WO1993022679A1

WO1993022679A1 - Improved wash technique for immunoassay

Info

Publication number: WO1993022679A1
Application number: PCT/US1993/004190
Authority: WO
Inventors: Linda Ann Mauck; Margaret Jeannette Smith-Lewis; Stuart Gilmour Macdonald
Original assignee: Eastman Kodak Company
Priority date: 1992-05-06
Filing date: 1993-05-04
Publication date: 1993-11-11
Also published as: EP0596074A1; JPH06509177A

Abstract

The present invention provides an improved wash technique for an immunoassay. The improved wash technique provides a more effective displacement of unbound label away from the detection area, thereby providing an assay having greater sensitivity. The improved wash technique of the invention comprises applying 8-10 νL of the wash liquid at a point up to 43 % of the distance in betweenthe center of the sample spot and the inner periphery of the area wetted by the sample. When applying the wash fluid in this manner, the central portion of the spot will be adequately washed of unbound label, while the excess label at the periphery will not be redistributed back towards the center.

Description

IMPROVED WASH TECHNIQUE FOR IM UNOASSAY FIELD OF THE INVENTION

The present invention relates generally to immunoassays and, in particular, to an improved wash technique for an immunoassay.

BACKGROUND OF THE INVENTION

Immunoassay is a well-recognized technique for qualitative or quantitative assay of antibodies and antigens. The basis for all immunoassay techniques is the unique, immunological phenomena whereby a specific antibody recognizes and binds to a specific antigen.

The immunoassay is conducted on a dry analytical element. Such elements typically include a porous spreading layer which contains antibodies immobilized therein that act as adsorption sites for an analyte of interest such as a drug (D) and chemical label (L) . In practice, a sample that contains drug D is applied to the spreading layer as a spot. The sample quickly penetrates the pores of the spreading layer and simultaneously reconstitutes (dissolves) the chemical label L which has either been previously coated and dried onto the top of the spreading layer or applied as a separate solution. Alternatively, the chemical label can be added to the sample and the mixture applied to the element. Ideally, the label L becomes uniformly distributed in the liquid sample regardless of how it is applied. Thus, when the sample application process is complete, the spreading layer is saturated with a uniform solution of D & L. Because immobilized antibody is present throughout the spreading layer, D & L begin to bind to these adsorption sites. In order to detect the level of D in the sample, there must be a competition between D & L for the limited number of antibody adsorption sites in the layer (competitive assay) . That is, the amount of D is determined from its relative ability to compete against L for adsorption sites available in the spreading layer.

In an alternative type of immunological assay, commonly referred to as a sandwich assay, an antibody is contacted with a sample containing an analyte (D) to cause the analyte to bind to the antibody. This complex is then contacted with a solution of a labeled antibody which reacts with the bound analyte. The amount of bound labeled antibody is thus directly proportional to the amount of bound analyte.

To measure D^rs success in competing for antibody sites, the amount of L bound to antibody is measured. The amount of adsorbed D can be obtained from a simple difference. Unfortunately, when L binds to antibody or ligand (sandwich assay) it does not develop any unique characteristic that distinguishes itself from unbound L, except of course that it is unable to move with the solution. Therefore, to measure the amount of L bound to the immobilized antibody, the L in solution has to be washed away from the region to be read by the detection device, which senses the presence of L. A wash step is generally included in the assay procedure which facilitates removal of unbound label. Typically, this wash step is carried out in conventional immunoassays by applying about 30 to 100 μl of wash fluid to the center of the area where the sample was spotted. As the wash fluid flows into the spreading layer, it displaces (pushes) the sample fluid containing dissolved L outward.

Ideally, the wash fluid is added fast enough to sweep away only the L that is in solution, without giving the adsorbed L a chance to dissociate from the immobilized antibody. The amount of L in this washed area then will be equal to the amount of adsorbed , which in turn is proportional to the amount of D originally in the sample. The amount of bound labeled ligand (antibody or antigen) can thus be correlated to the concentration of analyte in the biological fluid.

Unfortunately, the flow of wash fluid into the spreading layer does not follow the idealized format described above. Instead, wash fluid enters along the contact line and the unbound L is washed out primarily in the annular region of the spreading layer just below the contact line of the fluid meniscus (lens) created by the wash fluid on the top surface of the layer. Consequently, this flow process causes a significant fraction of unbound■L to remain in the center of the spot. This inefficient washing process sometimes referred to as "targeting" is accentuated if the wash fluid is applied to the top surface of the layer at a rate that is much faster than the rate of absorption by the spreading layer. In this case, the wash fluid forms a lens of substantial volume and the contact line is moved further away from the central region, thereby leaving an even larger unwashed region in the center.

