kinetic-QCL ^™ reagent):

• Reconstitute the co-lyophilized mixture of horseshoe crab (Limulus polyphemus) amebocyte lysate and chromogenic substrate with 2.6 ml LAL Reagent Water per vial immediately prior to use.

(2) Preparation of CSE stock solution (50 EU/ml, equivalent to standard S1):

• Reconstitute the CSE preparation (E. coli O55:B5-LPS, containing 50-200 EU of lyophilized endotoxin per vial) in the volume of LAL reagent water stated on the certificate of analysis, calculated to yield a solution containing 50 EU (or IU)/ml.

Vigorously shake the CSE stock solution on a shaker at high speed for at least 15 minutes.

Before use, warm the solution to room temperature and shake vigorously on a shaker at high speed for 15 minutes.

(3) Preparation of CSE standard series:

Dilute the CSE stock solution/standard S1 (step 1) with LAL reagent water at room temperature at a ratio of 1:10 to obtain a complete series of CSE standards (50, 5, 0.5, 0.05 and 0.005 EU/ml).

(4) LAL analysis in 96-well microplate ELISA format:

Carefully dispense 100 μl of LAL Reagent Water Blank, Endotoxin Standard, Product Sample, and Positive Product Control into the appropriate wells of the microplate.

Place the filled plate into the microplate reader and close the lid.

Pre-incubate the plate at 37°C ± 1°C for ≥ 10 minutes.

Using an 8-channel multifunctional pipette, dispense 100 μl of Kinetic-QCL ^™ reagent into all wells of the microplate, starting with the first column (A1-H1) and proceeding sequentially to the last column in use. Add the reagent as quickly as possible (avoid bubbles).

Immediately click the OK button on your computer keyboard to start the test.

(Note: Kinetic-QCL ^™ assays are performed with the microplate cover removed)

As used herein, "incorporate" means "add" or "provide." For example, "incorporating a known amount of CSE into a sample" means "adding a known amount of CSE to the sample" or "providing a known amount of CSE to the sample."

Endotoxins, also known as lipopolysaccharides (LPS), are macromolecules found in the outer membrane of Gram-negative bacteria that elicit a strong immune response in animals (e.g., humans). As described above, the present invention provides a method for determining (i.e., detecting and quantifying) bacterial endotoxins in a sample containing antibodies, wherein the method comprises steps (a) to (d) as described herein (preferably further comprising steps (a00), (a01), and (d00)).

In one embodiment, endotoxin can be E. coli endotoxin. Therefore, the endotoxin measured (i.e., detected and/or quantified) in step (d) of the endotoxin determination method provided herein can be E. coli endotoxin. For example, the endotoxin mixed into the sample during the LAL determination test can be E. coli endotoxin (i.e., endotoxin purified from E. coli). Preferably, the endotoxin is commercially available E. coli endotoxin (e.g., control standard endotoxin, CSE).

The WHO International Standard Endotoxin (I.S.) is an endotoxin preparation from Escherichia coli O113:H10:K- that is internationally recognized as the definitive calibrator for bacterial endotoxin testing. The current batch of the International Standard is known as the "Endotoxin WHO International Standard, 3rd I.S."

Reference Standard Endotoxin (RSE) is an endotoxin preparation that has been calibrated against the WHO International Standard Endotoxin. RSEs are established by national agencies (e.g. USP, EP, JP, ChP) and are provided for calibration of CSEs used in LAL assays (see below).

Control standard endotoxin (CSE) is a non-RSE endotoxin preparation that has been calibrated against RSE. CSE is a vendor-specific, highly purified endotoxin preparation produced from Escherichia coli O113:H10:K- (e.g., Associates of Cape Cod, Inc.) or other E. coli strains such as E. coli O55:B5 (e.g., Charles River, Lonza). Stabilizers such as human serum albumin, PEG, or starch may be added at the vendor's discretion. CSE is supplied in varying concentrations depending on its intended use.

The methods provided herein for determining (i.e., detecting and/or quantifying) endotoxins in a sample comprising an antibody; or the methods provided herein for preparing a sample comprising an antibody, have the beneficial effect of eliminating the LER effect in the LAL assay. Thus, one aspect of the present invention relates to the use of the sample preparation method provided herein or the endotoxin assay method provided herein for overcoming the LER effect when determining bacterial endotoxins in the LAL assay.

More specifically, the sample preparation methods provided herein or the endotoxin assay methods provided herein have the following advantageous effects: these methods render samples containing antibodies that exhibit the LER effect reactive toward Factor C in the LAL enzymatic cascade. Therefore, one aspect of the present invention relates to the use of the sample preparation methods provided herein or the endotoxin assay methods provided herein for rendering samples containing antibodies that exhibit the LER effect reactive toward Factor C in the LAL enzymatic cascade.

Here, the term "determination", in particular in "determination of bacterial endotoxins" or grammatical variants thereof, relates to the detection and/or quantification of endotoxins, preferably to the detection and quantification of endotoxins. In the context of the present invention, endotoxins (e.g. E. coli endotoxins such as CSE) are preferably determined by the LAL assay.

The terms "bacterial endotoxins test" or "bacterialendotoxin test" are used interchangeably herein to refer to a group of tests for the detection or quantification of endotoxins from Gram-negative bacteria. BET describes the pharmacopeial (i.e., related to a compendium that serves as a standard, such as the European or US Pharmacopoeia, or other national or international pharmaceutical standards) LAL assay (Limulus Amebocyte Lysate Assay). Furthermore, in the context of the present invention, it is preferred that the pH of the sample to be tested during the LAL assay has a pH of 5.7-8.0, preferably 6.0-8.0, more preferably 6.5-7.5.

The term "LAL assay" is well known in the art and is an in vitro endotoxin test for parenteral drugs, biological products, and medical devices in humans and animals. Specifically, the LAL assay is a test that uses amoebocyte lysates from horseshoe crabs (Limulus polyphemus or Tachypleus tridentatus) to detect and quantify endotoxins from Gram-negative bacteria. For example, during bacterial cell proliferation, cell division, vegetative dieback, and cell lysis, LPS molecules are released from the bacterial cell surface in a rather uncontrolled and nonspecific manner. The released LPS is a potent bacterial toxin and is primarily responsible for the toxic manifestations and deleterious effects (e.g., hyperthermia, hypotension, and irreversible shock) of severe infections with Gram-negative bacteria (Rietschel, 1994, FASEB J. 8: 217-225). The lipid A component is responsible for this biological activity of LPS. In dilute saline solutions, LPS forms macromolecular aggregates (micelles). The formation, size and dynamics of these micelles are related to the LPS concentration, various physicochemical parameters such as temperature, buffer concentration (ionic strength) and pH, and the structure of the O-chain (core-oligosaccharide of lipid A) (Aurell, 1998, Biochem. Biophys. Res. Comm. 253: 119-123). The lipid A portion of LPS is highly conserved among all Gram-negative bacteria and is the portion of the LPS molecule recognized by the LAL assay, making this test the gold standard and suitable method for studying endotoxin contamination from a wide range of Gram-negative bacterial sources (Takada (1988) Eur. J. Biochem; 175: 573-80).

The principle of the LAL assay is described below. In the LAL assay, LPS is detected by the gelation of LAL. This LAL-activated LPS activity is affected by several factors:

- Formation of LPS-LPS aggregates [Akama, 1984, "Bacterial Endotoxin" (J.Y.Homma, S.Kanegasaki, O.Lüderitz, T.Shiba and O.Westphal eds.), Chemie Publishers)

- Formation of protein-LPS aggregates, for example with human lipoprotein Apo A 1, lysozyme, ribonuclease A or human IgG (Emancipator, 1992, Infect Immun. 60: 596-601; Petsch, 1998, Anal. Biochem. 259: 42-47)

- Method for extracting LPS from bacterial cells [Galanos, 1984, "Bacterial Endotoxin" (edited by J.Y.Homma, S.Kanegasaki, O.Lüderitz, T.Shiba and O.Westphal), published by Chemie]

- Bacterial species; LAL activation activity varies 1000-fold within the Enterobacteriaceae [Niwa, 1984, "Bacterial Endotoxin" (J.Y.Homma, S.Kanegasaki, O.Lüderitz, T.Shiba and O.Westphal eds.), Chemie Publishers].

The LAL assay is harmonized between the pharmacopoeias of the United States (US), Europe (EP), and Japan (JP). In the harmonized pharmacopoeia chapters (USP<85>, Ph.Eur.2.6.14., and JP 4.01), three techniques for the LAL assay are described:

-Gel technology (based on endotoxin-induced gelation)

- Turbidity technology (based on gelation-induced turbidity)

- Chromogenic technology (based on coloration following decomposition of synthetic peptide-chromophore complexes).

These three techniques are used in six different methods:

Method A: Gel method, limit test

Method B: Gel method, semi-quantitative test

Method C: Dynamic Turbidity Method

Method D: Dynamic color development method

Method E: Chromogenic endpoint method

Method F: Turbidity Endpoint Method

According to Ph.Eur./USP/JP, these six methods are considered equivalent.

A preferred aspect of the present invention relates to the sample preparation method provided herein and the endotoxin determination method provided herein, wherein a dynamic color development method or a dynamic turbidity method is used to determine the bacterial endotoxin in the sample. Most preferably, a dynamic color development method is used in the sample preparation method and the endotoxin determination method provided herein. By using this technology, endotoxin can be detected by photometry. This technology is such an assay test in which the chromophore released from a chromogenic substrate (i.e., a suitable chromogenic peptide) by the reaction of endotoxin with LAL is measured. The dynamic color development assay can be a method in which the time (onset time) or color development rate required to reach a predetermined absorbance of the reaction mixture is measured. The test is performed at an incubation temperature (usually 37 ± 1 ° C) recommended by the lysate manufacturer. For example, in order to perform a dynamic color development LAL assay, the sample can be mixed with a reagent comprising LAL and a chromogenic substrate (i.e., a suitable chromogenic peptide such as Ac-Ile-Glu-Ala-Arg-pNA) and placed in an incubated plate reader. The color (e.g., yellow) that appears is then monitored over time. The time required before the color appears (reaction time) is inversely proportional to the amount of endotoxin present. That is, the reaction occurs rapidly when a large amount of endotoxin is present; the reaction time increases when a smaller amount of endotoxin is present. The concentration of endotoxin in the unknown sample can be calculated from the standard curve. During the LAL assay, i.e., in step (d) of the endotoxin assay method provided herein, quantification of endotoxin is preferably performed using a standard calibration curve that covers a range of at least two orders of magnitude (in one aspect of the invention, 0.005, 0.05, 0.5, 5.0, and 50.0 EU/ml).

For example, during the kinetic color development LAL technique, the following reaction may occur. Gram-negative bacterial endotoxins catalyze the activation of the zymogen in LAL. The initial activation rate is determined by the endotoxin concentration present. The activated enzyme catalyzes the separation of p-nitroaniline (pNA) from the colorless substrate Ac-Ile-Glu-Ala-Arg-pNA. The released pNA is measured continuously at 405nm by photometry during the entire incubation period. The endotoxin concentration in the sample is calculated from the reaction time by comparing the reaction time with a solution containing a known amount of endotoxin standard. For the LAL assay, the kit "Limulus Amoebocyte Lysate (LAL) Kinetic-QCL ^™ " (Cat. Nos.: 50-650U, 50-650NV, 50-650H; K50-643L, K50-643U) from LONZA can be used according to the manufacturer's instructions. When performing the LAL assay, the use of endotoxins contained in the kit used can be considered (e.g., E. coli O55:B5 endotoxin contained in the kit "Limulus Amoebocyte Lysate (LAL) Kinetic-QCL ^™ " from LONZA (Cat. Nos. 50-650U, 50-650NV, 50-650H; K50-643L, K50-643U)).

In the context of the present invention, the pH value of the sample to be tested during the LAL assay is preferably between 5.7 and 9.0. More preferably, the pH value of the sample to be tested during the LAL assay is between 5.8 and 8.0, even more preferably between 5.8 and 7.5, and even more preferably between 5.8 and 7.0. Most preferably, the pH value of the sample to be tested during the LAL assay is between 5.8 and 7.0. For example, the pH value of the sample to be tested during the LAL assay can be pH 5.7, pH 5.8, pH 5.9, pH 6.0, pH 6.1, pH 6.2, pH 6.3, pH 6.4, pH 6.5, pH 6.6, pH 6.7, pH 6.8, pH 6.9, or pH 7.0. Therefore, it is envisaged that, in the context of the present invention, the pH of the solution to be tested (i.e., dissolved solid sample or liquid sample) is adjusted to between pH 5.7 and pH 8.0, more preferably to between pH 5.8 and pH 7.0, before the LAL assay is performed. If necessary, the pH value is adjusted, for example, by dilution, addition of buffer and/or neutralization.

