MEANS AND METHODS FOR DIAGNOSING CARDIAC DISEASE IN A SUBJECT

Abstract

The present invention relates to the field of diagnostic methods. Specifically, the present invention contemplates a method for diagnosing a cardiac disease in a subject based on determining the amounts of at least three lipid metabolite biomarkers and at least one further cardiac biomarker. The invention also relates to tools for carrying out the aforementioned methods, such as diagnostic devices.

Claims

1. A method for diagnosing cardiac disease comprising the steps of: a. determining in a sample of a subject the amounts of at least three lipid metabolite biomarkers and of at least one additional cardiac biomarker, and b. comparing the amounts as determined in step a. to a reference, whereby cardiac disease is to be diagnosed.

2. The method of claim 1, wherein said at least three lipid metabolite biomarkers are: i. at least one triacylglyceride biomarker, at least one cholesterylester biomarker, and at least one phosphatidylcholine biomarker; ii. at least one triacylglyceride biomarker, at least one phosphatidylcholine biomarker, and at least one sphingomyelin biomarker; iii. at least one triacylglyceride biomarker, at least one cholesterylester biomarker, and at least one sphingomyelin biomarker; iv. at least one phosphatidylcholine biomarker, at least one cholesterylester biomarker, and at least one sphingomyelin biomarker; v. Cholesterylester C18:2, SSS and Cer(d17:1/24:0); vi. at least two sphingomyelin biomarkers selected from the group consisting of SM2, SM3, SM5, SM18, SM23, SM24, and SM28, and at least one triacylglyceride biomarker selected from the group consisting of SOP2, SPP1 and PPO1, (or alternatively at least one triacylglyceride biomarker selected from the group consisting of SOP2, SPP1 and PPP); vii. at least two triacylglyceride biomarkers selected from the group consisting of OSS2, SOP2, SPP1 and SSP2, and at least one sphingomyelin biomarker selected from the group consisting of SM23 and SM24; viii. SM18, SM24 and SM28; or ix. the biomarkers of panel 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, or 206 of Table 2.

3. The method of claim 2, wherein the at least one triacylglyceride biomarker in i., ii., and iii. is selected from the group consisting of SOP2, OSS2, SPP1, SSP2, PPO1 and PPP, the at least one cholesterylester biomarker in i., iii., and iv. is selected from the group consisting of cholesterylester C18:2 and cholesterylester C18:0, the at least one phosphatidylcholine biomarker in i., ii. and iv. is selected from the group consisting of PC4 and PC8, and the at least one sphingomyelin biomarker in ii., iii., and iv. is selected from the group consisting of SM18, SM24, SM23, SM21, SM28, SM5, SM3, SM29 and SM8.

4. The method of any one of claims 1 to 3, wherein the at least one additional cardiac marker is selected from the group consisting of at least one general lipid cardiac biomarker, at least one lipoprotein subfraction biomarker, at least one apolipoprotein biomarker and at least one inflammation biomarker.

5. The method of claim 4, wherein the at least one general lipid cardiac biomarker is selected from the group consisting of total cholesterol, HDL-cholesterol (High Density Lipoprotein Cholesterol), triglycerides, LDL-cholesterol (High Density Lipoprotein Cholesterol), the ratio of total cholesterol to HDL-cholesterol, and non-HDL cholesterol, preferably wherein the amount(s) of HDL cholesterol and/or LDL cholesterol is (are) determined, more preferably, wherein the amount of HDL cholesterol is determined as additional cardiac biomarker.

6. The method of claims 4 and 5, wherein the at least one lipoprotein subfraction biomarker is selected from LDL particles, small LDL particles, medium LDL particles and large HDL particles.

7. The method of any one claims 4 to 6, wherein the at least one apolipoprotein biomarker is selected from apolipoprotein B and lipoprotein (a).

8. The method of any one of claims 4 to 7, wherein the at least one inflammation biomarker is selected from C-reactive protein (CRP), in particular high sensitivity C-reactive protein (hsCRP) and Lipoprotein-associated phospholipase A2 (Lp-PLA2).

9. The method of any one of claims 4 to 8, wherein the amounts of total cholesterol, HDL-cholesterol (High Density Lipoprotein Cholesterol), triglycerides, LDL-cholesterol (High Density Lipoprotein Cholesterol), non-HDL cholesterol, LDL particles, small LDL particles, medium LDL particles, large HDL particles, hsCRP, Lp-PLA2, apolipoprotein B and lipoprotein(a) are determined as cardiac markers.

10. The method of claim 8, further comprising the determination of the ratio of total cholesterol to HDL-cholesterol.

11. The method of any one of claims 1 to 10, wherein the cardiac disease is selected from peripheral artery disease, coronary artery disease, atherosclerosis, cardiomyopathy, heart failure, and pulmonary heart disease.

12. The method of claim 11, wherein the cardiac disease is coronary artery disease.

13. The method of claim 11, wherein the cardiac disease is heart failure.

14. The method of claims 11 and 13, wherein the heart failure is heart failure with reduced left ventricular ejection fraction (HFrEF).

15. The method any one of the preceding claims, wherein at least the amounts of the biomarkers of i., ii., iii., vi, vii., or ix as defined in claim 2 are determined, and wherein the at least one triacylglyceride biomarker in i., ii., and iii. is selected from the group consisting of SOP2, OSS2, SSP2, PPO1 and PPP, and/or (in particular and) the least one cholesterylester biomarker in i., and iii. is cholesterylester C18:2, and/or (in particular and) the least one phosphatidylcholine biomarker in i. and ii. is PC4, and/or (in particular and) the least one sphingomyelin biomarker in ii. and iii. is selected from the group consisting of SM18, SM24, SM23, SM28, SM5, and SM3.

16. The method of any one any one of the preceding claims, wherein at least the amounts of the biomarkers of panel 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, or 56 in Table 2 are determined.

17. The method of any one of the preceding claims, wherein at least the amounts of the biomarkers of i., ii., iii, vi, or ix are determined, and wherein the at least one triacylglyceride biomarker in i., ii., and iii. is SOP2 and/or OSS2, and/or (in particular and) wherein the at least one cholesterylester biomarker in i., and iii. is cholesterylester C18:2, and/or (in particular and) wherein the at least one phosphatidylcholine biomarker in i. and ii. is PC4, and/or (in particular and) wherein the at least one sphingomyelin biomarker in ii. and iii. is selected from the group consisting of SM18, SM24, SM23, and SM3.

18. The method of any one of the preceding claims, wherein at least the amounts of the biomarkers of panel 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17 or 18 in Table 2 are determined.

19. The method of any one of the preceding claims, wherein at least the amounts of the biomarkers of i., ii., iii, vi, or ix are determined, and wherein the at least one triacylglyceride biomarker in i., ii., and iii. is SOP2 and/or OSS2, and/or (in particular and) wherein the at least one cholesterylester biomarker in i., and iii. is cholesterylester C18:2, and/or (in particular and) wherein the at least one phosphatidylcholine biomarker in i. and ii. is PC4, and/or (in particular and) wherein the at least one sphingomyelin biomarker in ii. and iii. is SM23.

20. The method of any one of the preceding claims, wherein at least the amounts of the biomarkers of panel 1, 2, 3 or 4 in Table 2 are determined.

21. The method of any one of the preceding claims, wherein at least the amounts of OSS2, PC4 and SM23 are determined, in particular wherein at least the amounts of OSS2, PC4 and SM23, and the amount of NT-proBNP or BNP are determined.

