WO2006083378A2

WO2006083378A2 - IN VIVO MONITORING OF β-AMYLOID PLAQUES AND NEUROFIBRILLARY TANGLES

Info

Publication number: WO2006083378A2
Application number: PCT/US2005/043236
Authority: WO
Inventors: Vladimir Kepe; Jorge R. Barrio; Gregory M. Cole; Gary W. Small
Original assignee: The Regents Of The University Of California; U.S. GOVERNMENT REPRESENTED by THE DEPARTMENT OF VETERANS' AFFAIRS
Priority date: 2004-11-29
Filing date: 2005-11-28
Publication date: 2006-08-10
Also published as: WO2006083378A3

Abstract

The invention provides, methods of monitoring the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of a human or other mammal as a function of time and/or medical treatment, and includes the steps of injecting a radiolabeled compound comprising [F-18]FDDNP or a derivative thereof into the bloodstream of a living; scanning the mammal's brain using positron emission tomography quantifying radiolabeled compound binding in a reference region and in each of one or more selected regions of interest (ROI) in the brain; calculating relative radiolabeled compound binding density in each selected ROI; and (e) repeating the process one or more times. Optionally, a medical treatment is administered, and its effect on the density of β-amyloid plaques and/or neurofibrillary tangles is measured.

Description

IN VIVO MONITORING OF β -AMYLOID PLAQUES AND NEUROFIBRILLARY TANGLES

ACKNOWLEDGEMENT OF GOVERNMENT SUPPORT

This invention was made with government support under grant no. DE-FC 0837- ER60615, awarded by the Department of Energy. The Government has certain rights in the invention.

BACKGROUND OF THE INVENTION

Alzheimer's disease is a progressive neurodegenerative disease that affects approximately 20-40% of the population over 80 years of age, the fastest growing age group in the United States and other post-industrial countries. Common features in the brain of patients with Alzheimer's disease include extensive loss of neurons from the vulnerable neuronal population, and the presence of neuropathological deposits, including β-amyloid senile plaques (SP) and neurofibrillary tangles (NFT)s As the disease worsens, the deposits spread throughout the brain in a predictable pattern. Related pathologies are seen in other forms of dementia and Down Syndrome. Since the initial deposits occur much earlier than the symptoms of the disease, early in vivo imaging of these deposits should have tremendous diagnostic value. A method using "[F-18]FDDNP" (2-(l-{6-[(2-fluoroethyl)(methyl)amino]-2-naphthyl) ethylidene) malononitrile), a [F-18]fluorine labeled probe, has been developed for in vivo detection of pathological deposits in Alzheimer's disease with positron emission tomography (PET). It has been used to determine in vivo levels of the pathological deposits present in brains of

Alzheimer's disease patients, and consistently shows higher levels of [F-18]FDDNP binding in the brain areas with known pathology when compared with the levels determined in the same areas in the brains of cognitively normal age matched subjects (Shoghi-Jadid et al., 2002; Kepe et al., 2004). In vitro results demonstrate the capacity of [F-18]FDDNP to bind to both major types of neuropathological deposits found in Alzheimer's disease: to extra cellular β-amyloid senile plaques (SP), fibrillary aggregates of β-amyloid proteins, and to intracellular neurofibrillary tangles (NFT), fibrillary aggregates of hyperphosphorylated protein tau. (Agdeppa et al., 2003a) This binding can be selective blocked by naproxen, a non-steroidal inflammatory drug. (Agdeppa et al., 2003b) Other types of radiolabeled compounds with reported specific binding to β-amyloid fibrillary aggregates suitable for PET have been developed, either in [C-l l]carbon labeled form or, in some cases, [F-18]fluorine labeled compounds. Based on their structure we can separate them into 5 classes for the purpose of this review:

• Benzothiazole derivatives: 2-[(4'-[C-l l]methylamino)phenyl]-6- hydroxybenzothiazole ([C-11]6-OH-BTA-1 or PIB) has been used in the recently published study on Alzheimer's disease patients and controls with PET (Klunk et al., 2004a). Two studies with this compound have been reported using β-amyloid transgenic mice and microPET which is in stark contrast with the published human study results (Klunk et al.,

2004b; Toyama et al., 2004). Both reported unsuccessful attempts to label β-amyloid in vivo in these animals. [F- 18] fluorine labeled benzothiazoles were also reported (Mathis et al., 2002; Vanderghiste et al., 2004) but their usefulness is limited by in vivo de-fluorination and lack of specific binding. Another drawback of [C-11]6-OH-BTA-1 is the fact that it gets sulfated at the 6-OH group, which makes it useless for the experiments in rats where this process occurs in the brain (Mathis et al., 2004).