Some substances (such as β-glucan) interfere with the LAL test to some extent (with the obvious exception of water samples). Interference can inhibit or enhance the LAL assay. In particular, interfering factors can enhance or weaken the LPS quantification obtained from the LAL test, and therefore enhance or weaken the quantification of endotoxins. Therefore, if the recovery of PPC in step (d) of the endotoxin determination method provided herein is not within the acceptable range of 50-200%, the interfering factors must be eliminated. This can be accomplished by diluting the sample in step (b) of the method provided herein. In particular, the sample can be diluted with endotoxin-free water or endotoxin-free buffer (preferably Tris/HCl buffer, pH ~ 7.0). The lowest sample dilution (highest product concentration) without inhibition/enhancement is called the "non-interfering concentration (NIC)". However, during sample dilution, the MVD (maximum effective dilution = the maximum possible dilution of the sample that can determine the endotoxin limit) must not be exceeded. In particular, based on the test results of different batches, a sample dilution (validated sample dilution or sample concentration) that covers all batches is selected. Or in other words, a sample dilution is selected in step (b) of the methods provided herein that results in 50-200% recovery in the PPC. To establish that the selected treatment can effectively eliminate interference without losing endotoxin (i.e., does not show a LER effect), an "interference factor test" can be performed by using a sample spiked with a defined concentration of endotoxin (i.e., PPC).

Thus, one aspect of the invention relates to an endotoxin assay as provided herein, wherein PPC is prepared and tested for endotoxin in step (d) of the endotoxin assay as provided herein. If the recovery of the control standard spiked with endotoxin reaches 50-200%, the sample is free of interfering factors.

Since BET according to USP/Ph.Eur./JP includes an internal control (PPC) which allows individual evaluation of each test result, method validation of BET according to USP/Ph.Eur./JP is not a prerequisite for correct endotoxin results.

The terms "low endotoxin recovery (LER)" or "LER effect" are known in the art and describe the endotoxin shielding specifically caused by a combination of polysorbate plus citrate or phosphate (Chen, J. and Williams, K.L., PDA Letter 10, 2013, 14-16). Endotoxin shielding can also be caused by any other buffer component or combination thereof. In order to identify a given substance (e.g., a buffer or therapeutic antibody sample) as exhibiting the LER effect, the endotoxin content can be monitored over time, for example, in an endotoxin retention time study. An endotoxin retention time study requires that endotoxin be spiked into an undiluted sample and the spiked sample be stored over time. For example, the sample can be stored for up to several days. Preferably, in a retention time study, the spiked sample is stored for several days (e.g., 7 to up to 28 days) and the LAL assay is performed at the specified time points. A recovery of less than 50% of the amount of spiked endotoxin indicates that the sample exhibits the LER effect. If endotoxin recovery is less than 50%, but occurs only at any of the intermediate time points and not at the terminal time point, the test sample cannot be considered to exhibit a shielding effect.

During the past few years, the FDA has recognized the LER phenomenon and issued guidance (see Hughes, P., et al., BioPharm.Asia March/April 2015, 14-25). These guidances define that once a certain amount of CSE is spiked into an undiluted sample, the acceptable limit of endotoxin recovery in a drug sample is between 50 and 200% (e.g., 5.0 EU/ml=100%). If a sample to be tested shows the LER effect, the recovery of the spiked endotoxin is less than 50% of the total amount of endotoxin spiked in.

In the methods of the present invention provided herein, the sample comprises an antibody, preferably a monoclonal antibody. The terms "sample", "test sample", "sample comprising an antibody" and "sample comprising an antibody to be tested" are used interchangeably herein and refer to a certain amount of liquid comprising an antibody to be tested for the presence and/or amount of endotoxin. In other words, the terms "sample", "test sample", "sample comprising an antibody" and "sample comprising an antibody to be tested" are used interchangeably herein and relate to a liquid to be tested for the presence and/or amount (preferably the presence and amount) of endotoxin, wherein the liquid comprises an antibody. The "sample comprising an antibody to be tested" is preferably a sample of a therapeutic antibody. The term "therapeutic antibody" relates to any antibody preparation intended for use in humans. Antibodies (e.g., therapeutic antibodies) are preferably formulated with polysorbate 80 or sodium citrate buffer, more preferably with polysorbate 80 and sodium citrate buffer. Most preferably, the antibody is formulated with about 25mM sodium citrate buffer and about 700mg/L polysorbate 80 and has a pH of about 6.5. In the context of the present invention, it is preferred that the antibody (e.g. therapeutic antibody) is a monoclonal antibody. Most preferably, the antibody (e.g. therapeutic antibody) is the anti-CD20 antibody rituximab. Thus, in the context of the present invention, the sample may be a sample of. It is envisaged that in the context of the present invention, the sample to be tested (i.e. the sample comprising the antibody to be tested) shows/displays the LER effect.

Here, the term "antibody" is used in the broadest sense and includes, in particular, intact monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies) formed from at least two intact antibodies, and antibody fragments, as long as they exhibit the desired biological activity. Human antibodies, humanized antibodies, camelized antibodies, or CDR-grafted antibodies are also included.

As used herein, the term "monoclonal antibody" refers to the antibody obtained from a substantially homogeneous antibody group, that is, except for the possible naturally occurring mutation present in a small amount, the single antibody of the antibody group is identical. Monoclonal antibodies are highly specific, for a single antigenic site. In addition, in contrast to the polyclonal antibody preparations comprising different antibodies for different determinants (epitopes), each monoclonal antibody is for a single determinant on the antigen. In addition to their specificity, the advantage of monoclonal antibodies is that they can be synthesized without being contaminated by other antibodies. The modifier "monoclonal" represents the feature of the antibody obtained from a substantially homogeneous antibody group, and should not be construed as needing to produce the antibody by any ad hoc method. For example, the monoclonal antibody included in the sample of the inventive method can be prepared by the hybridoma method described by Kohler, G. et al., Nature 256 (1975) 495, or can be prepared by a recombinant DNA method (see, for example, U.S. Patent number 4,816,567).

The monoclonal antibodies described herein are preferably produced by expression in host cells, most preferably Chinese hamster ovary (CHO) cells. For production, an isolated nucleic acid encoding an antibody (encoding an amino acid sequence comprising the VL of the antibody and/or an amino acid sequence comprising the VH of the antibody (e.g., a light chain and/or a heavy chain of the antibody)) is inserted into one or more vectors (e.g., an expression vector). These vectors are introduced into a host cell. The host cell comprises (e.g., has been transformed): (1) a vector comprising a nucleic acid encoding an amino acid sequence comprising the VL of the antibody and an amino acid sequence comprising the VH of the antibody, or (2) a first vector comprising a nucleic acid encoding an amino acid sequence comprising the VL of the antibody and a second vector comprising a nucleic acid encoding an amino acid sequence comprising the VH of the antibody. The host cell can be a eukaryotic cell, e.g., a Chinese hamster ovary (CHO) cell or a lymphoid cell (e.g., a Y0, NS0, Sp20 cell).

For recombinant production of antibodies, nucleic acids encoding antibodies (e.g., as described above) are isolated and inserted into one or more vectors for further cloning and/or expression in a host cell. Such nucleic acids can be easily isolated and sequenced using conventional procedures (e.g., by using oligonucleotide probes that can specifically bind to genes encoding the heavy and light chains of the antibody).

Suitable host cells for cloning or expressing vectors encoding antibodies include prokaryotic or eukaryotic cells as described herein. For example, antibodies can be produced in bacteria, particularly when glycosylation and Fc effector functions are not required. For expression of antibody fragments and polypeptides in bacteria, see, for example, US 5,648,237, US 5,789,199, and US 5,840,523. (See also Charlton, K.A., In: Methods in Molecular Biology, Vol. 248, Lo, B.K.C. (ed.), Humana Press, Totowa, NJ (2003), pp. 245-254, which describes expression of antibody fragments in E. coli). After expression, the antibody can be isolated from the bacterial cell paste as a soluble fraction and can be further purified.

In addition to prokaryotes, eukaryotic microorganisms such as filamentous fungi or yeast are suitable cloning or expression hosts for antibody-encoding vectors, including fungi and yeast strains whose glycosylation pathways have been "humanized" to produce antibodies with partially or fully human glycosylation patterns. See Gerngross, T.U., Nat. Biotech. 22 (2004) 1409-1414; and Li, H. et al., Nat. Biotech. 24 (2006) 210-215.

Suitable host cells for expressing glycosylated antibodies are also derived from multicellular organisms (invertebrates and vertebrates). Examples of invertebrate cells include plant and insect cells. Many baculovirus strains have been identified that can be used in conjunction with insect cells, particularly for transfecting Spodoptera frugiperda cells.

Plant cell cultures can also be used as hosts. See, for example, US 5,959,177, US 6,040,498, US 6,420,548, US 7,125,978, and US 6,417,429 (describing PLANTIBODIES ^™ technology for producing antibodies in transgenic plants).

Vertebrate cells can also be used as hosts. For example, suspension-adapted mammalian cell lines can be used. Other examples of useful mammalian host cell lines are SV40-transformed monkey kidney CV1 line (COS-7); human embryonic kidney line (293 or 293 cells, e.g., as described in Graham, FL et al., J. Gen Virol. 36 (1977) 59-74); baby hamster kidney cells (BHK); mouse Sertoli cells (TM4 cells, e.g., as described in Mather, JP, Biol. Reprod. 23 (1980) 243-252); monkey kidney cells (CV1); African green monkey kidney cells (VERO-76); human cervical carcinoma cells (HELA); canine kidney cells (MDCK); Buffalo rat liver cells (BRL3A); human lung cells (W138); human liver cells (HepG2); mouse mammary tumor (MMT060562); TRI cells, e.g., as described in Mather, JP et al., Annals NY Acad. Sci. 383 (1982) 44-68; MRC 5 cells; and FS4 cells. Other useful mammalian host cell lines include Chinese hamster ovary (CHO) cells, including DHFR ^- CHO cells (Urlaub, G. et al., Proc. Natl. Acad. Sci. USA 77 (1980) 4216-4220); and myeloma cell lines, such as Y0, NS0, and Sp2/0. For a review of certain mammalian host cell lines suitable for antibody production, see, for example, Yazaki, P. and Wu, AM, Methods in Molecular Biology, Vol. 248, Lo, BKC (ed.), Humana Press, Totowa, NJ (2004), pp. 255-268.

"Antibody fragments" include a portion of an intact antibody. The term "antibody fragment" includes an antigen-binding portion that retains the ability to bind to an antigen (such as CD20), i.e., an "antigen-binding site" (e.g., a fragment, a subsequence, a complementary determining region (CDR)), which comprises or alternatively consists of: for example, (i) a Fab fragment, a monovalent fragment consisting of a VL, VH, CL, and CH1 domain; (ii) a F(ab')2 fragment, a bivalent fragment comprising two Fab fragments linked by a disulfide bond at the hinge region; (iii) an Fd fragment consisting of a VH and CH1 domain; (iv) an Fv fragment consisting of the VL and VH domains of a single arm of an antibody, (v) a dAb fragment consisting of a VH domain (Ward; 1989; Nature 341; 544-546) and (vi) an isolated complementary determining region (CDR). Antibody fragments or derivatives also include F(ab')2, Fv or scFv fragments or single-chain antibodies.

Preferably, in the methods provided herein, the antibody (ie, the antibody contained in the sample) is rituximab.

The term "rituximab" (trade name) relates to a chimeric monoclonal antibody directed against the protein CD20. CD20 is found on the surface of cancerous and normal B cells. Rituximab destroys B cells and is therefore used, for example, to treat diseases characterized by excess B cells, overactive B cells, or dysfunctional B cells. This includes many lymphomas, leukemias, transplant rejection, and autoimmune diseases. For example, rituximab is used as a subcutaneous formulation for chronic lymphocytic leukemia. However, rituximab is usually administered by intravenous infusion. Stem cells in the bone marrow do not have the CD20 protein, allowing the B cell population to be restored after rituximab treatment. As used herein, the term "rituximab" also encompasses all anti-CD20 antibodies or anti-CD20 antibody fragments that meet the requirements for marketing authorization in a country or region selected from the United States, Europe, and Japan. Most preferably, the term "rituximab" refers to an antibody having heavy and light chain amino acid sequences as shown in SEQ ID NOs: 1 and 2, respectively. Those skilled in the art will readily know how to obtain the encoding nucleic acid sequence from a given amino acid sequence. Therefore, based on the knowledge of SEQ ID NOs: 1 and 2, the encoding nucleic acid sequence of rituximab can be readily obtained.

The trade name refers to a pharmaceutical preparation that contains epoetin beta as the active ingredient. Epoetin beta is a synthetic version of the naturally occurring hormone erythropoietin. Erythropoietin is produced by healthy kidneys and stimulates the bone marrow to produce red blood cells, which carry oxygen around the body. Epoetin beta is also used to treat symptomatic anemia in people who are undergoing chemotherapy for certain types of cancer. One of the side effects of chemotherapy is that it kills healthy blood cells along with the cancer cells. Injections of epoetin can increase the production of red blood cells and help relieve the symptoms of anemia. Because epoetin increases the production of blood cells, larger amounts of blood can be collected from a person receiving epoetin, and this blood can be stored for transfusion during or after surgery.

