Title of Invention

"METHOD OF QUANTIFYING THE SURFACTANT OF A PROTEIN PREPARATION"

Abstract A method for removing a proteinaceous component from a liquid-phase surfactant preparation characterized in that, it comprises - (a) providing a liquid-phase surfactant preparation containing a proteinaceous component; (b) adding a complexing agent to the preparation of step (a) and allowing the complexing agent to form a complex with the surfactant; (c) simultaneously with step (b), or subsequently, adding a miscible precipitating agent to the preparation of step (a) or the product of step (b), respectively, to form a liquid-phase reaction mixture and allowing the miscible precipitating agent to precipitate the proteinaceous component within the liquid-phase reaction mixture; and (d) separating the said complex from the precipitated proteinaceous component in the product of step (c) to provide a purified liquid-phase surfactant preparation; wherein the complex remains in solution within the liquid-phase reaction mixture, and wherein step (d) retains the complex in the liquid phase, wherein the said method comprises the optional step of non-covalently binding the complex to a solid phase, preferably a solid phase extraction (SPE) medium, wherein the said method comprises the optional step of determining the surfactant content of the purified liquid-phase surfactant preparation, or a further fraction derived therefrom.
Full Text ASSAY
The present invention relates to a method for removing components from a surfactant preparation, and for assaying for a surfactant in a preparation.
Particle formation following heat treatment of protein preparations, sucli as toman serum albumin (HSA) and recombmant human albumin (rHA.), is a known problem (EP 0 341 103). Panicles are thought to form through protein denatiiration at the air/liquid interface and at other hydrophobia surfaces (Manning, M.C., Patel, K. & Borchart, R.T. (1989), Pharmaceutical Research, 6, 903-918; Thurow, H. & Gersen, K. (1984), Diabetologia, 27, 212-218). Particle formation can be inhibited by the addition of surfactants to the protein preparation.
It is possible to use polysorbate 80 at a concentration of 10-20 µg.mL-1 or more as a forrnulant for rHA final product to prevent particle formation. EP 0 341 103 discusses the use of various surfactants at concentrations of up to 50 mg.L/1 for stabilising human albumin solutions. Many other pharmaceutical protein preparations include surfactants. For example, Orthoclone™ OKT3 (Janssen-Cilag GmbH, Germany) contains polysorbate 80 at about 0.2mg-mL1; Activase™ 50 (Genentech, Inc., CA., USA) contains polysorbate 80 at Thus, surfactants can represent a significant formulant in a protein preparation including pharmaceutical protein preparations. As such, there is a regulatory requirement to assay for them in the final product. The accuracy of the assay is particularly important in the case of pharmaceutical
preparations. However, it is not possible to assess surfactant content accurately in the presence of protein because the detection techniques employed for surfactants, such as spectroscopy, high performance liquid chromatograprry, interfacial tensiometry, capillary electrophoresis, total organic carbon (TOC) titrations, and TLC etc., also detect protein. Thus the protein content leads to an over-estimation of surfactant content.
Garewal (Anal. Biochem., 54, 319-324, 1973) provided a method for assa3>ing for the surfactant content of an aqueous protein solution. The protocol teaches, as a first step, the addition of ethanol to disrupt micelles, followed by the addition of ammonium cobaltothiocyanate (ACT). The method exemplified by Garewal used an aqueous solution of the Triton X-100 surfactant, to which ACT binds and forms a blue-coloured complex. Garewal then added a non-miscible organic phase (ethylene dichloride) in which the ACT-Triton X-100 complex is soluble. The complex migrates into the organic phase and the organic phase is separated from the aqueous phase. Finally, the Triton X-100 content of the organic phase is determined by recording the spectrum of the organic phase from 580nm to 700nm; the difference in absorbance at 622 nm and 687 nm is said to be proportional to the amount of Triton surfactant present.
Garewal investigated the effect on the efficacy of the method of introducing a protein, bovine serum albumin (BSA), into the aqueous solution of Triton X-100. BSA concentrations up to 666 µg.mL-1 were investigated . Use of lower BSA concentrations, e.g. 267µg.inL-1, caused a reduction in the extraction efficacy to about 85%, but increasing protein concentration up to 666µg.mL-1 was found not to cause any further significant reduction in extraction efficacy. Garewal concluded that, since poly(ethylene oxide) groups, with which ACT reacts, are rare in biological components (e.g.
proteins'), a minimum of interference is expected and the method described therein is appropriate for bio chemical assays.
The method of Garewal has remained the method of choice for surfactant quantitation in biological preparations for the last 30 years. With minor modifications., the method of Garewal was presented at WCBP 2002, 6th Symposium on the Interface of Regulatory and Analytical Sciences for Biotechnology Health Products (January 27-30, 2002) by Lanteigne, D. & Kobayashi, K. of Biogen, Inc., Cambridge., MA USA in a poster entitled -Quantitative Determination of Polysorbate in Formulated Protein-Based Biopharmacenticals by a Direct Colorimetric Method". The poster describes an assay for polysorbate 80 (sold under the trademark "Tween 80") in a monoclonal antibody preparation at 52 mg.mL"1. Lanteigne & Kobayashi state that, where samples contain a 'high' concentration of protein (e.g. 52 mg.rnL"1)., then it is necessary to use "a protein removal step to eliminate possible interference by the active drug substance" (i.e. by the protein). Lanteigne & Kobayashi address this by ethanol precipitation of the protein in a preparation, involving overnight incubation of the sample at minus 30 °C (plus centrifugation and isolation of the supernatant), prior to complexing the surfactant by the addition of ACT and extraction of the ACT-surfactant complex using dichloromethane as an organic liquid phase.
However, following investigations, we surprisingly found that, contrary to their teachings, the Lanteigne & Kobayashi method does not provide an accurate assay for the surfactant content of protein solutions. The accuracy of tliis method, and also of the basic Garewal method, is particularly poor at a higher protein concentration. For example, as described below (see Comparative Example IX the method of Garewal gave misleading results when a sample was tested in which the protein content of a surfactant solution was greater than 50 mg-mL1. The method of Garewal is not
expected to provide an accurate surfactant assay with solutions containing
protein at 200 mg.mL-1. This is because these methods fail to remove
protein components from the sample of surfactant at the point of analysis.
Moreover, we have demonstrated that the ethanol addition steps proposed
by Garewal and by Lanteigne & Kobayashi result in unacceptably high
losses of surfactant, and so provide unreliable data. We have also
demonstrated that removing the protein content of a surfactant preparation
leads to unacceptable surfactant losses when extracting the surfactant using
the ACT/dichloromethane process described by Lanteigne & Kobayashi.
To overcome this unexpected problem, we have devised a new method for separating protein and other components from a surfactant in a given sample, thereby to provide a more complete surfactant preparation which, when analysed, provides a result that is more representative of the actual surfactant content of the original sample from which it was taken. Moreover, the method of the present invention does not require the time-consuming step of overnight incubation of a sample in order to remove protein and so is a more efficient method to perform than that described by Lanteigne & Kobayashi.
Accordingly, in a first aspect of the present invention, there is provided a method for removing proteinaceous components from a liquid-phase surfactant preparation comprising -
(a) providing a liquid-phase surfactant preparation containing a
proteinaceous component;
(b) adding a complexing agent to the preparation of step (a) and
allowing the complexing agent to form a complex with the
surfactant;
(c) simultaneously with sieP (b), ox subsequently, adding a miscible precipitating ageni to tlie preparation of step (a) or the product of step (b), respectively, to form a liquid-phase reaction mixture, and allowing the miscible precipitating agent to precipitate the proteinaceous component within the liquid-phase reaction mixture; and
(d) separating the said complex from the precipitated proteinaceous component in the product of step (c) to provide a purified liquid-phase surfactant preparation;
wherein the complex remains in solution within the liquid-phase reaction mixture, and wherein step (d) retains the complex in the liquid phase.
Any surfactant type can be punfied by a method according to the first aspect of the present invention. A surfactant is a molecule that can act to reduce the surface tension of a liquid. Surface tension is the force acting on the surface of a liquid, tending to minimise the area of the surface; quantitatively,, it is the force that appears to act across a line of unit length on the surface. The surface tension of water is 72 dyne/cm when measured at room temperature (20°) using a tensiometer; a surfactant can reduce this value, typically to a surface tension of no more than 50 dyne/cm, for example about 30-50 dyne/cm.
Typically, the surfactant will be non-ionic, i.e. having an uncharged hydrophilic head group. Examples of non-ionic surfactants include surfactants having a poly(alkylene oxide) group, such as a polyethylene oxide) group, an alcohol group or another polar group. Suitable non-ionic surfactants may have a hydrophobia group and a reactive hydrogen atom, for example aliphatic alcohols, acids, amides or alkyl phenols with alkylene oxides, especially ethylene oxide either alone or with propylene oxide.
Thus the non-ionic surfactant may be a condensate between an alkylpbenol and an alkylene oxide; a polyoxyalkylene sorbitan oleate; or a polyoxyalkylene glycol.
Specific non-ionic surfactant compounds include alkyl (C6-C22) phenols-ethyl ene oxide condensates, the condensation products of aliphatic (C8-C18) primary or secondary linear or branched alcohols with ethylene oxide, and products made by condensation of ethylene oxide with the reaction products of propylene oxide and ethylenediamine. Other non-ionic surfactant compounds include long-chain tertiary amine oxides, long-chain tertiary pbosphrne oxides and dialkyl sulphoxides. A non-ionic surfactant may also be a sugar amide, such as a polysaccharide amide, such as one of the lactobionamides described in US 5,389,279 or one of the sugar amides described in US 5,009,814. Other typical surfactants of this type include Igepal DM 730, Igepal DM 530, Igepal DM 210, Igepal CO 880, Igepal CO 530, polyoxyethyleneglycols, including compounds sold under the Trade Mark Brij (such as polyoxyethylene (4) lauryl ether (Brij 30), lauryl ether (Brij 35), polyoxyethylene (20) cetyl ether (Brij 58), polyoxyethylene (20) stearyl ether (Brij 78) and polyoxyethylene (20) oleyl ether (Brij 92)), and polyoxyethylene fatty acid esters, including compounds sold under the Trade Mark Myrj (such as Myrj 51). Typical non-ionic surfactants include polyoxyethylene octyl phenol (such as Triton X-100); alkylplienoxypolyethoxy (3) ethanol, polyoxyethylene (20) sorbitan monolaurate (Tween 20), polyoxyethylene (20) sorbitan monopahnitate (Tween 40), polyoxyethylene (20) sorbitan monostearate (Tween 60), polyoxyethylene (20) sorbitan tristearate (Tween 65), polyoxyethylene (20) sorbitan monooleate (Tween 80), polyoxyethylene (20) sorbitan trioleate (Tween 85), polyoxyethylene (20) palmitate (G2079), polyoxyethylene (20) lauryl ether; polyoxyethylene (23), polyoxyethylene (25) hydrogenated
