EA049359B1 - ANALYSIS OF BLOOD SAMPLES TO DETECT TAUPATHY OR AMYLOIDOGENIC DISEASE - Google Patents

Abstract

A method for detecting p217+tau protein in subject samples obtained from blood with high sensitivity, accuracy, and precision is proposed. The assay comprises contacting the sample with a capture antibody directed to the p217+tau epitope to bind the capture antibody to p217+tau peptides in plasma to form antibody-peptide complexes, and separately contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes. The amount of p217+tau protein is determined by detecting the detector antibody. The amount of detected p217+tau protein is used to determine the presence of a tauopathy in a subject or the subject's susceptibility to developing a tauopathy, or the presence of an amyloidogenic disease in a subject or the subject's susceptibility to developing an amyloidogenic disease when the amount of p217+tau peptides is above a predetermined threshold. The method has improved sensitivity such that the specified threshold value is higher than the lower limit of quantification and/or the lower detection limit of the assay.

Description

List of sequences

This application contains a sequence listing provided in electronic form in ASCII format and incorporated herein by reference in its entirety. A copy of the sequence listing in ASCII format, created on June 30, 2021, is named JAB7064WOPCT_SL.txt and is 20,602 bytes in size.

Cross-references to related applications

This application claims priority to U.S. Provisional Patent Application No. 62/705,759, filed July 14, 2020, and U.S. Provisional Patent Application No. 63/200,399, filed March 4, 2021, the entire contents of which are incorporated herein by reference.

Field of application of the invention

The present application relates to methods for detecting tauopathy and/or amyloidogenic disease. In particular, the invention of the present application relates to methods for measuring the amount of p217+ tau protein species phosphorylated at one or more sites in samples obtained from blood, and uses thereof.

Prerequisites for the creation of an invention

Alzheimer's disease (AD) is a degenerative brain disorder clinically characterized by progressive loss of memory, cognitive function, reasoning, decision-making, and emotional stability, gradually leading to profound mental impairment and ultimately death. AD is a very common cause of progressive mental decline (dementia) in older adults. In the United States, more than 5 million people live with AD, and this number is growing as the population ages. Indeed, AD affects 10% of people over the age of 65, and it is the 5th leading cause of death in this population. Overall, in the United States, AD is the 6th leading cause of death (1 in 3 older adults dies from AD or another type of dementia), and it is estimated to have cost the country US$305 billion in 2020. AD also occurs in different ethnic groups around the world and is and will remain a major public health problem today.

The brains of patients with AD have characteristic lesions called senile (or amyloid) plaques, amyloid angiopathy (amyloid deposits in blood vessels), and neurofibrillary tangles. Patients with AD typically have large numbers of these lesions, particularly amyloid plaques and neurofibrillary tangles made of paired helical filaments, in several areas of the human brain important for memory and cognition.

Neurofibrillary tangles consist primarily of aggregates of hyperphosphorylated tau protein. The major physiological function of tau is to polymerize and stabilize microtubules. Binding of tau to microtubules occurs through ionic interactions between the positive charges at the tau-microtubule binding site and the negative charges on the microtubule meshwork (Butner and Kirschner, J Cell Biol. 115(3):717-30, 1991). Tau contains 85 possible phosphorylation sites, and phosphorylation at many of these sites interferes with the primary function of tau. Tau protein associated with the axonal microtubule network is in a hypophosphorylated state, whereas aggregated tau protein in AD is characterized by hyperphosphorylation and the formation of unique epitopes that differ from the pool of physiologically active tau protein (Iqbal et al., Curr Alzheimer Res. 7(8): 656–664, 2010).

Progression of tauopathy in the AD brain occurs in different ways. A hypothesis of transmission and spread of tauopathy has been described based on the Braak stages of tauopathy in the human brain and tauopathy spreading after injection of tau aggregates in preclinical tauopathy models (Frost et al., J Biol Chem. 284:12845–52, 2009; Clavaguera et al., Nat Cell Biol. 11:909–13, 2009). It is believed that tauopathy may spread like prions from one brain region to another. This spreading process may involve externalization of tau aggregates, which may be taken up by nearby neurons and may induce further tauopathy.

Numerous biochemical changes can be detected up to 20 years before symptoms appear. The National Institute on Aging-Alzheimer's Association (NIA-AA) Research Framework proposes a framework for diagnosing Alzheimer's disease (AD) based on measurements of markers that are related to the underlying disease pathology, β-amyloid (A), abnormal forms of tau (T), and neurodegeneration (N). Positron emission tomography with tau-specific radiotracers (tau-PET) is used to measure tau pathology with neurofibrillary tangles (NFT) in patients. However, tau-PET is expensive and labor-intensive, and the availability of tau-specific radiotracers may be limited.

Tau fragments in neurofibrillary tangles are released into the cerebrospinal fluid (CSF), where they can be collected and measured using sensitive assays. The presence of neurological disease can therefore be determined using assays that detect tau-derived fragments in the CSF. However, to obtain CSF, patients must undergo invasive lumbar puncture procedures, in which a doctor inserts a needle into the spinal canal.

- 1 049359 for the collection of CSF samples for use in analysis. Such procedures are uncomfortable and unpleasant for the patient, so they are not desirable to repeat frequently and are not suitable for routine monitoring of the disease course in patients.

Brief summary of the invention

One embodiment of the present invention relates to an assay method for detecting p217+tau in a subject. The method comprises contacting a plasma sample with a capture antibody directed to a p217+tau epitope to bind the capture antibody to p217+tau peptides in the plasma to form antibody-peptide complexes; and washing the antibody-peptide complexes. The method further comprises contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes. The method then comprises detecting the detector antibody to determine the amount of p217+tau peptides in the plasma sample.

A method for detecting tauopathy in a subject is also provided. The method includes obtaining a plasma sample from a subject and detecting a certain amount of p217+tau peptides present in the plasma sample using an analytical method. The assay uses a capture antibody directed to p217+mаuepiton to bind the capture antibody to p217+tau peptides in the plasma to form antibody-peptide complexes and a detector antibody to bind the detector antibody to the antibody-peptide complexes. The method further includes a step of determining whether the subject has tauopathy or whether the subject is at risk of developing tauopathy when the amount of p217+mау peptides is above a predetermined threshold value. In this case, the predetermined threshold value exceeds the lower limit of quantification (LLOQ) of the assay.

A method for detecting an amyloidogenic disease in a subject is also provided. The method includes obtaining a plasma sample from a subject and detecting a certain amount of p217+tau peptides present in the plasma sample by an analytical method. The assay uses a capture antibody directed to a p217+tau epitope to bind the capture antibody to p217+mau peptides in the plasma to form antibody-peptide complexes and a detector antibody to bind the detector antibody to the antibody-peptide complexes. The method further includes a step of determining whether the subject has an amyloidogenic disease or whether the subject is at risk of developing an amyloidogenic disease when the amount of p217+mau peptides is above a predetermined threshold value. In this case, the predetermined threshold value exceeds the lower limit of quantification (LLOQ) of the assay.

In another aspect of the present application, there is provided a method for detecting or predicting tauopathy in a subject. The method comprises detecting an amount of p217+mau peptides in a plasma sample by contacting the plasma sample with a capture antibody directed to p217+tau to bind the capture antibody to p217+tau peptides in the plasma to form antibody-peptide complexes, separately contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes, and generating tau data corresponding to the amount of p217+mau peptides detected. The method also comprises obtaining biomarker data corresponding to at least one biomarker determined in the patient, wherein the biomarker is selected from the group consisting of NFL, adiponectin, and leptin. The method further comprises comparing the tau protein data and additional biomarker data with a set of reference data using a machine learning module to determine or predict whether the subject has a tauopathy or is at risk of developing a tauopathy.

These and other aspects of the invention will become apparent to those skilled in the art upon reading the following detailed description of the invention, including the figures and the appended claims.

Brief description of graphic materials

Fig. 1a shows the data for the determination of p217+ tau protein in plasma at a dilution of 1:4, obtained using the assay described above.

Fig. 1b shows the data for determining p217+mau protein in plasma at a dilution of 1:16, obtained using the analysis described above.

Fig. 1c shows the data for the determination of p217+ tau protein in semi-denatured plasma samples obtained using the analysis described above.

Fig. 1g shows the data for determining p217+τau protein in immunoprecipitated plasma samples obtained using the analysis described above.

Fig. 2a shows data demonstrating the correlation between the CSF p217+mau protein values obtained using the assay described above and the CSF p217+tau protein values obtained using an example assay of the present invention.

Fig. 2b shows data comparing the serum levels of p217+mau protein determined using the assay described above and the serum levels of p217+mau protein determined using an example assay of the present invention.

Fig. 3a shows data demonstrating the correlation between CSF p217+mau protein values obtained using two different detector antibodies according to the examples

- 2 049359 implementation of the present invention.

Fig. 3b shows data demonstrating the correlation between serum p217+ tau protein measurements obtained using two different detector antibodies in accordance with embodiments of the present invention.

Fig. 3c shows data demonstrating the correlation between plasma p217+tau protein measurements obtained using two different detector antibodies in accordance with embodiments of the present invention.

Fig. 4a shows data comparing the effect of different sample diluents on granule aggregation in embodiments of the present invention.

Fig. 4b shows data comparing the effects of different sample diluents on p217+ tau protein levels determined in accordance with embodiments of the present invention.

Fig. 5a shows data on the p217+ tau protein detected in serum, obtained in accordance with an example of implementing the present invention.

Fig. 5b shows data on the p217+ tau protein detected in plasma, obtained in accordance with an example of implementing the present invention.

Figure 5c shows data demonstrating the correlation between p217+tau protein detected in serum, as shown in Figure 5b, and p217+tau protein detected in plasma, as shown in Figure 5c.

Figure 5d shows data demonstrating the correlation between p217+tau protein detected in plasma, as shown in Figure 5c, and p217+tau protein detected in serum, as shown in Figure 5b.

Fig. 6a shows examples of calibration curves generated using a calibration peptide for an example of implementing the analysis of the present application.

Fig. 6b shows data demonstrating the linearity upon dilution in an example of the assay of the present invention in serum and plasma.

Fig. 7a shows data demonstrating the single assay precision in an example of the plasma assay of the present invention.

Fig. 7b shows data demonstrating the precision across assays in an example of the assay of the present invention in plasma.

Fig. 7c provides additional data demonstrating the single assay precision in an example of the plasma assay of the present invention.

Fig. 8a shows data demonstrating the correlation of p217+tau protein detected in CSF and p217+tau protein detected in plasma using an example assay of the present invention in subjects with AD.

Fig. 8b provides additional data demonstrating the correlation of p217+tau protein detected in CSF and p217+tau protein detected in plasma using an exemplary assay of the present invention in subjects with AD.

Fig. 9a provides additional data demonstrating the correlation of p217+tau detected in CSF and p217+tau detected in plasma using an exemplary assay of the present invention in a validation cohort.

Fig. 9b shows a receiver operating characteristic (ROC) curve for the data shown in Fig. 7b, demonstrating the sensitivity of plasma measurements obtained using an example assay of the present invention in detecting brain pathologies due to tauopathy.

Figure 10a shows data demonstrating the correlation of p217+ tau protein detected in CSF with tau protein accumulation in brain tissue detected by positron emission tomography (PET) imaging.

Figure 10b shows the ROC curve for the data shown in Figure 7b, demonstrating the sensitivity of CSF p217+tau measurements in detecting tauopathy-related brain pathologies.

Fig. 11a provides additional data demonstrating the correlation of p217+tau detected in CSF and p217+tau detected in plasma using an exemplary assay of the present invention in a validation cohort.

Figure 11b shows a subset of the data from Figure 11a for amyloid-positive patients with a CSF Αβ42/40 ratio < 0.089.

Figure 11c shows a subset of the data from Figure 11a for patients who were amyloid negative and had a CSF Αβ42/40 ratio > 0.089.

Fig. 12a shows data demonstrating the correlation of p181 tau protein detected in CSF and p217+ tau protein detected in plasma using an example assay of the present invention.

Figure 12b shows the ROC curve for the data shown in Figure 12c, demonstrating the sensitivity of plasma p217+tau measurements in detecting CSF p217+tau levels.

Fig. 12c shows the data shown in Fig. 11a with threshold values for p217+tau.

- 3,049,359 plasma protein and p217+ tau protein in CSF to identify patients with or at risk of developing tauopathy versus patients not at risk of developing tauopathy.

Figure 12d shows the ROC curve for the data shown in Figure 12d, demonstrating the sensitivity of plasma p217+tau measurements in detecting CSF p217+tau levels.

Figure 12d shows a subset of the data shown in Figure 12c for subjects with normal cognitive function.

Figure 12e shows the ROC curve for the data shown in Figure 12g, demonstrating the sensitivity of plasma p217+tau measurements in detecting abnormal CSF p217+tau levels.

Figure 12g shows a subset of the data shown in Figure 12c for subjects with mild to moderate dementia.

Figure 13a shows the ROC curve for the data shown in Figure 13b, demonstrating the sensitivity of plasma p217+tau measurements in detecting abnormal CSF Ae42/40 ratios.

Fig. 13b shows data demonstrating the correlation of the Ae42/40 ratio in CSF and p217+tau protein detected in plasma using an example assay of the present invention.

Figure 13c shows the ROC curve for the data shown in Figure 13d, demonstrating the sensitivity of plasma p217+tau measurements in detecting abnormal CSF Ae42/40 ratios.

Figure 13d shows a subset of the data shown in Figure 13a for subjects with normal cognitive functions.

Figure 13e shows the ROC curve for the data shown in Figure 13e, demonstrating the sensitivity of plasma p217+tau measurements in detecting abnormal CSF Ae42/40 ratios.

Figure 13e shows a subset of the data shown in Figure 13a for subjects with mild to moderate dementia.

Fig. 14a shows data demonstrating the correlation of p217+tau protein detected in CSF and p217+tau protein detected in crude plasma using an example assay of the present invention.

Fig. 14b shows data demonstrating the correlation of p217+tau protein detected in CSF and p217+tau protein detected in chemically extracted plasma using an example assay of the present invention.

Fig. 14c shows data demonstrating the correlation of p217+tau protein detected in CSF and p217+tau protein detected in semi-denatured plasma using an example assay of the present invention.

Fig. 15a shows data on the p217+ tau protein detected in semi-denatured plasma samples, obtained in accordance with an example of implementing the analysis of the present invention.

Fig. 15b shows examples of calibration curves generated using a calibration peptide for another embodiment of the assay of the present invention, in which samples are semi-denatured prior to measurement.

Figure 15c shows data demonstrating the single-assay precision in an example assay shown in Figure 9b, in which samples are semi-denatured prior to measurement.

Fig. 16a shows a ROC curve for a machine learning method for detecting brain pathologies caused by tauopathy using serum levels of p217+ tau protein as a biomarker element in accordance with an embodiment of the present invention.

Fig. 16b shows a ROC curve for a machine learning method for detecting brain pathologies caused by tauopathy using serum p217+ tau protein levels and neurofilament light chain (NFL) data as biomarker elements in accordance with an embodiment of the present invention.

Fig. 16c shows a ROC curve for a machine learning method for detecting brain pathologies caused by tauopathy using serum p217+ tau protein levels, NFL data and adiponectin as biomarker elements in accordance with an embodiment of the present invention.

Fig. 16g shows a ROC curve for a machine learning method for detecting brain pathologies caused by tauopathy using serum p217+ tau protein levels, NFL, adiponectin and leptin data as biomarker elements in accordance with an embodiment of the present invention.

Fig. 16d shows the ROC curve for the machine learning method for detecting brain pathologies caused by tauopathy using data on NFL, adiponectin and leptin as biomarker elements in accordance with an embodiment of the present invention.

- 4 049359

Detailed description of the invention

All technical and scientific terms used in this document, unless defined otherwise, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Otherwise, defined terms used in this document have the meanings set forth in the specification. All patents, published patent applications, and publications cited in this document are hereby incorporated by reference as if fully set forth herein. It should be noted that throughout this document and the appended claims, the use of the singular includes plural references unless the context clearly dictates otherwise.

Unless otherwise specified, any numerical values such as a concentration or concentration range described herein are to be understood as modified in all instances by the term about. Thus, a numerical value will generally include ±10% of the stated value. For example, a concentration of 1 mg/mL includes 0.9 mg/mL to 1.1 mg/mL. Similarly, a concentration range of 1% to 10% (w/v) includes 0.9% (w/v) to 11% (w/v). As used herein, the use of a numerical range expressly includes all possible subranges, all individual numerical values within that range, including whole numbers within such ranges, and fractional values, unless the context clearly indicates otherwise.

As used herein, the term antibody or immunoglobulin refers to a specific protein capable of binding to an antigen or a portion thereof. These terms are used herein in a broad sense and include immunoglobulin or antibody molecules, including polyclonal antibodies, monoclonal antibodies (including murine, human, human-adapted, humanized, and chimeric monoclonal antibodies), and antibody fragments.

In general, antibodies are proteins or peptide chains that exhibit binding specificity for a particular antigen. The structures of antibodies are well known. Immunoglobulins can be classified into five main classes, namely IgA, IgD, IgE, IgG and IgM, depending on the amino acid sequence of the heavy chain constant domain. IgA and IgG are further divided into the isotypes IgA1, IgA2, IgG1, IgG2, IgG3 and IgG4. Accordingly, the antibodies of the present invention can be from any of the five main classes or the corresponding subclasses. The antibodies of the present invention are preferably IgG1, IgG2, IgG3 or IgG4. The light chains of antibodies of any vertebrate species can be classified, depending on the amino acid sequences of their constant domains, into one of two clearly distinct types, namely kappa and lambda. Accordingly, the antibodies of the present invention can comprise a kappa or lambda light chain constant domain. According to specific embodiments, the antibodies of the present invention comprise heavy and/or light chain constant regions of murine antibodies or human antibodies.

In addition to the constant domains of the heavy and light chains, antibodies contain variable regions of the light and heavy chains. The variable region of an immunoglobulin light or heavy chain consists of a framework region interrupted by antigen-binding sites. Antigen-binding sites are defined using various terms and numbering schemes described below.

(i) Kabat: Complementarity-determining regions or CDRs are based on sequence variability (Wu and Kabat, J Exp Med. 132:211-50, 1970). The antigen-binding site essentially has three CDRs in each variable region (e.g., HCDR1, HCDR2, and HCDR3 in the variable region of the heavy chain (VH) and LCDR1, LCDR2, and LCDR3 in the variable region of the light chain (VL));

(ii) Chothia: The term hypervariable region, HVR, refers to the regions of the variable domain of an antibody that are hypervariable in structure as defined by Chothia and Lesk (Chothia and Lesk, J Mol Biol. 196:901-17, 1987). The antigen-binding site essentially has three hypervariable regions in each of the heavy chain variable regions VH (H1, H2, H3) and light chain variable regions VL (L1, L2, L3). The numbering systems as well as the annotation of CDRs and HVRs have been revised by Abhinandan and Martin (Abhinandan and Martin, Mol Immunol. 45:3832-9, 2008).

(iii) IMGT: Another definition of the regions forming the antigen-binding site was proposed by Lefranc (Lefranc et al., Dev Comp Immunol. 27:55–77, 2003), based on a comparison of the V domains of immunoglobulins and T-cell receptors. The international ImMunoGeneTics (IMGT) database (http://www_imgt_org) provides a standardized numbering and definition of these regions. The correspondence between the CDR, HVR, and IMGT delimitations is described in Lefranc et al., 2003, Id.;

(iv) The antigen-binding site can also be defined by the publication Specificity Determining Residue Usage (SDRU) (Almagro, Mol Recognit. 17:132-43, 2004), which lists as SDR the amino acid residues of immunoglobulin that are directly involved in contact with the antigen.

The terms framework or framework sequences represent the remaining sequences of the variable region of an antibody that are distinct from those defined as antigen-binding sites. Since a specific definition of an antigen-binding site can be formulated based on different features, as described above, a specific framework sequence

- 5 049359 ity depends on the definition of the antigen-binding site. Framework regions (FRs) are the more highly conserved regions of the variable domains. The variable domains of the native heavy and light chains each contain four FRs (FR1, FR2, FR3, and FR4, respectively), which essentially adopt a beta-sheet configuration connected by three hypervariable loops. The hypervariable loops in each chain are held together in close proximity by the FRs and, together with the hypervariable loops from the other chain, contribute to the formation of the antigen-binding site of the antibody. Structural analysis of antibodies has revealed a relationship between the sequence and the shape of the binding site formed by the complementarity-determining regions (Chothia et al., J. Mol. Biol. 227: 799–817, 1992; Tramontano et al., J. Mol. Biol. 215:175–182, 1990). Despite the high sequence variability, five of the six loops have only a small set of main-chain conformations, called canonical structures. These conformations are determined, first, by the length of the loops and, second, by the presence of key residues at specific positions in the loops and in the framework regions that determine the conformation through their packing, hydrogen bonding, or ability to adopt unusual main-chain conformations.

As used herein, the term antigen-binding fragment refers to a fragment of an antibody such as, for example, a diabody, Fab, Fab', F(ab') ₂ , Fv fragment, disulfide-stabilized Fv fragment (dsFv), (dsFv) ₂ , bispecific dsFv (dsFv-dsFv'), disulfide-stabilized diabody (ds-diabody), single chain antibody molecule (scFv), single domain antibody (sdab), scFv dimer (bivalent antibody), bispecific or multispecific antibody formed from a portion of an antibody containing one or more CDRs, camel single domain antibody, nanobody, domain antibody, bivalent domain antibody, or any other fragment of an antibody that binds to an antigen but does not contain the complete antibody structure. The antigen-binding fragment has the ability to bind to the same antigen that the parent antibody or parent antibody fragment binds. According to particular embodiments, the antigen-binding fragment comprises a light chain variable region, a light chain constant region, and an Fd segment of a heavy chain constant region. According to other particular embodiments, the antigen-binding fragment comprises Fab and F(ab').

As used herein, the term epitope refers to a site on an antigen to which an immunoglobulin, antibody, or antigen-binding fragment thereof specifically binds. Epitopes may be formed from either contiguous amino acids or non-contiguous amino acids that are positioned adjacent to one another as a result of tertiary folding of the protein. Epitopes formed from contiguous amino acids are generally preserved upon exposure to denaturing solvents, whereas the structure of epitopes formed as a result of tertiary folding is generally disrupted upon treatment with denaturing solvents. An epitope typically includes at least 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 amino acids in a unique spatial conformation. Methods for determining the spatial conformation of epitopes include, for example, X-ray crystallography and 2-dimensional nuclear magnetic resonance. See, for example, Epitope Mapping Protocols in Methods in Molecular Biology, Vol. 66, G. E. Morris, Ed. (1996).