Unbound L that remains in the central region of the wetted area due to nonuniform or incomplete washing has a detrimental effect on the analytical technique, since the detection scheme cannot distinguish between the bound and unbound "free" label. A stronger signal is obtained in the center because there is more L there than expected. This not only reduces the sensitivity of the assay, it also makes the measurement sensitive to the position of the slide in the instrument during the optical scan and to the interaction of the wash fluid with the metering tip.

One approach to maximize wash-out of unbound label from the center of the spot during washing, is to add the wash liquid very slowly so that no lens of significant size is formed by the wash fluid. However, when wash liquid is added slowly, it dribbles onto the spreading layer as many small volume increments. Each time wash fluid touches the top of the spreading layer and detaches from the equipment (metering tip) , the associated surface stresses disrupt the physical integrity of the spreading layer. The cumulative damage done by these many volume increments to the center of the spreading layer further compromises the sensitivity of the assay.

Another approach to maximize wash-out of unbound label is disclosed in Japanese Kokai Patent Application No. Sho (1986)-126470 wherein wash solution (i.e., 30-60 μl) is applied at a position outside the periphery of the area wetted by the specimen. Japanese Kokai No. Sho 61 (1986)-62865 discloses a method of washing wherein relatively large volumes of (50-100 μl) of wash solution is dropped at a point 2/3 the distance between the central portion of the region where the specimen is applied to twice the distance between the central portion and the peripheral portion.

Applying the wash solution in the range suggested by the above Kokai causes some imprecision. Applying wash solution to a position outside the periphery of the area wetted by the specimen, unbound label at the periphery of the wetted area will be washed back into the center. This will reduce sensitivity of the assay as previously described.

Also, a substantial volume of wash liquid is necessary because wash liquid quickly flows away from the periphery into the dry area of the spreading layer of the element. However, such large volumes of wash solution cannot be used with dry multilayer immunoassay elements. The latter elements have a maximum volume capacity of 30 μl. Therefore, a need exists for a wash technique which overcomes the above problems. SUMMARY OF THE INVENTION

The present invention provides a new and improved wash technique for immunoassay of a fluid sample, wherein unbound label is effectively displaced away from the detection area, thereby increasing the level of sensitivity of the immunoassay.

The improved wash technique of the invention comprises applying the wash liquid at a position in between the center of the sample spot and the inner periphery of the area wetted by the sample. When applying the wash fluid in this manner, the central portion of the original sample spot will be adequately washed of unbound label, while the excess label at the periphery will not be redistributed back towards the center. In addition, any optional artifact caused by the stagnant zone immediately beneath the wash application will not be viewed by the optical read.

In a preferred embodiment, there is provided a method of conducting a solid phase immunoassay of a fluid sample containing an unknown level of analyte, comprising: applying a liquid sample containing an analyte to an analytical element of the type having at least one porous zone, said element containing an immobilized binding agent capable of reacting with the analyte, the sample forming a wetted area on the element, the wetted area being defined by a sample spot at the center thereof coinciding to the point of application of the sample and a ring at the periphery thereof; supplying a label to the element; applying 8 to 30 μL a stream of washing solution to the element at a position in between the center of said sample spot and up to 50% of the distance between the center of said spot and the inside periphery of said ring, said washing solution effectively displacing unbound label away from the center area of said spot; and measuring the extent to which bound label is present within the spot area.

It is understood that the label can be supplied to the element by either coating it in the element, providing it with the sample or as a separate solution, so long as inthe competitive assay, there is no pre-contact (binding) between the label and antibody.

BRIEF DESCRIPTION OF THE DRAWING FIGURES

Fig. 1 is a diagram which illustrates a wash process under ideal conditions.

Fig. 2 is a diagram which illustrates the distribution of unbound label where wash solution was applied at the center.

Fig. 3 is a diagram which illustrates the distribution of unbound label during sample addition where label is coated in the element.

Fig. 4 is a diagram which illustrates the distribution of unbound label where wash solution was applied at the periphery of the area wetted by the sample.

