WO2008133744A2

WO2008133744A2 - Sample preparation for antigen detection testing

Info

Publication number: WO2008133744A2
Application number: PCT/US2007/086894
Authority: WO
Inventors: William E. Hessman
Original assignee: Central States Testing LLC
Current assignee: Central States Testing LLC
Priority date: 2006-12-08
Filing date: 2007-12-08
Publication date: 2008-11-06
Anticipated expiration: 2009-06-08
Also published as: WO2008133744A3

Abstract

The present invention provides methods for reducing false negative errors in antigen capture tests of pooled and single samples. The methods are particularly useful for detecting antigen-containing samples when pooled with other samples that contain an inhibiting agent.

Description

l :

SAMPLE PREPARATION FOR ANTIGEN DETECTION TESTING

FIELD OF THE INVENTION

The present invention relates to tissue sample preparation methods for improving the ability to detect an analyte in a tissue using a test based on antigen capture technology.

BACKGROUND OF THE INVENTION

A challenging problem is reliably detecting an analyte that is present, if at all, at a low concentration in a very small fraction, e.g., less than one percent, of a population of samples to be tested. The various technologies that are available to detect an analyte present at a low concentration in a sample are generally based on specific high-affinity binding of a binding partner to the analyte. Examples include the specific binding of an antibody to an antigen or the binding of a probe, such as an oligonucleotide, to a nucleic acid molecule. Much effort has been devoted to improving detection of an analyte by improving the specificity and/or affinity of the interaction between the analyte and the binding partner. Often one or more amplification processes, such as enzymatic production of the detected molecule or specific, template-dependent synthesis of a nucleic acid analyte using the polymerase chain reactive (PCR), are used to enhance the sensitivity of the test. Several technologies are available to detect the presence of an analyte-binding partner complex using direct or indirect labels that are detectable as changes in optical, radiological, chemical, magnetic or electrical signals. Often improvements in specificity involve a loss in sensitivity and an increase in false negative errors.

The known approaches to solving the problem of detecting an analyte present only in a very small fraction of the population of samples to be tested can exacerbate the problem of detecting an analyte that if present, is present at a concentration near the threshold of detection. A known solution is to test first for the presence of the analyte in pooled aliquots of samples, then test separately the samples contributing to a pool that tests positive for the analyte. The suitable pool size generally varies by analyte and detection technology, and would likely differ for a PCR-based test and an antigen capture ELISA (enzyme-linked immunosorbent assay) test. Simple dilution of a single positive sample with several negative samples can reduce the concentration of the analyte below the detection threshold for the analyte and cause false negative errors. In addition, a negative sample may also contain one or more factors that can suppress or inhibit the interactions required to detect the analyte, also causing false negative errors.

Certain sample handling procedures have been found that reduce false negative errors and thereby improve the ability to detect an analyte that is present, if at all, at a low concentration in a only very small fraction, e.g., less than ten percent, and in preferred embodiments, less than one percent, of the population of samples to be tested.

SUMMARY OF THE INVENTION

The present invention provides methods for reducing false negative errors in antigen capture tests of pooled and single samples. The methods are particularly useful for detecting BVDV (bovine viral diarrhea virus) antigen-positive samples when pooled with other samples that contain an inhibiting agent, such as an antibody to a BVDV antigen or other interfering substance.

hi certain preferred embodiments, the invention provides a method of processing a tissue sample to provide a liquid sample to be tested for the presence of an analyte, comprising the steps of immersing a tissue sample in a buffered physiological saline solution; incubating the tissue sample in the buffered physiological saline solution at about 32-41 degrees Celsius for at least 1 hour; reducing the temperature of the tissue sample in buffered physiological saline solution by at least about fifteen degrees Celsius for at least fifteen minutes; and removing an aliquot of the buffered physiological saline solution thereby providing a liquid sample to be tested for the presence of an analyte.

Suitable buffered physiological saline solutions have a buffer, pH, inorganic components and organic components (if present) that are known in the art and are compatible with the tissue sample, the formation of the analyte/binding partner complex and the detection means. Where the test is an antigen capture ELISA test, a preferred buffered physiological saline solution is phosphate buffered saline solution.

In preferred embodiments, the temperature of the tissue sample in buffered physiological saline solution is reduced to room temperature after incubation at about 32-41 degrees Celsius. Alternatively, the temperature of the tissue sample in buffered physiological saline solution is reduced to about 2-7 degrees Celsius after being incubated at about 32-41 degrees Celsius. In further embodiments, the temperature of the tissue sample in buffered physiological saline solution is reduced to about —20 degrees Celsius after being incubated at about 32-41 degrees Celsius.

In other embodiments, the method further includes the steps of performing the test for the analyte on the liquid sample, and evaluating the test measurements using a standard that has been empirically determined using known positive and negative samples to minimize false negative errors, hi preferred embodiments, the liquid sample is a pooled sample that includes aliquots of 2 to 20 samples.

In other aspects, the present invention provides a method of increasing the sensitivity of a test that measures the presence or amount of an analyte located in a tissue based on the specific high-affinity binding of a binding partner to the analyte, comprising the steps of immersing a tissue sample in a buffered physiological saline solution; incubating the tissue sample in the buffered physiological saline solution at about 32-41 degrees Celsius for at least 1 hour; reducing the temperature of the tissue sample in buffered physiological saline solution by at least about fifteen degrees Celsius for at least fifteen minutes; removing an aliquot of the buffered physiological saline solution thereby providing a liquid sample to be tested for the presence of an analyte; testing the liquid sample using the test based on the specific high-affinity binding of a binding partner to the analyte; detecting the presence or amount of the analyte by measuring the formation of an analyte/binding partner complex; and evaluating the test measurements using a standard that has been determined empirically using known positive and negative samples to minimize false negative errors, thereby increasing the sensitivity of the test.

