GB2528393A - Improvements in or relating to detection of parasite infestations - Google Patents

Abstract

Disclosed is a method of detecting a tapeworm infestation in an equid animal subject such as a horse, the method comprising the step of performing, in vitro, (a) an assay on a saliva sample, obtained from the subject, to detect the presence and/or amount in the sample of a first analyte Specific ELISA which is or comprises antibody specific to antigen/s from the tapeworm such as Anaplocephala sp.; and (b) an assay on a saliva sample, obtained from the subject, to measure the amount and/or concentration of a second analyte such as the total immunoglobulin total ELISA, the result of which assay gives a measure of the dilution of the sample.

Description

Intellectual Property Office Application No. GB1513925.6 RTTVI Date:9 September 2015 The following terms are registered trade marks and should be read as such wherever they occur in this document: Tween (page 22, 23 & 24) EquiSal (page 30, 31&32) MedCalc (page 31)-protected Intellectual Property Office is an operating name of the Patent Office www.gov.uk/ipo Title: Improvements iii or Relating to Detection of Parasite Infestations

Field of the Invention

The present invention relates to a method of detecting parasite infestations, especially tapeworm infestation in horses.

Background of the Invention

Tapeworms of the genus Anaplocephala are common pathogens of the horse. Three tapeworm species have been identified in horses. By far the most common, and pathogenic, is Anaplocephala perfoliata. The other two known tapeworms in horses are Anaplocephala magna and Paranoplocephala mamilliana. For present purposes, all three species are considered members of the genus Anaplocephala. They inhabit the intestine mainly at the ileo-caecal junction (which is an intestinal valve), where they attach themselves to the gut lining by means of special suckers acting as is attachment structures, When numerous tapeworms are present, the junction can be obstructed and the mucosal surface can become seriously inflamed, leading to significant impairment of intestinal function. Drugs which can eliminate tapeworms are available, but these should not be used indiscriminately, and there is a strong movement within the veterinary, scientific and equitation communities to limit the use of anthelmintic drugs to situations when they are really necessary.

Such targeted drug use is possible only when there is a means by which the presence of damaging infections of tapeworms can be identified, but diagnosis of tapeworm infestations in live horses is difficult. Detection of their eggs in horse faeces is not reliable by standard techniques for determining presence of eggs of other internal parasites, such as nematodes, Therefore, not finding tapeworm eggs in faeces does not mean these parasites are actually absent from a horse, Tapeworm infestations can sometimes be verified by finding mature worm specimens in faeces after a horse has been treated with a drug active against these parasites.

It has been found that the presence of tapeworms in a horse can be inferred by detecting blood antibodies that specifically bind to the secretory antigens shed from the tapeworms into the gut lumen. These antibodies can be detected by standard methods, such as enzyme-linked immunosorbent assay (ELISA) techniques. The blood antibody method is a substantial improvement over egg counting, but it only infers that worms are present, rather than proving that they are present, because blood (systemic) antibodies normally remain in the circulation long after the stimulating substance i.e. (the tapeworm antigens) has gone. This is because there is a strong memory effect in the systemic antibody system, and this is why vaccines are effective.

For this reason, a blood antibody test for tapeworms can give a strong positive response even after the worm infection in fact has been eliminated, The interaction between parasite antigens released in the gut lumen and the systemic immune system io is complex, being affected by various factors including the presence of gut inflammation, the integrity of the mucosal epithelium, the past exposure of the immune system to the particular antigens, the concentration of the antigens in the lumen and the route of entry from the lumen into the systemic immune compartment.

A blood antibody response requires antigens from the tapeworm to pass through the is mucosal lining of the intestine in sufficient quantities to be taken up by dendritic cells of the systemic immune system eventually leading to the production of lgG by B-lymphocytes, following well-defined immunological pathways and intercellular interactions. It therefore follows that the mucosal lining must in some way allow antigen ingress before a blood antibody response can occur. A strong blood antibody response suggests that both a significant amount of antigen is, or has been, present (proportional to the number of tapewonris in the vicinity), together with enough local inflammation to render the mucosa permeable to those antigens. The extent to which local mucosal inflammation and other factors influence either new or repeat blood antibody responses is difficult to define, but a substantial blood concentration of specific antibody strongly infers that a damaging tapeworm infection is present, in accordance with the published specificity and sensitivity of available blood antibody tests.

The blood antibody test is a useful improvement over basic faecal egg counts for tapeworms. It is available commercially as a technical service (e.g. offered by Diagnosteq at the University of Liverpool, Veterinary School or Rossdales Diagnostic Laboratory, Newmarket, Suffolk) which is known and used by veterinary surgeons and informed horse owners. The dependence of the result on the presence of a -1, pathogenic tapeworm burden (rather than a low burden) is actually an advantage, because anthelmintic drugs should only be used when there is a clinical problem to be resolved. It is held by some authorities that low numbers of tapeworms that are not causing damage to the intestine are not detrimental to the health of the horse and do not need to be treated by anthelmintics, It must be recognised that once a horse's systemic immune system has been primed by a first encounter with tapeworm antigens, it can be in a state of immunological sensitisation in which subsequent encounters with the same antigens will result in more rapid and intense responses. This means that a given antibody level can result in from very different tapeworm numbers, depending on the responsiveness of the systemic immune system at the time of sampling.

At the practical and economic level, there is a further problem with the blood test approach, in that it is always necessary to take a blood sample for analysis. This adds cost and inconvenience to the process, which can discourage wide uptake of the test.

is There are very few anthelmintic drugs currently available, and the modes of action even more limited, so the emergence of drug-resistance would be a very serious problem. Accordingly, it is important that animals are treated with anti-parasitic drugs only when necessary, to reduce the risk of drug resistance emerging, and to reduce the impact of potentially toxic compounds on the animals and on the environment more generally.

For all these reasons, despite the inherent benefits of the blood testing services, there remains a major unmet need for an accurate, reliable and non-invasive method for detecting pathogenic burdens of parasites in horses and other animals.

The present invention aims to meet the unmet need noted above.

Summary of the Invention

In a first aspect the invention provides a method of detecting a tapeworm infestation in an equid subject, the method comprising the step of performing, in vitro, (a) an o assay on a saliva sample obtained from the subject, to detect the presence and/or amount in the sample of a first analyte which is or comprises antibody specific to antigen/s from the tapeworm; and (b) an assay on a saliva sample, obtained from the subject, to measure the amount and/or concentration of a second analyte, the result of which assay gives a measure of the dilution of the sample (e.g. as a consequence of the rate of saliva production by the subject at the time of sampling, and/or due to variation between individual horses).

It will be apparent to those skilled in the art that the sample used for steps (a) and (b) should preferably be the same (which includes use of different aliquots of the same sample) or, failing that, two samples which are obtained substantially io contemporaneously (e.g. within 10 minutes of each other, preferably within 5 minutes of each other).

