Diagnosis of Dementia by Vascular Magnetic Resonance Imaging

Abstract

A method of diagnosing a likelihood of onset or progression of Alzheimer's disease and related dementias (ADRD) in a subject is provided. The method requires determining vascularization changes in different regions of the brain on the basis of a quantitative cerebral blood volume (qCBV) map of the subject's brain. The qCBV is obtained from one or more quantitative ultrashort time-to-echo contrast-enhanced (QUTE-CE) MRI images of the brain. A method of treating a subject for ADRD is provided. Diagnostic markers for onset and progression of Alzheimer's disease are also provided.

Claims

1. A method of diagnosing onset or progression of Alzheimer's Disease or Related Dementias (ADRD) in a subject, the method comprising the steps of: (a) obtaining one or more quantitative ultrashort time-to-echo contrast-enhanced (QUTE-CE) MRI images of the subject's brain; (b) producing a quantitative cerebral blood volume (qCBV) map of the subject's brain from the image; (c) determining regions of hypervascularization and regions of hypovascularization in the subject's brain based on comparing the qCBV map obtained in step (b) to a pre-determined qCBV map representative of a normal brain; and (d) diagnosing the subject's likelihood of onset of ADRD or progression of ADRD based on an analysis of the hypervascularization and hypovascularization regions.

2. The method of claim 1, wherein in step (d), a greater number of hypervascularization regions relative to hypovascularization regions indicates that onset of ADRD has occurred in the subject.

3. The method of claim 1, wherein in step (d), a greater number of hypovascularization regions relative to hypervascularization regions indicates that progression of ADRD has occurred in the subject.

4. The method of claim 1 further comprising repeating the method at a later time, wherein a decrease in the number or extent of hypovascularization regions found in step (d) at the later time indicates progression of ADRD in the subject.

5. The method of claim 1, further comprising repeating the method at a later time, wherein an increase in the number or extent of hypervascularization regions found in step (d) at the later time indicates progression of ADRD in the subject.

6. The method of claim 1, wherein hypovascularization and/or hypervascularization is determined based on a measurement of microvasculature, capillary density, or mean vascularity.

7. The method of claim 1, wherein progression of ADRD is indicated in the subject, and wherein a degree of hypovascularization and/or hypervascularization indicates a degree of progression of ADRD in the subject.

8. The method of claim 1, wherein said hypervascularization is in one or more regions of the subject's brain selected from the group consisting of ventral tegmental area, raphe linear, reticulotegmental nucleus, raphe obscurus nucleus, habenula nucleus, median raphe nucleus, dorsomedial tegmental area, dorsal raphe, pontine nuclei, raphe magnus, ventral subiculum, motor trigeminal nucleus, copula of the pyramis, pontine reticular nucleus caudal, pontine reticular nucleus oral, trapezoid body, subiculum dorsal, parabrachial nucleus, reticular nucleus midbrain, retrosplenial caudal ctx, pedunculopontine tegmental area, red nucleus, sub coeruleus nucleus, PCRt, inferior colliculus, facial nucleus, 9th cerebellar lobule, gigantocellular reticular nucleus, principal sensory nucleus trigeminal, entorhinal ctx, root of trigeminal nerve, visual 1 ctx, 10th cerebellar lobule, prelimbic ctx, precuniform nucleus, infralimbic etx, superior colliculus, solitary tract nucleus, and periaqueductal gray thalamus.

9. The method of claim 1, wherein said hypovascularization is in one or more regions of the subject's brain selected from the group consisting of ventral tegmental area, raphe linear, reticulotegmental nucleus, raphe obscurus nucleus, habenula nucleus, median raphe nucleus, dorsomedial tegmental area, dorsal raphe, pontine nuclei, raphe magnus, ventral subiculum, motor trigeminal nucleus, copula of the pyramis, pontine reticular nucleus caudal, pontine reticular nucleus oral, trapezoid body, subiculum dorsal, parabrachial nucleus, reticular nucleus midbrain, retrosplenial caudal ctx, pedunculopontine tegmental area, red nucleus, sub coeruleus nucleus, PCRt, inferior colliculus, facial nucleus, 9th cerebellar lobule, gigantocellular reticular nucleus, principal sensory nucleus trigeminal, entorhinal ctx, root of trigeminal nerve, visual 1 ctx, 10th cerebellar lobule, prelimbic ctx, precuniform nucleus, infralimbic etx, superior colliculus, solitary tract nucleus, and periaqueductal gray thalamus.

10. The method of claim 1, wherein said hypervascularization is in one or more regions of the subject's brain selected from the group consisting of paraventricular nucleus, ventral subiculum, dorsal raphe, visual 2 ctx, dorsomedial tegmental area, inferior colliculus, motor trigeminal nucleus, primary somatosensory ctx trunk, triangular septal nucleus, ventral medial striatum, lateral preoptic area.

11. The method of claim 1, wherein said hypovascularization is in one or more regions of the subject's brain selected from the group consisting of paraventricular nucleus, ventral subiculum, dorsal raphe, visual 2 ctx, dorsomedial tegmental area, inferior colliculus, motor trigeminal nucleus, primary somatosensory ctx trunk, triangular septal nucleus, ventral medial striatum, lateral preoptic area.

12. The method of any of claims 1-11, wherein obtaining QUTE-CE MRI images comprises introducing a paramagnetic or superparamagnetic contrast agent into the brain of the subject.