Suitable buffered physiological saline solutions have a buffer, pH, inorganic components and organic components (if present) that are known in the art and that are compatible with the tissue sample, the formation of the analyte/binding partner complex and the detection means. Where the test is an antigen capture ELISA test, a preferred buffered physiological saline solution is phosphate buffered saline solution.

In preferred embodiments, the temperature of the tissue sample in buffered physiological saline solution is reduced to room temperature after incubation at about 32-41 degrees Celsius. Alternatively, the temperature of the tissue sample in buffered physiological saline solution is reduced to about 2-7 degrees Celsius after being incubated at about 32-41 degrees Celsius. In further embodiments, the temperature of the tissue sample in buffered physiological saline solution is reduced to about -20 degrees Celsius after being incubated at about 32-41 degrees Celsius.

In yet other aspects, the present invention provides a method of increasing the sensitivity of a test that measures the presence or amount of an antigen located in a tissue, where the antigen is likely to be detected in less that 10% of the samples to be tested, comprising the steps of immersing a tissue sample in a buffered physiological saline solution; incubating the tissue sample in the buffered physiological saline solution at about 32-41 degrees Celsius for at least 1 hour; reducing the temperature of the tissue sample in buffered physiological saline solution by at least about fifteen degrees Celsius for at least fifteen minutes; removing an aliquot of the buffered physiological saline solution thereby providing a liquid sample to be tested for the presence of an antigen; pooling an aliquot of b liquid samples to form a pooled sample, where b is an integer from 2 to 20 inclusive; testing the pooled sample for the presence and amount of the antigen; evaluating the test measurements for each pooled sample using a standard that has been determined empirically using known positive and negative samples to minimize false negative errors; retesting aliquots from individual liquid samples that contributed to pooled that were evaluated as other than negative for the presence of the antigen; and evaluating the test measurements for each individual liquid sample using a standard that has been determined empirically using known positive and negative samples to minimize false negative errors, thereby increasing the sensitivity of the test and reducing the total number of tests required to measure the presence or amount of an antigen located in a tissue, where the antigen is likely to be detected in less that 10% of the samples to be tested. In certain embodiments, the antigen is likely to be detected in less that 1% of the samples to be tested.

Where the test is a BVDV antigen capture ELISA test, a preferred standard for classifying the S/P (ratio of the sample absorbance to the positive control absorbance) result is:

S/P > 0.39 = antigen positive; S/P = 0.20-0.39 = suspect; S/P = 0.07-0.19 = requiring re-check; and

S/P < 0.07 = antigen negative.

In certain embodiments, the buffered physiological saline solution contains a non-interfering colorant. When the buffered physiological saline solution is a phosphate buffered saline solution, a preferred colorant is a mixture of FD&C Yellow 5 and FD&C Blue.

DETAILED DESCRIPTION OF THE INVENTION

In order to focus on the efficacy of changes in sample preparation procedures, known positive and negative samples were tested using a commercially available antigen capture ELISA test (ACE) (Herdchek™, IDEXX, Westbrook, Maine) for a low prevalence analyte, immunoreactive components of the bovine viral diarrhea virus.

Bovine viral diarrhea virus (BVDV) is a positive stranded RNA virus of the Pestivirus genus of the Flaviviridae family. This virus is found worldwide in ruminant animals both domestic and wild, including sheep, goats and deer as well as cattle. In addition to having several genetic subtypes (BVDV Ia, BVDV Ib and BVDV 2), the virus can exist in cytopathic and noncytopathic forms, producing a wide range of clinical disease. The interactions of this complex pathogen with the immune system at different stages of life are responsible for production of animals persistently infected (PI) with BVDV as well as a wide range of clinical diseases ranging from subclinical or clinically inapparent infections to a highly fatal disease known as mucosal disease (Baker, J. C, The clinical manifestations of bovine viral diarrhea infection, Vet Clin North Am Food Anim Pract, 11(3): 425-445, 1995). The initial description of the disease and pathogen were based on an enteric infection on a dairy farm, resulting in the names bovine viral diarrhea (BVD) and bovine viral diarrhea virus. It is now recognized that BVDV is responsible for a number of disease syndromes in cattle affecting the reproductive, respiratory tract, gastrointestinal, circulatory, immunologic, lymphatic, musculoskeletal, integumentary and the central nervous systems.

Persistently infected animals are persistently viremic animals. Cattle persistently infected with BVDV act as the major reservoir of BVDV and are the primary source of infection, with transiently infected cattle considered a less important source. See generally, Larson, R.L., et al., Bovine Viral Diarrhea (BVD): Review for Beef Cattle Veterinarians, Bov Pract 38:93-102, 2004. Persistently infected animals are much more efficient transmitters of BVDV than transiently infected animals because they secrete much higher levels of virus for a much longer period of time. BVDV is shed by a PI animal throughout life from virtually all secretions and excretions including nasal discharge, saliva, semen, urine, tears, milk, and to a lesser extent, feces. Traven, M., et al., Primary bovine viral diarrhoea virus infection in calves following direct contact with a persistently viraemic calf. J Vet Med B 38:453-462, 1991. Fetuses, placentae and fetal fluids, from BVDV-induced abortions can also contain BVDV. Horizontal transmission of BVDV to seronegative cattle has been shown to occur after only one hour of direct contact with a single PI animal. Over-the-fence contact with a PI animal from a neighboring herd can also introduce BVDV into a susceptible herd (Larson et al., 2004). Further discussion of the effects of PI animals and methods of livestock management can be found in patent applications U.S. 11/340,958 and PCT/US06/18945, the contents of which are incorporated herein by reference. Laboratory tests that are known in the art for the detection of the BVD virus include virus isolation with a microplate immunoperoxidase detection system, antigen capture ELISA (ACE), immunohistochemistry (IHC) and nuclear acid detection using polymerase chain reaction (PCR). See Drew, T., "Bovine Viral Diarrhoea," Chapter 2.10.6 in Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, 5^th Ed., 2004, World Organisation For Animal Health (Office International des Epizooties, OIE), http://www.oie.int/eng/ normes/mmanual/ A 00132.htm. Pooling of samples and retesting the individual samples contributing to a pooled sample testing positive has been used as a procedure to reduce the number of tests, and thus the time, needed to screen numerous samples for a low prevalence analyte. However, mixing samples can cause an increased number of false negative errors due to the effects of dilution factors as well as other factors. The present invention provides sample handling procedures and modified test procedures that minimize false negative errors.

