Alzheimer's disease (AD) is a neurodegenerative disorder characterized by the progressive decline of memory and cognitive functions. It is estimated that over five million Americans are living with this progressive and fatal disease. Alzheimer's destroys brain cells, causing memory loss and problems with thinking and behavior that decrease quality of life. AD has no known cure, but treatments for symptoms can improve the quality of life of the millions of people, and their families, suffering from AD. An early diagnosis of AD gives the patient time to make choices that maximize quality of life, reduces anxiety about unknown problems, gives more time to plan for the future, and provides a better chance of benefiting from treatment.
The inaccuracy of the diagnosis for AD, however, has made its therapeutic intervention difficult, particularly at early stages to prevent significant neurodegeneration and cognitive dysfunction. Thus, there is a need for highly sensitive and specific tests to diagnose Alzheimer's disease.
In earlier studies of human dermal fibroblast aggregation, the present inventors quantified cell aggregation by the area of the unit aggregate, which was the average area per number of aggregates (A/N) at 48 hours after plating on Matrigel. The present inventors discovered large aggregates for AD patients when compared with age-matched controls for which cellular aggregation was more evenly distributed.
The prevent inventors developed a new diagnostic method(s) relating to cell aggregation. Disclosed herein is a method for diagnosing Alzheimer's disease based on quantitatively measured cellular aggregation rate. The present inventors discovered that the rate of change of cell aggregation with increasing cell density is elevated in AD cells compared to non-AD cells. The trend of higher rate of change of cell aggregation is consistent with the inventors' previous studies in which it was reported that AD cells are consistently bigger and less adhesive in average than the AC/Non-ADD cells.
Thus, disclosed herein is a method of diagnosing Alzheimer's Disease in a human subject comprising the steps of
(a) obtaining one or more cells from a human subject;
(b) culturing the one or more cells for a time period;
(c) determining an average area of cell aggregates and dividing the average area by the number of aggregates to obtain an average area per number of aggregates;
(d) determining an aggregation rate by evaluating the rate of change of the average area per number of aggregates determined in step (c) as a function of cell density;
(e) comparing the determination of step (d) with an aggregation rate determined using non-Alzheimer's disease cells; and
(f) diagnosing the presence or absence of Alzheimer's Disease based on the comparison in step (e), wherein the diagnosis is positive for Alzheimer's disease if the aggregation rate determined in step (d) is increased compared to the aggregation rate determined using the non-Alzheimer's disease cells.
In some embodiments, the method further comprises confirming the diagnosis using at least one additional diagnostic assay. In certain embodiments, the at least one additional diagnostic assay is chosen from AD-Index and Morphology diagnostic assays.
Also disclosed herein is a simple-majority rule for diagnosing AD using at least three diagnostic assays. There is disclosed a method of diagnosing Alzheimer's disease in a human subject comprising
i) performing at least three diagnostic assays;
ii) evaluating whether each assay indicates the presence or absence of Alzheimer's disease; and
iii) diagnosing the presence or absence of Alzheimer's disease based on whether a simple majority of the diagnostic assays indicate the presence or absence of Alzheimer's disease,
wherein one of the diagnostic assays comprises:
(a) obtaining one or more cells from a human subject;
(b) culturing the one or more cells for a time period;
(c) determining an average area of cell aggregates and dividing the average area by the number of aggregates to obtain an average area per number of aggregates;
(d) determining an aggregation rate by evaluating the rate of change of the area per number of aggregates determined in step (c) as a function of cell density; and
(e) comparing the determination of step (d) with an aggregation rate determined using non-Alzheimer's Disease cells, wherein the presence of Alzheimer's disease is indicated if the aggregation rate determined in step (d) is increased compared to the aggregation rate determined using the non-Alzheimer's Disease cells.
Abbreviations: AD: Alzheimer's disease; AC: non-demented age-matched control; Non-ADD: non-Alzheimer's dementia.
As used herein, the singular forms “a,” “an,” and “the” include plural reference unless the context dictates otherwise.