13. The method of claim 12, wherein the paramagnetic or superparamagnetic contrast agent is selected from the group consisting of iron oxide nanoparticles, a gadolinium chelate, and a gadolinium compound.

14. The method of claim 13, wherein the iron oxide nanoparticles comprise a material selected from the group consisting of Fe.sub.3O.sub.4 (magnetite), y-Fe.sub.2O.sub.3 (maghemite), a-Fe.sub.2O.sub.3 (hematite), ferumoxytol, ferumoxides, ferucarbotran, and ferumoxtran.

15. The method of claim 14, wherein the iron oxide nanoparticles comprise ferumoxytol.

16. A method of treating ADRD in a human subject, the method comprising: performing the method of claim 1 to diagnose the onset or progression of ADRD in the human subject; and treating the human subject for ADRD.

17. The method of claim 16, wherein the step of treating comprises administering a cholinesterase inhibitor, such as donepezil, rivastigmine, galantamine, memantine; an antidepressant such as citalopram, fluoxetine, paroxeine, sertraline, or trazodone; an anxiolytic, such as lorazepam or oxazepam; or an antipsychotic, such as aripiprazole, clozapine, haloperidol, olanzapine, quetiapine, risperidone, or ziprasidone or any combination thereof.

18. The method of claim 16, wherein the step of treating the human subject comprising applying a behavioral therapy; such as changing an environment, redirecting attention, avoiding a confrontation, providing rest, or monitoring one or more of pain, hunger, thirst, constipation, full bladder, fatigue, infection, skin irritation, and room temperature; and any combination thereof.

Description

DESCRIPTION OF THE DRAWINGS

[0029] FIGS. 1A-IC show representative images of QUTE-CE MRI raw intensity rendered in 3DSlicer. FIG. 1A shows a pre-contrast image (before ferumoxytol contrast injection), and FIG. 1B a post-contrast image (after injection of ferumoxytol 14 mg/kg). Rendering parameters are equivalent. FIG. 1C shows a segmented brain from FIG. 1B. Relevant parameters include: 3D Radial UTE; FOV 3×3×3 cm.sup.3; matrix mesh size 200×200×200; TE 13 μs; TR 3.5 ms; and 0=20°, scan time=8 m 22 seconds, 2 averages.

[0030] FIG. 2 illustrates a QUTE-CE MRI image analysis pipeline and biomarker measurements in Sprague Dawley rats. Regarding the method: In the top-left image: Null-contrast and angiographic images are obtained pre- and post-ferumoxytol injection respectively. Note how the contrast in the maximum Intensity Projection (MIP) images is almost solely due to ferumoxytol presence at 7T, and blood flow effects are confined to the periphery. Field corrections for coil sensitivity (B1) and flip-angle distribution (B1+) are applied along with motion correction between the pre- and post-contrast images. In the middle-left image: The qCBV is calculated by cropping out the brain and applying the corresponding formula with the quantitative intensity values. In the bottom-left imnage: Regions are characterized by distributing the 500,000 voxels within the brain into a 173-region atlas via co-registration affine transform using EVA Software (Ekam Solutions, Boston, Mass. USA). In the middle image: A vascular atlas was constructed of Male Sprague Dawley rats (9-10 weeks old, n=11) and (right) CO.sub.2 can was applied to measure vascular responsivity. Of note, imaging was performed on awake, yet mechanically restrained, rats to simulate neurophysiological conditions for human-brain clinical imaging experiments. Thus, the direction and magnitude of vascular changes during isoflurane anesthesia were also determined for rats (labeled “ISO”).

[0031] FIG. 3A illustrates hypovascularity of wild-type (WT) female vs. APOE4 female rats at 24 months of age. FIG. 3B illustrates vascular density—mean of wild-type female vs. APOE4 female rats at 8 months and 24 months of age. FIG. 3C illustrates capillary density (or small vessels)—mode of wild-type female vs. APOE4 female rats at 8 months and 24 months of age.

[0032] FIG. 4A illustrates an APOE4 and WT comparison of mean vascular abnormality detected at 8 months. FIG. 4B illustrates an APOE4 and WT comparison of microvascular abnormality detected by region at 8 months.

[0033] FIG. 5A is a graph of vascular abnormalities for small vessels and all vessels at 8 months and at 2 years, illustrating a hypovascular trend at 2 years. FIG. 5B is a graph of CO.sub.2 challenge demonstrating metabolic dysfunction with CO.sub.2 channels in APOE4 and WT.

[0034] FIG. 6 is a table of results of the mean of structural differences at 8 months.

[0035] FIG. 7 is a graphical illustration of the results of FIG. 6.

[0036] FIG. 8 is a table of results of the mode of structural differences at 8 months.

[0037] FIG. 9 is a graphical illustration of the results of FIG. 8.

[0038] FIG. 10 is a table of results of the mean of structural differences at 2 years.

[0039] FIG. 11 is a graphical illustration of the results of FIG. 10.

[0040] FIG. 12 is a table of results of the mode of structural differences at 2 years.

[0041] FIG. 13 is a graphical illustration of the results of FIG. 12.

[0042] FIG. 14 is a table of results of WT mean differences in CO.sub.2 challenge states for WT rats at 2 years.