Fig. 5(a) is a top view of a an immunoassay element after sample has been added. Fig. 5(b) is a magnified edge view of

Fig. 5(a) illustrating distribution of label (L) and drug (D) in solution, surrounding beads that are coated with antibody (A) .

Fig. 6 is a diagram which illustrates the wash technique of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides an improved washing technique for immunoassays. The wash liquid is applied at a position "off-center" to the spot of application of the sample. As used herein, the term "off-center" means the wash liquid is applied at a -1- position in between the center of the sample spot and up to 50% of the distance between center of the sample spot and the inner periphery of the area wetted by the sample. As used herein, the term "ring" means the area on the element that defines the boundary between that portion of the element which is wetted by the sample and that portion of the element which is dry. As used herein, the term "spot" or "sample spot" means the area on the element coinciding to the point of sample application.

It is important to the understanding of the invention to first consider various wash procedures currently used in immunoassays. With reference to Figure 1, there is shown a diagram which illustrates the objective of the wash process under ideal conditions. The symbols L (label) and D (drug) which are connected by a line to the antibody (A) are bound species which are immobilized. The unbound or "free" label and drug are free to move with the wash solution to the periphery 18. Wash fluid is added using any conventional means 10 to the element 12 at 14. In a perfectly designed wash process, all free label should wash out from the center 16, the only remaining label in the center of the spot being that bound to antibody. The wash procedure most typically employed in currently available immunoassays consists of applying wash solution directly to the center of the element. That is, to the central area at a position directly overlapping the sample application point. With reference to Figure 2, there is illustrated the distribution pattern when wash fluid is applied at the center. It is apparent that nonuniform distribution of unbound label results from the natural flow of wash liquid into the spreading layer along the contact line 22 at the periphery of the fluid meniscus (lens) 24. Basically, as wash fluid 26 enters the test element, the diameter of the lens decreases at any point in time and fluid enters the spreading layer only at the periphery 22 (contact line) of this fluid lense. Thus, any label substantially within that diameter will not be displaced until the lens shrinks. By that time: (1) far less wash fluid remains; (2) displacement distance has increased; and (3) solubilization of unbound and bound fractions may have changed. The unbound label that remains in the central area diminishes the sensitivity of the analytical technique to the particular analyte because unbound label is indistinguishable from label that is bound to antibody (not shown) in the center.

It is understood that label can already be coated in the spreading layer before sample is added to the element. This is the preferred method of introducing the label to the element and is highly desirable to the alternatives of either premixing the label with the sample just prior to spotting the sample onto the element or adding the label as a separate solution.

The same non-uniform flow that occurs during the wash step (Fig. 2) can also occur during the sample application step. With reference to Figure 3, there is illustrated the non-uniform distribution of label L that is produced during sample addition. The liquid sample 28 enters the dry spreading layer in a non- uniform manner, the natural flow of liquid being primarily along the contact line 30 at the periphery of the lens 32 as shown. This flow pattern redistributes the label nonuniformly by washing out the label along the periphery. Consequently, there exists an annular region surrounding the center of the wetted area that generates a signal which is less sensitive to the amount of drug present in the sample because of the absence of sufficient label to compete effectively with analyte for the available adsorption sites on the antibody.

Referring to Figure 4, the effects of applying the wash liquid 36 at the periphery 34 of the area wetted by the sample is illustrated. Using this procedure, unbound label at the periphery is washed back into the center (shown by arrow) where it reduces the ability to detect the amount of bound label in this area, thereby reducing sensitivity. Basically, there are two features of the practical embodiments of the conventional wash techniques that diminish the ability to detect drug in a sample, and both features derive from the nonuniform flow pattern that inevitably occurs as liquid enters a spreading layer. Because the liquid applied to the top of a spreading layer does not enter the porous medium uniformly, but rather enters primarily along the circular edge of the lens on top of the layer: (1) label is washed out of an annular region near the center of the spot during sample addition, leaving a nonuniform distribution of label in the spreading layer to compete with a uniform distribution of drug for a limited number of binding sites; and (2) unbound label is not washed out of the center of the spot during the washing step, leaving excess label in the center that is unrelated to the amount of drug present.