22. The method of any one of the preceding claims, wherein at least the amounts of OSS2, cholesterylester C18:2 and SM23 are determined.

23. The method of any one of the preceding claims, wherein at least the amounts of the biomarkers of iii. are determined, and wherein the at least one triacylglyceride biomarker is SOP2 and/or OSS2, and/or (in particular and) wherein the at least one cholesterylester biomarker is cholesterylester C18:2, and/or (in particular and) wherein the at least one sphingomyelin biomarker is SM23.

24. The method of any one of the preceding claims, wherein at least the amounts of SOP2, OSS2, PC4, Cholesterylester C18:2, SM18, SM28, SM24, SSP2, and SM23 are determined.

25. The method of any one of the preceding claims, wherein the cardiac disease is heart failure with reduced left ventricular ejection fraction is DCMP (dilated cardiomyopathy).

26. The method of claim 25, wherein at least the amounts of the lipid metabolite biomarkers shown in i., ii., iii. or vii, in particular of in i., ii., or iii. are determined, and wherein the at least one triacylglyceride biomarker in i., ii. and iii. is selected from the group consisting of SOP2, OSS2, and SSP2, and/or (in particular and) wherein the at least one cholesterylester biomarker in i. and iii. is cholesterylester C18:2, and/or (in particular and) wherein the at least one phosphatidylcholine biomarker in i. and ii. is PC4, and/or (in particular and) wherein the at least one sphingomyelin biomarker in ii. and iii. is selected from the group consisting of SM24, SM23, SM28, and SM3.

27. The method of claims 25 and 26, wherein at least the amounts of the biomarkers of panel 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35 or 36 in Table 2 are determined.

28. The method of any one of claims 25 to 27, wherein at least the amounts of the lipid metabolite biomarkers shown in ii. or iii. are determined, and wherein the at least one triacylglyceride biomarker in ii. and iii. is SOP2 and/or OSS2, and/or (in particular and) wherein the at least one cholesterylester biomarker in iii. is cholesterylester C18:2, and/or (in particular and) wherein the at least one phosphatidylcholine biomarker in ii. is PC4, and/or (in particular and) wherein the at least one sphingomyelin biomarker in ii. and iii. is SM23 and/or SM24.

29. The method of any of claims 25 to 28, wherein at least the amounts of the biomarkers of panel 31, 32, 33, 34, 35 and 36 in Table 2 are determined.

30. The method of any one of the preceding claims, wherein the cardiac disease is heart failure with reduced left ventricular ejection fraction is ICMP (ischemic cardiomyopathy).

31. The method of claim 30, wherein at least the amounts of the lipid metabolite biomarkers shown in ii., iii. or vi, in particular in ii., or iii. are determined, and wherein the at least one triacylglyceride biomarker in ii. and iii. is SOP2 and/or OSS2 (in particular and), wherein the at least one cholesterylester biomarker in iii. is cholesterylester C18:2, and/or (in particular and) wherein the at least one phosphatidylcholine biomarker in ii. is PC4, and/or (in particular and) wherein the at least one sphingomyelin biomarker in ii. and iii. is selected from the group consisting of SM18, SM24, and SM23.

32. The method of claims 30 and 31, wherein at least the amounts of the biomarkers of panel 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55 or 56 in Table 2 are determined.

33. The method of any one of claims 30 to 32, wherein at least the amounts of the lipid metabolite biomarkers shown in iii. are determined, and wherein the at least one triacylglyceride biomarker is SOP2, and/or (in particular and) wherein the at least one cholesterylester biomarker is cholesterylester C18:2, and/or (in particular and) wherein the at least one sphingomyelin biomarker is SM18 and/or SM23.

34. The method of any one of claims 30 to 33, wherein at least the amounts of the biomarkers of panel 51, 52, 53, 54, 55 or 56 in Table 2 are determined.

35. The method of any one of the preceding claims, wherein the cardiac disease is HfrEF and wherein the HfrEP is asymptomatic (or wherein the subject does not show symptoms of heart failure).

36. The method of claim 36, and wherein at least the amounts of the biomarkers of, ii., iii. vi, vii., or ix are determined, and wherein the at least one triacylglyceride biomarker in ii., and iii. is selected from the group consisting of SOP2, OSS2, and PPO1, and/or (in particular and) wherein the at least one cholesterylester biomarker in iii. is cholesterylester C18:2, and/or (in particular and) the least one phosphatidylcholine biomarker in ii. is PC4, and/or (in particular and) wherein the at least one sphingomyelin biomarker in ii. and iii. is selected from the group consisting of SM24 and SM23.

37. The method of claims 35 and 35, wherein at least the amounts of the biomarkers of panel 7, 8, 9, 10, 11, 12, 19, 20, 21, 22, 23, 24, 37, 38, 39, 40, 41 or 42 in Table 2 are determined.

38. The method of any one of claims 35 to 37, wherein at least the amounts of the biomarkers of iii. or vi. (in particular of iii.) of claim 2 are determined, and wherein the at least one triacylglyceride biomarker is SOP2, and/or (in particular and) wherein the at least one cholesterylester biomarker is cholesterylester C18:2, and/or (in particular and) wherein the at least one sphingomyelin biomarker is selected from the group consisting of SM24 and SM23.

39. The method of claim 38, wherein at least the amounts of the biomarkers of panel 7, 8, 9, 10, 11, or 12 in Table 2 are determined.

40. The method of any one of claims 1 to 10, wherein the heart failure is heart failure with preserved ejection fraction (HFpEF).

41. The method of claim 40, wherein at least the amounts of the biomarkers of ii., vi, or vii. of claim 1 are determined, and wherein the at least one triacylglyceride biomarker in ii. is selected from the group consisting of SOP2, SSP2, SPP1 and PPO1, and/or (in particular and) the least one phosphatidylcholine biomarker in ii. is selected from the group consisting of PC4 and PC8, and/or (in particular and) the least one sphingomyelin biomarker in ii. is selected from the group consisting of SM18, SM24, SM23, SM28, SM5, and SM3.

42. The method of claim 41, wherein at least the amounts of the biomarkers of panel 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101 or 102, in Table 2 are determined.

43. The method of any one of the preceding claims, wherein the subject is a human subject.

44. The method of any one of the preceding claims, wherein the sample is blood, serum or plasma.

45. The method of any one of the preceding claims, wherein the method does not comprise the determination of NT-proBNP or BNP.

46. The method of any one of claims 1 to 44, further comprising determining the amount of NT-proBNP or BNP in a sample/the sample from the subject and comparing the amount of NT-proBNP or BNP to a reference.

47. The method of any one of claims 1 to 46, further comprising carrying out a correction for confounders.

48. The method of claim 47, wherein the confounders are age, BMI and/or gender, in particular age, BMI and gender.

49. The method of any one of the preceding claims, wherein in step b) a score is calculated based on the determined amounts of the at least three lipid metabolite biomarkers and the at least one additional cardiac biomarker, and wherein the reference is a reference score.

50. The method of any one of claims 1 to 49, wherein the reference is from a subject or group of subjects known not to suffer from cardiac disease.

51. The method of claim 50, wherein a value for each of the at least three lipid metabolite biomarkers and the at least one additional cardiac biomarker in the test sample being essentially identical as compared to the reference is indicative for the absence of cardiac disease.

52. The method of any one of claims 1 to 49, wherein the reference is from a subject or group of subjects known to suffer from heart failure.

