WO1995016787A1 - Diagnostic tests and reagents for detecting risk of alzheimer's disease and stroke - Google Patents

Abstract

The present invention makes available diagnostic assays and reagents for facilitating accurate diagnosis and subsequent monitoring of Alzheimer's disease and its progression, as well as determination of stroke occurrence in a patient. As described herein, the percentage T-lymphocytes of a unique T-lymphocyte phenotype (IgM+ T cells) is statistically correlated with Alzheimer's disease as well as the recent occurrence of a stroke.

Description

Diagnostic Tests and Reagents for Detecting Risk of Alzheimer's Disease and Stroke

Background of the Invention

Alzheimer's disease (AD) is a neurodegenerative disorder that affects a significant percentage of elderly individuals and is characterized by a progressive dementia. The classical neuropathological features of Alzheimer's disease include neuritic plaques, composed of neurites and extracellular amyloid deposits, and intracellular neurofibrallary tangles of hippocampus and cerebral cortex. Alterations in transmitter-specific markers have been described.

In assessing suspected AD patients, a medical history is typically taken from the patient and from an informant who is well acquainted with the affected individual. This approach is used to establish a history of progressive deterioration and for identifying tasks that the patient can no longer perform adequately. For example, a diary maintained by an observer is often used to document changes in various functions. Such histories can disclose abnormalities, including impaired memory and other cognitive functions, impaired activities of daily living, alterations in mood, often delusions and illusions, and sometimes hallucinations. For example, common complaints of patients or families include forgetfulness about appointments or errands; inability to find the way to an accustomed destination; inability to use money and instruments of daily living such as a telephone; deterioration in work or homemaking performance; difficulty adapting to changes in the workplace; difficulties in dressing, reading, and writing; and inability to recognize previously familiar individuals. The typical clinical examination provides data to fulfill inclusionary and exclusionary criteria for the diagnosis of Alzheimer's disease and to document symptoms such as delusions or depression that identify subgroups of patients important both for research studies and for patient care. Mental status testing, an essential component of the clinical examination, includes specific assessment of orientation, registration, attention, calculation, recent recall, naming, repeating, understanding, reading, writing, and ability to draw or copy.

Quantitative aids to the clinical examination include the Mini Mental State Examination (MMSE) (Folstein et al. (1975) J Psychiatric Res 12:189-198) for cognitive screening; the Blessed Dementia Scale (Blessed et al. (1968) Br J Psychiatry 114:797-811) for clinical symptoms and social function; the Hamilton Depression Scale (Hamilton et al. (1967) British J Soc Clin. Psychol 6:278-298) for severity of depression; the Present State Examination (Wing et al. The measurement and Classification of Psychiatric Symptoms. Cambridge: Cambridge University Press, 1974) for anxiety, depression, delusions, and hallucinations; and the Hachinski Scales S for estimating the likelihood of multi-infarct dementia. A complete psychiatric evaluation has generally been needed to exclude various other psychiatric disorders.

Complete examination of sensory and motor systems (including cranial nerves, tone, reflexes, coordination, gait, and proprioception) is often performed to exclude other neurologic disorders. In early stages, patients are alert and free of other neurologic changes related to the dementia except for the occasional presence of snout reflex, jaw jerk, rigidity, or myoclonus, all of which may also be encountered in nondemented elderly people. As the disease progresses, some patients become apathetic or show irritability, agitation, paranoid ideas, sleep disorders, or incontinence. In the very advanced stages, patients may become mute and lose all ability to communicate.

Neuropsychological tests are also used to provide additional information for the diagnosis of dementia. Because there are no normative population standards for many of these tests, abnormal performance can be determined only by comparison with a normal control group matched for age, sex, and local education. A score falling in the lowest fifth percentile of an individual's normal control group may be designated as "abnormal." One or more abnormal scores will identify an individual for research purposes who is highly likely to be cognitively impaired. Progressive worsening can be established by comparison with the patient's previous performance on these tests. Similar series of tests are used to assess less severely affected patients by increasing the complexity of the neuropsychological tests.

In longitudinal assessment, many patients with Alzheimer's disease show progressive loss of recent memory followed by disorders of language, praxis, or visual perception. In some patients with Alzheimer's disease, however, the first symptoms are difficulty in finding words, impaired visual perception, or apraxia, with memory impairment and other symptoms and signs appearing later.

However, despite the widespread use of such clinical examinations and neuropsychological testing, the clinical criteria for the diagnosis of Alzheimer's disease remain inadequate. Indeed, the need to refine clinical diagnostic criteria has been emphasized by a significant number of cases wherein patients clinically diagnosed with Alzheimer's disease are found, at autopsy, to have other neurogenic disorders and not Alzheimer's disease. Moreover, therapeutic trials can be meaningfully compared only if uniform and reliable criteria are used for diagnosis and response to treatment. Therefore, there presently exists a need for specific diagnostic laboratory tests for Alzheimer's disease, especially tests that can enhance the diagnostic accuracy.

Moreover, certain groups of individuals, particularly geriatric groups, are more susceptible to apoplexy, such as thromboembolic-mediated stroke, due to factors including age, medical conditions, and medications being taken. In some instances, it may difficult to confirm whether or not a patient has actually suffered from a stroke based on typical clinical evaluation, e.g. from outward appearances and motor coordination. There is thus a need to provide a diagnostic test to determine or confirm whether a patient has suffered some form of a stroke.

It is therefore an object of this invention to provide a diagnostic assay for Alzheimer's disease which is useful for enhancing the accuracy of clinical evaluation of patients and which can be used to provide uniform and reliable criteria for diagnosis of Alzheimer's disease.

It is another object of this invention to provide a diagnostic assay for detecting the occurrence of stroke in a patient so as to enhance the accuracy of clinical evaluation of stroke patients so as, for example, to facilitate prompt treatment of the condition, or to improve monitoring drug regimens for patients at risk of stroke.

Summary of the Invention

The present invention makes available diagnostic assays and reagents for facilitating accurate diagnosis and subsequent monitoring of Alzheimer's disease and its progression.

As described herein, the identification of a unique T-lymphocyte phenotype (IgM positive T cells) associated with Alzheimer's disease can be exploited to overcome the problems previously associated with diagnosis of Alzheimer's disease.

Accordingly, one embodiment of the present invention makes concerns a method for identifying an individual's risk of Alzheimer's disease. The method generally comprises determining the percentage of at least one of either IgM-positive (IgM+) or IgM binding protein positive (IgM-bp+) T-lymphocytes in a bodily fluid, and comparing the measured percentage of the IgM+ or IgM-bp+ T-lymphocytes with percentages of IgM+ T lymphocytes or IgM-bp+ T-lymphocytes measured in a normal population of individuals. A statistically significant increase (e.g. at least 25% increase, more preferably at least 50% increase, and most preferably at least a 75% increase) in the measured percentage of IgM+ T-lymphocytes or IgM-bp+ T-lymphocytes relative to levels in the normal population is indicative that the individual has Alzheimer's disease or has an increased risk of developing Alzheimer's disease. The subject method can be used to refine clinical criteria for diagnosis of Alzheimer's disease, and is especially useful for enhancing diagnostic accuracy in clinical evaluation of patients. The subject method can be used to provide uniform and reliable criteria for diagnosis of Alzheimer's disease as well as monitoring response to treatment.

The present invention further makes available diagnostic assays and reagents for facilitating accurate diagnosis and subsequent monitoring of apoplexy, e.g. stroke, e.g. stroke having thrombolic, embolic and/or occlusive cerebravascular lesion (i.e., hemorrhaging) as an etiological causative agent. In preferred embodiment, the diagnostic assay is employed to detect recent cerebral circulatory disturbances, e.g. recent stroke. As described herein, the occurrence of recent apoplectic events is statistically negatively correlated with the percentage of IgM+ T-lymphocytes in biological fluids of a patient.

Thus, another embodiment of the present invention concerns a method for augmenting a diagnosis of stroke, which generally comprises determining the percentage of IgM-positive (IgM+) T-lymphocytes in a sample of bodily fluid, and comparing the percentage of IgM+ T- lymphocytes determined in the sample with a percentage of IgM+ T lymphocytes from a standardized data set (e.g. non-stroke patients). A statistically significant decrease in the percentage of IgM+ T-lymphocytes in the sample relative to the percentage in the a non- stroke population is indicative that the individual has suffered a stroke. In preferred embodiments, a statistically significant decrease in IgM+ T-lymphocytes indicative of recent stroke is at least a 10-20 percent decrease in the percentage of IgM+ T-lymphocytes in the patient sample relative to the percentage of IgM+ T-lymphocytes in the a non-stroke population, and more preferably at least a 40-50 percent decrease, and even more preferably at least a 75-80 percent decrease.

Yet another embodiment of the present invention provides a diagnostic test kit for identifying at least one of (i) an individual's risk of Alzheimer's disease, or (ii) the occurrence of a recent apoplectic events suffered by the individual. The diagnostic test kit of the present invention comprises a first antibody for detecting cells displaying a T-lymphocyte marker, and a second antibody for measuring the percentage of cells in the sample displaying one of either an IgM class antibody or a 96,000 daltons T-lymphocyte IgM binding protein. Used together, the two antibodies can be employed to determine the percentage of IgM+ or IgM- bp+ cells characterized by the T-lymphocyte marker. For example, the first antibody can be directed to the general T-cell population, or, alternatively, to a particular T-cell subpopulation, such as CD4+ or CD8+ T-cells.

Brief Description of the Drawings

Figure 1 compares cognitive diagnoses of study patients with the percentage of IgM+ T-cells measured in each patient. The diagnoses were divided into 11 different categories. Categories 1-6 represented demented patients with the following diagnoses: 1) Probable AD (n=23); 2) Possible AD (n=l); 3) Multi-infarct dementia (MID) (n=6); 4) Mixed AD/MID (n=l); 5) Demented, diagnosis unclear (n=l); 6) Neurological illness associated with dementia (n=l; this patient had Parkinson's Disease). Categories 7-11 represented patients without dementia, with the following diagnoses: 7) Age-associated memory impairment (n=2); 8) Recent stroke with focal neurological and focal cognitive deficits (n=4); 9) Recent stroke with focal neurological deficits only (n=l); 10) Not demented, diagnosis unclear (n=l); 11) Normal older adults (n=12).

Figure 2 is a scatter plot comparing the percentage of IgM+ T-cells (y-axis, IgM+ T cell/total T-cell) with donor age (x-axis).

Figure 3 illustrates the association of the percentage of IgM+ T-cells to scores received on the Mini Mental Status Exam (MMSE). The bar graph indicates the scores received on the MMSE in descending order (x-axis) versus the percentage of IgM+ T-cells (y axis; IgM+ T cells/total T cells).

Figures 4A and 4B illustrate the development of a stepwise regression model which demonstrates that of age, Alzheimer's Disease, use of gastrointestinal medication, recent stroke (e.g. approx. 9-72 days) and remote stroke (e.g. 7 months to 23 years), only AD and recent stroke are regarded as being independently associated with the percentage of IgM+ T lymphocytes.