In the step (d) of the sample preparation or endotoxin assay method provided herein, the sample (i.e., the sample comprising the antibody) is dialyzed against an endotoxin-free aqueous solution, wherein the sample has a pH value between pH 5.7 and pH 8.0 (preferably between pH 6.0 and 8.0, more preferably between 6.5 and 7.5). In biochemistry, dialysis is a common method for separating molecules in solution based on the difference in diffusion rates of molecules through semipermeable membranes (such as dialysis tubing). Dialysis is a commonly used laboratory technique, and its operating principle is the same as medical dialysis. In life science research, the most common application of dialysis is to remove unwanted small molecules, such as salts, reducing agents, or dyes, from larger macromolecules such as antibodies. Dialysis is also commonly used in buffer exchange and drug binding studies.

Diffusion is the random thermal motion (Brownian motion) of molecules in solution, resulting in the net movement of molecules from higher concentration areas to lower concentration areas, until equilibrium is reached. In dialysis, sample and buffer solution (called dialysate) are separated by a semipermeable membrane, resulting in differential diffusion patterns, thereby allowing the separation of molecules in sample and dialysate. Due to the pore size of the membrane, the macromolecules (such as antibodies) in the sample cannot pass through the membrane, thereby limiting their diffusion from the sample chamber. On the contrary, small molecules (such as components of sodium citrate buffer) will freely diffuse through the membrane and obtain equilibrium in the entire solution volume, thereby changing the total concentration of these molecules in sample and dialysate. Once equilibrium is reached, the final concentration of the molecule depends on the volume of the solution involved, and if the dialysate after equilibrium is replaced (or exchanged) with fresh dialysate (see step below), diffusion will further reduce the concentration of small molecules in the sample.

For example, the following dialysis step can be used to remove sodium citrate buffer from a sample (i.e., from a sample containing antibodies):

1. Obtain and wash a membrane with a molecular weight cut-off of 10 kDa

2. Load the sample into the dialysis tube, cassette or device

3. Place the sample into the outer chamber with dialysate (stirred buffer)

4. Dialyze for 24 hours at room temperature; change the water twice during the 24 hours

By using an appropriate volume of dialysate and multiple exchanges of buffer, the concentration of sodium citrate buffer in the sample can be reduced to negligible levels (ie, 1-2% of the original content).

The present invention is further described with reference to the following non-limiting figures and examples. In the figures and examples, most experiments described are indicated by designated numbers. For example, the designation [Rituximab 117] refers to an experiment conducted with formulated rituximab and/or formulated rituximab placebo, and the experiment is designated by the reference number "117."

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 Time dependence of dialysis with phosphate and polysorbate 20 using a MWCO of 12-16 kDa. The weight of the internal dialysate (lyophilized and weighed) obtained after the indicated dialysis time at room temperature is shown. The data at the top of the gray bar show the average amount (%) of two measurements.

FIG2 shows dialysis of a 12-16 kDa MWCO membrane containing phosphate and polysorbate 20 using a 12-16 kDa MWCO membrane with or without (w/o) bovine serum albumin (BSA) treatment prior to dialysis. The phosphate (P) content in the internal dialysate obtained after the indicated time is shown. The left bar shows the P amount when the membrane was treated with 0.2% BSA prior to dialysis. The right bar corresponds to the P amount without BSA treatment. Photometric determination of phosphate recovered from the internal dialysate was performed according to Strominger (1959, J. Biol. Chem. 234: 3263-3267).

Figure 3: Recovery (%) of [Rituximab 115] and [Rituximab 117] using (A) rituximab and (B) rituximab placebo as samples, following the protocol described in Example 2.1 to overcome the LER effect. In the figure, "fast spin" and "slow spin" refer to the frequency of the stirrer (i.e., "fast spin" means the stirrer frequency is high). This exemplary protocol can also be used for routine quality control of other samples, preferably for quality control of samples containing sodium citrate buffer and polysorbate 80 as a detergent.

Figure 4: Schematic diagram of a modified protocol for overcoming the LER effect and the recoveries obtained by performing said protocol. (A) Schematic diagram of a protocol according to the present invention for overcoming the LER effect (e.g. in rituximab and rituximab placebo). The detailed protocol is described in Example 2.2. This exemplary protocol can also be used for routine quality control of other samples, preferably for samples containing sodium citrate buffer and polysorbate 80 as detergents. (B) [Rituximab 046] The recoveries (%) of rituximab and rituximab placebo obtained by the LER assay after performing the protocol according to Figure 4(A). For more details, please refer to the protocol described in Example 2.2.

Figure 5: [Rituximab 059] Recovery (%) obtained by carrying out the protocol described in Reference Example 2 using rituximab as a sample.

Figure 6: Recovery (%) obtained by carrying out the protocol described in Reference Example 3 using rituximab as sample. Recovery [Rituximab 061] obtained by carrying out the protocol described in Reference Example 2.2.

Figure 7: Recovery (%) obtained by performing the protocol described in Reference Example 4 using rituximab as a sample. (A) Recovery obtained by performing the protocol described in Reference Example 3.1 [Rituximab 062]. (B) Recovery obtained by performing the protocol described in Reference Example 3.2 [Rituximab 063].

Figure 8: Recovery (%) obtained by performing the protocol described in Reference Example 5 using rituximab as a sample. (A) Recovery obtained by performing the protocol described in Reference Example 4.1 [Rituximab 064]. (B) Recovery obtained by performing the protocol described in Reference Example 4.2 [Rituximab 065].

Figure 9: Recovery (%) obtained by carrying out the protocol described in Reference Example 6 using rituximab and rituximab placebo as samples [Rituximab 072].

Figure 10: Recovery (%) obtained using rituximab and rituximab placebo as samples in the protocol described in Reference Example 7. (A) [Rituximab 079] no incubation; (B) [Rituximab 080] incubation for 4 hours; (C) [Rituximab 081] incubation for 1 day; (D) [Rituximab 082] incubation for 3 days.

Figure 11: Recovery (%) obtained by performing the LAL assay described in Reference Example 8 using rituximab as a sample. (A) [Rituximab 002] LAL assay with different dilutions; (B) [Rituximab 004] Comparison of Lonza and ACC CSE incorporation; (C) [Rituximab 005] LAL assay with different dilutions and pH adjustments.

Figure 12: Recovery (%) obtained using rituximab as a sample in the LAL assay as described in Reference Example 8. Dialysis and dilution alone were not able to overcome the LER effect [Rituximab 011].

Figure 13: Time dependence of the LER effect. This figure shows the recovery (%) obtained by performing the incorporation and LAL assay described in Reference Example 1 using rituximab as a sample [Rituximab 027]. The shaking time after incorporation is shown (i.e., 2 seconds to 60 minutes).

Figure 14: Importance of the buffer system on the LER effect. Shown are the recoveries (%) obtained from the LAL assay performed as described in Reference Example 10. (A) LAL assay using rituximab or sodium citrate as the sample and diluted at a ratio of 1:2, 1:5, 1:10, or 1:20 [Rituximab 006]; (B) LAL assay using sodium citrate, polysorbate 80, or sodium citrate and polysorbate 80 as the sample and diluted at a ratio of 1:2, 1:5, or 1:10 [Rituximab 029].

Figure 15: Effect of _MgCl₂ on the LER effect. Shown are the recoveries (%) obtained from the LAL assay described in Reference Example 13. (A) _MgCl₂ was added to a concentration of 10 mM [Rituximab 030]; (B) _MgCl₂ was added to a concentration of 50 mM [Rituximab 031]; (C) MgCl₂ was added to a concentration of ₂₅ mM [Rituximab 032]; (D) _MgCl₂ was added to a concentration of 75 mM [Rituximab 033].

Figure 16: Effect of mechanical treatment on the LER effect. Shows the recovery (%) obtained by performing the LAL assay described in Reference Example 14 [Rituximab 034]. In the figure, "shaking" refers to shaking for 60 minutes.

The following examples are provided to aid the understanding of the present invention, the true scope of the invention being set forth in the appended claims. It will be appreciated that modifications may be made to the procedures described without departing from the spirit of the invention.

Example 1: Technical Equipment and Reagents

1. Technical equipment

1.1 Microplate Reading System (also referred to as "Reader" in this article)

PRO, multi-mode microplate reader; Tecan Switzerland/Tecan Deutschland GmbH, Germany, P/N: 30050303.

Magellan V.7.1 software

Costar ^™ cell culture plates, 96-well, Fisher Scientific, P/N: 07-200-89.

1.2 Shaker System and Glass Vials

Multi Reax; Heidolph, Germany, P/N: 545-10000-00.

1.5 ml screw-neck glass bottle (N8); Macherey-Nagel GmbH & Co. KG, Germany, P/N: 702004 (Qty. 100).

N 8PP screw cap, black, closed top; Macherey-Nagel GmbH & Co. KG, Germany, P/N: 70250 (Qty. 100).

4 ml screw-neck glass bottle (N13); Macherey-Nagel GmbH & Co. KG, Germany, P/N: 702962 (Qty. 100).

N 13PP screw cap, black, closed top; Macherey-Nagel GmbH & Co. KG, Germany, P/N: 702051 (Qty. 100).

1.3 Dialysis Equipment

Rotary Dialyzer ^™ , chamber volume 1000 μl; Harvard Apparatus, USA, P/N 740314 (Qty. 1) and 740306 (Qty. 5), local distributor: Hugo Sachs Elektronik Harvard Apparatus, GmbH, Germany, P/N SP1 74-0306 (Qty. 5). Note: Batch number used: 032613.

The dialyzer is a simple, single-sided device used to dialyze biological samples. A wide range of dialyzer sizes are available to accommodate sample volumes from 20 μl to 5 ml. For 1 ml (as used herein), catalog number Nb is 74-0314. Membrane MWCO ranges from 100 to 300,000 Da. The entire unit is made of PTFE, a virtually non-reactive material.

• Rapid spin dialyzer, chamber volume 1000 μl, Harvard Apparatus, U.S.A., P/N 740510 (Qty. 1) or 740504 (Qty. 5). Note: Double-sided membrane system, top and bottom membranes.

The dialyzer is a reusable sample chamber made of PTFE for high sample recovery and has been redesigned to provide a larger membrane surface area for even faster dialysis rates. The volume of the Ultra-Fast Dialyzers ranges from 50 μl to 1500 μl, with 1000 μl used in this article. For 1 ml (as used herein), catalog No. 74-0412.

Cellulose acetate membrane, 500 Da MWCO, Harvard Apparatus, U.S.A., P/N: SP1 7425-CA500, local distributor: Hugo Sachs Elektronik Harvard Apparatus GmbH, Germany, P/N: SP17425-CA500.

Cellulose acetate membrane, 10 kDa MWCO, Harvard Apparatus, U.S.A., P/N: SP1 7425-CA10K, local distributor: Hugo Sachs Elektronik Harvard Apparatus GmbH, Germany, P/N: SP17425-CA10K.

Note: This membrane was tested in addition to a “standard” 500Da MWCO membrane in the LER studies with rituximab and in LER experiments with .

Cellulose acetate membrane, 25 kDa MWCO, Harvard Apparatus, U.S.A., P/N: SP17425-CA25K, local distributor: Hugo Sachs Elektronik Harvard Apparatus GmbH, Germany, P/N: SP1 7425-CA25K.

Note: This membrane was tested in addition to a “standard” 500Da MWCO membrane in the LER studies with rituximab and in LER experiments with .

Aqua B.Braun, sterile, pyrogen-free water, 1 liter, B.Braun Melsungen AG, Germany, P/N: 14090586.

Crystallizing dish, 900 ml, OMNILAB, Germany, P/N: 5144008.

(Note: used for rinsing dialysis membrane)

Beaker, tall type, 2000 ml, OMNILAB, Germany, P/N: 5013163.

Beaker, tall, 250 ml, OMNILAB, Germany, P/N: 5013136.

1.4 Conventional laboratory equipment

High-pressure sterilization system (Note: used for dialysis room disinfection)

LoRetention-Reloads, PCR clean grade, 0.5-10 μl, Eppendorf, Germany, P/N: 0030072.057

LoRetention-Reloads, PCR clean grade, 2-200 μl, Eppendorf, Germany, P/N: 0030072.022

LoRetention-Reloads, PCR clean grade, 50-1000 μl, Eppendorf, Germany, P/N: 0030072.030

Individual, 5 ml, Paper/Plastic Wrap, Fisher Scientific, P/N: 10420201.

2. Reagents

2.1 Kinetic chromogenic LAL assay and LAL-related reagents

Kinetic-QCL ^™ kit; Lonza, Switzerland, P/N: 50-650U or 50-650H (ie, "Lonza kit").

CHROMO-LAL from Associates of Cape Cod (AAC) Inc., USA, P/N: C0031-5 (i.e., “ACC kit”).

• Endotoxin Escherichia coli O55:B5 for K-QCL; Lonza, Switzerland, P/N: E50-643.

Endotoxin Escherichia coli O55:B5, 2.5 mg/vial; Lonza, Switzerland, P/N: N185.