The off-center wash technique of the present invention is an improvement of the wash methods described above. In a preferred embodiment, and with reference to Figure 5, sample 38 containing drug (D) is spotted onto the spreading layer 40 of the element using conventional techniques. The magnified edge view (Figure 5 (b) ) , provides a representation of the label L and drug D in solution, surrounding beads that are coated with antibody (A) . During the initial incubation period, L and D compete for the limited number of antibody-adsorption sites available for binding. Because of the prevailing fluid flow pattern, the reconstituted label is redistributed. This effect can be minimized by making the spreading layer highly wettable, so that the lens of liquid spreads out to a thin film. After the system has been incubated (e.g., for about 5 minutes) to allow the competitive binding process between drug D and label L to reach equilibrium, wash fluid is applied to the spreading layer to wash the free label away from the label that is bound to the immobilized antibody. The off-center wash technique of the invention is described above as used in a competitive assay. It is understood that the off-center wash technique is applicable to any assay, for example, a sandwich assay, which requires a wash step.

As shown in Figure 6, the wash fluid 38 is applied at a position in between the center of the spot (application point of sample) and inside the periphery 42 of the area wetted by the drug sample. Using the present wash technique, the central portion of the spot is adequately washed of unbound label, while the excess label at the periphery is not redistributed towards the center. The benefits of using the wash technique of the invention include greater reproductibility (precision) because targeting effects due to sample addition are not complimented by additional targeting associated with the wash process. Furthermore, the targeting that does occur during washing is not in the optical read area, which makes the measurement less sensitive to small variations in the slide positioning and fluid metering processes. For some assays, greater sensitivity can be expected because more efficient washing of unbound label reduces background signal levels in the assay technique. In addition, the spreading layer can be made more porous or more wettable without great concern over the toughness of the layer's surface, because the inevitable damage done to the spreading layer by the application of wash fluid will be outside the viewing area of the optical sensor.

It is understood that the novel wash technique of the invention may be incorporated into the procedure of any solid phase immunoassay known in the art. Furthermore, any fluid sample can be analyzed using the novel wash technique, such as biological fluids of either animals or humans including, but not limited to, whole blood, plasma, sera, lymph, bile, urine, spinal fluid, sputum, perspiration and the like as well as stool secretions. It is also possible to assay fluid preparations of human or animal tissue such as skeletal muscle, heart, kidney, lungs, brains, bone marrow, skin and the like. In a preferred embodiment of the solid phase immunoassay, the liquid sample is applied to an analytical element of the type having at least one porous spreading zone or layer, said element containing an immobilized binding agent capable of reacting with the analyte of interest and a label (supplied in any manner previously described) which is capable of either competing with the analyte for binding sites on the binding agent coated therein or binding with a binding agent-analyte complex, forming a sandwich. It is desirable that the spreading zone be isotropically porous, meaning that the porosity is the same in each direction in the zone as caused by interconnected spaces or pores between particles, fibers or polymeric strands. The spreading zone can have two or more discrete zones, either in the same layer or as superimposed layers. The zones can be reagent zones, capture zones or registration zones as those zones are known in the art, additional spreading zones, radiation-blocking or filter zones, subbing zones, barrier zones, etc. The zones are generally in fluid contact with each other, meaning that fluids, reagents and reaction products (e.g., color dyes) can pass or be transported between superposed regions of adjacent zones. Preferably, the zones are separately coated layers, although two or more zones can be a single layer. The zones can be self supporting (i.e., composed of a material rigid enough to maintain its integrity) , but preferably are carried on a separate support. Such a support can be any suitable dimensionally stable, and preferably, nonporous and transparent (i.e., radiation trans issive) material which transmits electromagnetic radiation of a wavelength between about 200 and about 900 nm. A support of choice for a particular element should be compatible with the intended mode of detection (transmission or reflectance spectroscopy) . Useful supports can be prepared from paper, metal foils, polystyrene, polyesters [e.g..., poly(ethylene tere- phthalate) ] , polycarbonates, cellulose esters (e.g., cellulose acetate) and others known in the art. The porous spreading zone can be prepared from any suitable fibrous or non-fibrous material or mixtures of either or both. The void volume and average pore size of this zone can be varied depending upon the use intended.

Useful spreading zones can be prepared using fibrous materials, either mixed with a suitable binder material or woven into a fabric, as described in U.S. Patent No. 4,292,272 (issued Sept. 29, 1981 to Kitajima et al.), which disclosure is hereby incorporated by reference. Alternatively, and preferably, the spreading zone is prepared from polymeric compositions (for example, blush polymers) or particulate materials, as described in U.S. Patent Nos. 3,992,158 (issued Nov. 16, 1976 to Przybylowicz et al.), 4,258,001 (issued March 24, 1981 to Pierce et al.), and 4,430,436 (issued February 7, 1984 to Koyama et al.) and Japanese Patent Publication 57 (1982)-101760 (published June 24, 1982), which disclosures are hereby incorporated by reference.

