WO2000005580A2

WO2000005580A2 - Method of determining the carry-over of an analyte

Info

Publication number: WO2000005580A2
Application number: PCT/EP1999/005332
Authority: WO
Inventors: Thomas Augustinus Maria Beumer; Wilhelmus Marinus Carpay
Original assignee: Akzo Nobel N.V.
Priority date: 1998-07-24
Filing date: 1999-07-19
Publication date: 2000-02-03
Also published as: WO2000005580A3; WO2000005580A8; AU5287699A

Abstract

The present invention relates to a method of determining the carry-over of an analyte in an assay comprising consecutive transfer of analyte-comprising samples by the same transferring means. According to the invention, a model analyte is used to determine a model carry-over, said model carry-over being representative for many of the parameters determining the carry-over of a particular assay. To determine the carry-over for that particular assay, the model carry-over is corrected with a correction factor C, provided by the manufacturer of the assay.

Description

Method of determining the carry-over of an analyte

Immuno assays like Enzyme Linked Sorbent Assays (ELISA) generally make use of a highly specific, immobilized capture molecules to concentrate the analyte(-s) of interest onto a solid phase. Due to the high affinity of the chemical compounds it is possible to detect, even quantify, very low concentrations of particular biomolecules in a complex environment like serum or plasma. If the compound of interest is present in the sample (this sample is to cause a so called positive test result), the concentration variation that naturally occurs is large. Typical examples are the pregancy hormone hCG that is known to occur in ranges from less than lU/1 up to more than 100,000U/1, and the Hepatitis-B antigen (HBsAg) that is found from less then 0.01 up to much more than 100,000U/ml.

As immuno assays are generally available as disposables for single use there is no need to clean a test device inbetween testing. Yet, large scale users like clinical labs and blood bank laboratories have introduced dedicated instruments to manipulate samples and reagents. Examples of such instruments are samplers that transfer the numerous sample aliquots from primary tubes in which samples are collected from the donor or patient into subsequent test devices and washers that remove liquid from test devices and change additional reagents. In such instruments subsequent samples make physical contact with part of the instrument. It is here that one risks the transfer of trace amounts of a highly reactive sample into a next-in-line non-reacive sample. This contamination is generally referred to as 'carry-over'. Given the extremely low detection levels of current assays, such a carry-over may cause a false positive test result. In current blood bank policy e.g. any suspectedly reactive sample causes a sequence of costly actions, varying from re-testing the non-reactive sample with even more refined and expensive test systems up to complete disposal of the precious donated material. This is all done to rigidly reduce any risk of transferring diseases like Hepatitis- B or HIV into people who need transfusion of blood components. Typical contact based carry-over fractions in sample distribution and 'washing' equipment range from 100 down to 0.1 ppm. With the above examples of occuring concentration ranges it is easy to see that samples of the highest reactivity undoubtedly cause a positive reaction in the next sample, even if that is known to be negative. This amount of transferred reactive material depends on the intrinsic design of the instrument (materials, geometry) and is to some extent subject to its use (between-run washing procedures).

With end-users being responsible for the correctness of the test result there is an increasing need for a tool to quantify the amount of carry-over that may of help (1) keeping track cq. quantification of the current quality status of the instrument with respect to contamination, (2) in optimizing cleaning procedures aiming at reduction of carry-over and (2) to help and estimate the operational risks of current procedures.

The present invention relates to a method of determining the carry-over of an analyte in an assay comprising consecutive transfer of analyte-comprising samples by the same transferring means.

Such a method is generally known in the art and comprises the steps of

1 ) transferring a predetermined volume of a first liquid solution comprising an analyte at a first known concentration into a first container by means of a transferring means;

2) optionally rinsing said transferring means with a rinse solution;

3) transferring a predetermined volume of a second liquid solution, optionally comprising the analyte at a second known concentration, by means of said transferring means into a second container, said second concentration differing from the first concentration; 4) determining the concentration of the analyte in the first and second container;

5) determining the carry-over using the formula:

I [A]_2m - [A]_2t I carry-over =

[A]_lm where

[A]ι_m = the concentration of analyte measured in the first container;

[ A]_2m ⁼ the concentration of analyte measured in the second container; and

[A]_2l = the theoretical concentration of analyte in the second container, i.e. the concentration of analyte expected without carry-over. As in practice, for optimum determination of the carry-over, the second liquid solution usually does not contain any analyte, [A]_2t, is usually zero. The determination of the concentration analyte ([A]) may be direct or indirect. In the latter case a relative concentration is determined, as will usually be the case when the analyte is an enzyme, and the amount of substrate converted (or product formed) is a measure of the enzyme concentration. Absolute concentrations may be determined by preparing a dilution series of the analyte at a known concentration, which dilution series serves as a calibration curve for the determination of the absolute concentration.