[0043] FIG. 15 is a table of results of WT mode differences in CO.sub.2 challenge states for WT rats at 2 years.

[0044] FIG. 16 is a table of results of APOE mean differences in CO.sub.2 challenge states for APOE rats at 2 years.

[0045] FIG. 17 is a table ofresults of APOE mode differences in CO.sub.2 challenge states for APOE rats at 2 years.

DETAILED DESCRIPTION

[0046] The present technology utilizes an imaging modality termed quantitative ultra-short time-to-echo contrast-enhanced (QUTE-CE) MRI (Gharagouzloo et al; 2017; Gharagouzloo et al, 2015), which is able to overcome the semi-quantitative nature of the MRI signal. The preliminary data demonstrate that QUTE-CE MRI can provide a highly accurate quantitative map of the brain “vasculome” from small to large vessels, offering a firm basis for assessment of microvascular contribution to brain function and neurological disorders.

[0047] The present inventors have utilized an imaging modality, QUTE-CE MRI, to study the micro- and macro-vascular abnormalities in an APOE-ε4 knock-in rat model. It is noteworthy that the APOE-ε4 allele is the single most important genetic risk factor for AD. The study involved characterizing vascular changes in 173 regions of the brain. While the 173-region characterization revealed both hyper- and hypovascularization, the changes in microvascularity were almost entirely hypervascular early on (rats at 8 months), and hypovascular later in life (24 months).

[0048] The resolution and sensitivity are such that QUTE-CE can map the physiological CBV across the entire rat brain in 500,000 small volumes (or voxels), which can be distributed over 173 anatomically distinct 3D volumes for network-level analysis indicating capillary density, small vessel responsivity and vascular reserve. Importantly, QUTE-CE MRI can be immediately translated and there is an ongoing human clinical trial to map the healthy brain vascularity of a small group of individuals. The preliminary preclinical data in an ApoE4 human-knock in genetically modified rat model for AD reveals a trend of hyper- to hypo-vascularization early in development before cognitive decline.

[0049] QUTE-CE MRI is a method that utilizes a 3D UTE pulse sequence and an intra-vascular contrast agent (CA) to render high contrast-to-noise ratio (CNR) vascular images with quantitative signal. (See WO 2017/019812, incorporated by reference herein.) At this ultrashort TE, contrast is inverted from the typical negative-contrast obtained from superparamagnetic iron-oxide nanoparticles (SPIONs) to purely TI-enhanced positive-contrast. Quantitative measurements of the micro- and macro-vasculature can be mapped throughout the whole rat brain and functional changes of state can also be measured (Gharagouzloo et al., 2017). The absolute cerebral blood volume (qCBV) is calculated by simple partial volume calculations using a two-compartment model for signal from blood and tissue:

I.sub.M=f.sub.BI.sub.B+(I−f.sub.B)I.sub.T  (1)

where I.sub.M is the measured signal intensity at each voxel, I.sub.T is the tissue intensity, I.sub.B is the blood intensity and f.sub.B is the fraction of the voxel occupied by blood. qCBV is calculated voxel-wise by subtracting a pre-contrast intensity from post-contrast after catheter injection of ferumoxytol (Feraheme, AMAG Pharmaceuticals, Waltham, Mass., USA, 7 mg/kg). Solving for the blood volume fraction,

[00001]$\begin{array}{cc}\mathrm{qCBV}=f_B=\frac{I_M^{\prime }-I_M}{I_B^{\prime }-I_B}& \left(2\right)\end{array}$

is measured in whole brain, with blood reference signal intensity taken from the superior sagittal sinus (SSS), per image. Images are then fit to a 173 anatomic region atlas for segmented analyses using anatomical scans. The mean qCBV is reported per region, and the mode is taken as a proxy for microvessel density.

[0050] Quantitative vascular mapping of the rat brain begins with acquisition of pre- and post-ferumoxytol scans. A 3D UTE sequence with optimized parameters for blood contrast and quantification is utilized. Field corrections for coil sensitivity (B1−) and flip-angle distribution (B1+) are applied along with motion correction between the pre- and post-contrast images. A voxel-wise calculation for the quantitative CBV (qCBV) is performed to produce the qCBV map using a two-volume blood/tissue model with knowledge of blood intensity obtained from large vessels. See FIGS. 1A-1C for image contrast, see FIG. 2 for image processing pipeline. For the rat models described here, voxels are distributed into an anatomically segmented atlas with 173 regions for quantitative analysis of the whole brain, both in terms of the mean vascularity and the microvessel density obtained from a regional characterization of the model. Statistically significant abnormalities are found by comparing healthy, normal vasculature to genetically modified model of disease in rats.

[0051] The procedure for producing QUTE-CE MRI images of a subject is described in detail in WO 2017/019182. As applied to a subject's brain, a magnetic field is applied to the brain followed by application of a radio frequency pulse sequence at a selected repetition time (TR) and application of a magnetic field gradient to provide a selected flip angle (FA) to excite protons in the region of interest. Generally, the repetition time is less than about 10 ms, and the FA ranges from about 100 to about 30°, around the Ernst angle of doped blood or at the angle of maximum contrast between doped- and undoped-blood. A response signal is measured during relaxation of the protons at a selected time-to-echo (TE) and a T1-weighted signal acquired. The time to echo is an ultra-short time to echo and is set to less than about 300 μs. In this manner, an image of the brain is generated. Next, a paramagnetic or superparamagnetic CA is introduced into the brain of the subject by injecting the agent into vasculature. The acquired signal is representative of a concentration of the CA in the brain and the amount of blood in a particular region in that region of the brain.