castor oil (G1292) and polyoxyethylene (25) oxypropylene monostearate (G2162).
Other surfactants suitable for use in a method according to the first aspect of the invention may be:
amonic, with negatively charged head groups. Examples of aniomic
surfactants include long-chain fatty acids, sulphosuccinates, alky] sulphates,
phosphates and sulphonates, sucli as sodium dodecyl sulphate, sodium
cholate, sodium deoxycholate, and sodium taurocholate.
cationic, with positively charged head groups. Examples of cationic
surfactants include protonated long-chain amines and long-chain quaternary
ammonium compounds, such as hexadecyltrimethyl ammonium bromide
(Cetaylon), cetyltrimethyl ammonium bromide, and N-hexadecylpyridinium
chloride.
• amphoteric, with zwitterionic head groups. Examples of amprioteric
surfactants include betaines and certain lecithins.
The surfactants may have one or more alkylene oxide groups. Any alkylene oxide group may he present, such as ethylene oxide, propylene oxide, burylene oxide and the like. Ethylene oxide groups are common in commercially available surfactants. Multiple alkylene oxide groups may be present as a polymer (e.g. a homopolymer, co-polymer or block co-polymer), i.e. as a poly(alkylene oxide) group, such as the nomopolymeric poly(ethyl)'lene oxide) group. It is common for a surfactant to contain six or more alkylene oxide groups, although it is possible for this method to work with surfactants having fewer, such as 5, 4, 3, 2 or 1 alkylene oxide si-oup('s). The surfactant may be a non-ionic surfactant having one or more polyethylene oxide) groups, such as polysorbates, octyhphenol ethylene oxide condensate, ethylene oxide/polypropylene oxide block copolymers,
polyoxyalkylene glycols, polyoxy ethyl ene hardened castor oil, polyoxyethylene glycerol fatty acid esters, polyoxyethylene alkyl ethers, polyoxyethylene polyoxypropylene glycol, polyoxyethylene alkyl allyl ethers and the like.
Polysorbates (also known as polyoxyethylene sorbitan esters, as sold under
the Registered Trade Mark Tween) are non-ionic surfactants derived from
sorbitan esters (Becher, P. "Polyol Surfactants" in Nonionic Surfactants,
Schick, MJ. Ed. (Deldcer, New York, 1967), page 247-299; Chislett, L.R. &
Walford, J. (1976) Int Flavours Pood Addit., 7, 61; Vanna, RX et al
(1985) Arzneimittel-Forsch, 35, 804). Preferred polysorbates include
polysorbate 20, 21, 40, 60, 65, SO, 81, 85 and the like. A particularly
preferred surfactant is polysorbate 80, which has the general formula (I) -
(Formula I Removed)
Octylphenoxy polyethoxyethanol (also known as octoxynol, and sold under the Trade Marks of Triton X, Igepal CA and Polytergent G) is a non-ionic surfactant that may be prepared by reacting isooctylphenol with an alkylene oxide, such as ethylene oxide. The average number of ethylene oxide units (n) per molecule of common commercially available octoxynol typically varies between 5 and 15. The general formula is represented by formula (II) below -
Iii a typical such surfactant, sold as Triton X-l 00, n is about 9.5.
Polyethylene polypropy lene glycols (also known as poloxamers and sold under the registered trade mark Pluronic) are a series of nonionic surfactants
-with the general fomiula represented by formula (III) below —
HO(CH2CH20)a(CH-(CH3)CH20)b(CH2CH20)cH (III)
where b is at least 15 and (CH2CH2O)a+ (CH2CH2O)C is varied from 20 to 90% by weight, The molecular weight ranges from 1,000 to 16,000 g-mol-1 or more. For a review of poloxamers, see Schmolka, LR. (1967) Am. Perfumer Cosmet, 82(7), 25-30, Examples of particular poloxamers include 'Tluronic L62LF", wherein a = 7, b = 30, c = 7; "Pluronic F6S" wherein a = 75, b = 30, c = 75: and "Pluronic L 101" wherein a = 7, b = 54, c = 7.
Surfactants for use with a method according to the first aspect of the present invention may additionally contain one or more linear or branched hydrocarbon chains. Hydrocarbon chains typically found in commercially available surfactants include fatty acids. A fatty acid usually has at least six carbon atoms in the hydrocarbon backbone, and larger backbones are common, such as C16 and C18- Accordingly, the hydrocarbon chain may be an oleic acid (i.e. C16 fatty acid) group. The surfactant may contain both a poly(alkylene oxide) group and a hydrocarbon chain. For example, polysorbates contain both polyethylene oxide) groups and an oleic acid group; octoxynol comprises a branched hydrocarbon chain and
polyethylene oxide) groups.
The proteinaceous component may comprise any proteinaceous molecule that is undesired in any purified surfactant preparation that is prepared from the starting material. In particular the component may be one that interferes with the accuracy of any subsequent surfactant quantification. A component is proteinaceous if it comprises or consists of a peptide, polypeptide or protein. The phrase "peptide, polypeptide or protein" includes any polymer of amino acids, whether naturally occurring or artificial, preferably joined by peptide bonds. Preferably a peptide, polypeptide or protein will be at least 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 amino acids in length. The proteinaceous component may be a naturally occurring or recombinantly produced protein, such as albumin., an albumin fusion protein such as mentioned in WO 01/77137 (incorporated herein by reference), a monoclonal antibody, etoposide, a serum protein (such as a blood clotting factor), antistasm, tick anticoagulant peptide or any one or more of the albumin "fusion partners" disclosed in WO 01/77137, as an individual protein separate from albumin.
Unlike prior art methods, the method of the first aspect of the invention is capable of efficiently separating surfactant from a highly concentrated proteinaceous component, which for example, may be present in the liquid-phase surfactant preparation of step (a) at a concentration of at least 50, 75, 100, 150, 200 nig/ml, where component levels are measured in weight per volume of surfactant preparation.
It may be appropriate to measure the ratio of surfactant to proteinaceous component in the liquid-phase surfactant preparation of step (a). Accordingly', the ratio of surfactant to proteinaceous component, when expressed as mass of surfactant molecules per mass of proteinaceous component molecules (i.e. ppm) present in the liquid-phase surfactant preparation of step (a) may be less than 4.800 ppm, such as less than 4,500
DPm: 4:000 Ppm, 3,500 ppm, 3.000 pprn: 2,500 ppm, 2:000 ppm: 1,500 ppm. 1,000 ppm, 900 ppm, 300 ppm, 700 ppm, 600 PPm, 500 pprn, 400 ppm, 300 PPm, 200 ppm, 110 ppm, 100 ppm, 90 ppm, 80 ppm, 75 ppm, 70 ppm, 60 ppm, 50 ppm, 40 ppm, 30 ppm, 20 ppm, 18 ppm, 17 ppm, 16 ppm, 15 ppm, 14 ppm, 13 ppm, 12 ppm, 11 ppm, 10 ppm, 9 ppm, 8 ppm, 7 ppm, 6 ppm, 5 ppm or less.
The term "'liquid-phase surfactant preparation' includes any liquid-phase preparation comprising a surfactant. The preparation may be. aqueous.
By "providing", in the context of providing a liquid-phase surfactant preparation, we include taking a whole sample, an aliquot from a larger preparation, or one of batch of samples prepared from the same basic lot.
The term "complexing agent" includes any compound capable of modifying the hydrophobia characteristic of a surfactant through its ability to form weak bonds with on or more surfactant molecules. A surfactant as defined above under the conditions of the process. Typically the complexing agent will be a compound that contains a polyvalent metal ion, such as a transition metal ion. For example, the metal ion may be a group VI, VH VIII IX or X transition metal ion, such as yttrium, zirconium, niobium, molybdenum, technetmm, ruthenium, rhodium, palladium, silver, cadmium, hafnium, tantalum, tungsten, rhenium, osmium, iridium, platinum, gold, mercury, although preferred transition metal ions are 3d transition metal ions such as cobalt, iron, copper, zinc, nickel, manganese, chromium, vanadium, titanium and scandium. Cobalt compounds may be used as the complexing agent. Accordingly, the complexing agent may be ammonium cobaltothiocyanate (ACT). ACT is an appropriate complexing agent to use in order to complex a surfactant having an alkylene oxide or poly(alkylene oxide) (e.g. ethylene oxide or polyethylene oxide)) group.
Similarly, iron compounds, such as iron (III) thiocyanate, may be used as the complexing agent.
Unlike prior art methods, the method of the present invention does not rely on the formation of a colour complex to assess surfactant presence. Therefore, the complexing agents used in the present invention do not necessarily need to form coloured complexes.
An effective amount of the complexing agent is added to the liquid-phase surfactant preparation. In other words, the amount of complexing agent added is sufficient to complex substantially all of the surfactant in the liquid-phase surfactant preparation. Typically it is added in excess. The amount of complexing agent required to complex substantially all of the surfactant in the liquid-phase surfactant preparation can be determined by empirical testing of the complexing agent with an uncontaminated solution of the surfactant.
By "allowing the complexing agent to form a complex with the surfactant" we mean that at least some of the complexing agent complexes with at least some of the surfactant. Typically, after the complexing agent is added to the liquid-phase surfactant preparation, the preparation is mixed to disperse the complexing agent within the preparation. The optimum conditions to allow complexing to occur will depend on the nature of the surfactant and the nature of the complexing agent, and will typically include modification of temperature, pressure, pH and/or ionic strength of the liquid phase. Useful conditions for complexing may include neutral pH and low ionic strength (Crabb & Persinger, 1961, Journal of the American Oil Chemist's Society, 41, 752-755). For example; in the case of the surfactant being polysorbate 80 and the complexing agent being ACT, suitable conditions for allowing the complexing agent to form a complex with the surfactant are as set out below in the examples.
Following formation, the complex remains in solution within the liquid-phase. Hence under the conditions used for allowing the complexing agent to form a complex with the surfactant (but in the absence of any protein), substantially none of the complex forms a precipitate. A complex can be said to remain in solution if the amount of surfactant that can be collected in the pellet as a precipitate, by centniugation of the liquid phase at 47,800 g for 15 minutes at 4°C, is less than 20%, 15%, 10%, 5%, 2%, 1%, 0.5% or 0.1% by weight of the surfactant collected in the supernatant after centrifugation when determined using HPLC as described in the examples below. Lower percentage values are preferred.
A 'precipitating agent" is any agent that causes a component other than the surfactant to precipitate. The precipitating agent must be "miscible" within the liquid-phase surfactant preparation in order to perform its function. In other words, under the conditions used, the precipitating agent must not form a separate liquid or solid phase that is immiscible with the liquid-phase surfactant preparation. Preferably the precipitating agent is miscible in an aqueous liquid-phase surfactant preparation. To be water-miscible, a precipitating agent will commonly have a polar region. Typically the precipitating agent is an organic water-miscible solvent. Examples of water-miscible precipitating agents include polar protic solvents and polar aprotic solvents such as alcohols, cyanoalkyls, amines, amides, carboxylic acids, aldehydes, ketones, glycols, ethers, allcylhahdes and aromatic hydrocarbons. Preferred precipitating agents include acetone, acetonitrile, isopropanol, metlianol and ethanol. Acetonitrile provides a good halance between surfactant yield and contaminant cany-over. Moreover, acetoiiitrile has advantages over acetone including -
(a) the use of acetone requires the use of glassware for supernatant manipulation which could be contaminated with detergent from
cleaning, whereas the use of acetonitrile allows disposable plastic containers to be used, thereby minimising the risk of contamination; and