• Imidazo[l,2-a]pyridine derivatives: developed as radioiodinated derivatives, these compounds have been reported to label β-amyloid plaques in transgenic animals as demonstrated by ex vivo autoradiography. (Kung et al., 2004) Recently, the [F-18]fluorine labeled derivatives were reported, but the initial in vivo experiments indicate significant level of de-fluorination resulting in progressive bone uptake of the F- 18 label (Cai et al., 2004)

• Stilbene derivatives: 4-([C-l l]rnethylamino)-4'-hydroxystylbene has been developed for the PET studies in humans (Ono et al., 2003) and a small study with 3 Alzheimer's disease patients and 3 controls was reported (Verhoef et al., 2004). The results are comparable to the results obtained with [C-11]6-OH-BTA in the same subjects. No data on animal studies has been reported so far.

• Acridine derivatives: [F-18]Fluorine derivatives were prepared, but no animal or human PET studies have been reported so far (Suemoto et al., 2004).

• Styrylbenzoxazoles: these compounds were designed to incorporate both the features of benzothiazole and stilbene derivatives. The [F-18]fluorine labeled compound BF-

168 was prepared and used in a transgenic mouse expressing beta-amyloid plaques. After 3 hours, ex vivo autoradiography shows uptake in the cortical areas. No PET or microPET experiments have been reported (Okamura et al., 2004).

None of the above described tracers could be successfully used for in vivo imaging of β-amyloid deposits in the brains of any animal model of Alzheimer's disease using PET or microPET.

The need for a reliable and quantitative method for in vivo imaging of the beta- amyloid load in the brains of humans, other primates, dogs, rats, mice, and other mammals, as well as transgenic animals used as animal models of Alzheimer's disease, is of greatest importance. Ideally such a method would allow for in vivo monitoring of emerging therapeutics designed to target the neuropathological deposits (e.g. beta-amyloid antibodies, anti-aggregation drugs, etc). The present invention addresses this need.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will become better understood when considered in conjunction with the following detailed description and accompanying drawings, wherein:

FIG. 1 is a series of [F-18] FDDNP Logan plot parametric images of a beta-amyloid triple transgenic rat brain, prepared in accordance with the present invention;

FIG. 2 is a series of [F-18] FDDNP Logan plot parametric images of a control rat brain (i.e., a normal rat without human transgenes); and

FIG. 3 is a pair of [F-18] FDDNP Logan plot parametric images of a tripe transgenic rat brain before and after naproxen treatment, in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

According to the invention, methods of monitoring the density of [..-amyloid plaques and/or neurofibrillary tangles in the brain of a human or other mammal as a function of time and/or medical treatment, are provided. In one embodiment, a method of monitoring the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of time comprises the steps of (a) injecting a radiolabeled compound comprising [F- 18]FDDNP or a derivative thereof into the bloodstream of a living mammal whose brain is known or suspected to contain β-amyloid plaques and/or neurofibrillary tangles; (b) scanning the mammal's brain using positron emission tomography; (c) quantifying radiolabeled compound binding in the mammal's cerebellum (which serves as a reference point) and in each of one or more selected regions of interest (ROI) in the brain; (d) calculating relative radiolabeled compound binding density in each selected ROI; and (e) repeating steps (a) - (d) one or more times, wherein each repetition of the four steps (a) - (d) is separated by a selected time interval, for example, one week, one month, one year, three years, etc. In some embodiments, the selected time intervals are all substantially the same, whereas in other embodiments the selected time intervals are not the same.

Preferably, one or more additional steps are included. For example, in some embodiments it is advantageous to compare the relative radiolabeled compound binding density in each selected ROI determined after a selected time interval with the relative radiolabeled compound binding density in each selected ROI determined from the initial scan of the mammal's brain. As another example, it can be advantageous to compare the relative radiolabeled compound binding density in each selected ROI after each scan of the mammal's brain with the relative radiolabeled compound binding density in each selected ROI with all other scans of the mammal's brain. As a third example, it can be advantageous to observe, over time, any trend of increasing, decreasing, or unchanging relative radiolabeled compound binding density for each selected ROI.

In an alternate embodiment, rather than using the cerebellum as a reference region, arterial blood (or arterialized venous blood) is withdrawn from the mammal and its radiolabeled compound content is measured. The relative radiolabeled compound binding density in each selected ROI is then determined by comparing the radiolabeled compound binding in each of the ROIs with the radiolabeled compound content of the withdrawn arterial blood (or arterialized venous blood). Nonlimiting examples of the types of mammal with which the present invention is to be used include humans, other primates, transgenic rats or mice, and dogs. In one embodiment, the mammal is a triple homozygous transgenic rat containing the following three transgenes: (a) Synapsin I - hAPP695(K670N/M671L), (b) PDGF-beta- hAPP695(K670N/M671L/V717F) minigene, and (c) Synapsin I - hPS-l(M146V). To practice the invention, a radiolabeled compound comprising [F-18]FDDNP or a derivative thereof is injected into the bloodstream of a living mammal. Suitable derivatives of [F-18]FDDNP include radiolabeled compounds of formula (I):

wherein Ri is selected from the group consisting of -C(O)-alkyl, -C(O)-alkylenyl-R₄, -C(O)O-alkyl, -C(O)O-alkylenyl-R₄, -C=C(CN)₂-alkyl, -C=C(CN)₂-alkylenyl-R₄,

radical selected from th

yl, aryl and substituted aryl; R₅ is a radical selected from the group consisting of