As used herein, “AD-Index” assay refers to a diagnostic assay for Alzheimer's disease based on the ratio of phosphorylated first and second MAP kinase proteins. The “AD-Index” assay comprises (a) contacting cells from a subject with an agent that is a protein kinase C activator; (b) measuring the ratio of a phosphorylated first MAP kinase protein to a phosphorylated second MAP kinase protein wherein the phosphorylated first and second MAP kinase proteins are obtained from the cells after the contacting in step (a); (c) measuring the ratio of phosphorylated first MAP kinase protein to phosphorylated second MAP kinase protein in cells from the subject that have not been contacted with the agent that is a protein kinase C activator used in step (a); (d) subtracting the ratio obtained in step (c) from the ratio obtained in step (b); and (e) diagnosing the presence or absence of Alzheimer's disease in the subject based on the difference calculated in step (d), wherein the presence of Alzheimer's disease is indicated in the subject if the difference is a positive value. U.S. Pat. No. 7,595,167 and U.S. Patent Application Publication Number 2014/0031245 are incorporated herein by reference for their detailed disclosure of the AD-Index assay. Thus, the AD-Index assay may be performed as described in those publications. For example, in certain embodiments, the PKC activator is bradykinin and the first and second MAP kinase proteins are Erk1 and Erk2, respectively.
As used herein, “Morphology” assay refers to a diagnostic assay for Alzheimer's disease comprising (a) obtaining one or more cells from a human subject; (b) culturing the one or more cells for a time period; (c) determining the average area of cell aggregates and dividing the average area by the number of aggregates to obtain the average area per number of aggregates; and (d) comparing the determination of step (c) with the average area per number of aggregates determined using non-Alzheimer's disease cells, wherein the presence of Alzheimer's disease is indicated if the average area per number of aggregates determined in step (c) is greater than the average area per number of aggregates using the non-Alzheimer's disease cells. Cells may be cultured and the area and number of aggregates may be determined as described in U.S. Pat. No. 8,658,134, which is incorporated herein by reference for its disclosure of the “Morphology” assay, and as described below.
As used herein, the term “simple majority” refers to an instance where a majority of diagnostic assays, for example, two out of three diagnostic assays, yield the same diagnosis (i.e. the presence or the absence of Alzheimer's disease).
The present inventors have discovered a new method for diagnosing AD based on an elevated cell aggregation rate with increasing cell density in cells from AD patients when compared with AC and Non-ADD patients. Thus, there is disclosed a method of diagnosing Alzheimer's Disease in a human subject comprising the steps of
(a) obtaining one or more cells from a human subject;
(b) culturing the one or more cells for a time period;
(c) determining an average area of cell aggregates and dividing the average area by the number of aggregates to obtain an average area per number of aggregates;
(d) determining an aggregation rate by evaluating the rate of change of the average area per number of aggregates determined in step (c) as a function of cell density;
(e) comparing the determination of step (d) with an aggregation rate determined using non-Alzheimer's disease cells; and
(f) diagnosing the presence or absence of Alzheimer's Disease based on the comparison in step (e), wherein the diagnosis is positive for Alzheimer's disease if the aggregation rate determined in step (d) is increased compared to the aggregation rate determined using the non-Alzheimer's disease cells.
In some embodiments, the one or more cells obtained from the human subject are peripheral cells (i.e., cells obtained from non-CNS tissue). In some embodiments, the one or more cells are fibroblast cells. In certain embodiments, the fibroblast cells are skin fibroblast cells. Cell precursors of fibroblasts, such as induced pluripotent stem cells (IPSC) may also be used. For example, recent techniques for obtaining IPSC from human skin fibroblasts permitted differentiation of IPSO in cells such as neurons and showed imbalances in Ap in both skin fibroblasts and IPSO differentiated neurons. Whalley K., “Neurodegenerative disease: Dishing up Alzheimer's disease”, Nature Reviews Neuroscience 13, 149 (March 2012)|doi:10.1038/nrn3201.
In some embodiments, the one or more cells are chosen from skin cells, blood cells (lymphocytes), and buccal mucosal cells.
The non-Alzheimer's disease cells may be chosen, e.g., from an age-matched control. In some embodiments, the age-matched control is chosen from a non-AD non-demented population. In some embodiments, the age-matched control is chosen from a non-AD demented population, such as patients with Huntington Chorea, B12 deficiency, or hypothyroidism.
The one or more cells may be cultured in a cell media for growth, such as, for example, a protein mixture. In some embodiments, the protein mixture is a gelatinous protein mixture. A non-limiting exemplary gelatinous protein mixture is Matrigel™. Matrigel™ is the trade name for a gelatinous protein mixture secreted by the Engelbreth-Holm-Swarm (EHS) mouse sarcoma cells and marketed by BD Biosciences. This mixture resembles the complex extracellular environment found in many tissues and is used by cell biologists as a substrate for cell culture.