(b) acetone also has a flash point of -18°C, lower than typical centrifugation temperatures, and for safety it is better to use acetonitrile, which has a flash point of + 13°C.
The precipitating agent is added simultaneously or, more generally, after, but not before, the complexing agent is added to the liquid-phase surfactant preparation. This is an important difference between the present invention and the prior art. Both Garewal (op. cit.} and Lanteigne & Kobayashi (op. cit.} added a precipitating agent (ethanol) to the surfactant preparation before the complexing agent (ACT) is added. This causes some surfactant to be lost from solution, as it is carried into the precipitate. Hence, the resulting quantification of the surfactant in the supernatant is an inaccurate measure of the amount of surfactant in the starting preparation. Without being bound by theory, it is believed that by adding a precipitating agent simultaneously or, more generally, after, but not before, the complexing agent, the efficacy of the proteinaceous component removal is improved. The complexing agent keeps substantially all of the surfactant in solution whilst the proteinaceous component is precipitated.
In some cases, the subsequent addition of the precipitating agent may enhance the effect of the complexing agent and result in a greater degree of complex formation between the surfactant and the complexing agent. Without being bound by theory, we believe that this is because the precipitating agent further separates the surfactant from the proteinaceous component, thereby allowing improved complexing of the surfactant by the complexing agent.
When "allowing the miscible precipitating agent to precipitate the orotemaceous component witlnn the liquid-phase reaction mixture", the liquid-phase reaction mixture may be mcubated under conduion, that favour the precipitation of the protemaceons component but do not substantially disturb the complex. The actual conditions used will depend ON the identity of the particular components within the system in question. The person skilled in the art is capable of determining appropriate conditions for any given combination of system components by empirical testing.
The complex remains in solution withm the liquid-phase reaction mixture. In this context, the complex "remains in solution" if the amount of surfactant that can be collected in the pellet as a precipitate, by cessation of the liquid-phase reaction mixture at 47,800 g for 15 nmmes at 4°C, is less than 20%, 15%, 10%, 5%, 2%, 1%, 0.5% or 0.1% by weidit of the surfactant collected in the supernatant after centrifugation when determined using HPLC as described in the examples below. Lower percentage values are preferred.
The step of "separating the said complex from the precipitated proteinaceous component in the product of step (c)" can be effected by any suitable method known in the art for separating precipitate from a solution, so long as it "retains the complex in the liquid phase". Substantially all of the complex is retained in the liquid-phase product of step (c). For the avoidance of doubt, the complex is not retained in the liquid phase if it is partitioned into a separate non-miscible liquid phase. This is another nnportant difference between the method of the present invention and the methods of Garewal (op. cit) and Lanteigne & Kobayashi (op. cit.). The methods of Garewal (op. cit} and Lanteigne & Kobayasln (op. cit) isolate the complexed surfactant from an aqueous solution by the addrtion of an
immiscible organic phase (either ethylene dichloride or dichloromethane). Without being bound by theory, we believe that the partition of surfactant complex into a separate liquid phase results in a large contaminant carryover. By contrast, we do not rely on this form, of complex isolation and, consequently, achieve a greater removal of proteinaceous components.
The separating step is typically performed by centrifuging the reaction mixture, such that the precipitated proteinaceous component forms a pellet and the complex is retained in the supernatant, and separating the supernatant from the pellet. Optimal centrifugation parameters such as g and duration will vary depending on the nature of the precipitate formed. Guidance can be taken from the examples below, although the person skilled in the art is capable of detemining appropriate conditions by empirical testing.
However, the person skilled in the art will be aware that numerous other methods are available in the art to separate a liquid phase preparation from a precipitate, such as filtration.
The product of the separation step is a purified liquid-phase surfactant preparation. By "purified liquid-phase surfactant preparation" is included the meaning of a liquid-phase surfactant preparation that is substantially free of precipitated proteinaceous component. In this context, a liquid-phase surfactant preparation is substantially free of precipitated proteinaceous component if it can be applied to a hydrophobic solid phase extraction cartridge under conditions defined in the examples below without blocking the cartridge or significantly affecting the purity of the surfactant after SPE purification.
A method according to the first aspect of the present invention may comprise one or more additional purification steps to further purify the
In one embodiment, the method according to the first aspect of the present invention compose, the additional step of non-covalenfiy binding the complex m the purified liquid-phase surfactant preparation to a solid phase. Typically a hydrophobe solid phase is used, as this adsorbs the surfactant Alternatively, a hydroplnlic solid phase may be used, which adsorbs lemammg protemaceous component in the purified liquid-pnase surfactant preparation without retaining the surfactant thereby allowing the surfactant to be collected as an eluate.
It may be helpful if the complex is dissociated prior to exposure to the solid phase'. The sidled person is well aware of methods to dissociate the complex. The particular details depend on the nature of the surfactant and complexmg agent. For example, a delating agent may be used. Typically the chelating agent will compete with the surfactant to bind to the polyvalent metal ion of the complexmg agent. Accordingly, where the complexmg agent is ACT (i.e. the polyvalent metal ion is cobalt), a suitable method of dissociating the complex is by the addition of a chelating agent such as ethylenediamme tetra-acetic acid (EDTA) to the purified liquid-phase surfactant.
In one embodiment, the solid phase used in the additional step is a solid phase extraction (SPE) cartridge or disk.
The SPE cartridge or disk may be hydrophobic. Examples of hydrophobic SPE cartridges and disks include a polystyrene divinylbenzene (e.g. the Bakerbond SDB1 columns exemplified below, the Licrolut EN PDBV cartridges supplied by Merck, or StrartaX supplied by Phenomena) or a C2-24alkl cartridge.
In a method according to the first aspect of the invention, where the surfactant is non-covalently bound to a solid matrix, the solid matrix may be washed with a liquid that allows the bound surfactant to remain bound to the matrix whilst any remaining proteinaceous component is washed away. Suitable wash liquids are well known in the art and are commercially available. Suitable wash liquids include isopropanol, hexane and acetonitrile. It may be helpful for a wash to be acidic or alkaline. For example, acetic acid can be presented in hexane at an appropriate concentration, such as 0.1% (v/v), to provide an acidic wash. Ammonium., or triethylamine, can be presented in hexane at an appropriate concentration, such as 0.5% (v/v) ammonium or 1% (v/v) triethylamine, to provide an alkaline wash. The appropriate wash conditions can be determined by the skilled person dependent on the nature of the surfactant and the solid phase.
The matrix may be washed with a liquid that does not remove the surfactant from the matrix. Typically, an appropriate wash liquid may be sufficiently hydrophilic as to not disrupt the interaction of the surfactant with the matrix or alternatively may be sufficiently hydrophobic as to precipitate the surfactant in solution. A method of determining a suitable wash liquid can be performed as follows. A liquid is considered to precipitate a surfactant if at least 90%, 92%, 94%, 96%, 98%, 99% or substantially 100% of the surfactant (e.g. potysorbate 80) can be recovered from the matrix (e.g. a polystyrene divinylbenzene SPE cartridge, such as the Bakerbond SDB1 columns exemplified below, the Licrohit EN PDBV cartridges supplied by Merck, or StrattaX supplied by Phenomenex, or equivalents thereof) under the following conditions -
(a) The cartridge is prepared according to step 2(v)(a) of Example 2 below.
(b) The cartridge is loaded (at approximately 0.5 mL min-1 ) with 10mL
of a 15µg.mL-1 aqueous surfactant solution.
(c) The cartridge is washed with 3 x 1 mL of the wash liquid in question. The cartridge is fully dried by passing an through it under vacuum for at least 30 seconds.
(d) Surfactant on each cartridge is eluted and collected in accordance
with Example 2, step 2(v) (d)-(g) and surfactant recovery determined
using an HPLC apparatus set up in accordance with Example 2, step
2 (vi) according to the protocol laid down in Example 2, steps (vii)-
(IX).
(e) The recovery should be calculated from an extraction without the
solvent wash under investigation.
Thus a skilled person is able to select an appropriate wash liquid, depending on the nature of the surfactant and the nature of The solid matrix being used. An appropriate wash may be sufficiently strong to precipitate the surfactant on the solid phase or sufficiently weak so as to minimise or prevent elution of the surfactant. T3Tically the wash liquid will be a water-insoluble organic solvent or water-soluble organic solvent.
A suitable wash liquid, particularly in the case of a surfactant Laving a poly (alkylene oxide) (such as poly (ethylene oxide)) group (e.g. polysorbate. 80) may include nexane or the like, such as chloroform or toluene. A suitable wash liquid, particularly in the case of surfactant having a group that strondy binds the solid phase, such as a sorbitan group (e.g. polysorbate SO) can be a weak wash that does not elute the surfactant, such as acetorutnle, iscpropanol and/or tnethylamme. The dolled person will appreciate that where appropriate, these approaches can be combined. Fox-example, polysorbate 80 contains both a polyethylene oxide) group and a
sorbitan group, and so both strong and weak washes can be used. For example, we have found that the following wash can be suitable for polysorbate 80: 30% (v/v) acetonitrile followed by isopropanol, 1% (v/v) triethylamine in hexane and finally hexane.
Further washing steps may be employed depending on the nature of the surfactant, the nature of the matrix and the nature of the proteinaceous components to be removed.
Following the washing step(s), the surfactant is typically emted from the matrix and collected as an eluate. Any suitable eluent can be used. We have found a toluene :ethanol (1:1) mix provides good results in the exemplified system.
The purified liquid-phase surfactant preparation, or eluate derived therefrom, can be analysed in order to determine the surfactant content. The skilled person is well aware of methods to determine the surfactant content of a solution. For example, if a surfactant contains at least six alkylene oxide groups, then the surfactant can be complexed with ACT and surfactant concentration determined spectrophotometrically, for example as described by Garewal (op. cit.}. Alternatively, surfactant content can be determined by HPLC, or aqueous GPC, such as described in the examples below.
Due to the low levels of proteinaceous component in the tested sample, the results of the analysis correlate more closely to the actual surfactant content of the initial liquid-phase surfactant preparation than if the analysis was performed according to methods of the prior art. Preferably, the level of proteinaceous component in the tested sample is below detectable levels when assessed by HPLC using the method exemplified below.