Detailed Description of the Invention

Alzheimer's disease (AD) affects approximately four million Americans, and is the fourth leading cause of death among persons older than 65. AD is age related with approximately 5% of adults affected by age 65, and 25% to 40% by age 80 to 85. Since the number of persons 65 years of age and older in the U.S. is increasing rapidly and is expected to represent 24% of the population by 2030, as compared to 12% now, the number of persons with AD is expected to rise dramatically. Presently, the criteria for the diagnosis of Alzheimer's disease are (a) the presence of dementia, (b) insidious onset with a generally slowly progressing and deteriorating course (often determined by interviewing family members), and (c) exclusion of all other specific causes of dementia by history, physical examination or laboratory test. Thus, a diagnosis of Alzheimer's disease has typically weighed heavily on subjective reporting by family members, and on the absence of evidence for other causes of disease.

One aspect of the present invention concerns diagnostic assays and reagents for facilitating accurate diagnosis and subsequent monitoring of Alzheimer's disease and its progression. As described herein, the identification of a unique T-lymphocyte phenotype (e.g. IgM+) associated with Alzheimer's disease can be exploited in clinical test kits. The test kits can be, for example, in the form of an immunoassay.

In addition to the pervasiveness of Alzheimer's Disease in the elderly population, apoplexy (e.g., stroke) is a very important and frequently occurring disease in gerontology and geriatric practice and clinics. Apoplexy is the clinical term for sudden loss of consciousness, followed by paralysis resulting from a cerebral hemorrhage, occlusion of a cerebral artery, thrombosis, or embolism, with loss of cerebral function of the affected area. In world-wide mortality statistics, cerebral circulatory disturbances are the third most frequent cause of death. In particular, stroke resulting from thrombosis occur more frequently with increasing age. In every case of stroke, whether of ischemic origin or caused by hemorrhage, a complete diagnostic program is essential to clarify etiology and pathogenesis in order to establish an adequate therapeutic strategy and prognostic assessment. For instance, remission mainly takes place (more than 80 percent) within the first three months post-stroke. Remission after six months is very rare. Thus, early diagnosis can be critical for therapeutic intervention. Moreover, the present diagnostic assay can be used to improve assessment of those individuals who do not outwardly present conclusive physical signs of stroke, but who may have suffered "petite" stroke(s). Patients undergoing therapeutic regimens which increase the risk of stroke occurrences may be particularly susceptible to such apoplectic events. Routine monitoring of such patients in the course of a particular drug therapy, using the subject assay, can facilitate early detection of slight apoplectic conditions and thus permit timely withdrawal or reduction of the dosage of the drug administered to such stroke-prone patients. Furthermore, the present invention can be used to distinguish between recent stroke and presence of a brain tumor, both of which can display overlapping symptoms. Accordingly, another aspect of the present invention concerns diagnostic assays and reagents for facilitating accurate diagnosis and subsequent monitoring of apoplectic events and their progression, particularly recent cerebral circulatory disturbances, e.g. recent stroke, e.g. occlusive cerebravascular lesion within the previous 3 months. As described herein, the occurrence of recent apoplectic events is statistically negatively correlated with the percentage of IgM+ T-lymphocytes in biological fluids of a patient.

A T-lymphocyte phenotype is described herein which results from the binding of autochthonous IgM to the surface of peripheral T lymphocytes (hereinafter "IgM+ T- lymphocytes"). The percentage of T-lymphocytes with this phenotype is shown here to be statistically correlated with performance in mental status testing that is considered indicative of a dementing illness. It has been discovered that there is a diagnostically significant difference between the mean percentage of IgM+ T-lymphocytes in AD patients and the mean percentage of IgM+ T-lymphocytes in non-AD patients (p < 0.001). In addition, it has also been discovered that there is a diagnostically significant correlation between the mean percentage of IgM+ T-lymphocytes in patients having recently experienced some apoplectic event, e.g. recent stroke. However, there is no statistical significance between the mean percentage of IgM+ T-lymphocytes and other factors, including age, sex, race, absolute lymphocyte count, medications (except gastrointestinal (GI) medications), or other medical diagnoses.

For example, when a threshold value of 20-percent IgM+ T-lymphocytes (relative to the total population of T-lymphocytes) is used to predict the diagnosis of AD, the sensitivity of the present assay can be greater than 62 percent and the specificity in the range of 90%. The present invention furnishes a significant advantage in ability to discern between individuals suffering from AD and those demented from other causes, especially when used in conjunction with clinical examination, such as MMSE data.

Likewise, the clinical diagnosis of recent stroke (e.g. within the previous three month period of time) was shown to be independently associated with depressed levels of IgM+ T lymphocytes relative to non-recent stroke patients. Compared to standardized data from non- stroke patients, a statistically significant decrease in the IgM+ phenotype is indicative of a recent stroke. Such testing can be particularly effective for diagnosis of patients presenting minor symptoms of stroke, as well as in discerning potential causes of multi-infarct generated dementia. o

The present invention provides a method and reagents for generating specific diagnostic laboratory tests for Alzheimer's disease and apoplexy. The subject method can be used to refine clinical criteria for diagnosis of either condition, and is especially useful for enhancing diagnostic accuracy in clinical evaluation of patients. As described below, the subject method can be used to provide uniform and reliable criteria for diagnosis of Alzheimer's disease as well as monitoring response to treatment. Additionally, the ability to detect and augment diagnosis of recent apoplectic events, such as arising from thromboembloic complications or cerebral hemorrhaging, can permit early intervention and can provide, for example, significant benefits when used to monitor drug dosage regimens in stroke-prone patients when such pharmacophores as anti-hypertensive agents are used, e.g. cardioselective b-adrenoreceptor antagonists.

According to one embodiment, the present invention makes available a method for identifying an individual's risk of Alzheimer's disease. The method generally comprises determining the percentage of at least one of either IgM-positive (IgM+) or IgM binding protein positive (IgM-bp+) T-lymphocytes in a bodily fluid, and comparing the measured percentage of the IgM+ or IgM-bp+ T-lymphocytes with percentages of IgM+ T lymphocytes or IgM-bp+ T-lymphocytes measured in a normal population of individuals. An increase in the measured percentage of IgM+ T-lymphocytes or IgM-bp+ T-lymphocytes relative to levels in the normal population is indicative that the individual has Alzheimer's disease or has an increased risk of developing Alzheimer's disease.

In similar fashion, another embodiment of the present invention makes available a method for augmenting a diagnosis of stroke, which generally comprises determining the percentage of IgM-positive (IgM+) T-lymphocytes in a sample of bodily fluid, and comparing the percentage of IgM+ T-lymphocytes determined in the sample with a percentage of IgM+ T lymphocytes from a standardized data set (e.g. non-recent stroke patients). A statistically significant decrease in the percentage of IgM+ T-lymphocytes in the sample relative to the percentage in the a non-recent stroke population is indicative that the individual has suffered a stroke. Thus, patients which are believed to have suffered a stroke a some point in the past can be identified through the present invention as having suffered either a recent stroke or a remote stroke.

In another embodiment of the present invention, a diagnostic test kit for identifying an individual's of risk of Alzheimer's disease or of apoplexy is provided. The diagnostic test kit of the present invention comprises a first antibody for detecting cells displaying a T- lymphocyte marker, and a second antibody for measuring the percentage of cells in the sample displaying one of either an IgM class antibody or a 96,000 daltons T-lymphocyte IgM y

binding protein. Used together, the two antibodies can be employed to determine the percentage of IgM+ or IgM-bp+ cells characterized by the T-lymphocyte marker. For example, the first antibody can be directed to the general T-cell population, or, alternatively, to a particular T-cell subpopulation, such as CD4+ or CD8+ T-cells.

In particular embodiments of the invention, diagnosis of Alzheimer's disease, or conversely of apoplexy, can be augmented by measuring in a patient sample, preferably a bodily fluid containing peripheral blood mononuclear lymphocytes (PBMLs), the percentage of T-lymphocytes which display IgM bound to their surface. In other embodiments, the percentage of T-lymphocytes scoring positive for an IgM-binding protein (IgM-bp), as described below, can be measured and can be used to assist in the diagnosis of either condition. A diagnosis or risk assessment of such disorders can be made by comparing the measured levels of IgM+ and/or IgM-bp+ T-lymphocytes with levels measured for a standardized population.

The term "specific binding protein" as used herein refers to any substance, or class of substances, which has a specific binding affinity for a T-lymphocyte population as a whole (e.g. T-lymphocyte specific), or either the IgM+ or the IgM-bp+ sub-populations (e.g. AD- specific). In the majority of embodiments, the present invention will incorporate specific binding assay reagents which will interact with a patient sample, particularly certain T cell populations, in an immunochemical manner. That is, there will be an antigen-antibody relationship between the reagents and antigens associated with the cells in the T-lymphocyte population of interest. These assays are termed immunoassays and the interaction between the reagent and T-lymphocyte antigen is an immunochemical binding. The use of either polyclonal or monoclonal antibodies is contemplated unless otherwise indicated. Additionally, it is well understood in the art that other binding interactions, such as receptor- ligand binding between T-lymphocyte surface components and reagents useful in the present assay can be used to quantitate the levels of IgM+ or IgM-bp T-lymphocyte populations.

The T-lymphocyte specific binding protein can be an antibody which is selective for a general T-lymphocyte surface marker. Such antibodies are useful in establishing the overall population of T-lymphocytes from which the percentage of IgM+ or IgM-bp+ T-lymphocytes is calculated. The T-lymphocyte specific binding protein can alternatively be chosen to be more selective, and can have a designated specificity for a sub-population of T-cells, such as suppressor T-lymphocytes, helper T-lymphocytes, and the like. For example, several T- lymphocyte specific antibodies or ligands for particular receptors expressed on T- lymphocytes are known to be useful for detecting total T-lymphocyte populations, or sub- populations of T-lymphocytes, and are readily available commercially. The T-lymphocyte specific binding protein can be a monoclonal or polyclonal antibody specific for a T- lymphocyte marker such as CD2, CD3, CD4, CD5, CD7, CD8, CD28, or the T-lymphocyte receptor. Both monoclonal and polyclonal antibodies are generally available for each of these exemplary T-cell specific markers. For example, the 1993 Sigma Chemical Company Catalog lists antibodies to CD3 (Catalog No. C7048, and F0522 (FITC labeled), CD4 (Catalog No. C1805, and B7280 (FITC labeled), CD5 (Catalog No. C7173, and F8893 (FITC labeled), CD7 (Catalog No. C7298, and F0647 (FITC labeled), and CD8 (Catalog No. C7423, and F0772 (FITC labeled). Other T-cell marker antibodies can be obtained from the ATCC collection, such as ATCC CRL 8000, ATCC CRL 8001, ATCC CRL 8016, and ATCC HB2. In addition, antibodies for detecting IgM bound to the isolated T cells of the sample are generally available through commercial sources. For instance, the IgM-detecting antibody employed in the subject method can be specific for the m heavy chain constant region (Sigma Catalog No. A2189, F5384 (FITC labeled), P9295 (R-phycoerythrin labeled), T3095 (TRITC labeled), or, alternatively, for one of the light chain constant regions (k chain specific antibodies, Sigma Catalog No. F3761 (FITC), T7279 (TRITC); or 1 chain specific antibodies, Sigma Catalog No. F6774 (FITC), T9153 (TRITC)).