-NH₂, -OH, -SH, -NH-alkyl, -NHR₄, -NH-alkylenyl-R₄, -O-alkyl, -O-alkylenyl-R₄, -S-alkyl, and -S-alkylenyl-R₄; R₆ is a radical selected from the group consisting of -CN, -COOH, -C(O)O-alkyl, -C(O)O-alkylenyl-R₄, -C(O)-alkyl, -C(O)-alkylenyl-R₄, -C(O)-halogen, - C(O)NH₂, -C(O)NH-alkyl, -C(O)NH-alkylenyl-R₄; R₇ is a radical selected from the group consisting of O, NH, and S; R₈ is N; R₂ is selected from the group consisting of alkyl and alkylenyl-R₅ and R₃ is alkylenyl-R₅, and R₅ is selected from the group consisting of -OH, -OTs, halogen, spiperone, spiperone ketal, and spiperone-3-yl, or R₂ and R₃ together form a heterocyclic ring, optionally substituted with at least one radical selected from the group consisting of alkyl, alkoxy, OH, OTs, halogen, alkylenyl-R₅, carbonyl, spiperone, spiperone ketal and spiperone-3-yl. One or more of the hydrogen, halogen or carbon atoms is replaced with a radiolabel, specifically a positron emitter. For example, hydrogen or halogen can be replaced with ¹⁸F (sometimes denoted "[F-18]"), carbon can be replaced with ¹¹C; etc. A subclass of compounds of formula (I) are compounds of formula (II):

wherein R₂ is selected from the group consisting of alkyl and alkylenyl-Rio and R₃ is alkylenyl-Rio, wherein Rio is selected from the group consisting of -OH, -OTs, halogen, spiperone, spiperone ketal and spiperone-3-yl, or R₂ and R₃ together form a heterocyclic ring, optionally substituted with at least one radical selected from the group consisting of alkyl, alkoxy, OH, OTs, halogen, alkylenyl-Rio, carbonyl, spiperone, spiperone ketal and spiperone-3-yl, and R₉ is an alkyl, aryl or substituted aryl group, and to pharmaceutically acceptable salts and solvates thereof. As in formula (I), one or more of the hydrogen, halogen or carbon atoms is replaced with a radiolabel, specifically a positron emitter, such as

18p

In general, the in vivo monitoring described herein includes a scanning step and a quantitation of radiolabeled compound binding step. Nonlimiting embodiments of these steps will now be described.

Scanning

The mammal is placed on the PET camera bed (or microPET camera bed) in an appropriate position (e.g. conscious human subject in supine position, anesthetized primates in supine position, anesthetized rodents belly down). The head is brought in the center of the camera's field of view.

The animals are anesthetized either with injectable or gaseous anesthetics and are kept anesthetized throughout the whole experiment.

A solution of [F-18JFDDNP is injected as a bolus injection via an indwelling venous catheter (tail vein in the case of non-human mammals, a vein in the arm in the case of humans).

The solution for humans contains 5-10 mCi of [F-18JFDDNP in 10 mL of 12.5% human serum albumin, the solution for primates contains 1 - 10 mCi of [F-18]FDDNP in 1 - 5 mL of 10% ethanol in normal saline (0.9% NaCl), the solution for rodents contains 100 mCi - 4 mCi of [F-18]FDDNP in 0.1 - 0.8 mL of 10% ethanol in normal saline.

Sequential [F-18]FDDNP emission head scans were obtained with PET camera for human or primate scans starting at the time of the tracer injection start with the following scan sequence: six 30 sec scans, four 180 sec scans, five 600 sec scans, and three 1200 sec scans for the total duration of 2 hours. All PET images were reconstructed using filtered back-projection with measured attenuation correction. Persons skilled in the art will appreciate that positron emission tomography (PET) techniques and equipment used to image large mammals, such as humans and other primates, may not be appropriate for imaging smaller mammals, such as rats and mice. For small mammals, a smaller, specialized form of PET camera is useful. One such camera, sometimes referred to informally as a microPET camera, is available from Concord Microsystems, Inc. MicroPET is the company's registered trademark for its dedicated small animal PET scanner.