Accordingly, a method of the present invention can he useful wherein the liquid-phase surfactant preparation used is an aliquot of a larger preparation or one sample of a hatch of preparations and the method comprises the additional step of correlating the thus determined surfactant content of the purified liquid-phase surfactant preparation, or eluate derived therefrom, with the surfactant content of the larger preparation or other hatch members.
Having made this con-elation, the user can then appropriately label the larger preparation or the other batch members, or can supply appropriate quality control reports, to reflect the thus determined surfactant content.
Since a method of the present invention provides a more accurate method for determining surfactant content than methods of the prior art, a preparation that has been subject to analysis using a method of the present invention and labelled with the thus determined surfactant content will be distinguished from prior art preparations in that its label or other associated data more accurately and more precisely reflects the surfactant level in its contents. Accordingly such a product is better able to comply with regulatory requirements.
Accordingly, in a second aspect of the present invention, there is provided a labelled liquid-phase surfactant preparation obtainable by a method as denned above. In a preferred embodiment, the liquid-phase surfactant preparation comprises a proteinaceous component, such as discussed above. Preferably the component is present in the liquid-phase surfactant preparation of step (a) at a concentration of at least 50, 15, 100, 150, 200 nag/ml or more, where component levels are measured in weight per volume of surfactant preparation.
It may he appropriate to measure the ratio of surfactant to proteinaceous component in the liquid-phase surfactant preparation of step (a).
Accordingly, the ratio of surfactant to proteinaceous component, when expressed as mass of surfactant molecules per mass of proteinaceous component molecules (i.e. ppm) present in the liquid-phase surfactant preparation of step (a) may be less than 4,800 ppm, such as less than 4,500 ppm, 4,000 ppm, 3,500 ppm, 3,000 ppm, 2,500 ppm, 2,000 ppm, 1,500 ppm, 1,000 ppm, 900 ppm, 800 ppm, 700 ppm, 600 ppm, 500 ppm, 400 ppm, 300 ppm, 200 ppm, 110 ppm, 100 ppm, 90 ppm, 80 ppm, 75 ppm, 70 ppm, 60 ppm, 50 ppni, 40 ppm, 30 ppm, 20 ppm, 18 ppm, 17 ppm, 16 ppm, 15 ppm, 14 ppm, 13 ppm, 12 ppm, 11 ppm, 10 ppm, 9 ppm, 8 ppm, 7 ppm, 6 ppm, 5 ppm or less.
It will be apparent to the slalled reader that the methods described above are useful in the quality control of a batch of a surfactant-containing pharmaceutical preparation. Quality control is a system of maintaining standards in a manufactured product by testing a sample of the output of the process of manufacture, typically a lot or batch, against a standard specification, thereby ensuring the output product meets the required standards. This is particularly important in the manufacture of ' pharmaceutical products which need to match demanding regulatory requirements. Hence, the "component" in the context of the surfactant assay is generally the desired pharmaceutically active compound. Thus, quality control of the surfactant content of the preparation may be performed by determining the surfactant content of a sample of the preparation using a method as defined above.
Accordingly, in a third aspect of the present invention, there is provided a surfactant-containing pharmaceutical preparation that has been quality-controlled using a method as described above.
The invention will now be described in more detail by reference to the following Figures and Examples wherein:
Figure 1 shows the results of an assay of poly sorb ate 80 using metal ion complex formation and solvent extraction as described in Comparative Example 1.
Figure 2 shows HP SEC chromatograms of different albumin extractions as described in Example 1.
Figure 3 shows the HPLC set up used in Example 2.