As used herein, room temperature is a temperature of 18-25 degrees Celsius. The temperature of a refrigerator, when not stated explicitly, is 2-7 degrees Celsius. In ALL sample handling procedures, whether involving intermediate steps of freezing or heating, the samples were allowed to come to room temperature and mixed gently before a liquid sample aliquot was removed for testing.

Example 1

Modifications of the Manufacturer's Testing Protocol to Reduce False Negative Errors

The manufacturer's product insert describes the test kit (Herdchek™, EDEXX, Westbrook, Maine) as a simple, fast and inexpensive means of detecting BVDV antigen in either serum samples or ear notch biopsy samples that is based on ELISA methodology, using a BVDV antigen-specific monoclonal antibody as the capture antibody, a goat polyclonal anti-BVDV antibody as the detector, and a horseradish peroxidase-anti-goat conjugate. The ELlSA procedure is carried out at room temperature (18-25 degrees Celsius). The test kit insert instructions were followed with the following modifications to the sample preparation instructions and the guidelines for interpretation of results.

According to the instructions for the commercial test kit, an ear-notch biopsy sample approximately 1 cm x 1 cm should be suspended in phosphate buffered saline (PBS, 0.1 M phosphate buffer (pH 7.4) with 0.9% NaCl) to prepare a liquid sample extract for use in the assay.

1. Add 2 mL of PBS to the ear notch biopsy sample in a glass or plastic tube; mix it by gentle vortexing or inversion. The ear notch biopsy sample must not be treated with a fixative (e.g., formalin).

2. Allow the ear notch biopsy sample to soak in PBS for a minimum of 10 minutes at room temperature.

3. Mix the tissue suspension again immediately prior to running the ELISA. The liquid sample extract is used in the ELlSA in place of serum samples.

4. The processed ear-notch sample can be either refrigerated at 2°-7°C for short-term storage (1-2 days) or frozen (-20⁰C or colder) for long-term storage.

The ELISA procedure yields a colored reaction product in the presence of BVDV antigen, the test result for each sample is calculated as the sample to positive control (S/P) ratio, which is calculated as (sample A₄₅₀ - negative control mean A₄₅o)/(positive control mean A₄₅₀ - negative control mean A₄₅₀).

The manufacturer's product insert does not mention sample pooling. Pooled sample testing regimes have been described before. Some regimes have been based on complex statistical models, providing protocols having multiple steps of pooling, dividing and re-pooling samples and provided inconsistent teachings as to preferred number of samples to pool. Munoz-Zanzi, CA. , et al., Pooled-sample testing as a herd- screening tool for detection of bovine viral diarrhea virus persistently infected cattle, J Vet Diagn Invest. 12(3): 195-203, 2000, disclosed that Monte Carlo simulations showed that the protocol associated with the least cost per cow involved an initial testing of pools followed by re-pooling and testing of positive pools, with intermediate splitting and testing of positive pools before testing aliquots of the individual samples that contributed to a positive pool. In other optimized pooled sample test strategies, positive pools are retested before splitting (Kennedy, N.L., Hierarchical screening with retesting in a low prevalence population, Sankhyά: The Indian Journal of Statistics, 66: 779-790, 2004). Larson et al. have proposed a two-stage test strategy of PCR testing of pooled blood samples followed by immunohistochemical staining of skin biopsy specimens from animals that had contributed to a pooled blood sample that had tested positive by PCR test (Larson, R.L., et al., Economic costs associated with two testing strategies for screening feeder calves for persistent infection with bovine viral diarrhea virus, J Am Vet Med Assoc 226:249-254, 2005).

In one preferred testing strategy of the present invention, a first stage of testing pooled samples is followed by testing of individual samples that contributed to a positive pool. In the first stage aliquots of several samples are tested as a pooled sample. The pooled sample approach also has the advantage of increasing the speed at which samples can be tested, since at a PI prevalence rate of 0.3 - 0.4 percent or less, more than 99% of the samples are expected to be BVDV negative. In the second stage, aliquots of the samples contributing to the few positive pooled samples can be retested separately to identify the individual positive samples.

The basic principles can be shown by a simple algebraic discussion. The maximum number of tests y that would be required to screen x samples is given by Equation 1, y = Zi + Z2 (Equation 1) where zi, the maximum number of tests of pooled samples, is the integer that is greater or equal to x/b, where x is the number of samples and b is the number of aliquots from individual samples that are combined to make a pooled sample, and Z₂, the maximum number of tests of individual samples, is the integer that is greater or equal to abx, where a is the estimated prevalence of positive animals in the screened population, x is the number of samples and b is the number of aliquots from individual samples that are combined to make a pooled sample. With an estimated prevalence of 0.4%, four samples per thousand would be expected to be BVD positive, with as many as four positive pooled samples that would require retesting, although as few as one pooled sample might test positive and require retesting.

Preferably, aliquots of 100/6 μl are taken from each sample vial, where b is the number of samples to be tested in each pooled sample. Low values of b provide minimal cost and time advantage, but high values of b may reduce the sensitivity of the test due to increased dilution of antigen in the pooled samples. In preferred embodiments using the BVDV ACE test, b is suitably in the range of 2 to about 20. In certain preferred embodiments, b is 6-12.