In some embodiments, the one or more cells are cultured in a preparation comprising extracellular matrix proteins. In some embodiments, the preparation comprises laminin, collagen, heparin sulfate proteoglycans, entactin/nidogen, and/or combinations thereof. In some embodiments, the preparation is extracted from a tumor, such as the EHS mouse sarcoma. The preparation may further comprise a growth factor, such as TGF-beta, epidermal growth factor, insulin-like growth factor, fibroblast growth factor, tissue plasminogen activator, and/or other growth factors or combinations thereof. In certain embodiments, the growth factors occur naturally in the EHS mouse sarcoma. Extracellular matrix proteins may also contain numerous other proteins.
In some embodiments, the one or more cells are cultured in a basement membrane preparation. In some embodiments, the preparation is solubilized. In some embodiments, the basement membrane preparation is extracted from a tumor, such as the EHS mouse sarcoma—a tumor rich in extracellular matrix proteins. Its major component is laminin, collagen IV, heparin sulfate proteoglycans, and entactin/nidogen. In some embodiments, the preparation contains TGF-beta, epidermal growth factor, insulin-like growth factor, fibroblast growth factor, tissue plasminogen activator, and/or other growth factors which may or may not occur naturally in the EHS tumor. BD Matrigel Matrix Growth Factor Reduced (GFR) is found to be particularly well suited for applications requiring a more highly defined basement membrane preparation.
Within a short time after being cultured, measurable cellular networks form. This time period may vary in view of, for example, cell type and conditions, but generally, this time period ranges from about 1 hour or less, ranging from about 10 minutes to about 60 minutes, such as from about 10 minutes to about 45 minutes or any time in between. After a time, for example, approximately 5 hours, these networks start to degenerate and edges retract to leave behind measurable “clumps” or aggregates. In some embodiments, the time period for culturing the one or more cells is chosen from about 1 hour to about 72 hours, such as from about 12 hours to about 72 hours or from about 24 hours to about 48 hours. In certain embodiments, the time period is about 48 hours or any one hour increment subdivision thereof.
The method may further comprise imaging the cultured cells at the end of the time period. Images may be captured according to techniques known in the art. For example, images of the cellular networks may be captured with an inverted microscope, such as Western Digital AMID Model 2000, and controlled by a computer via image acquisition software at a desired magnification. Appropriate imaging techniques include, but are not limited to, confocal microscopy, phase contrast, bright field, fluorescence, differential interference contrast, and robotic systems.
The area of aggregates can be determined by any suitable method, e.g., by fitting an ellipse across the aggregate. The counting of aggregates as well as aggregate area determination can be performed manually or can be automated, e.g., by image processing techniques known in the art.
In some embodiments, cell density is measured based on the number of cells per μm2 or per field of view. In certain embodiments, cell density is measured by measuring the number of cells per 10× image. In certain embodiments, the rate of change of the average area per number of aggregates as a function of cell density is evaluated within the boundaries of 320 to 550 cells/10× image or such as, of 330 to 500 cells/10× image.
Cell aggregation rate is determined by evaluating the rate of change of the average area per number of aggregates as a function of cell density. In some embodiments, the rate of change of the average area per number of aggregates as a function of cell density is evaluated by determining the slope of a linear fit between the average area per number of aggregates and cell density.
The aggregation rate of the cultured cells obtained from the human subject is compared to the aggregation rate determined using non-Alzheimer's disease control cells. The diagnosis is positive for Alzheimer's disease if the aggregation rate of the cultured cells from the human subject is increased compared to the aggregation rate determined using the non-Alzheimer's disease control cells.
The AD diagnosis may be confirmed by at least one additional diagnostic assay. In some embodiments, the at least one additional diagnostic assay is performed using cells obtained from the human subject. That is, the Aggregation Rate assay and the at least one additional diagnostic assay use the same cell lines, i.e., patients. In certain embodiments, the at least one additional diagnostic assay is chosen from AD-Index and Morphology diagnostic assays. In some embodiments, the diagnosis is confirmed by both the AD-Index and Morphology assays. The Aggregation Rate assay may also be combined with other diagnostic technologies known in the art, including, but not limited to, clinical diagnosis, positron emission tomography (PET)—amyloid imaging, and/or cerebrospinal fluid (CSF) levels of Aβ or tau.