As used herein, the term tau or tau protein refers to an abundant protein of the central and peripheral nervous system that has multiple isoforms. In the human central nervous system (CNS), there are six major isoforms of tau protein, ranging in length from 352 to 441 amino acids, due to alternative splicing (Hanger et al., Trends Mol Med. 15:1129, 2009). The isoforms differ from each other by the regulated incorporation of 0-2 N-terminal insertions and 3 or 4 tandemly arranged microtubule-binding repeats termed 0N3R, 1N3R, 2N3R, 0N4R, 1N4R, and 2N4R. As used herein, the term control tau protein refers to the tau isoform of SEQ ID NO: 1, which lacks phosphorylation and other post-translational modifications. As used herein, the term tau protein includes proteins that contain mutations, such as point mutations, fragments, insertions, deletions, and splice variants, of the full-length wild-type tau protein. The term tau protein also includes post-translational modifications of the amino acid sequence of tau protein. Post-transcriptional modification includes, but is not limited to, phosphorylation.

Unless otherwise stated, in this document, the numbering of amino acids in a tau protein or fragment thereof is given according to the amino acid sequence presented in SEQ ID NO: 1.

As used herein, the terms "pin+tau peptides," "pin+'rav," or "pin+tau protein" refer to human tau protein or a tau fragment that is phosphorylated at residue 217 (pT217) of tau protein and may or may not be further phosphorylated at additional residues, such as, for example, residue 212 (pT212) of tau protein, wherein the numbering of the positions corresponds to the numbering in SEQ ID NO: 1.

As used herein, the term p217+tau epitope refers to a tau epitope comprising at least one of phosphorylated T217 and phosphorylated T212, wherein the numbering of the positions corresponds to the numbering in SEQ ID NO: 1. Examples of a p217+tau epitope include, for example, a pT3 epitope. As used herein, the term

- 6 049359 epitope pT3 refers to an epitope comprising amino acids 210-220 of human tau protein, which is phosphorylated at residue 217 and may or may not be further phosphorylated at additional residues, such as, for example, residue 212, wherein the numbering of the positions corresponds to the numbering in SEQ ID NO: 1.

As used herein, each of the terms long p217+tau peptides, long p217+tau protein, long form of p217+tau peptides or long fragment of p217+tau peptides has the same meaning and refers to p217+tau peptides that comprise an epitope of p217+tau protein and an epitope comprising amino acid residues 7-20 of tau protein. Long p217+tau peptides according to embodiments of the present invention can have different lengths. For example, the N-terminus of a long fragment of p217+tau peptides can be the 1st, 2nd, 3rd, 4th, 5th, 6th or 7th amino acid residue of tau protein. In one example, long p217+tau peptides can comprise amino acid residues 7-220 of p217+tau protein.

As used herein, each of the terms short p217+tau peptides, short p217+tau protein, short form of p217+tau peptides or short fragment of p217+tau peptides has the same meaning and refers to p217+tau peptides that comprise an epitope of p217+tau protein and an epitope comprising amino acid residues 119-126 of tau protein, but do not comprise an epitope comprising amino acid residues 7-20 of tau protein. Short p217+tau peptides according to embodiments of the present invention may have different lengths. For example, the N-terminus of the short p217+tau peptides may be any of the amino acid residues that are located between the epitope comprising amino acid residues 7-20 of tau protein and the epitope comprising amino acid residues 119-126 of tau protein. In one example, short p217+tau peptides may contain amino acid residues 119-220 of the p217+tau protein.

As used herein, each of the terms long tau peptide, long tau protein, long form of tau peptide or long fragment of tau peptide has the same meaning and means a tau peptide that comprises an epitope of tau protein recognized by a phosphorylation-independent capture antibody and an epitope comprising amino acid residues 7-20 of tau protein. Long fragments of tau peptides according to embodiments of the present invention can have different lengths. For example, the N-terminus of a long fragment of tau peptides can be the 1st, 2nd, 3rd, 4th, 5th, 6th or 7th amino acid residue of tau protein.

As used herein, each of the terms short tau peptide, short tau protein, short form of tau peptide or short fragment of tau peptide has the same meaning and means a tau peptide that comprises an epitope of tau protein recognized by a phosphorylation-independent capture antibody and an epitope comprising amino acid residues 119-126 of tau protein, but does not include an epitope comprising amino acid residues 7-20 of tau protein. Short fragments of tau peptides according to embodiments of the present invention can have different lengths. For example, the N-terminus of a short tau peptide can be any of the amino acid residues that are located between the epitope comprising amino acid residues 7-20 of tau protein and the epitope comprising amino acid residues 119-126 of tau protein.

As used herein, the term "capture antibody" refers to an antibody that binds to an antigen of interest and is directly or indirectly bound to a solid support. Examples of solid supports include, but are not limited to, microparticles or beads, such as magnetic beads or paramagnetic beads. Examples of capture antibodies include, but are not limited to, a monoclonal antibody that binds to an epitope of p217+ tau protein.

According to embodiments of the present invention, the capture antibody can be a monoclonal antibody comprising the HCDR1, HCDR2, and HCDR3 regions of an immunoglobulin heavy chain having the polypeptide sequences of SEQ ID NOs: 23, 24, and 25, respectively, and the LCDR1, LCDR2, and LCDR3 regions of an immunoglobulin light chain having the polypeptide sequences of SEQ ID NOs: 26, 27, and 28, respectively. In particular embodiments, the capture antibody is pT3. As used herein, the term pT3 refers to an antibody that binds to p217+tau peptides and has the heavy chain variable region amino acid sequence of SEQ ID NO: 19 and the light chain variable region amino acid sequence of SEQ ID NO: 20. In one embodiment, the pT3 monoclonal antibody is expressed by a murine hybridoma. In another embodiment, the capture antibody is a humanized antibody having the heavy chain variable region amino acid sequence of SEQ ID NO: 21 and the light chain variable region amino acid sequence of SEQ ID NO: 22.

According to other embodiments of the present invention, the capture antibody may be a monoclonal antibody that binds to an epitope between amino acids 150 and 250 of tau, preferably amino acids 211-221 or amino acids 159-163 of human tau, regardless of phosphorylation, and the numbering of the positions corresponds to the numbering in SEQ ID NO: 1. In particular embodiments, the capture antibody is hT7. As used herein, the term hT7 refers to a publicly available ...

- 7 049359 clonal antibody that binds to an epitope comprising amino acids 159-163 of human tau protein, wherein the numbering of the positions corresponds to the numbering in SEQ ID NO: 1. The monoclonal antibody to hT7 is available commercially, for example, from ThermoFisher (e.g., catalog no.: MN1000).

As used herein, the term "detection antibody" refers to an antibody that binds to an antigen of interest and has a detectable label or is linked to a secondary detection system. Examples of detectable labels include, but are not limited to, various enzymes, prosthetic groups, fluorescent materials, luminescent materials, bioluminescent materials, and radioactive materials. Examples of detection antibodies include, but are not limited to, a monoclonal antibody that binds to a tau protein, preferably an epitope comprising amino acids 7-20 or 116-127 of human tau protein, wherein the numbering of the positions corresponds to the numbering in SEQ ID NO: 1. When using a monoclonal antibody that binds to a tau protein at an epitope comprising amino acids 7-20 as a detection antibody for captured p217+ tau peptides, long fragments of the tau protein are detected. Using a monoclonal antibody that binds to the tau protein at an epitope containing amino acids 116-127 as a detector antibody for captured p217+ tau peptides, both short and long fragments of tau protein are detected.

According to embodiments of the present invention, the detector antibody may be a monoclonal antibody comprising the HCDR1, HCDR2, and HCDR3 regions of an immunoglobulin heavy chain having the polypeptide sequences of SEQ ID NOs: 10, 11, and 12, respectively, and the LCDR1, LCDR2, and LCDR3 regions of an immunoglobulin light chain having the polypeptide sequences of SEQ ID NOs: 13, 14, and 15, respectively. In particular embodiments, the detector antibody is hT43. As used herein, the term hT43 refers to a monoclonal antibody that binds to an epitope comprising amino acids 7-20 of human tau protein, wherein the numbering of the positions corresponds to the numbering in SEQ ID NO: 1, and the antibody comprises, in the variable region of the heavy chain, the amino acid sequence of SEQ ID NO: 16 and the amino acid sequence of the variable region of the light chain of SEQ ID NO: 17.

According to another embodiment, the detector antibody may be a monoclonal antibody comprising the HCDR1, HCDR2 and HCDR3 regions of an immunoglobulin heavy chain having the polypeptide sequences of SEQ ID NOs, respectively: 2, 3 and 4, and the LCDR1, LCDR2 and LCDR3 regions of an immunoglobulin light chain having the polypeptide sequences of SEQ ID NOs, respectively: 5, 6 and 7. In another specific embodiment, the detector antibody is pT82. As used herein, the term pT82 refers to a monoclonal antibody that binds to an epitope comprising amino acids 119-126, preferably 117-127, of human tau protein, wherein the numbering of the positions corresponds to the numbering in SEQ ID NO: 1, and the antibody comprises in the variable region of the heavy chain the amino acid sequence of SEQ ID NO: 8 and the amino acid sequence of the variable region of the light chain SEQ ID NO: 9.

As used herein, the term "pT3-based assay" refers to an assay that uses the pT3 antibody as a capture antibody. As used herein, the term "pT3xhT43" refers to an assay that uses the pT3 antibody as a capture antibody and the hT43 antibody as a detection antibody. As used herein, the term "pT3xpT82" refers to an assay that uses the pT3 antibody as a capture antibody and the pT82 antibody as a detection antibody.

As used herein, the term hT7-based assay refers to assays that use the hT7 antibody as a capture antibody. As used herein, the term hT7xpT82 refers to assays that use the hT7 antibody as a capture antibody and the pT82 antibody as a detector antibody.

As used herein, the term "subject" refers to an animal, preferably a mammal. According to particular embodiments, the subject is a mammal, including mammals other than primates (e.g., a camel, a donkey, a zebra, a cow, a pig, a horse, a goat, a sheep, a cat, a dog, a rat, a rabbit, a guinea pig, a marmoset, or a mouse) or non-primates (e.g., a monkey, a chimpanzee, or a human). In particular embodiments, the subject is a human.

As used herein, the term tauopathy encompasses any neurodegenerative disorder that is associated with abnormal aggregation of tau protein in the brain. In addition to hereditary and sporadic AD, other examples of tauopathy include frontotemporal dementia with parkinsonism associated with chromosome 17 (FTDP-17), progressive supranuclear palsy, corticobasal degeneration, Pick's disease, progressive subcortical gliosis, neurofibrillary tangle-predominant dementia, diffuse neurofibrillary tangles with calcification, dementia with argyrophilic grains, amyotrophic lateral sclerosis-parkinsonism-dementia complex, Down syndrome, Gerstmann-Sträussler-Scheinker disease, Hallervorden-Spatz disease, myositis with

- 8 049359 inclusion bodies, Creutzfeldt-Jakob disease, multiple system atrophy, Niemann-Pick disease type C, cerebral amyloid angiopathy with prion proteins, subacute sclerosing panencephalitis, myotonic dystrophy, non-Guam motor neuron disease with neurofibrillary tangles, postencephalitic parkinsonism, and chronic traumatic encephalopathy such as dementia pugulistica (dementia of the boxer) (Morris et al., Neuron, 70:410-26, 2011).

As used herein, the term amyloidogenic disease includes any disease associated with (or caused by) the formation or deposition of insoluble amyloid fibrils. Examples of amyloidogenic diseases include, but are not limited to, systemic amyloidosis, Alzheimer's disease, maturity-onset diabetes, Parkinson's disease, Huntington's disease, frontotemporal dementia, and prion-associated transmissible spongiform encephalopathies (kuru and Creutzfeldt-Jakob disease in humans, scrapie and bovine spongiform encephalopathy (BSE) in sheep and cattle, respectively). The various amyloidogenic diseases are defined or characterized by the nature of the polypeptide component of the fibrils that are deposited. For example, in subjects or patients with Alzheimer's disease, the characteristic polypeptide component of the amyloid deposits is β-amyloid protein (e.g., wild-type β-amyloid protein, a variant thereof, or a truncated β-amyloid protein). Accordingly, Alzheimer's disease is an example of a disease characterized by Ae deposits or a disease associated with Αβ deposits, for example, in the brain of a subject or patient. The terms β-amyloid protein, β-amyloid peptide, β-amyloid, Ae, and 'Άβ-peptide are used interchangeably herein.

In this document, the terms determination, measurement, evaluation, and analysis are used interchangeably and include both quantitative and qualitative determination. These terms refer to any form of measurement and include determining the presence or absence of a characteristic, attribute, or feature. Evaluation may be relative or absolute. The term presence evaluation includes determining the amount of a substance present as well as determining the presence or absence of that substance.

As used herein, the term diagnosis refers to the detection of a disease or disorder or the determination of the stage or extent of a disease or disorder, such as a tauopathy or an amyloidogenic disease. Typically, diagnosis of a disease or disorder is based on the assessment of one or more factors and/or symptoms indicative of the disease. Diagnosis may be based on the presence, absence, or amount of a factor indicative of the presence or absence of the disease or condition, such as p217+ tau protein. Each factor or symptom that is considered indicative of a particular disease need not be specific to that particular disease, i.e., differential diagnoses are possible that may be based on a diagnostic factor or symptom. Similarly, it is possible that a factor or symptom indicative of a particular disease is present in a subject who does not have that disease. The term diagnostic also includes determining the therapeutic effect of a drug therapy, such as anti-p217+tau therapy, or predicting the nature of the response to a drug therapy, such as anti-p217+tau therapy. Diagnostic methods may be used independently or in combination with other diagnostic and/or staging methods known to those skilled in the art for a particular disease or disorder, such as Alzheimer's disease.

As used herein, the terms increase and decrease refer to differences in the amount of a particular biomarker in a sample compared to the level of a control sample or standard. For example, the amount of a particular peptide may be increased or decreased in samples from patients with a disease compared to a standard level. In one embodiment, the increase in level or decrease in level may be a difference between the levels of the biomarker present in the sample and in the control sample of at least about 1%, at least about 2%, at least about 3%, at least about 5%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 35%, at least about 40%, at least about 50%, at least about 60%, at least about 75%, at least about 80% or more. In one embodiment, the increase in level or decrease in level may be a statistically significant difference between the levels of the biomarker present in the sample and in the control sample. For example, the difference may be statistically significant if the measured level of the biomarker is outside of about 1.0 standard deviation, about 1.5 standard deviation, about 2.0 standard deviation, or about 2.5 standard deviation from the mean of any group of the control sample or reference. The reference or control sample may be, for example, a sample from a healthy subject or a sample taken from the same subject at an earlier time point, such as a time point before drug administration or at an earlier time point during a therapeutic regimen.

- 9 049359

As used herein, the term isolated means that a biological component (e.g., a nucleic acid, peptide, or protein) has been substantially separated, obtained separately, or purified from other biological components of the organism in which the component naturally occurs, i.e., from other chromosomal and extrachromosomal DNA, RNA, and proteins. Thus, nucleic acids, peptides, and proteins that have been isolated include nucleic acids and proteins purified by standard purification methods. Isolated nucleic acids, peptides, and proteins may be part of a composition and still be considered isolated if such composition is not part of the original environment of the nucleic acid, peptide, or protein. The term also includes nucleic acids, peptides, and proteins produced by recombinant expression in a host cell, as well as chemically synthesized nucleic acids.

An isolated antibody that binds to a tau protein or an isolated anti-tau antibody as used herein refers to an antibody that specifically binds to a tau protein and is substantially free of other antibodies with specificities for other antigens (e.g., an isolated anti-tau antibody is substantially free of antibodies that specifically bind antigens other than a tau protein). However, an isolated anti-tau detector antibody may have cross-reactivity with other related antigens, such as species antigens (e.g., species homologues of a tau protein).

As used herein, the term "specifically binds" or "specific binding" refers to the ability of an anti-tau antibody of the invention to bind to a predetermined target with a dissociation constant ( _KD ) of about 1 x ^10-6 M or stronger, such as about ¹ x ¹⁰⁷ M or less, about 1 x ^10-8 M or less, about 1 x ^10-9 M or less, about 1 x 10-10 M or less, about 1 x ^10-11 M or less, about 1 x ^10-12 M or less, or about 1 x ^10-13 M or less. _KD is the ratio of _KD to Ka (i.e., Kd/Ka) and is expressed in units of molar concentration (M). _KD values for antibodies can be determined using the methods of the art related herein. For example, the K _D value for an anti-tau antibody can be determined using surface plasmon resonance, such as using a biosensor system such as a Biacore® instrument, a Proteon instrument (BioRad), a KinExA instrument (Sapidyne), an ELISA, or competitive binding assays known to those skilled in the art. Typically, an anti-tau antibody binds to a predetermined target (i.e., tau protein) with a K _D that is at least ten times lower than its K D for a non-specific target, as measured by surface plasmon resonance using, for example, a Proteon instrument (BioRad). However, anti-tau antibodies that specifically bind to tau may have cross-reactivity with other related targets, such as the same pre-specified target but from a different species (homologues), such as mouse, rat, marmoset, dog, or pig.

As used herein, the term polynucleotide, which is synonymous with the term nucleic acid molecule, nucleotides, or nucleic acids, refers to any polyribonucleotide or polydeoxyribonucleotide, which may be unmodified RNA or DNA or modified RNA or DNA. Polynucleotides include, but are not limited to, single- and double-stranded DNA, DNA that is a mixture of single- and double-stranded regions, single- and double-stranded RNA, RNA that is a mixture of single- and double-stranded regions, fusion molecules containing DNA and RNA that may be single-stranded or, as a rule, double-stranded or are a mixture of single- and double-stranded regions. In addition, a polynucleotide refers to triple-stranded regions containing RNA or DNA, or both RNA and DNA. The term polynucleotide also includes DNA or RNA containing one or more modified bases, and DNA or RNA with backbones modified for stability or other purposes. Modified bases include, for example, tritylated bases and atypical bases such as inosine. DNA and RNA may be subject to a variety of modifications; therefore, the term polynucleotide encompasses chemically, enzymatically, or metabolically modified forms of polynucleotides typically found in nature, as well as chemical forms of DNA and RNA characteristic of viruses and cells. The term polynucleotide also encompasses relatively short chains of nucleic acids, often called oligonucleotides.

As used herein, the terms modulation, alleviation, treatment or cure include the prevention of a physical and/or mental condition, or the alleviation or relief of an established physical and/or mental condition once it has been identified, or the alleviation of the characteristic symptoms of such a condition.

As used in this document, the term accuracy refers to the degree to which the obtained value is close to the true value of the assay.

As used in this document, the term precision refers to the closeness of agreement among a set of measurement values obtained from multiple samplings of the same homogeneous sample for analysis.

As used in this document, the term sensitivity refers to the lowest

- 10 049359 concentration of analyte in a sample that can be measured by analysis with acceptable accuracy and precision.

The present invention provides assays and methods for detecting ρ217+τ^-^πι^οβ phosphorylated at one or more sites in samples obtained from blood, in particular in plasma. Blood sampling is fast and easy and provides a reduced risk of infection or other complications compared to lumbar puncture used to collect CSF. The assays and methods of the present invention allow for the measurement of p217+tau peptides in samples obtained from blood with significant sensitivity, precision and accuracy. Thus, the present application provides an improved method for measuring and/or monitoring p217+tau protein levels in subjects compared to CSF-based assays, minimizing the burden on subjects when collecting samples and thereby allowing for more frequent analysis and monitoring of changes in p217+tau protein levels, which is particularly required for monitoring and assessing a response to treatment. The sample used in the assays and methods of the present invention may be a blood, serum, or plasma sample. The sample is preferably a plasma sample. More preferably, the plasma sample has not been immunoprecipitated to concentrate the p217+ tau peptides contained therein. In one embodiment, the sample is a crude plasma sample.

The assays and methods of the present invention relate to the measurement of p217+tau peptides in samples obtained from blood using a capture antibody that binds to p217+tau peptides in the sample. The capture antibody is preferably immobilized on a solid phase such that the capture antibody selectively binds to p217+tau peptides present in the sample and immobilizes them on the solid phase. In a separate step, the captured p217+tau peptides are contacted with a tau detector antibody that is labeled with a reporter element that allows detection of the captured p217+tau protein molecules. The assays and methods described herein can be used for various diagnostic purposes, such as diagnosing AD, other tauopathies, other diseases characterized by Ae deposits, or other amyloidogenic diseases in a subject, monitoring the effectiveness of treatment, identifying a subject for whom treatment with antibodies to ρ217+τ^^λ^ is acceptable, pre-screening subjects for PET imaging and CSF assays for further detection of AD, other tauopathies, other diseases characterized by Ae deposits, or other amyloidogenic diseases, identifying subjects for inclusion in clinical trials related to AD, other tauopathies, other diseases characterized by Ae deposits, or other amyloidogenic diseases, etc.

In one embodiment, the assays and methods of the present invention include the steps of contacting a sample obtained from blood with a capture antibody directed to p217+tau to bind the capture antibody to p217+tau peptides in the sample to form antibody-peptide complexes; and then washing the antibody-peptide complexes. The antibody-peptide complexes can be washed with any suitable solution that does not block the assay, such as, for example, a buffer solution (e.g., phosphate-buffered saline (PBS)). The washed antibody-peptide complexes can then be contacted with a detector antibody to bind the detector antibody to the antibody-peptide complexes. The detector antibody is then detected to determine the amount of p217+tau peptides in the sample.

U.S. Patent No. 10,591,492 to Kolb et al. (hereinafter referred to as the Kolb '492 patent), which is incorporated herein by reference in its entirety, describes an assay for measuring p217+ tau peptides in biological samples. The biological samples may include cerebrospinal fluid (CSF), blood, or brain tissue homogenate. The Kolb '492 patent noted that measurements of tau protein in crude serum or plasma do not demonstrate ideal diagnostic performance, and sensitivity may suffer and matrix interference may occur. As demonstrated below, measurements in crude serum showed that the assay described in Kolb'492 was unable to provide sufficient sensitivity, as most measurements were below the lower limit of quantification (LLOQ) for the assay, which refers to the smallest amount of analyte that can be quantified with acceptable precision and accuracy, and it was shown that detection of the analyte was only possible after samples were immunoprecipitated followed by heat denaturation of the eluent. Example 3 below demonstrates that the pT3xpT82 assay described in Kolb'492, which combines the sample with both the capture antibody and the detector antibody in a single step in the same mixture, lacks sensitivity and linearity upon dilution when used to measure p217+ tau peptides in plasma.