The carry-over depends on the particular assay performed, including the assay protocol and the characteristics of the type of apparatus used for transferring samples. Therefore the carry-over is determined, in practice, by using a strongly positive sample tested earlier, which is rather undefined. If this strongly positive sample does not comprise analyte in the maximum concentration that a sample may comprise, the carryover may erroneously be assumed to be satisfactorily low. Thus, accurate determination of the carry-over is impeded, not in the least because assays are often optimized for maximum sensitivity and, as an adverse consequence, show a poor linearity between a signal generated and the analyte concentration. In addition, adsorption of analyte may occur when ill-defined samples are used, resulting in underestimation of the carry-over. A further disadvantage is that the determination of the carry-over is costly due to the use of expensive assay reagents and rather time consuming, taking an hour or more, for each different assay or assay protocol, that is. The object of the present invention is to overcome the above disadvantages.

To this end, the method according to the present invention is characterized in that for a model analyte a model carry-over is established, which model carry-over is used to determine the carry-over for an assay for establishing the concentration of an analyte, said carry-over being determined by correcting the model carry-over with a correction factor C using the formula carry-over = C * model carry-over wherein

C is a member of the group consisting of Cj., C₂, C₃, C₄ and C₅ or a product of at least two members from this group; in which Ci = the ratio between the viscosity of a first rinse solution used in the assay and the viscosity of a second rinse solution used between transfers of the model analyte; C₂ = either the ratio between the radius of the analyte and the radius of the model analyte or, as an approximation, the square root of the ratio between the molecular weight of the analyte and the molecular weight of the model analyte;

C₃ = the ratio between the temperature of the second rinse solution used between transfers of the model analyte and the temperature of the first rinse solution used in the assay;

C₄ = the ratio between the viscosity of a sample comprising the analyte assayed in the assay and the viscosity of a liquid solution comprising the model analyte used for determining the model carry-over; and C₅ = a further assay protocol-specific value; said carry-over being used to determine for which of the samples assayed the analyte concentration is not reliably established due to entrainment of analyte from a preceding sample.

Using a model analyte, which may be very cheap, it is possible to establish a model carry-over which is characteristic for many of the parameters determining the carry-over for a particular assay. This model carry-over is corrected for the particular assay to be performed, most conveniently using data, that is correction factor C or at least one member thereof, supplied by the manufacturer of the assay. Thus, it may suffice to perform only one experiment to determine the model carry-over of the model analyte, and calculate the carry-over for each of one or more different assays, which includes different assay protocols for one particular analyte. The model carry-over can be determined very quickly, typically within 5 to 30 minutes.

Each of the members determining the correction factor C may be determined empirically or theoretically, and separately or as a combination of one or more other factors. For example, if the rinsing duration between different samples in the assay differs from that when establishing the model carry-over, the member C5 may account for this difference, but the consequence of this different duration may be established by a manufacturer of an assay for a particular analyte together with other differences such as those determined by one or more of the members Ci to C₄, without any need to deter- mine each member separately. C₅ may be used to correct for, for example, a different inner diameter of a needle of an apparatus used to transfer solutions, time during which is rinsed and any other factor affecting the carry-over.

According to a first embodiment of the method according to the present inven- tion, an enzyme-comprising analyte is used as the model analyte.