By way of explanation, the rate of saliva production in an equid animal can vary greatly, and high rates of production will result in a relatively dilute sample, in which is the concentration of analvtes is reduced, which would tend to give a misleading assay result, For instance, the rate of saliva production is greatly increased in equids by exercise with a bit in the mouth or by eating. The inventors propose to overcome this problem by measuring a second analyte, the concentration of which is, in large part, inversely proportional to the rate of saliva production at the time the sample was obtained from the subject. In this way, the effect of dilution of the sample can be allowed for when interpreting the assay result. Preferably the concentration of the second analyte is generally not affected by, or is substantially insensitive to, the parasitic tapeworm burden in the animal, The invention thus provides an effective method of detecting and diagnosing a tapeworm infestation in an equid (which term encompasses, /nter al/a, donkeys, mules, hinnies and zebras), more especially in a horse, The assay of the invention comprises the detection and/or measurement of the amount of antibody specific to antigen/s from tapeworm, especially mucosal antibody, For present purposes, "mucosal" antibodies may be defined as antibodies produced locally in a mucosal location by plasma cells of the mucosal immune system and actively transported through mucosal epithelial cells by a secretory mechanism involving a transport molecule, In the case of mucosal IgA and 1gM, the secretory transport molecule is the polymeric immunoglobulin receptor (PIgR), and in the case of mucosal IgG it is the so-called neonatal IgGFc receptor (FcRn), or its functional equivalent in species in which a close homologue may not yet have been defined. The FcRn is actually present and active in this role throughout life, despite the allusion to neonatal in its name. Mucosal antibodies can also be defined by the way they are induced within the mucosal immune system, which is a major compartment of the immune system, consisting of cells (mainly lymphocytes and dendritic cells) and their products (especially antibodies of the IgA, 1gM and lgG class) which are associated io with mucosal epithelia, generally as part of the so-called Mucosa-Associated Lymphoid Tissue (MALT). The participating cells (especially the antibody-producing B-lymphocytes) are characterised by their propensity to become located in mucosal sites and to actively home to such locations, Intestinal mucosal immune responses are triggered mainly in response to antigens that are absorbed across the is mucosal epithelium actively through epithelial cells or through specialised M-cells associated with Peyer's Patches in defined regions of the gut. Most of the activated mucosal B-cells originating in a particular mucosal site (e.g. in a Peyer's Patch of the gut or in the region of a localised inflammation/infection) will migrate into the lymphatic system and will make their way back into the submucosal tissues (near to the point at which they were activated) where they can most effectively deliver their antibodies to the mucosal epithelium for active secretion into the gut lumen, The few of these activated mucosal B-lymphocytes that do not return to the gut will migrate more widely to colonise distant mucosal sites such as the salivary gland, instead of returning to the region where their target antigen is present. These wider-roaming B-lymphocytes set-up small temporary colonies where they actively secrete mucosal antibodies into the secretion associated with the site they have occupied, for example, into the saliva, Although IgA and 1gM antibodies are usually considered to be the classical mucosal antibodies, there is no doubt that certain lgG subclasses can also operate as mucosal antibodies, utilising the FcRn (or its functional equivalent) as the secretory transport molecule with which to traverse epithelial cells.

For present purposes, IgG(T), (a subclass of IgO present in horses), from mucosal sites in horses is considered as a bona fide mucosal antibody with typical mucosal immune system characteristics with respect to induction, distribution, memory and persistence.

The antibody population detected in the method of the present invention may be specific to a single tapeworm antigen, or to two or more antigens, derived from or present in the tapeworm. For example, there may be an antigenic molecule produced by the parasite which, in vivo, is processed into two or more separate molecules or fragments, or the host may produce a variety of mucosal antibodies which are specific to one or more distinct tapeworm antigens and, if desirable and advantageous, an ELISA well or other assay surface could, for example, be coated with a mixture of two or more tapeworm antigens.

is The person skilled in the art recognises that an antibody being described as "specific" for a particular antigen relates to the fact that the antigen-binding site located within the variable domains of the light and heavy chains (defined as the "paratope") will bind more or less exclusively to the antigen which triggered the antibody. This usually involves high affinity binding, resulting from a variety of cooperative molecular interactions. However, there are other regions of the antibody molecule which can be bound by other molecules, such as Pc receptors, complement proteins and various bacterial proteins, such as protein-A from Staphj.'icoccus aureus.

Moreover, as a relatively large protein, an antibody molecule is prone to all the non-specific binding interactions that can cause proteins in general to bind more or less avidly to surfaces with which they come into contact. These non-specific binding interactions are generally shared among all antibodies of the same class, regardless of their antigen specificity. Thus, an antibody will generally bind to its cognate antigen with an affinity that is orders of magnitude greater than the affinity with which it binds to most other molecules present in a typical assay mixture. The antibody will always bind preferentially to the particular antigen for which it is specific, if the antigen is present and accessible, but in the presence of an excess of other antibodies having a different antigen specificity, the other, weaker binding interactions can result in an overall level of Ig binding to an immunoassay surface that swamps the specific binding of a relatively few specific antibodies binding with high affinity to their cognate antigen. For these reasons, it can be difficult or impossible to detect low levels of high affinity antibodies specific for a particular antigen in an environment dominated by surfaces to which proteins in general can bind by non-specific interactions.

Equally the person skilled in the art will recognise that the term "antibody" does not necessarily indicate that a single molecular species is required. Rather, the term can encompass a heterogeneous mixture or population of molecules, and a "specific io antibody" may refer to that portion of a mixture which is specific for the antigen in question.

In preferred embodiments, the specific antibody detected by the method of the invention is antibody specific for the 12/13KD antigen of Anapocephalus sp is (Proudman & Trees, Parasite Jinmunof, 1996;18 499-506).

The assay may detect or measure all the antibody (i.e. regardless of antibody class) specific for the tapeworm antigen or may involve the detection or measurement of a selected class or classes of antibody, especially classes of antibody functioning as mucosal antibodies as herein defined, In particular, the assay may be specific for a selected subclass or subclasses of IgG and/or IgA. More preferably, the assay of the invention is specific for antibody of the IgG class functioning as a mucosal antibody.

In the human immune system there are four IgG subclasses, designated as IgGI -IgG4, and in the horse there are seven. The number, nomenclature and function are variable between different species. In horses, although up to seven putative IgG subclasses or isotypes have been identified (Wagner ci at, J. Immunol. 2004; 173 3230-3242), the four most abundant IgO subclasses are commonly referred to as IgGa, IgGb, IgGc and IgG(T).

The second analyte can be any analyte which is suitable for measurement and is normally present in equid saliva, Preferably the analyte is measured by an ELISA.

(Conveniently both the first and second analvtes are measured by ELISA). The second analyte may conveniently be selected from the group consisting of: electrolytes; proteins, including mucus glycoproteins and enzymes, (especially proteases such as serine proteases, acid phosphatase and kallikreii); hormones such as free steroids, conjugated steroids, and protein hormones; thiocyanate; and epidermal growth factor.

Specific examples of second analytes or second analyte assays include: alpha-amylase concentration; total protein concentration; cystatin-C concentration; serum albumin concentration; creatine concentration; and dehydroepi andosterone sulfate (DEIEA-S) concentration. Assays for measurement of the foregoing analytes are generally commercially available (e.g. from Salimetrics Europe Limited, Newmarket, Suffok, CB8 7SY).

In a preferred embodiment, the second analyte comprises an immunoglobulin or an immunoglobulin component in the saliva sample, such as a particular immunoglobulin chain, The second analyte assay may, for example, comprise an assay for total antibody concentration in the sample, (i.e. irespective of the antigen-binding specificity of the antibody) or an assay for the total concentration of a particular class or subclass of immunoglobulin (again, irrespective of the antigen-binding specificity of the immunoglobulin). Examples include assays for total secretory IgA or immunoglobulin J chain component. A preferred embodiment of the invention utilises an assay for total IgG(T) concentration, which the inventors have found to give especially reliable results.

A second analyte assay which involves measurement of an immunoglobulin or immunoglobulin component or the like is generally to be preferred for present purposes since an assay of this sort will not only give a measure of the dilution of the sample, but is also partly dependent on the animal's ability to secrete immunoglobulin into saliva, which will vary from animal to animal (e.g. dependent on the number of antibody secreting cells in the animal's salivary glands), Assay methods generally suitable for the quantitative or semi-quantitative measurement of antibodies in biological samples are well-known. One of the most convenient is the enzyme-linked immunosorbent assay or ELISA, and the method of the present invention preferably utilises ELISA format assays throughout, but other assay formats could in principle be used, such as RLk, and the like, and even liquid phase assays such as fluorescence polarization assays.