In contrast, although the levels of p217+ tau peptides in blood-derived samples are substantially lower than in CSF, the assays and methods of the present invention were unexpectedly able to measure p217+ tau peptides in human serum and/or plasma samples with improved sensitivity without first concentrating the ρ217+ τ^-^π^οβ in the samples via immunoprecipitation prior to measurement. Immunoprecipitation is a labor-intensive and imprecise process. Thus, in

- 11 049359 The present application provides improved assays and methods that have improved sensitivity for measuring p217+ tau peptides without cumbersome pre-treatment of samples by immunoprecipitation prior to measurement. In particular, it was unexpectedly discovered that the separate steps of (1) binding a capture antibody to p217+ tau peptides present in serum and/or plasma samples to form antibody-peptide complexes and (2) binding the antibody-peptide complexes to a detector antibody can successfully reduce interference from other components (e.g., endogenously produced or exogenously administered interfering antibodies) in the sample such that the assay is sufficiently sensitive to detect p217+ tau peptides in serum and/or plasma. In one embodiment, the sample may be first contacted with a capture antibody to bind the capture antibody to the p217+ tau peptides in the sample, and then washed to remove any unbound components that may interfere with the assay. Following washing, the captured p217+ tau peptides are contacted with a detector antibody to bind the detector antibody to the captured p217+ tau peptides. Example 4 below further demonstrates that when the assay described in Kolb '492, in which the sample is combined with both the capture antibody and the detector antibody in a single step in the same mixture, is used to measure p217+ tau peptides in serum, an artifactual signal corresponding to interfering components is observed that is not present in the case of CSF. However, this artifactual signal is absent in p217+tau measurements obtained using an example method comprising separate steps of binding a capture antibody and a detector antibody to p217+tau peptides as described herein.

Furthermore, it was also unexpectedly found that the assays and methods of the present invention were unexpectedly more sensitive in measuring p217+ tau peptides in human plasma samples compared to measurements in human serum samples due to the unexpectedly elevated level of p217+ tau protein detectable in plasma, as further shown in Example 7 below. The assays and methods of the present invention are capable of measuring p217+ tau peptides and provide accurate and precise quantification results in both healthy subjects and subjects with or at risk of developing a tauopathy, and more particularly AD. The assays and methods of the present invention are also capable of measuring p217+ tau peptides and provide accurate and precise quantification results in healthy subjects and subjects with or at risk of developing an amyloidogenic disease, in particular subjects with dementia (e.g., mild to moderate dementia). In particular, the assays and methods measure p217+tau peptides in plasma samples from both groups of subjects at levels above the LLOQ of the assays of the present invention, demonstrating an acceptable and reliable level of sensitivity. The LLOQ of the assays can be within 15-25% of the coefficient of variation (CV) of the assay, within 15-20% of the CV, or preferably within 20% of the CV. Additionally, the p217+tau measurements obtained according to the assays and methods of the present invention in plasma samples from healthy subjects are numerically distinguishable from similar measurements made in subjects with AD. Thus, the assays and methods of the present invention provide accurate and precise measurements that can be used to distinguish healthy subjects from subjects with AD.

In one embodiment, the assays and methods of the present invention measure p217+tau protein in a plasma sample from a subject and subsequently determine that the subject has a tauopathy and/or an amyloidogenic disease or is at risk of developing a tauopathy and/or an amyloidogenic disease when the amount of p217+tau protein determined in the plasma sample is above a predetermined threshold. The predetermined threshold may be any suitable threshold for distinguishing those subjects who have a tauopathy and/or an amyloidogenic disease or are at risk of developing a tauopathy and/or an amyloidogenic disease from those subjects who are healthy and not at risk of developing a tauopathy and/or an amyloidogenic disease. A given cutoff value may be defined as the plasma p217+tau concentration that: distinguishes those patients who have higher levels of tau in the brain or brain regions measured by PET imaging from those who have lower levels; distinguishes those patients who have higher levels of tau (eg, phosphorylated tau such as p181 or p217+tau) in the CSF from those who have lower levels; distinguishes those patients who have higher levels of β-amyloid (eg, Αβ40 or Αβ42), eg, in CSF or plasma; distinguishes those patients who have a higher ratio of Αβ42 to Αβ40, eg, in CSF or plasma, from those who have a lower ratio; and distinguishes patients with normal cognition from those with dementia. Due to the unexpectedly higher sensitivity of the assays and methods of the present invention when using human plasma, the predetermined cutoff value is above the LLOQ of the assay, which provides a sensitive, quantifiable cutoff value for identifying those subjects who have a tauopathy and/or an amyloidogenic disease or are at risk of developing a tauopathy and/or an amyloidogenic disease and separating them from healthy subjects. In particular, the predetermined cutoff value is above the LLOQ and/or the lower limit

- 12 049359 the LLOD of the assay, which is the smallest amount of analyte that can be reliably detected. For example, the specified threshold may be at least 3, 4, 5, 7, or 10 times the LLOQ of the assay and/or at least 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, or 120 times the LLOD of the assay.

Subjects determined to have a tauopathy and/or an amyloidogenic disease or to be at risk for developing a tauopathy and/or an amyloidogenic disease may be referred for additional clinical testing, such as, for example, CSF sampling and/or PET imaging, to further evaluate brain pathologies in these subjects. In another embodiment, subjects determined to have a tauopathy and/or an amyloidogenic disease or to be at risk for developing a tauopathy and/or an amyloidogenic disease may be administered an active agent for treating cognitive decline or a tauopathy and/or an amyloidogenic disease, such as AD. Active agents for the treatment of tauopathy may include anti-tau antibodies, anti-p217+tau antibodies, anti-human tau small interfering RNA (siRNA), anti-p217+tau siRNA, cholinesterase inhibitors, N-methyl D-aspartate (NMDA) receptor antagonists, etc. Active agents for the treatment of amyloidogenic disease may include anti-amyloid antibodies, beta-secretase inhibitors, gamma-secretase inhibitors, anti-human β-amyloid small interfering RNA (siRNA), cholinesterase inhibitors, N-methyl D-aspartate (NMDA) receptor antagonists, etc.

In some embodiments, the predetermined threshold value may correspond to a baseline value or a value that is significantly greater than the baseline value. As used herein, the term significantly greater refers to a value that is greater than statistically significant by chance, and for which the p-value is 0.05 or less. A significantly greater value may be at least about 1%, 2%, 5%, or 10% higher than the value in healthy volunteers, with a p-value of less than 0.05, 0.04, 0.03, 0.01, 0.005, 0.001, etc. The baseline value may correspond to an average level in a population of healthy subjects. The baseline value may also correspond to an average value of previous levels determined in the same subject.

In one embodiment, the capture antibody is a monoclonal antibody directed to a p217+tau epitope, and the detector antibody is a monoclonal antibody directed to an epitope comprising amino acid residues 7-20 of the tau protein. In another embodiment, the capture antibody is a monoclonal antibody directed to a p217+tau epitope, and the detector antibody is a monoclonal antibody directed to an epitope comprising amino acid residues 119-126, preferably 116-127, of human tau protein. In one embodiment, the capture antibody is a monoclonal antibody comprising the HCDR1, HCDR2, and HCDR3 regions of an immunoglobulin heavy chain having the polypeptide sequences of SEQ ID NOs: 23, 24, and 25, respectively, and the LCDR1, LCDR2, and LCDR3 regions of an immunoglobulin light chain having the polypeptide sequences of SEQ ID NOs: 26, 27, and 28, respectively, and the detector antibody is a monoclonal antibody comprising the HCDR1, HCDR2, and HCDR3 regions of an immunoglobulin heavy chain having the polypeptide sequences of SEQ ID NOs: 10, 11, and 12, respectively, and the LCDR1, LCDR2, and LCDR3 regions of an immunoglobulin light chain having the polypeptide sequences of SEQ ID NOs: 13, 14, and 15, respectively. More particularly, the capture antibody is pT3 and/or the detector antibody is hT43.

According to an embodiment of the present invention, p217+tau peptides in a sample of interest are captured by a capture antibody directed to a p217+tau epitope. The captured p217+tau peptides, each comprising ρ217+τ^^ππτοπ, may have different lengths, which may be determined using detector antibodies that bind to different epitopes. For example, a detector antibody directed to an epitope containing amino acid residues 7-20 of tau protein can detect only captured p217+tau peptides or their fragments that still contain amino acid residues 7-20 of tau protein (long ρ217+τ^-^πτ^βι), and a detector antibody directed to an epitope containing amino acid residues 119-126 of tau protein can detect not only long ρ217+τ^-^πτ^βι but also short p217+tau peptides. The captured p217+tau peptides can be contacted with a detector antibody directed to an epitope comprising amino acid residues 7-20 or 116-127 of the tau protein, and thus the amount of long ρ217+τ^-^πτ^οβ or ρ217+τ^-^πτ^οβ (long and short p217+tau peptides) in the sample can be detected and measured. The amount of short ρ217+τ^-^πτ^οβ in the sample is calculated by subtracting the amount of long p217+tau peptides from the amount of ρ217+τ^-^πτ^οβ.

In view of the unexpectedly improved sensitivity values observed in the detection of p217+tau peptides in plasma using the assays and methods of the present invention, it is believed that the improved sensitivity values are equally applicable to the detection of short p217+tau peptides and/or long p217+tau peptides described above. Thus, in another embodiment of the present invention, the assays and methods of the present invention comprise measuring short p217+tau peptides and/or long p217+tau peptides in serum samples and/or

- 13 049359 plasma. In particular, the assays and methods of the present invention can measure short p217+tau peptides and/or long p217+tau peptides in human plasma with increased sensitivity. In a further embodiment, the assays and methods of the present invention can measure short p217+tau peptides and/or long p217+tau peptides in a plasma sample from a subject and subsequently determine that the subject has a tauopathy or is at risk of developing a tauopathy, wherein the amount of short p217+tau peptides and/or long p217+tau peptides determined in the plasma sample is above a predetermined threshold value(s). The predetermined threshold value(s) exceeds the LLOQ of the assays.

According to another embodiment of the present invention, in addition to capturing and measuring the amount of p217+ tau peptides in a sample, total tau peptides in the sample are captured with a phosphorylation-independent capture antibody, such as an antibody directed to an epitope between amino acids 150 and 250 of tau protein, preferably an epitope comprising amino acids 159-163 of tau protein. The captured total tau peptides can be contacted with a detector antibody directed against an epitope comprising amino acid residues 7-20 or 116-127 of tau protein, and thereby detecting and measuring the amount of long total tau peptides or total tau peptides (long and short fragments of tau peptides) in the sample. The amount of short total tau peptides in the sample is calculated by subtracting the amount of long total tau peptides from the amount of total tau peptides.

According to embodiments of the present invention, a value related to p217+tau peptides in a sample, such as the amount of p217+tau peptides and the amount of long p217+tau peptides, optionally the amount of total tau peptides and the amount of total long tau fragments in a sample, as well as data based on measured amounts, such as an estimated amount of short p217+tau peptides and short total tau peptides, or a ratio based on p217+tau peptides, such as the ratio of short tau peptide fragments to long tau peptide fragments, the ratio of short p217+tau peptides to total short tau fragments, the ratio of long p217+tau peptides to total long tau fragments, etc., can be used for one or more diagnostic purposes. In one embodiment, it is determined that the subject has a tauopathy if a ratio based on p217+tau peptides, such as the ratio of short p217+tau peptides to long p217+tau peptides, is significantly greater than the corresponding baseline ratio, and the measured amount of p217+tau peptides, short p217+tau peptides, and/or long p217+tau peptides exceeds the LLOQ of the assays.

In one embodiment, the method of the present invention comprises (i) contacting a sample obtained from blood, preferably a plasma sample, with a capture antibody directed to an epitope comprising phosphorylated p217+tau protein to capture p217+tau peptides in the sample, (ii) separately contacting the captured p217+tau peptides with a detector antibody directed to an epitope comprising amino acid residues 7-20, preferably after washing the captured p217+tau peptide, thereby measuring the amount of long p217+tau peptides, and/or with a detector antibody directed to an epitope comprising amino acid residues 119-126 of tau protein, preferably after washing the captured p217+tau peptide, thereby measuring the amount of long and short p217+tau peptides in the sample, and (iii) determining whether the patient has a tauopathy or the risk of developing tauopathy or not, based on the amount of p217+ tau peptides or the ratio of short p217+ tau peptides to long p217+ tau peptides. Diagnosis can be made by comparing the amount or concentration of p217+ tau peptides in a sample from a subject with the corresponding predetermined cutoff levels. Diagnosis can also be made by comparing the ratio of short p217+ tau peptides to long p217+ tau peptides in a sample from a subject with the corresponding baseline ratios, wherein the amounts of short p217+ tau peptides and long p217+ tau peptides are greater than the LLOQ of the corresponding assays.

In another embodiment, the method of the present invention comprises (i) contacting a sample obtained from blood, preferably a plasma sample, with a capture antibody directed to a p217+tau epitope to capture p217+tau peptides in the sample and/or with a phosphorylation-independent capture antibody directed to a tau epitope between amino acids 150 and 250 of the tau protein to capture total tau peptides in the sample, (ii) separately contacting the captured p217+tau peptides or the captured total tau peptides with a detector antibody directed to an epitope comprising amino acid residues 116-127 of the tau protein, preferably after washing the captured p217+tau peptides or the captured total tau peptides, thereby measuring the amount of long and short p217+tau peptides or the amount of total short tau peptides in the sample and (iii) determining whether the patient has a tauopathy or is at risk of developing a tauopathy based on the ratio of short p217+tau peptides to total short tau peptides in the sample. The diagnosis can be made by comparing the ratio of short p217+tau peptides to total short tau peptides containing the same region of the tau protein recognized by the anti-

- 14 049359 pT3 body, i.e. amino acids 211-221 of tau protein, in a sample taken from a subject with corresponding baseline values, and the number of short ρ217+τ^-^πι^οβ exceeds the LLOQ of the analysis.

In another embodiment, the method of the present invention comprises (i) contacting a sample obtained from blood, preferably a plasma sample, with a capture antibody directed to an epitope of p217+tau to capture p217+tau peptides in the sample, (ii) separately contacting the captured p217+tau peptides with a detector antibody directed to an epitope comprising amino acid residues 7-20, preferably after washing the captured p217+tau peptides, thereby measuring the amount of long p217+tau peptides, and/or with a detector antibody directed to an epitope comprising amino acid residues 116-127 of a tau protein, preferably after washing the captured p217+tau peptides, thereby measuring the amount of long and short p217+tau peptides in the sample, and (iii) determining the efficiency treatment of a subject based on the amount of p217+tau peptides or the ratio of short p217+tau peptides to long p217+tau peptides, wherein the measured amounts of p217+tau peptides, short p217+tau peptides and/or long ρ217+τ^-^π^οβ exceed the LLOQ of the respective assays.

In another embodiment, the method of the present invention comprises (i) contacting a sample obtained from blood, preferably a plasma sample, with a capture antibody directed to a p217+ tau epitope to capture ρ217+ τ^-^^^οβ in the sample and/or with a phosphorylation-independent capture antibody directed to an epitope of a tau protein between amino acids 150 and 250 of a tau protein to capture total tau peptides in the sample, (ii) separately contacting the captured ρ217+ τ^peptides or the captured total tau peptides with a detector antibody directed to an epitope comprising amino acid residues 116-127 of a tau protein, preferably after washing the captured p217+ tau peptides or the captured total tau peptides, thereby measuring the amount of long and short p217+tau peptides or the amount of total short tau peptides in the sample and (iii) determining the efficacy of treatment in a subject based on the amount or ratio of short p217+tau peptides to total short tau peptides in the biological sample, wherein the measured amount of short p217+tau peptides exceeds the LLOQ of the assay.

In another embodiment, the efficacy of treatment in a subject is determined by monitoring the amount of ρ217+τ^-^^^οβ, the ratio of short p217+tau peptides to long p217+tau peptides, or the ratio of short p217+tau peptides to total short tau peptides before, during, or after treatment, wherein the amount of p217+tau peptides, short p217+tau peptides, and/or long p217+tau peptides exceeds the LLOQ of the respective assays. A decrease in the values compared to the baseline values indicates a positive response to treatment. The values may also increase transiently in biological fluids as the half-life of pathological forms of tau protein increases from the circulation and/or as pathological forms of tau protein are cleared from the brain.

According to a particular aspect, tauopathy includes, but is not limited to, one or more conditions selected from the group consisting of Alzheimer's disease (including familial Alzheimer's disease and sporadic Alzheimer's disease), frontotemporal dementia with parkinsonism related to chromosome 17 (FTDP-17), progressive supranuclear palsy, corticobasal degeneration, Pick's disease, progressive subcortical gliosis, dementia with predominant neurofibrillary tangles, diffuse neurofibrillary tangles with calcifications, dementia with argyrophilic grains, amyotrophic lateral sclerosis-parkinsonism-dementia complex, Down syndrome, Gerstmann-Sträussler-Scheinker disease, Hallervorden-Spatz disease, inclusion body myositis, Creutzfeldt-Jakob disease, multiple systemic atrophy, Niemann-Pick disease type C, prion cerebral amyloid angiopathy, subacute sclerosing panencephalitis, myotonic dystrophy, non-Guam motor neuron disease with neurofibrillary tangles, postencephalitic parkinsonism, chronic traumatic encephalopathy, and dementia pugulistica (boxer's dementia).

Tauopathy is preferably Alzheimer's disease (including familial Alzheimer's disease and sporadic Alzheimer's disease), FTDP-17, and progressive supranuclear palsy.

Most commonly, tauopathy is Alzheimer's disease (including familial Alzheimer's disease and sporadic Alzheimer's disease).

According to one embodiment, the method of the present invention comprises (i) contacting a sample obtained from blood, preferably a plasma sample, with a capture antibody directed to a p217+tau epitope to capture p217+tau peptides in the sample, (ii) separately contacting the captured p217+tau peptides with a detector antibody directed to an epitope comprising amino acid residues 7-20, preferably after washing the captured p217+tau peptides, thereby measuring the amount of long p217+tau peptides and/or with a detector antibody directed to an epitope comprising amino acid residues 116-127 of a tau protein, preferably after washing the captured p217+tau peptides, thereby measuring the amount of long and short p217+tau peptides in the sample, and (iii) determining whether the subject is suitable for treatment antithesis

- 15 049359 whether the assay is p217+tau or not, based on the amount of p217+tau peptides or the ratio of short p217+tau peptides to long p217+tau peptides, where the amount of p217+tau peptides, short p217+tau peptides and/or long p217+tau peptides is above the LLOQ of the respective assays.

According to a particular aspect, it is determined that the subject is suitable for treatment with an antibody to p217+ tau protein if the amount of p217+ tau peptides in a sample obtained from blood, in particular in a plasma sample, or the ratio of the amount of short p217+ tau peptides to the amount of long p217+ tau peptides in a sample obtained from blood or a plasma sample is significantly higher than the corresponding baseline value, wherein the corresponding baseline value is higher than the LLOQ of assays for measuring p217+ tau peptides, short p217+ tau peptides and/or long p217+ tau peptides, or the amount of p217+ tau peptides, short p217+ tau peptides and/or long p217+ tau peptides is higher than the LLOQ of the corresponding assays.

According to another particular aspect, the method of the present invention comprises (i) contacting a sample obtained from blood, preferably a plasma sample, with a capture antibody directed to a p217+tau epitope to capture p217+tau peptides in the sample or with a phosphorylation-independent capture antibody directed to an epitope of a tau protein between amino acids 150 and 250 of a tau protein to capture total tau peptides in the sample, (ii) separately contacting the captured p217+tau peptides or the captured total tau peptides with a detector antibody directed to an epitope comprising amino acid residues 116-127 of a tau protein, preferably after washing the captured p217+tau peptides or the captured total tau peptides, thereby measuring the amount of long and short p217+tau peptides or the amount of total short tau peptides in the sample and (iii) determining whether the subject is suitable for treatment with an antibody to p217+ tau protein or not based on the ratio of the amount of short p217+ tau peptides to the amount of total short tau peptides in the biological sample, where the amount of short p217+ tau peptides exceeds the LLOQ of the assay.

According to one embodiment, a subject is determined to be suitable for treatment with an antibody to p217+ tau protein if the ratio of the amount of short p217+ tau peptides to the amount of total short tau peptides is significantly greater than the corresponding baseline value, wherein the amount of short p217+ tau peptides is above the LLOQ of the assay.

In one embodiment, the method of the present invention comprises (i) contacting a sample obtained from blood, preferably a plasma sample, with a capture antibody directed to an epitope comprising phosphorylated p217+tau protein to capture p217+tau peptides in the sample, (ii) separately contacting the captured p217+tau peptides with a detector antibody directed to an epitope comprising amino acid residues 7-20, preferably after washing the captured p217+tau peptide, thereby measuring the amount of long p217+tau peptides, and/or with a detector antibody directed to an epitope comprising amino acid residues 119-126 of tau protein, preferably after washing the captured p217+tau peptide, thereby measuring the amount of long and short p217+tau peptides in the sample, and (iii) determining whether the patient has tauopathy or the risk of developing tauopathy or not, based on the amount of p217+tau peptides. The diagnosis can be made by comparing the amount or concentration of p217+tau peptides in a sample taken from a subject with the corresponding predetermined cut-off levels, where the amount or concentration of p217+tau peptides is above the LLOQ of the assay used in steps (i) and (ii).

The present invention also relates to measuring the amount of p217+ tau protein that is complexed with an antibody in a sample obtained from blood, as well as the amount of free, non-antibody-bound p217+ tau protein in the sample. In one embodiment, total antibodies are captured using affinity processing methods followed by denaturing conditions, including chaotropes, heat inactivation, or other protein destruction methods. The p217+ tau protein is separated from the antibody using an rPHPLC system and measured using the methods of the present invention, which allows for the quantification of antibody-bound p217+ tau protein.