In some embodiments, an AD diagnosis is based on a simple majority of at least three diagnostic assays. Thus, there is disclosed a method of diagnosing AD in a human subject using an aggregation rate, such as comprising: i) performing at least three diagnostic assays; ii) evaluating whether each assay indicates the presence or absence of AD; and iii) diagnosing the presence or absence of AD based on whether a simple majority of the at least three diagnostic assays indicate the presence or absence of AD, wherein one of the diagnostic assays comprises: a) obtaining one or more cells from a human subject; b) culturing the one or more cells for a time period; c) determining an average area of cell aggregates and dividing the average area by the number of aggregates to obtain an average area per number of aggregates; d) determining an aggregation rate by evaluating the rate of change of the area per number of aggregates determined in step (c) as a function of cell density; and e) comparing the determination of step (d) with an aggregation rate determined using non-AD cells, wherein the presence of AD is indicated if the aggregation rate determined in step (d) is increased compared to the aggregation rate determined using the non-AD cells.
In some embodiments, two of the at least three diagnostic assays are AD-Index and Morphology assays.
The methods described herein will be further described by the following examples, which are meant to illustrate, but not limit, the scope of the disclosure.
Materials and Methods
Banked and Fresh Cell Lines Used in This Study. Example experiments were carried out using skin fibroblast samples from 38 patients—9 banked cases (Supplementary Table 1) provided by the Coriell Institute for Medical Research (Camden, N.J.), and 29 cases (Supplementary Table 2) from the clinic provided by Marshall University (Huntington, W. Va.). The analysis showed a clear separation between AD and AC/Non-ADD using cellular aggregation rate as a diagnostic tool.
The fibroblast cells were plated on a thick layer (˜1.8 mm) of 3-D matrix (Matrigel, BD Biosciences, San Jose, Calif.) on 12 well plates. See Chirila F. V. et al., “Spatiotemporal Complexity of Fibroblast Networks Screens for Alzheimer's Disease,” J. Alzheimer's Disease, 33, 165-176 (2013). The available patient information is posted on Coriell web site (http://ccr.coriell.org/). The cell lines analyzed (38) were largely from the clinic (29/38) serving also as a validation for the previous studies with banked samples. See id. The age-matched control (AC) cases were not demented at the date of skin biopsy extraction. All the samples were taken antemortem. The banked skin fibroblast cells were frozen stocks under liquid nitrogen. Primary cultures were established after thawing those frozen samples and followed through successive passaging. See Zhao W. Q. et al., “MAP kinase signaling cascade dysfunction specific to Alzheimer's disease in fibroblasts,” Neurobiol. Dis., 11, 166-183 (2002); Khan T. K. et al., “An internally controlled peripheral biomarker for Alzheimer's disease: Erk1 and Erk2 responses to the inflammatory signal bradykinin,” Proc. Natl. Acad. Sci. U.S.A., 103(35), 13203-7 (2006); Khan T. K. et al., “Early diagnostic accuracy and pathophysiologic relevance of an autopsy-confirmed Alzheimer's disease peripheral biomarker,” Neurobiol. Aging, 31(6), 889-900 (2008); Khan T. K. et al., “A cellular model of Alzheimer's disease therapeutic efficacy: PKC activation revesres Abeta-induced biomarker abnormality on cultured fibroblasts,” Neurobiol. Dis. 34(2), 332-9 (2009); Chirila F. V. et al., “Spatiotemporal Complexity of Fibroblast Networks Screens for Alzheimer's Disease,” J. Alzheimer's Disease, 33, 165-176 (2013). All cell lines used were primary cell lines and were not treated in order to be immortalized.
Freshly taken fibroblasts were obtained as follows. Punch-biopsies (2-3 mm, upper arm) of skin tissues from patients and controls were obtained. The method of isolating fibroblasts from skin biopsies was followed according to the protocol in Takashima, A., “Establishment of Fibroblast Cultures,” Current Protocols in Cell Biology, vol. 2, 2.1.1-2.1.12 (1998). Cells with passages between 5 and 15 were used.
The initial cell density was controlled to be 50 cells/mm3 and was homogenized with 1.5 ml Dulbecco Modified Eagle Medium with 10% fetal bovine serum and 1% penicillin/streptomycin for each well. Cells were kept in a CO2 water-jacket incubator (Forma Scientific) up to 7 days after plating.