A water-insoluble adhesive can be employed in the invention to bond the organo-polymeric particles to one another to provide the coherent, three-dimensional structure of the spreading zone. The adhesive is composed of an organic polymer different from the specific polymer contained in the spreading zone, although quite commonly the adhesive represents a polymer containing many repeating units which are identical or similar to some of those present in the polymer composition of the spreading zone. Preferred water-insoluble adhesives for use in the invention are addition homopolymers and copolymers, particularly addition copolymers prepared from an addition polymerizable blend of monomers, although any known adhesive in the art may be used herein. For example, reference is made to U.S. Patent No. 4,258,001, which disclosure is hereby incorporated by reference. A leuco dye can also be incorporated into the spreading zone. Any suitable leuco dye can be used in the practice of this invention as long as it is capable of providing a detectable dye when oxidized in the presence of the label. Examples of useful leuco dyes include, but are not limited to, imidazole derivatives such as those described in U.S. Patent No. 4,089,747 and references noted therein, European Patent Application No. 122,641 (published October 24, 1984) and Japanese Patent Publication No. 58 (1983) -045, 557, and the triarylmethanes described, for example, in U.S Patent No. 4,670,385, which disclosures are hereby incorporated by reference.

In a preferred embodiment, the present invention is useful for the determination of an immunologically reactive ligand, which is a substance that will complex specifically with a corresponding reactant. As used herein, the term "ligand" is interchangeable with analyte. Such ligands include, but are not limited to, antigens, haptens, antibodies, toxins, hormones, therapeutic drugs, natural and synthetic steroids, proteins, viruses, bacteria, peptides, nucleotides, etc. The ligand (analyte)-to be determined and corresponding labeled ligand analog compete for, or form a sandwich with, a fixed amount of binding agent. The ligand analog can comprise a ligand covalently bonded to a suitable label. Labeled ligand analogs useful in this invention can be prepared using any signal generating label and suitable technique known to one skilled in the art. For example, conventional labels include radioactive tags, enzymes, chromophores, fluorophores, and enzyme cofactors and effectors. Enzymes may be measured by reacting the labeled ligand with a substrate, which by the action of the enzyme, releases a chromogenic or fluorogenic substance that can be measured by conventional techniques or that activates a chemiluminescence substance which gives off light with an intensity proportional to the analyte concentration. Ligands labeled with enzyme cofactors or effectors can be detected similarly by their effect on enzyme action on a substrate. Compounds labeled with chromophores, fluorescent and chemiluminescen compounds may be directly measurable, e.g. by fluorescence, ultraviolet spectroscopy or other spectroscopic means.

The labeled ligand analog and corresponding binding agent can be incorporated into the spreading zone prior to use, or added at the time of the assay. Preferably, both are incorporated into the spreading zone prior to use. More particularly, the binding agent can be immobilized within the porous spreading zone on a carrier material, such as glass or polymeric beads or other particles, resins, fibers and the like. Alternatively, the binding agent can be incorporated into an intermediate layer or zone separate from the spreading zone. Additionally, the labeled ligand analog can be incorporated into a separate water- soluble zone or layer in order to isolate it from the reactant. The binding agent immobilized in the spreading zone or alternative intermediate layer or zone can comprise any known binding agent in the art, for example, phosphorylcholine, avidin, biotin, thyroxin, thyroxin binding globulin, a polysaccharide etc., an antigen, an antibody, any known receptors such as estrogen receptor, and preferably is an antibody. Elements can be configured in a variety of forms, including elongated tapes of any desired width, sheets, slides or chips. Further, the assay can be manual or automated. In general, in using the dry elements, analyte (ligand) determination is made by taking the element from a supply roll, chip packet or other source and physically contacting it with a sample (e.g., 1-30 μl) of the liquid suspected of containing the analyte so that the sample and reagents within the element become mixed. Such contact can be accomplished in any suitable manner, e.g., dipping or immersing the element into the sample or, preferably, by spotting the element by hand or machine with a drop of the sample with a suitable dispensing means.