In the examples described and illustrated herein, using rodents, the emission head scan was performed continuously for one hour with microPET Concorde Focus 220 camera. The microPET images were reconstructed using filtered back-projection with the microPETManager software provided with the microPET camera by CTI Concorde Microsystems.

Quantitation of radiolabeled compound binding

It is possible to use various analytic techniques to calculate radiolabeled compound binding density in the regions of interest in the scanned mammal's brain. Among these are included the "rSUV" approach and the "rDV" approach. In addition, with the rDV approach, one can use either the Logan plot graphical analysis with the cerebellum as the reference region, or Logan plot graphical analysis with arterial input function. Each of these techniques will now be described.

1. Relative standardized uptake values (rSUV):

A set of regions of interest (ROIs) is drawn on all brain areas known to contain b- amyloid plaques and/or neurofibrillary tangles and also on cerebellum; time activity curves (TACs; showing the level of [F-18]FDDNP in a specific ROI as a function of time), are extracted for all ROIs including cerebellum; the values for the time points between 65 and 125 minutes (humans and primates) or between 30 and 60 minutes (rodents) are summed for each ROI; the resulting value for a specific ROI is normalized by dividing with the cerebella value; the resulting number presents a relative level of [F-18]FDDNP binding in a specific ROI over the cerebellum (relative standardized uptake value).

2. Relative distribution volume (rDV) using the cerebellum as the point of reference:

An ROI is drawn on a cortical area that is known to contain β-amyloid plaques and/or neurofibrillary tangles, another one is drawn on cerebellum; the time activity curves (TACs) are extracted and used for the Logan plot graphical analysis with cerebellum as reference region (Logan, 2003); once the linear relationship between these two TACs has been established then the slope of the resulting linear curve is extracted; the resulting value is distribution volume (DV), which describes the level of [F-18]FDDNP binding in the specific ROI; the same mathematical approach is used to determine distribution volume (DV) for every voxel in the 3D image, and a parametric image is generated containing the DV value for every voxel; a set of regions of interest (ROIs) is drawn on all brain areas known to contain β-amyloid plaques and/or neurofibrillary tangles and also on cerebellum; the average DV value for a specific ROI was normalized by dividing with the cerebellar DV value; the resulting relative distribution values (rDVs) were used to describe the level of [F-18]FDDNP in the specific ROI.

3. Relative distribution volume (rDV) using arterial (or arterialized venous) blood as the point of reference:

The methodology is similar to that described above, but arterial blood (or arterialized venous blood) is periodically withdrawn from the mammal; its radiolabeled compound content is measured; and it serves as the reference region for normalizing average DV values for specific ROIs. When using the rDV approach one can either use the Logan plot graphical analysis with the cerebellum as the reference region or one can use Logan plot graphical analysis with arterial input function.

The end result of Logan plot graphical analysis with arterial input function is similar to Logan plot graphical analysis with the cerebellum as the reference region: it is also the distribution volume of [F-18]FDDNP in the analyzed ROI. In this case, however, its calculation is based on the level of [F-18JFDDNP in the plasma instead on the level of [F- 18]FDDNP in the cerebellum. Logan plot graphical analysis with arterial input function requires that the research subject has a catheter in an artery through which blood is drawn at specific time points to determine the level of [F-18]FDDNP in the plasma (in humans we used ~ 24 time points). (Alternatively, a technique for arterializing venous blood is employed: a subject's hand is warmed, e.g., by immersion is warm water or by contact with a heating pad, causing capillaries in the tissues to open and allowing blood to be withdrawn as if it were arterial blood.)

A description of Logan plot graphical analysis with arterial input function is as follows: The TACs values are extracted for a specific ROI in the cortex. The level of [F- 18]FDDNP in the plasma is determined for each time point and is plotted out as a time activity curve (plasma input function). The values are used for the Logan plot graphical analysis with arterial input function. In brief, the method compares the integral of the [F- 18]FDDNP binding in the cortical ROI over time, divided by the amount of [F-18]FDDNP binding in the same ROI at the end point of the integral period, with the integral of the arterial plasma [F-18]FDDNP level over time, divided by the amount of [F-18]FDDNP binding in the cortical ROI at the end point of the integral period. A linear relationship can be found and the slope of such a function is the distribution volume of [F-18]FDDNP in the cortical ROI analyzed.

In the second step, the method calculates distribution volume (DV) value for each voxel and generates a parametric image in which every voxel has its DV value instead of actual [F-18]FDDNP activity density. One can draw the ROI on any brain region and directly extract the average value for DV and after normalizing it by the cerebellar DV value

(normalization is also performed with cerebellum despite the fact that arterial plasma input function was used for the DV calculation) the resulting relative distribution volume can be directly used for the comparison with other time point values in the same animal.