Figure 4A shows the calibration data.
Figure 4B shows the linear calibration curve generated in Example 2.

Figure 5 shows a chiomatographic profile for calculating the theoretical plate number.
Figure 6 shows a chromatographic profile for calculating peak tailing.

Figure 7 shows a chromatographic profile for calculating resolution.
Comparative Example 1
The following example is based on the method of Garewal (op. cit}.
Samples of rHA preparation at. 5 and 25% (w/v) were spited with polysorbate 80 ("Tween 80" from Sigma) to a final concentration of 15µg.mL-1 and 200µL aliquots were mixed with 2mL of ACT reagent (17.8g ammonium thiocyanate and 2.Sg"cobalt nitrate in 100mL Milli Q
water).
The mixture was then extracted with 2mL of chloroform by mixing for 15 minutes at room temperature. The chloroform was then collected and the extraction repeated with a further four 1mL aliquots of chloroform.
The absorbance at 600nm of each chloroform extract was measured and the total absorbance for each sample calculated (i.e. the total ACT complex extracted under the defined conditions above).
This extraction procedure was repeated for standard polysorbate 80 solutions (0, 0.5, 5.0 and 50.Omg.mL"1) prepared in both ultrapure Avater ("Milli Q™" water from Millipore Corp.) and rHA (5% w/v).
The results are shown in Table 1.
(Table 1 Removed)
ACT reagent when mixed with polysorbate 80 in aqueous solutions results in the formation of a coloured (blue) insoluble salt that can be extracted into organic buffers. The ACT reagent-solvent extraction of rHA produced higher background absorbances than those obtained for water (Figure 1). This indicates that rHA contains substances other than polysorbate SO that react with the ACT reagent producing a higher than expected absorbance in this method.
The estimation of polysorbate 80 in rHA final products is in excess of 4000 fold greater than the known concentration by this method (Table 1) demonstrating that the method is not accurate. The high absorbance m the unspiked rHA samples combined with the variability of the. assay presumably accounts for this result. The high and variable interfering response obtained with unspiked rHA makes tins method unsuitable for the direct assay of polysorbate. 80 in rHA final products.
Comparative Example 2
In an attempt to overcome the contaminant problem experienced when using the method of Comparative Example 1, the effect of including an additional purification step, using a C18 SPE cartridge, was assessed.
Polysorbate 80 was prepared in ultrapure water at a final concentration of 50mg-.mL-1 and in rHA at 15µg.mL1. To 200µL of each of these samples