[0052] The TE can, for example, be set at less than 180 μs, 160 μs, 140 μs, 120 μs, 100 μs, 90 μs, 80 μs, 70 μs, 60 μs, 50 μs, 40 μs, 30 μs, 20 μs, or 10 μs. Also, the time to echo can be set to less than a time in which blood volume displacement in the region of interest in the brain is about one order of magnitude smaller than a voxel size. The TR can be set to a value from about 2 to about 10 ms. The image of the ROI can have a contrast to noise ratio (CNR) of at least 4, at least 5, at least 10, at least 15, at least 20, at least 30, at least 40, at least 50, or at least 60. The CNR is determined between an ROI represented in a post-contrast image and a pre-contrast image. For example, a CNR is found by examining an ROI in the SSS and taking the difference between the two signal to noise ratios (SNRs)—which itself is defined by the mean of any given ROI divided by the standard deviation of the noise in that respective image. The response signal can be measured along trajectories in k-space in which total acquisition time can be longer than TE. The magnetic field can have a strength ranging from 0.2 T to 14.0 T.

[0053] It should be noted that the region of interest (ROI), such as a particular area of the brain, can comprises a volume fraction occupied by blood and a volume fraction occupied by tissue. Determining the volume fraction occupied by blood comprises, prior to introducing the CA to the ROI, applying the radio frequency pulse sequence at the selected TR to excite protons in the region of interest, and measuring a response signal during relaxation of the protons at the selected TE to acquire a signal from the ROI; and comparing signal intensities of the ROI prior to introducing the CA and after introducing the contrast agent.

[0054] Any paramagnetic formulation that provides contrast in MRI can serve as a CA for the QUTE-CE method. Compounds containing paramagnetic iron-oxide nanoparticles, gadolinium-based contrast agents (GBCAs), such molecular chelates or nanoparticles, or manganese nanoparticles, can serve as CAs. If the CA is ferumoxytol, for example, the CA is introduced in the blood at a concentration of 0.1 to 15 mg/kg. The nanoparticles can be delivered by a bolus intravenous or intraarterial injection, which may be optionally repeated. Paramagnetic and superparamagnetic nanoparticles can serve as the CA. Paramagnetic molecular chelates and superparamagnetic nanoparticles can serve as the CA. Example paramagnetic nanoparticles can be iron oxide, gadolinium, or manganese nanoparticles. Iron oxide nanoparticles can be Fe.sub.3O.sub.4 (magnetite), y-Fe.sub.2O.sub.3 (maghemite), a-Fe.sub.2O.sub.3 (hematite), ferumoxytol, ferumoxides, ferucarbotran, or ferumoxtran. Iron oxide nanoparticles can be coated with a carbohydrate and have a diameter from about 1 nm and about 999 nm, or from about 2 nm and about 100 nm, or from about 10 nm and about 100 nm, measured with dynamic light scattering. Nanoparticle CAs can have other coatings that allow them to circulate in the blood. Nanoparticles CAs can have various sizes that allow them to either be excreted by the kidneys or by the liver. Some gadolinium compounds include gadofosveset trisodium, gadoterate meglumine, gadoxetic acid disodium salt, gadobutrol, gadopentetic dimeglumine, gadobenate dimeglumine, gadodiamide, gadoversetamide, or gadoteridol.

[0055] Blood volume fraction in an ROI is determined as described in the following. Initially, a magnetic field is applied. A radio frequency pulse sequence is applied next at a selected TR and at a magnetic field gradient to provide a selected flip angle to excite protons in the region of interest. The TR is less than about 10 ms and the flip angle ranges from about 100 to about 30°. A response signal is measured during relaxation of the protons at a selected TE to acquire a T1-weighted signal from the ROI. The TE is an ultra-short time to echo. It is less than about 300 μs. A first image is generated without CA using 3D UTE sequences. Next, a paramagnetic or superparamagnetic CA is introduced into the blood. Then, A second image of the ROI is generated. Determining blood volume fraction comprises comparing signal intensities of the region of interest prior to introducing the CA and after introducing the CA. Specifically, determining the blood volume fraction comprises determining a difference in total signal intensities between the first image and the second image and determining a difference in blood signal intensities between the first image and the second image, wherein the blood volume fraction comprises a ratio of the total signal intensity difference to the blood signal intensity difference.

[0056] QUTE-CE MRI is quantitative, leading to direct assay of the CA concentration for quantitative MRI, as with nuclear imaging but without radiation toxicity or the other complications associated with radio-pharmaceuticals. Because the acquired signal is quantitative, the technique can be used for partial blood volume measurements using two volume methods. To date, there are no reported techniques that can potentially make absolute measurements of cerebral blood volume (qCBV) throughout the brain. QUTE-CE MRI can be used for identifying hyper- or hypo-vascularization, small vessel density, and vascular reserve, vascular responsivity to CO.sub.2 challenge, perfusion defects and standardized uptake values or organ absorbed dose, at the individual voxel and regional levels using an anatomical or functional atlas. Thus, QUTE-CE MRI provides an advantageous set of imaging biomarkers or diagnostic markers for assessing function and state.