Example 2

Reduction Of Apparent Antigen Levels In Known Positive Samples By Inhibiting Agent In Negative Samples

It has been found that pooling aliquots of known positive samples with aliquots of particular known negative sample caused a suppression of the apparent measurable viral antigen in positive samples below what would be expected based on a simple dilution effect. While not being held to a specific theory, it is believed that samples from a subset of animals that test negative using the BVDV antigen capture ELISA test contain one or more agents that interfere with the interaction between the BVDV viral antigen and the ELISA test antibody, producing false negative errors in pooled samples, but not in individual samples. The degree of interference has also been found to depend on the time elapsed between tissue sampling and testing. More interference was found with fresher samples. As used herein, a "fresh sample" is a sample collected from the animal less than three hours previously.

The first adjustment to the standard kit protocol was the introduction of an additional category for pooled samples. As provided in the manufacturer's instructions, the presence or absence of BVDV antigen is determined by the sample to positive control (S/P) ratio for each sample, which is calculated as S/P = (sample A₄₅₀ - negative control mean A₄5₀ )/(positive control mean A_45O - negative control mean A₄₅₀). The tables below provide the S/P ratio.

The standard IDEXX kit protocol for single samples provides the following categories for S/P values:

Positive S/P > 0.39

Suspect S/P = 0.20 - 0.39

Negative S/P < 0.2

The manufacturer's instructions provide that samples with S/P values falling within the suspect range should be assayed again, using a different "modified" working detector reagent containing no detector antibody. The suspect sample and the kit controls should be included in the same run using the standard working reagents. The S/P ratio for that sample is then calculated using a different formula, S/P = (A₄₅₀ sample (with detector antibody) - A₄₅₀ sample (without detector antibody))/(ρositive control mean A₄₅₀ - negative control mean A₄₅0), but compared again to the same breakpoint of S/P = 0.2.

These breakpoints defined in the manufacturer's instructions are referred to herein as "IDEXX" breakpoints. For example, in a recently reported study, detection of BVDV antigen in skin (ear notch) specimens was performed by use of a commercially available kit following the manufacturer's recommendations. Fulton, R.W., et al., Evaluation of diagnostic tests used for detection of bovine viral diarrhea virus and prevalence of subtypes Ia, Ib, and 2a in persistently infected cattle entering a feedlot, J. Amer. Vet. Med. Assoc, February 15, 2006, 228(4): 578-584. For samples that were retested, S/P values < 0.20 were considered negative and those with values > 0.20 were considered positive. For quality control for each run to be acceptable, the raw OD (optical density) values for negative and positive controls in the kit must have been < 0.5 and > 0.8, respectively.

As a result of further studies on known positive samples in pooled samples described below, the adjusted breakpoints and categories that have been developed for S/P values rounded to two decimal places are:

Positive S/P > 0.39

Suspect S/P = 0.20 - 0 .39

Re-check" S/P = 0.07-0. 19

Negative S/P < 0.07

These adjusted category boundaries, or "breakpoints," that provide an additional category are referred to herein as "CST" breakpoints. If the result from a pooled sample fell in either the "Positive," "Suspect," or "Re-check" categories, the samples that contributed to that pooled sample were re-tested individually.

Routine pooling of up to ten non-fresh samples including a known positive sample resulted in S/P values that while reduced, still fell in the "positive" or "suspect" ranges. Initial studies using more than seventy-five known positive samples in pools of various sizes showed that false negative errors can occur in tests of pools of eighteen and more samples. Similarly, studies of sample pools of < 100 samples reported false negative in a few samples that were negative by antigen capture ELISA for both pooled and individual samples but positive by PCR tests. See Kennedy, J. A., Diagnostic efficacy of a reverse transcriptase-polymerase chain reaction assay to screen cattle for persistent bovine viral diarrhea virus infection, J. Amer. Vet. Med. Assoc, November 1 , 2006, 229(9): 1472-1474.

An additional consequence of sample pooling was discovered while testing a series of samples from a group of cattle in a feedlot that were primarily (>80%) from the same herd. In some cases, a pool of a known positive sample (with a positive single sample result as high as S/P = 1.125) and six negative samples from particular animals would produce a pooled sample result as low as 0.03 - 0.09, substantially below the kit manufacturer's negative cutoff value of S/P = 0.2. In such a case, two of the six negative samples were found to interfere with the measurement of the antigen level in the known positive samples. In general, this effect was more prominent in fresh (<3 hours old) samples.

Table 1, below summarizes a study of this effect using nineteen known positive samples pooled with three different groups of samples from BVDV antigen negative animals (Group I, animals 1641, 1642, 1643, 1644, 1645, 1646; Group II, animals 4711, 4712, 4713, 4714, 4715, 4716; Group m, animals 4676, 4677, 4678, 4679, 4680, 4681) showing the effect of pooling a known positive sample with samples containing an interfering agent from negative animals. In all cases, the fresh samples were incubated in PBS for less than an hour before aliquots of the PBS were taken for testing.

In each of the three replications with different pools of negative samples, the majority of the 19 known positive samples would have been scored as false negative errors using the IDEXX breakpoints for single samples, while the adjusted breakpoints eliminated 65% to 100% of the false negative errors. Although the adjusted breakpoints improved the sensitivity dramatically, several false negative errors would still have been obtained.

Table 1

Group Group Il Group i

AVG S/P 0.19 0.10 0.18

False negative errors 14 17 13 Example 3

The Effect of Sample Handling On The Reduction of Apparent Antigen Levels In Known Positive Samples By Interfering Agent In Negative Samples

The effects of sample handling procedures were tested using the nineteen known positive samples in the same pools as described in Example 3. However, in this series of measurements, the samples were incubated in PBS for 4-8 hours, frozen overnight at - 20° Celsius (-5° Fahrenheit), thawed to room temperature, aliquoted & pooled and tested. The results are shown in Table 2, below. This sample handling procedure resulted in false negative errors in each pooled sample using the manufacturer's breakpoints, but not with the adjusted breakpoints.