was added 800µL of ethanol followed by 2mL of ACT reagent. The mixtures were then extracted by the addition of 5mL of chloroform followed by mixing at room temperature for 15 minutes. The chloroform extracts were then removed and extracted on C18 SPE as follows:
(Table Removed)
SPE skates were dried using centrifugal evaporation and resusp ended in either 1mL (water extracts) or 0.5mL (rHA extracts) of tetraliydrofuran (THF). The absorbance at 600nm of each resuspended eluate was then measured.
As a control a 200µL aliquot of rHA containing no polysorbate 80 was also extracted using the procedure described above, The results are shown in Table 2.
Table 2:
(Table 2 Removed)
"ND" means not determined. The recoveries of these samples were not calculated as the absorption of the unspiked rHA is equivalent to the SOmg.mL"1 standard. I.e. there is a high background from rHA samples.
Recovery of the ACT-polysorbate 80 complex from water is in excess of 80% (Table 2). This recovery of. ACT-polysorbate SO complex is maintained with SPE elution volumes from 0.25 to 1.0mL (Table 2). The use of low elution volumes may be beneficial in reducing the drying time of the eluates prior to resuspension.
faction of polvsorbate 80 from rHA using ACT complex formation, solvent enaction and SPE resulted in essential identical absences for sampbs with and without polysorbate 80 (Table 2). This indicates thai the colour produced in these extracts ts not totally related to the presence of polysorbate 80 and may be produced by rHA contacts, excipients or the protein itself.
Thus, although polysorbate 80-ACT complex can be extracted from chloroform using C,, SPE, a high background response rs also produced.
Comparative Example 3
The method of assaying surfactant levels in solutions contaimng 'high' concenttations of protein (e.g. 52 mg.mL-') described by Lanteigne. & Kobayashi (op. cit) involves ethanol precipitation of protein, rnoludrng overrent mcubaton rf the sample at rmnns 30 °C (plus centrifugation and eolation of the supernatant), prior to oomplexing the surfactant by the addvtion of ACT and extraction of the ACT-surfaotant complex using dichloromethane as an organic liquid phase.
We have found that the initial ethanol precipitation step proposed by Lanteigne & Kobayaslu results ur unacceptable levels of surfactant losses.
Ahquots (10mL) of rHA (5% w/v + lOµg.mL-1 polysorbate 80 were treated with 40mL of cold ethanol. Samples were oentrituged in a Sorval RC5C centrifuge (rotor = SS34) for 20 minutes at 20,000rpm. Supematants were to dried using rotar)- evaporation (to approxtmately 2mL) and then extracted on C18 SPE cartridges.
Removal of rHA by ethanol precipitation followed by C18 SPE did not improve the polysorbate 80 recovery above 35% (data not shown). This indicates that the precrpitanon of protem results in losses of polysorbate 80 as
the recovery on C18 SPE is lower than that obtained from water extracts (i.e. with, no protein).
The method of Garewal (op. czY.) also includes, as a first step, the addition of ethanol, albeit without the extensive overnight incubation of the sample at minus 30 °C as described by Lanteigne & Kobayashi. We have found that the initial ethanol addition step proposed by Garewal, or alternatively the addition of a similar solvent (in this case methauol or isopropanol). also results in unacceptable levels of surfactant losses.
rHA (5% w/v) + 10µg-mL-1 polysorbate 80 was prepared and 10mL aliquots mixed with 5mL of either methanol, isopropanol or ethanol. The treated samples (15mL) were then extracted as above on C18 SPE cartridges, eluates being assayed for polysorbate 80 as per Example 1. Pre-trearment of rHA final product with 30% isopropanol, methanol or ethanol prior to C18 SPE resulted in recoveries of polysorbate 80 of 7, 25 and 51% respectively. These recoveries are unacceptably low for a regulatory assay and would lead to misleading results.
Comparative Example 4
In addition to the surfactant losses observed as a result of ethanol precipitation when using the method of Lanteigne & Kobayashi (op. cit.}, described in Comparative Example 3, we have also demonstrated that the step of complexing the surfactant by the addition of ACT and extraction of the ACT-surfactant complex using dichloromethane as an organic liquid phase additionally causes surfactant loss.
Recovery of polysorbate 80 was compared between a lOmL aliquot of rHA (5% w/v) + 10µg.mL"1 Tween 80 and a 10mL aliquot of ultrapure water + 10µg.mL-1 polysorbate 80. Two additions of 70mL of ACT reagent (17.8g
ammonium thiocyanate and 2.8g cobalt nitrate hexahydrate in 100mL nitrate water, were made to each sample, prior to mixing with 5mL of dichloromethane. The samples were inclubated overnight Stores were centrifuged at 3000rpm for 5 minutes and the top aqueous phase iscarded. A few crystals of anhydrous ammonium sulphate were added and the samples were mixed and re-centnfuged as above. The dichloromethane was then n-ansferred to a dean tube and dried under a swam of helium. The rescue was then resided m 1mL memanol and dned by centrifugal evaporaton before, being resuspended in 100mL THF. These resuspended samples were to assayed forpolysorbate 80 as described in Example 1.
The recover,, of polysorhate 80 from me water sample was 82%. The re-overy of polysorbate 80 from the rHA sample was only 21%. This demonstrates that the surfactant recovery protocol of Lanteigne & Kobayashi cannot efficiently extract surfactant in the presence of
proteinaceous contaminant.
Example 1
Significant modifications to the methods used m Comparative Examples
include -
. a protenr-precipitating agent was added after, rather than before, the
addrtionof ACT ; and
. the ACT-polysorbate 80 complex was initially separated from proternaceons components usrng centrrfagation rather than solvent
extraction.
Six lots ("A" to "F") of rHA were examined. Lot F was deliberately spiked with 15µg-mL-1 polysorbaie 80. To 10mL aliquots of the rHA (250mg.mL), 2mL of ACT reagent followed by 18mL of acetone was added.
The samples were then vortex mixed and centrifuged at 47,800g for 15 minutes at 4°C. The supematants were removed and diluted with 30mL of 100mM EDTA in 0.5M Tris./HCl buffer pH 8.0 (pre-extracted on a Bakerbond SDB 200mg/3mL column). These diluted samples were then extracted by solid phase extraction (SPE) using 50mg Bakerbond SDBl columns as follows:
The SPE eluates were then dried by rotary evaporation, resuspended in 200µL of tetrahydrofuran (THF) and analysed by HPSEC, as follows:
(Equation Removed)

Only slight contamination was observed in a couple of batches (E and D') ('Figure 2). However, this can be negated in order to facilitate an accurate and precise assay by quantifying using height rather than area and/or rising a standard curve in unformulated rHA (i.e. the standard curve can be prepared using the rHA prior to Polysofbate 80 being added).
Example 2
1. Pharmaceutical preparations tested
Orthoclone™ QKT3: The product literature for Orthoclone™ OKT3 ((muromonab-CD3)-Janssen-Cilag GmbH. Germany) states that each 5mL ampoule contains amongst other ingredients 1mg potysorbate 80. For analysis, 1.25mL of product was assayed, equivalent to 0.25mg polys orb ate 80.
Vepesid™ J 100: ((etoposide)-Bristol Laboratories NJ. USA) contains amongst other ingredients 400mg polysofbate SO. For polys orb ate 80 analysis, 5uL of product was assayed.
NovoSeven™ 240: ((Coagulation factor Vila recombinant) Novo Nordisk A/S, Denmark) contains amongst other ingredients 0.65mg polysofbate 80. Reconstitution was performed as described in the product literature by the addition of 8.5mL of Sterile Water for Injection, USP. For polysofbate 80 analysis. 3mL of reconstituted product was assayed.
2. Pofysorbate 80 extraction and analysis Polysorbate 80 analysis was performed as follows:
(i) The following equipment was used: 1mL, 50mg Bakerbond SDB1
SPE cartridges (Mallinckrodt Baker B.V.); 3mL, 200mg Bakerbond
SDB1 SPE cartridges (Mallinckrodt Baker B.V.); Analytical HPLC
system with autoinjector fitted with 50uL sample loop, system
controller and integrator; Refractive index detector suitable for
HPLC system above; Phenomenex Phenogel 50A, 5µm columns
(300 x 7.8mm); Phenomenex Phenogel 50A 5µm guard column (50 x
7.8mm); HPLC column heater and control module (Waters, without
inserts); Sterilin containers (70mL); glass screw top vials (2mL -
12mm x 46mm) with lids; Univap rotary evaporative concentrator
with rotor for 12mm x 46mm vials; 250µL glass HPLC sample vials
with crimp top seals.
(ii) The following reagents were used: ammonium thiocyanate, AR grade (Fisher Chemicals); cobalt nitrate hexahydrate, AR grade (Fisher Chemicals); acetonitrile, far UV grade (Fisher Chemicals); ethylenediaminetetraacetic acid (disodium salt), Sigma Ultra grade (Sigma); tris(hydroxymethyl)aminomethane, Sigma grade (Sigma); hydrochloric acid (concentrated), SLR grade (Fisher Chemicals); hexane, Distol grade (Fisher Chemicals); tetrahydrofuran, GPC grade (Fisher Chemicals); triethylamine, AR grade (Fisher Chemicals); isopropanol, HPLC grade (Fisher Chemicals); methanol, HPLC grade (Fisher Chemicals); chloroform, HPLC grade (Fisher Chemicals); water, laboratory grade; toluene, GPC grade (Fisher Chemicals); ethanol, AR grade (Fisher'Chemicals); polysorbate SO, CAPP Raw material 34 (Surfachem); hexadecanoic acid, Sigma Ultra grade (Sigma).