According to a general aspect, the invention relates to a method for monitoring treatment with an antibody to p217+ tau protein in a subject, the method comprising the steps of: the present application relates to a method for monitoring treatment with an antibody to p217+ tau protein in a subject, the method comprising the steps of: (i) obtaining a sample obtained from blood, in particular a plasma sample, from the subject, (ii) obtaining a semi-denatured sample from the sample obtained from blood containing total p217+ tau protein, (iii) contacting the semi-denatured sample with a capture antibody directed to a p217+ tau epitope to capture p217+ tau peptides in the semi-denatured sample, and (iv) separately contacting the captured p217+ tau peptides with a detector antibody directed to an epitope comprising amino acid residues 7-20, preferably after washing the captured p217+tau peptide, thereby measuring the amount of long p217+tau peptides and/or with a detector antibody directed to an epitope comprising amino acid residues 119-126 of the tau protein, preferably after washing the captured p217+tau peptide, thereby measuring the amount of

- 16,049,359 quantities of long and short p217+tau peptides in a semi-denatured sample, with the measured quantities of p217+tau peptides, short p217+tau peptides and/or long p217+tau peptides being above the LLOQ of the respective assays.

A semi-denatured sample is obtained from a sample obtained from blood containing p217+tau peptides by destroying antibodies and/or other blood components that block binding of a capture antibody or a detector antibody to p217+tau peptides or block detection of a detector antibody bound to p217+tau peptides, while p217+tau peptides present in the sample obtained from blood are not destroyed. In one embodiment, the semi-denatured sample is obtained by heating the sample obtained from blood at a predetermined temperature at which denaturation of antibodies occurs for a predetermined time. The predetermined temperature can be from 75°C to 100°C, from 80°C to 90°C, or 85°C. The predetermined time may be from 0.1 to 30 minutes, from 1 to 15 minutes, from 2 to 10 minutes, from 3 to 9 minutes, or 7 minutes. After heat denaturation, the sample may optionally be cooled to a temperature that is acceptably stable for the p217+tau peptide (e.g., at 4°C or below) to stop further degradation of proteins in the semi-denatured sample. In one embodiment, the semi-denatured sample is prepared by heating a sample obtained from blood at 85°C for 7 minutes and then cooling in an ice bath at 4°C for 10 minutes.

According to another general aspect, the present application relates to a method for monitoring treatment with an antibody to p217+ tau protein in a subject, the method comprising the steps of: (i) obtaining a sample obtained from blood, in particular a plasma sample, from the subject, (ii) obtaining a semi-denatured sample from the sample obtained from blood containing total ρ217+ tau, wherein the semi-denatured sample is heated to denature the antibodies in the sample, (iii) contacting the semi-denatured sample with a capture antibody directed to a p217+ tau epitope to capture ρ217+ tau peptides in the semi-denatured sample, (iv) separately contacting the captured p217+ tau peptides with a detector antibody directed to an epitope comprising amino acid residues 7-20, preferably after washing the captured p217+tau peptides, thereby measuring the amount of long p217+tau peptides, or with a detector antibody directed to an epitope comprising amino acid residues 116-127 of tau, preferably after washing the captured ρ217+tau peptides, thereby measuring the amount of long and short ρ217+λ^-^^^οβ in a semi-denatured sample, (v) calculating the amount of antibody-bound p217+tau protein in the sample by subtracting the amount of non-antibody-bound p217+tau protein from the amount of total p217+tau protein, (vi) calculating the ratio of antibody-bound p217+tau protein to the amount of non-antibody-bound p217+tau protein, and (vii) performing a control for treatment with an antibody to p217+tau protein in the subject based on the calculated ratio, In this case, the measured amounts of p217+tau peptides, short p217+tau peptides and/or long p217+tau peptides are above the LLOQ of the corresponding assays.

According to a particular aspect, the efficacy of treatment of a subject is determined by monitoring the amount of antibody-bound and non-antibody-bound p217+ tau peptides before, during or after treatment. A decrease in the amount of non-antibody-bound p217+ tau protein relative to baseline or an increase in the amount of antibody-bound p217+ tau protein relative to baseline and, therefore, an increase in the ratio of antibody-bound p217+ tau protein to non-antibody-bound p217+ tau protein relative to baseline signals a positive response to treatment. Non-antibody-bound ρ217+ λ^^λκοβ values may also increase transiently in blood-derived fluids, such as plasma, as the half-life of abnormal tau protein forms is increased from the circulation and/or as abnormal tau protein forms are cleared from the brain.

In a further aspect of the present invention, the assays and methods can be used to monitor p217+tau levels in a patient undergoing any treatment for a tauopathy, including but not limited to administration of exogenous anti-tau antibodies, more particularly anti-p217+tau antibodies, or any treatment for an amyloidogenic disease. Detection of p217+tau levels in samples obtained from blood, particularly plasma samples, can be used for a variety of different purposes, including use as a decision-making tool in determining whether to increase or decrease a dose level or dose interval to ensure that effective or safe levels of a drug are achieved or maintained; use as an adjunct tool in initiating anti-tau antibody therapy by providing evidence that trough pK levels have been achieved; and use as a tool to indicate that a patient should be excluded from or included in a clinical trial and as an aid in subsequent monitoring of patient compliance with the medicinal product within the clinical trial.

According to some aspects, the capture antibody in the method of the present invention is first bound to a solid support before being contacted with the sample. The capture antibody may be provided in a diagnostic kit for measuring p217+ tau protein in a sample obtained

- 17 049359 from blood (in particular, in a plasma sample), pre-bound to a solid phase, such as wells of a microtiter plate or magnetic beads. The detector antibody may contain any detectable label (e.g., a fluorescent molecule, biotin, etc.), detectable directly or detectable by a secondary reaction (e.g., a reaction with streptavidin), or be attached to it. In an alternative embodiment, a second reagent containing a detectable label can be used, wherein the second reagent has binding specificity for the primary antibody. In specific embodiments, the detector antibody is biotinylated.

In some aspects, the amount of p217+tau peptides measured in the methods of the present invention may be determined using any suitable methods known in the art, including solid-phase ELISA and single-molecule array platforms. In certain aspects, the methods of the present invention use a highly sensitive array platform, such as Quanterix Simoa or MSD S-plex, to measure the amount of p217+mau peptides in a sample obtained from blood (particularly a plasma sample) that has lower concentrations of p217+tau peptides than CSF.

In a further aspect of the present invention, the assays and methods of the present invention provide a bead-based assay for measuring p217+mau peptides in blood-derived samples (e.g., blood, serum, and/or plasma), wherein the assay has reduced interference from assay reagents and is therefore more accurate and precise. It has been discovered that certain assay reagents, when measuring p217+tau peptides in blood-derived samples, interact in a manner that interferes with the assay measurements. In particular, it has been discovered that in a bead-based assay, wherein the capture antibody is bound to a magnetic bead prior to contact with a blood-derived sample, the sample diluent used when preparing the blood sample adversely affects the accuracy and precision of the assay in blood-derived samples. For example, as shown in Example 7 below, the sample diluent from the Simoa Homebrew Assay Starter Kit, Cat. No. 101351 (available commercially from Quanterix) exhibits a reduction in bead counts, which is believed to be due to aggregation of the magnetic beads used as supports for the capture antibody. However, the assays and methods of the present invention employ sample diluents that reduce the interference caused by bead aggregation. In particular, the sample diluents of the present invention comprise a non-ionic surfactant. More specifically, the non-ionic surfactant comprises a hydrophilic chain of polyethylene oxide and/or a lipophilic or hydrophobic group of an aromatic hydrocarbon. In particular, the non-ionic surfactant is Triton X-100. The sample diluent may also comprise tris(hydroxymethyl)aminomethane (Tris), which has been found to further reduce interference with the assay. Example 7, described in detail below, demonstrates that sample diluents containing Triton X-100 reduced the observed interference, and that Tris-buffered saline-based sample diluents provided a greater reduction in interference compared to phosphate-buffered saline-based sample diluents. An example of the diluents of the present invention may further include other suitable components that do not interfere with the measurement of tau protein in blood-derived samples, such as NaCl, ethylenediaminetetraacetic acid (EDTA), a heterophile blocker, and/or bovine serum albumin.

In another general aspect, the present application relates to a kit for detecting p217+ tau protein in samples obtained from blood (e.g., blood, serum, plasma), comprising (a) a capture antibody directed to a p217+ tau epitope, optionally a phosphorylation-independent capture antibody directed to an epitope of tau protein between amino acids 150 and 250 of tau protein, (b) magnetic beads for conjugating the capture antibody thereto, (c) a sample diluent comprising a non-ionic surfactant, and (d) at least one detector antibody directed to an epitope of tau protein comprising amino acid residues 7-20 or 116-127 of tau protein. The kit is used to measure the amount of p217+tau peptides, the ratio of the amount of short p217+mau peptides to the amount of long p217+mau peptides and/or the ratio of the amount of short p217+mau peptides to the amount of total short tau peptides in a sample.

In another aspect of the present invention, the p217+tau measurements for blood-derived samples (e.g., blood, serum, plasma) obtained according to the assays and methods described above are further analyzed by a computer device to detect and/or predict a tauopathy in a subject. In particular, the p217+tau measurements for blood-derived samples are analyzed by a computer device in combination with data corresponding to measurements obtained for other biomarkers that can also be determined in blood-derived samples to provide a further improved method for detecting and/or predicting a tauopathy in a subject. The improved ability to detect and/or predict a tauopathy, in particular AD, using a biomarker(s) that can be suitably measured in blood-derived samples can be used for various diagnostic purposes, such as for diagnosing

- 18 049359 detection of AD or other tauopathies in a subject, monitoring the effectiveness of treatment, identifying a subject suitable for treatment with antibodies to tau protein or p217+ tau protein, preliminary screening of subjects for PET imaging and/or CSF analysis for the purpose of further detection of AD or other tauopathies, identifying subjects for inclusion in clinical trials associated with AD or other tauopathies, etc.

In one embodiment, a method for detecting or predicting tauopathy in a subject is provided. The method includes detecting an amount of p217+tau peptides in a sample obtained from blood (e.g., blood, serum, plasma) using an assay. The assay may be any of the assay examples described in the present application. In particular, the assay measures the amount of p217+tau peptides in samples obtained from blood, in particular plasma, by contacting the sample with a capture antibody that binds to p217+tau peptides in the sample and, in a separate step, by contacting the captured p217+tau peptides with a detector antibody to tau protein that is labeled with a reporter element that allows detection of the captured p217+tau protein molecules. In particular, the assay allows for the determination of the amount of p217+tau peptides in a plasma sample by contacting the plasma sample with a capture antibody directed to the p217+tau epitope to cause the capture antibody to bind to the p217+tau peptides in the plasma to form antibody-peptide complexes and separately by contacting the antibody-peptide complexes with a detector antibody to cause the detector antibody to bind to the antibody-peptide complexes.

The computing device receives measurements of p217+tau protein detected using the assay to generate tau data corresponding to the amount of p217+tau peptides. The tau data may represent the amount of p217+tau peptides determined using the assay. Alternatively, the tau data may represent a binary status (yes/no) indicating whether the amount exceeds a predetermined threshold. The assay is sufficiently sensitive, as described above, such that the predetermined threshold is above the LLOQ of the assay method. The computing device may also receive medical data of the subject, such as, for example, demographic information (e.g., age, gender), medical history, electronic medical records (EMR), pharmacy data corresponding to patient medical records, etc. In particular, the computing device may receive biomarker data corresponding to measurements or binary statuses of at least one biomarker determined in the patient. The biomarker may be an acceptable tauopathy biomarker. Preferably, the biomarker is detectable in samples obtained from blood, in particular in plasma samples, of the subject. For example, the biomarker may be selected from the group consisting of amyloid β (Ae), neurofilament light chain (NFL), adiponectin, leptin and other inflammatory or metabolic markers. More particularly, the biomarker is selected from NFL, adiponectin and leptin. The computer device analyzes tau protein data and biomarker data using a machine learning module to determine or predict whether the subject has a tauopathy or whether the subject is at risk of developing a tauopathy. The machine learning module is trained using a set of reference data. The machine learning module compares the tau protein data and biomarker data with the set of reference data to determine or predict whether the subject has a tauopathy or whether the subject is at risk of developing a tauopathy. The reference data set includes tau and biomarker data, as well as data relevant to brain pathology-tauopathy (e.g. disease stage, amount of p217+ tau detected in CSF, PET measurements of tau in brain tissue, etc.), for a reference group of patients.

The machine learning module may be a supervised and/or unsupervised machine learning module. The machine learning module may be a machine learning classifier for identifying a data set as correlated with one or two categories. The machine learning module may include a support vector machine, a random forest, a logistic regression, a gradient boosting module, or multiple modules thereof. In one embodiment, the machine learning module is a multiple module comprising at least one of a support vector machine, a random forest, a logistic regression, and/or a gradient boosting module.

It will be appreciated by those skilled in the art that the computer-implemented embodiments described herein may be implemented in any number of ways, including a single software module, a combination of hardware and software, or the like. For example, the example methods may be embodied in one or more programs stored on a non-volatile storage medium and comprising lines of code that, when compiled, may be executed by one or more processor cores or a single processor. A system in accordance with one embodiment comprises a plurality of processor cores and a set of instructions executable by the plurality of processor cores to implement the example methods described above. The processor cores or the single processor may be embedded in or in communication with any suitable electronic device, such as internal nodes.

- 19 049359 processing units in the device or with processing units external to the device, such as a mobile computing device, smartphone, computing tablet, computing device, etc., which can exchange data with at least part of the device.

Examples

The following examples are provided to further illustrate the nature of the invention. It should be understood that the following examples do not limit the invention and that the scope of the invention is defined by the appended claims.

Example 1. Highly sensitive assay for the detection of p217+ tau protein in plasma.

Example I provides an example of an improved assay of the present invention. In the example embodiment of Example I, a bead-based enzyme-linked immunosorbent assay (ELISA) is used to detect and/or quantify the presence of p217+ tau peptides in a sample. Specifically, Example I utilizes a Single Molecule Array (SiMoA) digital bead-based ELISA system purchased from Quanterix Corp. (Boston, MA). The SiMoA assay utilizes femtoliter reaction chamber arrays to digitally enumerate individual immunocomplexes. Assay-specific reagents were prepared as described in more detail below and loaded into the SiMoA analyzer to react with p217+ tau peptides to detect p217+ tau peptides in samples. The assay-specific reagents include 2.7 μm paramagnetic capture beads, buffers, and reagents from the Simoa Homebrew Assay Starter Kit, Cat. number 101351 (available from Quanterix), Wash Buffer 1 (available from Quanterix), magnetic beads (available as Simoa Homebrew Helper Bead Vial (918), Cat. No. 101732, from Quanterix), a capture antibody, and a detector antibody. The capture antibody in Example 1 is the mouse monoclonal antibody (mAb) pT3. The detector antibody in Example 1 is the mAb hT43.

Each sample analyzed in Example I is diluted with a sample diluent. Examples of sample diluents used in Example I include 50 mM Tris buffer, 100 mM NaCl, 5 mM EDTA, 2% (v/v) bovine serum albumin, and 0.5% (v/v) Triton X-100, heterophile blocking agent HBR-9 (available under Cat. No. 3KC564 from Scantibodies Laboratory, Santee, CA). The sample diluent has a pH of 7.4.

The assay in Example I was calibrated using a custom-made calibration peptide from New England Peptide. The calibration peptide is a peptide containing the hT43 and pT3 epitopes linked by a PEG4 linker and has a molecular weight of 4357 g/mol. The calibration peptide has the amino acid sequence of SEQ ID NO: 18.

Obtaining a reagent.

In the first step, paramagnetic capture beads were coated with 0.3 mg/mL of capture antibody, which in Example I is mAb pT3, following the protocol described in the Quanterix manual for attaching capture antibody to beads. The coated capture beads were diluted in the bead diluent buffer from the Simoa Homebrew Assay Starter Kit to 100,000 beads/mL and 300,000 beads/mL of auxiliary beads were added to give a total bead concentration of 400,000 beads/mL.

The detector antibody, which in Example I is mAb hT43, was biotinylated at 60x following the protocol described in the Quanterix manual and diluted in the sample diluent described above such that the detector solution had a detector antibody concentration of 0.9 μg/mL.

Streptavidin-β-galactosidase (SBG) concentrate was diluted to 200 pM in SBG diluent from the Simoa Homebrew Assay Starter Kit.

The calibration peptide was reconstituted to a concentration of 5 mg/mL in 0.1% phosphoric acid/water, divided into 20 µL aliquots, and frozen. When ready for use, aliquots of the calibration peptides were thawed and diluted 1:1000 (e.g., 1.5 µL in 1498.5 µL), and these diluted solutions were further diluted 1:1000 with sample diluent such that the final peptide concentration was 5000 pg/mL. The assay in Example I was calibrated using different concentrations of the calibration peptide, with a standard curve generated over the following concentration range: 30, 10, 3.33, 1.11, 0.37, 0.186, 0.093, 0.046, 0.023, 0.012, 0.006, and 0 pg/mL.

The plasma samples analyzed in Example I were diluted in sample diluent in a ratio of 1:2.

Analysis in the SiMoA system.

A customized assay was developed for the SiMoA system, including a three-step protocol. This three-step protocol includes the following steps: contacting the sample to be analyzed with the pT3 mAb attached to the capture beads prepared as described above with the sample to be analyzed for 35 minutes, then washing the capture beads with Tween-20 in phosphate-buffered saline (PBS), in particular Wash Buffer 1 developed for the Simoa HD-1 instrument (available from Quanterix), then incubating the capture beads with the detection antibody for 5 minutes, then washing the capture beads with Wash Buffer 1 a second time, incubating the capture beads with SBG for 5 minutes, washing the capture beads again

- 20 049359 zero with a PBS-based solution, specifically Wash Buffer 1, and finally adding 25 µl of 100 µM resorufin-b-O-galactopyranoside (RGP) before applying to the measuring disks for imaging and measurement on the SiMoA HD-1 instrument. Each reaction was performed in Simoa cuvettes and contained 25 µl of the bead solution containing the pT3 mAb attached to paramagnetic beads and auxiliary beads prepared as described above, 172 µl of the diluted sample or calibration sample, 100 µl of the detection solution containing the biotinylated detection antibody prepared as described above, 100 µl of the SBG solution and 25 µl of 100 µM RGP.

The SiMoA analyzer uses high-resolution fluorescence imaging to detect, on reaction chamber arrays in femtoliter volumes, the fraction of beads whose fluorescence emission corresponds to the fraction of beads bound to at least one enzyme and the fluorescence intensity in each reaction chamber. Based on these measurements, the SiMoA analyzer generates an output signal reflecting the average number of enzymes per bead (AEB).

Example 2. Detection of competing antibodies present in plasma.

Provided that additional pre-treatment of the samples is performed, the highly sensitive assay of Example 1 can be used to measure p217+ tau levels even in the presence of assay competing antibodies that compete with the assay reagents of Example 1, such as, for example, assay competing antibodies produced endogenously in the subject or assay competing antibodies exogenously administered to the subject. The method described in Example 2 herein can be used to analyze the pharmacodynamics of therapeutic antibodies to p217+ tau, such as humanized mAb pT3, present in plasma. For example, the method of Example 2 can be used to measure the effect of an active agent for treating a tauopathy, in particular a monoclonal antibody to p217+ tau (e.g., humanized mAb pT3), on peripheral tau levels present in a plasma sample obtained from a human individual.

An aliquot of the plasma sample was first diluted 1:3 in 0.1 M NaOAc and then heated at 85°C for 7 minutes, followed by cooling in an ice bath (4°C) for 10 minutes. The heated and then cooled sample fluid was centrifuged at 14,000 x g for 10 minutes at 4°C, and the supernatant was separated from the pellet. 1 M Tris base was added to the supernatant in an amount of 7% by volume to achieve a neutral pH of the supernatant, and a semi-denatured sample example was obtained. At the same time, a second aliquot of the plasma sample was cooled in an ice bath throughout the preparation of the semi-denatured sample fluid. The second aliquot in Example 2 is also referred to as the non-denatured sample fluid herein.

The output signal generated by the SiMoA analyzer for the semi-denatured fluid signal corresponds to the total amount of p217+ tau protein present in the plasma sample, while the output signal generated for the non-denatured fluid corresponds to the unbound p217+ tau protein in the plasma sample. The output signals are correlated with a standard curve constructed using a calibration peptide to determine the concentrations of total p217+ tau protein and unbound p217+ tau protein present in the plasma sample, respectively. By subtracting the latter value from the former, a quantitative indicator of the amount of unbound p217+ tau protein present in the plasma sample is obtained.

The precise heating times and temperatures for Example 2 were determined using CSF and exogenously added antibodies as a combination of heating times and temperatures sufficient to irreversibly modify any interfering antibodies in the sample such that they no longer interfere with the binding of antibodies to p217+tau in the assay, while not affecting the p217+tau signal itself. Specifically, the data obtained using CSF and exogenously added antibodies showed that the p217+tau signal was stable when heated at 85°C for at least 10 minutes, and that the antibodies were denatured after only 2 minutes of heating at 85°C. Thus, the results obtained using the semi-denatured fluid and the non-denatured fluid in Example 2 do not directly measure whether the capture antibodies bind to p217+tau protein, but rather demonstrate that there are competing antibodies in the plasma that interfere with the ability of the p217+tau assay to quantify p217+tau protein.

Example 3. The prior art method for analyzing p217+tau protein is not suitable for use with plasma.

As discussed above, the Kolb '492 patent previously described an assay for measuring p217+ tau peptides in biological samples. However, the Kolb '492 patent does not provide any examples of quantifying p217+ tau peptides present in human serum or plasma. Instead, the Kolb '492 patent states that unpurified serum or plasma samples are problematic due to interferences and cannot be measured and quantified with sufficient sensitivity without prior immunoprecipitation of the samples followed by thermal desorption.

- 21 049359 eluate turation.

Example 3 evaluates the assay described in Example 1 of the Kolb '492 patent using a set of human plasma samples obtained from 5 healthy volunteers (HV) as control subjects and 5 subjects diagnosed with Alzheimer's disease (AD). In this example, mAb pT3 is used as the capture antibody and mAb pT82 is used as the detector antibody. The data obtained according to Example 3 and generated by the SiMoA analyzer are shown in Figs. 1a-1d in AEU units. The left side of each graph shows the data corresponding to the subjects with AD, and the right side of each graph shows the data corresponding to HV. For each category of subjects, the mean is shown as the longer horizontal line with the ±standard deviation (SD) of the data set shown as the shorter lines above and below the mean line.