Here the same formula would apply as in the case of Logan plot graphical analysis with arterial input function: β-amyloid plaque and/or neurofibrillary tangle density = rDV = ([F-18]FDDNP distribution volume in an ROI)/( [F-18]FDDNP distribution volume in the cerebellum).

A voxel is the unit of space in the 3D tomographic image and is roughly 2 mm x 2 mm x 2 mm in size with the current human PET cameras - it is ~ 1.5 mm x 1.5 mm x 1.5 mm in size in the case of the Concorde Focus microPET camera. β-amyloid plaque and/or neurofibrillary tangle density in a specific region of the brain is proportional to relative distribution volume of [F-18]FDDNP in the same region. It is also proportional to the relative standardized uptake value of [F-18]FDDNP in that region. β-amyloid plaque and/or neurofibrillary tangle density = rSUV value = (summed [F- 18]FDDNP binding density over specific time in an ROI)/(summed [F-18]FDDNP binding density over specific time in cerebellum)

Specific time in humans and primates is 65 - 125 minutes; in rodents it is 30 - 60 minutes β-amyloid plaque and/or neurofibrillary tangle density = rDV = ([F-18JFDDNP distribution volume in an ROI)/( [F-18]FDDNP distribution volume in cerebellum).

A more detailed discussion of the graphical methods and analytical techniques used with radioisotopes in connection with PET is provided in J. Logan, A Review of Graphical Methods for Tracer Studies and Strategies to Reduce Bias, Nuclear Medicine and Biology 2003, 30, 833-844, the entire contents of which is incorporated by reference herein. In another aspect of the invention, an in vivo method of monitoring the density of β- amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of medical treatment is provided, and comprises (a) injecting a radiolabeled compound comprising [F-18]FDDNP or a derivative thereof into the bloodstream of a living mammal whose brain is known or suspected to contain β-amyloid plaques and/or neurofibrillary tangles; (b) scanning the mammal's brain using positron emission tomography; (c) quantifying radiolabeled compound binding in the mammal's cerebellum and in each of one or more selective regions of interest (ROI) in the brain; (d) calculating relative radiolabeled compound binding density in each selected ROI; (e) administering to the mammal a first medical treatment; (f) repeating steps (a) - (d); (g) administering to the mammal one or more additional medical treatments; and (h) repeating steps (a) - (d). The process can be continued with subsequent administrations of medical treatment, PET scanning, and quantitation/evaluation of effect on β-amyloid plaques and/or neurofibrillary tangles. Also included within the scope of the invention are variations in which, before or after administering a particular medical treatment, one or more additional steps are carried out. For example, in some embodiments it is advantageous to compare the relative radiolabeled compound binding density in each selected ROI determined after a selected time interval or medical treatment with the relative radiolabeled compound binding density in each selected ROI determined from the initial scan of the mammal's brain. As another example, it can be advantageous to compare the relative radiolabeled compound binding density in each selected ROI after each scan of the mammal's brain with the relative radiolabeled compound binding density in each selected ROI with all other scans of the mammal's brain. As a third example, it can be advantageous to observe, as a function of time and/or medical treatment, any trend of increasing, decreasing, or unchanging relative radiolabeled compound binding density for each selected ROI.

The medical treatments referred to include, without limitation, any intervention that could potentially affect, directly or indirectly, the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of the subject mammal. Non-limiting examples include treatment with antibodies directed against the β-amyloid protein or hyperphosphorylated tau protein; treatments focused at decreasing the neuroinflammation levels within the brain (and thereby retarding or slowing down other neuropathological processes); treatment with substances having demonstrated anti-aggregation effect, either with β-amyloid aggregates and/or with hyperphosphorylated tau protein aggregates; etc. As one example, it is known that Naproxen inhibits [F-18]FDDNP binding to β-amyloid fibrils in vitro. In the method described above, Naproxen can be used as the medical treatment, and its effect on β-amyloid plaques and/or neurofibrillary tangles can be monitored. Regardless of whether a medical treatment is administered to the subject mammal, in vivo monitoring of the density of β- amyloid plaques and/or neurofibrillary tangles can proceed using either the cerebellum as the reference region, or arterial blood withdrawn from the mammal.

The following describes one embodiment of a method of monitoring the density of β- amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of medical treatment. For convenience, the radiolabeled compound is identified as [F- 18]FDDNP, but it will be understood that a derivative of [F-18]FDDNP can also be used. • A research subject receives a [F-18]FDDNP brain scan before the start of a selected medical treatment (for example, administration of naproxen). The data is analyzed as described above and the resulting quantitative values are used as the baseline values (controls).

• The treatment regimen is started.

• The subject receives another [F-18]FDDNP brain scan after a pre-determined period of time and the data is analyzed as described above. The resulting values are compared with the baseline values to gauge the effectiveness of the treatment. • The treatment regimen is continued and the [F-18JFDDNP brain scans were performed at regular intervals.