The amount of tapeworm-specific antibody!ies (typically mucosal antibodies) present in the sample may be measured in absolute terms or, more conveniently, in relative terms. For example, the amount of antibody may be compared to one or more threshold values by comparing an OD reading or the like obtained from an ELISA.

io The OD reading may be directly compared with one or more predetermined threshold OD values, which comparison indicates the presence or absence of a parasite infestation and, optionally, gives a semi-quantitative indication of the size of the parasite burden in the subject (e.g. low, medium or high).

is Alternatively, the one or more threshold values may not be predetermined, but may be determined by, for example, performing an equivalent assay using one or more calibration reagents containing known (in absolute or relative terms) amounts of tapeworm-specific antibody, which calibration may be performed on each occasion the assay is performed.

Desirably a parallel calibration control assay is performed, which gives an indication of the amount of assay signal ("background signal") which might be generated by the assay format in question, even in the abseuce of any tapeworm-specific antigen. In a prefered embodiment the method of the invention utilises an ELISA to measure the amount of tapeworm antigen-specific antibody in the saliva sample, which assay generates an assay test signal in terms of e.g. an OD reading. Preferably a parallel calibration control assay is performed (with no sample added), which calibration control assay will generate a calibration control signal, in terms of e.g. an OD reading.

In a preferred embodiment the reading of the calibration control assay (e.g. an OD reading) is subtracted from that of the test signal, to give a corrected test result (e.g. OD) i.e. corrected test result = test result -calibration control result, In principle, this corrected result could be compared to one or more threshold values to give an assay outcome. Preferably however the corrected test OD (or other test reading) is converted to a test score' by comparing the corrected test OD or other test reading with a calibration curve obtained by using reagent preparations of known protein concentration (eg. by constructing a graph of concentration of protein or IgU [if known] against OD). This gives a test score, which can be compared to one or more threshold scores to yield an assay outcome.

As noted above, the method of the present invention is performed using a saliva sample from the subject. By utilising saliva samples, the method of the present io invention avoids the invasive blood-sampling step required by the prior art blood-based detection methods. However, the use of saliva involves a number of significant difficulties and it is very surprising that the method of the present invention is successful.

is The choice of saliva as a sample is far from obvious to the person skilled in the art for several reasons, Firstly, gut parasites such as tapeworms obviously live in the gut, not the mouth. Accordingly, it would be more obvious to consider testing for antibodies in the gut lumen or faeces. Although the phenomenon of distribution (seeding) of mucosal antibody-producing lymphocytes from the site of induction to unaffected remote mucosal sites is known, that very remoteness, together with the apparent quantitative randomness of the seeding process, and the sparseness of the B-lymphocyte colony that settles in the salivary gland, all suggest that the presence of tapeworm antigen-specific antibodies in saliva is too indirectly connected to the tapeworm infestation events to be a reliable marker, Secondly, the concentration of tapeworm-specific antibodies in saliva is so low that it is preferable to test the saliva undiluted, In practice, the saliva is normally used at a low dilution of about I in 3. Consequently, due to saliva viscosity and the promotion of non-specific binding, the ability to reliably detect diagnostically-significant amounts of specific antibody, especially a single antibody class, and more especially a single sub-class of lgG, in saliva is surprising. Even though the exact sequence of events and underlying mechanisms (and their relative contributions) responsible for the presence of these antibodies in saliva are too complex to be defined exactly in any

II

individual case, in practice the test result is found to correlate closely with the tapeworm burden, giving a diagnostically useful result.

Thirdly, the concentration of antibodies in saliva is very variable from one animal to another, in particular, there are big variations in the intrinsic seeding of antibody-producing cells (plasma cells) in the salivary glands, which means that (unlike for blood samples) there is a very wide range in the "normal" values of antibody concentrations (i.e. when tapeworms are absent), making it difficult to decide meaningful scores or cut-off values. Moreover, the concentration of antibodies in io saliva varies greatly over time (especially following salivation on exercise or eating) in the same individual animal due to the great variation in the rate of saliva production.

Finally, unlike blood (semm or plasma), the presence of which tends to diminish is background binding, some saliva components can adsorb strongly to solid-phase surfaces, making substrates such as ELISA wells sticky' for other proteins (such as IgG) to a variable extent, which means that salivary antibody (ELTSA) tests are subject to large, and variable, non-specific binding (NSB) effects, which would be expected to cause false results.

The inventors have found that this latter problem can be overcome to a significant extent by including, in the method of the invention, an assay to take account of the non-specific binding effects of saliva.

Accordingly, in some preferred embodiments, the method of the invention preferably comprises the steps of (ai) performing, in vitro, an assay on a saliva sample, obtained from the subject to detect the presence and/or amount in the sample of a first analyte which is antibody specific to antigen/s from the tapeworm; (afl) performing, in vitro, an assay on the saliva sample to obtain a measure of non-specific binding occuring as a result of the presence of the saliva; and (b) performing, in vitro, an assay on the saliva sample to measure the amount and/or concentration of a second analyte, the result of which gives a measure of the dilution of the sample.

Such an assay as (au) above may be referred to for present purposes as a "saliva control assay". Assay (a i) may be referred to as an "antigen-specific assay". The purpose of the saliva control assay is to provide an indication of the amount of signal' in the antigen-specific assay (ai) which might be an artefact, arising from non-specific binding occurring due to the presence of saliva in the assay mixture.

Thus, for example, where assay (ai) might give a certain numerical result (e.g. an OD reading in an ELISA), the result of assay (au) might be subtracted from the result of assay (ai) to give a "saliva-adjusted" result.

More specifically, the result of the saliva control assay (e.g. an ELISA OD reading) is may conveniently be converted to a saliva control score', by reference to a specific calibration control curve (e.g. a graph of lgG concentration against OD), The saliva control assay score may be subtracted from the calibrated specific score, or corrected specific test score, to give a corrected, saliva-adjusted, specific test score.

In preferred embodiments the method of the invention comprises the performance of a "calibration control" assay. The purpose of a calibration control assay is to provide an indication of the amount of assay signal which is attributable to reactions occurring in the antigen-specific assay and/or the second analyte assay other than those due to the presence of antigen-specific antibody, and other than those due to the presence of saliva. This can be achieved by omitting the saliva sample but including every other component of the antigen-specific assay. For example, where the relevant assay format is an ELISA, a calibration control assay may use an equivalent volume of buffer or other irrelevant aqueous liquid in place of the saliva sample. The result of the calibration control assay is conveniently subtracted from the result of the antigen-specific assay, Surprisingly, this simple and crude adjustment of the antigen specific absorbance (but not the saliva control absorbance), can provide significantly improved assay accuracy.

H

In some preferred embodiments the method of the invention comprises the performance of both a saliva control assay and a calibration control assay. Thus, in preferred embodiments, the method of the invention may comprise the steps of: (ai) performing, ii ri/rn, an assay on a saliva sample, obtained from the subject to detect the presence and/or amount in the sample of antibody specific to antigen/s from the tapeworm; (au) performing, in vitro, an assay on the saliva sample to determine the amount of non-specific binding occurring as a result of the presence of io the saliva; (aiii) performing, in vitro, a specific assay calibration control; and (b) performing, in vitro, a second analyte assay as aforesaid.

The specific assay calibration control may not necessarily be performed on every is occasion: for example, the results of one calibration control assay may be used to correct the result of a plurality of subsequent batches of specific assays, but more preferably a fresh calibration control assay lI be performed each time the method of the invention is performed.