In various embodiments of the present assays, antibodies can be used in immunoassays to detect and quantitate IgM+ T-lymphocyte populations. Such antibodies are specifically contemplated to include functional binding fragments thereof containing the binding region (e.g. scFv, Fv, dab, Fab, F(ab')2). The immunoassay systems include, but are not limited to, fluorescence-activated cell sorting and other fluorescent immunoassays, radioimmunoassays, ELISA (Enzyme-linked immunosorbent assay), agglutination assays, complement fixation assays, and immunoradiometric assays to name but a few. In an illustrative embodiment, diagnosis can be facilitated by detecting the immunospecific binding of an antibody, or derivative or fragment thereof, directed against an epitope of one of either IgM or the IgM-bp, in a population of T-lymphocytes.

In a preferred embodiment, fluorescence-activated cell sorting (FACS) is used to determine the percentage of IgM+ or IgM-bp+ T-lymphocytes. For example, two fluorescently-labeled antibodies can be used in detection; one antibody directed to either IgM or IgM-bp, and one T-lymphocyte specific antibody for a more widely occurring T- lymphocyte marker. Each of the AD-specific and T-lymphocyte specific antibodies are labeled such that discrimination in detection of each is possible. Thus, cells having two different fluorescent labels are distinguished from cells having only the T-lymphocyte specific fluorescent label and unlabeled cells using FACS techniques. The proportion of IgM+ or IgM-bp+ T-lymphocytes can be calculated based on the fraction of doubly-labeled cells relative to the total population of cells labeled with the T-lymphocyte specific antibody. In each instance, the antibody is labeled with a dye, such as a fluorochrome, that facilitates FACS. Examples of suitable dyes for FACS analysis and/or separation are well known. Those dyes are described, for example, in Practical Flow Cytometry, 2nd Ed, by Howard M. Shapiro and Alan R. Liss (1988), at pages 115-198. Preferred dyes are fluorochromes including fluorescein (e.g., fluorescein isothiocyanate - FITC), rhodamine (e.g., tetramethylrhodamine isothiocyanate -TRITC), phycoerythrin (PE), allophycocyanin (APC) and Texas Red (Molecular Probes, Eugene OR). The combinations of fluorochromes used for labeling are chosen so that distinguishable wavelengths of light are emitted. A preferred combination is a fluorochrome that emits green light together with one that emits red or orange light, e.g., FITC with PE or Texas red.

The antibodies useful in the present assay can be labeled with the fluorochrome, directly or indirectly, by well known methods. The conjugation methods for attaching labels to antibodies generally can be used for these purposes. Direct labeling methods for dyes such as FITC are well known. For dyes such as PE which are more difficult to attach while maintaining antibody activity, the antibody can be labeled with biotin. The dye can be attached by incubation of the biotinylated antibody with PE-avidin or PE-strepavidin (which are commercially available) either concurrently with or, preferably, following incubation of the antibody with the cells.

To prepare the cells with the labeled T-lymphocyte specific and AD-specific antibodies, the cells are incubated with the antibodies for a time sufficient for substantially complete antibody binding. An excess amount of antibody is preferably used. In a preferred embodiment, the cells are incubated at about 4°C. for about 30 minutes. These conditions are usually sufficient for substantially complete antibody binding while maintaining cell integrity. Preferably, the incubation is performed in the dark when using a fluorochrome label. The sample is preferably mixed, during the incubation to ensure contact of the antibodies with the cells. Mixing can be effected using a hematology blood rocking device. Secondary reactions (e.g. incubation of fluorochrome-labeled avidin with biotin-labeled cells) are performed in the same manner.

In another embodiment, the present invention contemplates the use of sandwich immunoassays to quantitate the percentage of IgM+ or IgM-bp+ T-lymphocytes in a sample of bodily fluid. Such assays can include (1) a solid phase support, (2) a T-lymphocyte specific antibody which is capable of binding to a T-lymphocyte determinant, (3) an AD- specific antibody capable of binding to one of either IgM or IgM-bp, and (4) a detectable label coupled with, or able to associate with, at least one of the antibodies. In one embodiment, the solid support is a microtitre plate having two different types of wells. One well is coated with an AD-specific antibody and one is coated with a T-lymphocyte specific antibody. T-lymphocytes from a patient sample that is incubated with the support are bound to the solid support by the coating antibodies. The bound T-lymphocytes are subsequently reacted with yet another T-lymphocyte specific antibody, the binding of which is detected to determine the number of T-lymphocytes bound in each well. The number of cells detected in the wells coated with AD-specific antibodies is compared to the number of cells in the wells coated with T-lymphocyte specific antibodies.

The present invention also contemplates assay kits for measuring the percentage of

IgM+ or IgM-bp+ T-lymphocytes in a bodily fluid sample. The assay kits generally provide a first antibody for detecting a T-cell specific marker, and a second antibody for detecting either an IgM class antibody or the 96kd IgM binding protein associated with T-cells. Both of the antibodies can be derivatized with a label group that can be ultimately detected, as for example, by spectrophotometric techniques or radiographic techniques. For instance, the label can be any one of a number of radioisotopes, fluorescent compounds, enzymes, and enzyme co-factors. To illustrate, the label group can be a functional group selected from the group consisting of horseradish peroxidase, alkaline phosphatase, β-galactosidase, luciferase, urease, fluorescein and analogs thereof, rhodamine and analogs thereof, allophycocyanin, R- phycoerythrin, erythrosin, europiam, luminol, luciferin, coumarin analogs, 125^ 131 ₅ 3^, 35_S, 14_C and 32p.

In certain embodiments of the subject assay kit, sample data from at least one standardized population of individuals can be provided to accomplish statistical comparison of the measured percentage of IgM+ or IgM-bp+ T-lymphocytes. The sample data can be used to establish predictive values (e.g. values of IgM+ or IgM-bp+ T cells) against which the patient's data can be compared to determine risk of Alzheimer's disease, or to detect recent apoplectic events. The standardized data can be derived, for example, from a "normal" population study, as described below. In addition, the standardized data can provide data obtained from an "abnormal" population study in which, for example, the presence or risk of Alzheimer's disease has been determined in a clinical trial using the method of the present invention. Such abnormal population studies can include a population of persons which uniformly suffer from a particular disease or injury. Measurement of IgM+ or IgM-bp+ percentages in a patient also suffering from such a disease or injury can be compared to this "specialized" standardized data set in order to more accurately predict the onset or risk of Alzheimer's disease in that patient. In similar fashion, standardized data sets can be derived from prediction of stroke in which other disease states are accounted for in the data. For instance, such specialized data sets can be used to account for other circulatory disturbances. Assay kits provided according to the invention may include a selection of several different types of AD-specific and/or T-lymphocyte specific antibodies. The antibodies may be in solution or in lyophilized form. In some embodiments, the T-lymphocyte specific antibodies may come pre-attached to the solid support, or they may be applied to the surface of the solid support when the kit is used. The labeling means may come pre-associated with the T-lymphocyte specific antibody, or may require combination with one or more components, e.g. buffers, antibody-enzyme conjugates, enzyme substrates, or the like, prior to use. Many types of detectable labels are available and could make up one or more components of a kit. Narious detectable labels are known in the art, and it is generally recognized that a suitable label group is one which emits a detectable signal. Narious label groups can be used, depending on the type of sandwich immunoassay conducted. Useful labels include those which are fluorescent, radioactive, phosphorescent, chemiluminescent, bioluminescent, and free radical. Also, the label groups may include polypeptides (e.g., enzymes or proteins), polymers, polysaccharides, receptors, cofactors, and enzyme inhibitors. Kits of the invention may also include additional reagent. The additional reagent can include blocking reagents for reducing nonspecific binding to the solid phase surface, washing reagents, enzyme substrates, and the like. The solid phase surface may be in the form of microtiter plates, microspheres, or the like, composed of polyvinyl chloride, polystyrene, or the like materials suitable for immobilizing proteins. Such materials having solid phase surfaces are referred to herein as "support means".

The patient sample may consist of any body fluid, including but not limited to peripheral blood, plasma, cerebrospinal fluid, lymphatic fluid, peritoneal fluid, or pleural fluid or any body tissue. Binding may be accomplished and/or detected in vitro. In vitro binding may be performed using histologic specimens or subtractions of tissue or fluid, i.e., substantially purified T-lymphocytes. Peripheral blood lymphocytes (PBLs) (which contain about 80%) T-lymphocytes) are a particularly useful as a source of T-lymphocytes for the present assay. PBLs can be obtained by standard techniques (e.g., Boyum et al. (1968) Scand. J. Clin. Lab. Invest 21:77). If desired, a fraction of cells enriched in T-lymphocytes (to about 95%ι) can be obtained using standard techniques, such as by eliminating B cells by rosetting. Generally, PBLs will be obtained from fresh blood by Ficoll-Hypaque density gradient centrifugation.

The results of two related studies are provided below in Examples I and II. The primary objective of each of these studies was to determine whether there was an association between the percentage of IgM+ T-lymphocytes and dementia. Example I describes the expression of an IgM binding protein with bound autochthonous IgM on freshly isolated peripheral blood T-lymphocytes from 113 individuals between 20 and 89 years of age. This new T-lymphocyte phenotype was detected in subjects as early as the third decade and increased in percentage and prevalence with each twenty year age group thereafter. In 49 of the 113 donors, no obvious correlation was detected between the number of IgM bound T- lymphocytes per total T-lymphocytes (percentage of IgM+ T-lymphocytes) and immunologic cell phenotypes examined including the percentage of CD3, CD4 or CD8 cells, B cells, CD5+ B cells, and levels of serum IgM or IgM rheumatoid factor. In another 48 of the 113 donors, it was found that the IgM+ T-lymphocytes increased with decreasing scores obtained on the Mini-Mental State Examination (r = -0.64 and p < 0.001). After adjusting for age, this correlation remained significant (r = -0.43 and p = 0.003).

As presented in Example II, in order to investigate this phenomena further, a sample of older adults was studied to determine whether there was an association between percentage of IgM+ T-lymphocytes and dementia, and in particular, Alzheimer's Disease. Blood from a diverse group of demented and nondemented older adults was tested in this study, and the negative correlation between scores received on the MMSE and the percentage of IgM+ T- lymphocytes found in the Example I study was confirmed. Since the MMSE is only a screening test for cognitive dysfunction and does not identify specific cognitive disorders, the cognitive diagnosis of each patient in this study was determined from review of the medical records. From these diagnoses, the mean percentage of IgM+ T-lymphocytes of patients with AD was significantly higher than seen in those without AD. 83.3% of those individuals with percentage of IgM+ T-lymphocytes greater than 20% had AD. In contrast, 25.7% of those with a percentage of IgM+ T-lymphocytes < 20% had AD (c2 = 15.9 and p = < 0.001). This study strongly supports the assertion that an elevated percentage of IgM+ T-lymphocyte level above 20%) is associated with Alzheimer's Disease.