• The effectiveness of the treatment is gauged based on the comparison of the [F-18]FDDNP binding values at the specific time point with the baseline value (using either rDV or rSUV approach); if the treatment is effective it will either cause the decrease of total β-amyloid and/or neurofibrillary tangle load (pathology load) in the specific brain region (or in the whole brain) or it will prevent the increase in the brain pathology load. In the first case a reduction of [F-18]FDDNP rDV or [F-18JFDDNP rSUV values as a function of time will be observed. In the second case the [F-18]FDDNP binding values (rDVs or rSUVs) will remain unchanged. • In the case of ineffective treatment, the total brain pathology load will increase with time and one will see the increase of [F-18]FDDNP rDV or [F-18]FDDNP rSUV values as a function of time.

• The treatment regimen and [F-18]FDDNP brain PET (or microPET) scanning is continued until enough time points are established for a satisfactory determination of the treatment effectiveness (increased, decreased or unchanged [F-18]FDDNP binding values (rDVs or rSUVs).

The following, non-limiting example illustrates the invention.

Example 1.

We have visualized for the first time β-amyloid load in the living brain of transgenic rodents. We used the triple transgenic (TG+) rat model of β-amyloid deposition, recently developed by Cephalon Inc. with [F-18]FDDNP and microPET. Three TG+ rats (15 months) and two control rats (> 9 months) were injected with [F-18]FDDNP and dynamic scans (up to 60 min) were performed with the Concorde Focus microPET (two TG+ and one control were scanned twice). Sequential brain images were analyzed using Logan plot graphical analysis with cerebellum as reference region. We observed only background cortical binding in control rats, comparable to binding observed in cerebellum (rDV: frontal 1.04 ± 0.01; hippocampus 0.99±0.04). In contrast, all transgenic animals had binding in cerebral cortex elevated, most prominently in the frontal lobe (1.32±0.04) and in the hippocampal region (1.23±0.05). In separate experiments cortical [F-18]FDDNP binding in TG+ animals (N=2) was blocked with naproxen (3 doses/8 mg starting 26 h before scanning) lowering cortical rDV values to the levels observed in controls. These results mimic previous in vitro results with β-amyloid neurofibrils and brain specimens of Alzheimer's disease patients (Agdeppa et al, Neuroscience 117, 723, 2003).

Figure 1 presents [F- 18] FDDNP Logan plot parametric images of a beta-amyloid triple transgenic rat brain. [F-18]FDDNP microPET dynamic data sets were analyzed using Logan plot graphical analysis with the cerebellum as the reference region. The resulting parametric image shows distribution volume (DV) values. The left side of the panel shows brain in horizontal (A), sagittal (C) and transverse (E) cuts through the rat's head. The right side of the panel shows the [F-18JFDDNP parametric images laid over the brain images (B, D, and F). The anatomical reference allows for identification of the regions with increased [F-18]FDDNP binding (elevated DV values) in the cortex and hippocampus. In contrast, the cerebellum has only background level of binding. Figure 2 presents [F-18JFDDNP Logan plot parametric images of a control rat brain

(normal rat without human transgenes). The left side of the panel shows the brain in horizontal (A) and sagittal (C) cuts through the rat brain. [F-18]FDDNP parametric image of a controlled brain (B and D) shows no significant increase in [F-18JFDDNP distribution volumes in the cortex and hippocampus when compared with the cerebellum. The color scale is identical to the color scale in Figure 1.

Figure 3 presents [F-18JFDDNP Logan plot parametric images of a triple transgenic rat brain before and after naproxen treatment. [F-18]FDDNP distribution volumes in the cortex and hippocampus were increased when compared to the cerebellum (left), after the oral treatment with naproxen [F-18JFDDNP, distribution volumes in cortex and hippocampus were significantly decreased in the same animal (right). The color scale is identical to the color scales in Figures 1 and 2.

Conclusions: We have performed the first successful microPET visualization of brain pathology in the living brain of any rodent transgenic model of β-amyloid or tau deposition in Alzheimer's disease. The in vivo [F-18]FDDNP binding is consistent with the know distribution of cortical pathology in these animals and is blocked by naproxen in the same regions.

The invention has been described with reference to various embodiments and examples, but is not limited thereto. Persons skilled in the art will appreciate that other modifications can be made without departing from the invention or its fair scope, which is limited only by the appended claims.

Claims

WHAT IS CLAIMED IS:

1. An in vivo method of monitoring the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of time, comprising:

(a) injecting a radiolabeled compound comprising [F-18]FDDNP or a derivative thereof into the bloodstream of a living mammal whose brain is known or suspected to contain β-amyloid plaques and/or neurofibrillary tangles;

(b) scanning the mammal's brain using positron emission tomography;

(c) quantifying radiolabeled compound binding in the mammal's cerebellum and in each of one or more selected regions of interest (ROI) in the brain;

(d) calculating relative radiolabeled compound binding density in each selected ROI; and

(e) repeating steps (a) - (d) one or more times, wherein each repetition is separated by a selected time interval.