53. The method of any one of claim 50, wherein a value for each of the at least three lipid metabolite biomarkers and the at least one additional cardiac biomarker in the test sample being essentially identical as compared to the reference is indicative for the presence of the heart failure.

54. The method of any one of claims 1 to 53, wherein the amounts of the at least three lipid metabolite biomarkers are determined by mass spectrometry (MS).

55. The method of claim 54, wherein the mass spectrometry is LC-MS, in particular LCMS/MS, or HPLC-MS, in particular HPLC-MS/MS.

56. The method of claims 54 and 55, wherein the mass spectrometry comprises an ionization step in which the at least three lipid metabolite biomarkers are ionized

57. The method of claim 56, wherein the ionization step is carried out by electrospray ionization, in particular by positive ion mode electrospray ionization.

58. A diagnostic device for carrying out the method according to any one of claims 1 to 57, comprising: a) an analysing unit comprising at least one detector for the at least three lipid metabolite bimarkers and at least one detector for and the at least one additional cardiac biomarker in connection with the present invention detected by the at least one detector, and, operatively linked thereto; b) an evaluation unit comprising a computer comprising tangibly embedded a computer program code for carrying out a comparison of the determined amounts of the at least three lipid metabolite biomarkers and the at least one additional cardiac biomarker with the reference amounts, and a data base comprising said reference amounts for the said biomarkers, whereby it will be diagnosed whether a subject suffers from a cardiac disease.

59. A diagnostic device for carrying out the method according to any one of claims 1 to 57, comprising: a) an analysing unit comprising at least one detector for the at least three lipid metabolite biomarkers and at least one detector for and the at least one additional cardiac biomarker in connection with the present invention detected by the at least one detector, and, operatively linked thereto; b) an evaluation unit comprises a computer comprising tangibly embedded a computer program code for calculating a score based on the determined amounts of the at least three lipid metabolite biomarkers and the at least one additional cardiac biomarker, and for carrying out a comparison of the calculated score and the reference score, wherein said evaluation unit further comprises a data base comprising said reference score, whereby it will be diagnosed whether a subject suffers from a cardiac disease.

60. Use of at least three lipid metabolite biomarkers and the at least one additional cardiac biomarker as set forth in the preceding claims in a sample of a subject for diagnosing cardiac disease.

61. The method of any one of claims 1 to 57, the device of claim 58 or 59 or the use of claim 60, wherein the cardiac disease is HFrEF with a left ventricular ejection fraction of lower than 50% but larger than 35%.

62. The method of any one of claim 1 to 57 or 61, the device of claim 58, 59 or 61, or the use of claim 60 or 61, wherein the subject is overweight.

63. The method of any one of claims 1 to 57 and 61 and 62, the diagnostic device of claim 58, 59, or 61 or the use of claims 60 and 61, wherein said at least three lipid metabolite biomarkers are the biomarkers of panel 1, and wherein said at least one additional cardiac biomarker is HDL cholesterol and/or LDL cholesterol, preferably wherein at least one additional cardiac biomarker is HDL cholesterol.

64. The method of claim 63, further comprising the determination of the amount of NT-proBNP in a sample/the sample from the subject and comparing the amount of NT-proBNP or BNP to a reference.

Description

EXAMPLES

Example 1: Study Design

[0688] A multicentric study with three clinical centers and in total 843 subjects was conducted. The study comprised 194 male and female DCMP, 183 male and female ICMP and 210 male and female HFpEF patients as well as 256 male and female healthy controls in an age range from 35-75 and a BMI range from 20-35 kg/m.sup.2. NYHA (New York Heart Association) scores of the patients ranged from I to III. Patients and controls were matched for age, gender and BMI. For all patients and controls, a blood sample was collected. Plasma was prepared by centrifugation, and samples were stored at 80 C. until measurements were performed.

[0689] Three subgroups of CHF (congestive heart failure) (DCMP, ICMP and HFpEF) were defined on the basis of echocardiography and hemodynamic criteria:

[0690] a) Subgroup DCMP: is hemodynamically defined as a systolic pump failure with cardic dilation (echocardiographic enhancement of the left ventricular end diastolic diameter >55 mm and a restricted left ventricular ejection fraction (LVEF) of <50%) without the presence of >50% stenosis

[0691] b) Subgroup ICMP: is hemodynamically defined as systolic pump failure due to a coronary insufficiency (>50% coronary stenosis and LVEF of <50%)

[0692] c) Subgroup heart failure with preserved ejection fraction (HFpEF): concentric heart hypertrophy (echocardiography: cardiac septum >12 mm and posterior myocardial wall >11 mm) and with a diastolic heart failure (non or mildly impaired pump function with LVEF of 50%) without a cardiac septum thickness >18 mm.

[0693] NYHA IV patients were excluded as well as patients suffering from apoplex, patients who had myocardial infarction within the last 4 months before testing, patients with altered medications within the last 4 weeks before testing as well as patients who suffered from acute or chronic inflammatory diseases and malignant tumours.

Example 2: Determination of Metabolites for the at Least Three Lipid Metabolite Biomarker Panels Shown in Table 2

[0694] Significantly altered lipid metabolite biomarkers in CHF vs healthy respectively HFrEF vs healthy patients have been identified. Lipid metabolite biomarkers have been selected according to their accessibility with a one-shot LC-MS/MS measurement with a simplified preanalytical process (see analytical method description, below).

[0695] An LC-MS/MS method for the analysis of the at least three lipid metabolite biomarkers was established. This method is capable to analyze all of the biomarkers as listed in Table 1, where the biomarkers are characterized by the combination of retention time and multiple reaction monitoring (MRM) transitions, and further potential lipid metabolites.

[0696] Each biomarker of Table 1 may contain more than one analyte, whereby the analytes contained in the same biomarker have the same total number of carbon atoms and the same total number of double bonds.