[0057] In some aspects, the CA can be ferumoxytol, an ultra-small superparamagnetic iron oxide nanoparticles (USPION) with a dextran coating. Since the size exceeds the cutoff (˜6 nm) for glomerular filtration, ferumoxytol is not cleared by the kidney, and instead is an excellent blood pool contrast agent with a long intravascular half-life of ˜15 h (Bremerich et al., 2007). Numerous clinical MRI studies using ferumoxytol have been conducted in children and adults, demonstrating no major adverse effects (Muehe et al., 2016); thus QUTE-CE can be readily used in the clinic to study SVD.

[0058] Comparison to Prior Clinical Image Techniques.

[0059] Currently, clinical imaging of cerebral SVD is indirect and mostly represents its sequelae, such as ischemic (white matter hyperintensity (WMH) (Reijmer et al., 2016), lacunar infarcts) and hemorrhagic (cerebral microbleeds (CMBs)) lesions (Shi et al., 2016; Greenberg et al., 2009). Such consequences of SVD can be detected in the clinical setting on conventional MRI sequences, including T2/FLAIR (WMH, chronic infarcts), DWI (acute infarcts), and susceptibility weighted imaging (SWI) (CMBs) (Wardlaw et al., 2013). However, these techniques are limited in specificity, accuracy, and reliability in the assessment of total, quantitative burden of microvascular disease state (Wey et al., 2013; Brunser et al., 2013). Furthermore, current MRI techniques provide no insight regarding the microarchitecture of the brain's global network of cerebral small vasculature, or vasculome (Guo et al., 2012), which may play a crucial role in understanding of the underlying SVD pathology and mechanisms of disease in patients with stroke as well as apparently healthy aging adults with SVD that could be silent but unremittingly progressive and, ultimately, disabling. Thus, novel diagnostic methods to reliably quantify the total extent of SVD are urgently needed to address the growing burden of SVD-related disability.

[0060] Comparison to Other Prior Art Approaches for Measuring CBV.

[0061] Other MRI methods for SPION imaging utilize long-range susceptibility induced effects (Cunninham et al., 2005; Stuber et al., 2007; Seppenwoolde et al., 2003), whereas QUTE-CE MRI avoids them by making T.sub.1-enhanced measurements with TEs 1000 times shorter than standard modalities. Note that regular T.sub.1-weighted imaging is not quantitative and does not lead to the detailed images obtained with QUTE-CE. Dynamic susceptibility contrast (DSC), or perfusion-weighted MRI is commonly used for measuring CBV values (Barbier et al., 2001), but requires accurate determination of the arterial input function (AIF) (Rempp et al., 1994; Yankeelov et al., 2009), or gadolinium based contrast agent (GBCA) concentration versus time curve, which is typically 15-30% inaccurate (Walker-Samuel et al., 2007; Schabel et al., 2008). Other techniques for measuring the CBV, such as steady-state susceptibility contrast mapping (SSGRE), steady state CBV (SS_CBV), and ΔR2 (Troprès et al., 2001; Christen et al., 2012) all utilize T.sub.2 and T.sub.2* effects, which are susceptible to intra- and extra-voxular dephasing, flow artifacts and vessel size, density and orientation (Kim et al., 2012). Iron fMRI using SPIONs differs from QUTE-CE in that it is T.sub.2* weighted, requires high CA doses and is sensitive to extra-vascular space (Stuber et al., 2007; Mandeville, 2012). QUTE-CE is the only MR imaging technique that leads to positive contrast imaging without susceptibility-induced signal dropout.

[0062] Thus, qCBV measurements derived from QUTE-CE MRI can be used as a quantitative diagnostic marker for ADRD.

EXAMPLES

[0063] All animal experiments were conducted in accordance with institutional IACUC approved protocols. QUTE-CE measurements were made on 5 wild-type (WT) and 6 APOE-ε4 knock-in female rats of 7 months of age showing signs of mild cognitive impairment.

[0064] The data suggests that initial hypervascularization may be a coping mechanism to compensate for metabolic dysfunction in aging and dementia.

[0065] Methods

[0066] Behavioral Testing

[0067] The measure of cognitive behavior routinely performed at the Center for Translational Neuroimaging (CTNI) are Barnes maze for spatial memory and the Novel Object Preference (NOP) for object memory. Both the Barnes and NOP are hippocampus dependent but involve different “learning” strategies (McLay et al., 1997; Assini et al., 2009; Larkin et al., 2014; Pardo et al., 2016). While one test can be enough, at least two tests preferably are performed that recruit the same function/area (e.g. memory/hippocampus). The focus on hippocampal-dependent functions is desirable because of its involvement in psychiatric disorders and similarities across species (Squire et al., 1992).

[0068] Animal Experiments

[0069] QUTE-CE MRI Biomarkers

[0070] The QUTE-CE MRI measurement pipeline is illustrated in FIG. 2. QUTE-CE Vascular Biomarkers: The (1) voxel-based, physiological blood fraction (qCBV) is measured from 0-1, with 1 being an artery or a vein. Regional (2) macro- and (3) micro-vascular measurements are obtained by considering the mean or the mode of regional distributions, respectively. Regional volumes of interest (VOIs) detailed by a high-resolution, 173-region rat anatomical atlas (Ekam Solutions, Boston, Mass. USA) that is fit digitally to the brain using an affine transform. (4) Regions can be classified for vascular heterogeneity by considering distance between the mean and the mode. (5) Dynamic functional tests, such as a hypercapnic challenge with 5% CO.sub.2, can be applied to examine the responsivity of the vascular reserve.