As noted previously, the concentration of antibody in a saliva sample is affected by, inter a/ia, the rate of production of saliva: the higher the rate of saliva production, the lower the effective concentration of antibody in saliva. Accordingly it is necessary to perform an assay to measure a second analyte, the concentration of which is preferably largely or substantially independent of the level of tapeworm burden in the subject. A preferred second analyte is "total antibody" (which, in this context, may mean total IgG(T) antibody or other Ig class or subclass), irrespective of antigen-specificity, Thus in preferred embodiments, the method of the invention comprises the step of performing, in vitro, an assay on a saliva sample to determine the total amount of antibody in the sample. The amount' of antibody (especially "mucosal" antibody) may be determined in absolute or relative terms and optionally includes determination of a concentration.

N

By performing an assay for a second analyte, such as a "total" antibody assay, it is possible to validate the sample and the result of the antigen-specific assays and the diagnostic outcome of the method by checking that there is sufficient antibody present in the saliva to obtain meaningful results. For example, if the second analvte (e.g. total antibody) is present below a particular threshold, it is likely that the saliva sample was obtained at a time when the rate of saliva production was too high (e.g. shortly after a horse has eaten). In this way the "total" antibody or other second analyte assay acts as a quality control measure, as well as a definitive quantification tool.

A fbrther benefit of including a measure of total antibody in the sample is that it can be used to provide standardisation of the assay scores, Accordingly, monitoring of assay scores from individual animals over a period of time can legitimately be performed. This information can show, for example, if animals are likely to have is become re-infected and/or whether they are responding to a dose of anthelmintic drug or other anti-parasite treatment.

Those skilled in the art will appreciate that the second analyte assay, such as a "total" antibody assay, may not be intended to detect all the immunoglobulins present in the sample, Rather, it may be sufficient to detect all the immunoglobulin of a particular class or subclass, regardless of the antigen-specificity of the antibody. Thus, for example, if the parasite antigen-specific assay detects or measures immunoglobulin of a particular class or subclass, then the "total" antibody assay may simply provide a measure of the total amount of antibody of that class or subclass (regardless of antigen specificity) in the sample.

The performance of the second analyte assay allows the specific antibody assay result to be "normalised" using the result of the second analyte assay. For example, the specific antibody assay result may be divided by the result of the second analyte assay (and conveniently then multiplied by e.g. 100), In a particular embodiment, the method of the invention comprises the steps of (ai) performing, in viiro, an assay on a saliva sample obtained from the subject to detect the presence and/or amount in the sample of antibody specific to antigen/s from the tapeworm; (afl) performing, in vitro, an assay on the saliva sample to determine the amount of non-specific binding occurring as a result of the presence of the saliva; (aiii) performing, in vitro, a specific assay calibration control; and to (bi) performing, in vitro, an assay on a saliva sample to determine the total amount of antibody in the sample.

In the same way that the specific assay (ai) preferably includes a specific calibration control assay (aiii), where the method of the invention comprises a "total" antibody ts assay (bi), it will preferably also include a "total" calibration control assay (bii). As with the specific calibration control assay, it is preferred to include a "total" calibration control assay for each batch of testing.

Preferably, the method of the invention further comprises a step of subtracting the result of (aiii) from the result of(ai), to provide a corrected specific assay result. This is considered to be an especially desirable feature of the invention to ensure that a negative value result is obtained for a horse with no tapeworms present. Such a negative value is especially needed where the second analyte assay yields a very low result (e.g. because the saliva sample is very dilute). In this situation, and without the inclusion of the above subtraction step, then the specific test score will be a positive value and may be artificially exaggerated by dividing by a low second analyte assay score, Accordingly it is a preferred feature of the invention that a specific antibody assay result for a horse with no tapeworms present generates a negative score so that, even when normalised, the normalised result is negative and therefore remains below the threshold for a low diagnosis.

In preferred embodiments, the method of the invention further comprises the step of converting the corrected specific assay result to a calibrated specific assay score, by performing a step of comparing the specific assay result with a specific calibration curve. The specific calibration curve is typically constructed using calibration reagents of known concentration (e.g. purified serum immunoglobulin of known protein concentration). The calibration curve may be physically plotted on paper or, more conveniently, is effectively formed "in silico", For example, a corrected specific assay result (such as an ELISA O.D.) may be read against the calibration curve to provide a calibrated specific assay score.

It will be apparent to those skilled in the art that the type of assay used in step (a ii) io and/or (a iii) and (bii) should preferably be substantially identical to that used in step (a i), in order for meaningful conclusions to be drawn. Typically each of assays (a i), (a ii), (a iii), (bi) and (bii) is an ELISA.

It will be appreciated by those skilled in the art that referring to the various assay is steps as "(a i)", "(a ii)", "(a iii)", "(bi)" etc. does not imply that the assays must be performed in any particular order (nor that all of the assay steps must necessarily be performed in any particular embodiment of the invention, unless the context dictates otherwise). Indeed, the assays will normally be performed in parallel and substantially simultaneously, typically using automated apparatus adapted and configured for this task, and typically using different aliquots of a single sample.

In addition, a determination of the amount or concentration of total Ig or total IgG (bi), and parasite-specific Ig or IgG, may be made in absolute terms or in relative terms, More especially, the total and parasite-specific assays may measure just a particular IgO subclass, such as IgG(T).

In detail, a preferred embodiment of the invention comprises the steps of: (ai) performing, in vitro, an assay on a saliva sample obtained from the subject to detect the presence and/or amount in the sample of antibody specific to antigen/s from the tapeworm; (afl) performing, in vitro, an assay on the saliva sample to determine the amount of non-specific binding occuning as a result of the presence of the saliva; (aiii) performing, in vitro, a specific assay calibration control; Vi (hi) performing, in vitro, an assay on a saliva sample to determine the amount of total antibody or other second analyte in the sample; (bii) performing, in vitro, a total antibody assay or other second analyte assay calibration control; (ci) subtracting the result of assay (aiii) from the result of assay (ai) to give a "corected specific assay result"; (cii) subtracting the result assay (bii) from the result of assay (bi) to give a "corrected total antibody (or other second analyte) assay result"; (ciii) comparing the corrected specific assay result from step (ci) with a specific io calibration curve to give a specific calibrated score; (civ) comparing the result of the saliva control assay (au) with the aforementioned specific calibration curve to give a saliva control score; (cv) subtracting the saliva control score (obtained from step civ) from the specific calibrated score (obtained from step ciii) to give a saliva-corrected specific is score; and (cvi) comparing the corrected total antibody (or other second analyte) assay result (obtained from step cii) with a total calibration curve, to give a calibrated total assay score.

In a final step the "Saliva Score" for the subject is calculated by dividing the saliva corrected specific score (obtained from step cv) by the calibrated total assay score (obtained from step cvi), and multiplying by 100.

It will be apparent to those skilled in the art, with the benefit of the present disclosure, that simpler embodiments can be practised, in which some of the steps described above can be omitted, but doing so is likely to have an adverse effect on the accuracy or reliability of the assay result and/or its usefulness.

For example, it is possible to omit steps (bi) and (bii), but it is then not possible to compare with confidence the assay results obtained from an individual animal from two or more distinct time points, or to ensure that a true "low parasite burden" result has been obtained, as opposed to a result obtained where there is not enough antibody present to provide a reliable result (.e.g the saliva is too dilute).

Further, the foregoing refers to an "assay result" -The person skilled in the art will appreciate that the method of the invention will preferably involve the testing of replicate aliquots of a sample, and the "assay result" will normally be an average (typically a mean) of the results from various replicate aliquots.