Example 4

The Effect of Particular Negative Sample On The Reduction of Apparent Antigen Levels In Known Positive Samples By Interfering Agent

The Table 3 summarizes the results of a study in which two known BVDV positive samples (A78, single sample test result S/P = 0.99, and 4718, single sample test result S/P = 1.13) were pooled with twelve different groups of six negative samples. The groups of six negative samples are denoted by the number of the first of the six samples; Groups I, II and III of Examples 2 and 3 correspond to 1641, 471 1 and 4676. Group I had two animals with a moderate inhibitory effect, Groups II and III each had one animal with a substantial inhibitory effect, 4715 and 4676, respectively. In all cases, the fresh samples were incubated in PBS for less than an hour before aliquots of the PBS were taken for testing.

Table 3

A78 as the positive sample 4718 as the positive sample

Neg group S/P IDEXX CST S/P IDEXX CST

1641

4690 4697 4704 4711 4699 4725 4732 4739 4662 4676 4689

AVG S/P 0.16 0.30

False negative errors

Table 4, below, summarizes the results of a study using the same samples, but using a different sample handling procedure. In this series of measurements, the samples were incubated in PBS for 2 days, frozen overnight at -20° Celsius (-5° Fahrenheit), thawed to room temperature, aliquoted & pooled and tested.

Table 4

A78 as the positive sample 4718 as the positive sample

Ne rou S/P IDEXX CST S/P IDEXX CST

This sample handling procedure resulted in false negative errors in Groups I, π, and HI using the IDEXX breakpoints, but not with the adjusted CST breakpoints. The average S/P ratios obtained with this sample handling procedure (Table 4) are larger that those obtained with the fresh samples (Table 3). A further study, summarized in Table 5, below, examined whether the freeze/thaw treatment of the negative samples in the pooled sample was critical to the results. The known positive sample was fresh (incubated about 1.5 hours in PBS), while the negative samples were not fresh (3 days in PBS, frozen for the two intervening nights, but thawed each day). False negative errors were obtained for the groups with the inhibiting agent for both the manufacturer's breakpoints and the adjusted breakpoints. Table 5

30/30/30/20 50/50/50/10

Neg group 1641 4690 4697 4704 4711 4699 4725 4732 4739 4662 4676 4689

AVG S/P 0.10 0.10

False negative errors 12 12

Example 5

The Effect of Sample Handling Procedure on Both Negative Samples And Known Positive Samples

The negative samples from the same set of cattle used in Example 4 were used in this study, including Groups I, II and III, that included samples having interfering agents. Samples from a known PI animal (4718) were used as known positive samples. The results are summarized in Table 6, with the data presented in Tables 7 and 8. Eight sample handing procedures were studied.

Table 7 presents the data from three sample handling procedures in which a fresh sample was subjected to three hours incubation in PBS at room temperature (A), refrigeration in PBS for three hours (B) or freezing - storage at -20° C for three hours followed by thawing to room temperature (C) before testing. Using the manufacturer's breakpoints produced 10-11 false negative errors out of 12 samples, while the adjusted breakpoints produced three false negative errors for the three pooled samples (Groups I, II, and EI) having interfering agents. Table 7 also provides the results for tissue handling procedure D, which was 48 hours incubation in PBS at room temperature, frozen overnight at -20° C, and allowed to thaw at room temperature before testing. Tissue handling procedure D using the IDEXX breakpoints produced 3 false negative errors out of 12 samples, while the adjusted breakpoints produced zero false negative errors. Table 8 presents the data from four tissue sample handling procedures in which a fresh sample was subjected to various treatments that lasted about two days or longer before testing. Treatments E and F involved a long period of refrigeration in PBS (overnight or two days) either preceded by 12 hours in PBS at room temperature or followed by heating for one hour at 35° C. Treatments G and H involved two overnight freeze-thaw cycles; in H, the overnight freeze-thaw cycles were preceded by 2 days in PBS at room temperature. Treatment H produced the greatest reduction in false negative errors, but at a cost of four days added to the testing process.

Table 6 Tissue Handling

(All Tissue Sam les Fresh, <3 Days Since Collection At Start

Group I

Group Il

Group III

Group I

Group Il

Group III

Example 6

The Effect of Sample Handling Procedures of Positive Samples In Different Sized Sample Pools

As shown in Table 9, below, sample handling procedures, such as holding a sample overnight at -20° C, could further reduce false negative errors. The tissue sample was incubated in PBS at 4-5° C for 24 hours in a refrigerator, and then stored overnight in a -20° C freezer.

Table 9

Comparison of Fresh Sample Results to Results From Sample Incubated 24 Hours in PBS and Frozen Overnight (-20° C) Fresh Frozen

Pool Size S/P IDEXX CST S/P IDEXX CST

AVG S/P 0.10 0.45

False Negative Errors

Example 7 Identifying The Source of "Inhibition"

All samples were incubated in PBS at 4-5° C in a refrigerator for about 24 hours prior to testing. The positive sample used was 4718, which had single sample test results of 0.98 - 1.125. The negative samples containing an inhibitor were identified in a two- step procedure. First, pools of the positive sample 4718 with two negative samples were tested. The three sample pools showing the greatest "inhibition," defined as reduced OD values or S/P ratios, were retested as pools of the positive sample, 4718, with each single negative sample. The results are summarized in Table 10, below, where each row should be followed left to right. The negative samples listed in boldface italics, 4715 and 4676, caused the most "inhibition." The negative samples listed in boldface, 4677, 1641, 1646 and 4712, caused moderate "inhibition" and all other negative samples were considered to be minor "inhibitors" or cause no "inhibition". Table 10

Identifying The Samples Having "Inhibitor" Positive Sample 4718

Pool of Two Negative Pool of One Negative Negative Sample Samples Plus 4718 Sample Plus 4718 Only

Table 1 1 compares the degree of "inhibition" according to results in Table 10 and the serum neutralizing antibody titers to each BVDV subtype. The sample from the BVDV positive animal used in these studies, 4718, was subtyped and determined to be type Ib. The samples producing the largest "inhibition," 4715 and 4676, also had the highest type Ib antibody titers. Those samples producing moderate "inhibition" had moderate type Ib antibody titers and those negative samples causing low or no "inhibition" had the lowest type Ib antibody titers. These results suggest that the "inhibition" may be caused by a subtype-specific neutralizing antibody.