Each of these plasma samples was diluted according to the Kolb '492 patent to obtain two different dilutions, 1:4 and 1:16. The results obtained using both dilutions of the plasma samples are shown in Figs. 1a and 1b, respectively. The data shown in Fig. 1a demonstrate that at the 1:4 dilution, 40% of the plasma samples were below the LLOQ of the assay described in Example 1 of the Kolb '492 patent, while 20% were at the LLOQ and another 40% were significantly higher than all other samples. In particular, at the 1:4 dilution, 6/10 samples from Example 3 were at or below the LLOQ (AEB = signal = 0.035 = signal/noise (S/N) > 2 and CV < 20%), while the remaining 4 samples had a significantly higher signal (11-745x above the LLOQ). As shown in Fig. 1b, at the 1:16 dilution, all plasma samples, including the 40% that had significantly higher values at the 1:4 dilution, measured values below the LLOQ of the assay described in Example 1 of the Kolb '492 patent. More specifically, at the 1:16 dilution, all samples in Example 3 were below the LLOQ, indicating that the high signal at the 1:4 dilution in 4/10 subjects in Example 3 mentioned above was substantially non-linear and thus can be considered an artifact of the plasma matrix. The data shown in Figs. 1a and 1b demonstrate the lack of sensitivity and linearity of the assay at dilution for detecting p217+ tau peptides in plasma using the assay described in Example 1 of the Kolb '492 patent, and therefore such an assay is not suitable for determining the amount of p217+ tau peptides present in plasma.

A set of plasma samples from the same 5 DD and 5 BA subjects were denatured according to Example 2 above to generate 10 different semi-denatured sample fluids. The method for generating semi-denatured sample fluids described in Example 2 modifies the interfering antibodies such that they no longer interfere with the binding of anti-ρ217+τ^λ^ antibodies to ρ217+τ^ peptides in the sample without attenuating the signal from p217+ tau detected by the anti-ρ217+τ^λ^ antibodies. Each of the semi-denatured sample fluids was diluted 1:6 and measured using the assay described in Example 1 of the Kolb '492 patent. The results are shown in Fig. 1c. As can be seen in Fig. 1c, the method for producing semi-denatured sample fluids was able to eliminate all measurable signals in 40% of the samples that previously showed values significantly above the LLOQ at a 1:4 dilution without the denaturation step (see Fig. 1b). The method for producing semi-denatured sample fluids described in Example 2 does not result in attenuation of the ρ217+τ^ protein signal detected by the p217+tau antibody. Thus, the data shown in Fig. 1c indicate that the signal detected in plasma and presented in Fig. 1a may be contaminated by interference and/or artifacts from components other than p217+tau peptides. In particular, the high p217+tau signal in 4/10 plasma samples shown in Fig. 1a disappeared after denaturation, indicating that this signal is not a true tau signal. The 1:16 dilution data for crude plasma shown in Figs. 1a and 1b show a higher signal in AD subjects compared to DD subjects, demonstrating that matrix interference removal steps can release the p217+tau signal relevant to plasma biomarkers. However, the low sensitivity precludes the use of the assay described in Example 1 of the Kolb '492 patent as a viable method for measuring p217+tau in plasma samples.

The Kolb '492 patent acknowledges that the analysis of unpurified serum or plasma may be subject to decreased sensitivity and interference from matrix structures, and describes that an immunoprecipitation enrichment strategy can be used in combination with the assay described in Example 1 of the Kolb '492 patent to obtain measurements of abnormal forms of tau proteins in samples obtained from blood. It is further demonstrated that immunoprecipitation in combination with the assay described in Example 1 of the Kolb '492 patent provides improved sensitivity and allows discrimination of DD from AD subjects.

A set of plasma samples from the same 5 MD and 5 AD subjects were immunoprecipitated for p217+ tau signal using pT3 antibodies before measurement according to the assay described in Example 1 of Kolb '492. The immunoprecipitated samples were diluted 1:4 with sample diluent. The results obtained for these immunoprecipitated samples at the 1:4 dilution are shown in Fig. 1d. Comparison of the results shown in Fig. 1d with those in Fig. 1a, which were obtained at the same dilution ratio but without the initial immunoprecipitation with pT3 antibodies, shows that the immunoprecipitation steps resulted in improved

- 22 049359 sensitivity of the assay such that all values for plasma samples from AD subjects are within the linear range and clearly distinguishable from the PD samples. Figure 1d shows that the amount of p217+ tau peptides present in plasma obtained from PD samples is slightly above the LLOQ of the assay described in Example 1 of the Kolb '492 patent, but does not provide results greater than the LLOQ of the assay (e.g., the LLOQ is within the standard deviation range of the PD samples) for reliable quantification. These results indicate that purification and concentration of the p217+ tau signal in plasma may provide a useful assay for p217+ tau in plasma. However, immunoprecipitation is a labor-intensive and imprecise method that is unnecessarily cumbersome and may introduce additional uncertainty into the assay. The assays and methods of the present invention do not require a separate immunoprecipitation step to concentrate the signal from the p217+ tau protein in order to obtain results that are sufficiently sensitive to detect p217+ tau peptides in human plasma samples.

Example 4. Comparison of the previous assay with the highly sensitive assay of the present invention for the detection of p217+mау peptides.

The assays and methods of the present invention comprise separate steps for contacting a plasma sample with a capture antibody to bind the capture antibody to p217+mau peptides in the plasma to form antibody-peptide complexes and for contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes after washing, as explained in Example 1. The first step of contacting the plasma sample with the capture antibody and then washing the antibody-peptide complexes before contacting with the detector antibody allows for separation of the signal from the p217+mau protein from the influence of interfering components in the plasma sample. In particular, the p217+tau peptides in the sample are bound to the capture antibody, while the interfering components of the sample are washed out before the addition of the detector antibody to the antibody-peptide complexes. Assays and methods comprising these separate steps are also referred to herein as 3-step assays. In contrast, in the assay described in Example 1 of the Kolb '492 patent, the biological sample is combined with both the capture antibody and the detector antibody, allowing both antibodies to bind to the biological sample before the wash step. The assay described in Example 1 of the Kolb '492 patent is also referred to herein as a 2-step assay. This example also increased the incubation time and sample loading volume in the first step of the 3-step assay compared to the 2-step assay to ensure maximum signal capture. As further demonstrated in Example 4 below, it was unexpectedly found that the 3-step assay of the present invention provides improved sensitivity compared to the 2-step assay for measuring p217+mаy peptides in serum, which is not observed when measurements are made in CSF.

Example 4 compares the 3-stage assay of the present invention with a 2-stage assay in measuring p217+ tau peptides in CSF and serum. Measurements were made on a set of CSF samples from multiple subjects with different cognitive function statuses using both the 2-stage and 3-stage assays. Specifically, 96 CSF samples from 21 subjects with mild to moderate dementia in a clinical study were measured using both the 2-stage and 3-stage assays. In Fig. 2a, for each sample, the results (in pg/mL of p217+ tau detected) obtained using the 2-stage assay (plotted on the x-axis) are plotted against the result obtained using the 3-stage assay (plotted on the y-axis). In addition, measurements were performed using both the 2-stage and 3-stage assays on a set of serum samples from the same 5 DD and 5 BA subjects as in Example 3. The results (in pg/mL of detected p217+mау protein) are presented in the bar graph in Fig. 2b, where the left bar for each sample corresponds to the result obtained using the 2-stage assay and the right bar corresponds to the result obtained using the 3-stage assay.

As shown in Fig. 2a, the 2-stage and 3-stage assays showed a high correlation (r2=0.94) in measuring p217+ tau peptides in the CSF sample. However, Fig. 2b shows that no such correlation was observed when the assays were used to measure p217+ tau peptides in serum samples. Furthermore, while Fig. 1a shows a subset of samples with measured values that were much higher than the rest of the samples when the 2-stage assay was used to measure p217+ tau peptides in plasma, Fig. 2b shows that the high signals of these outliers were significantly reduced when the 3-stage assay was used to measure p217+ tau peptides in serum. The data in Fig. 2b show that the high p217+mau protein signal outliers observed in the 2-step assay are not similarly observed when the same samples are measured using the 3-step assay. Thus, the data in Fig. 2b demonstrate that the high p217+mau protein signal outliers observed in the 2-step assay are due to interference and/or assay artifacts and are not accurate measures of the amount of p217+mau peptide present in plasma. These data indicate that there is negligible matrix interference in the CSF but significant positive interference in blood-derived products.

- 23 049359

Example 5. Comparison of detector antibodies using the highly sensitive assay of the present invention for the detection of p217+ tau peptides.

Example 5 evaluates the compliance and relative fragmentation in three types of biological fluids, i.e., CSF, serum, and plasma, using two different detector antibodies and the example assay described in Example 1. Specifically, Example 5 compares the difference between hT43 and pT82 as the detector antibodies used in the assay of Example I (except that different detector antibodies are used, as indicated in Example 5). In both embodiments, the capture antibody is pT3. Thus, the two example assays compared in Example 5 are pT3xhT43 and pT3xpT82.

In CSF, serum, and plasma from 18 subjects with AD (specifically, mild to moderate dementia) in a clinical study, p217+ tau was measured using the assay of Example 1 and a modified assay similar to that described in Example 1 except that the detector antibody was replaced with mAb hT43. The results (in pg/mL p217+ tau detected) are presented in Figs. 3a, 3b, and 3c for CSF, serum, and plasma, respectively. For each type of biological fluid, the results (in pg/mL p217+ tau detected) obtained using the pT3xhT43 assay (plotted on the x-axis) are plotted against the result obtained using the pT3xpT82 assay (plotted on the y-axis) in Figs. 3a, 3b, and 3c for CSF, serum, and plasma, respectively. In each of Figs. 3a-3c show the linear regression line along with the R ² value for this linear regression line. As shown in Fig. 3a-3c, in all three types of samples, the results obtained using the hT43 detector antibody and the pT82 antibody showed a high agreement (R 2 = 0.82-0.95). The slopes of the linear regression lines shown in Fig. 3a-3c were 2.83, 2.41, and 2.05, respectively. Additionally, the results obtained with pT3xpT82 were ~2.5-fold higher than those obtained with pT3xhT43 in CSF, serum, and plasma. As discussed above, pT3 recognizes an epitope at amino acids 210-220 of human tau protein. The hT43 detector antibody recognizes amino acids 7-20 of human tau, and the pT82 detector antibody recognizes amino acids 116-127 of human tau. Thus, pT82 recognizes an epitope located closer to pT3 than the epitope recognized by the hT43 antibody, which allows pT82 to recognize short fragments of p217+ tau peptides in addition to long fragments of p217+ tau peptides, compared to hT43. The pT3xpT82 assay is thought to give higher concentrations than the pT3xhT43 assay because tau protein in CSF is known to be highly fragmented, and pT82 is able to detect shorter (and therefore more abundant) fragments of p217+ tau peptides than hT43. It is interesting to note that similarly higher levels with the pT3xpT82 assay are also observed in serum and plasma, indicating that the pattern of tau fragmentation in crude samples (between amino acids 20 and 116) in blood components (e.g., serum and plasma) may be similar to that in CSF and that the higher degree of fragmentation between the hT43 and pT82 epitopes is not present in blood products. Since the pT3xhT43 and pT3xpT82 assays have been shown to give highly consistent results for CSF, serum, and plasma, indicating that the level of p217+ tau fragmentation reflects the fragmentation in CSF, the two assays evaluated in Example 5 are considered interchangeable. However, the pT3xhT43 assay provides higher sensitivity than the pT3xhT43 assay.

Example 6. Comparison of different sample diluents for use in an assay for the detection of p217+ tau peptides.

In Example 6, various sample diluents for use in the assay of Example I (except that different sample diluents are used as described in Example 6) are evaluated to determine which diluent will reduce bead aggregation and/or the amount of artifact in the p217+ tau signal obtained by the assay. The effect of buffer type (PBS versus Tris), NaCl concentration, and the presence of a heterophilic blocker (e.g., a human antibody-mouse antibody blocker) are measured in terms of the contribution to the artifact signal from p217+ tau in serum and the effect on the amount of beads determined at the end of the ELISA method. The various sample diluents evaluated in Example 6 are listed below in Table 1.

- 24 049359

Table 1

Sample diluent Composition 1 Sample Diluent from the Simoa Homebrew Assay Starter Kit 2 50 mM phosphate buffer, 50 mM NaCl and 0.5% (v/v) Triton X-100 3 50 mM phosphate buffer, 100 MM NaCl and 0.5% (v/v) Triton X-100 4 50 mM Tris buffer, 50 mM NaCl and 0.5% (v/v) Triton X-100 5 50 mM Tris buffer, 100 mM NaCl and 0.5% (v/v) Triton X-100 6 50 mM phosphate buffer, 50 mM NaCl, 0.5% (v/v) Triton X-100 and HBR9 7 50 mM phosphate buffer, 100 mM NaCl, 0.5% (v/v) Triton X-100 hHBR-9 8 50 mM Tris buffer, 50 mM NaCl, 0.5% (v/v) Triton X-100 and HBR-9 9 50 mM Tris buffer, 100 mM NaCl, 0.5% (v/v) Triton X-100 and HBR-9

Although not listed in Table 1, each sample diluent also contains 5 mM EDTA and 2% (v/v) bovine serum albumin.

Three pooled serum samples obtained from human individuals with high tau levels and a 1 pg/mL calibration peptide solution (as a negative control) were analyzed using a modified assay similar to that described in Example 1 with different sample diluents as described herein in Example 6. The SiMoA analyzer was used to determine the number of beads loaded onto the SiMoA disk and the AEB value for each sample and diluent combination. Figure 4a shows the number of beads loaded onto the SiMoA disk for the three serum samples and the calibration peptide in each of the 9 different sample diluents. Figure 4b shows the fluorescent signals detected by the SiMoA analyzer in AEB units for the three serum samples and the calibration peptide in each of the 9 different sample diluents. Figure 4a and 4b the data are presented as bar graphs for each sample measured in the different sample diluents, which are shown in the same order as they are listed in Table 1.

As can be seen in Fig. 4a, Sample Diluent 1, which is the sample diluent from the Simoa Homebrew Assay Starter Kit, provides significantly lower bead counts in all three serum samples compared to the calibration peptide solution (1688-2921 vs. 5378). Sample Diluents 2-9 provide better bead counts in serum and significantly reduce the artifactual signal observed in two of the samples analyzed using the Simoa Homebrew Diluent (Sample Diluent 1). This observed decrease in bead counts in serum is believed to be caused by the aggregation of the paramagnetic beads used as supports for the capture antibodies in the assay when the assay is used to study serum samples. Aggregation of paramagnetic beads is undesirable since it negatively affects the accuracy and precision of the assay for the detection of p217+ tau in serum. The data in Fig. 4a demonstrate that sample diluents 2-9, all of which contain the detergent Triton X-100, provide increased bead loads and, therefore, reduced bead aggregation, which would otherwise have a negative impact on the accuracy and precision of the assay. Furthermore, Fig. 4a shows that the Tris-based sample diluents provided higher bead loads than the phosphate-based buffers, demonstrating that Tris-based buffers are useful in reducing the interference with the accuracy and precision of the assay caused by bead aggregation. As can be seen in Fig. 4b, all sample diluents provided similar levels of fluorescent signals for the calibration peptide solution. However, Sample Diluent 1 also resulted in the detection of significant levels of fluorescent signals that were not detected with the other sample diluents (although Sample Diluents 2, 3, and 5 also allowed some fluorescent signals to be detected). These increased fluorescent signals observed with Sample Diluent 1 are believed to be due to interference and/or assay artifacts, similar to what was demonstrated with the 2-step assay in Figure 2b, and are not an accurate measure of the p217+ tau peptides present in the serum samples. Figure 4b also shows that the addition of HBR-9, which is a heterophile blocker designed to reduce the binding of anti-mouse IgG antibodies to the assay reagents, further reduces blocking and/or assay artifacts. In particular, as seen in Figure 4b, sample diluents 6-9 showed lower fluorescent signals than sample diluents 2-5 in one of the serum samples, indicating that the addition of HBR-9 provides additional reduction in interference and/or assay artifacts. The data in Figs. 4a and 4b show that of the sample diluents evaluated in Example 4, sample diluent 9, which is the same sample diluent described in Example 1, provided both minimal bead aggregation and minimal artifact signal. Given the data discussed above, particular improvement was observed with the use of Tris buffer, lower NaCl concentration, and heterophile blocker.

Example 7. Comparison of p217+tau peptide detection in serum and plasma.

Example 7 evaluates the detection of p217+ tau peptides in serum and plasma using the exemplary assay described in Example 1. In a set of serum samples from 10 3D (also referred to as in

- 25 049359 healthy volunteers and healthy controls in this document) and the same 16 AD subjects as in Example 5 were measured using the example assay described in Example 1; the results are shown in Fig. 5a. The samples from the DV subjects were obtained from a blood collection service and the donors were assumed to have normal cognitive function. A set of plasma samples from a subset of 12 DV and 18 AD subjects from the subject group shown in Fig. 5a were measured using the example assay described in Example 1; the results are shown in Fig. 5b. In Figs. 5a and 5b, the data corresponding to the DV subjects are shown on the left side of each graph and the data corresponding to the AD subjects are shown on the right side of each graph. For each subject category, the mean is shown as the longer horizontal line with the ±standard deviation (SD) of the data set shown as the shorter lines above and below the mean line. Also included in each of Figs. 5a and 5b is a dotted line showing the LLOQ of the assay for serum and plasma, respectively. The data generated according to Example 7 are shown in Figs. 5a and 5b in pg/mL. Additionally, for each subject for whom both serum and plasma samples are presented in Figs. 5a and 5b (i.e., 10 BD and 16 AD subjects), in Figs. 5c and 5d, the results (in pg/mL detected p217+ tau) obtained using plasma from the subject are plotted against the results obtained using serum from the same subject. As can be seen in Fig. 5a and 5b, the values in both serum and plasma samples from AD subjects were significantly higher than those in serum and plasma samples from DD subjects. These data indicate that both serum and plasma can be used for diagnostic purposes. Surprisingly, the concentration values obtained for plasma were ~2-3 times (specifically, 2.3 times) higher than those obtained for serum, which was determined by calculating the mean ratios of plasma p217+ tau to serum p217+ tau determined for all subjects. However, as shown in Fig. 5d, the linear regression line fitted to the data exhibits a slope of 1.9, indicating that the values obtained for plasma are 1.9 times higher than those obtained for serum. Because the detectable level of p217+ tau analyte in serum is low, many serum samples from DD subjects were below the assay LLOQ. However, because p217+ tau detection is higher in plasma, all plasma samples from DD and AD patients were at or above the assay LLOQ. Therefore, the exemplary assay described in Example 1 can be used to detect and quantify p217+ tau peptides in plasma from both DD and AD subjects. Additionally, as can be seen in Fig. 5b, the range of p217+ tau peptides detected in DD subjects does not overlap significantly with the range of p217+ tau peptides detected in AD subjects. All serum samples from DD patients were below the assay LLOQ (shown in Fig. 5a), but most (11 of 12) plasma samples from the same DD patients were above the assay LLOQ and in the linear range (shown in Fig. 5b). Plasma and serum p217+ tau concentrations were well correlated (r2=0.82), but plasma measurements were on average ~1.9-fold higher than serum measurements, as shown in Fig. 5c. Thus, it was unexpectedly found that the assay of the present invention provides quantitative data that can be used to distinguish DD subjects from AD subjects. In particular, a subject may be determined to have a tauopathy or to be at risk of developing a tauopathy, particularly Alzheimer's disease, when the amount of p217+ taupeptides detected in plasma is above a predetermined threshold (e.g., ~0.1 pg/mL based on the data shown in Fig. 5b).

Example 8. Linear range of the assay for detection of p217+ tau protein in plasma.

Linear range for calibration material.

The calibration peptides described in Example 1 were prepared. The calibration peptides contained core epitopes for the pT3 and hT43 antibodies separated by a PEG4 linker and were used to generate standard curves corresponding to the ratio between the AEB output signals from the SiMoA analyzer and the concentrations of the calibration peptides. Examples of standard curves obtained in 5 separate runs of the analysis of different dilutions of the calibration peptides as described in Example 1 are shown in Fig. 6a. The 4-parameter curve fitting (4PL, 1/y2 weighted) data preprocessing method was used to generate the calibration curves. The lower limit of detection (LLOD) of the assay example from Example 1 was defined as the calculated calibration level giving an AEB value equal to the mean of the zero calibrator + 2.5 standard deviations (SD), including a 10% coefficient of variation (CV). By these criteria, representative data showed an LLOD of ~0.002 pg/mL. The linear range of the assay between the LLOQ and the upper limit of quantification (ULOQ) was defined as the interval between the minimum and maximum points of the standard curve that achieved a CV of <20% and a recovery of 80-120% of expected. By these criteria, the linear range of the assay in Example 1 was 0.012 to 30 pg/mL. The calibration curves for the 5 individual runs were in good agreement (showing stable signal over a wide dynamic range of 5-30,000 fg/mL) and showed increasing signal across the range, but in some cases showed saturation at the upper point, indicating

- 26 049359 indicates that the dynamic range for the assay example from Example 1 is 0.005-10 pg/mL. The mean, SD, CV, and signal-to-background ratio for each dilution in the 5 separate runs shown in Fig. 6a are given below in Table 2.

Table 2

pg/ml calibration peptide Average AEV SO CV(%) Signal/background 0 0,015 0,003 20.4 1 0,006 0,026 0,003 13.1 1,7 0,012 0,037 0,003 7.5 2,5 0,023 0,059 0,004 6.8 3.9 0,046 0.104 0,005 4.7 6.9 0,093 0,188 0,017 9.0 12.5 0,186 0.344 0,030 8.7 22.8 0.37 0,662 0,044 6,7 43.9 1,11 1,968 0,200 10.2 130.4 3.33 5,515 0.387 7.0 365,5 10 14,228 1,440 Yu,1 942.8 30 23,524 0,187 0.8 1558,7

LLOD = background + 2.5 SD = 0.15 + (2.5 x 0.003) = 0.0228 AEV, which is calculated to correspond to a theoretical concentration of 0.002 pg/mL. LLOQ = first point on the calibration curve with CV < 20% and signal-to-noise ratio (S/N) > 2 = 0.012 pg/mL.

Linearity when diluted to plasma.