The amount of parasite antigen-specific antibody in the saliva sample is typically very low. As a result, the saliva sample must be tested at low dilution in order for the assay to have sufficient sensitivity. Conversely, the saliva sample must be relatively io highly diluted when testing for the total amount of antibody or total selected IgG sub-class, in order to provide reliable results.

The differences in sample dilutions between a "total" Ig ELISA (assay hi), and the tapeworm antigen-specific assay, would appear to be a problem to a person of is ordinary skill in the art, since the "stickiness" factor of saliva may dilute differently from sample to sample, and the ratio of stickiness to analyte is likely to be different at different dilutions, making it impossible to predict the differential impact of sample NSB in the two tests. Surprisingly, the inventors have found that it is possible to extract meaningful calibrated values that correlate with tapeworm burden numbers with high statistical significance.

Preferably the method of the invention involves comparing at least the (a i) and (a ii) assay results with a calibration curve, so as to derive a corresponding calibrated assay value. This step may be performed by physically comparing the raw assay data with a drawn graph but, more conveniently, it is performed on a computer or by some other device comprising a microprocessor or the like and performing a suitable calibration algorithm, which is preferably programmed into the computer or microprocessor device.

By way of explanation, the inventors created a calibration standard by pooling antibodies from a variety of naturally infected horses, to provide a mixed antibody preparation comprising a range of affinities. This was used to create a calibration curve that allows the determination of meanin'ul values from direct ELISA OD readings. The infected horses were examined post-mortem and confirmed as having a tapeworm burden. Similar calibration reagents for other host/parasite systems could readily be prepared in analogous manner, given the benefit of the present invention.

Accordingly, performance of the method of the present invention in particular allows for the distinction between tapeworms being absent (or absent below a clinically-significant threshold, such that no treatment is necessary), and tapeworms being present at or above a significant threshold, at which anthelmintic drugs are indicated.

The step of obtaining a saliva sample may be performed by a veterinaiian, or by the in owner or other person supervi sing the health of a horse or other animal, Conveniently an absorbent swab is used to obtain the saliva sample. It is important that the swab is formed from materials which will not adsorb and retain immunoglobulins (especially IgG). Suitable swabs are commercially available (including those from Alere Toxicology).

To obtain sufficiently high antibody levels in saliva samples for testing, it is highly desirable that the subjects (typically horses) do not eat, drink, or exercise for 30 minutes before the sample is taken. The test will not give reliable resulls if the horse is producing too much saliva. An excessive saliva flow will result in the need for a retest, using a further sample of saliva, Typically the volume of saliva required to perform the method of the invention is about imI. The saliva swab is placed into 2 ml preservative solution, thereby providing a first dilution step, This volume is advmtageously divided into aliquots such that at least duplicates of the various assays can be performed, (e.g, at least two saliva control assays on each sample, and at least two antigen-specific assays performed on each sample) in order to increase reliability, The sample is further diluted in order for at least two total antibody assays to be performed. It is straightforward to run parallel duplicates of ELISAs in microtitre assay plates. Where duplicate or other multiple assays are performed, it will normally be desirable to derive a mean assay value, optionally disregarding obvious "outlier" or erroneous results, It will also be appreciated that the saliva sample may be subjected to some initial processing steps before being assayed. Conveniently, for example, debris or other solid particulate matter is removed from the sample by centrifugation and/or filtration.

The method preferably involves the performance of various control assays. For example, in some embodiments it will usually be necessary, or at least desirable, to include a "calibration control". For example, where the assay is an ELISA, a calibration control assay may be performed in which no saliva sample is used, for io both a "total" IgG subclass assay and for an antigen-specific IgO subclass assay.

In a diagnostic assay with two outcomes (e.g. infected or not infected), the predictive power of the test depends on its sensitivity and its specificity. The sensitivity (or "tme positive rate") is the proportion of actual positives (e.g. infected subjects) is correctly identified as such, The specificity (or "tme negative rate") is the proportion of negatives (e.g. uninfected subjects) correctly identified as such. Thus, the perfectly predictive diagnostic test would, in theory, be 100% sensitive and 100% specific. In practice, no diagnostic test achieves this. In simple terms, the accuracy of a diagnostic assay may be stated to be the mean of its sensitivity and specificity.

In the present invention, the method preferably has an accuracy of at least 70%, more preferably at least 75%, and most preferably at least 77%. In preferred embodiments, the diagnostic method of the invention has an accuracy of at least 78%, preferably at least 80% and most preferably at least 82%, Such levels of accuracy are surprisingly high, when compared to the accuracy achieved by other tests for parasite infections, including the long-established faecal egg counts for round worms, The invention will now be further described by way of illustrative example and with reference to the accompanying drawings, in which: Figure 1 is a schematic illustration of an algorithm that may be employed to check that performing the method of the invention has generated a meaningful result which can be reported to an end-user; Figures 2-5 are schematic illustrations of various salivary assay methods, some of which are in accordance with the present invention; and Figures 6-10 are case studies showing graphs of saliva antibody score against time for various horses, monitoring the level of tapeworms in the animals, using a diagnostic test in accordance with the invention.

Examples

Example I -Post mortem sample collection and examination A total of 139 horses of mixed type and age were examined at an abattoir in the UK during August and November 2013. The prior history of the horses was not available.

is At the point of slaughter, 30m1 blood samples were collected in SOml sterile tubes (Fisher Scientific UK, Leicestershire, UK). These were stored at room temperature and allowed to clot overnight.

Saliva samples were obtained from 104 of the t39 horses. The saliva samples were collected using three sterile polystyrene/viscose swabs (Fisher Scientific UK) per horse, After saliva collection the swabs were returned to the protective tube and immediately placed on ice. The ileocaecal junction, caecum and adjacent ileum were examined for the presence of tapeworms and, where present, the tapeworms were counted and recorded.

It was found that 58% of the horses were infected with tapeworms, which is in line with previous reports (Fogarty eta!., Vet, Rec. 1994 134, 5l5-58; Owen eal Vet, Rec. 1988 123, 562-563; Morgan eta!., Vet. Rec. 2005 156, 597-600; and Pittaway ci aL, Vet. Parasitol. 2014 199, 32-42).

Example 2-Detailed description of one embodiment of the Invention 2.1 Preparation of purified horse IgG calibrator reagent All reagents were equilibrated to 25°C before use. Horse serum, collected from horses with confirmed tapeworm burdens at post-mortem (>20 tapeworms present), as described above, was pooled arid 2 ml added to 4 ml acetate buffer (50 mM, pH 40).

During continuous mixing, caprylic acid was added slowly to a final concentration of 6% followed by mixing for 60 minutes at 25°C. The mixture was centriffiged at 4,000 x g for 15 minutes at 25°C to pellet the precipitated contaminating protein. The resulting supernatant, containing purified IgG, was then buffer exchanged to acetate buffer (0.]M, pH 5,6) using a 10 ml Zeba column (Thenrio Scientific) as described in io the manufacturer's instructions. IgG concentration was determined by absorbance at A280 and the purity assessed by SDS-PAGE (not shown).

2.2 Total IgG(T) ELISA (Total antibody assay, hi) The ELISA described below detects equine IgG(T), [an IgO subclassl, regardless of is the antigen-specificity of the antibody, Calibration solutions were prepared using serial dilutions of the purified lgG reagent in sample buffer (PBS, 0,05% Tween, 0.5% BSA, pH 7.4) at concentrations of 15 ng/ml, S ng/mI and ng/ml purified lgG, as well as a buffer control.