TABLE-US-00001 TABLE 1 Lipid metabolite biomarkers used for composition of the at least three lipid metabolite biomarker panels shown in Table 2 and their analytical characteristics. Transition (parent/ Retention fragment) Time Biomarker Analyte 1 Analyte 2 Analyte 3 Group (Da) (min) Cer(d16:1/24:0) Cer(d16:1/24:0) Ceramide 622.6/236.2 2.83 Cer(d17:1/24:0) Cer(d17:1/24:0) Ceramide 636.6/250.2 2.95 Cer(d18:1/23:0) Cer(d18:1/23:0) Ceramide 636.5/264.2 2.95 Cer(d18:1/24:1) Cer(d18:1/24:1) Ceramide 648.6/264.2 2.83 Cer(d18:2/24:0) Cer(d18:2/24:0) Ceramide 648.6/262.2 2.9 CE C18:0 CE C18:0 Cholesterylester 670.7/369.3 4.72 CE C18:2 CE C18:2 Cholesterylester 666.7/369.3 4.47 Glutamic acid Glutamic acid amino acid 148.0/84.1 0.23 PC4 PC (C16:0 C18:2) Phosphatidylcholine 758.6/184.1 1.72 PC8 PC (C18:0 C18:2) PC (C18:1 C18:1) Phosphatidylcholine 786.6/184.1 1.88 SM10 SM(d18:1/18:0) SM(d16:1/20:0) Sphingomyeline 731.5/184.1 1.7 SM18 SM(d18:1/21:0) SM(d16:1/23:0) SM(d17:1/22:0) Sphingomyeline 773.5/184.1 1.97 SM2 SM(d18:1/14:0) SM(d16:1/16:0) Sphingomyeline 675.5/184.1 1.47 SM21 SM(d17:1/23:0) SM(d18:1/22:0) SM(d16:1/24:0) Sphingomyeline 787.5/184.1 2.05 SM23 SM(d18:1/23:1) SM(d18:2/23:0) SM(d17:1/24:1) Sphingomyeline 799.5/184.1 2.02 SM24 SM(d18:1/23:0) SM(d17:1/24:0) Sphingomyeline 801.5/184.1 2.16 SM28 SM(d18:1/24:0) Sphingomyeline 815.5/184.1 2.27 SM29 SM(d18:2/17:0) Sphingomyeline 715.5/184.1 1.57 SM3 SM(d17:1/16:0) Sphingomyeline 689.5/184.1 1.52 SM5 SM(d18:1/16:0) SM(d16:1/18:0) Sphingomyeline 703.5/184.1 1.57 SM8 SM(d18:2/18:1) Sphingomyeline 727.5/184.1 1.55 SM9 SM(d18:1/18:1) SM(d18:2/18:0) Sphingomyeline 729.5/184.1 1.62 OSS2 TAG C18:1 C18:0 C18:0 Triacylglyceride 906.9/605.4 4.87 PPO1 TAG C16:0 C16:0 C18:1 Triacylglyceride 850.8/577.5 4.64 PPP TAG C16:0 C16:0 C16:0 Triacylglyceride 824.8/551.5 4.61 SOP2 TAG C18:0 C18:1 C16:0 Triacylglyceride 878.9/577.6 4.76 SPP1 TAG C18:0 C16:0 C16:0 Triacylglyceride 852.9/579.6 4.74 SSP2 TAG C18:0 C18:0 C16:0 Triacylglyceride 880.9/579.6 4.85 SSS TAG C18:0 C18:0 C18:0 Triacylglyceride 908.9/607.6 4.96

[0697] In Table 2, the at least three lipid metabolite biomarker panels (for combination with at least one cardiac biomarker and optionally NT-proBNP) according to the present invention are listed. In column 1 the respective panel number is given, in column 2 the composition of each at least three lipid metabolite biomarker panel, and in column 3 the number of lipid metabolite biomarkers is given.