To assess the dilution linearity and determine the minimum required dilution (MRD) for testing serum and plasma samples, a set of 3 pooled serum samples from AD subjects with high tau levels and 1 pooled plasma sample from AD subjects with high tau levels were titrated with dilutions from 1:2 to 1:6 (i.e., 1:2, 1:3, 1:4, 1:5, and 1:6 dilutions) in sample diluent and measured according to the assay in Example 1. The amount of p217+ tau detected at each dilution was then recalculated to obtain the estimated concentration of p217+ tau present in the undiluted serum or plasma sample and compared to the estimated concentration of p217+ tau calculated from the data determined in the serum or plasma sample at the 1:3 dilution. The data obtained are presented in Fig. 6b for each of the samples as % of the predicted p217+ tau concentration based on the 1:3 dilution sample. The results obtained for each of the 3 serum samples are presented in Fig. 6b as circles (•), squares (), and triangles (▲). The results obtained for the plasma sample are presented in Fig. 6b as an inverted triangle (▼). As can be seen in Fig. 6b, the results for each of the sample dilutions were within 20% of the other dilutions for all serum and plasma samples shown. These data demonstrate acceptable linearity over the dilution range of 1:2 to 1:6. Since the serum and plasma samples contain low amounts of p217+ tau, a 1:2 dilution may be preferable.

Example 9a. Verification of compliance with the specifications of an assay for the detection of p217+ tau protein in plasma.

Precision.

To assess the precision of replicate measurements of plasma p217+ tau, measurements were performed on a cohort of 232 plasma samples (including samples from both DD and AD subjects) according to the assay example in Example 1 in quadruplicate. Based on the standard curve of the calibration peptides, the LLOQ of the assay was determined to be 10 fg/mL. Taking into account the 1:2 dilution of the samples, the LLOQ was corrected to 20 fg/mL. However, the data indicate that imprecision increases starting below ~40 fg/mL (Fig. 7a), based on a plot of % CV for each quadruplicate measurement set versus the mean amount of p217+ tau detected (in fg/mL) in each sample. The data in Fig. 7a show that 93% of samples (216/232) were above the assay LLOQ (the concentration at which CV < 20% = 40 fg/mL in this example; shown by the dashed lines). Moreover, all but 4 samples were within the acceptable limits of 20% CV for the research-only (RUO) assay. The average single-assay precision for all 232 plasma samples was 7.1% CV, and the average single-assay precision for all samples above the LLOQ was 6.7% CV.

To assess the interassay precision in the assay of Example 1, a panel of 3 quality control (QC) samples containing low, medium, and high concentrations of the calibration peptide in the assay diluent, as well as a pooled plasma sample from AD subjects with high tau levels and a pooled serum sample from AD subjects with high tau levels were prepared. These samples were then measured according to the assay example of Example 1 in 5 separate runs. The dilution-corrected amounts of p217+ tau detected in these samples are shown in Fig. 7b. For each sample, the mean is shown as the longer horizontal line, with the ± standard deviation (SD) of the data set shown in

- 27 049359 as shorter lines above and below the mean line. For these 5 samples, the interassay precision was determined to be 5-15% CV.

Transferability between laboratories.

To assess the precision of p217+ tau assays across sites, a set of plasma samples from PD and AD subjects can be tested using the same reagent lot at two separate sites. If the values for these samples are highly similar at both sites, this confirms that the assay is transferable between laboratories.

Stability of the analyte.

The stability of the endogenous p217+ tau epitope in plasma can be assessed at different temperatures by making aliquots of pooled plasma from subjects with AD and storing each aliquot at 4°C, 22°C, or 37°C for 1, 2, or 4 h. Additionally, a subset of aliquots can be freeze-thawed (-80°C to 22°C) 1, 2, 3, or 4 times. If no significant signal changes occur upon storage under these different conditions, this would indicate that the tau molecules (and in particular the epitopes) recognized by the assay example in Example 1 are sufficiently stable to permit standard storage/testing procedures.

Example 9b. Verification of compliance with the specifications of an assay for the detection of p217+ T-protein in plasma.

Precision.

Additionally, the results for 227 plasma samples (157 mild-to-moderate subjects, 70 cognitively normal subjects) from Example 9a were also analyzed to assess the precision of the replicate plasma p217+ tau measurements; the results are shown in Fig. 7c. All samples shown in Fig. 7c were detectable (showing signal > LLOD) with acceptable precision (< 25% CV, mean CV = 7.1%). Indeed, 223 of the 227 samples (98.2%) showed CV < 20%. To establish a more robust lower limit of quantification (LLOQ), a cutoff value of 37 fg/mL was set based on the point below which plasma measurements were more likely to show > 20% CV. As shown in Fig. 7c, 94.7% of all samples had values above this LLOQ.

Example 10. Clinical validation of an assay for the detection of p217+ tau in plasma versus p217+ tau in CSF and tau-PET.

To assess the utility of the assay example from Example 1 for the diagnosis and staging of AD, three cohorts of plasma samples were obtained for measurement of p217+tau using this assay. The results of these measurements were analyzed for correlation with CSF p217+tau levels and/or SUVR on tau-PET, as explained in detail below in the discussion of Cohort 3.

Cohort 1. Correlation of plasma p217+tau with CSF p217+tau in a cohort with BA.

In Cohort 1, the correlation of p217+ tau peptides detected in plasma was assessed with p217+ tau peptides detected in corresponding CSF from the same AD subject. Lumbar fluid (LF) CSF samples from each of 16 AD subjects (who had a clinically confirmed diagnosis of mild to moderate dementia, with a Clinical Dementia Rating Scale score of 1+) in the clinical study were measured using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent. Plasma samples from the same 16 AD subjects were measured according to the assay example from Example 1 described above. For each subject, the amount of p217+ tau (in pg/mL) detected in the corresponding CSF sample (plotted on the x-axis) is plotted against the amount of p217+ tau detected in the corresponding plasma sample (plotted on the y-axis) in Fig. 8a and Fig. 8b, which shows the data from Fig. 8a on logarithmic coordinates. The linear regression line (R2=0.43, slope=0.007, p=0.006) along with the ^R2 value for the linear regression line are shown in Fig. 8a. Plasma p217+ tau concentrations were 1.95+/-0.23% (mean+/-SEM) of CSF p217+ tau concentrations.

Cohort 2. Correlation of plasma p217+tau with CSF p217+tau in the validation cohort.

Cohort 2 evaluates the correlation of plasma p217+ tau peptides with CSF p217+ tau peptides in a larger cohort of subjects diagnosed with mild to moderate dementia (Clinical Dementia Rating Scale score 1+) in a clinical trial (n=159), in addition to a cohort of asymptomatic subjects from another clinical trial (n=70). LF CSF samples from Cohort 2, which consisted of 229 subjects, were measured using the pT3xIT43 assay described in Example 1 of the Kolb '492 patent. Plasma samples from the same 229 subjects were measured according to the assay example from Example 1. For each subject, the amount of p217+tau (in pg/mL) detected in the corresponding CSF sample (plotted on the x-axis) is plotted against the amount of p217+tau (in fg/mL) detected in the corresponding plasma sample (plotted on the y-axis) in Fig. 9a. The linear regression parameters (not shown) were ^R2 = 0.35, slope = 6, and p < 0.0001.

- 28 049359

Figure 9a also shows a vertical dotted line representing the first cutoff value (6.6 pg/mL) above which CSF samples would indicate that the subject has a tauopathy or is at risk of developing a tauopathy (e.g., as demonstrated by elevated p217+ tau levels in CSF and/or elevated SUVR on tau PET, as explained in more detail below in the discussion of Cohort 3), and a corresponding horizontal dotted line representing the second cutoff value (104 fg/mL) above which plasma samples would indicate that the subject has a tauopathy or is at risk of developing a tauopathy. The upper right quadrant, labeled True+, corresponds to subjects with values in both CSF and plasma above the first and second cutoff values, indicating that both CSF and plasma measurements are indicative of the subject having a tauopathy or being at risk for developing a tauopathy. The lower left quadrant, labeled True-, corresponds to subjects with values in both CSF and plasma below the first and second cutoff values, indicating that both CSF and plasma measurements consistently identify the subject as not being at risk for developing a tauopathy. The upper left quadrant, labeled False+, corresponds to subjects with values in CSF above the first cutoff value but values in plasma below the second cutoff value, indicating that plasma measurements are indicative of the subject having a tauopathy or being at risk for developing a tauopathy but CSF measurements are not. The lower right quadrant, labeled False-, corresponds to subjects whose CSF values are below the first cutoff but whose plasma values are above the second cutoff, indicating that the CSF measurements indicate that such subjects have a tauopathy or are at risk for developing a tauopathy, but the plasma measurements do not. The number of subjects identified in each quadrant in Fig. 9a is given below in Table 3.

Table 3

Truth+ Truth- Lie+ Lie- Number of subjects 133 74 10 12 % of subjects 58 32 4 5

The data from Fig. 9a and Table 3 were used to generate the receiver operating characteristic (ROC) curve shown in Fig. 9b, demonstrating the ability of the plasma measurements to discriminate between subjects having CSF measurements above the second cutoff, indicating that the subject has a tauopathy or is at risk for developing a tauopathy, or below the second cutoff, indicating that the subject is healthy or is not at risk for developing a tauopathy. The plasma assay from Example 1 demonstrated good specificity and sensitivity with an AUC of 0.943 (95% CI: 90.9; 97.8).

The Kolb '492 patent reported that the 2-stage assay described therein was able to distinguish CSF samples from subjects having biopsy+ brain biopsies from CSF samples from subjects having biopsy- brain biopsies, indicating that measurements of p217+ tau in CSF can be indicative of the presence of a tauopathy clinical pathology, particularly AD, in a patient. Based on the data presented in Figs. 9a and 9b and Table 3 herein, measurements in plasma according to the assays and methods of the present invention can also be indicative of the presence of a tauopathy clinical pathology in a patient, are consistent with increased p217+ tau levels in CSF, and can be used as predictive biomarkers for the detection of tauopathy in patients.

Cohort 3. Correlation of CSF p217+ tau with PET tau values.

Cohort 3 evaluates the correlation of p217+ tau peptides detected in CSF with ¹⁸ F-T807 ( ¹⁸ F-AV-1451) retention in brain tissue measured by PET images. F-T807 retention measurements by PET (tau-PET measurements) correspond to tau protein accumulation in brain tissue, which is a key indicator for distinguishing subjects with or at risk for tauopathy from healthy subjects not at risk for tauopathy.

Cohort 3 includes 178 subjects with different cognitive statuses (controls without cognitive impairment, mild cognitive impairment, AD dementia, and several other neurodegenerative diseases). Tau PET measurements were made in regions of interest (ROIs) of the Braak stages I-IV brain of each subject in Cohort 3. LF CSF samples collected from these subjects simultaneously or in parallel were measured using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent. For each subject, the amount of p217+ tau (in pg/mL) detected in the CSF sample (plotted on the x-axis) is plotted against the standardized uptake rate ratio (SUVR) for the ^F18 -labeled tracer T807 in the corresponding subject in Fig. 10a. In cohort 3 subjects, tau PET values were shown to correspond to a specific range of SUVR for the ^F18 -labeled tracer T807, as shown on the y-axis in Fig. 10a. Linear regression parameters (not shown) were ^R2 = 0.722 and p < 0.0001.

- 29 049359

Cohort 3 subjects can also be divided into two different subgroups: those subjects who were found to be correlated with increased or absent increases in amyloid-β (Άβ) deposits in brain tissue as measured by PET. Those who had increased amounts of Ae are referred to in Figure 10 as Άβ+, while those who did not are referred to below as Ae-. In Figure 10a, subjects who are Άβ+ are shown with darker icons, while Άβ- subjects, who are overrepresented in the lower left quadrant of the figure, are shown with lighter icons. The linear regression parameters (not shown) for the plot of SUVR for F ¹⁸ -labeled T807 versus CSF p217+ tau measurements were for the Άβ+ subgroup: R ² = 0.740 and p < 0.0001. The linear regression parameters (not shown) for the plot of SUVR for F ¹⁸ -labeled T807 versus CSF p217+ tau measurements were for the Άβ+ subgroup: R ² = 0.091 and p = 0.532. The data in Fig. 10a was used to generate a ROC curve demonstrating the ability of CSF measurements to discriminate between patients having an SUVR for the ^F18 -labeled tracer T807 above a cutoff value (e.g., 1.25), indicating that the subject had a tauopathy or was at risk of developing a tauopathy, or below a cutoff value, indicating that the subject was healthy or not at risk of developing a tauopathy. The CSF measurements obtained using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent showed good specificity and sensitivity in predicting this high T807 SUVR with ΆUC = 0.905 (95% CI: 86, 94.9), and also allowed us to define 6.6 pg/mL as the required cutoff value above which CSF p217+ tau would indicate that a subject has a tauopathy or is at risk for developing a tauopathy. Thus, this required cutoff value was used in the analysis of the data obtained in Cohort 2, as discussed above, to analyze the correlation between plasma p217+ tau and tau PET data. Accordingly, the data presented in Fig. 10a and 10b, when considered in combination with the data obtained in Cohort 2, further demonstrate that plasma measurements obtained according to the assays and methods of the present invention can be indicative of the presence of a tauopathy clinical pathology in a patient, are consistent with increased tau protein accumulation in brain tissue, and can be used as predictive biomarkers for the detection of tauopathy in patients.

Example 11. Clinical validation of an assay for the detection of p217+ tau in plasma versus p217+ tau in CSF and p181-tau in CSF.

To evaluate the utility of the assay example from Example 1 for the diagnosis and staging of AD, plasma samples were obtained for measurement of p217+tau using this assay. The results of these measurements were analyzed for correlation with CSF p217+tau levels and/or CSF p181-tau levels. The level of p181-tau in CSF corresponds to the amount of human tau or tau fragment detected in CSF that is phosphorylated at residue 181 of tau.

Correlation of plasma p217+tau with CSF p217+tau in the validation cohort.

The same cohort used in Example 9b was used to evaluate the correlation of p217+ tau peptides detected in plasma with p217+ tau protein detected in CSF. LF CSF samples from this cohort, which consists of 227 subjects, were measured using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent. Plasma samples from the same 227 subjects were measured according to the assay example from Example 1 described above. For each subject, the amount of p217+ tau protein (in pg/mL) detected in the corresponding CSF sample (plotted on the x-axis) is plotted against the amount of p217+ tau protein (in fg/mL) detected in the corresponding plasma sample (plotted on the y-axis) in Fig. 11a. Linear regression parameters (not shown): R ² = 0.35. Plasma p217+tau concentrations were 1.87+/-0.11% (mean +/- SEM) of CSF p217+tau concentrations.

Figures 11b and 11c show the data from Figure 11a divided according to the amyloid status (e.g., A+ or A−) of the patients. A+ indicates patients who are considered amyloid positive, having a CSF Άβ42/40 ratio < 0.089. A− indicates patients who are considered amyloid negative, having a CSF Άβ42/40 ratio > 0.089. Note that 17 subjects from Figure 11a are not included in either Figure 11b or Figure 11c because CSF amyloid data were not available for these patients. Figure 11b shows data for the subset of patients who are A+ (n=160), and Figure 11c shows data for the subset of patients who are A− (n=50). Linear regression parameters (not shown) for Fig. 11b: R ² = 0.27. Linear regression parameters (not shown) for Fig. 11c: R ² = 0.01. Plasma p217+ tau concentrations in A+ patients were 1.63 ± 0.08% (mean ± SEM) of CSF p217+ tau concentrations. Plasma p217+ tau concentrations in A+ patients were 2.73 ± 0.39% (mean ± SEM) of CSF p217+ tau concentrations. As the above data show, the ratio of plasma p217+tau to CSF p217+tau was slightly but significantly lower in the amyloid-positive cohort (p < 0.0001 using an unpaired t-test).

Correlation of p217+ tau protein in CSF with p181-tau protein in CSF.

The correlation of p217+ tau peptides detected in CSF with p181 levels detected in CSF was assessed. In CSF samples from subjects with mild to moderate dementia (n=286; 89%

- 30 049359

A+) measurements were performed using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent and the Innotest p181tau assay to determine the concentrations of p217+ tau and p181, respectively, in CSF samples. For each subject, the amount of p181 tau (in pg/mL) detected in the CSF sample (plotted on the x-axis) is plotted against the amount of p217+ mav (in pg/mL) detected in the CSF sample (plotted on the y-axis) for the corresponding subject in Fig. 12a. Those patients with elevated CSF p181 tau levels can be identified as T+, indicating that the patients have or are at risk for tauopathy. Those patients whose CSF p181 levels are below a certain cutoff value can be identified as T-, indicating that the patients are healthy or not at risk of developing a tauopathy. In this example, a CSF p181-tau concentration > 52 pg/mL is identified as T+, while a CSF p181-tau concentration < 52 pg/mL is designated as T-. Linear regression of the data shown in Fig. 12a is used to relate this cutoff CSF p181-tau concentration to the CSF p217+ tau concentration. Based on these data, a CSF p181-tau cutoff value of 52 pg/mL is associated with 11.4 pg/mL CSF p217+ tau. Therefore, a CSF p217+tau concentration > 11.4 pg/mL can be used to identify those patients who are T+, whereas a CSF p217+tau concentration < 11.4 pg/mL can be used to identify those patients who are T-.

The data from Fig. 11a were used to construct a ROC curve (Fig. 12b) demonstrating the ability of CSF p217+tau measurements to discriminate T+ from T- patients. CSF measurements obtained using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent showed high accuracy in predicting whether a patient would be T+ or T-, with an AUC of 0.9469. ROC curve analysis using the Youden index determined the cutoff value of plasma p217+tau to discriminate T+ from T- patients to be 124.6 fg/mL.

The data from Fig. 11a are also shown in Fig. 12c, where the vertical dotted line represents the first cutoff value (as defined above, equal to 11.4 pg/mL) above which CSF samples will indicate that they are from a T+ patient, and the corresponding horizontal dotted line represents the second cutoff value (as defined above, equal to 124.6 fg/mL) above which plasma samples will indicate that they are from a T- patient. The upper right quadrant, labeled True+, corresponds to subjects in whom both CSF and plasma samples are above both the first and second cutoff values, indicating that both CSF and plasma measurements indicate that the subject has a tauopathy or is at risk for developing a tauopathy. The lower left quadrant, labeled True-, corresponds to subjects whose CSF and plasma measurements are both below the first and second cutoff values, indicating that both CSF and plasma measurements consistently identify the subject as not at risk for developing a tauopathy. The upper left quadrant, labeled False+, corresponds to subjects whose CSF measurements are above the first cutoff value but whose plasma measurements are below the second cutoff value, indicating that plasma measurements indicate that such subjects have a tauopathy or are at risk for developing a tauopathy, but CSF measurements do not. The lower right quadrant, labeled False-, corresponds to subjects whose CSF measurements are below the first cutoff but whose plasma measurements are above the second cutoff, indicating that the CSF measurements indicate that such subjects have a tauopathy or are at risk for developing a tauopathy, but the plasma measurements do not. Figure 12c shows the low frequencies of False cases of +/- 10 and 2%, respectively. The numbers of subjects identified in each quadrant of Figure 12c are listed below in Table 4.

Table 4

Truth+ Truth- Lie+ Lie- Number of subjects 108 91 24 4 % of subjects 48 40 10 2

Figures 12e and 12g show the data from Figure 12c divided into patients with normal cognitive function and patients with mild to moderate dementia. Figure 12e shows the data for the subgroup of patients with normal cognitive function, and Figure 12g shows the data for the subgroup of patients with mild to moderate dementia.

The data from Fig. 12e were used to construct a ROC curve (Fig. 12d) demonstrating the ability of CSF p217+tau measurements to discriminate T+ from T− patients in the cognitively normal subgroup (n=70). CSF measurements in the cognitively normal subgroup showed a similar level of accuracy with an AUC of 0.9045. ROC curve analysis using the Youden index was used to determine the cutoff value of plasma p217+tau to discriminate T+ from T− patients. Also shown in Fig. 12e is a vertical dotted line representing the first cutoff value (defined above as 11.4 pg/mL) above which CSF samples will indicate that they are from a T+ patient, and a corresponding horizontal line

- 31 049359 dashed line representing the second threshold (determined from the ROC curve in Fig. 12d) above which plasma samples will indicate that they are from patient T-. The upper right quadrant, labeled True+, corresponds to subjects whose CSF and plasma values are both above the first and second thresholds. The lower left quadrant, labeled True-, corresponds to subjects whose CSF and plasma values are both below the first and second thresholds. The upper left quadrant, labeled False+, corresponds to subjects whose CSF values are above the first threshold but whose plasma values are below the second threshold. The upper left quadrant, labeled False-, corresponds to subjects whose CSF values are below the first threshold but whose plasma values are above the second threshold. In Fig. Figure 12e shows low false frequencies of +/- 23 and 0%, respectively. The number of subjects identified in each quadrant of Figure 12e is given below in Table 5.

Table 5

Ispshna+ Truth- Lie+ Lie- Number of subjects 5 49 16 0 % of subjects 7 70 23 0

The data from Fig. 12g were used to construct a ROC curve (Fig. 12e) demonstrating the ability of CSF p217+tau measurements to discriminate T+ from T- patients in the subgroup of patients with mild to moderate dementia (n=157). CSF measurements in the subgroup of patients with mild to moderate dementia showed a similar level of accuracy with an AUC of 0.9254. ROC curve analysis using the Youden index was used to determine the cutoff value of plasma p217+tau to discriminate T+ from T- patients.

Also shown in Fig. 12g is a vertical dotted line representing the first cutoff value (defined above as 11.4 pg/mL) above which CSF samples will indicate that they are from a T+ patient, and a corresponding horizontal dotted line representing the second cutoff value (defined by the ROC curve in Fig. 12e) above which plasma samples will indicate that they are from a T- patient. The upper right quadrant, labeled True+, corresponds to subjects whose CSF and plasma values are both above the first and second cutoff values. The lower left quadrant, labeled True-, corresponds to subjects whose CSF and plasma values are both below the first and second cutoff values. The upper left quadrant, labeled False+, corresponds to subjects whose CSF values are above the first cutoff value but whose plasma values are below the second cutoff value. The upper left quadrant, labeled False-, corresponds to subjects whose CSF values are below the first cutoff value but whose plasma values are above the second cutoff value. Figure 12g shows the low frequencies of False+/- events of 9 and 4%, respectively. The numbers of subjects identified in each quadrant in Figure 12g are listed below in Table 6.