Though the model carry-over may be determined using other methods, such as those employing a radioactive, fluorescent or chemiluminescent analyte, for example fluorescein-labelled bovine serum albumin, an enzyme-comprising analyte offers the advantageous possibility of enhanced sensitivity due to the enzymatic activity of the label If hazardous labels are used, such as radioactive labels, determination of the carry-over increases the amount of radioactive waste to be disposed of

According to a preferred embodiment, the method according to the invention is characterized in that an adsorption preventing blocking agent is present at a concentration in excess of the model analyte. Thus adsorption of model analyte is virtually completely eliminated Moreover, the blocking agent, such as bovine serum albumin, casein or polyethylene glycol (PEG), may also give the model analyte containing solution a viscosity similar to that of a sample to be tested in the assay, which may eliminate a variable to be reckoned with by the manufacturer of an assay kit Because of its ubiquity, the preferred assay for which the carry-over is established is an immunoassay

If the assay is an immunological assay using an enzyme label, the enzyme used for determining the carry-over is preferably the same as in the assay This eliminates the need to use different reagents for detecting the enzyme when determining the model carry- over Thus, advantageously horseradish peroxidase is used as the enzyme label The enzyme label may be a carrier-bound enzyme label, the carrier for example being bovine serum albumin

When the enzyme label and the enzyme used for determining the carry-over are the same, the number of reagent solutions necessary for establishing the (relative) enzyme concentration, such as substrate solution, co-factor comprising solution etc is reduced, saving time

Preferably the model analyte is used at a concentration of at least e _t, where nest ^— l*m Alo / r_ac0 wherein

F_m = a multiplication factor of at least 10;

Aiow = lowest concentration of model analyte detectable; and Faco = acceptable carry-over factor.

The acceptable carry-over factor F_ac0 depends on the particular assay, such as an immunoassay, to be performed. The F_aco can easily be established by calculating the ratio between a) the threshold concentration, i.e. - in case the analyte, for example a virus such as HIV, should not be present, the concentration of analyte at which a sample is considered positive, and

- in case the analyte, such as a insulin, is present in samples of healthy subjects, the absolute difference in concentration between the concentration of analyte at which a sample is considered positive and the mean concentration of analyte present in healthy subjects, and b) the highest concentration of analyte that can be encountered in a sample.

Thus a very reliable determination of the carry-over to at least one subsequent container is possible.

Currently, if a carry-over is determined with an analyte-comprising standard solution provided with an assay kit is used, the standard solution with the highest concentration of analyte is used. However, as the assay usually is aimed at determining the presence of an analyte at a concentration just above the detection limit, the highest concentration in the standard solution provided with the assay kit is often considerably lower than the highest concentration of analyte that may be present in a sample. For example, a typical commercially available assay kit for hepatitis B comprises a standard solution of about 5 units antigen per ml, whereas a sample containing a high concentration of antigen may comprise 100,000 to 1 ,000,000 units per ml. This may lead to an underestimation of the carry-over. Due to the linearity of the method according to the present invention, a high concentration of model analyte can be used, cost of the analyte being insignificant with respect to a typical standard solution.

According to a favourable embodiment, the molecular weight of the model- analyte is generally approximately equal to and or smaller than, for example, 0,7 times the molecular weight of the analyte, preferably 0,5 times or less and more preferably 0,3 times or less.

Such lower molecular weights of the model analyte allow the determination of the carry-over to be determined quicker, and/or more precisely. The invention also relates to a solution suitable for determination of a model carry-over, said solution comprising a model analyte chosen from the group of enzymes, fluorescent molecules, molecules comprising a fluorescent, radio-active, enzyme or chemiluminescent label, together with an effective concentration of an adsorption- preventing agent. Such a solution improves quality assurance procedures, allowing in addition people from different laboratories to more easily compare their results. A person skilled in the art can easily determine the effective concentration for a particular adsorption- preventing agent, such as polyethylene glycol or a protein such as bovine serum albumin. As long as a higher concentration of the adsorption-preventing agent results in a larger signal, the effective concentration is not reached. Examples of large fluorescent molecules can easily be found in the literature, example are phycobiliproteins such as allophycocyanin.

Finally, the invention relates to the use of a correction value specific for a particular assay for correcting a model carry-over value determined using a standardized method to obtain a carry-over value for the particular assay.

The particular assay is preferably an immunoassay. Using a model analyte and a standardized method for determining the carry-over with that particular model analyte, results in a model carry-over, which comprises particulars such as characteristics of an apparatus for dispensing sample and reagent solutions, as well as those regarding the model analyte. A manufacturer of a particular assay can determine a correction factor, which can be used to correct the model carry-over resulting in the carry-over for that particular assay.