TABLE-US-00002 TABLE 2 at least three lipid metabolite biomarker panels for combination with at least one additional cardiac biomarker and optionally BNP or NT-proBNP, in particular NT-proBNP Panel No. of Number Composition of the .sub.at least three lipid metabolite biomarker panel Biomarkers 1 SM23; OSS2; PC4 3 2 OSS2; SM23; CE C18:2 3 3 SOP2; OSS2; PC4; CE C18:2; SM18; SM28; SM24; SSP2; SM23 9 4 OSS2; CE C18:2; SM23 3 5 SOP2; OSS2; SM18; CE C18:2; SM24; SM28; Cer(d16:1/24:0); PC4 8 6 SM18; OSS2; CE C18:2 3 7 SM23; SOP2; CE C18:2 3 8 SM23; SOP2; CE C18:2 3 9 SOP2; SM24; SSP2; SM23; CE C18:2; SM28; SM18; PC4 8 10 SM23; SOP2; SM28 3 11 SM24; SOP2; CE C18:2; SM28; SSP2; SM18; OSS2; SM2; Cer(d16:1/24:0) 9 12 SM24; SOP2; CE C18:2 3 13 OSS2; PC4; SM23 3 14 OSS2; CE C18:2; SM3 3 15 SM18; SM28; PC4; OSS2; SOP2; CE C18:2; PPO1; SM10 8 16 CE C18:2; SOP2; SM18 3 17 SM18; OSS2; SOP2; CE C18:2; PC4; SPP1; SM24; Cer(d16:1/24:0); SM28; SM21 10 18 SM18; CE C18:2; OSS2 3 19 SM23; SOP2; CE C18:2 3 20 OSS2; SM23; CE C18:2 3 21 CE C18:2; OSS2; SM23; SM24; SSP2 5 22 OSS2; SM23; CE C18:2 3 23 SM23; CE C18:2; OSS2; SSP2; SM24 5 24 SM24; SSP2; OSS2 3 25 SOP2; PC4; SM3 3 26 OSS2; PC4; SM3 3 27 SSP2; PPO1; SM18; CE C18:2; OSS2; SOP2; PC4; SM28 8 28 SOP2; SM28; PC4 3 29 Cer(d16:1/24:0); SM28; PC4; SM24; CE C18:2; SPP1; OSS2; SOP2 8 30 OSS2; CE C18:2; SM24 3 31 SOP2; SM23; PC4 3 32 OSS2; CE C18:2; SM23 3 33 SM24; SOP2; OSS2; CE C18:2; SSP2; PC4; SM28 7 34 SOP2; SM24; PC4 3 35 OSS2; CE C18:2; SOP2; SM18; SM24; SSP2; SM28; PC4 8 36 OSS2; SM24; CE C18:2 3 37 SM5; PPO1; SM23 3 38 SM5; SOP2; SM24 3 39 SSP2; CE C18:2; SM18; SM23; Cer(d16:1/24:0); PC4; SOP2; PPO1; SM28; SM24 10 40 SM24; SOP2; PC4 3 41 PC4; SM5; CE C18:2; SM2; PPO1; SOP2; SM28; Cer(d16:1/24:0); SM24 9 42 SM24; CE C18:2; SOP2 3 43 Cer(d18:1/24:1); SOP2; CE C18:2; SM18 4 44 SM18; CE C18:2; SOP2 3 45 SOP2; SM24; CE C18:2; SM18 4 46 SM18; CE C18:2; SOP2 3 47 SSP2; PC4; Cer(d18:1/24:1); SM28; SM24; SOP2; CE C18:2; SM18 8 48 SM18; CE C18:2; SOP2 3 49 SM28; CE C18:2; SOP2; SM18 4 50 SM18; CE C18:2; SOP2 3 51 SM23; CE C18:2; SOP2 3 52 SM23; CE C18:2; SOP2 3 53 SOP2; SM24; CE C18:2; SM18; PC4; SM28; PPO1; Cer(d16:1/24:0); OSS2; Cer(d18:1/24:1) 10 54 SM18; CE C18:2; SOP2 3 55 SOP2; SM18; SM24; CE C18:2; Cer(d16:1/24:0); SM28; OSS2; PC4 8 56 SM18; CE C18:2; SOP2 3 57 SOP2; SM23; PC4 3 58 SOP2; SSP2; SM5; SPP1; SM2; SM3; PC4; CE C18:2; Glutamic acid 9 59 SOP2; SM5; SM2 3 60 SOP2; SM23; SSP2; SM24; CE C18:2; PC4; OSS2 7 61 SOP2; SM23; CE C18:2 3 62 SOP2; SSP2; SM2; CE C18:2; SM24; OSS2; SM18; PC4; SM28 9 63 SOP2; SM18; CE C18:2 3 64 SOP2; PC4; OSS2; Cer(d18:1/24:1); SM24; PC8; SM21; Cer(d17:1/24:0) 8 65 SOP2; PC4; SM24 3 66 SOP2; SM18; PC4; SM21 4 67 SOP2; SM18; PC4 3 68 SOP2; PC4; SM28; SM18; Cer(d18:1/24:1); Cer(d18:2/24:0); SM9 7 69 SOP2; PC4; SM18 3 70 PC4; SOP2; SM18; SM24; OSS2; SPP1; SM21; Cer(d16:1/24:0); SM28; CE C18:2 10 71 SM18; SOP2; PC4 3 72 SOP2; PC4; SM5 3 73 SOP2; SM3; PC4; SM21 4 74 SOP2; SM3; PC4 3 75 SOP2; SM18; PC4; SM28; OSS2; SSP2; CE C18:2 7 76 SOP2; SM18; PC4 3 77 SOP2; OSS2; SM18; SM24; PC4; CE C18:2; Cer(d16:1/24:0); SM28 8 78 OSS2; SM18; CE C18:2 3 79 SPP1; SOP2; SM5; PC4; SM3; SSP2 6 80 SPP1; SM5; PC4 3 81 SPP1; SM3; SM5; SSP2; PC4; Glutamic acid; CE C18:0; OSS2; SOP2 9 82 SPP1; SM3; SM5 3 83 SPP1; SOP2; SSP2; PC4; SM23; SM3; SM3; CE C18:2 8 84 SPP1; SM23; SOP2 3 85 SM3; SPP1; SSP2; SM5; SOP2; PC4 6 86 SPP1; SM3; SM5 3 87 SOP2; Cer(d18:1/24:1); PC8; PPO1; Cer(d18:1/23:0); PC4; SM24; SM28; SM9 9 88 SOP2; PC4; SM24 3 89 Cer(d18:1/23:0); SM24; SOP2; Cer(d18:1/24:1); PC4; PPO1; PCS; SSS; SM18 9 90 SM24; SOP2; PC4 3 91 PC4; SM28; SM24; Cer(d18:1/24:1); PC8; SOP2; SM9; PPO1; SM21; SSP2 10 92 PC4; SM28; SOP2 3 93 SM24; PC4; SM18; SOP2; SM21; SPP1; Cer(d18:1/23:0); SSP2 8 94 SM24; SM18; SOP2 3 95 SOP2; PC8; SPP1; PC4; SSP2; PPO1; SM5; SM18; Cer(d17:1/24:0) 9 96 SOP2; PC8; SM5 3 97 SPP1; PC4; SM3; SOP2; SSP2; Cer(d18:1/23:0); SM24; SM5; Glutamic acid 9 98 SPP1; SM3; PC4 3 99 SOP2; PC4; SPP1; PC8; SSP2; SM28; SM24; PPO1; SM18; Cer(d17:1/24:0) 10 100 SOP2; PC4; SM24 3 101 PC4; SPP1; SM24; SSP2; SM3; SOP2; SM18; Glutamic acid; Cer(d17:1/24:0); Cer(d18:1/23:0) 10 102 SPP1; SM18; PC4 3 103 CE C18:2; SSS; Cer(d17:1/24:0) 3 104 PC4; SOP2; CE C18:2 3 105 CE C18:2; PC4; SM29; SOP2 4 106 CE C18:2; PC8; SM29; SSP2 4 107 CE C18:2; PC4; SM8; SSP2 4 108 CE C18:2; PC8; SM8; SSP2 4 109 CE C18:2; PC4; SM29; PPO1 4 110 CE C18:2; PC8; SM29; PPO1 4 111 CE C18:2; PC4; SM8; PPO1 4 112 CE C18:2; PC8; SM8; PPO1 4 113 CE C18:2; PC4; SM29; PPP 4 114 CE C18:2; PC8; SM29; PPP 4 115 CE C18:2; PC4; SM8; PPP 4 116 CE C18:2; PC8; SM29; SOP2 4 117 CE C18:2; PC8; SM8; PPP 4 118 CE C18:2; PC4; SM8; SOP2 4 119 CE C18:2; PC8; SM8; SOP2 4 120 CE C18:2; PC4; SM29; SPP1 4 121 CE C18:2; PC8; SM29; SPP1 4 122 CE C18:2; PC4; SM8; SPP1 4 123 CE C18:2; PC8; SM8; SPP1 4 124 CE C18:2; PC4; SM29; SSP2 4 125 PC4; SOP2; CE C18:2; SM18 4 126 PC4; SOP2; CE C18:0; SM18 4 127 PC4; SOP2; CE C18:2; SM21 4 128 PC4; SOP2; CE C18:0; SM21 4 129 PC4; SOP2; CE C18:2; SM23 4 130 PC4; SOP2; CE C18:0; SM23 4 131 PC4; SOP2; CE C18:2; SM24 4 132 PC4; SOP2; CE C18:0; SM24 4 133 SOP2; CE C18:0; SM18; PC4; SPP1 5 134 SOP2; CE C18:2; SM21; PC4; SPP1 5 135 SOP2; CE C18:0; SM21; PC4; SPP1 5 136 SOP2; CE C18:2; SM23; PC4; SPP1 5 137 SOP2; CE C18:0; SM23; PC4; SPP1 5 138 SOP2; CE C18:2; SM24; PC4; SPP1 5 139 SOP2; CE C18:0; SM24; PC4; SPP1 5 140 SOP2; CE C18:2; SM18; PC4; SSP2 5 141 SOP2; CE C18:0; SM18; PC4; SSP2 5 142 SOP2; CE C18:2; SM21; PC4; SSP2 5 143 SOP2; CE C18:0; SM21; PC4; SSP2 5 144 SOP2; CE C18:2; SM23; PC4; SSP2 5 145 SOP2; CE C18:0; SM23; PC4; SSP2 5 146 SOP2; CE C18:2; SM24; PC4; SSP2 5 147 SOP2; CE C18:0; SM24; PC4; SSP2 5 148 SOP2; CE C18:2; SM18; PC4; PPO1 5 149 SOP2; CE C18:0; SM18; PC4; PPO1 5 150 SOP2; CE C18:2; SM21; PC4; PPO1 5 151 SOP2; CE C18:0; SM21; PC4; PPO1 5 152 SOP2; CE C18:2; SM23; PC4; PPO1 5 153 SOP2; CE C18:0; SM23; PC4; PPO1 5 154 SOP2; CE C18:2; SM24; PC4; PPO1 5 155 SOP2; CE C18:0; SM24; PC4; PPO1 5 156 SOP2; CE C18:2; SM18; PC4; PPP 5 157 SOP2; CE C18:0; SM18; PC4; PPP 5 158 SOP2; CE C18:2; SM21; PC4; PPP 5 159 SOP2; CE C18:0; SM21; PC4; PPP 5 160 SOP2; CE C18:2; SM23; PC4; PPP 5 161 SOP2; CE C18:0; SM23; PC4; PPP 5 162 SOP2; CE C18:2; SM24; PC4; PPP 5 163 SOP2; CE C18:0; SM24; PC4; PPP 5 164 SOP2; CE C18:2; SM18; PC4; SPP1 5 165 PC4; PPP; CE C18:0; SM18 4 166 PC4; SPP1; CE C18:2; SM21 4 167 PC4; SSP2; CE C18:2; SM21 4 168 PC4; PPO1; CE C18:2; SM21 4 169 PC4; PPP; CE C18:2; SM21 4 170 PC4; SPP1; CE C18:0; SM21 4 171 PC4; SSP2; CE C18:0; SM21 4 172 PC4; PPO1; CE C18:0; SM21 4 173 PC4; SPP1; CE C18:2; SM18 4 174 PC4; PPP; CE C18:0; SM21 4 175 PC4; SPP1; CE C18:2; SM23 4 176 PC4; SSP2; CE C18:2; SM23 4 177 PC4; PPO1; CE C18:2; SM23 4 178 PC4; PPP; CE C18:2; SM23 4 179 PC4; SPP1; CE C18:0; SM23 4 180 PC4; SSP2; CE C18:0; SM23 4 181 PC4; PPO1; CE C18:0; SM23 4 182 PC4; SSP2; CE C18:2; SM18 4 183 PC4; PPP; CE C18:0; SM23 4 184 PC4; SPP1; CE C18:2; SM24 4 185 PC4; SSP2; CE C18:2; SM24 4 186 PC4; PPO1; CE C18:2; SM24 4 187 PC4; PPP; CE C18:2; SM24 4 188 PC4; SPP1; CE C18:0; SM24 4 189 PC4; SSP2; CE C18:0; SM24 4 190 PC4; PPO1; CE C18:0; SM24 4 191 PC4; PPO1; CE C18:2; SM18 4 192 PC4; PPP; CE C18:0; SM24 4 193 PC4; PPP; CE C18:2; SM18 4 194 PC4; SPP1; CE C18:0; SM18 4 195 PC4; SSP2; CE C18:0; SM18 4 196 PC4; PPO1; CE C18:0; SM18 4 197 PC4; SOP2; CE C18:2; SM28 4 198 PC4; SOP2; CE C18:0; SM28 4 199 SM18; SM24; SM28 3 200 OSS2; SM23; CE C18:2; PC4 4 201 CE C18:2; SM18; SSS 3 202 CE C18:2; PC8; SM18; SM2; SM24; SSS 6 203 CE C18:2; SM18; SM21; SM24; SSS 5 204 SM24; CE C18:2; SM2 3 205 SM2; SSS; CE C18:2 3 206 SM24; SSS; CE C18:2 3 Panels 1 to 206 were previously shown to allow for the diagnosis of heart failure (e.g. in combination with NT-proBNP). The results are shown in WO2016/034600. The International patent application claims the priorities of U.S. 62/044,367, EP 14183105.7 and U.S. 62/128,586. The International patent application as well as the three priority applications are herewith incorporated by reference in their entirety.