LAL reagent water – 100 ml; Lonza, Switzerland, P/N: W50-100.

• _MgCl2 , 10 mM solution for use with LAL, 30 ml vial; Lonza, Switzerland, P/N: S50-641.

Magnesium chloride hexahydrate for analysis ISO, Reag. Ph. Eur., 250 g; Merck, Germany, P/N: 1.05833.0250.

• Tris buffer, 50 mM solution for use with LAL, 30 ml vial; Lonza, Switzerland, P/N: S50-642.

2.2 Protein reagents

Albumin bovine fraction V, very low endotoxin, fatty acid free, 25 g; Serva, Germany, P/N: 47299.04.

Albumin, human serum, fraction V, high purity; 1 g; Merck, Germany, P/N: 126658-1GM.

3. Drugs tested

For the examples described herein, rituximab (which comprises RIBA®) and RIBA® (which comprises epoetin-β) were used. Additionally, rituximab and respective placebos were also used in the methods described herein.

The placebo for each sample was identical to the sample except that the active therapeutic ingredient was absent, ie, the rituximab placebo did not contain rituximab but contained all other components of the formulation.

Example 2: Method of overcoming the LER effect of the present invention

Example 2.1: Solution to Overcoming the LER Effect

In this example, rituximab and rituximab placebo were used as samples. However, as described below, the protocol described herein can be used to overcome LER in all typical formulations of drug antibodies.

Materials used in this example

-membrane:

10 kDa cellulose acetate (CA) membrane, from Harvard Apparatus, USA, P/N: SP1 7425-CA10K

- Dialysis Machine:

FastSpin DIALYZER, chamber volume 1000 μl, Harvard Apparatus, USA, P/N 740510 (Qty. 1) or 740504 (Qty. 5)

-Sample vials:

1.5 ml screw-neck glass bottle (N8); Macherey-Nagel GmbH & Co. KG, Germany, P/N: 702004

N 8PP nut, black, closed top; Macherey-Nagel GmbH & Co. KG, Germany, P/N: 70250

-Crystallizing dish:

900ml, Duran, VWR Germany, P/N: 216-1817

-MgCl ₂ - stock solution:

1M MgCl2 dissolved in water (magnesium chloride hexahydrate, for analysis ACS, ISO, Reag. Ph. Eur., 250 g; Merck, Germany, P/N: 1.05833.0250)

- Tris-buffered saline, 50 mM solution for use with LAL (ie, endotoxin-free), 30 ml vial; Lonza, Switzerland, P/N: S50-642

-sample:

Rituximab placebo and LAL water

Phased implementation plan :

Step 1 : Prepare the sample

1x rituximab placebo 900 μl + 100 μl LAL water

1x rituximab placebo 900 μl + 100 μl CSE concentration 50 EU/ml = final concentration 5.0 EU/ml

1x LAL water 1000 μl

1x LAL water 900μl + 100μl CSE concentration 50EU/ml = final concentration 5.0EU/ml

Shake the sample for 60 minutes at RT (room temperature) [i.e. in a Heidolph Multi Reax shaker, high speed (2037 rpm)],

Step 2 : Washing the dialysis membrane

Use 10 10 kDa cellulose acetate (CA) membranes and place them in a crystallization dish with 300 ml of Aqua Braun (distilled water from the manufacturer B. Braun, Melsungen),

Shake it for 1 hour (shaker SG 20, IDL GmbH, Germany, or equivalent, 50 to 300 rpm, preferably 100 rpm)

Transfer the membrane to a new crystallizing dish with fresh Aqua Braun (also 300 ml),

Shake it for 1 hour (shaker SG 20, IDL GmbH, Germany, or equivalent, 50 to 300 rpm, preferably 100 rpm)

Step 3 : Add MgCl2 to a final MgCl2 concentration of approximately 50 mM MgCl2

Add 50 μl of 1M _MgCl2 stock solution to the sample in step 1

Shake for 1 minute [i.e., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature]

Incubate samples at room temperature for 60 minutes

Shake for 1 minute [i.e., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature]

Step 4 : Dilution

Take one of the samples from step 3 and dilute it 1:10 with a pH-7 buffer (i.e., 50 mM Tris/HCl buffer pH-7).

o 895 μl 50 mM Tris buffer + 105 μl sample

• Perform the assay twice to replicate (i.e., in duplicate):

o 2x rituximab placebo, 1:10 with Tris buffer

o 2x rituximab placebo 5.0EU/ml, 1:10 with Tris buffer

o 2x LAL water, 1:10 with Tris buffer

o 2x LAL water 5.0EU/ml, 1:10 with Tris buffer

Step 5 : Dialysis

All diluted samples were shaken for 1 minute [i.e., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature],

Transfer to a fast-spin dialyzer

Each beaker (i.e., beaker, high model, 2000 ml, OMNILAB, Germany, P/N: 5013163) was placed on a dialyzer on a magnetic stirrer plate. The stirrer frequency was adjusted to high (i.e., "fast spin"). High stirrer frequency refers to 50-300 rpm, preferably 200-300 rpm. The stirrer was a heat-sterilized (250°C, 4 hours) magnetic stirrer with a length of approximately 40 mm and a diameter of approximately 14 mm.

Pour 200ml of Aqua Braun into the beaker

Dialyze for 24 hours at room temperature (21 ± 2°C), switching to Aqua Braun after 2 and 4 hours.

After dialysis, transfer the sample to a new 1.5 ml screw cap vial

Step 6 : Oscillation

Shake the sample for 20 minutes [i.e., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature]

Step 7 : Prepare a LER positive control (i.e., a positive LER control) and an additional water control

Prepare LER positive control 1 hour before the end of dialysis

1. Rituximab placebo 900 μl + 100 μl LAL water

2. Rituximab placebo 900 μl + 100 μl CSE concentration 50 EU/ml = final concentration 5.0 EU/ml

3. LAL water 1000 μl

4. LAL water 900μl + 100μl CSE concentration 50EU/ml = final concentration 5.0EU/ml

Shake for 1 hour [i.e., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature]

Dilute the sample with LAL water at a ratio of 1:10 (sample:LAL water)

Shake [i.e., in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature] for 1 minute

Step 8 : LAL assay

• Prepare standards according to the manufacturer's instructions (Kinetic-QCL ^™ assay; Lonza) and initiate LAL assay experiments.

Results and Discussion

As can be seen in Figures 3(A) (i.e., [Rituximab 117]) and 3(B) (i.e., [Rituximab 115]), the above method is able to overcome the LER effect. Furthermore, by using this method, the LER effect can be overcome in rituximab as well as rituximab placebo. This indicates that the above protocol is independent of the formulation comprising a specific monoclonal antibody, but can be used to eliminate the LER effect in any formulation comprising polysorbate 80 and a chelating buffer (e.g., sodium citrate). This formulation is typical for antibodies, particularly monoclonal antibodies. Therefore, it is expected that the above method can be used to overcome the LER effect in any antibody formulation.

It has been found that Mg2 ⁺ is the divalent cation of choice for restoring LAL reactivity in formulations containing chelating buffers such as sodium citrate and exhibiting the LER effect.

To remove the chelating buffer (e.g. sodium citrate buffer), a second step (after addition of Mg ²⁺ ) is performed by dialysis. The Harvard Rotary Dialyzer ^™ is the preferred apparatus for dialysis.

Detergents (e.g., polysorbate 80) are a second cause of the LER effect. Typically, the presence of detergents (e.g., polysorbate 80) in biological samples leads to micelle formation when the critical micelle concentration (CMC) of the detergent is reached (usually in the μM range). Micelles can inhibit LPS-mediated activation of factor C (a serine protease that is the first enzyme in the LAL cascade) (Nakamura (1988a) J. Biochem. 103: 370-374). In monoclonal antibody preparations, undiluted samples are typically above the CMC to obtain functional dissolution of the antibody. In the products studied in this article, the CMC of the detergents did exceed their CMC (polysorbate 80: 700 mg/l (50-fold excess)), leading to the speculation that polysorbate 80 exists in micellar form. In the above protocol, the detergent concentration is reduced by dilution, such that the detergent concentration approaches/drops below the CMC value (polysorbate 80: 14 mg/l or 10.6 μM). Dilution of the detergent to a concentration close to the CMC eliminates the compartmentalization of the micelles and thus makes the incorporated CSE molecules accessible to the LAL enzyme.

Thus, the problem of the LER effect (eg in the case of sodium citrate and polysorbate 80 used to formulate drug products) can now be considered solved. In summary, a safe, stable and reproducible test method for drug products is provided herein.

In summary, in rituximab and rituximab placebo, the above regimen surprisingly overcomes the LER effect. In contrast, the same regimen for (which does not comprise an antibody but comprises epoetin-β) cannot reveal satisfactory results, indicating that the methods provided herein are particularly useful for antibody formulations, preferably for formulations with monoclonal antibodies, citrate buffer, and polysorbate 80.

Example 2.2: Modified Scheme (1) for Overcoming the LER Effect

In this example, a modification scheme to overcome the LER effect was used. Compared with Example 2.1, the most important changes are as follows:

1. In Example 2.2, a rotary dialyzer was used. In contrast, in Example 2.1, a fast rotary dialyzer was used, which has a more efficient dialysis chamber and improves the efficiency of dialysis (the membrane is on both sides of the column).

2. In Example 2.2, the MWCO of the dialysis membrane is 500 Da. In contrast, in Example 2.1, the MWCO of the dialysis membrane is 10 kDa.

3. In Example 2.2, the sample was diluted 1:10 with endotoxin-free water. In contrast, in Example 2.1, the sample was diluted 1:10 with Tris buffer, pH ~7 (i.e., Tris/HCl buffer, pH ~7). The pH of the sample was adjusted to approximately 6.0 by diluting the sample 1:10 with endotoxin-free water.

4. In Example 2.2, the dialysis time was 4 hours. In contrast, in Example 2.1, the dialysis time was 24 hours.

In this example, rituximab and rituximab placebo were used as samples. However, for the same reasons discussed in Example 2.1, this approach can be used to overcome LER in all typical antibody formulations.

Specifically, the protocol used in Example 2.2 is detailed below.

Solution Overview

Step 1: "Set up the LER effect" (see also "LER positive control" below):

Rituximab and rituximab placebo samples were spiked with 5 EU/ml or 0.5 EU/ml (CSE; Lonza, E. coli 0055:B5), and the mixtures were shaken at maximum speed for 60 min at room temperature [shaker: Heidolph Multi Reax, high speed (2037 rpm)] to obtain “positive LER effect” samples.

Step 2: Add MgCl ₂ : Before dialysis, add 2M MgCl ₂ stock solution to make the final concentration about 50 mM MgCl ₂ ; shake for 1 minute as in step 1.

Step 3: Dilute 1:10 [one sample remains undiluted as a reference]; shake for 1 minute as in step 1.

Step 4: Dialyze using a 500Da membrane for 4 hours (pre-incubated with 0.2% BSA for 30 minutes; optional but not mandatory), changing the water once every 2 hours. Transfer the solution from the dialysis chamber to a glass vial and shake at RT (room temperature, i.e., 21 ± 2°C) for 20 minutes as in step 1.

Step 5: Dynamic LAL-assay measurement.

Detailed plan

Step 1 : Prepare the sample

Preparation of antibody solution (rituximab) for 50 mM MgCl ₂ : Inject 877.5 μl of rituximab + 97.5 μl of CSE into one tube (50 (5) EU CSE/ml stock solution → 5 (0.5) EU/ml final concentration).

Unincorporated control: 877.5 μl rituximab placebo + 97.5 μl water

Unspiked water control: 975 μl water (for blank subtraction)

Vial used: Transparent flat bottom small opening 1.5ml Macherey & Nagel, Ref.Nr.70213

- Shake at high speed (2037 rpm) on a Heidolph Multi Reax for 1 hour at room temperature.

Step 2 : Add _MgCl2 to a final concentration of 50 mM _MgCl2

• Stock solution 1M MgCl ₂ ·6H ₂ O: Add 50 μl of 1M MgCl ₂ stock solution to the spiked samples as well as to the unspiked samples (blank).

Step 3 : Dilution

Rituximab sample containing 50 mM _MgCl2 : prepare a 1:10 dilution by adding 900 μl endotoxin-free water (i.e., LAL water) + 100 μl sample

Replace rituximab: dilute 1:10 and treat the water control in the same manner as endotoxin-free water (i.e., LAL water).

Step 4 : Dialysis

• Shake for 1 minute before dialysis [ie, in a Heidolph Multi Reax shaker at high speed (2037 rpm) at room temperature].

• The sample is placed in a 1 ml dialyzer chamber (Harvard rotary dialyzer) in which a membrane with a MWCO of 500 Da is immobilized (optionally pre-incubated with 0.2% BSA for 30 minutes).

Dialysis at 24°C for 4 h against 1 l of Aqua Braun (i.e. sterile, pyrogen-free water; supplied by Braun, Melsungen); water exchange after 2 h. The exchanged water has been adjusted to 24°C.