Table 11 BVDV Strain Antibody Titer of Samples Having "Inhibitor"

Serum Neutralizing Antibody Titer Sam le S/P 1b 1a 2

Largest Inhibition

Moderate Inhibition

Low or no Inhibition

Example 8 Effect Of Tissue Handling Procedures Table 12 summarizes the tissue handling procedures used in the following studies. Table 13 shows the outcomes of using different tissue handling procedures with inhibitors at 2:1 and 7:1 dilutions and compares the percentage improvement in the S/P ratio by a tissue handling procedure.

A sample from animal No. 4715 which was shown to have a large "inhibition" effect above was used as a negative sample in the 2: 1 and the 7: 1 pools. The negative samples used for the pools were the same for all tissue handling procedures.

In Table 13 those positive results that would have been found using the standard test kit protocol breakpoints are indicated by (b), positive results that would have been found using the CST breakpoints but not by the standard test kit protocol breakpoints are indicated by (a) and false negative errors by both are without an indicator letter.

The charts to the right of the outcomes illustrates the increase in values obtained when comparing the means for each tissue handling procedure to incubation for 10 minutes in PBS at room temperature procedure (standard test kit protocol) which was used as control group. The tissue handling procedure highlighted showed the greatest increase in S/P.

Table 13

Different tissue procedures with inhibitor at 2:1 and 7: comparing percentage improvement in S/P

S/P

Test found positive by CST procedures (with adjusted breakpoints)

Test found positive by routine procedures Tables 14 and 15 summarize the results of studies in the effects of higher temperatures and heating times were examined.

Table 14

HEATING TIMES AND TEMPERATURES 4718 POSITIVE AS SINGLE AND WITH INHIBITORS 4715 AND 4676

Table 15

INHIBITOR

Table 16 summarizes the results of studies in the effects of higher temperatures followings by freezing for various times were examined. All samples (including the controls) were incubated in PBS for one hour at 105 F° (41° C), and then frozen for the indicated time.

Table 16

Comparing The Results Obtained With Various Freezing Times After The Samples Were Incubated In PBS 1 hr @ 105 F° (41° C)

Inhibitor Samples Are 4715 And 4676 And The Positive Samples Are The Pl Fetus Samples