Each well in a 96-well flat-bottom high binding ELISA plate was sensitised by incubating so il of ag/mi goat anti-horse IgG(T) antibody (AbD Serotec) in PBS (pH 7.4) at 25°C for t h. Following this incubation, wells were washed 3 times with ial wash buffer (PBS, 0, 1?/b Tween, pH 7,4), Remaining binding surfaces were blocked by incubation with 50 jal blocking buffer (PBS, 0,1% Tween, 0.5% BSA, pH 7.4) for 1 h at 25°C. Wells were washed 3 times as described above, followed by addition of 50 tl samples (diluted 1/1000 or 1/2000 in sample buffer) and calibration standards in duplicate. The plate was again incubated for 1 h at 25°C followed by washing 3 times as described above, Goat anti-horse IgG(T), conjugated to horse radish peroxidase (AbD Serotec), was diluted 1/10,000 in blocking buffer, added to each well (50 al) and incubated for 1 hour at 25°C, Following a further washing step as described above, 50 j.tl stabilized 3, 3', 5, S"-tetramethylbenzidine (TMB Stabilized Chromogen, Invitrogen) was added to each well and incubated at 25°C for 10 mm.

The enzyme reaction was stopped by the addition of a stop solution (50 1 1 lvi HCl).

Finally, the absorbance was measured at a wavelength of 450 nm using an automated plate reader (FLTJOstar, BMG Labtech). The buffer control absorbance was subtracted from each of the absorbance readings. The slope of the standard curve was calculated (linear model) using the mean of each calibrator duplicate. As the ELISA specifically detects IgG(T), which is present as a proportion of the purified lgG calibration reagent used, the results are reported as "scores" rather than a definitive concentration.

in 2.3 Tapeworm-specific IgG(T) ELISA (Antigen specific assay, a i) The tapeworm-specific ELISA detects equine IgG(T) antibodies specific to the excretory/secretory 12/13 kDa tapeworm antigens, The 12/13 kDa antigens were prepared for use in the ELISA as described by Proudman & Trees, Parasite finmunol.

1996; 18, 499-506. The purified equine igO (described above) was used as a is calibration reagent in the ELISA. Calibration solutions were prepared using serial dilutions of the purified IgG reagent in sample buffer (PBS, 0.05% Tween, 0.5% BSA, pH 7,4) at concentrations of 0 ag/mI, S jig/mI and I jag/mI purified lgG, as well as a buffer control, zo Each well in a 96-well flat-bottom high binding ELISA plate was sensitised by incubating 50 p1 of 0.5 jig/mI 12/13 kDa tapeworm antigen in 0.05M bicarbonate buffer (pH 9.6) at 25°C for 1 h. Following this incubation, wells were washed 3 times with 200 p.1 wash buffer (PBS, 0.1% Tween, pH 7,4), Remaining binding surfaces were blocked by incubation with 50 p.1 blocking buffer (PBS, 0.1% Tween, 0.5?/b BSA, pH 7.4) for 1 h at 25°C. Wells were washed 3 times as described above, followed by addition of 50 p.1 samples (diluted 1/5 or 1/10 in sample buffer) or calibration standards in duplicate. The plate was again incubated for I h at 25°C followed by washing 3 times as described above. Goat anti-horse IgG(T), conjugated to horseradish peroxidase (AbD Serotec), was diluted 1/1000 in blocking buffer, o added to each well (50 p1) and incubated for 1 hour at 25°C. Following a further washing step as described above, 50 p1 stabilized 3, 3', 5, 5"-tetramethylbenzidine (TMB Stabilized Chromogen, Invitrogen) was added to each well and incubated at 25°C for 45 mm. The enzyme reaction was stopped by the addition of a stop solution (50 d I M MCI). Finally, the absorbance was measured at a wavelength of 450 nm using an automated plate reader (FLUOstar, BMG Labtech). The buffer control absorbance was subtracted from each of the calibrator and sample absorbance readings. The slope of the standard curve (linear model) was calculated using the mean of each calibrator duplicate. As the ELISA detects lgG(T) specific to tapewonri antigens, which is present only as a proportion of the purified IgG calibration reagent used, the results are reported as "scores" rather than a definitive concentration.

io The format of the assays is illustrated schematically in Figure 5 (described in greater detail below), together with example absorbance readings (OD 45O) for the ELISAs and the corresponding IgG(T) scores after comparison with the standard calibration curve.

is 2.4 Non-specific binding [LISA (Saliva control assay, a ii) Due to the potential for non-specific binding (NSB) with the saliva samples, NSB levels for each saliva sample were determined. NSB absorbances are converted into values using the calibration curve derived from the purified IgG(T) calibration reagents as described for the Tapeworm Specific IgG(T) ELISA, 21) Bicarbonate buffer (0,OSM, pH 9.6) was added (50 p1) to each well of a 96-well flat-bottom high binding ELISA plate and incubated at 25°C for 1 h. Following the incubation, wells were washed 3 times with 200 p.1 wash buffer (PBS, 0.1% Tween, p1-I 7,4), Binding surfaces were blocked by incubation with 50 p.1 blocking buffer (PBS, 0.1% Tween, 0.5% BSA, pH 7.4) for 1 h at 25°C. Wells were washed 3 times as described above, followed by addition of 50 p.1 samples (diluted 1/5 or 1/10 in sample buffer) and calibration standards in duplicate. The plate was again incubated for I h at 25°C followed by washing 3 times as described above. Goat anti-horse IgG(T), conjugated to horse radish peroxidase (AbD Serotec), was diluted 1/1000 in blocking buffer, added to each well (50 p.1) and incubated for 1 hour at 25°C.

Following a further washing step as described above, 50 p.1 stabilized 3, 3', 5, 5"-tetramethylbenzidine (TTVIB Stabilized Chromogen, Invitrogen) was added to each well and incubated at 25°C for 45 mm, The enzyme reaction was stopped by the addition of a stop solution (30 uI I Ni HCI), Finally, the absorbance was measured at a wavelength of 450 nm using an automated plate reader (FLUOstar, BMG Labtech).

The buffer control absorbance was subtracted from the calibrator absorbance readings only. The slope of the standard curve (linear model) was calculated using the mean of each calibrator duplicate. The results are reported as "scores" rather than as a concentration.

2.5 Calculation of tapeworm saliva scores io For each saliva sample, the total and specific scores were derived using the following calculations, in which: S = Antigen-specific test absorbance average (a i) R = Antigen-specific test calibration control absorbance average (a iii) C = Antigen-specific test calibration curve intercept (a i) IV1 = Antigen-specific test calibration curve slope (a i) N = Saliva control assay absorbance average (au) T = Total antibody test absorbance average (b i) U = Total antibody calibration control absorbance average (b ii) D = Total antibody calibration curve intercept (b i) L = Total antibody calibration curve slope (b i)

T-U-D

Total Score = L Specific Score = (S -R-C) -(N-C) The following calculation was used to determine the overall tapeworm saliva score, taking into account salivation rate at the time of sampling, by expressing the Specific Score as a percentage of Total IgG(T) in the sample Saliva Score = ((5 -(T-U (N_ C)) x 100 This can be simplified mathematically to Saliva Score = ((S -R-N) (T --100 The minimum total score required of a sample for the specific ELISA to be able to pick up the difference between a low diagnosis and a borderline or moderate/high diagnosis is 4,4, An algorithm is used to determine whether a result is correct for reporting a diagnosis to an end user. A suitable algorithm is illustrated in Figure 1.

Example 3

The preceding example relates to a detailed description of a preferred embodiment of the method of the invention, in which method there are performed, inter a/ia, in vitro assays of: (i) the total amount or concentration of (typically a selected class or subclass of) immunoglobulin in the saliva sample, regardless of the antigen-specificity of the antibody (assay bi); (ii) the amount or concentration of (typically a selected class or subclass of) immunoglobulin which is parasite antigen-specific (assay ai); and (iii) the amount of non-specific binding occurring as a result of the presence of saliva in the sample, (assay au).