• Distribute the rotary dialyzer (depending on the number of dialysis chambers) over several 2 1 beakers filled with 1 1 of Aqua Braun and stir (magnetic Teflon stirrer).

• Up to 5 dialyzers in one 2 l beaker.

Step 5 : Prepare LER positive control

Also in this example, a LER positive control is used in the LAL assay. This LER positive control can be prepared at any time, as long as it is ready when the LAL assay begins. Advantageously, the LER positive control is prepared 1 hour before the end of the 4-hour dialysis so that all samples are ready for testing at the same time. To prepare the LER positive control, the following protocol is used:

· Rituximab 900 μl + CSE 100 μl → final CSE concentration: 5.0 EU/ml.

- Shake at high speed (2037 rpm) for 1 hour at RT in a Heidolph Multi Reax. Only under these conditions was a maximum LER effect achieved (recovery < 1%).

Prepare the following blanks in parallel:

o Water with 5.0 EU/ml CSE

o Water with 5.0 EU/ml CSE; 1:10 dilution (0.5 EU/ml).

Step 6 : LAL assay

·Start the test after all samples are prepared.

• From all samples, two aliquots of 100 μl were used for repeated determinations in plates (2x, ie duplicate determinations), where the plates were incubated at 37° C. for 10 minutes in a Tecan Reader.

• 100 μl of LAL reagent (Kinetic-QCL ^™ assay; Lonza) was added to each sample in a clearly defined order (according to the machine readout).

Results and Discussion

The protocol described in Example 2.1 yielded the best reproducible recoveries (again relative to the water control). However, the protocol described in Example 2.2 yielded good CSE recoveries for both CSE concentrations spiked, ranging from 50% to 95% (see Figure 4(B) [Rituximab 046]). Therefore, it can be concluded that the protocol used in Example 2.2 is functionally equivalent to the protocol described in Example 2.1.

Reference Example 1: Time dependence of the LER effect

In the prior art, the LER effect is assumed to occur immediately after spiking a sample with a defined amount of CSE (C. Platco, 2014, “Low lipopolysaccharide recovery versus low endotoxin recovery in common biological product matrices.” American Pharmaceutical Review, September 1, 2014, pp. 1-6). Therefore, the sample is initially shaken for a very short time, approximately 2-10 minutes, at room temperature after LPS incorporation. However, this incorporation has proven ineffective, as some experiments have shown that the shielding effect of the substance has not yet reached its maximum within such a short time interval (<10 minutes). The discovery of the incorporation mechanism is one of the essential processes for correctly analyzing the LER effect (see, for example, FIG. 13 [Rituximab 027]). Based on these data, the LER effect is a dynamic phenomenon that requires time to shield the CSE molecule, for example, by penetrating into the micelles of the formulation mixture. Therefore, shaking for 2-10 minutes before the next step of analyzing the LER effect, as is conventional practice, is inappropriate because the conditions for the formation of the LER effect have not yet been reached and cannot be considered representative of the LER effect. Therefore, an internal standard to test for a "positive LER effect" (defined as present when 0% recovery was achieved by the LAL test) was included in the experiment. A kinetic study of the LER effect in rituximab confirmed that a positive LER effect required an incubation time of ≥60 minutes.

In particular, it was analyzed how long shaking [in a 1.5 ml clear glass crimp-top flat-bottom container, in a Heidolph MultiReax shaker, at room temperature (21°C ± 2°C), high speed (2037 rpm), maximum frequency (i.e., vortex)] was required to achieve the maximum LER effect. Therefore, rituximab samples were spiked with CSE in vials to obtain 0.5 and 5.0 EU/ml (vials prepared by Macherey-Nagel, 1.5 ml). After incorporation, the samples were shaken for 60 minutes, 30 minutes, 10 minutes, 5 minutes, or 2 seconds, respectively. Afterwards, a 1:10 dilution was prepared by mixing 900 μl of endotoxin-free water (i.e., LAL water) and 100 μl of sample. After dilution, the sample was shaken for another minute. Subsequently, the samples were tested in duplicate in the LAL assay. Specifically, 100 μl of each sample was applied to the plate and incubated in a reader at 37°C for 10 minutes. Then, 100 μl of chromogen was added to each sample and measured. In this experiment, all solutions were at room temperature. As shown in Figure 13 [Rituximab 027], if the sample was diluted at a ratio of 1: 10, the LER effect was lower (i.e., the recovery rate was higher) compared to the corresponding undiluted sample. In addition, after 2 seconds of oscillation, the recovery value of the diluted sample with 5.0 EU/ml endotoxin was still approximately 50%. However, increasing the oscillation (i.e., vortex) time resulted in a continuous decrease in the recovery rate (except for the value of 30 minutes), see Figure 13 [Rituximab 027]. In contrast, the undiluted sample showed the maximum LER effect after 2 seconds. However, the diluted sample also showed a significant LER effect in the sample that had been oscillated (i.e., vortexed) for 60 minutes.

From this result it was concluded that the incorporation requires time to shield the LPS molecules into the detergent micelles. The "positive LER effect" is complete when approximately 100% shielding or a CSE recovery of <0.5% is obtained. This process requires shaking for at least 1 hour at room temperature [e.g. shaker: Heidolph Multi Reax, high speed (2037 rpm), at room temperature, in a 1.5 to 5 ml clear glass jaw flat bottom for 1 hour] or storage at 4°C for >24 hours longer. In all figures, the resulting "positive LER control" is shown as a bar on the right side of the figure.

Reference Example 2: Effect of Human Serum Albumin (HAS) and Different _MgCl2 Concentrations on Recovery

In order to determine the effect of HSA and different _MgCl2 concentrations on the recovery rate of rituximab samples spiked with endotoxin, the following experiment was performed. In addition, the effect of dialysis on the recovery rate was also analyzed in this experiment. More specifically, the rituximab-spiked sample was shaken for 60 minutes to obtain a "positive LER effect". Before dialysis, 10-75mM _MgCl2 was added, followed by dilution. A membrane without BSA blocking was used. After dialysis, 0.01μg/ml HSA was added or not. Subsequently, the sample was shaken for 20 minutes. In addition, some samples were not dialyzed at all. Specifically, the different samples tested in the LAL assay are shown in Figure 5 (i.e., [Rituximab 059]). In this experiment, the LER effect could be overcome in some samples that were not dialyzed. However, further experiments showed that without dialysis, the LER effect could not be overcome reproducibly. Or in other words, without dialysis, the LER effect was sometimes overcome and sometimes not. Therefore, dialyzing the sample provides a more robust method for overcoming the LER effect.

Samples were prepared in 1.5 ml Macherey-Nagel screw-cap vials.

Step 1 : Prepare the sample

Preparation for _spiked rituximab for 10 mM MgCl2: 897 μl rituximab + 99.8 μl CSE to obtain 5.0 EU/ml

Preparation for _spiked rituximab for 50 mM MgCl2: 889 μl rituximab + 98.8 μl CSE to obtain 5.0 EU/ml

Preparation for _spiked rituximab in 75 mM MgCl2: 883 μl rituximab + 98.1 μl CSE to obtain 5.0 EU/ml

Preparation of spiking water for 10 mM _MgCl2 : 897 μl water + 99.8 μl CSE to obtain 5.0 EU/ml

Preparation of spiking water for 50 mM _MgCl2 : 889 μl water + 98.8 μl CSE to obtain 5.0 EU/ml

Preparation of spiking water for 75 mM _MgCl2 : 883 μl water + 98.1 μl CSE to obtain 5.0 EU/ml

Shake for 60 minutes [Heidolph Multi Reax shaker, high speed (2037 rpm) at room temperature]

Step 2 : Add MgCl2

• A 4M MgCl ₂ stock solution was used (ie 511.437 mg MgCl ₂ ·6H ₂ O in 0.629 ml water).

o For 10 mM MgCl ₂ , add 2.5 μl of 4 M solution to the spiked sample.

o For 50 mM MgCl ₂ , add 12.5 μl of the 4 M solution to the spiked sample.

o For 75 mM MgCl ₂ , add 19 μl of the 4 M solution to the spiked sample.

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 3 : Dilution

Prepare a 1:10 dilution as follows:

Preparation of rituximab 1:10: always 900 μl LAL water + 100 μl sample

• Due to insufficient dialyzers available, water was not diluted 1:10.

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 4 : Dialysis

The sample was placed in a 1 mL dialyzer. 500 Da membrane. However, the membrane was washed in LAL water.

Dialysis was performed against 1 L of Aqua Braun at 24°C for 4 hours, with the water being replaced after 2 hours. The new water also had a temperature of 24°C.

• The dialyzer was placed in three 2 L beakers and swirled with a long stirrer (ie, stir bar) in each beaker.

There are always 4 dialyzers in each beaker.

Step 5 : Add HSA after dialysis

After dialysis, the samples were divided into portions. To prepare HSA samples, 396 μl of each sample was added to a separate vial. To prepare samples without HSA, 400 μl was added to a separate vial.

To obtain an HSA concentration of 0.01 μg/ml, add 4 μl of the 1 μg/ml solution to 396 μl of sample.

· HAS stock solution is freshly prepared.

Shake for 20 minutes [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 6 : Prepare LER positive control

Prepare the LER positive control 1 hour before the end of the 4-hour dialysis so that it can be prepared at the same time as the other samples.

900 μl of rituximab + 100 μl of CSE from different CSE stock solutions to obtain 5.0 EU/ml]

Shake at room temperature for 1 hour [Heidolph Multi Reax shaker, high speed (2037 rpm)]

Step 7 : LAL assay

In duplicate assays, apply 100 μl of each sample to the plate

• Incubate in reader at 37°C for 10 minutes.

• Apply 100 μl of chromogen to each sample.

Start measurement in the reader.

Results and Discussion

The results are shown in Figure 5 [Rituximab 059]. This experiment demonstrated that HSA treatment reduced the recovery rate and is therefore less useful in the methods provided herein. In addition, the results show that BSA treatment of the dialysis membrane is unnecessary for obtaining satisfactory recovery rates. In addition, this experiment also demonstrated that 50mM _MgCl2 was the optimal value for recovery, with 10 and 75mM _MgCl2 resulting in lower recoveries. However, satisfactory recoveries were also obtained with 75mM _MgCl2 . In addition, this experiment showed that even without dialysis, the addition of _MgCl2 resulted in a recovery within a satisfactory range (70-100%). However, as mentioned above, the LER effect cannot be reproducibly overcome without dialysis. Therefore, dialysis of the sample provides a more robust method for overcoming the LER effect. In experiment [Rituximab 059], the water control values were high (partial values >220%). The LER positive control was satisfactory; i.e., 0% recovery.

Reference Example 3: 4 hours incubation time after adding _MgCl2

In this experiment, rituximab samples were shaken for 60 minutes to achieve a "positive LER effect." After adding _MgCl₂ , the undiluted samples were incubated at room temperature for 4 hours. Following incubation, the samples were shaken for 2 minutes. Figure 5(B) shows the different samples [rituximab 061] tested in the LAL assay.

Samples were prepared in 1.5 ml Macherey-Nagel screw-cap vials.

Step 1 : Sample preparation

Preparation for _spiked rituximab/water for 10 mM MgCl2: 897 μl rituximab/water + 99.8 μl CSE to obtain 5.0 EU/ml

Preparation for _spiked rituximab/water for 50 mM MgCl2: 889 μl rituximab/water + 98.8 μl CSE to obtain 5.0 EU/ml

Preparation for _spiked rituximab/water for 75 mM MgCl2: 883 μl rituximab/water + 98.1 μl CSE to obtain 5.0 EU/ml

Preparation for _spiked rituximab/water for 100 mM MgCl2: 877 μl rituximab/water + 97.5 μl CSE to obtain 5.0 EU/ml

Preparation for _spiked rituximab/water for 150 mM MgCl2: 866 μl rituximab/water + 96.3 μl CSE to obtain 5.0 EU/ml

Shake for 60 minutes [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 2 : Add _MgCl2

• A 4M MgCl2 _stock solution was used (ie 534.661 mg _MgCl2 · _6H2O in 0.657 ml water).

• For 10 mM MgCl ₂ , add 2.5 μl of a 4 M solution to the spiked sample.

• For 50 mM MgCl ₂ , add 12.5 μl of a 4 M solution to the spiked sample.

• For 75 mM MgCl ₂ , add 19 μl of a 4 M solution to the spiked sample.

• For 100 mM MgCl ₂ , add 25 μl of a 4 M solution to the spiked sample.

• For 150 mM MgCl ₂ , add 37 μl of 4 M solution to the spiked sample.