2. An in vivo method as recited in claim 1, wherein the selected time interval ranges from 1 week to 3 years.

3. An in vivo method as recited in claim 1, wherein all selected time intervals are the same.

4. An in vivo method as recited in claim 1, further comprising comparing the relative radiolabeled compound binding density in each selected ROI determined after a selected time interval with the relative radiolabeled compound binding density in each selected ROI determined from the initial scan of the mammal's brain.

5. An in vivo method as recited in claim 1, further comprising comparing the relative radiolabeled compound binding density in each selected ROI after each scan of the mammal's brain with the relative radiolabeled compound binding density in each selected ROI with all other scans of the mammal's brain.

6. An in vivo method as recited in claim 1, further comprising observing over time any trend of increasing, decreasing, or unchanging relative radiolabeled compound binding density for each selected ROI.

7. An in vivo method as recited in claim 1, wherein the [F-18]FDDNP derivative comprises a radiolabeled compound having the formula (I):

radical selected from the group ^^"^ consisting of alkyl, substituted alkyl, aryl and substituted aryl; R₅ is a radical selected from the group consisting of

-NH₂, -OH, -SH, -NH-alkyl, -NHR₄, -NH-alkylenyl-R₄, -O-alkyl, -O-alkylenyl-R₄, -S-alkyl, and -S-alkylenyl-R₄; R₆ is a radical selected from the group consisting of -CN, -COOH, -C(O)O-alkyl, -C(O)O-alkylenyl-R₄, -C(O)-alkyl, -C(O)-alkylenyl-R₄, -C(O)-halogen, - C(O)NH₂, -C(O)NH-alkyl, -C(O)NH-alkylenyl-R₄; R₇ is a radical selected from the group consisting of O, NH, and S; R₈ is N; R₂ is selected from the group consisting of alkyl and alkylenyl-R₅ and R₃ is alkylenyl-Rs, and R₅ is selected from the group consisting of -OH, -OTs, halogen, spiperone, spiperone ketal, and spiperone-3-yl, or R₂ and R₃ together form a heterocyclic ring, optionally substituted with at least one radical selected from the group consisting of alkyl, alkoxy, OH, OTs, halogen, alkylenyl-Rs, carbonyl, spiperone, spiperone ketal and spiperone-3-yl, wherein one or more hydrogen, halogen or carbon atoms is replaced with a positron-emitting radiolabel.

8. An in vivo method as recited in claim 1, wherein relative radiolabeled compound binding density is calculated using rSUV or rDV methodology.

9. An in vivo method as recited in claim 1, wherein the mammal is a primate.

10. An in vivo method as recited in claim 1, wherein the mammal is a human.

11. An in vivo method as recited in claim 1, wherein the mammal is a transgenic rat or mouse.

12. An in vivo method as recited in claim 1, wherein the mammal is a triple homozygous transgenic rat containing the following three transgenes:

(a) Synapsin I - hAPP695(K670N/M671L),

(b) PDGF-beta-hAPP695(K670N/M671I.yV717F) minigene, and

(c) Synapsin I - hPS-l(M146V).

13. An in vivo method as recited in claim 1, wherein the mammal is a dog.

14. An in vivo method of monitoring the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of time, comprising:

(a) injecting a radiolabeled compound comprising [F-18JFDDNP or a derivative thereof into the bloodstream of a living mammal whose brain is known or suspected to contain β-amyloid plaques and/or neurofibrillary tangles;

(b) scanning the mammal's brain using positron emission tomography;

(d) calculating relative radiolabeled compound binding density in each selected ROI;

(f) observing over all time periods any trend of increasing, decreasing, or unchanging relative radiolabeled compound binding density for each selected ROI.

15. An in vivo method of monitoring the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of medical treatment, comprising: (a) injecting a radiolabeled compound comprising [F-18]FDDNP or a derivative thereof into the bloodstream of a living mammal whose brain is known or suspected to contain β-amyloid plaques and/or neurofibrillary tangles;

(b) scanning the mammal's brain using positron emission tomography;

(e) administering to the mammal a first medical treatment;

(f) repeating steps (a) - (d);

(g) administering to the mammal one or more additional medical treatments; and

(h) repeating steps (a) - (d).

16. An in vivo method as recited in claim 15, further comprising comparing the relative radiolabeled compound binding density in each selected ROI determined after the first medical treatment and each additional medical treatments with the relative radiolabeled compound binding density in each selected ROI determined after the initial scan of the mammal's brain.

17. An in vivo method as recited in claim 15, further comprising observing over time any trend of increasing, decreasing, or unchanging relative radiolabeled compound binding density for each selected ROI, as a function of medical treatment.