Image Capture. Images of the cellular networks were captured with an inverted microscope (Westover Digital AMID Model 2000, Westover Scientific, Bothell Wash.), controlled by a computer via an image acquisition software (Micron 2.0.0), using a 10× and a 4× objective. Five to nine images captured per well and four wells per cell line were typically used. In the first day, images were acquired at 30 min after seeding, the second day at 24 hours, and the third day at 48 hours. Images were processed with ImageJ, a freely available software from NIH (http://rsbweb.nih.gov/ij/).
The 5 images per well were initially taken using the same standard pattern, center (1), up (2), down (3), left (4), right (5), by moving one image field with respect to the central image. See Chirila F. V. et al., “Spatiotemporal Complexity of Fibroblast Networks Screens for Alzheimer's Disease,” J. Alzheimer's Disease, 33, 165-176 (2013). Later in the process, the number of images was increased from 5 to 9 by filling the corners of the rectangle with images from 6 to 9, in order to increase the area investigated and further improve the coefficient of variation without affecting the diagnostic discriminability. See id. Image 1 was always in the center of the well. To determine the center of the well, one of the following methods was used: a) the live image under 4× magnification should be symmetric, i.e., the shadows in the four corners should have equal areas for an aligned microscope; b) mark the center with a needle; or c) use gridded plates (Pioneer Scientific; Shrewsbury, Mass.) where the central square is always the 6th, in the central row or column. For image quantification, two sets of tools were used: initially manual as provided by Micron, software which came with the microscope; later automated with ImageJ.
For the initial cell count, a custom ImageJ plug-in was used in which “despeckle” was run three times; the image was filtered three times with a minimum filter of radius 0.5; and “Subtract Background” was run with a rolling radius of 20. Finally, the image was made binary and “Analyze Particles” was run in the size range 200-10000. All of these ImageJ commands were run inside a loop to permit analysis of all the images from one cell line automatically. The ImageJ plug-in was tuned by using manual cell counts on the same images and the relative error was below 7%.
The target number of cells per 10× image was 417 which corresponded to an initial cell concentration of 50 cells/ml. See Bikfalvi A. et al., “Phenotypic modulations of human umbilical vein endothelial cells and human dermal fibroblasts using two angiogenic assays,” Biol. Cell, 72, 275-278 (1991). A variation of cell concentration between 45 and 60 cells/ml was permitted. To minimize heterogeneity of the cell distribution in the image, images outside of the range 320-550 cells per 10× image were eliminated. For cellular aggregates at 48 h, manual ellipse fitting with the Micron software was used.
Average area per number of aggregates (A/N). Average area per number of aggregates was calculated in the following manner. For each image (i), an average aggregate area, <A>i, was calculated and the number of aggregates, Ni, was counted. Then, for each image, the ratio <A>i/Ni was evaluated. Typically, nine images per well were used and an average area per number for each well was evaluated as the
An addition average was performed over the four wells:
For simplicity, A/N is used instead of A
i/Ni
and A/N is called area of the unit aggregate. The aggregates were manually fitted with ellipses using Micron 2.0 software and their area and number were recorded. An automatic script for ImageJ was developed which agreed well with the manual ellipse fitting. These two approaches were within one standard deviation of each other.
Aggregation rate. The dependence of A/N on cell density was evaluated (
For data analysis, Gnuplot 4.4, freely available software (http://www.gnuplot.info), was used. A built in fit function from Gnuplot, which used an implementation of the nonlinear least-squares (NLLS) Marquardt-Levenberg algorithm, fit the raw data points. Unless otherwise specified, the error-bars are standard errors of the mean (SEM).
Probability distribution of cellular aggregates. For all the AD and AC patients, the values for slope(s) and intercept (int) of aggregates were binned into equal intervals, fit with Gaussian functions for each variable, and then integrated into a normalized two-dimensional distribution.
As shown in the inventors' previous studies, cell aggregation is enhanced in Alzheimer's disease (AD) patients when compared with age-matched controls (AC) and non-Alzheimer's disease demented patients (Non-ADD). See Chirila F. V. et al., “Spatiotemporal Complexity of Fibroblast Networks Screens for Alzheimer's Disease,” J. Alzheimer's Disease, 33, 165-176 (2013). Cell aggregation was previously quantified by the area of the unit aggregate, which was the average area per number of aggregates (A/N) at 48 hours, after plating on Matrigel. See id. Here, a new way of screening Alzheimer's disease patients (AD) from age-matched controls (AC) and from non-Alzheimer's disease demented patients (Non-ADD) is based on the inventors' discovery of increased cell aggregation when the cell density increases, referred to here as the slope intercept representation of cell aggregation.