Oscillate for 1 minute [at room temperature and high speed (2037 rpm)]

Step 3 : Dilution

Prepare dilution at a ratio of 1:10: 900 μl LAL water + 100 μl sample (i.e., rituximab sample or water sample)

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 4 : Prepare LER positive control

Rituximab 900 μl + 100 μl CSE to obtain 5.0 EU/ml

Prepare another LER positive control by mixing 900 μl rituximab + 100 μl CSE to obtain 5.0 EU/ml; then dilute the sample 1:10 with endotoxin-free water

Step 5 : Oscillation

Shake all samples and the LER positive control at room temperature for 1 hour [Heidolph Multi Reax shaker, high speed (2037 rpm)]

Step 6 : LAL assay

In duplicate assays, apply 100 μl of each sample to the plate

• Incubate in reader at 37°C for 10 minutes.

• Apply 100 μl of chromogen to each sample.

Start measurement in the reader.

Results and Discussion

The results of this experiment are shown in Figure 6(B) (i.e., [Rituximab 061]). This figure again demonstrates that 50mM _MgCl2 is the reproducible optimal value for CSE recovery; 10, 75, and 150mM show slightly worse results. After adding _MgCl2 and implementing an incubation time, the undiluted rituximab sample did not result in CSE recovery at all, see Figure 6(B) (i.e., [Rituximab 061]). However, a 1:10 dilution resulted in approximately 50-60% recovery (especially when ₁₀ , 50, 75, or 100mM MgCl2 was added). Both the water control values and the LER positive control were satisfactory. In this experiment, dialysis was not performed. However, some experiments have shown that dialysis is necessary to reproducibly overcome the LER effect.

Reference Example 4: Comparison of 2-hour and 4-hour incubation times after addition of _MgCl2

Rituximab samples were shaken for 60 minutes to achieve a "positive LER effect." After adding _MgCl2 , undiluted samples were incubated for 2 or 4 hours, then diluted 1:10 and measured in the LAL assay. Figures 7(A) and 7(B) show different samples tested in the LAL assay (i.e., [Rituximab 062] and [Rituximab 063], respectively).

Samples were prepared in 1.5 ml Macherey-Nagel screw-cap vials.

Step 1 : Sample preparation

_• Preparation of rituximab/water for 10 mM MgCl2: 897 μl rituximab/water + 99.8 μl different CSE stocks to obtain 0.5 and 5.0 EU/ml.

• Preparation of rituximab/water for 50 mM _MgCl2 : 889 μl rituximab/water + 98.8 μl different CSE stocks to obtain 0.5 and 5.0 EU/ml.

• Preparation of rituximab/water for 75 mM _MgCl2 : 883 μl rituximab/water + 98.1 μl different CSE stocks to obtain 0.5 and 5.0 EU/ml.

Step 2 : Prepare two LER positive controls

• Rituximab 900 μl + 100 μl CSE to obtain 0.5 and 5.0 EU/ml.

• Dilute one of the LER positive controls at a ratio of 1:10 with endotoxin-free water.

Step 3 : Oscillation

Shake all samples and the LER positive control for 1 hour: [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 4 : Add MgCl2

Use 4M _MgCl2 stock solution.

o For 10 mM MgCl ₂ , add 2.5 μl of 4 M solution to the spiked sample

o For 50 mM MgCl ₂ , add 12.5 μl of 4 M solution to the spiked sample

o For 75 mM MgCl ₂ , add 19 μl of 4 M solution to the spiked sample

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 5 : Incubation time

• The (undiluted) samples (and the LER positive control) were divided into portions. Half of each sample (approximately 500 μl) was incubated for 2 hours and the other half for 4 hours.

Step 6 : Dilution

Shake for 2 minutes [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Prepare dilutions at a ratio of 1:10: 900 μl LAL water + 100 μl sample (ie, rituximab sample or water sample).

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 7 : LAL assay

In duplicate assays, apply 100 μl of each sample to the plate

• Incubate in reader at 37°C for 10 minutes.

• Apply 100 μl of chromogen to each sample.

Start measurement in the reader.

Results and Discussion

The results are shown in Figures 7(A) and 7(B) (i.e., [Rituximab 062] and [Rituximab 063], respectively). The recoveries of 0.5 and 5.0 EU/ml endotoxin were measured here. When 10-75 mM MgCl ₂ was added to the sample, the recovery in the sample incubated for 2 hours (Figure 7(A), i.e., [Rituximab 062]) and the sample incubated for 4 hours (Figure 7(B), i.e., [Rituximab 063]) was the same. In both experiments, the recoveries were very similar. In addition, in the sample spiked with 5.0 EU/ml endotoxin, a satisfactory recovery (80-90%) was obtained even without dialysis. In the sample spiked with 0.5 EU/ml endotoxin, the recovery was approximately 35-45%. Importantly, without dilution (at a ratio of 1:10), a complete LER, i.e., 0% recovery, was observed, also in the presence of 10-75 mM MgCl ₂ . The water control and the LER positive control were satisfactory. In these experiments, dialysis was not performed. However, further experiments showed that the LER effect was sometimes overcome and sometimes not overcome without dialysis. Therefore, dialysis of the samples provides a more robust method to overcome the LER effect.

Reference Example 5: Comparison of adding different amounts of _MgCl2 and incubating for 2 hours with adding different amounts of _MgCl2 and not incubating

Rituximab samples were shaken for 60 minutes to achieve a "positive LER effect." After adding _MgCl2 , the undiluted samples were either not incubated or incubated for 2 hours. They were then diluted 1:10 and measured in the LAL assay. Figures 8(A) and 8(B) show different samples tested in the LAL assay (i.e., [Rituximab 064] and [Rituximab 065], respectively).

Refer to Example 5.1: No incubation time after adding _MgCl2

In this experiment, _MgCl2 was added to the samples without incubation.

Samples were prepared in 1.5 ml Macherey-Nagel screw-cap vials.

Step 1 : Prepare the sample

_• Preparation of rituximab/water for 10 mM MgCl2: 897 μl rituximab/water + 99.8 μl different CSE stocks to obtain 0.5 and 5.0 EU/ml.

• Preparation of rituximab/water for 25 mM _MgCl2 : 895 μl rituximab/water + 99.4 μl different CSE stocks to obtain 0.5 and 5.0 EU/ml.

• Preparation of rituximab/water for 50 mM _MgCl2 : 889 μl rituximab/water + 98.8 μl different CSE stocks to obtain 0.5 and 5.0 EU/ml.

Step 2 : Prepare two LER positive controls

• Rituximab 900 μl + 100 μl CSE to obtain 0.5 and 5.0 EU/ml.

• Dilute one of the LER positive controls at a ratio of 1:10 with endotoxin-free water.

Step 3 : Oscillation

Shake (i.e., vortex) all samples and the LER positive control for 60 minutes: [Heidolph Multi Reax shaker, high speed (2037 rpm) at room temperature]

Step 4 : Add _MgCl2

Use 4M _MgCl2 stock solution.

o For 10 mM MgCl ₂ , add 2.5 μl of 4 M solution to the spiked sample

o For 25 mM MgCl ₂ , add 6.25 μl of 4 M solution to the spiked sample

o For 50 mM MgCl ₂ , add 12.5 μl of 4 M solution to the spiked sample

Shake for 1 minute [Heidolph Multi Reax shaker, high speed (2037 rpm) at room temperature]

Step 5 : Dilution

Shake for 1 minute [Heidolph Multi Reax shaker, high speed (2037 rpm) at room temperature]

Prepare dilutions at a ratio of 1:10.

Shake for 1 minute [Heidolph Multi Reax shaker, high speed (2037 rpm) at room temperature]

Step 6 : LAL assay

In duplicate assays, apply 100 μl of each sample to the plate

• Incubate in reader at 37°C for 10 minutes.

• Apply 100 μl of chromogen to each sample.

Start measurement in the reader.

Results and Discussion

The results are shown in Figure 8(A) (i.e. [rituximab 064]). For discussion of the results, see reference

Example 4.2.

Refer to Example 5.2: Add _MgCl2 and incubate for 2 hours

In this experiment, the sample was incubated for 2 hours after the addition of _MgCl2 . Steps 1 to 4 were performed as described above with reference to Example 4.1. However, after the addition of _MgCl2 , the undiluted sample was incubated for 2 hours at room temperature (21°C). After incubation, steps 5 and 6 below were performed. Figure 8(B) shows different samples (i.e., [Rituximab 065]) tested in the LAL assay.

Step 5 : Dilution

Shake for 2 minutes [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Prepare dilution at a ratio of 1:10: 900 μl LAL water + 100 μl sample (i.e., rituximab sample or water sample)

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 6 : Prepare two LER positive controls

Rituximab 900 μl + 100 μl CSE to obtain 0.5 and 5.0 EU/ml

Dilute one of the LER positive controls at a ratio of 1:10 with endotoxin-free water

Step 7 : Oscillation

Shake all samples and the LER positive control for 1 hour [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 8 : LAL assay

• In duplicate assays 100 μl of each sample was applied to the plate.

• Incubate in reader at 37°C for 10 minutes.

• Apply 100 μl of chromogen to each sample.

Start measurement in the reader.

Results and Discussion

The results of Reference Example 4.1 are shown in FIG8(A) (i.e., [Rituximab 064]); the results of Reference Example 4.2 are shown in FIG8(B) (i.e., [Rituximab 065]). In both experiments, dilution and LAL measurement were performed immediately after the _addition of MgCl2 (FIG8(A), [Rituximab 064]) or after the samples were allowed to stand for 2 hours after the addition _of MgCl2 (FIG8(B), [Rituximab 065]). All recoveries were very similar, and in the samples spiked with 5.0 EU/ml CSE, satisfactory recoveries (60-80%) were obtained even without dialysis. Interestingly, after a 2-hour incubation time, a 25 mM _MgCl2 concentration resulted in 100% recovery. Therefore, the incubation time after the _addition of MgCl2 appears to be a valuable measure to reduce the LER effect. However, the recovery values of the samples spiked with 0.5 EU/ml CSE were low, approximately 20-35%. This indicates that, in addition to the addition of Mg2 ⁺ and dilution, dialysis is a necessary step to reliably overcome the LER effect. In these experiments, the values of the water control were satisfactory. The undiluted LER positive control was also satisfactory, i.e., 0%.

Reference Example 6: Comparison of different dilutions using rituximab and rituximab placebo samples

After incorporation, rituximab and rituximab placebo samples were shaken for 60 minutes to achieve a "positive LER effect." After adding _MgCl , the undiluted samples were shaken for 1 hour and diluted at a ratio of 1:2, 1:5, 1:10, or 1:20. The LAL assay was then performed. FIG9 shows the different samples tested in the LAL assay (i.e., [rituximab 072]).

Specifically, the following experiments were conducted:

Step 1 : Prepare the sample

Preparation of Rituximab/Rituximab Placebo/Water for ₂₅ mM MgCl2: 895 μl Rituximab/Rituximab Placebo/Water + 99.4 μl different CSE stocks to obtain 0.5 and 5.0 EU/ml.

Step 2 : Prepare three LER positive controls

• Rituximab placebo 450 μl + 50 μl CSE to obtain 0.5 and 5.0 EU/ml (first LER positive control).

• Rituximab 450 μl + 50 μl CSE to obtain 0.5 and 5.0 EU/ml (second LER positive control).

• Rituximab 450 μl + 50 μl CSE to obtain 0.5 and 5.0 EU/ml. Afterwards, this sample was diluted at a ratio of 1:10 (third LER positive control).

Step 3 : Oscillation

All samples and the LER positive control were shaken (i.e., vortexed) for 60 minutes [Heidolph Multi Reax shaker, high speed (2037 rpm) at room temperature]

Step 4 : Add _MgCl2

Use 4M _MgCl2 stock solution.

o For 25 mM MgCl ₂ , add 6.25 μl of 4 M solution to the spiked sample

Shake for 1 minute [Heidolph Multi Reax shaker, high speed (2037 rpm) at room temperature]

Step 5 : Dilution

The samples were diluted as follows:

o Dilute at a ratio of 1:5: 400 μl water + 100 μl sample

o Dilute at a ratio of 1:10: 450 μl water + 50 μl sample

o Dilute at a ratio of 1:20: 475 μl water + 25 μl sample

• Dilute one of the three LER positive controls at a ratio of 1:10.

Do not dilute with water that does not contain _MgCl2 .

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 6 : LAL assay

In duplicate assays, apply 100 μl of each sample to the plate

• Incubate in reader at 37°C for 10 minutes.

• Apply 100 μl of chromogen to each sample.

Start measurement in the reader.

Results and Discussion

The results are shown in Figure 9 (i.e., [Rituximab 072]). Importantly, there was no significant difference in the results for rituximab and rituximab placebo. This suggests that the LER effect in rituximab is primarily based on the buffer system (i.e., citrate buffer with polysorbate 80) and that the antibody (i.e., rituximab) has no significant effect on the LER effect. However, in this experiment, the recoveries for both rituximab and rituximab placebo were unsatisfactory. In the above experiment, the values for the water control were satisfactory. The undiluted LER positive control was also satisfactory, with a recovery of 0%.