Ill)

The foil owing solutions were used:
(a) Orthcdone™ OK.T3, Vepesid™ J 100, and NovoSeven™ 240 as
defined above, made up to 10mL. with laboratory grade water;
(b) Aqueous solution of reconibinant human albumin 25% (w/V)
containing 15µg.mL-3 Polys-orb ate 80 (10mL).
(c) ACT reagent (71.2g of ammonium thiocyanate and 11.2g of
cobalt nitrate hexahydrate dissolved in 20mL of laboratory grade
water, volume made up to 10OmL).
(d) Buffered EDTA Solution (37.22g of ethylenediaminetetraacetic
acid, dis odium salt (EDTA), and 60.55g of
tris(hydroxyniethyl)aniinomethane dissolved in approximately
900mL of laboratory grade water. pH adjusted to 8.0 by addition of
concentrated hydrochloric acid and volume made up to 1L). The
solution was purified using solid phase extraction as follows:

* A 3mL. 200mg Bakerbond SDB1 SPE cartridge was washed
with 6mL of THF followed by 6mL of laboratory grade water
allowing the solutions to flow through under gravity.
* A Pharmacia PI pump was used to pass 6mL of the buffered
EDTA solution through the SPE column at approximately 4
mL.min-1 and the solution discarded.
The remaining buffered EDTA solution was pumped through
the SPE column at approximately 4 mL.min"1 and collected
for use in the assay.
(e) 30% (v/V) acetonitrile (30mL of acetonitrile was mixed with 70mL. of laboratory grade water):

(f) 1% (v/v) triethylamine in hexane (200µL of triethylamine was
dissolved in 19.8mL ofliexane);
(g) toluene:ethanol (1:1) (lOmL of toluene was mixed with lOmL of
ethanol);
(h) polysorbate 80 standard solution (0.5000±0.0005g polysorbate 80 was dissolved in a final volume of 50mL of laboratory grade water in a grade A volumetric flask). Final concentration = 1 Omg.mL-1; and
(i) "System Suitability" Standard Solution (0.10g hexadecanoic acid and 0.10g polysorbate 80 was dissolved in a final volume of 10mL of . THF in a volumetric flask: Final concentration = 10mg.mL"1 polysorbate 80 and 10mg.mL-1 hexadecanoic acid; stored in 200L aliquots in glass vials at minus 20°C).
(iv) Protein precipitation and removal was performed as follows:
(a) To all tubes (standards and tests) 4mL of ACT reagent was added
and the mixture agitated gently to mix.
(b) To all tubes (standards and tests) 20mL of acetonitrile was added.
Tubes were capped and shaken vigorously to break up viscous
precipitates and vortex mixed for at least 1 minute. The rubes were
incubated at room temperature for 15 minutes.
(c) Following incubation, each sample was vortex mixed for a further
minute and then centrifuged at 47, 800g (Sorvall RC5C Centrifuge
and SS34 rotor at 20,000 r.p.m.) at 4°C for 20 minutes.
(d) To 10 Sterilin pots (70mL), 17mL of buffered EDTA solution was added.
(e) Following centrifugation.. all of the supernatant from each tube
was transferred into separate aliquots (17mL) of buffered EDTA
solution (prepared above}.
(f) Each centrifuge tube was rinsed with a further 17mL of buffered
EDT'A solution, which was added to the appropriate Sterilin pot.
This represents the purified surfactant preparation.
(Y) The purified surfactant preparation was further purified by Solid Phase. Extraction (SPE). SPE was performed as follows:
(a) Following the manufacturer's instructions, ten 1mL., 50mg Baker-bond SDB1 SPE cartridges were fitted to the SPE manifold and washed with 2mL chloroform followed by 2mL me-thanol and finally 2mL of 30% acetonitrile each.
(b)Each column was loaded (at approximately 0.5mL.min"1) with a purified surfactant preparation obtained as outlined above.
(c) Each column was washed with 2mL of 30% (v/v) acetonitrile
followed by 1mL isopropanol, 1mL of 1% (v/v) triethylamine in
hexane and finally ImL of hexane. The SPE cartridges were fully
dried by passing air through each cartridge under vacuum for at least
30 seconds.
(d) Glass screw cap 2mL vials were fitted into the SPE manifold for
eluate collection.
(e) Each column was eluted with 2 aliquots of 1000mL of
toluene:ethanol (1:1) at approximately 0.5mL.min"1. After each
aliquot the eluate was expelled into the collection tube by passing a
10mL syringe full of air through the SPE column.
(f) All eluates were dried using centrifugal evaporation at 50°C
under vacuum and then resuspended in 200µL of 1HF.
(g) Each sample was transferred to 250µL glass HPLC vials and
sealed with crimp top lids.
(vi) The HPLC apparatus was set up as shown in Figure 3. The mobile phase reservoir, containing 2L of tetrahydrofuran (THF), was placed in a thermostatically controlled water bath set to 25°C. A suction filter was connected to the HPLC Pump inlet pipe and the mobile phase primed the line up to the guard column placed to expel air. The autoinjector was connected to the guard column and then connected to the analytical columns. The guard and analytical columns were placed in the thermostatically controlled oven at 25°C as set on the Waters HPLC oven control module. One hour was allowed for temperature equilibration following installation of the columns. The outlet from the analytical column was connected to the refractive index detector inlet port, the refractive index detector reference was directed, and outlets purged into a waste container. The pump flow rate was set to 1.0 mL.min" and the refractive index detector was set to a sensitivity of 256 and a time constant of 10 seconds. The detector oven was set to 35°C. The HPLC system controller and integrator were set to collect and integrate the chromatographic data following the manufacturer's instructions. Prior to analysis, system suitability tests were run (see below).
(vii) The following procedure was used for HPLC analysis: refractive index detector was purged for at least one hour prior to use and monitored for a steady baseline; buffered EDTA solution was prepared and its extraction started; all other assay buffers were prepared; extraction of tests and standards was started; the HPLC
system was started and equilibrated; the baseline on HPLC was checked and test solution prepared; a system suitability test was run; while samples were being processed through SPE, the HPLC system was examined to ensure that the system suitability test was acceptable; when extraction was completed, and system, suitability-was acceptable, the samples were run.
(viii) AD extracted standard curve of polys orb ate 80 was prepared from 0.00, 0.10, 0.20, 0.30, 0.40 and 0.50mg polysorbate 80 in 10mL lab oratory water.
(ix) Polysorbate 80 was quantified by HPLC as follows:
(a) Immediately after performing the system suitability7 test, 50µL of
each sample was injected onto the HPLC under the standard
conditions described above.
(b) After chromatograplry of all samples and the integration of the
polysorbate 80 peaks, a linear calibration curve of polysorbate 80
peak height against standard concentration (mg.mL"1) was
constructed for the standards by perfonning linear regression to
calculate slope (m) and intercept on the x-axis (c) for the standard
curve. These regression data were used to calculate the
concentration of polysorbate 80 in the test samples as follows:
Line of best fit for standard curve is polysorbate 80 peak height in test = mx + c
Where x = polysorbate 80 concentration (µg.mL-3)
Thus: (Equation Removed)
for the test replicates was then calculated.
Results
The extracted standard curve generated a linear calibration curve with a regression line R2 of 0.999 and a percentage CV for the normalised peak heights of 3.4% (Figure 4). Comparison of the measured polysorbate 80 mass against the stated formulation mass using the calibration curve showed
close agreement for Albumin,
Table 3:
(Table 3 Removed)
The method of quantifying polysorbate 80 in rHA final product as described above showed itself to be suitable for the quantitation of all products with no modification to the methodology.
Sysr.au SuiiabUm- Tcsis Procedure:
1. 50µL of the System Suitability Standard Solution was injected onto the
HPLC running under the standard conditions.
2. The test sample was evaluated by calculating the theoretical plates,
tailing and resolution for polysorbate 80 and hexadecanoic acid (see
below).
3. If any one of the parameters, either for the polysorbate 80 or
hexadecanoic acid, was below the expected value the. columns were
replaced with a new set of columns.
Evaluation Of Test Sample
1. The theoretical plate number was calculated for both the polysorbate 80 (first eluting peak) and hexadecanoic acid (second eluting peak) peaks using Equation I, with reference to Figure 5.
(Equation Removed)

2. The peak tailing was calculated for both the polysorbate 80 and hexadecanoic acid peaks using Equation 2, with reference to Figure 6.