An example of the dependence of cell aggregation (A/N) on cell density (# cells/10× image field) is illustrated in
A small number of banked cases from Coriell (9 cell lines)—3AD, 3AC, and 3 Non-ADD—showed the same separability between AD and the other two groups (
The data also showed that the AD-AC/Non-ADD separation increased for the 48-h time point when compared with the 24-h time point (
The aggregation rate (slope) and intercept changed with age as depicted in
The data showed that the slope intercept representation for cell aggregation is a more refined representation of cell aggregation than a simple average, i.e., A/N. It may be a useful tool to resolve the AD/AC/Non-ADD cases which are too close to the cut-off line in the A/N representation. Typically, the cut-off line for the AD and AC populations is at the intersection of the two Gaussian distributions that fit the biomarker outputs. In a narrow region near the but-off line, the tails of the Gaussian distributions coexist with certain probability therefore defining a gray zone of false positive/negative. Unknown cases falling in the gray zone for the A/N measure might be removed by using the rate of change of A/N.
The discrimination of AD cases from AC cases using the slope intercept analysis was further investigated on the 29 samples from the clinic. Among these samples, 5 were AD cases and 24 were AC cases. A higher slope and more negative intercept for the AD cases was confirmed (
The slope and intercept of A/N versus cell density were analyzed independently. Looking at the intercept for the AC, AD and Non-ADD groups both at large scale (
As shown in
The value of an AD bio-marker can be assessed by using the probability distributions for the two groups, AD and AC. Based on the probability distributions, one can estimate the extent of possible overlapping probability. The two groups of data (8 AD, 27 AC) were binned in slope and intercept. The probability distributions were estimated based on the frequency of occurrence in each bin divided by the total number of occurrences. The raw probability data sets were fitted with Gaussian curves in variable, slope and intercept. Then, the two dimensional Gaussian probabilities were plotted using Gnuplot. The AC probability distribution was narrower than the AD probability distribution (
The population of 38 cases (8 AD, 27 AC, and 3 Non-ADD) were evaluated from the point of view of the efficacy of the two bio-markers quantified by the average area of the unit aggregate (A/N) and aggregation rate (slope). The A/N measure of cell aggregation was plotted against the aggregation rate (slope) in
AD is a complex disease with a multifactorial structure in which many pathways are disrupted and/or affected. Furthermore, the sporadic form of AD has an onset above 65 years old when the likelihood of co-morbidity with other diseases is very high. Therefore, a single bio-marker may be unlikely to detect with high precision sporadic AD cases in their early progression when the drug efficacy for AD might be very high. The disclosed biomarker quantifying the fibroblast aggregation rate, when cross-validated with other two more mature biomarkers, AD-Index and Morphology assays, showed an overlap of approximately 92%.
Cross-validation of the aggregation rate with the AD-Index. A comparison of the aggregation rate with a developed assay such as the AD-Index can be a measure of performance. See Hadley M A et al., “Extracellular matrix regulates Sertoli cell differentiation, testicular cord formation and germ cell development,” J. Cell Biol., 101, 1511-1522 (1985); Ingber D. et al., “Mechanochemical switching between growth and differentiation during fibroblast growth factor-stimulated angiogenesis in vitro. Role of extracellular matrix,” J. Cell Biol., 109, 317-331 (1989); Furukawa K S et al., “Tissue-engineered skin using aggregates of normal human skin fibroblasts and biodegradable material,” J. Artif. Organs, 4, 353-356 (2001); Furukawa K S et al., “Formation of human fibroblast aggregates (spheroids) by rotational culture,” Cell Transplantation, 10, 441-445 (2001). Out of the 38 cases tested with the aggregation rate assay, 26 were also tested with the AD-Index assay western blot. Out of these 26 cases tested with both assays, 24 gave the same diagnosis, representing 92.3% overlap between the two biomarkers (
Cross-validation of the aggregation rate with the Morphology assay. A comparison of the aggregation rate with the Morphology assay, a studied assay, can also be a measure of performance. See Chirila F. V. et al., “Spatiotemporal Complexity of Fibroblast Networks Screens for Alzheimer's Disease,” J. Alzheimer's Disease, 33, 165-176 (2013). All of the 38 cases tested with the aggregation rate were also tested with the simpler assay of average area per number of aggregates (NN). Out of these 38 samples tested with both assays, 35 gave the same diagnosis, representing 92.1% overlap between the two bio-markers (
