Diagnostic methods based on lipid profiles

A method for discriminating between non-alcoholic steatohepatitis (NASH) and steatosis and correspondingly treating a subject previously diagnosed with non-alcoholic fatty liver disease (NAFLD), the method comprising:

• (i) fractionating by liquid chromatography a serum sample isolated from the subject;
• (ii) determining by mass spectrometry the levels of one or more metabolic markers selected from the group consisting of:
• (a) metabolic marker TG(44:1) comprising at least one of analytes TG(14:0+18:1+12:0) and TG(16:0+16:1+12:0);
• (b) metabolic marker TG(48:2) comprising at least one of analytes TG(14:0+18:1+16:1) and TG(16:0+18:2+14:0);
• (c) metabolic marker TG(49:1) comprising analyte TG(15:0+18:1+16:0);
• (d) metabolic marker TG(50:1) comprising analyte TG(16:0+18:1+16:0);
• (e) metabolic marker TG(51:1) comprising analyte TG(16:0+17:0+18:1);
• (f) metabolic marker TG(51:2) comprising analyte TG (16:0+17:1+18:1);
• (g) metabolic marker TG(51:3) comprising analyte TG(18:2+18:1+15:0);
• (h) metabolic marker TG(52:0) comprising analyte TG(16:0+18:0+18:0);
• (i) metabolic marker TG(52:3) comprising analyte TG(16:0+18:1+18:2);
• (j) metabolic marker TG(52:4) comprising analyte TG(16:0+18:2+18:2);
• (k) metabolic marker TG(53:3) comprising at least one of analytes TG(18:2+18:1+17:0) and TG(18:1+18:1+17:1);
• (l) metabolic marker TG(54:2) comprising analyte TG(18:0+18:1+18:1);
• (m) metabolic marker TG(54:3) comprising at least one of analytes TG(20:2+20:1+14:0), TG(20:2+18:1+16:0), TG(20:1+18:2+16:0), and TG(18:2+18:1+18:0);
• (n) metabolic marker TG(54:5) comprising analyte TG(18:2+18:2+18:1);
• (o) metabolic marker TG(54:6) comprising analyte TG(20:4+18:2+16:0);
• (p) metabolic marker TG(56:3) comprising analyte TG(18:1+20:1+18:1);
• (q) metabolic marker TG(56:7) comprising at least one of analytes TG(22:5+18:2+16:0) and TG(20:4+18:2+18:1);
• and
• (r) metabolic marker TG(56:8) comprising analyte TG(22:6+18:2+16:0);
• and
• (iii) comparing each of the levels obtained in (ii) to a reference value to obtain a relative value;
• (iv) generating a predictive score based on the relative values of the metabolic markers in a logistic regression model to discriminate between NASH and steatosis, wherein the predictive score is given by formula (I)
• predictive
• ⁢
• score
• =
• 1
• 1
• +
• e
• -
• z
• (
• I
• )
• wherein
• Z
• =
• ∑
• i
• ⁢
• c
• i
• ⁡
• [
• A
• i
• 1
• ]
• ⁡
• [
• A
• i
• 2
• ]
• ci is the coefficient estimated for the model for the parameter i, wherein each parameter i of the model includes Ai1 and Ai2 wherein Ai1 is a first metabolic marker of (a)-(r) determined in (ii) or a first metabolic marker logarithm, and Ai2 is a second metabolic marker of (a)-(r) determined in (ii), second metabolic marker logarithm, body mass index, or body mass index logarithm and [Ai1] and [Ai2] are corresponding relative values of Ai1 and Ai2;
• and
• (v) treating the patient for NASH when the predictive score is ≥0.5, and treating the patient for steatosis when the predictive score is <0.5.