Table 6

Truth+ Truth- Lie+ Lie- Number of subjects 100 36 14 7 % of subjects 64 23 9 4

The data shown in Figs. 12e and 12g show that plasma measurements according to the assays and methods of the present invention provide similar predictive ability in each of the subgroups: those with normal cognitive functions and those with mild to moderate dementia.

Example 12. Clinical validation of an assay for the detection of p217+ tau protein in plasma versus β-amyloid in CSF.

To assess the utility of the assay example in Example 1 for the diagnosis and staging of AD, plasma samples were obtained for measurement of p217+ tau using this assay. These measurements were analyzed for correlation with CSF β-amyloid levels.

Correlation of plasma p217+ tau protein with CSF β-amyloid levels.

A cohort of 210 patients was used to evaluate the correlation of p217+ tau peptides detected in plasma with CSF β-amyloid levels. CSF samples from this cohort were measured to determine the amount of Ae42 and the amount of Ae40 present in the samples. Plasma samples from the same 227 subjects were measured according to the assay example in Example 1 described above. The ratio of the amount of Ae42 detected to the amount of Ae40 is a key indicator for distinguishing subjects with or at risk of developing an amyloidogenic disease from healthy subjects not at risk of developing an amyloidogenic disease. For each subject, the ratio of the amount of Ae42 detected to the amount of Ae40 (Ae42/40 ratio) in the corresponding CSF sample (plotted on the x-axis) is plotted on

- 32 049359 plot of the amount of p217+ tau protein (in fg/mL) detected in the corresponding plasma sample (plotted on the y-axis) in Fig. 13a. Patients having a decreased Ae42/40 ratio can be identified as A+, indicating that the patients have an amyloidogenic disease or are at risk of developing an amyloidogenic disease. Patients having an increased Ae42/40 ratio can be identified as A-, indicating that the patients are healthy or are not at risk of developing an amyloidogenic disease. In this example, an Ae42/40 ratio < 0.089 is identified as A+, while an Ae42/40 ratio > 0.089 is designated as A-.

The data from Fig. 13b were used to construct a ROC curve (Fig. 13a) demonstrating the ability of plasma p217+tau measurements to discriminate between T+ and T- patients. Plasma p217+tau showed high accuracy in predicting whether a patient would be A+ or A-, with an AUC of 0.8964. ROC curve analysis using the Youden index determined the cutoff value of plasma p217+tau to discriminate between A+ and A- patients to be 103.9 fg/mL.

Also shown in Fig. 13b is a vertical dotted line representing a first cutoff value (defined above as 0.089) below which the Ae42/40 ratio would indicate that the patient is A+, and a corresponding horizontal dotted line representing a second cutoff value (defined above as 103.9 fg/mL) below which the plasma p217+ tau levels would indicate that the patient is A-. In the upper left quadrant, labeled True+, the Ae42/40 ratio is below the first cutoff value and the plasma p217+ tau levels are above the second cutoff value, indicating that both the Ae42/40 ratio and the plasma p217+ tau levels are indicative of the subject having an amyloidogenic disease or being at risk of developing an amyloidogenic disease. The lower right quadrant, labeled True-, corresponds to subjects with both the Ae42/40 ratio above the first cutoff and plasma p217+tau levels below the second cutoff, indicating that both the Ae42/40 ratio and plasma p217+tau levels consistently identify the subject as not at risk of developing an amyloidogenic disease. The upper right quadrant, labeled False+, corresponds to subjects with both the Ae42/40 ratio and plasma p217+tau levels above the first and second cutoffs, respectively, indicating that plasma p217+tau levels are indicative of the presence of an amyloidogenic disease or of the susceptibility of such subjects to developing an amyloidogenic disease, whereas the Ae42/40 ratio does not. The lower left quadrant, labeled False-, corresponds to subjects with an Ae42/40 ratio below the first cutoff and measured plasma p217+tau levels below the second cutoff, indicating that the Ae42/40 ratio indicates that such subjects have or are at risk for developing an amyloidogenic disease, whereas the plasma p217+tau levels do not. Figure 13b shows the low false event rates of +/- 1 and 18%, respectively. The numbers of subjects identified in each quadrant of Figure 13b are listed below in Table 7.

Table 7

Truth+ Truth- Lie+ Lie- Number of subjects 123 48 2 37 % of subjects 58 23 1 18

Figures 13d and 13e show the data from Figure 13b divided into patients with normal cognitive function and patients with mild to moderate dementia. Figure 13d shows data for the subgroup of patients with normal cognitive function, and Figure 13e shows data for the subgroup of patients with mild to moderate dementia.

The data in Fig. 13d were used to construct a ROC curve (Fig. 13c) demonstrating the ability of plasma p217+tau measurements to discriminate A+ from A- patients in the cognitively normal subgroup (n=70). The ROC curve for plasma p217+tau measurements in the cognitively normal subgroup has an AUC of 0.6554. ROC curve analysis using the Youden index was used to determine the cutoff value of plasma p217+tau to discriminate A+ from A- patients.

Also shown in Fig. 13d are a vertical dotted line representing the first cut-off value (defined above as 0.089) below which the Ae42/40 ratio would indicate that it is an A+ patient, and a corresponding horizontal dotted line representing the second cut-off value (defined from the ROC curve in Fig. 13c) below which plasma p217+tau levels would indicate that they are an A− patient. In the upper left quadrant, labeled True+, the Ae42/40 ratio is below the first cut-off value and the plasma p217+tau levels are above the second cut-off value. The lower right quadrant, labeled True−, corresponds to subjects in whom both the Ae42/40 ratio is above the first cut-off value and the plasma p217+tau levels are above the second cut-off value.

- 33 049359 measurements and plasma p217+tau levels below the second cutoff value. The upper right quadrant, labeled False+, corresponds to subjects with Ae42/40 ratio and plasma p217+tau levels above the first and second cutoff values, respectively. The lower left quadrant, labeled False-, corresponds to subjects with Ae42/40 ratio below the first cutoff value and measured plasma p217+tau levels below the second cutoff value. Figure 13d shows the False incidence rates of +/- 10 and 24%, respectively. The numbers of subjects identified in each quadrant of Figure 13d are listed below in Table 8.

Table 8

Truth+ Truth- Lie+ Lie- Number of subjects 21 25 7 17 % of subjects 30 36 10 24

The data from Fig. 13e were used to construct a ROC curve (Fig. 13d) demonstrating the ability of plasma p217+tau measurements to discriminate A+ from A- patients in the subgroup of patients with mild to moderate dementia (n=140). Plasma p217+tau measurements provided high accuracy in the subgroup of patients with mild to moderate dementia with AUC=0.9832. ROC curve analysis using the Youden index was used to determine the cutoff value of plasma p217+tau to discriminate A+ from A- patients.

Also shown in Fig. 13e is a vertical dotted line representing the first cutoff value (defined above as 0.089) below which the Ae42/40 ratio would indicate that it is an A+ patient, and a corresponding horizontal dotted line representing the second cutoff value (defined by the ROC curve in Fig. 13d) below which the plasma p217+tau levels would indicate that they are an A− patient. In the upper left quadrant, labeled True+, the Ae42/40 ratio is below the first cutoff value and the plasma p217+tau levels are above the second cutoff value. The lower right quadrant, labeled True−, corresponds to subjects who have both the Ae42/40 ratio above the first cutoff value and the plasma p217+tau levels below the second cutoff value. The upper right quadrant, labeled False+, corresponds to subjects with Ae42/40 ratio and plasma p217+tau levels above the first and second cutoff values, respectively. The lower left quadrant, labeled False-, corresponds to subjects with Ae42/40 ratio below the first cutoff value and measured plasma p217+tau levels below the second cutoff value. Figure 13e shows the False +/- incidence rates of 0 and 6%, respectively. The numbers of subjects identified in each quadrant of Figure 13e are listed below in Table 9.

Table 9

Truth+ Truth- Lie+ Lie- Number of subjects FROM 18 0 9 % of subjects 81 13 0 6

The data shown in Figs. 13d and 13e show that plasma measurements according to the assays and methods of the present invention provide improved prognostic accuracy in a subgroup of patients with mild to moderate dementia.

Example 13. Correlation of plasma p217+tau protein with CSF p217+tau protein with biochemical clearance.

An additional cohort of 36 patients was tested to evaluate the correlation of p217+ tau peptides detected in plasma with p217+ tau protein detected in CSF. CSF samples from this cohort were measured using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent. Plasma samples from the same subjects were measured in three different ways. First, crude plasma samples were measured according to the assay example of Example 1. Second, tau peptides were chemically extracted from plasma samples and measured according to the assay example of Example 1. Third, plasma samples were semi-denatured according to Example 2 such that interfering proteins were denatured by heating. For each subject, the amount of p217+tau (in pg/mL) detected in the corresponding CSF sample (on the x-axis) is plotted against the amount of p217+tau (in fg/mL) detected in the corresponding crude plasma sample (on the y-axis) in Figure 14a. The linear regression parameters (not shown) for Figure 14a: R2=0.6418. The amount of p217+tau (in pg/mL) in the CSF sample (on the x-axis) is plotted against the amount of p217+tau (in fg/mL) detected in the corresponding plasma sample after chemical extraction (on the y-axis) in Figure 14b. The linear regression parameters (not shown) for Figure 14b: R2=0.6748. The amount of p217+tau (in pg/mL) in the CSF sample (plotted on the x-axis) is plotted against the amount of p217+tau (in fg/mL) detected in the corresponding semi-denatured plasma sample (plotted on the y-axis) in Fig. 14c. The linear regression parameters (not shown) for Fig. 14c were: R2=0.5484.

- 34 049359

Example 14. Quantification of total p217+ tau protein in subjects receiving treatment with exogenous anti-tau antibodies.

The assays and methods of the present invention can be used to determine total p217+ tau in plasma samples obtained from subjects receiving treatment with an anti-tau antibody, particularly an anti-ρ217+ τ^^λ^ antibody. However, when detecting p217+ tau in plasma samples from subjects receiving treatment with an anti-tau antibody, interference and/or artifacts caused by the presence of the therapeutic antibody in the plasma samples may occur. Therefore, the steps for producing a semi-denatured sample fluid described in Example 2 can be used to process a plasma sample obtained from subjects receiving treatment with an anti-tau antibody to reduce interference from the therapeutic antibody while maintaining the signal from p217+ tau in the semi-denatured sample fluid.

Example 14 modifies the assay example of Example 1 by using steps to denature the samples in the same manner as described for the preparation of semi-denatured sample fluids according to Example 2. The modified assay example is evaluated using plasma samples from DD and AD subjects. These plasma samples were heated and measured according to the assay example of Example 10; the results (in pg/mL p217+ tau protein detected) are shown in Fig. 15a. The left side of Fig. 15a shows the data corresponding to DD subjects, and the right side of each graph shows the data corresponding to AD subjects. For each subject category, the mean is shown as the longer horizontal line with the ±standard deviation (SD) of the data set shown as the shorter lines above and below the mean line. As can be seen in Fig. 15a, the values in semi-denatured plasma samples from subjects with BA are significantly higher than those in samples obtained from subjects with DD, reflecting the results observed for crude plasma samples when crude plasma samples were analyzed as described in Example 1.

To study and evaluate the sensitivity and precision, a modified assay example from Example 14 was used to measure a panel of 585 plasma samples from a phase 1 clinical study of treatment with an anti-ρ217+τ^^λ^ antibody. Example standard curves were generated from 8 separate runs of semi-denatured sample fluids prepared at different dilutions of the calibration peptide as described in Example 1 and are shown in Fig. 15b. The linear range between the LLOQ and ULOQ for the assay of semi-denatured samples as described in Example 14 was defined as the interval between the minimum and maximum points of the standard curve achieving a CV < 20% and a recovery of 80-120% of expected, followed by correction for a 1:6 sample dilution. With these criteria, the linear range of the modified assay example from Example 10 was from ~0.24 to 180 pg/mL. However, to evaluate the precision of the modified assay example of Example 14 when analyzing real samples, the average amount of p217+ tau (in fg/mL) detected in each semi-denatured plasma sample was plotted against the % CV for each sample (Figure 15c). The data in Figure 15c show that the CV is in the range of 0-141% with a mean of 14%. Moreover, 82% of the samples shown in Figure 15c have a CV < 20%, and 66% of the samples are in the linear range. The vertical dotted line represents the concentration in semi-denatured samples at which the imprecision appears to increase, and therefore this line is the determined LLOQ for the samples of ~0.2 pg/mL using the method of Example 14.

In view of these data, it is further contemplated that the assays and methods of the present invention may be combined with pre-analytical manipulation of plasma samples to measure p217+ tau levels in plasma samples obtained from subjects having exogenously administered anti-tau antibodies to monitor the pharmacological effects of anti-tau antibody treatment on plasma p217+ tau levels.

Example 15. A computer-implemented method for detecting and/or predicting tauopathy.

Example 15 describes an example of a computer-implemented method for analyzing tauopathy biomarker measurements present in blood for improved detection and/or prediction of tauopathy in a subject. In particular, one of the biomarker measurements used in Example 15 is p217+ tau protein levels analyzed in serum samples. However, it is contemplated that the method described in Example 15 is equally applicable to p217+ tau protein levels measured in plasma, such as those obtained using the example assay of Example 1.

In Example 15, a panel of 23 blood biomarkers were analyzed in plasma and serum samples from 199 subjects with mild to moderate AD in a Phase III clinical trial. The twenty-three (23) blood biomarkers included P217+ τ^^λοκ, NFL, adiponectin, leptin, and other inflammatory and metabolic markers. In addition, each of these subjects also had p217+ tau levels measured in their CSF using the pT3xhT43 assay described in Example 1 of the Kolb '492 patent. A subject was defined as T-positive if the amount of p217+ tau peptides measured in the subject's CSF was greater than 21 pg/mL; This concentration corresponds to the standard cutoff value of 70 pg/mL in the Innotest p181-tau assay for T-status determination. The data corresponding to the 23 biomarker measurements and the measurements

- 35,049,359 CSF p217+ tau, along with patient demographics (e.g., age and gender) for all 199 subjects were split into two separate datasets: training data (n=150) and holdout data (n=49). The training data was analyzed to select features that were more highly correlated with elevated CSF levels in subjects identified as T-positive. The selected features included blood-derived measurements for p217+ tau, NFL, adiponectin, and leptin. Multiple machine learning modules were trained using this training data. Specifically, a support vector machine module, a random forest module, a logistic regression module, and a gradient boosting module were trained using this training data. The ensemble of all these trained machine learning modules was used to generate the output.

To evaluate the sensitivity and accuracy of the results generated by the ensemble machine learning module, the holdout data were analyzed by the ensemble machine learning module to generate a decision on whether each subject’s data in the holdout data corresponds to a T-positive status. The decisions generated by the ensemble machine learning module were then compared to the actual T-positive status of the subject in the holdout data to evaluate the sensitivity and accuracy of the ensemble machine learning module. The ensemble machine learning module was evaluated in this manner using different subsets of biomarkers, as shown below in Table 10.

Table 10

Data subgroups Biomarker Elements Control Without a biomarker, the analysis uses the subject's age and gender 1 p217+tau protein in serum 2 p217+tau protein in serum and NFL 3 p217+serum tau, NFL and adiponectin 4 p217+tau in serum, NFL, adiponectin and leptin 5 NFL, adiponectin and leptin

The control data subset includes the data from the holdout data excluding any biomarker data. Specifically, the control data subset was analyzed by the ensemble machine learning module using the non-biomarker elements of age and gender for each subject. The analysis results generated by the ensemble machine learning module for the control data subset were used to construct a ROC curve demonstrating the ability of the ensemble machine learning module to discriminate the T-positive status of subjects in the holdout data without any biomarker information, which is shown by the dashed lines in Figs. 16a-16d. The AUC of the ROC curve for the ensemble machine learning module analyzing the control data subset without any biomarker information is 0.59.

Each of the data subsets 1–5 includes a subset of the control data and data from the holdout data corresponding to the biomarkers listed above in Table 10. The results of the analyses generated by the ensemble machine learning module for each of the data subsets 1–5 are summarized below in Tables 5–9. In Tables 5–9, subjects with CSF p217+ tau measurements above 21 pg/mL are listed as Observed+, and those with measurements below are listed as Observed-. In Table 5–9, subjects identified by the ensemble machine learning module as matching the T-positive status are listed as Predicted+, and those not identified as matching the T-positive status are listed as Predicted-.

The results of the analysis of data subset 1, in which serum p217+tau measurements are used as an element, are shown in Table 11. The data from Table 11 were used to construct a ROC curve demonstrating the ability of the ensemble machine learning module to discriminate between subjects with T-positive status in the delayed serum p217+tau measurements, as shown by the solid line in Fig. 16a. The AUC of the ROC curve is 0.87.

Table 11

Predictable+ Predicted- Observed+ 9 9 Observed- 0 31

The results of the analysis of data subset 2, which used serum p217+ tau measurements and NFL data as elements, are shown in Table 12. The data from Table 12 were used to

- 36 049359 ROC curve plotting the ability of the ensemble machine learning module to discriminate between T-positive status of subjects in the delayed serum p217+tau data and NFL data, as shown by the solid line in Fig. 16b. The AUC of the ROC curve is 0.89.

Table 12

Predictable+ Predicted- Observed+ 14 4 Observed- 1 30

The results of the analysis of subset 3 data, which uses serum p217+tau measurements and NFL and adiponectin data as elements, are shown in Table 13. The data from Table 13 were used to construct a ROC curve demonstrating the ability of the ensemble machine learning module to discriminate between subjects' T-positive status in the lagged serum p217+tau measurements and NFL and adiponectin data, as shown by the solid line in Fig. 16c. The AUC of the ROC curve is 0.92.

Table 13

Predictable+ Predicted- Observed+ AND 7 Observed- 0 31

The results of the analysis of subset 4 data, which uses serum p217+tau measurements and NFL, adiponectin, and leptin data as elements, are shown in Table 14. The data from Table 14 were used to construct a ROC curve demonstrating the ability of the ensemble machine learning module to discriminate between subjects' T-positive status in the lagged serum p217+tau measurements and NFL, adiponectin, and leptin data, as shown by the solid line in Fig. 16d. The AUC of the ROC curve is 0.96.

Table 14

We predict ye+ Predicted- Observed+ AND 7 Observed- 1 30

The results of the analysis of subset 5, which uses NFL, adiponectin, and leptin data as elements, are shown in Table 15. The data from Table 15 were used to construct a ROC curve demonstrating the ability of the ensemble machine learning module to discriminate between subjects' T-positive status in the lagged data based on NFL, adiponectin, and leptin data, but not on serum p217+ tau measurements, as shown by the solid line in Fig. 16e. The AUC of the ROC curve is 0.78.

Table 15

Predictable+ Predicted- We observe ye+ 8 10 We observe ye- 4 27

The biomarker element set used in Example 15 consisted of serum p217+tau, NFL, adiponectin, and leptin, each of which showed correlations with CSF p127+tau levels with Spearman correlation coefficients of 0.47, 0.37, 0.16, and -0.23, respectively. Machine learning analysis using serum p217+tau, age, and gender as elements improved performance, showing an AUC of 0.87, compared to 0.59 in the control subset. Increasing the complexity of the machine learning analysis by sequentially adding NFL, adiponectin, and leptin as elements at each step improved performance, showing AUCs of 0.89, 0.92, and 0.96, respectively. When the machine learning analysis included all 4 biomarkers (serum p217+-tau measurements and NFL, adiponectin, and leptin data) as features, the accuracy was 0.84, which was significantly above the no information rate (NIR) (p < 0.005). When excluding serum p217+-tau measurements as a feature, the AUC decreased to 0.78, thus demonstrating that p217+-tau measurements are a significant biomarker for predicting T-positivity.

When all 4 biomarkers were used, the accuracy was 0.84, which was significantly higher than the rate when no information was available (p < 0.005). Excluding serum tau from the full model reduced the AUC to 0.78.

Thus, biomarkers present in the blood can be used to detect tau

- 37,049,359 positive subjects. Serum p217+ tau was the best single analyte for predicting tauopathy or brain pathology due to tauopathy (e.g., the amount of p217+ tau detected in CSF) in subjects with mild to moderate AD.

The invention described and claimed herein should not be limited to the specific embodiments described herein, since these embodiments are intended to illustrate several aspects of the invention. Any equivalent embodiments are intended to be within the scope of the invention. Indeed, various modifications of the invention in addition to those shown and described herein will be apparent to those skilled in the art from the foregoing description. Such modifications are also intended to be included within the scope of the appended claims. All publications cited herein are incorporated herein by reference in their entirety.