Control Control 15 Min 30 Min 45 min 60 min

4715/6176 0.19 0.20 0.70 1.09 1.09 0.70

0.34 0.26 0.80 1.21 1.17 1.71

0.38 0.22 0.84 1.31 1.26 0.88

4715/6184 1.61 1.63 1.51 1.42 1.31 1.41

1.62 1.63 1.39 1.45 1.35 1.62

1.50 1.45 1.52 1.53 1.40 1.69

4715/13106 0.55 0.50 0.68 0.89 1.09 0.44

0.52 0.52 0.61 0.99 1.10 0.68

0.69 0.52 0.60 1.02 1.17 0.72

4676/6176 0.36 0.35 0.92 1.34 1.26 1.16

0.51 0.42 1.17 1.31 1.38 1.34

0.59 0.41 1.18 1.23 1.24 1.20

4676/6184 1.43 1.79 1.62 1.62 1.46 1.65

1.63 1.82 1.59 1.67 1.40 1.73

1.44 1.65 1.45 1.46 1.32 1.46

4676/13106 0.63 0.84 1.00 1.16 1.22 0.91

0.87 0.83 0.94 1.29 1.30 1.23

0.79 0.80 0.95 1.15 1.24 1.15

6176 1.12 1.07 1.56 1.40 1.51 1.43

1.27 1.30 1.41 1.45 1.51 1.52

1.12 1.17 1.42 1.46 1.41 1.40

6184 1.96 2.05 1.85 1.83 1.69 1.57

1.74 1.92 1.82 1.77 1.62 1.71

1.47 1.88 1.70 1.63 1.55 1.59

13106 1.44 1.45 1.46 1.65 1.47 1.13

1.39 1.39 1.48 1.53 1.34 1.45

1.32 1.40 1.34 1.39 1.38 1.36

Control Control 15 Min 30 Min 45 min 60 min

S/P Mean 1.05 1.24 1.38 1.34 1.29

% difference

15 Min 15.6%

30 Min 28.5%

45 Min 25.1 %

60 Min 20.2% Table 17

S/P

4715/4676 sample 4715 (2:1 ) ALONE 4715 (7:1 ) (7:1 ) MEAN

1 10-7 Rt 10 min 0.22(b) 0.52(b) 0.07(a) 0.05 0.21

Rt 1 hr 0.17(a) 0.67(b) 0.06 0.03 0.23

Rt 1 hr/fr 0.22(b) 0.70(b) 0.08(a) 0.06 0.25

Rt 1 hr/di 0.19(a) 0.65(b) 0.08(a) 0.05 0.23

41 °C 1 hr 0.13(a) 0.65(b) 0.10(a) 0.06 0.22

41 ⁰C 1 hr/fr 0.19(a) 0.62(b) 0.07(a) 0.05 0.23

41 ⁰C 1 hr/di 0.22(b) 0.68(b) 0.10(a) 0.05 0.27

115-60 Rt 10 min 0.03 0.1 1 (a) 0.02 0.01 0.06

Rt 1 hr 0.06 0.23(b) 0.03 0.01 0.10

Rt 1 hr/fr 0.05 0.16(a) 0.03 0.02 0.08

Rt 1 hr/di 0.08(a) 0.23(b) 0.03 0.02 0.11

41 ⁰C 1 hr 0.37(b) 0.25(b) 0.03 0.01 0.18

41 "C 1 hr/fr 0.07(a) 0.24(b) 0.03 0.01 0.08

41 ⁰C 1 hr/di 0.07(a) 0.27(b) 0.04 0.01 0.09

129-15 Rt 10 min 0.09(a) 0.35(b) 0.05 0.04 0.12

Rt 1 hr 0.31 (b) 0.75(b) 0.11 (a) 0.08(a) 0.29

Rt 1 hr/fr 0.19(a) 0.66(b) 0.08(a) 0.07(a) 0.23

Rt 1 hr/di 0.22(b) 0.79(b) 0.13(a) 0.07(a) 0.28

41 ⁰C 1 hr 0.23(b) 0.85(b) 0.10(a) 0.08(a) 0.29

41 ⁰C 1 hr/fr 0.29(b) 0.87(b) 0.12(a) 0.09(a) 0.32

41 °C 1 hr/di 0.23(b) 0.70(b) 0.09(a) 0.06 0.25

(a) (b)Test found positive by CST procedures (with adjusted breakpoints) (b)Test found positive by routine procedures

mean7/60/15 % > contr.

Rt 10 min 0.13 0.0% Rt 1 hr 0.21 61.4% Rt 1 hr/fr 0.19 47.4% Rt 1 hr/di 0.21 62.1 %

41⁰C 1 hr/fr 0.22 69.6% 41 °C 1 hr/di 0.21 61.4%

Table 17, above, summarizes the results of studies in the effects of room temperature incubations of various durations with or without freezing for various times, or higher temperatures with or without freezing for various times were examined. Samples processed according to the manufacturer's recommended test kit protocol (room temperature for 10 minutes) were used as the control. As in Table 13, those positive results that would have been found using the standard test kit protocol breakpoints are indicated by (b), positive results that would have been found using the CST breakpoints but not by the standard test kit protocol breakpoints are indicated by (a) and false negative errors by both are without an indicator letter. The chart below the data outcomes illustrates the increase in S/P values obtained when comparing the means for each tissue handling procedure to control. The tissue handling procedure highlighted showed the greatest increase in S/P value.

Table 18, below, summarizes the results of studies of the effects of higher temperatures with or without subsequent freezing or room temperature incubation.

Table 18

Comparison of Results Obtained By 1 hr @ 41⁰C Followed By 40 Minutes @ RT or 30 Minutes @ - 20⁰C S/P

Sample 1 hr41°C 1 hr41°C, 40min RT 1 hr 41°C, 30min -20"C

Table 18 continued

Comparison of Results Obtained By 1 hr @ 41 °C Followed By 40 Minutes @ RT or 30 Minutes @ - 20⁰C C S/P

Sam le 1 hr41°C 1 hr41°C,40minRT 1 hr41°C 30min -20°C

Table 18 continued

Comparison of Results Obtained By 1 hr @ 41⁰C Followed By 40 Minutes @ RT or 30 Minutes @ - 20°C S/P

Sam le 1 hr41°C 1 hr41°C 40min RT 1 hr 41°C 30min -20"C

Table 19 compares the results obtained after testing 190 known positive samples after six different sample handling procedures (described in Table 12, above) using both the IDEXX and the CST breakpoints. Pooled samples consisted of an aliquot of one known positive sample and aliquots of six known negative samples. Note that the manufacturer's protocol was particularly likely to produce false negative errors when known positive samples were pooled with negative samples that produced "inhibition," i.e., samples 3, 18, 30, 32, 65, 120, 125, 150, and 173. As above, positive results that would have been found using the CST breakpoints but not by the standard test kit protocol breakpoints are indicated by "(a)". Here false negative errors produced by both sets of criteria are indicated by "(n)".

Table 20, below, summarizes the effectiveness of different sample handing procedures and the adjusted breakpoints in reducing the numbers of false negative errors

Example 9 Use Of A Visible Colorant In Phosphate Buffered Saline

It has been found that a dilute solution of a visible colorant in phosphate buffered saline is useful in positioning a sample in the wells on a microtiter plate. The colorant does not interfere with the accuracy of the testing procedure since the colored PBS is replaced during the testing protocol before an optical measurement is made. Table 21 provides the S/P results for known positive samples in PBS with colorant ("+ Dye") and without a colorant ("No Dye"). In a preferred embodiment, the colorant is a green mixture of FD&C Yellow 5 and FD&C Blue at a final concentration of 0.005% v/v.

Table 21 Colorant Study Results

Known Positive + Dye No Dye Known Positive + Dye No Dye Samples (S/P) (S/P) Samples (S/P) (S/P)

84 0.45 0.52 1 0.198 0.206

45 0.45 0.52 1 0.397 0.436

1 0.53 0.46 45 0.219 0.206

76 0.35 0.40 45 0.406 0.494

39 0.46 0.39 76 0.214 0.127

28 0.58 0.50 76 0.298 0.411

57 0.31 0.40 18 0.215 0.222

38 0.31 0.34 18 0.379 0.390

Example 10 Use of a Sample Pool Size of Twenty

The results of using a sample pool size of twenty with 46 known positive samples processed using tissue handling treatment R are shown in Table 22, below. The handling treatment plus the adjusted CST breakpoints produced no false negative errors, while the IDEXX S/P criteria produced 13 false negative errors.

Table 22

Results Obtained Using A Sample Pool Size of 20

All Positive Samples Treated Using CST Tissue Handling Protocol "R" (12 hr soak in PBS) IDEXX

Criteria CST Protocol

Table 22 (Continued)

Results Obtained Using A Sample Pool Size of 20 All Positive Samples Treated Using CST Tissue Handling Protocol "R" (12 hr soak in PBS)

IDEXX CST

Criteria Protocol

S/P

Sample VALUE Results Results

study

Example 11 Effect of Increasing Cut Sample Surface

Increasing the cut surface of the tissue sample increased the S/P values, which would help increase the test sensitivity. In this pairs of punch samples or notch samples were taken adjacent to each other, minimizing the effects of the differences in the thickness of the various ear notch or punch biopsies. One tissue sample was treated as harvested and the partner adjacent tissue sample was divided and then subjected to the same sample handling protocol. The data in Table 23 are from studies of 5mm punch biopsies. The data in Table 24 are from ear notch tissue samples. These data show that for paired samples of substantially the same volume, an increase in the cut surface area is associated with higher S/P ratios.