In order to decide what type of assays needed to be included in the method in order to confer a desirable level of accuracy, the inventors performed a number of different versions of the method, and compared the sensitivity and specificity of the assay results and the associated Spearman's rank (which is a measure of the statistical dependence between the two: a Spearman's rank of 1.0 results when all data points with greater X' values than that of a given data point will also all have a greater Y' value as well. Accordingly, a Spearman's rank as close to 1.0 as possible is desirable in the present context).

The different versions of the method (Vi, V4, V7, V8) are described in Table I below, together with their associated accuracies and Spearman's rank. Vi is the preferred method.

Calculation Tapeworm cut off Spearmans version Calculations used Fl-cut 5+cut 1O+cut 20+cut rank ___________ __________________ off off off off __________ VI Statistics onsaliva Sensitivity 0.83 0.89 0,88 0.86 0,76 score -specific Specificity 0.80 0.77 0.80 0.79 calculaled by: TesI -calibration

background -saliva

control V4 Stalislics on specific Sensilivily 0.84 0.89 0,88 0.83 0,76 score -specific Spccificit 0.83 0.77 0,77 0.81 calculated by: Test -calibration

background -saliva

control V7 Statistics on specific Sensitivity 0.64 0,58 score-specific Specificity 0.78 calculated by: Test -calibration

background

VS Statistics on saliva Sensitivity 0.69 0,6 score -specific Specificity 0.74 calculated by: Test -calibraUon

background

The simplest version of the method is "Version 7", and is outside the scope of the present invention, Figure 2 shows, in schematic form, the assays performed in this to version. In this version, an antigen-specific ELISA (2) is performed (in which the ELISA wells are sensitised with the relevant tapeworm antigen preparation, shown as triangular symbols) in parallel with a calibration control ELISA (4), in which no saliva sample is added. The average OD value of the calibration control (4) is subtracted from the average OD value of the antigen-specific ELISA (2), which step is indicated by reference numeral (6). The resulting "corrected specific assay result" (in this instance, an OD) is shown at (8).

In addition, a specific calibration curve (10) was constructed from the results of parallel calibration curve control assays (12), in which the calibration reagent (purified horse serum Ig of known protein concentration) is tested in place of the saliva sample. The calibration curve (10) is illustrated as a graph of IgO concentration against absorbance (OD units). In practice this calibration curve may be formed "in silico". The corrected specific assay OD (8) is read against the calibration curve (10) to provide a corresponding IgG concentration. As noted elsewhere, the calibration reagent also contains irrelevant (i.e. non antigen-specific) 1g. Accordingly the test IgG concentration is referred to as a "score" (14) rather than a concentration.

Referring to Table, it is seen that this relatively crude version of the method of the invention gives a sensitivity of 64% (or 0.64) and a specificity of 78% (and thus an is accuracy, as herein defined, of 71%). These results are almost as good as some current serological tapeworm detection assays, but are not ideal.

A more sophisticated version, which is an embodiment of the invention, is Version 8', which is illustrated schematically in Figure 3. Some parts of this version are common to version 7, and like parts/steps are denoted by common reference numerals in Figure 3.

On the right hand side of Figure 3, tapeworm antigen-specific assays are conducted (steps 2-12), to provide an antigen-specific score (14), exactly as described previously in relation to version 7. To the left of the broken vertical line a series of "second analyte" (in this instance "total Ig", i.e. not antigen-specific) assays are performed.

These include a "total test" assay (6), in which the ELISA wells are sensitised with an anti-horse IgG(T) antibody (indicated by Y' shaped symbols). A parallel "Total calibration control" assay (18) is performed, in which no saliva sample is present.

Analgous to the antigen specific assay, the average OD reading from total calibration control is subtracted from the average OD reading of the total test assay, which step is indicated at (20), to give a "calibrated Total test" OD (22).

Essentially minoring the specific tests, a "Total calibration" curve (24) is prepared from the results of a Total Calibration assay (26), which curve is a graph of Ig concentration against absorbance (in OD units). The calibrated Total test OD reading (22) is plotted against the calibration curve (24) to derive a Total Ig concentration which, as explained elsewhere, is referred to as a Total Ig score (26). The overall assay score (28) is calculated by (specific Ig score Total Ig score) x 100.

Referring to Table I, it can be seen that version 8 of the method gives a sensitivity of 69% and a specificity of 74?/o, Thus, the overall accuracy of version 8(71,5%) is not io markedly higher than that of version 7, but it permits the discrimination of true negative' results from those due to use of saliva samples which are too dilute.

A relatively simple version of the method is that of version 4, which is illustrated schematically in Figure 4. As seen in Figure 4, version 4 is akin to version 7, in that is there are no "total Ig" assays performed. Steps common to versions 4 and 7 are denoted by common reference numerals in Figure 4. Version 4 is outside the scope of the invention, Relative to version 7, version 4 contains an additional "Saliva Control" assay (30). Tn this assay, the ELISA is performed using unsensitised assay wells, any assay signal (OD) arising primarily as a result of non-specific binding (predominantly due to the presence of the saliva sample). The saliva control assay (30) gives an average OD reading (32). This is converted to an equivalent saliva Ig score (34) by reading off the specific calibration curve. The overall specific assay score (36) is calculated by subtracting the saliva Ig score (34) from the specific Ig score (14). This step is denoted by reference (38), As seen in Table 1, version 4 provides significantly improved sensitivity, improved specificity, and thus significantly improved accuracy (82-83%) overall, relative to both versions 7 and 8, It is concluded that the inclusion of the "Saliva Control" assay, and making use of the results thereoL provide a significant advantage in terms of assay accuracy, Version 1 of the method represents the preferred embodiment of the invention. That version is illustrated schematically in Figure 5. Version 1 is similar to version 8 illustrated in Figure 3, and like parts/steps are denoted by common reference numerals. Version 1 also includes, however, the saliva control assay of version 4 (illustrated in Figure 4), arid like parts/steps common to that portion of version 4 are denoted by common reference numerals. In version 1 the specific assay score (36) is converted to a final saliva score (50) (numerical value = -0.1) according to the calculation: saliva score = (specific score + total score) x 100. As the specific score (36) has a negative value (due to the subtraction of the saliva control score (34) from io the specific calibrated score (14)), then however small the total score, the final saliva score (50) will be negative and, importantly, a sample from a horse with no tapeworms will remain below the threshold for a "low burden" diagnosis.

Version 1 is the preferred embodiment of the invention since, although it does not offer any significant improvement in assay accuracy relative to, say, the much simpler version 4 (see Table 1), it is more reliable since it includes a "total" antibody assay which can detect if the levels of antibody in the sample are too low for the assay to provide meaningful results (i.e. the saliva sample is too dilute), and can be used to compare assay results from individual animals over time, thus enabling monitoring of the parasite burden in a particular animal and allowing an assessment of the efficacy of anthelmintic treatment.

Example 4

An embodiment of the diagnostic method of the present invention (referred to herein as "EquiSal" i.e. Version 1 as described above) was compared with a prior art serological ELISA-based technique.

Validation was carried out by reference to the ideal gofti standard of diagnosis which, in this case, is the counting of actual tapeworms present in the gut. The strength of the relationship between tapeworm burden and ELISA scores from the serological ELISA and the EquiSal tapeworm Test was evaluated using Spearman's rank correlation coefficient, Both test results were also directly compared with each other using Spearman's rank correlation coefficient. Tapeworm burdens were classified into three groups, low (no tapeworm), moderate (1 to 19 tapeworms) and high (20+ tapeworms). Data for the average tapeworm score (median) and scatter, based on the interquartile range, were assessed for each group. The accuracy of the ADB IgG(T) Serological ELISA and the EquiSal® Tapeworm Test to diagnose tapeworm burden s levels was assessed using the Mann-Whitney test and Receiver-operator characteristic (ROC) curves. The Mann-Whitney tests were carried out using Prism (GraphPad Software, San Diego, CA), and MedCalc software (Medcalc, Ostend, Belgium) was used for calculation of ROC curves, as well as sensitivity and specificity calculations.