After sample application, the element is exposed to any conditioning, such as incubation, heating or the like, that may be desirable to quicken or otherwise facilitate obtaining any test result to form a complex between the ligand and ligand analog and the binding agent. Once this reaction has taken place, the wash technique of the present invention is applied off-center as previously described. The wash solution can comprise any known solution in the art used for washing, and in some embodiments is preferably a buffer solution including a buffering agent and optionally an electron transfer agent, peroxide, chelating agent and/or surfactant.

The amount of ligand in the test sample is then determined by using the appropriate detecting apparatus, depending on the particular label used. As previously mentioned, any label and detection method known in the art may be used herein.

The following examples are provided to illustrate the practice of the invention.

EXAMPLE I

C-Reactive Protein (CRP) Assay The following formulation is for a C-reactive protein element in accordance with the present invention:

For purposes of the above element, the following definitions apply:

TES = N-[Tris(hydroxymethyl) ethyl] -2- aminoethane-sulfonic acid buffer.

Dimedone = 5,5-Dimethyl-l,3-cyclohexanedione. Triarylimidazole leuco dye = 4,5-Bis(4-dimethylaminophenyl)-2-

(4-hydroxy-3,5-imethoxphenyl)imidazole blue forming leuco dye.

Adhesive = Poly(methyl acrylate-co-sodium 2- acrylamido-2-methylpropanesulfonate- co-2-acetoacetoxy-ethyl methacrylate) . Polymer beads Poly(m-&-p-vinyltoluene-co- methacrylicacid) .

Zonyl FSN = a nonionic, fluorinated surfactant (DuPont deNemours) .

Polymer bead reagent = Poly[styrene-co-m-&-p- (2- chloroethylsulfonyl- ethyl)styrene] .

Antibody-polymer bead reagent = Appropriate immunoreactive antibody covalently bound to the polymer bead reagent.

TX-100 = Triton X-100, an octylphenoxy polyethoxyethanol nonionic surfactant (Rohm & Haas) Hardener = Bis(vinylsulfonylmethyl) ether gelatin hardener.

A CRP element, as described above, was spottedwith 10 μL of sample containing 25.6 mg/L CRP and 3 nM monoclonal antibody-horseradish peroxidose label (MAb-HRP) at the center of the slide and incubated at 37°C for 5 minutes to allow the sandwich to form. CRP assay is a sandwich type enzyme immunoassay using a pair of antibodies. An anti-CRP monoclonal antibody (MAb) was labeled with horseradish peroxidase (HRP) . MAb-HRP was used as the signal antibody and a second goat anti-CRP antibody (PAb) covalently bound to the surface of poly(styrene-co-m-&- p-(2-chloroethylsulfonyl-methyl)styrene) polymer beads was used as the capture antibody. The slide was removed from the incubator, and 10 μL of wash solution (hydrogen peroxide lOmM, sodium phosphate, pH 6.8, lOiriM, '-hydroxyacetanilide 5mM, diethylenetriaminepentaacetic acid O.OliriM) was applied either at the center of the slide (at point of sample application) or at a point "off-center" in between the spot (position at which sample was applied) and the inner periphery of the area wetted by the sample, (approximately 2.8 mm away from the center which is 43% of the distance betweeen the center of sample spot and the inner peripery area) to remove unbound MAb-HRP conjugate from the observation area (center area of slide) . The slide was then placed back into the incubator at 37°C. The amount of formed PAb—CRP—MAb- HRP sandwich was determined by measuring the rate of the triarylimidazole leuco dye oxidation by hydrogen peroxide in the observation area in a 0 to 1 minute read time window. The results are shown below in Table I.

The imprecision was measured as a percent coefficient of variance, and the lower the variance the lower the imprecision. N equals the number of replications (the mean rate and precision are the means for this number of replications) . The results of this Example show that the lower mean rate for the 2.8 mm from Center wash of the present invention, is excluding the reading of contamination (unbound label) read in the Center wash assay. EXAMPLE II Diσoxin Assay

The following formulation is for a digoxin, element in accordance with the present invention:

For purposes of the above element, the following definitions apply.

HRP = orseradish peroxidase (label) .

MOPS =3-morpholinopropanesulfonic acid buffer.

TES = N- [Tris(hydroxymethyl)methyl]-2- aminoethane-sulfonic acid buffer.

Dimedone = 5,5-Dimethyl-l,3-cyclohexanedione.