In an attempt to determine whether there were further associations with the percentage of IgM+ T-lymphocytes, a number of other variables were examined, including age, sex, race, medications used in greater than 10% of patients, medical diagnoses occurring in greater than 10% of patients, and absolute lymphocyte counts (ALC). With the exception of the medical diagnosis of stroke and GI medications, no other association besides the diagnosis of AD correlated with percentage of IgM+ T-lymphocytes. However, when stroke and GI medication were controlled for in a general linear model, AD remained independently associated with the percentage of IgM+ T-lymphocytes.

As discussed below in Example II, the mean percentage of IgM+ T-lymphocytes was found to be significantly different between those patients with a medical diagnosis of stroke and those without this diagnosis. However, unlike AD patients, stroke patients were found to have significantly smaller percentage of IgM+ T-lymphocytes when compared to non-stroke patients (17.6% ± 31.9% versus 27.7% ± 29.7%, p < 0.002). Moreover, when recent stroke patient (i.e. where stroke had occurred in the preceding 9-72 days were assessed independent of remote stroke patients (e.g. were apoplexy occurred 9 months to 23 years prior to trial), the diminishment in value of the percentage of IgM positive T lymphocytes was even more pronounced (4.2%) ± 4.3%) and, as described below, was also found to be independently associated with occurrence of a recent apoplectic event. While it is not yet apparent whether particular cerebral circulatory disturbances can be discerned within this set, it may be possible to utilize the present invention to distinguish even between stroke caused by thrombolic, embolic, or hemorrhaging events, which can be important in, for example, assessment of multi-infarct dementia. A larger sample group will be required to address this issue.

Moreover, the predictive value to the present invention in discerning between various dementia, particularly where more than one causative agent may be at work, e.g. stroke and Alzheimer's, a standardized set of IgM+ T-lymphocyte data for stroke victims can be assembled, and appropriate threshold values for prediction of Alzheimer's disease established for this specialized set. Where a new patient is diagnosed as having suffered a stroke, data obtained by subsequent application of the subject method can be compared with the standardized "stroke" data rather than with the general population data, and diagnosis of Alzheimer's disease made accordingly.

Moreover, the mean percentage of IgM+ T lymphocytes was also found to be lower in those patients taking GI medications and those not taking any such medications (9.6% ± 16.8%), n=10 versus 27.5 ± 29.9%, n=43, respectively, pθ.01). Nine out of the ten patients on GI medications were non-AD patients. GI medication are diverse category including antacids, cathartics, laxatives and H₂ receptor antagonists. The results attributed to GI medications, while potentially spurious, may nevertheless be useful in generating standard data sets which account for GI medications.

When a cutoff of greater than 20% IgM+ T-lymphocytes was utilized to support the diagnosis of AD, the specificity was 89.7% and the sensitivity was 62.5%). Using this criteria, only three patients without a diagnosis of AD in the Example II study were found to have elevated percentages of IgM+ T-lymphocytes. Interestingly two of these patients were felt to be demented, and thus may represent misclassification. For example, as shown in Figure 1 , the patient in category 5 had a history of closed head trauma with resulting chronic subdural hematomas. Although the family felt the dementia began after the head trauma, review of the medical record revealed changes in cognition typical of those of AD patients which occurred prior to the accident. The second patient was felt to have Multi-infarct Dementia (MID). However since AD and MID coexist in approximately 10-15%) of patients, it is possible that this patient actually had both diseases. Moreover, this patient had an undiagnosed anemia and had refused bone marrow examination. It is possible that an immunological disorder existed in this patient that could have interfered with the assay. This latter hypothesis may also explain the third patient, who had normal cognition but suffered from idiopathic thrombocytopenia. This test, which identifies an unusual T cell phenotype present in the immune system, may give a high percentage of false positives in individuals with certain autoimmune disorders, though such patients may be controlled for.

The significance of the elevated percentage of IgM+ T-lymphocytes in AD patients is unclear. Recent work has suggested that inflammatory mechanisms may be involved in the pathogenesis of AD. Both T and B lymphocytes can be detected in AD brain tissue, with T- lymphocytes outnumbering B cells. Whether the circulating percentage of IgM+ T- lymphocytes contribute to the pathology of AD or whether they represent an abnormality induced by the Alzheimer's Disease itself is unknown. Moreover, not all of the AD patients had elevated levels of these T-lymphocytes. This did not appear to be related to severity of disease as measured by MMSE scores or duration of AD, since neither of these variables correlated to percentage of IgM+ T-lymphocytes. However, AD is clearly a heterogeneous illness, which is likely to have more than one etiology. It is thus possible that a subset of AD patients may express the IgM binding protein on their T-lymphocytes in higher numbers or the IgM binding protein may bind IgM more avidly for reasons that relate to etiology. From the present data, it appears that this new T-lymphocyte phenotype occurs normally in levels less than 20%), and that either primary or secondary immunological aberrations likely occur in AD to cause this ratio to increase.

The data from the study of Example I reveals a significant correlation between the percentage of IgM+ T-lymphocytes and MMSE scores (r = -0.64 and p = O.001). An analysis of the sample in the study of Example II showed a significant association of percentage of IgM+ T-lymphocytes and MMSE, but the correlation coefficient was lower than in the previous study (r = -0.33 and p = 0.02). The previous study contained more younger adults with higher MMSE scores than this study.

Also, in the study of Example I, the percentage of IgM+ T-lymphocytes increased with age to give an average of 6.5%, 8.7%, 11% and 18.9% for the 20-39, 40-59, 60-79 and

> 80 year age groups respectively. In the study of Example II, no correlation was found between %IgM+ T cells and age. We did not include any patients in the 20-35 year age group in this study, but the mean %IgM+ T cells was %21.6 (n=6), %26.3 (n=31), %20.7 (n=16) for the 40-59, 60-79 and >80 age groups respectively. The reason for the differences between the two studies with regard to percentage of IgM+ T-lymphocytes correlation with age may derive from the different patient samples. It is unknown what percentage of patients had AD in the Example I study. However, it is probable, given the prevalence of AD in the aging population, that some patients of the Example I study had AD and some of these patients had elevated percentage of IgM+ T-lymphocyte levels which could account for the elevated average percentage of IgM+ T-lymphocyte levels in the first study compared with the adjusted averages in the current study. Thus the apparent correlation of the percentage of IgM+ T-lymphocytes with increasing decades seen in the Example I study may have been due to the increasing numbers of AD patient in the older age ranges.

Although the mean absolute lymphocyte count (ALC) was decreased in the AD patients, the difference between the ALC in the AD versus non-AD patients was not statistically different. There was no correlation of ALC with increasing age in these studies.

Other associations with percentage of IgM+ T-lymphocytes may emerge as more data is collected. Although other brain dysfunctions, such as stroke, Parkinson's Disease, and other dementias that might be confused with AD were included in the Example I study, the sample sizes were small. Given this, however, the test may provide a mechanism to discriminate MID from AD, which is often difficult on clinical grounds.

The IgM binding protein present on the surface of an increasingly high proportion of T-lymphocytes from aged individuals likely represents a new member of the Fc IgM receptor family. This Fc receptor, detected in vivo with significant autochthonous Ig bound to the receptor represents a unique opportunity to study the effects of the Ig-Fc receptor complex. Since the binding of IgM to freshly isolated T-lymphocytes could be detected, the surface molecules which facilitated the IgM binding was isolated. A human monoclonal IgM coupled to sepharose 4B beads was used to precipitate a 96 kDa IgM binding protein from 125ι_E-rosetted T-lymphocytes from a 76 year old donor (donor X) under reducing conditions. No band was detected in the B cells from donor X or unseparated peripheral blood mononuclear cells from a young donor. In order to determine the specificity of the IgM binding membrane component, the precipitations were then performed in the presence of an excess of IgM or non-IgM immunoglobulin isotypes. E-rosetted T-lymphocytes from donor X and a different young donor, W, were precipitated as follows. Samples were precipitated with the monoclonal anti-human IgM as described above and the 96 kDa band was again detected from X, but not from the young donor, W. Other samples were precipitated in the same way except for the addition of a 100 fold excess of purified IgM, IgA or IgG respectively. The precipitation of the 96 kDa band was inhibited only with soluble excess IgM and not with equivalent amounts of purified IgA or IgG. These results clearly demonstrate that the 96 kDa IgM binding protein is specific for immunoglobulins of the IgM isotype.

The IgM binding protein of the present invention can be used to produce anti-IgM-bp antibodies by known techniques. Both monoclonal and polyclonal antibodies (Ab) directed against IgM-bp, and antibody fragments such as Fv, scFv, Fab and F(ab')2, can be used in the present immunoassays. The purified IgM-bp can be used directly as an immunogen, or can be linked to a suitable carrier protein by conventional techniques, including by chemical coupling means as well as by genetic engineering using a cloned gene of the IgM-bp. The purified IgM-bp can also be covalently or noncovalently modified with non-proteinaceous materials such as lipids or carbohydrates to enhance immunogenecity or solubility. The IgM- binding protein can also be coupled with or incorporated into a viral particle, a replicating virus, or other microorganism in order to enhance immunogenicity. The IgM-bp may be, for example, chemically attached to the viral particle or microorganism or an immunogenic portion thereof. The present invention is understood to include all such chemical modifications of the IgM-binding protein, so long as the modified peptide antigens retain substantially all the antigenic/immunogenic properties of the parent mixture.

In an illustrative embodiment, the IgM binding protein, or a fragment thereof (e.g. produced by limited proteolysis) is conjugated to a carrier which is immunogenic in animals. Preferred carriers include proteins such as albumins, serum proteins (e.g., globulins and lipoproteins), and polyamino acids. Examples of useful proteins include bovine serum albumin, rabbit serum albumin, thyroglobulin, keyhole limpet hemocyanin, egg ovalbumin and bovine gamma-globulins. Synthetic polyamino acids such as polylysine or polyarginime are also useful carriers. With respect to the covalent attachment of IgM-bp or IgM-bp fragment to a suitable immunogenic carrier, there are a large number of chemical cross- linking agents that are known to those skilled in the art. For the present invention, the preferred cross-linking agents are heterobifunctional cross-linkers, which can be used to link proteins in a stepwise manner. Heterobifunctional cross-linkers provide the ability to design more specific coupling methods for conjugating proteins, thereby reducing the occurrences of unwanted side reactions such as homo-protein polymers. A wide variety of heterobifunctional cross-linkers are known in the art. These include: succinimidyl 4-(N- maleimidomethyl) cyclohexane- 1-carboxylate (SMCC), m-Maleimidobenzoyl-N- hydroxysuccinimide ester (MBS); N-succinimidyl (4-iodoacetyl) aminobenzoate (SIAB), succinimidyl 4-(p-maleimidophenyl) butyrate (SMPB), l-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC); 4-succinimidyloxycarbonyl- a-methyl-a-(2-

^• pyridyldithio)-tolune (SMPT), N-succinimidyl 3-(2-pyridyldithio) propionate (SPDP), succinimidyl 6-[3-(2-pyridyldithio) propionate] hexanoate (LC-SPDP). Those cross-linking agents having N-hydroxysuccinimide moieties can be obtained as the N- hydroxysulfosuccinimide analogs, which generally have greater water solubility. In addition, those cross-linking agents having disulfide bridges within the linking chain can be synthesized instead as the alkyl derivatives so as to reduce the amount of linker cleavage in vivo.