The Mann-Whitney test is a non-parametric test to determine whether two populations io are significantly different. The lower the resulting p value, the greater the significant difference between the groups. ROC curves are often used to determine the threshold laboratory value that separates a clinical diagnosis of "normal" from one of "abnormal". In this case, the thresholds attributed to 1+ tapeworms and 20+ tapeworms provide cut-offs for low, moderate and high burdens. The ROC curve is displays the relationship between the true positive rates (sensitivity) and the false positive rates (I 00-specificity) over a range of possible cut-off values. The area under the ROC curve (AUC) quantifies the overall ability of the test to discriminate between tapeworm burden levels. A perfect test, that has no false positives or false negatives, hasanareaofl.O,whereasapoortestwillhaveanareaofO.5 Thepvalueteststhe zo null hypothesis that the area under the curve equals 0.5, so if the p value is small, it can be concluded that the test does discriminate between groups.

Statistical analysis of both the EquiSal® Tapeworm Test and the ADB IgG(T) Serological ELISA revealed extremely significant p values (<0.0001) in Mann-Whitney tests and statistically significant ROC AUC values for both 1+ and 20+ tapeworm cut-offs (p = <0.001), as well as high sensitivity and specificity values. For detection of 1+ tapeworm burdens, an EquiSalt tapeworm score cut-off of -0.09 provided optimum sensitivity and specificity (83% and 85%, respectively), whereas an ADB IgG(1) Serological ELISA score cut-off of 2.7 provided optimum sensitivity and specificity (85% and 78%, respectively). Both the EquiSal® Tapeworm Test and the ADB IgG(T) Serological ELISA had strong positive correlations to tapeworm number (Spearman's rank, 0.74 and 0.78, respectively). Additionally, there was a strong positive correlation between the two tests (Spearman's rank, 0.87), demonstrating that tapeworm-specific antibodies levels present in saliva correlate well to the level of tapeworm specific IgG(T) present in blood. These data provide definitive statistical evidence that both the EquiSal Tapeworm Test and the ADB IgG(T) Serological ELISA are able to reliably diagnose low burdens from moderate/high burdens (1+ tapeworm cut-off) and that the two tests have strong positive correlation with each other.

Thus, the method of the present invention provides a diagnostic test that is essentially at least as reliable as the existing serological test, but avoids the need for obtaining a io blood sample, which requires an invasive technique and a trained veterinarian to perform the initial sampling.

Example 5 Case Studies is Trial details Saliva was collected from horses by means of EquiSal® saliva swabs and placed into preservative solution. The saliva was tested to determine the EquiSal® tapeworm saliva score and tapeworm burden diagnosis. The frequency of testing was every 2 weeks for the majority of horses, Horses with an established low burden status were subsequently tested every month. Horses were monitored for various time periods between 2 to 14 months.

Data are presented here as case studies, for selected horses of interest, The results are shown in Figures 6-10, which are plots of EquiSal® saliva score against time.

With reference to Figure 6, there are shown results for horse B', which are typical for those in horses kept in a well-managed paddock (e.g. with removal of horse faecal matter several times a week), and the surrounding paddocks were also well-managed.

The horse initially had a borderline tapeworm burden diagnosis. The horse was given a double dose of pyrantel embonate, an anthelmintic drug (indicated by the downward arrow), and the saliva score reduced to a low' diagnosis (i,e, saliva score of -0,09 or less) within 3-4 weeks, and stayed at that level over the course of the monitored time period.

Figure 7 shows the results for horse S', which was kept in a well-managed paddock, but which was adjacent to a poorly managed paddock.

Horse S initially had a borderline burden diagnosis. Within 2 weeks of worming with an anthelmintic drug (praziquantel) the saliva score reduced to a low diagnosis.

The saliva scores remained in the low diagnosis range for approximately 3,5 months, at which point the saliva score increased to moderate/high burden, clearly indicating io that the horse had been re-infected. Following worming with praziquantel, the saliva score dropped to a low diagnosis within 3 weeks.

Horse S lives in a well-managed paddock, however, it is adjacent to a very poorly managed paddock containing horses that are never wormed. It is thought that close is proximity to the poorly maintained paddock could be the reason for re-infection following a period of low burden status, especially as oribatid mites (the intermediate host) are known to travel between paddocks.

Figure 8 shows the results for horse P. Horse P lives in the same paddock as horse S. Horse P initially had a moderate/high burden diagnosis. Within 2 weeks of worming (with praziquantel), the saliva score reduced to a low diagnosis. The saliva scores remained mainly in the low diagnosis range for approx. 3 months at which point the saliva score increased to moderate/high burden and continued to increase.

Horse P was wormed with praziquantel 6 weeks after re-infection was diagnosed.

However, Horse P spat out a large amount of the dose. The saliva score continued to increase following this incorrect worming dose so it was decided to treat again.

Following worming with the correct dose of praziquantel, the saliva score strikingly dropped to a low diagnosis within 2 weeks. As mentioned above, Horse P lives with Horse S, and both horses' diagnosis changed to moderate/high at a similar time.

Figure 9 shows the results for Horse R. Horse R initially had a moderate/high burden diagnosis. Following worming (with a double dose of pyrantel), the saliva score reduced to a low diagnosis within 3 months.

This is the longest timescale we have observed for a reduction in tapeworm-specific antibodies following worming. However, the reduction was sequential, therefore indicating that pyrantel had been effective prior to the saliva score reaching low diagnosis.

Horse R lives in a well-managed paddock and the surrounding paddocks are also well managed, so re-infection is an unlikely explanation for the extended persistence of the tapeworms. It is more likely that the previous tapeworm infection history of Horse R had resulted in a primed immune system, such that a more vigorous immune response was mounted against this particular episode of parasite infection, involving more persistent B4ymphocyte (plasma cells) colonisation of the salivary gland. It is is recognised in the art that plasma cells can be longer-lived or short-lived, depending on the immune induction pathway followed, including the involvement or not of germinal centre reactions, and immunological priming through pervious responses to the same antigen.

Figure 10 shows the result for horse Bo, Horse Bo lived in poorly managed paddock. Removal of faecal matter was carried out on a r2mdom basis but the field was never completely cleared. Additionally, the muck heap was situated within the grazing area.

Horse Bo lived with two other horses, and all three animals had a moderate/high burden diagnosis when initially tested.

Within 3 weeks of worming (with praziquantel), the saliva score reduced, but not to a low diagnosis. The saliva score subsequently increased over 5 weeks at which point Horse Bo was again wormed with praziquantel, The saliva score dropped substantially within 4 weeks but, again, not to a low diagnosis. The saliva score subsequently increased over 2 months, after which point the investigators no longer had access to the horse.

It is clear, from the results obtained for Horse Bo, that monitoring horses after treatment for tapeworm is complicated through re-infection by tapewonn larvae. In a poorly managed paddock, reinfection is perhaps an obvious outcome. However, given that the tapeworm's life cycle requires an intermediate host (an oribatid mite), even well managed paddocks containing horses with high tapeworm burdens will contain a high level of infected oribatid mites resulting in a high risk of reinfection of horses grazing there.

Conclusions

The data presented here indicate that salivary antibodies are reliable biomarkers of tapeworm infection and unlikely to produce false positives due to persistent secretion is after resolution of infection within the current recommended testing schedules.