Sequence Listing <110> JANSSEN PHARMACEUTICALS NV <120> ANALYSIS OF BLOOD SAMPLES FOR THE DETECTION OF TAUPATHY OR AMYLOIDOGENIC DISEASE <130> JAB7064 <140>

<141>

<150> 63/200,399 <151> 2021-03-04 <150> 62/705,759 <151> 2020-07-14 <160> 28 <170> PatentIn, version 3.5 <210> 1 <211> 441 <212> Protein <213> Artificial sequence <220>

<223> Description of artificial sequence: synthetic polypeptide <220>

<223> tau protein 2N4R <400> 1

Met Ala Glu Pro Arg Gln Glu Phe Glu Val Met Glu Asp His Ala Gly

5 1015

Thr Tyr Gly Leu Gly Asp Arg Lys Asp Gln Gly Gly Tyr Thr Met His 20 2530

Gln Asp Gln Glu Gly Asp Thr Asp Ala Gly Leu Lys Glu Ser Pro Leu 35 4045

Gln Thr Pro Thr Glu Asp Gly Ser Glu Glu Pro Gly Ser Glu Thr Ser

- 38 049359

Asp Ala Lys Ser Thr Pro Thr Ala Glu Asp Val Thr Ala Pro Leu Val

70 7580

Asp Glu Gly Ala Pro Gly Lys Gln Ala Ala Ala Gln Pro His Thr Glu

9095

Ile Pro Glu Gly Thr Thr Ala Glu Glu Ala Gly Ile Gly Asp Thr Pro

100 105110

Ser Leu Glu Asp Glu Ala Ala Gly His Val Thr Gln Ala Arg Met Val

115 120125

Ser Lys Ser Lys Asp Gly Thr Gly Ser Asp Asp Lys Lys Ala Lys Gly

130 135140

Ala Asp Gly Lys Thr Lys Ile Ala Thr Pro Arg Gly Ala Ala Pro Pro

145 150 155160

Gly Gln Lys Gly Gln Ala Asn Ala Thr Arg Ile Pro Ala Lys Thr Pro

165 170175

Pro Ala Pro Lys Thr Pro Pro Ser Ser Gly Glu Pro Pro Lys Ser Gly

180 185190

Asp Arg Ser Gly Tyr Ser Ser Pro Gly Ser Pro Gly Thr Pro Gly Ser

195 200205

Arg Ser Arg Thr Pro Ser Leu Pro Thr Pro Pro Thr Arg Glu Pro Lys

210 215220

Lys Val Ala Val Val Arg Thr Pro Pro Lys Ser Pro Ser Ser Ala Lys

225 230 235240

Ser Arg Leu Gln Thr Ala Pro Val Pro Met Pro Asp Leu Lys Asn Val

245 250255

Lys Ser Lys Ile Gly Ser Thr Glu Asn Leu Lys His Gln Pro Gly Gly

260 265270

Gly Lys Val Gln Ile Ile Asn Lys Lys Leu Asp Leu Ser Asn Val Gln

275 280285

Ser Lys Cys Gly Ser Lys Asp Asn Ile Lys His Val Pro Gly Gly Gly

- 39 049359

290 295 300

Ser Val Gln Ile Val Tyr Lys Pro Val Asp Leu Ser

305 310 315

Lys Val Thr Ser

320

Lys Cys Gly Ser Leu Gly Asn Ile His His Lys Pro

325 330

Gly Gly Gly Gly

335

Val Glu Val Lys Ser Glu Lys Leu Asp Phe Lys Asp

340 345

Arg Val Gln Ser

350

Lys Ile Gly Ser Leu Asp Asn Ile Thr His Val Pro

355 360

Gly Gly Gly Asn 365

Lys Lys Ile Glu Thr His Lys Leu Thr Phe Arg Glu

370 375 380

Asn Ala Lys Ala

Lys Thr Asp His Gly Ala Glu Ile Val Tyr Lys Ser

385 390 395

Pro Val Val Ser

400

Gly Asp Thr Ser Pro Arg His Leu Ser Asn Val Ser

405 410

Ser Thr Gly Ser

415

Ile Asp Met Val Asp Ser Pro Gln Leu Ala Thr Leu

420 425

Ala Asp Glu Val

430

Ser Ala Ser Leu Ala Lys Gln Gly Leu

435 440 <210>

<211>

<212>

Protein <213> Artificial Sequence <220>

<223> Description of the artificial sequence of the peptide synthetic <220>

<223> CDR-H1 sequence of mAb PT82 <400> 2

Gly Phe Thr Phe Ser Asn Tyr Trp Met Asn 1 5 10 <210> 3 <211> 19 <212> Protein <213> Artificial sequence

- 40 049359 <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> CDR-H2 sequence of mAb PT82 <400> 3

Gln Ile Arg Leu Gln Ser Asp Asn Tyr Ala Thr Arg Tyr Ala Glu Ser 1 5 10 15

Val Lys Gly <210> 4 <211> 4 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> CDR-H3 sequence of mAb PT82 <400> 4

Gly Ile Thr Tyr <210> 5 <211> 11 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> CDR-L1 sequence of mAb PT82 <400> 5

Lys Ala Ser Gln Asn Val Gly Thr Ala Val Ala

5 10 <210> 6 <211> 7 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide

- 41 049359 <220>

<223> CDR-L2 sequence of mAb PT82 <400> 6

Ser Ala Ser Ile Arg Tyr Thr

5 <210> 7 <211> 9 <212> Protein <213> Artificial sequence <220>

< 223> Description of the artificial sequence: synthetic peptide <220>

< 223> CDR-L3 sequence of mAb PT82 <400> 7

Gln Gln Phe Ser Ser Tyr Pro Tyr Thr

5 <210> 8 <211> 115 <212> Protein <213> Artificial sequence <220>

<223> Description of artificial sequence: synthetic polypeptide <220>

<223> VH sequence of mAb <400> 8

Glu Val Lys Leu Glu Glu Ser Gly 1 5

Ser Met Lys Leu Ser Cys Val Ala 20

Trp Met Asn Trp Ile Arg Gln Ser 35 40

PT82

Gly Gly Leu Val Gln Pro Gly Gly

15

Ser Gly Phe Thr Phe Ser Asn Tyr 25 30

Pro Glu Lys Gly Leu Glu Trp Val 45

Ala Gln Ile Arg Leu Gln Ser Asp

55

Asn Tyr Ala Thr Arg Tyr Ala Glu 60

Ser Val Lys Gly Arg Phe Thr Ile 65 70

Ser Arg Asp Glu Ser Lys Thr Ser

80

Val Tyr Leu Gln Met Asn Asn Leu 85

Arg Thr Glu Asp Thr Gly Ile Tyr

95

- 42 049359

Tyr Cys Thr Gly Gly Thr Thr Tyr Trp Gly Gln Gly Thr Leu Val Thr

100 105 110

Val Ser Ala

115 <210> 9 <211> 107 <212> Protein <213> Artificial sequence <220>

<223> Description of artificial sequence: synthetic polypeptide <220>

<223> VL sequence of mAb <400> 9

Asp Ile Val Met Thr Gln Ser Gln 1 5

PT82

Lys Phe Met Ser Thr Ser Val Gly

15

Asp Arg Val Ser Ile Thr Cys Lys 20

Ala Ser Gln Asn Val Gly Thr Ala 25 30

Val Ala Trp Tyr Gln Gln Lys Pro

40

Gly Gln Ser Pro Lys Leu Leu Ile 45

Tyr Ser Ala Ser Ile Arg Tyr Thr

55

Gly Val Pro Asp Arg Phe Thr Gly 60

Ser Gly Ser Gly Thr Asp Phe Thr 65 70

Leu Thr Ile Asn Tyr Met Gln Ser

80

Glu Asp Leu Ala Asp Tyr Phe Cys 85

Gln Gln Phe Ser Ser Tyr Pro Tyr

95

Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys

100 105 <210> 10 <211> 10 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide

- 43 049359 <220>

<223> HCDR1 mAb HT43 <400> 10

Gly Phe Thr Phe Arg Ser Tyr Gly Met Ser 1 5 10 <210> 11 <211> 17 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> HCDR2 mAb HT43 <400> 11

Thr Ile Asn Ser Asp Gly Ser Tyr Thr Tyr Tyr Pro Asp Ser Val Lys 1 5 10 15

Gly <210> 12 <211> 8 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> HCDR3 mAb HT43 <400> 12

Ser Trp Asp Gly Ala Met Asp Tyr

5 <210> 13 <211> 16 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> LCDR1 mAb HT43 <400> 13

Arg Ser Ser Gln Ser Ile Leu His Ser Asn Gly Asn Thr Tyr Phe Glu

- 44 049359 <210> 14 <211> 7 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> LCDR2 mAb HT43 <400> 14

LysValSerAsnArgPheSer

5 <210> 15 <211> 9 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> LCDR3 mAt HT43 <400> 15

Phe Gln Gly Ser Leu Val Pro Trp Thr

5 <210> 16 <211> 117 <212> Protein <213> Artificial sequence <220>

< 223> Description of the artificial sequence: synthetic polypeptide <220>

< 223> Heavy chain variable region of mAb HT43 <400> 16

Glu Val Lys Leu Val Glu Ser Gly Gly Asp Leu Val Lys Pro Gly Gly

5 1015

Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Arg Ser Tyr 20 2530

Gly Met Ser Trp Val Arg Gln Thr Pro Asp Lys Arg Leu Glu Trp Val

4045

- 45 049359

Th

Th 50

Ile

Asn

Ser

Asp

Gly 55

Ser

Tyr

Th

Tyr

Tyr 60

Pro Asp Ser Val

Lys 65

Gly

Arg

Phe

Th

Ile

Ser

Arg

Asp

Asn

Ala 75

Lys

Asn Thr Leu Tyr 80

Leu

Gln

Met

Ser

Ser 85

Leu

Lys

Ser

Glu

Asp 90

Th

Ala

Met Tyr Tyr Cys 95

Ala

Val

Ser

Trp 100

Asp

Gly

Ala

Met

Asp 105

Tyr

Trp

Gly

GlnGlyThrSer

110

Val

Th

Val

115

Ser

Ser <210> 17 <211> 112 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial polypeptide sequence <220>

<223> Light chain variable region mAb HT43 synthetic <400> 17

Asp Val Leu Val 1

Thr Gln Thr Pro 5

Leu Ser Leu Pro

Val Ser Leu Gly

Asp Gln Ala Ser 20

Ile Ser Cys Arg

Ser Ser Gln Ser 25

Ile Leu His Ser

AsnGlyAsnThr35

Tyr Phe Glu Trp 40

Tyr Leu Gln Arg

Pro Gly Gln Ser 45

Pro Lys Leu Leu 50

Ile Tyr Lys Val 55

Ser Asn Arg Phe 60

Ser Gly Val Pro

Asp Arg Phe Ser 65

Gly Ser Gly Ser 70

Gly Thr Asp Phe 75

Th Leu Lys Ile

Ser Arg Val Glu

Ala Glu Asp Leu 85

Gly Val Tyr Tyr 90

CysPheGlnGly95

Ser Leu Val Pro

Trp Thr Phe Gly

Gly Gly Th Lys

Leu Glu Ile Lys

- 46 049359

100

105

110 <210> 18 <211> 40 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial polypeptide sequence <220>

<223> pT3xhT43 calibration peptide <220>

<221> SITE <222> (20)..(21) < 223> dPEG4 Linker <220>

< 221> MOD_RES < 222> (26).. (26) < 223> phosphorylated threonine <220>

< 221> MOD_RES < 222> (31)..(31) < 223> phosphorylated threonine <220>

< 221> MOD_RES < 222> (40)..(40) < 223> C-terminal amide < 400> 18

Pro Arg Gln Glu Phe Glu Val Met Glu Asp His Ala

510 synthetic

Leu Gly Asp Arg Gly Ser Arg Ser Arg Thr Pro Ser 2025

Gly Thr Tyr Gly 15

Leu Pro Thr Pro 30

Pro Thr Arg Glu Pro Lys Lys Val

3540 <210>

<211>

<212>

118 Protein <213> Artificial sequence <220>

<223> Description of the artificial polypeptide sequence <220>

<223> VH mouse mAb PT3 synthetic

- 47 049359 <400> 19

Glu Val Lys Leu 1

Sir Leu Lys Leu

Val Glu Ser Gly 5

Ser Cys Ala Ala

Gly Asp Leu Val 10

Lys Pro Gly Gly 15

Ser Gly Phe Thr 25

Phe Ser Ser Tyr 30

Ala Met Ser Trp 35

Val Arg Gln Asn 40

Pro Glu Lys Arg

Leu Glu Trp Val 45

Ala Ser Ile Ser 50

LysGlyGlyAsn 55

Thr Tyr Tyr Pro 60

Asn Ser Val Lys

Gly Arg Phe Thr 65

Ile Ser Arg Asp 70

Asn Ala Arg Asn 75

Ile Leu Tyr Leu 80

Gln Met Ser Ser

Leu Arg Ser Glu 85

Asp Thr Ala Leu 90

Tyr Tyr Cys Ala 95

ArgGlyTrpGly

100

Asp Tyr Gly Trp

Phe Ala Tyr Trp 105

Gly Gln Val Thr

110

Leu Val Thr Val Ser Ala

115 <210> 20 <211> 107 <212> Protein <213> Artificial sequence <220>

<223> Description of artificial sequence: synthetic polypeptide <220>

<223> VL of mouse mAb PT3 <400> 20

Asp Ile Lys Met Thr Gln Ser Pro Ser Ser Met

5 10

Tyr Ala Ser Leu Gly

Glu Arg Val

Thr Ile Thr Cys Lys Ala Ser Gln Asp 20 25

Ile Asn Arg Tyr

Leu Asn

TrpPhe 35

GlnGlnLys

Pro Gly Lys

Ser Pro Lys Thr Leu Ile 45

Tyr Arg Ala Asn Arg Leu Leu Asp Gly Val Pro

Ser Arg Phe Ser Gly

- 48 049359

Ser Gly Ser Gly Gln Asp Tyr Ser Leu Thr Ile Ser 65 70 75

Ser Leu Asp Tyr 80

Glu Asp Met

Gly

Ile Tyr Tyr Cys Leu Gln Tyr Asp 85 90

Glu Phe Pro Leu

Thr Phe Gly Asp Gly Thr Lys Leu Glu Leu Lys

100 105 <210> 21 <211> 118 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial polypeptide sequence <220>

<223> VH humanized mAb PT3 synthetic <400> 21

Gln Val Gln Leu 1

Val Glu Ser Gly 5

Gly Gly Val Val 10

Gln Pro Gly Arg 15

Ser Leu Arg Leu 20

Ser Cys Ala Ala

Ser Gly Phe Thr 25

Phe Ser Ser Tyr 30

Ala Met Ser Trp

Val Arg Gln Ala 40

Pro Gly Lys Gly

Leu Glu Trp Val 45

Ala Ser Ile Ser 50

LysGlyGlyAsn 55

Thr Tyr Tyr Ala 60

Asp Ser Val Lys

Gly Arg Phe Thr 65

Ile Ser Arg Asp 70

Asn Ser Lys Asn 75

Thr Leu Tyr Leu 80

Gln Met Asn Ser

Leu Arg Ala Glu 85

Asp Thr Ala Val 90

Tyr Tyr Cys Ala

ArgGlyTrpGly

100

Asp Tyr Gly Trp

Phe Ala Tyr Trp 105

Gly Gln Val Thr

110

Leu Val Thr Val

115

Ser Ser

- 49 049359 <210> 22 <211> 107 <212> Protein <213> Artificial sequence <220>

<223> Description of artificial sequence: synthetic polypeptide <220>

<223> VL of humanized mAb PT3 <400> 22

Asp Ile Gln Met Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 1 5 1015

Asp Arg Val Thr Ile Thr Cys Lys Ala Ser Gln Asp Ile Asn Arg Tyr 20 2530

Leu Asn Trp Phe Gln Gln Lys Pro Gly Lys Ala Pro Lys Ser Leu Ile 35 4045

Tyr Arg Ala Asn Arg Leu Leu Asp Gly Val Pro Ser Arg Phe Ser Gly 50 5560

Ser Gly Ser Gly Thr Asp Phe Thr Leu Thr Ile Ser Ser Leu Gln Pro 65 70 7580

Glu Asp Phe Ala Thr Tyr Tyr Cys Leu Gln Tyr Asp Glu Phe Pro Leu 85 9095

Thr Phe Gly Gln Gly Thr Lys Leu Glu Ile Lys

100105 <210> 23 <211> 5 <212> Protein <213> Artificial sequence <220>

< 223> Description of the artificial sequence: synthetic peptide <220>

< 223> HCDR1 of mouse mAb PT3, numbering according to Kabat < 400> 23

Ser Tyr Ala Met Ser

5 <210> 24 <211> 16

- 50 049359 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <220>

<223> HCDR2 of mouse mAb PT3, Kabat numbering <400> 24

Ser Ile Ser Lys Gly Gly Asn Thr Tyr Tyr Pro Asn Ser Val Lys Gly 1 5 10 15 <210> 25 <211> 10 <212> Protein <213> Artificial sequence <220>

< 223> Description of the artificial sequence: synthetic peptide <220>

< 223> HCDR3 of mouse mAb PT3, Kabat numbering <400> 25

Gly Trp Gly Asp Tyr Gly Trp Phe Ala Tyr

5 10 <210> 26 <211> 11 <212> Protein < 213> Artificial sequence <220>

< 223> Description of the artificial sequence: synthetic peptide <220>

< 223> LCDR1 mouse mAb PT3, Kabat numbering <400> 26

Lys Ala Ser Gln Asp Ile Asn Arg Tyr Leu Asn

5 10 <210> 27 <211> 7 <212> Protein <213> Artificial sequence <220>

<223> Description of the artificial sequence: synthetic peptide <223> LCDR2 of mouse mAb PT3, numbering according to Kabat <220>

-

Claims (28)

<400> 27 Arg Ala Asn Arg Leu Leu Asp 1 5 <210> 28 <211> 9 <212> Protein <213> Artificial sequence <220> <223> Description of the artificial sequence: synthetic peptide <220> <223> LCDR3 mouse mAb PT3, Kabat numbering <400> 28 Leu Gln Tyr Asp Glu Phe Pro Leu Thr 1 5 CLAUSE OF INVENTION 1. An analytical method for detecting p217+ tau peptides in a subject, the method comprising the steps of: obtain a plasma sample from the subject; contacting the plasma sample with a capture antibody directed to a p217+ tau epitope to cause the capture antibody to bind to p217+ tau peptides in the plasma sample to form antibody-peptide complexes; the antibody-peptide complexes are washed; contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes; and detecting the detector antibody to determine the amount of p217+ tau peptides in the plasma sample. 2. The method according to claim 1, wherein the capture antibody is immobilized on a solid phase. 3. The method according to claim 2, wherein the solid phase is a magnetic granule. 4. The method of claim 1, wherein the capture antibody binds to an epitope comprising amino acids 210-220 of human tau protein. 5. The method according to claim 1, wherein the detector antibody binds to an epitope comprising amino acids 7-20 or 116-127 of human tau protein. 6. The method of claim 4, wherein the capture antibody is pT3. 7. The method of claim 5, wherein the detector antibody is pT82. 8. The method of claim 3, wherein the plasma sample is diluted with a sample diluent prior to contact with the capture antibody, wherein the sample diluent comprises at least one of a nonionic surfactant and tris(hydroxymethyl)aminomethane. 9. A method for detecting tauopathy in a subject, the method comprising the steps of: obtain a plasma sample from the subject; detecting an amount of ρ217+τ^-^πτ^οβ in a plasma sample by an assay, wherein the assay comprises (i) contacting the plasma sample with a capture antibody directed to a p217+tau epitope to bind the capture antibody to p217+mау peptides in the plasma sample to form antibody-peptide complexes, (ii) washing the antibody-peptide complexes, (iii) contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes, and (iv) detecting the detector antibody; and determining whether a subject has a tauopathy or is at risk of developing a tauopathy when the amount of p217+tau peptides is above a predetermined threshold, wherein the predetermined threshold is above the lower limit of quantification (LLOQ) of the assay. 10. The method of claim 9, wherein the analysis does not concentrate p217+tau peptides from the plasma sample by immunoprecipitation prior to measuring the amount of p217+tau peptides present. - 52 049359 11. The method according to claim 9, wherein the plasma sample is unpurified plasma. 12. The method according to claim 9, wherein the LLOQ corresponds to a coefficient of variation (CV) of the assay of 15-25%. 13. The method of claim 12, wherein the LLOQ corresponds to a CV of the assay of 20%. 14. The method according to claim 9, wherein the specified threshold value is at least 3 times greater than the LLOQ. 15. The method of claim 9, wherein the specified threshold value is at least 10 times the lower detection limit of the assay. 16. The method of claim 9, wherein the tauopathy is selected from the group consisting of familial Alzheimer's disease, sporadic Alzheimer's disease, frontotemporal dementia with parkinsonism related to chromosome 17 (FTDP-17), progressive supranuclear palsy, corticobasal degeneration, Pick's disease, progressive subcortical gliosis, dementia with predominant neurofibrillary tangles, diffuse neurofibrillary tangles with calcification, dementia with argyrophilic grains, amyotrophic lateral sclerosis-parkinsonism-dementia complex, Down syndrome, Gerstmann-Sträussler-Scheinker disease, Hallervorden-Spatz disease, inclusion body myositis, Creutzfeldt-Jakob disease, multiple system atrophy, Niemann-Pick disease type C, cerebral amyloid angiopathy with prion proteins, subacute sclerosing panencephalitis, myotonic dystrophy, non-Guam motor neuron disease with neurofibrillary tangles, postencephalitic parkinsonism, chronic traumatic encephalopathy, and dementia pugulistica (boxer's dementia). 17. The method of claim 16, wherein the tauopathy is Alzheimer's disease. 18. The method of claim 16, wherein the tauopathy is progressive supranuclear palsy. 19. A method for detecting an amyloidogenic disease in a subject, the method comprising the steps of: obtain a plasma sample from the subject; detecting an amount of p217+ tau peptides in a plasma sample by an assay, wherein the assay comprises (i) contacting the plasma sample with a capture antibody directed to a p217+ tau epitope to bind the capture antibody to p217+ tau peptides in the plasma sample to form antibody-peptide complexes, (ii) washing the antibody-peptide complexes, (iii) contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes, and (iv) detecting the detector antibody; and determining whether a subject has an amyloidogenic disease or is at risk of developing an amyloidogenic disease when the amount of p217+ tau peptides is above a predetermined threshold, wherein the predetermined threshold is above the lower limit of quantification (LLOQ) of the assay. 20. The method of claim 19, wherein the analysis does not concentrate p217+tau peptides from the plasma sample by immunoprecipitation prior to measuring the amount of p217+tau peptides present. 21. The method according to claim 19, wherein the plasma sample is unpurified plasma. 22. The method of claim 19, wherein the LLOQ corresponds to a coefficient of variation (CV) of the assay of 15-25%. 23. The method according to claim 22, wherein the LLOQ corresponds to a CV of the assay of 20%. 24. The method according to claim 19, wherein the specified threshold value is at least 3 times greater than the LLOQ. 25. The method according to claim 19, wherein the specified threshold value is at least 10 times greater than the lower limit of quantification of the assay. 26. The method of claim 19, wherein the amyloidogenic disease is Alzheimer's disease. 27. A method for detecting or predicting tauopathy in a subject, the method comprising the steps of: detecting a certain amount of p217+tau peptides in a plasma sample, the method comprising (i) contacting the plasma sample with a capture antibody directed to a p217+tau epitope to bind the capture antibody to the p217+tau peptides in the plasma sample to form antibody-peptide complexes, (ii) washing the antibody-peptide complexes, (iii) contacting the antibody-peptide complexes with a detector antibody to bind the detector antibody to the antibody-peptide complexes, and (iv) detecting the detector antibody; generate tau protein data corresponding to the number of p217+tau peptides detected; obtaining biomarker data corresponding to at least one biomarker determined in a subject, wherein the biomarker is selected from the group comprising neurofilament light chain (NFL), adiponectin and leptin; and comparing the tau protein data and additional biomarker data with a set of reference data using a machine learning module to determine or predict whether the subject has a tauopathy or is at risk of developing a tauopathy. 28. The method of claim 27, wherein the machine learning module comprises at least one of the following: -