Example 3: Analytical Method for the Determination of the at Least Three Lipid Metabolite Biomarkers

[0698] Human plasma samples were prepared and subjected to HPLC-MS/MS analysis as described in the following:

[0699] 10 l plasma were mixed with 1500 l extraction solvent containing methanol/dichloromethane (in a ratio of 2:1, v/v) and 10 l internal standard mixture in a 2 ml safelock microcentrifuge tube (Eppendorf, Germany). Lipid standards were purchased from Avanti Polar Lipids, CA, U.S.A., Larodan Fine Chemicals, Sweden, Sigma-Aldrich, MO, U.S.A., or Tokyo Chemical Industry, Japan.

[0700] Ultrapure water (Milli-Q water system, Millipore) and analytical grade chemicals were used for extraction, dilution or as LC solvents. Quality control and reference sample were prepared from commercially available human plasma (RECIPE Chemicals+Instruments GmbH). Delipidized plasma (Plasma, Human, Defibrinated, Delipidized, 2 Charcoal treated, Highly Purified; USBio) was used for the preparation of calibrators and blanks.

[0701] After thoroughly mixing at 20 C. for 5 min, the precipitated proteins were removed by centrifugation for 10 min. An aliquot of the liquid supernatant was transferred to an appropriate glass vial and stored at 20 C. until analysis by LC-MS/MS. This sample preparation method uses protein precipitation as the only purification step to capture all lipids of interest (PC, SM, Cer, CE, TAG), so that a more comprehensive and complex extraction of lipids (as described e.g. by Folch, or Bligh & Dyer) was not necessary. The HPLC-MS/MS systems consisted of an Agilent 1100 LC system (Agilent Technologies, Waldbronn, Germany) coupled to an ABSciex API 4000 triple quadrupole mass spectrometer (ABSCIEX, Toronto, Canada). HPLC analysis was performed at 55 C. on commercially available reversed phase separation columns with C18 stationary phases (Ascentis Express C18 column (5 cm2.1 mm, 2.7 m, Phenomenex, Germany)).

[0702] Up to 5 L of the crude extract were injected and separated by gradient elution using a mixture of solvents consisting of methanol, water, formic acid, 2-propanol and 2-methoxy-2-methylpropan:

[0703] Solvent A: 400 g methanol, 400 g water, 1 g formic acid

[0704] Solvent B: 400 g tert-butyl methyl ether (tBME), 200 g 2-propanol, 100 g methanol, 1 g formic acid

[0705] Mass spectrometry was carried out by electrospray ionization in positive ion mode using multiple-reaction-monitoring (MRM). The source parameters were: nebulizer gas, 50; heater gas, 60; curtain gas, 25; CAD gas, 4; ion spray voltage, 5500 V; temperature, 400 C.; pause between mass ranges, 5 ms; resolution Q1 and Q3, unit. To enhance the ionization efficiency in the electrospray process for some lipids (CE, TAG), ammonium formate buffer dissolved in methanol (Solvent C: 200 g methanol, 30 g 0.1M ammonium formate solution in water) was added post column during the elution time of CE and TAG. The method is intended to be compatible with e.g. the ABSciex 3200MD benchtop LC-MS/MS system.

[0706] Furthermore, as some highly abundant lipids are out of the dynamic range of the MS detector, MS parameterslike collision energywere changed for said highly abundant lipids to get lower signal intensities due to lower fragmentation efficiencies.

[0707] Using electrospray isonization (ESI), PC 8 and several SMs with equal numbers of carbons and double bonds were detected together (as to say without separation of said species), since these isobaric species were not separated chromatographically. Chromatography was required to detect the low-concentrated ceramides and to separate phosphatidiyl cholines and sphingomyelins from each other, because the C13-impact of PC to SM or SM to PC disturbs the metabolite performance.

[0708] Positive and negative controls were prepared by lyophilization different amounts of commercially available plasma to meet the positive and negative cutoffs for all down regulated metabolites. For the triacylglycerides (TAG) the corresponding TAG was added into the plasma before the lyophilisation process but avoiding the protein to precipitate. The lyophilisated samples were stored in the freezertill sample preparation.

[0709] Quality control samples were prepared by extracting commercially available plasma with extraction solvent. Several aliquots of these extracts were stored in the freezer and used for the daily quality control of the instrument performance and sample preparation.

[0710] Plasma samples were analyzed in randomized analytical sequence design. Following comprehensive analytical validation steps, the resulting peak areas were divided by the peak areas of an internal standard with similar analytical behavior to reduce analytical variation. The resulting ratios were log 10-transformed to achieve normal distribution.

Example 4: Data Analysis and Statistical Evaluation

[0711] For each lipid metabolite biomarker listed in Table 1, the direction of change in CHF patients relative to healthy control subjects was calculated by ANOVA (see Tables 1A and 1B). The direction Up means that the levels of the biomarker are higher in CHF patients than in healthy control subjects, the direction Down means that the levels of the biomarker are lower in CHF patients than in healthy control subjects. The direction of change was found to be the same for all CHF patients compared to healthy control subjects taken together [ANOVA model: CHF+CENTER+(GENDER+AGE+BMI)2], for HFpEF patients compared to healthy control subjects, and for HFrEF patients compared to healthy control subjects [CHF_SUBGROUP+CENTER+(GENDER+AGE+BMI)2].