In addition to the heterobifunctional cross-linkers, there exists a number of other cross-linking agents including homobifunctional and photoreactive cross-linkers. Disuccinimidyl suberate (DSS), bismaleimidohexane (BMH) and dimethylpimelimidate.2 HCI (DMP) are examples of useful homobifunctional cross-linking agents, and bis-[β-(4- azidosalicylamido)ethyl]disulfιde (BASED) and N-succinimidyl-6-(4'-azido-2'-nitrophenyl- amino)hexanoate (SANPAH) are examples of useful photoreactive cross-linkers for use in this invention. For a recent review of protein coupling techniques, see Means et al. (1990) Bioconjugate Chemistry 1 :2-12, incorporated by reference herein.

One particularly useful class of heterobifunctional cross-linkers, included above, contain the primary amine reactive group, N-hydroxysuccinimide (NHS), or its water soluble analog N-hydroxysulfosuccinimide (sulfo-NHS). Primary amines (lysine epsilon groups) at alkaline pH's are unprotonated and react by nucleophilic attack on NHS or sulfo-NHS esters. This reaction results in the formation of an amide bond, and release of NHS or sulfo-NHS as a by-product.

Another reactive group useful as part of a heterobifunctional cross-linker is a thiol reactive group. Common thiol reactive groups include maleimides, halogens, and pyridyl disulfides. Maleimides react specifically with free sulfhydryls (cysteine residues) in minutes, under slightly acidic to neutral (pH 6.5-7.5) conditions. Halogens (iodoacetyl functions) react with -SH groups at physiological pH's. Both of these reactive groups result in the formation of stable thioether bonds.

The third component of the heterobifunctional cross-linker is the spacer arm or bridge. The bridge is the structure that connects the two reactive ends. The most apparent attribute of the bridge is its effect on steric hindrance. In some instances, a longer bridge can more easily span the distance necessary to link the IgM-bp, or fragment thereof, with a suitable carrier protein. For instance, SMPB has a span of 14.5 angstroms. Preparing protein-protein conjugates using heterobifunctional reagents is a two-step process involving the amine reaction and the sulfhydryl reaction. For the first step, the amine reaction, the protein chosen should contain a primary amine. This can be lysine epsilon amines or a primary alpha amine found at the N-terminus of most proteins. The protein should not contain free sulfhydryl groups. In cases where both proteins to be conjugated contain free sulfhydryl groups, one protein can be modified so that all sulfhydryls are blocked using for instance, N-ethylmaleimide (see Partis et al. (1983) J. Pro. Chem. 2:263, incorporated by reference herein). Ellman's Reagent can be used to calculate the quantity of sulfhydryls in a particular protein (see for example Ellman et al. (1958) Arch. Biochem. Biophys. 74:443 and Riddles et al. (1979) Anal. Biochem. 94:75, incorporated by reference herein).

The reaction buffer should be free of extraneous amines and sulfhydryls. The pH of the reaction buffer should be 7.0-7.5. This pH range prevents maleimide groups from reacting with amines, preserving the maleimide group for the second reaction with sulfhydryls.

The NHS-ester containing cross-linkers have limited water solubility. They should be dissolved in a minimal amount of organic solvent (DMF or DMSO) before introducing the cross-linker into the reaction mixture. The cross-linker/solvent forms an emulsion which will allow the reaction to occur.

The sulfo-NHS ester analogs are more water soluble, and can be added directly to the reaction buffer. Buffers of high ionic strength should be avoided, as they have a tendency to "salt out" the sulfo-NHS esters. To avoid loss of reactivity due to hydrolysis, the cross-linker is added to the reaction mixture immediately after dissolving the protein solution.

The reactions can be more efficient in concentrated protein solutions. The more alkaline the pH of the reaction mixture, the faster the rate of reaction. The rate of hydrolysis of the NHS and sulfo-NHS esters will also increase with increasing pH. Higher temperatures will increase the reaction rates for both hydrolysis and acylation.

Once the reaction is completed, the first protein is now activated, with a sulfhydryl reactive moiety. The activated protein may be isolated from the reaction mixture by simple gel filtration or dialysis. To carry out the second step of the cross-linking, the sulfhydryl reaction, the protein chosen for reaction with maleimides, activated halogens, or pyridyl disulfides must contain a free sulfhydryl, usually from a cysteine residue. Free sulfhydryls can be generated by reduction of protein disulfides. Alternatively, a primary amine may be modified with Traut's Reagent to add a sulfhydryl (Blattler et al. (1985) Biochem 24:1517, incorporated by reference herein). Again, Ellman's Reagent can be used to calculate the number of sulfhydryls available in protein.

In all cases, the buffer should be degassed to prevent oxidation of sulfhydryl groups.

EDTA may be added to chelate any oxidizing metals that may be present in the buffer. Buffers should be free of any sulfhydryl containing compounds.

Maleimides react specifically with -SH groups at slightly acidic to neutral pH ranges (6.5-7.5). A neutral pH is sufficient for reactions involving halogens and pyridyl disulfides. Under these conditions, maleimides generally react with -SH groups within a matter of minutes. Longer reaction times are required for halogens and pyridyl disulfides.

The first sulfhydryl reactive-protein prepared in the amine reaction step is mixed with the sulfhydryl-containing protein under the appropriate buffer conditions. The protein- protein conjugates can be isolated from the reaction mixture by methods such as gel filtration or by dialysis.

To prepare polyclonal rabbit antibodies to IgM-bp epitopes, the immunogen (1 mg in 1 ml) is emulsified with an equal volume of Freund's Complete Adjuvant and injected intradermally into each rabbit. The process is repeated after two weeks. Two weeks later, monthly subcutaneous booster injections are begun with 0.5 mg in 0.5 ml of the immunogen and 0.5 ml of Freund's Incomplete Adjuvant per animal. The rabbits are bled biweekly by a marginal ear vein technique beginning six weeks after the primary immunization. The blood collected is refrigerated, allowing clots to form, and the supernatant (antisemm) retained. The antisemm from each rabbit is collected and stored, either at -20o C. without preservative, or at 4o C. after addition of sodium azide to a final concentration of 0.1%. The same schedule is followed for each immunogen.

Monoclonal antibodies to the IgM-bp can be generated by applying generally known fusion cell techniques (cf. G. Kohler, C. Milstein, Vol 6, Eur J Immunol, pp 511-519 (1976) and M. Shulmen et al, vol 276, Nature pp 269-270 (1978) herein incorporated by reference) to obtain a hybridoma producing the antibody. Monoclonal antibodies are prepared by obtaining mammalian lymphocytes (preferably spleen cells), committing the lymphocytes to produce antibodies (e.g., by immunizing the mammal with the particular IgM-bp antigenic determinant beforehand), fusing the lymphocytes with myeloma (or other immortal) cells to form hybrid cells, and then culturing a selected hybrid cell colony in vivo or in vitro to yield antibodies which are identical in structure and specificity. In particular, monoclonal antibodies to the IgM-bp can be raised by employing IgM- bp, or fragments thereof, as set out above in the production of polyclonal antibodies. Mice or other animals can be challenged by injection with a solution of IgM-bp in complete Freund's adjuvant at weekly intervals. After the initial injection, the booster injections can be administered without adjuvant or emulsified in incomplete Freund's adjuvant.

Serum samples from the immunized animal can be taken and analyzed by an immunoassay to detect the presence of antibodies cross-reactive with the IgM-bp. Animals that exhibit antibody titers for IgM-bp are sacrificed and their spleens homogenized. Alternatively, the spleen cells can be extracted and the antibody-secreting cells expanded in vitro by culturing with a nutrient medium. The spleen cells are then fused with myeloma (or other immortal) cells by the above-referenced procedure of Kohler and Milstein. The hybridomas so produced are screened (i.e., cloned by the limiting dilution procedure of the above-referenced Baker et al. article) to select a cell producing antibodies which react specifically with IgM-bp. Large scale antibody production can be obtained from such cell lines by various techniques, including the induction of ascites tumors (e.g., after priming with pristane) and the purification of such antibodies from the ascites fluid by Protein A- Sepharose affinity chromotography.

The present invention will now be illustrated by the following examples, which are not intended to be limiting in any way.

Example I IgM Binding Protein with Bound Authchthonous IgM or T Cells from Aged Adults

Peripheral blood leukocytes (PBLs) were obtained after receiving informed consent from 60-89 year old donors attending two separate outpatient clinics. Forty-nine blood samples were obtained from consecutive ambulatory donors from the Alabama Respiratory Disease Clinic (ARDC) and twenty-two samples were obtained from donors at a University of Alabama at Birmingham (UAB) Geriatric Primary Care Clinic (GPCC). Together these populations consisted of 40 black males, 11 white males, 7 black females and 13 white females. Blood was also obtained from employees at UAB between the ages of 20-59. This younger population consisted of 22 white females, 1 black female, 16 white males, 1 Asian female and 2 Asian males.

Mental status was evaluated using the MMSE in the 22 donors from the GPCC and in 27 of the 42 UAB employees. This test was administered by a study physician, a study investigator or a nurse. PBLs were isolated from freshly obtained heparinized blood over Ficoll-Hypaque (Pharmacia Fine Chemicals, Piscataway, NJ) density gradient. Buffy coat containing the mononuclear cell fraction was then washed in PBS with 5% FCS and stained as described below. The plasma was collected and IgM purified as described below. To remove IgM from the T-lymphocytes, PBLs were cultured in RPMI (Gibco) with 10% FCS for 2 hours at 37° C. The cells were washed once and stained as described below to demonstrate the IgM binding protein. Purification of T-lymphocytes from PBLs was performed by E-rosetting. Briefly, T-lymphocytes were pelleted with neuraminidase treated sheep red blood cells, incubated for 1 hour on ice followed by separation of the rosetted T-lymphocytes from the non-rosetted B cells by density gradient centrifugation.

The CD5-PE, CD3-PE, CD4-PE and CD8-PE were obtained from Becton Dickinson, (Mountainview, CA). SA-DA4-4, a monoclonal anti-human IgM was made here. FITC goat anti-human IgM, FITC goat anti-mouse IgM and Streptavidin-FITC (SA-FITC) were obtained from Southern Biotechnology (Birmingham, AL). The human IgA, IgG and two IgM antibody preparations were purified by affinity chromatography from sera of unrelated individuals with multiple myeloma and are described by Ohno et al. (1990) J Exp Med 172:1165. One of the IgM preparations, LP, is known to bind to intermediate filaments, however the specificity of the other myeloma IgM, EE, is unknown. The mouse IgMK, Cla, is specific for chicken MHC Class II.

