TY - JOUR
T1 - Morphometric imaging biomarker identifies Alzheimer’s disease even among mixed dementia patients
AU - Chirila, Florin V.
AU - Xu, Guang
AU - Fontaine, Dan
AU - Kern, Grant
AU - Khan, Tapan K.
AU - Brandt, Jason
AU - Konishi, Yoshihiro
AU - Nebe-von-Caron, Gerhard
AU - White, Charles L.
AU - Alkon, Daniel L.
N1 - Funding Information:
We want to thank Mr. Paul Tanico, Synaps Dx, for the continuous financial support of these studies since 2016 and for supporting the gold standard autopsy confirmation. We would also like to thank him for understanding and accepting the expansion of the science behind these studies. We would also like to thank Mr. Frank Amato, Synaps Dx, for supporting the completion of the trial and supporting the data analysis under blind conditions. We thank Mr. William MacTurk, BS, for his valuable contribution to this study. We are thankful for all patient enrolment sites: Mary Quiceno, MD, and Ramon Diaz-Arrastia, MD, U T Southwestern Medical Center at Dallas, Dallas, TX 75390-9129. Alan Anderson, MD, Johns Hopkins University School of Medicine—The Gardens at William Hill Manor, 545 Cynwood Drive, Easton, MD 21601. Vassilis Koliatsos, MD, Johns Hopkins University School of Medicine—The Retreat at Sheppard Pratt. 6501 North Charles Street, Towson, MD 21204-6819. Autopsy Pathology Sites: Department of Pathology, Johns Hopkins University (under the supervision of Prof. Juan Troncoso). Finally, we appreciate the help of Dr. Shirley Neitch from Marshall University, Huntington, West Virginia, for her oversight and clinical diagnoses for the apparently healthy controls.
Funding Information:
Funding for the study was provided by Synaps Dx.
Publisher Copyright:
© 2022, The Author(s).
PY - 2022/12
Y1 - 2022/12
N2 - A definitive diagnosis of Alzheimer’s disease (AD), even in the presence of co-morbid neuropathology (occurring in > 50% of AD cases), is a significant unmet medical need that has obstructed the discovery of effective AD therapeutics. An AD-biomarker, the Morphometric Imaging (MI) assay on cultured skin fibroblasts, was used in a double-blind, allcomers (ages 55–90) trial of 3 patient cohorts: AD dementia patients, N = 25, all autopsy confirmed, non-AD dementia patients, N = 21—all autopsy or genetically confirmed; and non-demented control (AHC) patients N = 27. Fibroblasts cells isolated from 3-mm skin punch biopsies were cultured on a 3-D Matrigel matrix with movement dynamics quantified by image analysis. From counts of all aggregates (N) in a pre-defined field image and measures of the average area (A) of aggregates per image, the number-to-area ratios in a natural logarithmic form Ln(A/N) were determined for all patient samples. AD cell lines formed fewer large aggregates (cells clustered together) than non-AD or AHC cell lines. The cut-off value of Ln(A/N) = 6.98 was determined from the biomarker values of non-demented apparently healthy control (AHC) cases. Unequivocal validation by autopsy, genetics, and/or dementia criteria was possible for all 73 patient samples. The samples were collected from multiple centers—four US centers and one center in Japan. The study found no effect of center-to-center variation in fibroblast isolation, cell growth, or cell aggregation values (Ln(A/N)). The autopsy-confirmed MI Biomarker distinguished AD from non-AD dementia (non-ADD) patients and correctly diagnosed AD even in the presence of other co-morbid pathologies at autopsy (True Positive = 25, False Negative = 0, False Positive = 0, True Negative = 21, and Accuracy = 100%. Sensitivity and specificity were calculated as 100% (95% CI = 84 to 100.00%). From these findings, the MI assay appears to detect AD with great accuracy—even with abundant co-morbidity.
AB - A definitive diagnosis of Alzheimer’s disease (AD), even in the presence of co-morbid neuropathology (occurring in > 50% of AD cases), is a significant unmet medical need that has obstructed the discovery of effective AD therapeutics. An AD-biomarker, the Morphometric Imaging (MI) assay on cultured skin fibroblasts, was used in a double-blind, allcomers (ages 55–90) trial of 3 patient cohorts: AD dementia patients, N = 25, all autopsy confirmed, non-AD dementia patients, N = 21—all autopsy or genetically confirmed; and non-demented control (AHC) patients N = 27. Fibroblasts cells isolated from 3-mm skin punch biopsies were cultured on a 3-D Matrigel matrix with movement dynamics quantified by image analysis. From counts of all aggregates (N) in a pre-defined field image and measures of the average area (A) of aggregates per image, the number-to-area ratios in a natural logarithmic form Ln(A/N) were determined for all patient samples. AD cell lines formed fewer large aggregates (cells clustered together) than non-AD or AHC cell lines. The cut-off value of Ln(A/N) = 6.98 was determined from the biomarker values of non-demented apparently healthy control (AHC) cases. Unequivocal validation by autopsy, genetics, and/or dementia criteria was possible for all 73 patient samples. The samples were collected from multiple centers—four US centers and one center in Japan. The study found no effect of center-to-center variation in fibroblast isolation, cell growth, or cell aggregation values (Ln(A/N)). The autopsy-confirmed MI Biomarker distinguished AD from non-AD dementia (non-ADD) patients and correctly diagnosed AD even in the presence of other co-morbid pathologies at autopsy (True Positive = 25, False Negative = 0, False Positive = 0, True Negative = 21, and Accuracy = 100%. Sensitivity and specificity were calculated as 100% (95% CI = 84 to 100.00%). From these findings, the MI assay appears to detect AD with great accuracy—even with abundant co-morbidity.
UR - http://www.scopus.com/inward/record.url?scp=85141103678&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85141103678&partnerID=8YFLogxK
U2 - 10.1038/s41598-022-21796-y
DO - 10.1038/s41598-022-21796-y
M3 - Article
C2 - 36319674
AN - SCOPUS:85141103678
SN - 2045-2322
VL - 12
JO - Scientific Reports
JF - Scientific Reports
IS - 1
M1 - 17675
ER -