[0071] Rationale for Data Format and Analysis

[0072] In these measurements, the qCBV was calculated by using the pre- and post-contrast UTE images. Throughout the entire rat brain, about 500,000 voxels at 150 micro-meter isotropic resolution were obtained in about 8 minutes. However, it can be noticed that while outstanding vascular images are produced, the quantification at the voxel level still has a high degree of error. Considering the voxel-based error and given the slight variation in neuroanatomy from one animal to the next, it was preferable to quantify regional vascular measurements in these small animals. In order to accomplish this, each rat's brain was fit to an anatomical atlas with 173 regions using a manually adjusted affine transform. In the atlas, the left and right halves of the brain were lumped into a single region by default. Thus, the 500,000 voxels were distributed into the 173 regions, and the mean and mode were calculated for each region.

[0073] To test for inter-group comparison, the list of means and modes for one group was compared to the means or modes of another group for all 173 regions. Statistical significance was achieved through a t-test for difference (P<0.05 or P<0.01) between the two lists.

[0074] Microvascular, or small vessel, changes were associated with the mode; the rationale for this is that most of the brain volume can be expected to be filled with mainly smaller vessels, and it was a way to remove influence from large vessels that are 100% filled with blood.

[0075] QUTE-CE MRI measurements were made on female wild-type (WT) and APOE4+ Sprague Dawley (SD) rats at 8 months of age and at 2 years of age.

[0076] Experiments at 8 Months Old (all Female SDs)

[0077] Structural QUTE-CE

[0078] Group 1 (n=5): WT

[0079] Group 2 (n=5): APOE4+

[0080] Experiments at 2 y Old (all Female SDs)

[0081] Structural QUTE-CE

[0082] Group 1 (n=5): WT

[0083] Group 2 (n=5): APOE4+

[0084] Dynamic CO.sub.2 Challenge at 2 y (all Female (SDs)

[0085] Dynamic QUTE-CE (Multiple CO2 challenges)

[0086] Group 1 (n=5): WT

[0087] Group 2 (n=5): APOE4+

[0088] Results

[0089] The structural and functional changes of the vasculature were measured using QUTE-CE MRI longitudinally in aging of APOE4 knock-in female rats. The method utilized an FDA approved contrast agent and was compatible with existing clinical scanners, indicated that it can be implemented in humans for routine screening of CNS diseases.

[0090] Hypervascularization Found in ApoE4 Rats at 8 m of Age

[0091] Small Vessels: 44 of 173 regions changed (P<0.05), of which 39 showed increases. See FIGS. 6-9. A pattern of mean vascular changes was also observed consisting of 11 increasing and 25 decreasing regions (p<0.01).

[0092] Trend Towards Hypovascularization at 2 Years of Age

[0093] In this model, AD advanced quicker for males than females, and though females did not yet exhibit statistically significant cognitive impairment, static QUTE-CE MRI revealed an age-dependent hyper- to hypo-microvascular remodeling trend (FIGS. 5A, 5B). This male/female trend in impairment due to the APOE4 gene has been validated in rodents (www.ncbi.nlm.nih.gov/pmc/articles/PMC4687024/pdf/nihms740272.pdf). The hyper-microvascular trend was found to highly correlate with brain regions exhibiting hyper-connectivity, as measured by Echo-Planar Imaging (EPI). The females were found to exhibit cognitive impairment later at 2 years of age, as measured by the Barnes Maze and Novel Object Recognition tests (p<0.05).

[0094] ApoE4 Rats: Hypersensitive and Hysteretic Vascular Dysfunction

[0095] Dynamic QUTE-CE MRI revealed hypersensitive recruitment of the vascular reserve. (FIG. 5B) A very significant response to breathing in 5% CO.sub.2 gas was observed for the ApoE4 rats, and unlike the WT, they did not recover during the 1 min rest period. These data could suggest that initial hypervascularization may be a coping mechanism to compensate for metabolic dysfunction in aging and dementia.

[0096] Structural Differences at 8 m

[0097] The tables in FIGS. 6-9 list the results regarding the structural differences, both in terms of the mean (p<0.01, top) and the mode (p<0.05, bottom). The P-value is shown, and the mean qCBV along with the STD is available between all animals in the two groups (see Methods, group comparison at 8 m of age). The region number is the assigned region number in the anatomical atlas, and the rows have been organized such that the largest qCBV (APOE4 mean or mode—WT mean or mode) is at the top (hypervascularized APOE4). On the right of the bottom table comparing the modes, the region names and their p-values are listed in order of most to least significant. Of the 173 regions, non-significant regions have been omitted.