(Equation Removed)

4. Tile resolution between polys orb ate 80 and hexadecanoic acid was calculated using Equation 3, with reference lo Figure 7.
(Equation Removed)



WE CLAIM:
1. A method for removing a proteinaceous component from a liquid-phase surfactant
preparation characterized in that, it comprises -
(a) providing a liquid-phase surfactant preparation containing a proteinaceous
component;
(b) adding a complexing agent to the preparation of step (a) and allowing the complexing
agent to form a complex with the surfactant;
(c) simultaneously with step (b), or subsequently, adding a miscible precipitating
agent to the preparation of step (a) or the product of step (b), respectively, to form a
liquid-phase reaction mixture and allowing the miscible precipitating agent to precipitate
the proteinaceous component within the liquid-phase reaction mixture; and
(d) separating the said complex from the precipitated proteinaceous component in the
product of step (c) to provide a purified liquid-phase surfactant preparation;
wherein the complex remains in solution within the liquid-phase reaction mixture, and wherein step (d) retains the complex in the liquid phase,
wherein the said method comprises the optional step of non-covalently binding the complex to a solid phase, preferably a solid phase extraction (SPE) medium,
wherein the said method comprises the optional step of determining the surfactant content of the purified liquid-phase surfactant preparation, or a further fraction derived therefrom.
2. A method as claimed in claim 1 wherein the surfactant contains one or more alkylene oxide groups.
3. A method as claimed in claim 1 or 2 wherein the surfactant is nonionic, and is preferably a condensate between an alkylphenol and an alkylene oxide; a polyoxyalkylene sorbitan oleate; or a polyoxyalkylene glycol.
4. A method as claimed in claim 2 or 3 wherein the complexing agent comprises a polyvalent metal ion, preferably a transition metal ion.
5. A method as claimed in any preceding claim wherein the precipitating agent is an aqueous organic miscible solvent, such as an alcohol, cyanoalkyl, amine, amide, carboxylic acid, aldehyde, ketone, glycol, ether, alkylhalide or aromatic hydrocarbon, for example acetone, acetonitrile or ethanol.
6. A method as claimed in to any preceding claim wherein the proteinaceous component comprises a peptide, polypeptide or protein.
7. A method as claimed in claim 6 wherein the proteinaceous component comprises albumin, an albumin-containing fusion protein, a monoclonal antibody, etoposide or a blood clotting factor.
8. A method as claimed in any preceding claim wherein the protein concentration of the
proteinaceous component in the liquid-phase surfactant preparation of step (a) of Claim 1 is at
least 50 mg/ml.
9. A method as claimed in any preceding claim wherein the surfactant is present at less than
4800 ppm relative to the proteinaceous component in the liquid-phase surfactant preparation of
step (a) of Claim 1.
10.A method as claimed in any one of the preceding claims, wherein the step of providing a purified liquid-phase surfactant preparation comprises centrifuging the reaction mixture, such that the precipitated proteinaceous component forms a pellet and the complex is retained in the supernatant, and separating the supernatant from the pellet.
11. A method as claimed in claim 1, wherein the complex is optionally dissociated prior to the step of non-covalent binding of the surfactant to the solid phase.
12.A method as claimed in claim 11, wherein the complex is dissociated by the addition of a chelating agent to the purified liquid-phase surfactant.
13. A method as claimed claim 1, wherein the solid phase is a hydrophobic SPE medium.
14. A method as claimed in claim 13, wherein the SPE medium is a polystyrene divinylbenzene or a C2-24 alkyl medium.
15.A method as claimed in claim 1, wherein the surfactant that is bound to the solid phase is optionally washed to remove residual proteinaceous component.
16.A method as claimed in claim 15, wherein the matrix is washed with a water-soluble organic solvent, such as acetonitrile, isopropanol and/or triethylamine.
17.A method as claimed in claim 15, wherein the solid phase is washed with a liquid that would precipitate the surfactant if it were in solution.
18.A method as claimed in claim 17 wherein the solid phase is washed with a water-insoluble organic solvent, such as hexane, chloroform or toluene.
19.A method as claimed in claim 1, wherein the surfactant is optionally eluted from the solid phase and collected as an eluate.
20. A method as claimed in claim 1, wherein the liquid-phase surfactant preparation of step (a) is an aliquot of a larger preparation or is one sample of a batch of preparations and the method comprises the optional step of correlating the thus determined surfactant content of the purified liquid-phase surfactant preparation, or a further fraction derived therefrom, with the surfactant content of the larger preparation or other batch members.
2 LA method as claimed in claim 20 comprising the optional step of appropriately labelling the larger preparation or the other batch members to reflect the thus determined surfactant content.
22. A method as claimed in any one of claims 1 to 21, whenever used for quality control of a batch of a surfactant-containing pharmaceutical preparation comprising determining the surfactant content of a sample of the preparation.


Documents:

4995-DELNP-2005-Abstract-(01-09-2009).pdf

4995-DELNP-2005-Abstract-(20-01-2009).pdf

4995-delnp-2005-abstract.pdf

4995-DELNP-2005-Claims-(01-09-2009).pdf

4995-DELNP-2005-Claims-(20-01-2009).pdf

4995-delnp-2005-claims.pdf

4995-DELNP-2005-Correspondence-Others-(20-01-2009).pdf

4995-delnp-2005-correspondence-others.pdf

4995-DELNP-2005-Description (Complete)-(20-01-2009).pdf

4995-delnp-2005-description (complete).pdf

4995-DELNP-2005-Drawings-(20-01-2009).pdf

4995-delnp-2005-drawings.pdf

4995-DELNP-2005-Form-1-(20-01-2009).pdf

4995-delnp-2005-form-1.pdf

4995-delnp-2005-form-18.pdf

4995-DELNP-2005-Form-2-(20-01-2009).pdf

4995-delnp-2005-form-2.pdf

4995-DELNP-2005-Form-3-(20-01-2009).pdf

4995-delnp-2005-form-3.pdf

4995-delnp-2005-form-5.pdf

4995-DELNP-2005-GPA-(20-01-2009).pdf

4995-delnp-2005-gpa.pdf

4995-delnp-2005-pct-101.pdf

4995-delnp-2005-pct-210.pdf

4995-delnp-2005-pct-220.pdf

4995-delnp-2005-pct-237.pdf

4995-delnp-2005-pct-304.pdf

4995-delnp-2005-pct-308.pdf

4995-DELNP-2005-Petition-137-(20-01-2009).pdf


Patent Number 236302
Indian Patent Application Number 4995/DELNP/2005
PG Journal Number 43/2009
Publication Date 23-Oct-2009
Grant Date 15-Oct-2009
Date of Filing 02-Nov-2005
Name of Patentee NOVOZYMES BIOPHARMA UK LIMITED
Applicant Address KROGAHOEJVEJ 36,DK-2880 BAGSVAERD,DENMARK
Inventors:
# Inventor's Name Inventor's Address
1 PHILIP HARVEY MORTON DELTA BIOTECHNOLOGY LIMITED, CASTLE COURT, 59 CASTLE BOULEVARD, NOTTINGHAM NG7 1FD, GREAT BRITAIN
2 JASON CAMERON DELTA BIOTECHNOLOGY LIMITED, CASTLE COURT, 59 CASTLE BOULEVARD, NOTTINGHAM NG7 1FD, GREAT BRITAIN
PCT International Classification Number G01N 31/22
PCT International Application Number PCT/GB2004/001992
PCT International Filing date 2004-05-07
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 0310347.0 2003-05-07 U.K.