TABLE-US-00003 TABLE 1A Results of the ANOVA analysis as described above regarding the different lipid metabolite biomarkers from Table 1. CHF_ALL HFpEF HFrEF Biomarker Direction p < 0.05 Ratio p-Value Direction p < 0.05 Ratio p-Value Direction p < 0.05 Ratio p-Value Cer(d16:1/24:0) Down Yes 0.76 3.08E09 Down Yes 0.84 1.74E03 Down Yes 0.72 2.72E11 Cer(d17:1/24:0) Down Yes 0.78 1.88E09 Down Yes 0.82 7.38E05 Down Yes 0.76 4.17E10 Cer(d18:1/23:0) Down Yes 0.86 2.22E06 Down Yes 0.87 1.62E04 Down Yes 0.86 6.23E06 Cer(d18:1/24:1) Down No 0.98 5.77E01 Down No 0.97 4.07E01 Down No 0.99 7.62E01 Cer(d18:2/24:0) Down Yes 0.86 4.78E05 Down Yes 0.90 1.24E02 Down Yes 0.85 1.53E05 CE C18:0 Down No 0.94 1.55E01 Down No 0.99 8.34E01 Down No 0.92 5.65E02 CE C18:2 Down Yes 0.87 1.21E13 Down Yes 0.92 1.31E04 Down Yes 0.85 2.72E17 Glutamic acid Down No 0.93 1.31E01 Down Yes 0.89 3.62E02 Down No 0.96 3.62E01 OSS2 Up Yes 1.75 2.67E11 Up Yes 1.54 2.03E05 Up Yes 1.87 2.13E12 PC4 Down Yes 0.90 1.31E08 Down Yes 0.92 1.09E04 Down Yes 0.89 5.37E09 PC8 Down Yes 0.92 6.89E05 Down Yes 0.93 4.44E03 Down Yes 0.92 6.56E05 PPO1 Up Yes 1.49 1.29E07 Up Yes 1.35 1.16E03 Up Yes 1.57 1.81E08 PPP Up Yes 1.61 1.16E06 Up Yes 1.46 1.39E03 Up Yes 1.70 4.38E07 SM10 Down Yes 0.95 2.18E02 Down No 0.96 8.57E02 Down Yes 0.95 2.44E02 SM18 Down Yes 0.79 1.41E18 Down Yes 0.84 8.68E08 Down Yes 0.76 1.55E21 SM2 Down Yes 0.81 3.77E13 Down Yes 0.85 3.14E06 Down Yes 0.79 1.83E14 SM21 Down Yes 0.84 4.51E15 Down Yes 0.89 1.22E05 Down Yes 0.81 2.54E18 SM23 Down Yes 0.83 2.44E15 Down Yes 0.88 1.70E06 Down Yes 0.81 1.07E17 SM24 Down Yes 0.81 2.42E17 Down Yes 0.86 5.47E07 Down Yes 0.78 1.92E20 SM28 Down Yes 0.83 1.22E13 Down Yes 0.90 2.92E04 Down Yes 0.79 8.08E18 SM29 Down Yes 0.86 4.18E07 Down Yes 0.88 3.22E04 Down Yes 0.85 3.49E07 SM3 Down Yes 0.81 4.76E14 Down Yes 0.83 8.12E08 Down Yes 0.79 2.51E14 SM5 Down Yes 0.89 1.94E10 Down Yes 0.90 2.20E06 Down Yes 0.88 3.05E10 SM8 Down Yes 0.87 5.27E07 Down Yes 0.89 2.50E04 Down Yes 0.87 5.99E07 SM9 Down No 0.97 2.21E01 Down No 0.98 4.67E01 Down No 0.97 1.89E01 SOP2 Up Yes 1.70 7.93E11 Up Yes 1.50 3.73E05 Up Yes 1.81 6.21E12 SPP1 Up Yes 1.75 3.54E09 Up Yes 1.59 5.96E05 Up Yes 1.85 1.32E09 SSP2 Up Yes 1.64 2.51E09 Up Yes 1.50 6.41E05 Up Yes 1.73 7.37E10 SSS Up Yes 1.27 1.11E04 Up Yes 1.22 1.06E02 Up Yes 1.31 6.26E05

TABLE-US-00004 TABLE 1B Results of the ANOVA analysis for CHF subgroups ICMP and DCMP as described above regarding the different lipid metabolite biomarkers from Table 1. Only the lipid metabolite biomarkers from Panel 1 in Table 2 are listed. ICMP DCMP Biomarker Direction p < 0.05 Ratio p-Value Direction p < 0.05 Ratio p-Value OSS2 Up Yes 1.77 5.01E08 Up Yes 1.96 2.85E11 PC4 Down Yes 0.86 1.52E10 Down Yes 0.92 9.52E05 SM23 Down Yes 0.78 6.64E19 Down Yes 0.84 3.67E10

[0712] As comparison for the combination of the lipid metabolite biomarkers of Panel 1 with NT-proBNP and the additional cardic biomarkers HDL cholesterol and LDL cholesterol as mentioned in Example 8, the data of the study described in Example 1 were utilized for the evaluation of the diagnostic power for the classification of CHF subgroups compared with controls in combination with the peptide NT-proBNP. CHF patients were subdivided based on CHF subtype [HFpEF, DCMP, ICMP, or alternatively the joined DCMP+ICMP group named HFrEF (heart failure with reduced ejection fraction)] and a measure for the severety of the disease. Two different measures for the severety of the disease were used: NYHA class and LVEF. To this end, CHF patients were subdivided into those having no or only mild symptoms (NYHA class I and early stages of NYHA class II; sometimes referred to as asymptomatic) and those having more severe symptoms (NYHA classes II and III, sometimes referred to as symptomatic). Alternatively, CHF patient were subdivided into those with severely reduced LVEF (LVEF <35%) and those with mildly reduced or non-reduced LVEV (LVEF35). The latter subgroup comprises HFrEF patient with 35% LVEF <50% as well as all HFpEF patients (LVEF 50%).

[0713] A total set of about 850 samples was split into a training or identification set (also referred to as ID Data set; about 66% of the samples) and a testing set (also referred to as VD Data set; about 33% of the samples). The split was done group-wise, with groups being defined by the combination of gender, type of heart failure and NYHA levels.

[0714] The total sample numbers and compositions were as follows:

TABLE-US-00005 ID Data Set: VD Data Set: CHF: 374 samples CHF: 205 samples CHF, NYHA I-I/II: 169 samples CHF, NYHA I-I/II: 93 samples CHF, NYHA II-III: 205 samples CHF, NYHA II-III: 112 samples CHF, LVEF 35%: 275 samples* CHF, LVEF 35: 145 samples HFpEF: 134 samples HFpEF: 74 samples ICMP: 118 samples ICMP: 65 samples ICMP, NYHA I-I/II: 44 samples ICMP, NYHA I-I/II: 24 samples ICMP, NYHA II-III: 74 samples ICMP, NYHA II-III: 41 samples DCMP: 122 samples DCMP: 66 samples DCMP, NYHA I-I/II: 46 samples DCMP, NYHA I-I/II: 25 samples DCMP, NYHA II-III: 76 samples DCMP, NYHA II-III: 41 samples HFrEF: 240 samples HFrEF: 131 samples HFrEF, NYHA I-I/II: 90 samples HFrEF, NYHA I-I/II: 49 samples HFrEF, NYHA II-III: 150 samples HFrEF, NYHA II-III: 82 samples HFrEF, 35% LVEF < 50%: 141 samples* HFrEF, 35% LVEF < 50%: 71 samples HFrEF, LVEF < 35%: 98 samples* HFrEF, LVEF < 35%: 60 samples Controls: 167 samples Controls: 87 samples** *One CHF patient with missing LVEF measurement was not included in these subgroups **3 controls were subsequently excluded due to HF-related medication; re-analysis showed no significant changes with regard to the diagnostic performance of the at least three lipid metabolite biomarker panels