[0098] Structural Differences at 2 v

[0099] The tables in FIGS. 10-13 list the results regarding the structural differences, both in terms of the mean (p<0.01, left) and the mode (p<0.05, right). The H-values and P-values are presented, and the mean qCBV for WT and APOE4 are available along with the STD. Differences were calculated between the two groups and standard deviations of the differences were calculated using error propagation for subtraction of means. The region number is the assigned region number in the anatomical atlas, and the rows have been organized such that the highest decrease in is at the top (hypovascularized APOE4) and so on. The regions have been colored dark gray and darker gray if those same regions were also statistically significant for change in the 8 m-old rats. If they were hypervascularized, the color on the left representing the 8 m state is dark grey (red in original); if hypovascularized it is darker gray (blue in original). The color to the right corresponds to the hypo- or hyper-vascularization of the 2 y-old state.

[0100] For the mean, it can be observed that of the 66 regions that are statistically significant for change at 2 y, 18 were also significant at 8 m-old. Thirteen of those regions were hypovascularized and stayed hypovascularized, three were hypervascularized and stayed hypervascularized, two were hypervascularized and went hypovascularized, but none went from hypo- to hyper-vascularization.

[0101] For the mode, it can be observed that of the 32 regions that are statistically significant for change at 2 y, six were also significant at 8 m-old. Two of those regions were hypovascularized and stayed hypovascularized, two were hypervascularized and stayed hypervascularized, two were hypervascularized and went hypovascularized, but none went from hypo- to hyper-vascularization.

[0102] Dynamic CO.sub.2 at 2 Y

[0103] The tables in FIGS. 14-17 list the results for the two groups that were tested at 2 y of age. The tables list the mean and mode differences in the CO.sub.2 challenge states. That means that M1 and its STD represent the mean of the differences that each animal had per region. By following the mean of the intra-animal differences rather than the absolute values, statistical significance can be tested while neglecting inter-animal comparison.

[0104] Everything was compared to the first post-contrast scan in which the qCBV is calculated (scan 1). Scans 2-4 were performed with CO.sub.2 on (M1 is the mean difference), with CO.sub.2 back off (M2 is the mean difference), and finally with CO.sub.2 toggled back on (M3 is the mean difference). The h- and p-values are also presented. It is noted that the comparisons h1, p1 correspond to the first CO.sub.2 state; h2, p2 is when the CO.sub.2 is toggled back off, h3, p3 is when the CO.sub.2 is toggled back on.

[0105] As used herein, “consisting essentially of” allows the inclusion of materials or steps that do not materially affect the basic and novel characteristics of the claim. Any recitation herein of the term “comprising,” particularly in a description of components of a composition or in a description of elements of a device, can be exchanged with “consisting essentially of” or “consisting of”.

[0106] While the present technology has been described in conjunction with certain preferred embodiments, one of ordinary skill, after reading the foregoing specification, will be able to effect various changes, substitutions of equivalents, and other alterations to the compositions and methods set forth herein.