To purify human IgM from plasma, affinity columns were constmcted using activated Sepharose 4-B beads (Pharmacia) coupled to SA-DA4-4 (a monoclonal anti-human IgM antibody). The plasma was applied over the column followed by washing with borate saline with monitoring. The IgM was eluted with glycine-HCl, pH 2.7 into borate saline and the pH was quickly adjusted to physiologic pH, if necessary. The IgM was dialysed, concentrated and tested for anti-T cell reactivity.

To detect T-lymphocytes binding autochthonous IgM, viable PBLs (0.25x10^) were incubated for 10 minutes on ice with lOul of CD5-PE or CD3-PE, washed and then stained with lOul of either FITC anti-human IgM (O.lmg/ml) or SA-DA4-4 biotin (O.lmg/ml) followed by SA-FITC (1/50) To detect the IgM binding protein, T-lymphocytes were cultured in RPMI 1640 to remove bound IgM as described above. Cells were then stained with 20ul human IgM (EE, O.lmg/ml) or 20ul of mouse IgMK (Cla, O.lmg/ml) followed by either FITC goat anti-human or -mouse IgM. To determine the phenotype of IgM+ T- lymphocytes, cells were co-stained with CD4-PE or CD8-PE followed by FITC anti-human IgM. Cells were analyzed by flow cytometry using a FACScan (Becton Dickinson) and the Consort 30 analysis program.

Associations between different quantitative parameters were conducted using the Spearman rank correlation (Biostatistical Analysis. 2nd. Zar, J.H., Prentice Hall, Inc.: NJ, 1984) and student's T-test.

T-lymphocytes with autochthonous bound IgM (IgM+ T-lymphocytes) were identified by staining peripheral blood mononuclear cells with either PE anti-CD5 or PE anti- CD3 (both pan T-lymphocyte markers) followed by polyclonal FITC anti-human IgM. Similar results were obtained when PBLs from selected individuals were stained using a monoclonal anti-human IgM antibody, SA-DA4-4 biotin, followed by SA-FITC in conjunction with PE-anti-CD5.

PBLs from a total of 113 donors from the ARDC (n=49), GPCC (n=22) as well as employees from UAB (n=31) were analyzed to determine how the IgM+ T/T ratio changes with age. As can be seen in the scatter plot in Figure 2, the IgM+ T/T ratio increases with age to give an average of 6.5%, 8.7%, 11% and 18.9% for the 20-39 year (n=25), 40-59 (n=18), 60-79 (n=40) and >80 (n=30) year age group, respectively. In addition, the percentage of individuals with a ratio of IgM+ T/T of greater than 15% increased from 6.7%) in the 20-39 year age group an extra 5% each 20 year period thereafter until the greater than 80 age when almost half of the individuals had greater than 15% IgM+ T-lymphocytes in the total T- lymphocyte population. These results are suggestive that not only do the IgM+ T/T ratio increase with age, but also the incidence of this phenotype.

PBLs from 22 consecutive donors from the GPCC were analyzed to determine the phenotype of T-lymphocytes with associated autochthonous IgM. The percentages of IgM+ CD3, CD4 and CD8 T-lymphocytes as well as the CD4:CD8 ratio were measured. In 17/22 donors {11%), the percentage of IgM+ CD8 T/total CD8 T-lymphocytes was at least twice as high and up to 24 times greater than the percentage of IgM+ CD4 T/total CD4 T- lymphocytes. In 2/22 donors (9%), however, there was almost twice as many CD4+ T- lymphocytes with bound IgM than CD8+ T-lymphocytes, while 3 donors (14%) had few detectable IgM+ T-lymphocytes of either type. The CD4:CD8 ratio varied from 0.9 to 7.0 in these individuals, however greater than 90% of them had more CD4 than CD8 T- lymphocytes. Therefore a higher proportion of CD8 T-lymphocytes appeared to bind autochthonous IgM than CD4 T-lymphocytes, despite the fact that there was a greater percentage of CD4 T-lymphocytes in most of the donors.

Freshly isolated PBLs from a 76 year old donor (donor X) were stained with FITC anti-human IgM followed by PE anti-CD3 to detect IgM+ T-lymphocytes. In addition, an aliquot of unstained cells was incubated at 37° C for 2 hours in complete RPMI media and restained as above. Incubation of the PBLs in vitro eliminated detectable bound autochthonous IgM, however at least some of the receptor for the IgM remained intact since restaining of the surface IgM-cleared T-lymphocytes with either monoclonal human or mouse IgM restored IgM binding on the T-lymphocytes. These results demonstrated the presence of a functional IgM binding protein on the surface of these cultured T-lymphocytes although the binding to heterologous IgM was always less than that of autochthonous IgM present on the freshly isolated T-lymphocytes. These reassociation experiments also clearly show that both human as well as mouse IgM can bind to the human IgM binding protein present on PBLs from aged individuals.

To determine if any of the IgM detected on T-lymphocytes was the result of anti-T cell reactivity present in the sera of the PBL donors, IgM was purified from the plasma of 13 consecutive donors age 69-89 from the GPCC by affinity chromatography using SA-DA4-4 coupled to sepharose. Purified IgM (0.1 mg/ml) was then used to stain T-lymphocytes from a 25 year old individual (donor Y) and a 69 year old individual (donor X) with 1% and 40% IgM+ T-lymphocytes, respectively. Prior to staining with the panel of purified IgM preparations, the autochthonous IgM on T-lymphocytes from donorY and X was removed by incubation of the cells for 2 hours at 37° C as described above. The ages of the donors from which the IgM was purified as well as the percentage of T-lymphocytes in these donors with bound IgM at the time the plasma was collected are shown in Table I below. Donors 1-6 each contained greater than 15% IgM+ T/T cells while donors 7-13 had <10%. If the IgM detected on peripheral T-lymphocytes is the result of autoantibody activity then higher levels of binding by IgM purified from donors 1-6 would be expected than from donors 7-14. However, it is clear from Table I that IgM from donors 1-6 does not bind purified T- lymphocytes from either donor Y or X, in fact, even the IgM preparation from donor X did not bind to cells from donor X, despite the fact that 40% of his T-lymphocytes originally bound autochthonous IgM. Some anti-T cell reactivity is detected in donors 7, 8, and 11 although little autochthonous T-lymphocyte associated IgM was detected on the freshly purified PBLs from these individuals . These results suggest that the IgM detected on T- lymphocytes from these aged individuals is not the result of IgM auto-antibodies against T- lymphocytes.

Table I

Staining of 25 Staining of 69

% IgM+ year old donor year old donor

Donor Age T/T T cells *# T cells

1 89 32

2 69 40 - -

3 86 41 - -

4 72 32 - -

5 79 17 - -

6 80 56 +/- -

7 82 8 +/- +

8 81 1 + +

9 82 3 - -

10 80 9 - -

11 83 9 + +

12 78 3 - -

13 83 1 - -

This 25 year old donor had <l/% IgM+ T/T cells.

** This 69 year old donor had 40% IgM+ T/T cells.

# The autochthonous IgM was removed from these T cells prior to the staining by incubation at 37 degrees Cfor 2 hours.

Analysis of the original 49 donors from the ARDC, showed that there was no statistically significant correlation between the levels of IgM+ T-lymphocytes and percentage of CD3, CD4 or CD8 T-lymphocytes, B cells,CD5+ B cells, levels of semm IgM or IgM rheumatoid factor. However, it was noted that many of the individuals from the GPCC clinic were cognitively impaired, thus the MMSE was used to measure cognitive function in the individuals from the GPCC (n=22) as well as the donors from UAB (n=27). There was a statistically significant correlation between the IgM+ T/T ratio and the scores received on the MMSE (r=-0.640, pO.OOl) using Spearmans rank correlation among these donors. In general, donors receiving higher MMSE scores presented with a lower IgM+ T/T as indicated in the bar graph in Figure 3. It should be stressed that while MMSE scores alone cannot provide a clinical diagnosis of dementia, studies have shown that this test is 87% sensitive and 82% specific in accordance with the diagnosis of ward patients by psychiatrists when scores of 23 or less indicate the presence of either dementia or delirium. Therefore, in general, non-demented individuals received scores of 24 or greater, while cognitively impaired individuals generally score less than 24. It is of interest that individuals receiving scores of 24 or greater (n=37) had an average of 6.4 +/- 1.0 percentage of IgM+ T/T while the average percentage of IgM+ T/T rose to 24.1+/- 4.6 in individuals with scores of 23 or less (n=l 1) (two-tailed t-test, t=5.886, pθ.001).

Example II Association between Alzheimer's Disease and Bound Autochthonous IgM on T Cells

The proper evaluation of any new diagnostic test requires that the sample studied include patients of varying disease severity in the disease of interest as well as other patients who might have other diagnoses often confused with the disease of interest. Given the desire to evaluate this test as a potential marker for dementia, patients were recmited from various sites to form a sample pool comprising patients with differing types and severities of dementia, stroke, psychiatric disorders as well as cognitively normal individuals. The subjects were recmited from four different inpatient and outpatient sites at the University of Alabama at Birmingham Medical Center, including: 1.) the geriatric primary care clinic (n=28); 2.) the inpatient unit at the rehabilitation center, where either post-stroke or post- fracture rehabilitation was provided (n=8); 3.) the inpatient geriatric psychiatric unit (n=3); and 4.) the outpatient psychiatric clinic (n=14) after subjects had completed their participation in clinical trials of investigational medications for the treatment of AD or AD with depression. In the fourth group, all patients had been free of investigational medications for at least 30 days prior to obtaining the blood samples.

Sixty-five adults were asked to participate in the study. Fifty-three patients met the eligibility criteria of willingness to participate in the study, to complete the Mini-Mental State Examination (MMSE) and to donate a five milliliter sample of blood. Twelve patients were excluded for the following reasons: two patients refused the blood donation; two blood samples were mistakenly not drawn; two patients were uncooperative with the MMSE; in four patients, a cognitive diagnosis could not be determined due to inadequate information available in the medical records; in one patient, there were technical problems during the isolation of the white blood cells; and, in one patient, poor antibody preparation during the staining procedure prevented accurate interpretation of results. Data were collected on variables that could conceivably be correlated with the percentage of IgM+ T-lymphocytes. Age, sex, race, medications, medical diagnoses, MMSE scores, cognitive diagnosis, duration of AD (if applicable) and absolute lymphocyte counts (ALC) were obtained from the medical records at the time of venipuncture, or, were not available on the charts, a study investigator administered this test. Patients were considered to have a medical condition if it was mentioned in the medical history and documented in the chart.

Cognitive diagnosis was determined by retrospective chart review. In the patients from the first three recmitment sites, two study investigators, a board certified neurologist with expertise in dementing illnesses (LH) and a geriatrician (SK), independently reviewed the patient's cognitive history and medical evaluations to determine a diagnosis.