Reference Example 7: Effect of incubation time before addition of _MgCl2 on the recovery of rituximab and rituximab placebo samples

In the following experiment, it was tested whether the incubation time before adding _MgCl2 had an effect on the recovery of rituximab and rituximab placebo samples. Specifically, rituximab and rituximab placebo samples were shaken for 60 minutes to achieve a "positive LER effect". The samples were then incubated at 4°C for 0 hours to 3 days. Afterwards, _MgCl2 was added to a concentration of 50mM to dilute the samples. The samples were then dialyzed using a dialysis membrane without BSA blocking. Figure 10 shows different samples tested in the LAL assay (i.e., [rituximab 079] and [rituximab 082]).

Samples were prepared in 1.5 ml Macherey-Nagel screw-cap vials.

Step 1 : Prepare the sample

• Preparation of Rituximab/Rituximab Placebo/Water for 50 mM _MgCl2 : 5,346 μl Rituximab/Rituximab Placebo/Water + 596 μl of different CSE stocks to obtain 0.5 or 5.0 EU/ml.

Vortex for 60 minutes [Heidolph MultiReax shaker, high speed (2037 rpm) at room temperature]

For further steps, transfer 1 ml of each sample to a new vial

Step 2 : Incubation time

• Place the samples in a 4°C refrigerator for 0 hours, 4 hours, 1 day, 3 days, or 7 days.

• After an incubation time of 1 day, 3 days or 7 days, the samples were shaken for 2 minutes [Heidolph MultiReax shaker, high speed (2037 rpm) at room temperature].After 0 hours and 4 hours of incubation, the samples were not shaken.

Step 3 : Add _MgCl2

Use 5M _MgCl2 stock solution (i.e. 0.9055g _MgCl2 in 0.891ml water)

o For 50 mM MgCl ₂ , add 10 μl of 5 M solution to the spiked sample

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)

Step 4 : Dilution

Prepare dilutions at a ratio of 1:10: 900 μl LAL water + 100 μl sample (i.e., rituximab sample, rituximab placebo sample, or water sample)

Do not dilute with water that does not contain _MgCl2 .

Shake for 1 minute [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)

Step 5 : Dialysis

Add 1 ml of sample to the dialyzer. Dialyze against 1 liter of Aqua Braun at 24°C for 4 hours, changing the water after 2 hours. The new water also has a temperature of 24°C. Use a 500 Da membrane that has not been previously incubated with BSA.

• The dialyzer was placed in three 2 L beakers and rotated due to a long stirrer (ie, stirring bar) in each beaker.

After dialysis, transfer the sample to a new 1.5 ml vial.

Step 6 : Oscillation

Shake for 20 minutes [Heidolph Multi Reax shaker, room temperature, high speed (2037 rpm)]

Step 7 : Prepare LER positive control

• Prepare the LER control 1 hour before the end of the 4-hour dialysis so that it can be prepared at the same time as the other samples.

Rituximab or rituximab placebo 900 μl + 100 μl CSE to obtain 5.0 EU/mL CSE

Shake at room temperature for 1 hour [Heidolph Multi Reax shaker, high speed (2037 rpm)]

Step 8 : LAL assay

In duplicate assays, apply 100 μl of each sample to the plate

• Incubate in reader at 37°C for 10 minutes.

• Apply 100 μl of chromogen to each sample.

Start measurement in the reader.

Results and Discussion

The results are shown in Figure 10 (A) (i.e., [Rituximab 079], no incubation), Figure 10 (B) (i.e., [Rituximab 080], incubation for 4 hours), Figure 10 (C) (i.e., [Rituximab 081], incubation for 1 day), and Figure 10 (D) (i.e., [Rituximab 082], incubation for 3 days). Results for 7 days of incubation are not shown. The results again demonstrate that good recoveries can be obtained for rituximab and rituximab placebo samples using the protocol described herein. In addition, this experiment demonstrates that incubation time before adding _MgCl2 does not improve recovery. Specifically, incubation times of 0-4 hours resulted in recoveries within the desired range (50-200%), while incubation times of 1 day and 3 days resulted in lower recoveries (approximately 20-30%). Without incubation (i.e., 0 hours), very good recoveries (70-80%) were obtained for rituximab. The recovery of rituximab placebo spiked with 5.0 EU/ml CSE was also satisfactory. The recovery of rituximab placebo spiked with 0.5 EU/ml CSE was negative due to the very high blank. This may indicate that the blank sample was contaminated with endotoxin. The recovery of the water control (80-120%) and the LER positive control (~0%) was also satisfactory in this experiment.

Reference Example 8: Prior Art Protocol ("LAL Assay") and Its Modifications Fail to Overcome the LER Effect in Rituximab or Rituximab Placebo

In this example, it was determined whether the commonly known LAL assay was capable of detecting endotoxin in rituximab and rituximab placebo preparations. Therefore, the following materials were used:

[Rituximab 002]: Lonza CSE + Lonza reagent (i.e., Lonza kit)

[Rituximab 004]: ACC CSE or Lonza CSE + ACC reagent (i.e., ACC kit)

[Rituximab 005]: Lonza CSE+ACC reagent (i.e., ACC kit)

The LAL assay was performed exactly as described by the manufacturer.

As can be seen from Figures 11(A)-(C) (i.e., [Rituximab 002], [Rituximab 004], and [Rituximab 005], respectively), the standard LAL assay did not result in satisfactory recovery, even when several different dilutions were tested. Specifically, in one experiment (i.e., [Rituximab 002]), rituximab was pipetted into a 96-well plate (i.e., microtiter plate) and spiked with Lonza CSE to a final concentration of 0.5 EU/ml or 5.0 EU/ml. Subsequently, dilutions with water were performed in the 96-well plate as shown in Figure 11(A) (i.e., [Rituximab 002]). The LAL assay was then performed. However, as can be seen in Figure 11(A) (i.e., [Rituximab 002]), 50% recovery was not achieved.

In similar experiments (i.e., [Rituximab 004]), rituximab was pipetted into microtiter plate wells and spiked with Lonza CSE and ACC CSE to a final concentration of 0.5 EU/ml or 5.0 EU/ml. Subsequently, dilution with water was performed in a 96-well plate as shown in Figure 11(B) (i.e., [Rituximab 004]). Measurements were then performed. However, as shown in Figure 11(B), the recovery rate using ACC CSE exceeded 200%, while the Lonza CSE spike did not achieve satisfactory recovery.

In further experiments, the effect of pH adjustment on the LAL assay was analyzed. Specifically, in one experiment [Rituximab 005], rituximab was pipetted into a microtiter plate, where it was spiked with Lonza CSE. Subsequently, dilution with water or pH adjustment was performed, as shown in Figure 11(C) [Rituximab 005]. However, neither dilution nor pH adjustment resulted in 50% recovery (see Figure 11(C) [Rituximab 005]).

Dialysis alone also failed to achieve satisfactory recoveries. More specifically, in further experiments, rituximab was spiked with CSE to obtain final concentrations of 0.5 and 5.0 EU/ml (i.e., 900 μl of rituximab solution was mixed with 100 μl of CSE). Subsequently, the sample was dialyzed at 4°C for 4 hours in a 1 ml rotary dialyzer (in a 1 ml Teflon chamber), with water being changed once after 2 hours. The dialysis membrane has a MWCO of 100 Da. Dilutions as shown in Figure 12 (i.e., [rituximab 011]) were then performed in the plate. Subsequently, endotoxin recovery was measured by using the LAL assay. However, only the water control achieved good recoveries. In the case of rituximab, the maximum recovery was <5% (see Figure 12 [rituximab 011]). Therefore, dialysis and dilution alone cannot overcome the LER effect.

Reference Example 9: Retention Time Study

In order to identify and monitor the LER effect, endotoxin levels were monitored over time in endotoxin retention time studies. Therefore, endotoxin was spiked into undiluted samples of various buffers and stored over time (up to 28 days). Acceptable endotoxin recovery values in PPC after spike-in with the appropriate sample mixture were defined as being within the range of 50-200% of the theoretical spike-in value (100%). The LER effect is indicated by a significant loss of endotoxin over time. Specifically, an unfavorable trend in endotoxin values <50% of the theoretical spike-in value indicates a LER effect.

Several formulation buffer compositions were investigated in the endotoxin retention time studies (see table below for results).

Table 1: Hold Time Study

n.d. = not determined

As can be seen from the above table, buffers containing polysorbate 20 and _Na2HPO4 ; polysorbate ₂₀ and _NaH2PO4 ; polysorbate 20, _Na2HPO4 ⁺ and _NaH2PO4 ; sodium citrate, polysorbate ₈₀ and NaCl; sodium citrate and polysorbate ₈₀ ; _and polysorbate 80 and NaCl exhibit the LER effect.

Reference Example 10: Effect of buffer and detergent on LER effect

In several experiments, the effect of citrate and/or polysorbate 80 on the LER effect was analyzed. Specifically, in one experiment, rituximab and 25 mM sodium citrate buffer were used as samples. Prior to incorporation, the pH was adjusted to pH 7. Subsequently, CSE was incorporated into the plate and the sample was diluted with water. As can be seen in Figure 14(A) (i.e., [Rituximab 006]), satisfactory recoveries were obtained with sodium citrate using a 1:10 dilution. However, a 50% recovery could not be achieved with rituximab.

In another experiment, 25 mM sodium citrate buffer, polysorbate 80, and a combination of the two were used as samples. Specifically, the concentrations present in rituximab (i.e., polysorbate 80: 0.7 mg/ml; sodium citrate: 9 mg/ml) were used. These buffer systems were spiked with 0.5 and 5.0 EU/ml of Lonza CSE or Cape cod CSE (except sodium citrate, which was only spiked with Lonza, as sodium citrate buffer spiked with ACC had already been performed in the above experiments and is shown in Figure 14(A) (i.e., [Rituximab 006]). After spike-in, 1:2 or 1:5 dilutions were performed in the plate with water. In the case of polysorbate 80, several samples gave satisfactory recoveries between 50% and 200%. In contrast, in the case of sodium citrate buffer, only a 1:10 dilution gave recoveries between 50% and 200%. This indicates that the citrate buffer has a more significant effect on the LER effect than polysorbate 80. In addition, in this experiment, the LER effect could not be overcome if a combination of sodium citrate and polysorbate 80 was used as a sample. This experiment shows that in monoclonal antibody preparations, the LER effect is caused by the buffer preparation, i.e., the combination of sodium citrate buffer and polysorbate 80.

These results have been confirmed by another experiment in which several different dilutions were tested. Specifically, samples containing 25mM sodium citrate buffer (pH 6.5), 700mg/L polysorbate 80 or both were prepared (i.e., rituximab formulations). These formulations and water controls were spiked with Lonza CSE to a final concentration of 0.5 or 5.0EU/ml. All samples were shaken in a vortex machine for 1 hour at room temperature [shaker: Heidolph Multi Reax, high speed (2037rpm), in a 1.5 transparent glass crimp flat-bottom container. Subsequently, dilutions as shown in Figure 14 (B) (i.e., [rituximab 029]) were performed in 1.5ml vials with endotoxin-free water and shaken (as before) for 1 minute. After shaking, the LAL assay was performed. Specifically, 100μl of each sample was added to a 96-well plate and incubated in a reader at 37°C for 10 minutes. Then, 100μl of chromogen was added to each sample and measured. As shown in Figure 14 (B) (i.e. [rituximab 029]), the LER effect of the buffer (i.e. sodium citrate buffer) is stronger than the LER effect of the detergent (i.e. polysorbate 80). Polysorbate 80 shows a relatively constant recovery rate (~40-90%, see Figure 14 (B), [rituximab 029] from the right column 2-5), whereas the LER effect in citrate buffer depends on the dilution. Most importantly, if polysorbate 80 is combined with citrate buffer (as in the case of monoclonal antibody preparations), a strong and reproducible LER effect is exhibited, see Figure 14 (B) (i.e. [rituximab 029]). In these samples, high dilutions were also used, and the recovery rate was only ~5-10%. Therefore, this experiment demonstrates how a positive LER effect can be obtained. This result is groundbreaking in the field of endotoxin determination because it allows testing the ability of several means and methods to overcome the LER effect.

When the effects of buffer and detergent were analyzed separately, the effect of the buffer on the LER effect was more pronounced in both Rituximab and NeoRecormon (see, e.g., Figure 14(B), i.e., [Rituximab 029]). These data surprisingly indicate that removing the buffer is more important than removing the detergent. Considering that the concentration of the buffer was comparable in both formulations (Rituximab: 25 mM sodium citrate and NeoRecormon: 27.8 mM sodium phosphate), the cause of this effect must be sought in the structure of the buffer and/or its physico-chemical properties. Sodium citrate is a well-known chelating anion, whereas in phosphates this effect is less pronounced. Therefore, these observations may also explain why the addition of Mg ²⁺ is important for overcoming the LER effect, since Mg ²⁺ is complexed by the chelating buffer, reducing its concentration in the LAL test.

Reference Example 11: Standard physical and biochemical methods cannot recover endotoxins shielded by the LER effect

In order to overcome the LER effect (in the buffer identified as having the LER effect with reference to Example 9), different physical and biochemical approaches were tested:

The endotoxin-spiked samples were frozen at -30° C. This study was based on the initial finding that the LER was more pronounced at room temperature than at 2-8° C. Results: Freezing the endotoxin-spiked samples did not overcome the LER.