EXAMPLES

1. In an immunoassay the residual concentration after washing is determined of a fraction using horseradish peroxidase (HRP) as an enzyme label (model analyte according to the present invention) and a fraction using normal human serum (NHS) (of which human serum albumin (HSA) is an important component and frequently used in common enzyme linked immunosorbent assays (ELISA)).

Both molecules have a molecular weight of approximately 40 kD (kiloDalton). In figure 1 the results are shown from which is clear that after washing with larger volumes, the residual concentration of both HRP and NHS decreased in a similar fashion. In this experiment a freshly made solution of HRP was made which gave excellent reproducible detection levels and quantitation of carry-over. However, an excellent alternative to this fresh solution is to lyophilize ('freeze dry') the HRP (or Bovine Serum Albumine (BSA)-HRP mixture) in an appropriate sugar matrix. Both the stock solutions and the pre-diluted reference concentrations were shown to have a shelve life well over 6 months and reproducibility equivalent to fresh solutions.

2. The results from example 1 are used to compare and determine the correction-factor for the assay. As presented in figure 2, the data from the ELISA method (X-axes) are compared with the data from the model analyte according to the present invention (HRP- molecule)(Y-axes). From figure 2, it is clear that the method is comparable for a factor less than 2. The accuracy is also within this range which means that they do not differ significantly. Therefore, this method is perfectly fit for determining the carry-over of an analyte in an assay.

Claims

1. Method of determining the carry-over of an analyte in an assay comprising consecutive transfer of analyte-comprising samples by the same transferring means, characterized in that for a model analyte a model carry-over is established, which model carry-over is used to determine the carry-over for an assay for establishing the concentration of an analyte, said carry-over being determined by correcting the model carry-over with a correction factor C using the formula carry-over = C * model carry-over wherein

C is a member of the group consisting of Ci, C₂, C₃, C₄ and C₅ or a product of at least two members from this group; in which Ci = the ratio between the viscosity of a first rinse solution used in the assay and the viscosity of a second rinse solution used between transfers of the model analyte;

C₂ = either the ratio between the radius of the analyte and the radius of the model analyte or, as an approximation, the square root of the ratio between the molecular weight of the analyte and the molecular weight of the model analyte; C₃ = the ratio between the temperature of the second rinse solution used between transfers of the model analyte and the temperature of the first rinse solution used in the assay;

C₄ = the ratio between the viscosity of a sample comprising the analyte assayed in the assay and the viscosity of a liquid solution comprising the model analyte used for determining the model carry-over; and C₅ = a further assay protocol-specific value; said carry-over being used to determine of which samples assayed the analyte concentration is not reliably established due to entrainment of analyte from a preceding sample.

2. Method according to claim 1, characterized in that as the model analyte an enzyme-comprising analyte is used.

3. Method according to claim 1 or 2, characterized in that the model analyte is used at a concentration of at least where test 1 m -^Mow l aco wherein F_m = a multiplication factor of at least 10;

A_)ow = lowest concentration of model analyte detectable; and Faco ⁼ acceptable carry-over factor.

4. Method according to any of the preceding claims, characterized in that an adsorption preventing blocking agent is present at a concentration in excess of the model analyte.

5. Method according to any of the preceding claims, characterized in that an immunoassay is used to determine the analyte.

6. Method according to claim 5, characterized in that in the immunoassay an enzyme is used as a label and the enzyme used for determining the model carry-over is the same as the enzyme used as a label.

7. Solution suitable for determination of a model carry-over, said solution comprising a model analyte chosen from the group of enzymes, fluorescent molecules, molecules comprising a fluorescent, radio-active, enzyme or chemiluminescent label, together with an effective concentration of an adsorption-preventing agent.

8. Method according to any of the preceding claims, characterized, in that the molecular weight of the model-analyte is 0,5 times or less the molecular weight of the analyte and more preferably 0,3 times or less.

9. Use of a correction value specific for a particular assay for correcting a model carry-over value determined using a standardized method to obtain a carry-over value for the particular assay.