In one study, roughly cylindrical tissue samples were cut from the animals' ears using a round biopsy punch having a 5 mm diameter that produced a tissue plug about 15.7 mm in circumference and having a height of the thickness of the ear tissue.

Samples were prepared from the plugs and from the corresponding adjacent plugs after being bisected in a plane perpendicular to the ear surface and containing the long axis of the plug. Average S/P values obtained from 36 samples prepared from biopsy plugs before and after bisection showed an increase in S/P from 0.17 to 0.31 for a increase in cut surface edge length from 15.7 mm to 25.7 mm. Similarly, ear notch tissue was used to prepare fifty samples from tissue biopsies having a cut edge length of about 26 mm and after a cut that split the corresponding adjacent biopsies to produce a cut edge length of about 52 mm, resulting in measured S/P levels of 1.32 and 1.53, respectively.

The claims should not be read as limited to the described order or elements unless stated to that effect. Therefore, all embodiments that come within the scope and spirit of the following claims and equivalents thereto are claimed as the invention.

Claims

CLAIMS What is claimed:

1. A method of processing a tissue sample to provide a liquid sample to be tested for the presence of an analyte, comprising the steps of: immersing a tissue sample in a buffered physiological saline solution; incubating the tissue sample in the buffered physiological saline solution at about 32-41 degrees Celsius for at least 1 hour; reducing the temperature of the tissue sample in buffered physiological saline solution by at least about fifteen degrees Celsius for at least fifteen minutes; and removing an aliquot of the buffered physiological saline solution thereby providing a liquid sample to be tested for the presence of an analyte.

2. The method of claim 1 wherein the buffered physiological saline solution is phosphate buffered saline solution.

3. The method of claim 1 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to room temperature after incubation at about 32-41 degrees Celsius.

4. The method of claim 1 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to about 2-7 degrees Celsius after being incubated at about 32-41 degrees Celsius.

5. The method of claim 1 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to about -20 degrees Celsius after being incubated at about 32-41 degrees Celsius.

6. The method of claim 1 further comprising the steps of performing the test for the analyte on the liquid sample, and evaluating the test measurements using a standard that has been empirically determined using known positive and negative samples to minimize false negative errors.

7. The method of claim 6 wherein the liquid sample is a pooled sample.

8. A method of increasing the sensitivity of a test that measures the presence or amount of an analyte located in a tissue based on the specific high-affinity binding of a binding partner to the analyte, comprising the steps of: immersing a tissue sample in a buffered physiological saline solution; incubating the tissue sample in the buffered physiological saline solution at about 32-41 degrees Celsius for at least 1 hour; reducing the temperature of the tissue sample in buffered physiological saline solution by at least about fifteen degrees Celsius for at least thirty minutes; removing an aliquot of the buffered physiological saline solution thereby providing a liquid sample to be tested for the presence of an analyte; testing the liquid sample using the test based on the specific high-affinity binding of a binding partner to the analyte; detecting the presence or amount of the analyte by measuring the formation of an analyte/binding partner complex; and evaluating the test measurements using a standard that has been determined empirically using known positive and negative samples to minimize false negative errors, thereby increasing the sensitivity of the test.

9. The method of claim 8 wherein the buffered physiological saline solution is phosphate buffered saline solution.

10. The method of claim 8 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to room temperature after incubation at about 32-41 degrees Celsius.

11. The method of claim 8 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to about 2-7 degrees Celsius after being incubated at about 32-41 degrees Celsius

12. The method of claim 8 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to about -20 degrees Celsius after being incubated at about 32-41 degrees Celsius.

13. A method of increasing the sensitivity of a test that measures the presence or amount of an antigen located in a tissue, where the antigen is likely to be detected in less that 10% of the samples to be tested, comprising the steps of: immersing a tissue sample in a buffered physiological saline solution; incubating the tissue sample in the buffered physiological saline solution at about 32-41 degrees Celsius for at least 1 hour; reducing the temperature of the tissue sample in buffered physiological saline solution by at least about fifteen degrees Celsius for at least fifteen minutes; removing an aliquot of the buffered physiological saline solution thereby providing a liquid sample to be tested for the presence of an antigen; pooling an aliquot of b liquid samples to form a pooled sample, where b is an integer from 2 to 20 inclusive; testing the pooled sample for the presence and amount of the antigen; evaluating the test measurements for each pooled sample using a standard that has been determined empirically using known positive and negative samples to minimize false negative errors; retesting aliquots from individual liquid samples that contributed to pooled that were evaluated as other than negative for the presence of the antigen; and evaluating the test measurements for each individual liquid sample using a standard that has been determined empirically using known positive and negative samples to minimize false negative errors, thereby increasing the sensitivity of the test and reducing the total number of tests required to measure the presence or amount of an antigen located in a tissue, where the antigen is likely to be detected in less that 10% of the samples to be tested.

14. The method of claim 13 wherein the buffered physiological saline solution is phosphate buffered saline solution.

15. The method of claim 13 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to room temperature after incubation at about 32-41 degrees Celsius.

16. The method of claim 13 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to about 2-7 degrees Celsius after incubation at about 32-41 degrees Celsius.

17. The method of claim 13 wherein the temperature of the tissue sample in buffered physiological saline solution is reduced to about -20 degrees Celsius after incubation at about 32-41 degrees Celsius.

18. The method of claim 13 wherein the standard for classifying the result is: S/P > 0.39, antigen positive;

S/P = 0.20-0.39, suspect;

S/P = 0.07-0.19, requiring re-check; and

S/P < 0.07, antigen negative.

19. The method of any one of claims 1, 8, or 13 wherein the buffered physiological saline solution contains a non-interfering colorant.