The endotoxin-spiked samples were heated at 70°C for 30 minutes. This study was conducted because heating has been shown to overcome the endotoxin shielding effect of some products (Dawson, 2005, LAL update. 22: 1-6). Results: Heat treatment of the endotoxin-spiked samples did not overcome the LER.

Endotoxin-spiked samples were diluted to the maximum effective dilution (MVD). This study was performed because sample dilution is a standard method for overcoming LAL inhibition. Results: As can be seen in Figure 11 (i.e., [Rituximab 002], [Rituximab 004], [Rituximab 005]), dilution alone was not able to overcome the LER effect.

EndoTrap columns were used for samples spiked with endotoxin. These columns are used to remove endotoxin from solutions by affinity chromatography. Tests were performed using aqueous solutions of endotoxin. Results: No endotoxin could be recovered from the columns.

Reference Example 12: Detergent Removal by Dialysis

As described below, several commercially available dialysis chambers and membranes (including different sizes of molecular weight cut-off, MWCO) have been tested.

Suitable commercially available membranes for dialysis chambers are, for example, cellulose acetate (MWCO 100 to 300,000 Da), regenerated cellulose (MWCO 1,000 to 50,000 Da) or cellulose esters (MWCO 100 to 500 Da). Cellulose acetate and cellulose esters are preferred, with cellulose acetate being most preferred.

The test sample (i.e., rituximab) was diluted prior to dialysis to approach the CMC and produce increasing levels of detergent monomers (which are expected to diffuse through the dialysis membrane). Investigations into the recovery of incorporated CSE revealed that, in the case of rituximab, regenerated cellulose was not as effective as cellulose acetate. In a series of experiments using rituximab, a MWCO of ~10 kDa was identified as being preferred. This size is preferred because it is believed to i) speed up the dialysis process and ii) also allow aggregates of higher oligomers of the detergent (but not micelles) to pass through the membrane (provided that the hydrophobic nature of cellulose acetate (acetyl esters on the glucose polymers) does not inhibit this transport process).

For the situation of simulation, the experiment of determining optimal dialysis was carried out. As previously mentioned, buffer and detergent are those compounds that mainly affect the LER effect in the sample preparation. For the preparation of simulation, phosphate buffer (2.7mg) of a certain amount was prepared in the presence of 0.1mg/ml polysorbate 20, in a total volume of 0.5ml, and dialyzed (in a rotary dialyzer). A very simple example of this experiment is shown in Figure 1, wherein a cellulose acetate dialysis membrane with a MWCO of 12-16kDa is used. The material remaining in the inner dialysate after a given time is shown here, thus illustrating the efficiency of dialysis. Specifically, the weight of the material remaining in the inner dialysis chamber for 72 hours (3d) is analyzed. The result is quite surprising because it shows the complete effective dialysis that can be achieved only after a long dialysis time (>24-48 hours) at room temperature. Based on this experiment, the preferred dialysis time is 20 hours to overnight (e.g., 24 hours). Furthermore, this result suggests that the Harvard Rapid Dialyzer is superior to the Harvard Rotary Dialyzer, which has a dual-zone dialysis membrane, resulting in more rapid dialysis.

In Figure 2, the dialysis efficiency is shown when phosphate is placed in the inner dialysate compartment of the dialysis. In this experiment, the dialysis membrane (MWCO 12-16 kDa) was washed with 0.2% BSA (30 minutes) before use to avoid nonspecific absorbance of CSE spiked into the sample. However, in the inventive methods provided herein, dilution of the sample (e.g., at a 1:10 ratio) can reduce the concentration of the detergent.

Reference Example 13: Effect of _MgCl2 on the LER Effect

It has been found that the LER effect can be reduced by adding _MgCl2 to the sample (see, e.g., Figure 15 (i.e., [Rituximab 031]). In particular, the best results were observed when the concentration of Mg2 ⁺ was twice that of sodium citrate (i.e., 50 mM Mg2 ⁺ ).

Specifically, samples containing 25 mM sodium citrate buffer pH 6.5 (i.e., sodium citrate buffer, pH 6.5), 0.7 mg/ml polysorbate 80, or both (i.e., rituximab formulations) were prepared. Lonza CSE was added to these formulations, as well as to a water control, to a final concentration of 0.5 or 5.0 EU/ml. All samples were shaken for 1 hour at room temperature [shaker: Heidolph Multi Reax, high speed (2037 rpm), in a 1.5 transparent glass crimp-top flat-bottom container]. MgCl _was then added to the samples to reach a concentration of 10 mM, 25 mM, 50 mM, or 75 mM. Subsequently, dilutions as shown in Figures 15 (A), 18 (B), 18 (C), and 18 (D) (i.e., [rituximab 030-rituximab 033]) were performed in 1.5 ml vials with endotoxin-free water and shaken (i.e., vortexed as before) for 1 minute. After shaking, the LAL assay was performed. Specifically, 100 μl of each sample was added to a 96-well plate and incubated in a reader at 37°C for 10 minutes. Afterwards, 100 μl of chromogen was added to each sample and measured. As shown in Figure 15(A) (i.e., [Rituximab 030]), _MgCl2 (10 mM) neutralized the complexing effect of citrate in all diluted samples. Magnesium ions reduced the LER effect in samples containing polysorbate 80 and citrate buffer. In this case, a recovery of approximately 50% was achieved with 5.0 EU/ml and approximately 25% with 0.5 EU/ml. The values for the water control were within the range of the theoretically expected values (i.e., 70-130%). In addition, a comparison of Figures 15(A), 15(B), 15(C), and 15(D) (i.e., [Rituximab 030], [Rituximab 031], [Rituximab 032], and [Rituximab 033]) shows that _a MgCl concentration twice that of the citrate buffer (i.e., 50 mM _MgCl ) yielded the best recovery. Although 25 mM and 75 mM _MgCl were not optimal concentrations for _MgCl , these concentrations could still overcome the LER of the citrate buffer (recovery: 75-190%). In similar experiments using rituximab as a sample, it was demonstrated that the addition of _MgCl2 alone to a concentration of 10 mM, ₅₀ mM or 75 mM MgCl2 and subsequent dilution at a ratio of 1:10 (without dialysis) resulted in satisfactory recovery of endotoxin spiked to a final concentration of 5.0 EU/ml (see Figure 5, i.e. [Rituximab 059]).

Reference Example 14: Effect of Mechanical Treatment on LER Effect

It was tested whether mechanical treatments such as shaking and sonication could be used to disperse the micelles and thereby reduce the LER effect.

Specifically, endotoxin-free water (i.e., LAL water) and rituximab were spiked with Lonza CSE to achieve final concentrations of 0.5 and 5.0 EU/ml. The samples were then sonicated for 1 hour or shaken for 1 hour [i.e., in a Heidolph Multi Reax shaker in a 1.5 ml clear glass crimp-top container, vortexed at high speed (2037 rpm) at room temperature]. A 1:10 (sample:water) dilution was then prepared with endotoxin-free water. The diluted samples were then dialyzed using a 12-16 kD membrane (which had been incubated in 0.2% BSA for 30 minutes prior to dialysis). Dialysis was performed in two 2-liter beakers for 4 hours. The external dialysate was 1 liter of AquaBraun, and the water was changed after 2 hours of dialysis. After dialysis, _MgCl₂ (as shown in Figure 16, i.e., [Rituximab 034]) was added to some samples to achieve _a final MgCl₂ concentration of 50 mM. After the addition of MgCl ₂ , all samples (including samples without MgCl ₂ ) were shaken for 20 minutes (e.g., vortexed as before). Subsequently, the LAL assay was performed. Specifically, 100 μl of each sample was added to a 96-well plate and incubated in a reader at 37° C. for 10 minutes. Then, 100 μl of chromogen was added to each sample and measured. As can be seen from Figure 16 (i.e. [Rituximab 034]), neither oscillation nor ultrasound treatment improved the recovery. It can therefore be concluded that mechanically dispersing the micelles that cause the LER by oscillation or ultrasound treatment is relatively ineffective. However, the addition of MgCl ₂ (50 mM) reduced the LER effect, as it resulted in improved recovery values of up to 5-20%. In addition, this experiment also confirmed that the order of the different implementation steps is important for overcoming the LER effect. Specifically, in the above-mentioned experiment (shown in FIG16 , i.e., [Rituximab 034]), the order of the steps was (a) dilution, (b) dialysis, and (c) addition of MgCl ₂ . This order did not result in satisfactory recovery (see FIG16 , i.e., [Rituximab 034]). However, as shown in FIG3 and FIG4 (i.e., [Rituximab 046], [Rituximab 115], and [Rituximab 117]), the order of (a) addition of MgCl ₂ , (b) dilution, and (c) dialysis resulted in a recovery that met FDA requirements (i.e., 50%-200%).

The present invention relates to the following nucleotide and amino acid sequences:

SEQ ID NO: 1: Rituximab heavy chain amino acid sequence

QVQLQQPGAELVKPGASVKMSCKASGYTFTSYNMHWVKQTPGRGLEWIGAIYPGNGDTSYNQKFKGKATLTADKSSSTAYMQLSSSLTSEDSAVYYCARSTYYGGDWYFNVWG AGTTVTVSAASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCD KTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEK TISKAKGQPREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK

SEQ ID NO: 2: Rituximab light chain amino acid sequence

QIVLSQSPAILSASPGEKVTMTCRASSSVSYIHWFQQKPGSSPKPWIYATSNLASGVPVRFSGSGSGTSYSLTISRVEAEDAATYYCQQWTSNPPTFGGGTKLEIK RTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

Claims (25)

1. A method for preparing a sample containing an antibody for the determination of horseshoe crab cell lysates, wherein the method comprises the following steps in the following order: (a) Add magnesium ions to the sample. (b) Dilute the sample, and (c) Samples with pH values of 5.7-8.0 compared to endotoxin-free aqueous solutions. 2. The method of claim 1, wherein in step (a), magnesium ions are added to the sample in the form of _MgCl2 . 3. A method for determining bacterial endotoxins in antibody-containing samples exhibiting low endotoxin recovery effects, wherein the method comprises the following steps in sequence: (a) Add magnesium ions to the sample. (b) Dilute the sample. (c) Samples with pH values of 5.7–8.0, compared to endotoxin-free aqueous solutions, and (d) Bacterial endotoxins in the sample were determined by using the horseshoe crab amoeboid lysate assay. 4. The method of claim 3, wherein in step (a), magnesium ions are added to the sample in the form of _MgCl2 . 5. The method of claim 1, wherein the antibody is a therapeutic antibody. 6. The method of claim 3, wherein the antibody is a therapeutic antibody. 7. The method of claim 1, wherein the antibody is formulated with polysorbate 80. 8. The method of claim 3, wherein the antibody is formulated with polysorbate 80. 9. The method of claim 1, wherein the antibody is prepared with citrate buffer. 10. The method of claim 3, wherein the antibody is prepared with citrate buffer. 11. The method of claim 1, wherein the antibody is prepared with 25 mM sodium citrate buffer and 700 mg/L polysorbate 80 and has a pH of 6.5. 12. The method of claim 3, wherein the antibody is prepared with 25 mM sodium citrate buffer and 700 mg/L polysorbate 80 and has a pH of 6.5. 13. The method of claim 1, wherein the antibody is an anti-CD20 antibody, rituximab. 14. The method of claim 3, wherein the antibody is an anti-CD20 antibody, rituximab. 15. The method according to any one of claims 1-14, wherein magnesium ions are added in step (a) to a final concentration of 25 to 75 mM. 16. The method according to any one of claims 1-14, wherein in step (b) the pH of the sample is adjusted by diluting the sample with 10-50 mM Tris/HCl buffer at pH 6.0-9.0. 17. The method of claim 16, wherein in step (b) the pH of the sample is adjusted by diluting the sample with 10-50 mM Tris/HCl buffer at pH 6.0-8.0. 18. The method according to any one of claims 1-14, wherein the sample is diluted at a ratio of 1:10 in step (b). 19. The method according to any one of claims 1-14, wherein during dialysis in step (c), the sample has a pH of 6.0-8.0. 20. The method according to any one of claims 1-14, wherein in step (c) dialysis is performed at room temperature for 24 hours. 21. The method according to any one of claims 1-14, wherein for dialysis in step (c), a membrane with a molecular weight cutoff of 10 kDa is used. 22. The method according to any one of claims 1-14, wherein a cellulose acetate membrane is used for dialysis in step (c). 23. The method according to any one of claims 1-14, wherein the water is changed twice for dialysis in step (c). 24. The method of claim 3, further comprising generating a low endotoxin recovery positive control by incorporating a known amount of endotoxin into an aliquot of the sample and agitating the aliquot of the endotoxin-incorporated sample for 60 minutes to 2 hours. 25. Use of the method according to any one of claims 1-24 for making an antibody-containing sample reactive to factor C in a horseshoe crab cell lysate enzymatic cascade.