Disagreements in diagnosis were resolved by a second independent chart review, followed by a consensus opinion. Overall, disagreement occurred in less than 8% of cases. The diagnosis of probable and possible AD was based on the National Institute of Neurological and

Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders

Association's (NINCDS/ADRDA) criteria. Other cognitive diagnoses were based on DSM-

IIIR criteria and/or standard neurological classification. Patients from the fourth recmitment site were considered to have probable AD since inclusion into an investigational pharmaceutical trial requires a diagnosis of probable AD based on NINCDS/ADRDA criteria.

The cognitive diagnoses were divided into 11 different categories, as shown in Figure 1. Categories 1-6 represented demented patients with the following diagnoses: 1) Probable AD (n=23); 2) Possible AD (n=l); 3) Multi-infarct dementia (MID) (n=6); 4) Mixed AD/MID (n=l); 5) Demented, diagnosis unclear (n=l); 6) Neurological illness associated with dementia (n=l; this patient had Parkinson's Disease). Categories 7-11 represented patients without dementia, with the following diagnoses: 7) Age-associated memory impairment (n=2); 8) Recent stroke with focal neurological and focal cognitive deficits (n=4); 9) Recent stroke with focal neurological deficits only (n=l); 10) Not demented, diagnosis unclear (n=l ); 11 ) Normal older adults (n=l 2).

Peripheral blood was collected by venipuncture into purple-top vacutainers (EDTA, anticoagulant) and prepared within 24 hours of the collection time as follows. The blood was diluted with an equal volume of phosphate buffered saline (PBS) and layered on a Ficoll Hypaque density gradient (Pharmacia LKB, Piscataway, NJ) followed by centrifugation at 1800 rpm for 20 minutes. The white cells were collected from the buffy coat, washed twice in cold PBS-FCS, (phosphate buffered saline, pH 7.4, 5% fetal calf serum (HyClone, Logan, UT), 0.2%) sodium azide) and prepared for staining as follows. IO⁶ cells were stained with 20 ul of 0.05 mg/ml polyclonal FITC anti-human IgM (Southern Biotechnology Associates, Birmingham, AL) for 20 minutes on ice followed by washing with PBS-FCS buffer. Next, cells were stained with 10 ul of PE anti-CD3 (Becton Dickinson, Mountainview, CA) for 20 minutes on ice followed by a final wash with PBS-FCS. Cells were then fixed in 1% paraformaldehyde, pH 7.4, and analyzed by flow cytometry using a Consort 30 program.

The data were analyzed using the Statistical Analysis System (SAS) Programs. For the main outcome variable, percentage of IgM+ T-lymphocytes, a log-transformation was utilized in all statistical analyses. Univariate analysis was performed using correlation coefficients (r) for continuous variables and Chi square (χ²) and Fisher's Exact Test for categorical variables. Differences between groups were compared using general linear models. In all calculations, test results were considered to be statistically significant if p < 0.05 by a two-tailed test.

Of the 53 patients, 18 were male and 35 were female. Thirty-nine were white and 14 were African American. There were no significant differences in the mean percentage of IgM+ T-lymphocytes between males and females (27.0% ± 32.1% versus 22.6% ± 27.1%, p = 0.77) or whites and African Americans (23.9% ± 28.9% versus 24.8% ± 29.0%, p = 0.83). The mean age was 73.2 ± 9.1 years with a range from 48 to 89 years old. Univariate analyses showed no correlation between age and percentage of IgM+ T-lymphocytes (r = 0.02 and p = 0.87).

The mean percentage of IgM+ T-lymphocytes for the 53 patients was 24.1 ± 28.7% and the mean MMSE score was 20.4 ± 8.1 with the total possible score of 30. There was a significant negative correlation between MMSE scores and the percentage of IgM+ T- lymphocytes (r = -0.33 and p = 0.02) and MMSE scores and age (r = -0.27 and p = 0.048). The mean absolute lymphocyte count (ALC) in 51 patients was 1906 ± 623/mm³ with a range between 874/mm³ and 3280/mm³. The ALC was not significantly correlated with percentage of IgM+ T-lymphocytes (r = -0.03 and p = 0.84) or age (r = 0.08 and p = 0.56).

The relationship between cognitive diagnoses and the percentage of IgM+ T- lymphocytes is presented in Figure 1. Visual inspection of this graph suggested that AD patients had a greater percentage of their total T-lymphocytes bearing IgM when compared to all other cognitive diagnostic categories. Because there was no statistical difference in the mean percentage of IgM+ T-lymphocytes (t = 1.65 and p = 0.10) between the non-AD demented patients (cognitive diagnoses 3-6) and the nondemented patients (cognitive diagnoses 7-11), these groups were combined and then compared to the patients with AD (cognitive diagnoses 1-2). This analysis revealed a significant difference in percentage of IgM+ T-lymphocytes between the AD and non-AD groups (35.6 ± 30.2% versus 14.6 ± 23.9%, p < .001). As shown in Table II, when a cutoff of greater than 20% IgM+ T- lymphocytes is utilized to predict the diagnosis of probable or possible AD, the sensitivity was 62.5% and the specificity was 89.7% (χ2 = 15.9 and p < 0.001). Using this data base sample prevalence, the positive predictive value was 83.3% and the negative predictive value was 74.4%ι. This analysis revealed that only three non-AD patients had IgM+ T-lymphocytes values greater than 20%.

Table II Association of %IgM+ T Cells with Alzheimer's Disease

%IgM+ T Cells Alzheimer's Disease Non-Alzheimer's Disease

Patients, n Patients, n

>20% Ϊ5 3 <20% 9 26 χ = 15.9 and p< 0.001

To determine whether other medical conditions correlated with the percentage of IgM+ T-lymphocytes, statistical analysis was performed on those medical diagnoses that occurred in greater than 10% of patients. Table III below summarizes those findings. There were no significant differences except for the diagnosis AD and stroke. The mean percentage of IgM+ T-lymphocytes was significantly decreased in those with a history of stroke (17.6 ± 31.9 versus 26.7 ± 27.3, p = 0.01). However, after controlling for stroke, AD remained independently associated with the percentage of IgM+ T-lymphocytes (p = 0.01).

Table III

Mean %IgM+ T Cells Compared in Diagnoses

Occurring in >10% of Patients

Diagnosis Diagnosis

Present Absent n (Mean %IgM+ T n 24 (Mean %Igm+ T

Diagnosis Cells ± SD) Cells ± SD) p Value

Alzheimer's Disease 24 (35.6 ± 30.2) 29 (14.6 ± 23.9) O.001

Cardiovascular 19 (21.3 ± 25.2) 24 (25.7 ± 30.7) 0.23

Disease

Hypertension 27 (28.5 ± 33.4) 26 (19.5 ± 22.5) 0.88

Hypothyroidism 7 (30.6 ± 32.7) 46 (23.1 ± 28.3) 0.24

Stroke 15 (17.6 ± 31.9) 38 (26.7 ± 27.3) 0.01

Depression 6 (59.3 ± 46.0) 47 (19.6 ± 22.7) 0.12

To determine whether medication usage was correlated with the percentage of IgM+ T-lymphocytes, medications were grouped into categories according to the American Hospital Formulary Service Classification System. For each group of medications that were being taken by greater than 10% of patients, the mean percentage of IgM+ T-lymphocytes was compared in those on therapy versus those not on therapy. There were no significant differences in the six groups tested except GI medications: cardiovascular dmgs (p = 0.88), analgesics and antipyretics (p = 0.63), psychotherapeutic dmgs (p = 0.09), hormones and synthetic substitutes (p = 0.55), gastrointestinal dmgs (p < 0.01) and diuretics (p = 0.44).

Table IV summarizes the clinical characteristics of AD versus non-AD patients. There was a significant difference in MMSE scores but no difference in age, sex, race or ALC. There were no significant differences in those diagnoses that occurred in greater than 10% of AD versus non-AD patients except stroke. There was no difference in medication usage between AD and non-AD patients in the six groups of medications that greater than 10% of patients were using except gastrointestinal medication: cardiovascular dmgs (p = 0.11), analgesics and antipyretics (p = 0.45), psychotherapeutic dmgs (p = 0.44), hormones and synthetic substitutes (p = 0.68), gastrointestinal dmgs (p = 0.02) and diuretics (p = 0.57). As illustrated in Figures 4 A and 4B, a stepwise regression model was developed in order to control for the variables of age, Alzheimer's Disease, use of gastrointesinal medication, recent stroke (e.g. approx. 9-72 days) and remote stroke (e.g. 7 months to 23 years). After controlling for GI medications (step 1, Figure 4A), remote stroke (step 2, Figure 4B), and age (step 3, Figure 4B), only AD and recent stroke are regarded as being independently associated with the percentage of IgM+ T lymphocytes. For instance, after controlling for GI medications and all forms of stroke, AD remained independently associated with the percentage of IgM+ T lymphocytes (p=0.008).

Table IV

Variables Compared in Alzheimer's Disease

Versus Non- Alzheimer's Disease Patients

Variable Alzheimer's Non- p Value Disease Alzheimer's Disease n=24 n=29

Age, mean years ± SD 73.7 ± 6.8 72.8 ± 10.7 0.72

Race, % whites 83.3 65.5 0.14

Mini Mental State

Examination, mean 16.5 ± 7.3 23.6 ± 7.5 0.001 score ± SD, out of 30

Absolute lymphocyte count

(n=51), mean (per mm³) ± 1875 ± 555 1933 ± 687 0.74

SD

Cardiovascular disease, n 7 12 0.36

Hypertension, n 11 16 0.50

Hypothyroidism, n 4 3 0.69

Depression, n 3 3 1.00

Stroke, n 0 15 O.001 Of those with information available (n=22), the average duration of probable or possible AD was 4.6 ± 2.7 years with a range from one to 11.5 years. In those with AD, there was no significant correlation between the percentage of IgM+ T-lymphocytes and duration of AD (r = 0.10 and p = 0.65) or MMSE scores (r = 0.10 and p = 0.65). There was a negative correlation between duration of AD and MMSE scores (r = -0.49 and p = 0.02).

Example III Isolation of IgM Binding Protein from T-lymphocytes

Purified T-lymphocytes (3x10⁷) were incubated at 37°C for 2 hours to remove the bound IgM. The cells were then labeled with lmCi 125j (Amersham, Corp., Arlington Heights, IL) by the lactoperoxidase method and lysed in 1 ml of 1% NP-40 in 50mM Tris-HCl, pH 7.4. The lysate was centrifuged at 15,000 for 20 minutes at 4° C and cleared with lOOul sepharose 4B beads coupled to human albumin (5mg/ml beads). The IgM binding molecule was isolated from the cleared lysates by precipitation with 20ul sepharose 4B beads coupled with an IgMK (LP) (2.5 mg/ml beads) overnight at 4° C. After washing in lysis buffer, the IgM binding protein was dissociated from beads by addition of Laemli sample buffer and run on 10%) SDS-PAGE. To demonstrate the specificity of the IgM binding protein for IgM, cells were labeled and lysed as above and precipitated in four different ways (1) 5ul (2.5 mg/ml) IgMK (LP) sepharose beads, (2) 5ul IgMK (LP) sepharose 4B beads plus 1.5 mg IgM (EE) (3) 5ul IgMK (LP) sepharose beads plus 1.25 mg IgA (4) 5ul IgMk LP plus 1.25 mg IgG. Beads were washed and processed as above.