Triarylimidazole leuco dye = 4,5-Bis(4-dimethylaminophenyl)-2- (4- hydroxy-3,5-dimethoxphenyl) imidazole blue forming leuco dye. Adhesive = Poly(methyl acrylate-co-sodium 2- acrylamido-2-methylpropanesulfonate- co-2-acetoacetoxy-ethyl methacrylate)

Polymer beads = Poly(m-&-p-vinyltoluene-co- methacrylicacid) .

Zonyl FSN = a nonionic, fluorinated surfactant

(DuPont de Nemours) .

Polymer bead reagent = Poly[styrene-co-m-&-p-(2-chloroethyl- sulfonylmethyl)styrene] . Antibody-polymer bead reagent = Appropriate immunoreactive antibody covalently bound to the polymer bead reagent. TX-100 = Triton X-100, an octylphenoxy polyethoxyethanol nonionic surfactant (Rohm & Haas)

Hardener = Bis(vinylsulfonylmethyl) ether gelatin hardener.

The digoxin element described above was spotted with 10 μl of sample containing digoxin and incubated at 37°C for 5 minutes to allow the competitive reaction to occur. The concentration of dioxin used in the "center vs. periphery" experiment was 1.6 ng/mL; and in the "center" vs. "2.8 mm from center" (present invention) experiment it was 3.1 ng/mL. Drug and drug-labeled HRP competed for binding sites on the immobilized antibody. The slide was removed from the incubator and washed with 10 μl of wash solution (hydrogen peroxide 0.03%, sodium phosphate, pH 6.8, lOmM, 4'-hydroxyacetanilide 5mM, diethylenetriaminepentaacetic acid lOiriM and hexadecylpyridinium chloride, 0.1%). The wash was applied off-center at a point, as in example 1, in between the spot (position at which sample was applied) and the inner periphery of the area wetted by the sample (2.8 mm from center which is 43% of the distance between the center and the periphery) , to the center of the spot or on the periphery of the area wetted by the spot, to remove unbound drug-HRP outward to the edges of the slide. The slide was placed back into the incubator at 37°C. The amount of bound digoxin-HRP was determined by measuring the rate of the triarylimidazole leuco dye oxidation by hydrogen peroxide in the observation area, which is at the center, in a 0 to 100 second time window.

The results are shown in Table II as follows:

TABLE II

Wash position Center Periphery Mean rate 0.0563 0.0630 Standard deviation 0.00164 0.000967

Wash position Center 2.8 mm from Center Mean rate 0.0659 0.0606 Standard deviation 0.00169 0.00114

The results of this Example show that the lower mean rate for the "2.8 mm from Center" wash of the present invention, is excluding the reading of contamination (unbound label) read in the center and periphery wash assays. The difference between the standard deviations (lower standard deviation desired) of the center and the wash technique of the invention (2.8 mm) are significant.

The invention has been described in detail with particular reference to preferred embodiments thereof, but it will be understood that variations and modifications can be effected within the spirit and scope of the invention.

Claims

WE CLAIM:

1. A method of conducting a solid phase immunoassay of a liquid sample containing an unknown level of analyte, comprising: applying a liquid sample to an analytical element of the type having at least one porous zone, said element containing an immobilized binding agent capable of reacting with said analyte, said sample forming a wetted area on said element, said wetted area being defined by a sample spot at the center thereof coinciding to the point of application of said sample and a ring at the periphery thereof; supplying a label to said element; applying 8 to 30 μL a stream of washing solution to the element at a position in between the center of said sample spot and up to 50% of the distance between the center of said spot and the inside periphery of said ring, said washing solution effectively displacing unbound label away from the center area of said spot; and measuring the extent to which bound label is present within said spot area.

2. The method of claim 1, wherein said immunoassay is a chemiluminescence immunoassay.

3. The method of claim 1, wherein said immunoassay is a colorimetric rate immunoassay.

4. The method of claim 1, wherein said liquid sample -i-»s human or animal biological fluid.

5. The method of claim 4, wherein said analyte is selected from the group consisting of digoxin, C-reactiveprotein, thyroxine, phenytoin, phenobarbital and carbamazepine.

6. The method of claim 1, wherein said wash solution is a buffer solution.

7. The method of claim 1, wherein said wash solution is applied at a position between 25 to 45% of the distance from said spot to the inner periphery of said ring.

8. The method of claim 1, wherein up to 10μL said wash solution is applied at a position 43% ofthe distance from said spot to the inner periphery of said ring.