Simple-majority rule. Considering the 26 cases that were tested with all three bio-markers and upon use of a simple majority rule, i.e. 2/3 assays give the same result, the agreement with the clinical diagnostic increased to 100% (
Quantification of cross-validation and majority rule. The three biomarker values were normalized between 0 and 1 as
where x was the biomarker value to be normalized, max was the maximum value, and min was the minimum value of the data set of 26 cases. This normalization brought all three biomarkers within the same range, 0 to 1, making the comparison easier (
The curve fitting for all three biomarker values was done in two steps. First the fitting was done with a linear function, f(x)=a*x+b, for the AC values and with an exponential function, g(x)=c*exp(d*x), for the AD values. Second, the fitting was done with a glued function h(x)=f(x)+g(x)=a*x+b+c*exp(d*x). For the starting values of the parameters a, b, c, and d, the end values from the previous fit were used. After fitting with the glued function, h(x), the new fit parameters were recorded and used for the plots in
The absolute difference per case, |b1−b2|, |b2−b3|, and |b3−b1| where b1=AggR, b2=ADI, b3=A/N, were used to assess which pair of biomarkers was best correlated. The average values of the absolute differences were then calculated for all the 19 AC cases,
where j,k=1,3, and j≠k, and all the 7 AD cases,
where j,k=1,3, and j≠k. The overall averages for the absolute differences were also calculated for all 26 values,
The smaller these group differences were, the closer the two biomarkers were correlated
For quantifying the majority rule, once more the bio-marker values, x, were normalized with respect to the specific Cutoff value,
Examples are shown in
In
The absolute difference between pairs of values were considered on a case by case basis to measure the closeness of the three biomarkers (
An overview of the closeness of the three biomarkers is represented in
The same majority-rule presented herein could be applied to more than 3 biomarkers. The three biomarker values for the 7 AC and 7 AD patients were normalized as (x-Cutoff)/|x-Cutoff| as described above, where x was the current value of the biomarker and the Cutoff was the separation between the AC and AD groups. The cutoff values were determined at the intersection of the two Gaussian distributions fitting the AC and AD data.
The normalized values as described above can have one of the three possible states −1 for AC and Non-ADD, 1 for AD, or uncertain (U) for the cutoff value. The normalized AD values for the three AD biomarkers are shown in
The test-retest reliability was verified for the Aggregation Rate assay with two cell lines (1 AC, 1 AD) (
In summary the slope intercept representation of the human skin fibroblast aggregation showed significant separation in the slope but not in the intercept, suggesting that the elevated rate of change of cell aggregation with increasing cell density in AD was the driving force in the experiments for this new biomarker.
Analyses on 38 cases, out of which 9 were banked cases and 29 were cases from the clinic, suggested a clear separation between AD and AC/Non-ADD when using the aggregation rate (slope) as a biomarker.
The analysis of the probability distributions of the slope and intercept for 35 samples indicated that the probability distribution for the AC group (27) was narrower and resided on the tail of the probability distribution of the AD group (8) which was wider. This suggested that for larger data sets the estimated overlapping probability for the two groups, AD and AC, is less than 10%.
The new biomarker, aggregation rate (A), was cross-validated with two more mature assays, AD-Index (B), and area per number of aggregates (C). The cross-validation resulted in 92% overlap with each of the assays and in >97% overlap with the clinical diagnosis. The 92% agreement of the new biomarker A with the two more mature biomarkers B and C, which were also tested with hyper-validated samples, established this new biomarker via Euclid's common notions as a hyper-validated biomarker. See Euclid, Elements: Books I-XIII-Complete and Unabridged, (2006). That is, if biomarkers B,C=hyper-validated and if B,C=A (92%) then A=hyper-validated (92%).
Further, under a simple majority rule, i.e. when two out of the three diagnostic assays agree, the biomarkers agreed with clinical diagnosis 100%.
Supplemental Tables:
This application claims priority to U.S. Provisional Patent Application No. 61/923,373, filed Jan. 3, 2014, the contents of which are incorporated herein by reference.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/US2015/010066 | 1/2/2015 | WO | 00 |
Number | Date | Country | |
---|---|---|---|
61923373 | Jan 2014 | US |