All of the above-cited references and publications are hereby incorporated by reference.

Equivalents

Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, numerous equivalents to the specific assay and reagents described herein. Such equivalents are considered to be within the scope of this invention and are covered by the following claims.

Claims

1. A method for augmenting a diagnosis of Alzheimer's disease comprising measuring, in a sample of bodily fluid containing peripheral blood mononuclear lymphocytes, a level of at least one of an IgM antibody and an IgM-binding protein being displayed on the cell surface of T-lymphocytes in the sample.

2. The method of claim 1, wherein the level of IgM antibody on the T-lymphocytes is measured by detecting cells displaying both a T-lymphocyte marker and a surface- bound IgM antibody; and the level of IgM-binding protein on T-lymphocytes is measured by detecting cells displaying both a T-lymphocyte marker and an IgM- binding protein having an apparent molecular weight in the range of 96,000 daltons.

3. The method of claim 2, wherein the T-lymphocyte marker is selected from a group consisting of CD2, CD3, CD4, CD5, CD7, CD8, CD28, and a T-lymphocyte receptor (TCR) protein.

4. The method of claim 2, wherein the detection of T-lymphocytes displaying both the T- lymphocyte marker and at least one of the IgM antibody and IgM-binding protein is accomplished by an immunoassay technique using a first antibody which specifically binds the T-lymphocyte marker, and a second antibody selected from a group consisting of an antibody which specifically binds the IgM antibody, an antibody which specifically binds the IgM-binding protein, and a combination of antibodies which collectively bind each of the IgM antibody and the IgM-binding protein.

5. The method of claim 4, wherein at least one of the first and second antibodies has attached thereto a label group able to be detected, and the binding of the labeled antibody is measured by detecting the label group.

6. The method of claim 5, wherein the label group is selected from a group consisting of radioisotopes, fluorescent compounds, enzymes, and enzyme co-factors.

7. The method of claim 6, wherein each of the first and second antibodies have attached thereto a fluorescent compound, and the immunoassay technique comprises detecting the binding of each of the first and second antibodies by fluorescence-activated cell sorting.

8. The method of claim 4, wherein the binding of at least one of the first and second antibodies is measured by a sandwich assay.

9. The method of claim 1 , wherein the bodily fluid is blood plasma or serum.

10. A method for identifying an individual's of risk of Alzheimer's disease, comprising i) providing a sample of a biological fluid from the individual, ii) measuring, by an immunoassay technique, a percentage of at least one of IgM- positive (IgM+) T-lymphocytes and IgM binding protein positive (IgM-bp+) T- lymphocytes of a population of T-lymphocytes in a bodily fluid, and iii) comparing the measured percentage of IgM+ T-lymphocytes or IgM-bp+ T- lymphocytes with a percentage of IgM+ T lymphocytes or IgM-bp+ T- lymphocytes measured in a normal population of individuals, wherein an increase in the measured percentage of IgM+ T-lymphocytes or IgM-bp+ T- lymphocytes relative to the percentage measured in the normal population is indicative that the individual has Alzheimer's disease or has an increased risk of developing Alzheimer's disease.

11. The method of claim 10, wherein the immunoassay technique comprises i) admixing with the sample a first antibody which specifically binds a T- lymphocyte marker, and a second antibody selected from a group consisting of an antibody which specifically binds an IgM class antibody and an antibody which specifically binds a T-lymphocyte IgM binding protein, and ii) detecting the binding of each of the first and second antibodies, wherein the level of IgM+ T-lymphocytes or IgM-bp+ T-lymphocytes is calculated from a ratio of cells in the sample which bind both the first and second antibodies relative to the cell which bind only the first antibody.

12. The method of claim 11, wherein the T-lymphocyte marker is selected from a group consisting of CD2, CD3, CD4, CD5, CD7, CD8, CD28, and a T-lymphocyte receptor (TCR) protein.

13. The method of claim 11, wherein the T-lymphocyte IgM-binding protein has an apparent molecular weight in the range of 96,000 daltons.

14. The method of claim 11, wherein each of the first and second antibodies has attached thereto a fluorescent group able to be detected, and the binding of each of the antibodies is detected by fluorescence-activated cell sorting.

15. A method for identifying an individual's of risk of Alzheimer's disease, comprising i) providing a sample of a biological fluid from the individual, ii) mixing with the sample a first antibody which specifically binds a T-lymphocyte marker and a second antibody which specifically binds an IgM class antibody, iii) determining a level of IgM-positive (IgM+) T-lymphocytes in the sample by measuring the binding of each of the first and second antibodies via an immunoassay technique, and iv) comparing the measured levels of IgM+ T-lymphocytes with levels of IgM+ T lymphocytes measured in a normal population of individuals, wherein an increase in the measured levels of IgM+ T-lymphocytes relative to the level of IgM+ T-lymphocytes measured in the normal population is indicative that the individual has Alzheimer's disease or has an increased risk of developing Alzheimer's disease.

16. The method of claim 15, wherein the binding of at least one of the first and second antibodies is measured by a sandwich assay.

17. The method of claim 15, wherein at least one of the first and second antibodies has attached thereto a label group able to be detected, and the binding of the labeled antibody is measured by detecting the label group.

18. The method of claim 17, wherein the label group is selected from a group consisting of radioisotopes, fluorescent compounds, enzymes, and enzyme co-factors.

19. The method of claim 17, wherein the label group a fluorescent group and the binding of the labeled antibody is detected by fluorescence-activated cell sorting.

20. The method of claim 15, wherein the first antibody binds a T-lymphocyte marker selected from a group consisting of CD2, CD3, CD4, CD5, CD7, CD8, CD28, and a T- lymphocyte receptor (TCR) protein.

21. The method of claim 15, wherein the second antibody binds a μ-chain of an IgM class antibody.

22. The method of claim 15, wherein the second antibody binds one of either a K or λ light chain of an antibody.

23. A diagnostic test kit for identifying an individual's of risk of Alzheimer's disease, comprising a first antibody for measuring, in a sample of bodily fluid provided from the individual, a level of cells displaying a T-lymphocyte marker, and a second antibody for measuring a level of cells in the sample displaying one of either an IgM class antibody or a 96,000 daltons T-lymphocyte IgM binding protein.

24. The diagnostic kit of claim 23, wherein at least one of the first and second antibodies has attached thereto a label group able to be detected, the label group being selected from a group consisting of radioisotopes, fluorescent compounds, enzymes, and enzyme co-factors

25. The diagnostic kit of claim 24, wherein the label group is a functional group selected from the group consisting of horseradish peroxidase, alkaline phosphatase, β- galactosidase, luciferase, urease, fluorescein and analogs thereof, rhodamine and analogs thereof, allophycocyanin, R-phycoerythrin, erythrosin, europiam, luminol, luciferin, coumarin analogs, 125^ 131j₅ 3ji₅ 35g₅ \4Q g^ 32p

26. The diagnostic kit of claim 23, further comprising standarized data, from at least one standard population of individuals, for statistical comparison with levels of IgM positive T-lymphocytes or IgM-bp positive T-lymphocytes measured in the sample using the first and second antibodies.

27. A method for augmenting a diagnosis of stroke, comprising i) providing a sample of a bodily fluid containing mononuclear lymphocytes from an individual, ii) determining the percentage of IgM-positive (IgM+) T-lymphocytes in the sample, and iii) comparing the percentage of IgM+ T-lymphocytes determined in the sample with a percentage of IgM+ T-lymphocytes from a standardized data set, wherein a statistically significant decrease in the percentage of IgM+ T-lymphocytes in the sample relative to the percentage of IgM+ T-lymphocytes in the a non-recent stroke population is indicative that the individual has suffered an acute stroke.

28. The method of claim 27, wherein the percentage of IgM+ T-lymphocytes in the sample is determined by an immunoassay technique.

29. The method of claim 28, wherein the immunoassay technique comprises i) admixing with the sample a first antibody which specifically binds a T- lymphocyte marker, and a second antibody which specifically binds an IgM class antibody, and ii) detecting the binding of each of the first and second antibodies, wherein the level of IgM+ T-lymphocytes is determined as the fraction of cells in the sample which bind both the first and second antibodies relative to the cell which bind only the first antibody.

30. The method of claim 29, wherein the T-lymphocyte marker is selected from the group consisting of CD2, CD3, CD4, CD5, CD7, CD8, CD28, and a T-lymphocyte receptor (TCR) protein.

31. The method of claim 29, wherein each of the first and second antibodies has attached thereto a fluorescent group able to be detected, and the binding of each of the antibodies is detected by fluorescence-activated cell sorting.

32. The method of claim 27, wherein the bodily fluid is blood plasma or serum.

33. The method of claim 27, wherein said statistically significant decrease in IgM+ T- lymphocytes is in the range of a 40-50 percent decrease in the percentage of IgM+ T- lymphocytes in the sample relative to the percentage of IgM+ T-lymphocytes in the a non-recent stroke population.

34. The method of claim 27, wherein said statistically significant decrease in IgM+ T- lymphocytes is in the range of a 75-80 percent decrease in the percentage of IgM+ T- lymphocytes in the sample relative to the percentage of IgM+ T-lymphocytes in the a non-stroke population.

35. A method for identifying an individual's of risk of having suffered from a recent stroke, comprising i) providing a sample of a biological fluid from the individual, the biological fluid comprising mononuclear lymphocytes, ii) admixing with the sample a first antibody which specifically binds a T- lymphocyte marker and a second antibody which specifically binds an IgM class antibody, iii) determining a percentage of IgM-positive (IgM+) T-lymphocytes in the sample by measuring the binding of each of the first and second antibodies via an immunoassay technique, and iv) , comparing the percentage of IgM+ T-lymphocytes in the sample with percentage of IgM+ T-lymphocytes measured in a non-stroke population of individuals, wherein a statistically significant decrease in the percentage of IgM+ T-lymphocytes in the sample relative to the percentage of IgM+ T-lymphocytes in the non-stroke population is indicative that the individual has recently suffered an acute stroke.

36. The method of claim 35, wherein the binding of at least one of the first and second antibodies is measured by a sandwich assay.

37. The method of claim 35, wherein at least one of the first and second antibodies has attached thereto a label group able to be detected, and the binding of the labeled antibody is measured by detecting the label group.

38. The method of claim 37, wherein the label group is selected from a group consisting of radioisotopes, fluorescent compounds, enzymes, and enzyme co-factors.

39. The method of claim 38, wherein the label group is a fluorescent group and the binding of the labeled antibody is detected by fluorescence-activated cell sorting.

40. The method of claim 35, wherein the first antibody binds a T-lymphocyte marker selected from the group consisting of CD2, CD3, CD4, CD5, CD7, CD8, CD28, and a T-lymphocyte receptor (TCR) protein.

41. The method of claim 35, wherein the second antibody binds a m-chain of an IgM class antibody.

42. The method of claim 35, wherein the second antibody binds one of either a k or 1 light chain of an antibody.