18. An in vivo method as recited in claim 15, wherein the first medical treatment comprises treatment with antibodies directed against a β-amyloid protein or hyperphosphorylated tau protein.

19. An in vivo method as recited in claim 15, wherein the first medical treatment comprises decreasing neuroinflammation in the mammal's whole brain or one or more ROI.

20. An in vivo method as recited in claim 15, wherein the first medical treatment comprises treatment with a substance having a demonstrated anti-aggregation effect against β-amyloid aggregates and/or hyperphosphorylated tau protein aggregates.

21. An in vivo method as recited in claim 15, wherein the [F-18]FDDNP derivative comprises a radiolabeled compound having the formula (I):

radical selected from th

-NH₂, -OH, -SH, -NH-alkyl, -NHR₄, -NH-alkylenyl-R₄, -O-alkyl, -O-alkylenyl-R₄, -S-alkyl, and -S-alkylenyl-R₄; R₆ is a radical selected from the group consisting of -CN, -COOH,

-C(O)O-alkyl, -C(O)O-alkylenyl-R₄, -C(O)-alkyl, -C(O)-alkylenyl-R₄, -C(O)-halogen, - C(O)NH₂, -C(O)NH-alkyl, -C(O)NH-alkylenyl-R₄; R₇ is a radical selected from the group consisting of O, NH, and S; R₈ is N; R₂ is selected from the group consisting of alkyl and alkylenyl-R₅ and R₃ is alkylenyl-R₅, and R₅ is selected from the group consisting of -OH, -OTs, halogen, spiperone, spiperone ketal, and spiperone-3-yl, or R₂ and R₃ together form a heterocyclic ring, optionally substituted with at least one radical selected from the group consisting of alkyl, alkoxy, OH, OTs, halogen, alkylenyl-R₅, carbonyl, spiperone, spiperone ketal and spiperone-3-yl, and wherein one or more hydrogen, halogen or carbon atoms is replaced with a positron-emitting radiolabel.

22. An in vivo method as recited in claim 15, wherein relative radiolabeled compound binding density is calculated using rSUV or rDV methodology.

23. An in vivo method as recited in claim 15, wherein the mammal is a primate.

24. An in vivo method as recited in claim 15, wherein the mammal is a human.

25. An in vivo method as recited in claim 15, wherein the mammal is a dog.

26. An in vivo method as recited in claim 15, wherein the mammal is a transgenic rat or mouse.

27. An in vivo method as recited in claim 15, wherein the mammal is a triple homozygous transgenic rat containing the following three transgenes:

(a) Synapsin I - hAPP695(K670N/M671L),

(b) PDGF-beta-hAPP695(K670N/M671L/V717F) minigene, and

(c) Synapsin I - hPS-l(M146V).

28. An in vivo method of monitoring the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of time, comprising:

(b) scanning the mammal's brain using positron emission tomography;

(c) withdrawing arterial blood, or arterialized venous blood, from the mammal and measuring its radiolabeled compound content;

(d) quantifying radiolabeled compound binding in each of one or more selected regions of interest (ROI) in the brain;

(e) calculating relative radiolabeled compound binding density in each selected ROI; and

(f) repeating steps (a) - (e) one or more times after a selected time interval.

29. An in vivo method of monitoring the density of β-amyloid plaques and/or neurofibrillary tangles in the brain of a mammal as a function of medical treatment, comprising:

(b) scanning the mammal's brain using positron emission tomography;

(d) quantifying radiolabeled compound binding in the mammal's withdrawn blood and in each of one or more selected regions of interest (ROI) in the brain;

(e) calculating relative radiolabeled compound binding density in each selected ROI;

(f) administering to the mammal a first medical treatment;

(g) repeating steps (a) - (e);

(h) administering to the mammal one or more additional medical treatments; and

(i) repeating steps (a) - (e).

30. An in vivo method as recited in any one of claims 1-29, wherein the [F-

18]FDDNP derivative comprises a radiolabeled compound having the formula (II):

wherein R₂ is selected from the group consisting of alkyl and alkylenyl-R₁₀ and R₃ is alkylenyl-Rio, wherein Rio is selected from the group consisting of -OH, -OTs, halogen, spiperone, spiperone ketal and spiperone-3-yl, or R₂ and R₃ together form a heterocyclic ring, optionally substituted with at least one radical selected from the group consisting of alkyl, alkoxy, OH, OTs, halogen, alkylenyl-Rio, carbonyl, spiperone, spiperone ketal and spiperone-3-yl, and R₉ is an alkyl, aryl or substituted aryl group, and pharmaceutically acceptable salts and solvates thereof; and wherein one or more of hydrogen, halogen or carbon atoms is replaced with a positron emitting radiolabel.