REFERENCES

[0107] Alzheimer's Association. 2018 Alzheimer's Disease Facts and Figures. Alzheimers Dement. 14(3):367-429 (2018). [0108] Assini F L, Duzzioni M, Takahashi R N. Object location memory in mice: Pharmacological validation and further evidence of hippocampal CA1 participation. Behav Brain Res. 2009; 204(1):206-211. doi:10.1016/j.bbr.2009.06.005 [0109] Barbier E L, Lamalle L, Décorps M. Methodology of brain perfusion imaging. J Magn Reson Imaging. 2001; 13(4):496-520. doi:10.1002/jmri.1073 [0110] Bremerich J, Bilecen D, Reimer P. M R angiography with blood pool contrast agents. Eur Radiol. 2007; 17(12):3017-3024. doi:10.1007/s00330-007-0712-0 [0111] Brunser A M, Hoppe A, Illanes S, et al. Accuracy of diffusion-weighted imaging in the diagnosis of stroke in patients with suspected cerebral infarct. Stroke. 2013. doi:10.1161/STROKEAHA.111.000527 [0112] Chen, C.-C. V, Chen, Y.-C., Hsiao, H.-Y., Chang, C. & Chem, Y. Neurovascular abnormalities in brain disorders: highlights with angiogenesis and magnetic resonance imaging studies. J. Biomed Sci. 20, 47 (2013). [0113] Christen T, Ni W, Qiu D, et al. High-resolution cerebral blood volume imaging in humans using the blood pool contrast agent ferumoxytol. Magn Reson Med. 2012; Im:705-710. doi: 10.1002/mrm.24500 [0114] Cunningham C H, Arai T, Yang P C, McConnell M V., Pauly J M, Conolly S M. Positive contrast magnetic resonance imaging of cells labeled with magnetic nanoparticles. Magn Reson Med. 2005; 53(5):999-1005. doi:10.1002/mrm.20477 [0115] Gharagouzloo C A, McMahon P N, Sridhar S. Quantitative contrast-enhanced MRI with superparamagnetic nanoparticles using ultrashort time-to-echo pulse sequences. Magn Reson Med. 2015; 74(2):431-441. doi: 10.1002/mrm.25426 [0116] Gharagouzloo C A, Timms L, Qiao J, et al. Quantitative vascular neuroimaging of the rat brain using superparamagnetic nanoparticles: New insights on vascular organization and brain function. Neuroimage. 2017; 163:24-33. doi:10.1016/j.neuroimage.2017.09.003 [0117] Greenberg S M, Vernooij M W, Cordonnier C, et al. Cerebral microbleeds: a guide to detection and interpretation. Lancet Neurol. 2009. doi:10.1016/S1474-4422(09)70013-4 [0118] Guo S, Zhou Y, Xing C, et al. The Vasculome of the Mouse Brain. PLoS One. 2012. doi:10.1371/journal.pone.0052665 [0119] Huang, Y. & Mucke, L. Alzheimer mechanisms and therapeutic strategies. Cell 148, 1204-1222 (2012). [0120] Kim S M, Kim M J, Rhee H Y, et al. Regional cerebral perfusion in patients with Alzheimer's disease and mild cognitive impairment: effect of APOE Epsilon4 allele. Neuroradiology. 2013; 55(1):25-34. doi:10.1007/s00234-012-1077-x [0121] Larkin M C, Lykken C, Tye L D, Wickelgren J G, Frank L M. Hippocampal output area CA1 broadcasts a generalized novelty signal during an object-place recognition task. Hippocampus. 2014; 24(7):773-783. doi:10.1002/hipo.22268 [0122] Li L, Jiang Q, Zhang L, et al. Angiogenesis and improved cerebral blood flow in the ischemic boundary area detected by MRI after administration of sildenafil to rats with embolic stroke. Brain Res. 2007; 1132(1):185-192. doi:10.1016/j.brainres.2006.10.098 [0123] Mandeville J B. IRON fMRI measurements of CBV and implications for BOLD signal. Neuroimage. 2012; 62(2):1000-1008. doi:10.1016/j.neuroimage.2012.01.070 [0124] McLay R N, Freeman S M, Harlan R E, Ide C F, Kastin A J, Zadina J E. Aging in the hippocampus: Interrelated actions of neurotrophins and glucocorticoids. Neurosci Biobehav Rev. 1997; 21(5):615-629. doi:10.1016/S0149-7634(96)00046-2 [0125] Muehe A M, Feng D, Von Eyben R, et al. Safety Report of Ferumoxytol for Magnetic Resonance Imaging in Children and Young Adults. Invest Radiol. 2016; 51(4):221-227. doi: 10.1097/RLI.0000000000000230 [0126] Pardo J, Uriarte M, Cónsole G M, et al. Insulin-like growth factor-I gene therapy increases hippocampal neurogenesis, astrocyte branching and improves spatial memory in female aging rats. Eur J Neurosci. 2016; 44(4):2120-2128. doi:10.1111/ejn.13278 [0127] Reijmer Y D, van Veluw S J, Greenberg S M. Ischemic brain injury in cerebral amyloid angiopathy. J Cereb Blood Flow Metab. 2016. doi:10.1038/jcbfm.2015.88 [0128] Rempp K A, Brix G, Wenz F, Becker C R, Gückel F, Lorenz W J. Quantification of regional cerebral blood flow and volume with dynamic susceptibility contrast-enhanced MR imaging. Radiology. 1994; 193:637-641. doi:10.1148/radiology.193.3.7972800 [0129] Schabel M C, Parker D L. Uncertainty and bias in contrast concentration measurements using spoiled gradient echo pulse sequences. Phys Med Biol. 2008; 53(9):2345-2373. doi:10.1088/0031-9155/53/9/010 [0130] Seppenwoolde J H, Viergever M A, Bakker C J G. Passive tracking exploiting local signal conservation: The white marker phenomenon. Magn Reson Med. 2003; 50(4):784-790. doi:10.1002/mrm.10574 [0131] Shi Y, Wardlaw J M. Update on cerebral small vessel disease: a dynamic whole-brain disease. BMJ. 2016; 1(3):83-92. doi:10.1136/svn-2016-000035 [0132] Squire L R. Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans. Psychol Rev. 1992; 99(2):195-231. doi:10.1037/0033-295X.99.2.195 [0133] Stuber M, Gilson W D, Schär M, et al. Positive contrast visualization of iron oxide-labeled stem cells using inversion-recovery with ON-resonant water suppression (IRON). Magn Reson Med. 2007; 58:1072-1077. doi:10.1002/mrm.21399 [0134] Troprès I, Grimault S, Vaeth A, et al. Vessel size imaging. Magn Reson Med. 2001; 45:397-408. doi:10.1002/1522-2594(200103)45:3<397::AID-MRM1052>3.0.CO; 2-3 [0135] Walker-Samuel S, Leach M O, Collins D J. Reference tissue quantification of DCE-MRI data without a contrast agent calibration. Phys Med Biol. 2007; 52:589-601. doi:10.1088/0031-9155/52/3/004 [0136] Wardlaw J M, Smith E E, Biessels G J, et al. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol. 2013; 12(8):822-838. doi:10.1016/S1474-4422(13)70124-8 [0137] Wey H-Y, Desai V R, Duong T Q. A review of current imaging methods used in stroke research. Neurol Res. 2013. doi:10.1179/1743132813Y.0000000250 [0138] WHO|Dementia. WHO Fact sheet Updated May 2017 (2017). doi:/entity/mediacentre/factsheets/fs297/en/index.html [0139] Yankeelov T, Gore J. Dynamic contrast enhanced magnetic resonance imaging in oncology: theory, data acquisition, analysis, and examples. Curr Med Imaging Rev. 2009; 3(2):91-107. doi:10.2174/157340507780619179.Dynamic [0140] Zlokovic, B. V. Neurovascular pathways to neurodegeneration in Alzheimer's disease and other disorders. Nat. Rev. Neurosci. (2011). doi:10.1038/nrn31 I.