Methods for improving inflammatory bowel disease diagnosis
AssigneeNestec S.A.
InventorsFred Princen, Steven Lockton, Lisa J. Croner, Frederick A. Fletcher, Thomas Stockfisch, Sharat Singh
Life of this patent
Estimating the term…
Abstract
The present invention provides methods and systems to predict and diagnose inflammatory bowel disease (IBD) and subtypes such as ulcerative colitis (UC) and Crohn's disease (CD) by detecting the presence, absence, level, and/or genotype of one or more sero-genetic-inflammation markers. Advantageously, with the present invention, it is possible to provide a diagnosis of IBD versus non-IBD, to rule out IBD that is inconclusive for CD and UC, and to differentiate between CD and UC with increased accuracy.
Claims30
A method for diagnosing inflammatory bowel disease (IBD) and/or a clinical subtype thereof in an individual, said method comprising:
- (a) analyzing a sample obtained from said individual to determine the presence or level or genotype of at least each of the following markers to obtain a marker profile:
- (i) the presence or level of each of the serological markers ASCA-A, ASCA-G, ANCA, pANCA, anti-OmpC antibody, anti-CBir1 antibody, anti-FlaX antibody, and anti-A4-Fla2 antibody;
- (ii) the presence or level of each of the inflammation markers VEGF, ICAM, VCAM, SAA, and CRP;
- and (iii) the genotype of each of the genetic markers ATG16L1, ECM1, NKX2-3, and STAT3;
- (b) applying a first random forest statistical analysis to said marker profile to compute a first model score and comparing said first model score to a first cut-off value to obtain a decision whether said sample is an IBD sample or a non-IBD sample;
- (c) if said sample is an IBD sample, then applying a decision tree to said IBD sample to determine if said IBD sample is or is not an inconclusive sample,
- wherein said IBD sample is an inconclusive sample when said IBD sample has an ANCA level greater than a quartile score of 3 (>Q3), is pANCA2 positive, and has a level of anti-CBir1 antibody or anti-A4-Fla2 antibody or anti-FlaX antibody >Q3, or
- wherein said IBD sample is an inconclusive sample when said IBD sample is pANCA2 positive and expresses two out of three markers selected from anti-CBir1 antibody, anti-A4-Fla2 antibody, and anti-FlaX antibody >Q3, and
- wherein the presence or level of pANCA is used to determine the value of pANCA2;
- and
- (d) if said IBD sample is not an inconclusive sample, then applying a second random forest statistical analysis to said IBD sample to compute a second model score and comparing said second model score to a second cut-off value to determine a clinical subtype of IBD.
A method for diagnosing inflammatory bowel disease (IBD) and/or a clinical subtype thereof in an individual, said method comprising:
- (a) analyzing a sample obtained from said individual to determine the presence or level or genotype of at least each of the following markers to obtain a marker profile:
- (i) the presence or level of each of the serological markers ASCA-A, ASCA-G, ANCA, pANCA, anti-OmpC antibody, anti-CBir1 antibody, anti-FlaX antibody, and anti-A4-Fla2 antibody;
- (ii) the presence or level of each of the inflammation markers VEGF, ICAM, VCAM, SAA, and CRP;
- and (iii) the genotype of each of the genetic markers ATG16L1, ECM1, NKX2-3, and STAT3;
- (b) applying a first random forest statistical analysis to said marker profile to determine whether said sample is an IBD sample or a non-IBD sample with a sensitivity of at least 70%;
- (c) if said sample is an IBD sample, then applying a decision tree to said IBD sample to determine if said IBD sample is or is not an inconclusive sample,
- wherein said IBD sample is an inconclusive sample when said IBD sample has an ANCA level greater than a quartile score of 3 (>Q3), is pANCA2 positive, and has a level of anti-CBir1 antibody or anti-A4-Fla2 antibody or anti-FlaX antibody >Q3, or
- wherein said IBD sample is an inconclusive sample when said IBD sample is pANCA2 positive and expresses two out of three markers selected from anti-CBir1 antibody, anti-A4-Fla2 antibody, and anti-FlaX antibody >Q3, and
- wherein the presence or level of pANCA is used to determine the value of pANCA2;
- and
- (d) if said IBD sample is not an inconclusive sample, then applying a second random forest statistical analysis to said IBD sample to determine a clinical subtype of IBD with a sensitivity of at least 85% for Crohn's disease (CD) and a sensitivity of at least 95% for ulcerative colitis (UC).
Dependencies are inferred from the claim text.