Example 5

[0715] Prediction probabilities for each patient were calculated with a logistic regression model fitted using the elastic net algorithm as implemented in the R package glmnet (Zou, H. and Hastie, T., 2003: Regression shrinkage and selection via the elastic net, with applications to microarrays. Journal of the Royal Statistical Society: Series B, 67, 301-320; Friedman, J., Hastie, T., and Tibshirani, R, 2010: Regularization Paths for Generalized Linear Models via Coordnate Descent. J. Stat. Softw. 33) based on the at least lipid metabolite biomarker panel 1 as well as based on the at least lipid metabolite biomarker panel 1+NT-proBNP. The fitting was performed on the data from the ID cohort. The L1 and the L2 penalties were given equal weight. Log-transformed peak area ratios were centered and scaled to unit variance before the anaysis. The prediction probability was calculated using the formula

[00004]$p=\frac{1}{1+e^{-\left(w_0+{}_{i=1}^n.Math.w_i.Math.{\hat{x}}_i\right)}},$

[0716] with the feature {circumflex over (x)}.sub.i being

[00005]${\hat{x}}_i=\frac{x_i-m_i}{s_i}$

[0717] wherein x.sub.i are the log-transformed peak area ratios (or concentrations, e.g. of NT-proBNP, if taken into account) and m.sub.i, s.sub.i are feature specific scaling factors and w.sub.i are the coefficients of the model [w.sub.0, intercept; w.sub.1, coefficient for NT-proBNP (where applicable); w.sub.2 . . . w.sub.n, coefficients for the features; n, number of lipid metabolite biomarkers+NT-proBNP, if determined or taken into account].

[0718] For example the coefficients for Panel 1+NT-proBNP are: WO=0.7602899 (Intercept), w.sub.1=1.5745605 (NT-proBNP), w.sub.2=0.7158906 (SM23), w.sub.3=0.6762354 (OSS2), w.sub.4=0.4594183 (PC4), n=4. The respective scaling factors are m1=2.2023800, m2=0.5379916, m3=2.1316954, m4=0.8736775 and s1=0.6364517, s2=0.1159420, s3=0.3863649, s4=0.0834807.

[0719] A sample with a prediction probability larger than the cutoff of the respective panel is said to be tested positive, and a sample with a prediction probability smaller than or equal to the cutoff is said to be tested negative. The sensitivity is the fraction of HF patients in the respective subgroup that are tested positive, while the specificity is the fraction of healthy control subjects that are tested negative.

[0720] The cutoff values can e.g. be determined to maximize the Youden index for the detection of the respective disease state.

[0721] Alternatively, the cutoff values can be determined to achieve e.g. the same specificity for the detection of the respective disease state in the ID dataset when using theat least three lipid metabolite biomarker panel as when using e.g. NT-proBNP alone (fixed sensitivity).

Example 6: Validation of Performance in Testing Test

[0722] The performance of a at least three lipid metabolite biomarker panel identified in the training data when applied to the validation data was assessed by the Area Under the Curve (AUC) of a receiver operating curve, which was modelled with the binormal model.

Example 7: Determination of Biomarkers for the at Least One Additional Cardiac Biomarker

[0723] In addition to at least three lipid metabolite biomarkers e.g. as described above, at least one additional cardiac biomarker described in the following is determined from a biological sample, e.g. blood plasma or blood serum sample.

[0724] (1) General Lipid Cardiac Biomarkers: Total Cholesterol, HDL Cholesterol, Triglycerides, LDLCholesterol, the Ratio of Total Cholesterol to HDL-Cholesterol, Non-HDL Cholesterol.

[0725] Total Cholesterol is determined by using commercially available tests, e.g. by a enzymatically based test using in addition a colorimetric method standardized to NIST (National Institute of Standards and Technology).

[0726] HDL cholesterol is measured by using commercially available tests, e.g. by separating from other lipoprotein fractions using e.g. either ultracentrifugation or chemical precipitation of other lipoprotein fractions with divalent ions such as Mg.sup.2+, within the resulting HDL fraction the lipoprotein-cholesterol/cholesterylester complex will be destroyed to allow the release of cholesterol and cholesterylesters. The cholesterylesters will be cleaved enzymatically with cholesteryl esterase to yield cholesterol which will be oxidised together with the already present cholesterol via cholesteroloxidase and the hereby produced H.sub.2O.sub.2 will be determined in an indicator reaction. HDL cholesterol is alternatively measured by an automated homogeneous analytical methods in which lipoproteins containing apo B are blocked using antibodies to apo B, then a colorimetric enzyme reaction measures cholesterol in the non-blocked HDL particles. HDL cholesterol is alternatively measured by HPLC or a turbidimetric method.

[0727] Triglycerides are measured by commercially availabale tests, e.g. determined enzymatically by first contacting the sample with lipase under conditions and for a time sufficient to allow conversion into glycerol and free fatty acids; then the sample comprising the glycerol is contacted with glycerokinase under conditions and for a time sufficient to allow conversion into glycerol-3-phosphate. Subsequently the sample comprising the glycerol-3-phosphate is contacted with glycerophosphate oxidase under conditions and for a time sufficient to allow conversion into dihydroxyacetone phosphate and H.sub.2O.sub.2; finally the amount of H.sub.2O.sub.2 is enzymatically or chemically determined.

[0728] LDL-cholesterol is determined by using commercially available tests or e.g. determined by using the Friedewald equation (Warnick G R, Knopp R H, Fitzpatrick V, Branson L (January 1990). Estimating low-density lipoprotein cholesterol by the Friedewald equation is adequate for classifying patients on the basis of nationally recommended cutpoints. Clinical Chemistry 36 (1): 15-9.PMID 2297909), by which the amount of LDL cholesterol is determined by subtraction of other cholesterol sources from total cholesterol. Friedewald equation: LDL cholesterol is approximately equal total cholesterol minus HDL-cholesterol concentrations minus triacylglycerols. The concentration of triacylglycerols is multiplied by a factor which is 0.20 if the quantities are measured in mg/dl and 0.45 if in mmol/l.

[0729] Non-HDL cholesterol is determined by subtracting HDL cholesterol from total cholesterol.

[0730] The ratio of total cholesterol to HDL-cholesterol is determined by determining total cholesterol as described above, by determining HDL-cholesterol as described above and by determining the ratio thereof.

[0731] (2) Lipoprotein subfraction biomarkers: LDL particles (herein also referred to as total LDL particles), small LDL particles, medium LDL particles and large HDL particlesQuantification and particle count of lipoprotein subfractions The lipoprotein subfraction biomarkers can be determined by ion mobility analysis. How to carry out such a ion mobility analysis is e.g. described in the publication Caulfield et al. (Clinical Chemistry August 2008 vol. 54 no. 8 1307-1316) which herewith is incorporated by reference with respect to the entire disclosure content. Ion mobility analysis preferably uses gas-phase electrophoresis to separate lipoproteins on the basis of size. This test is also commercially available from Quest Diagnostis Incorporated (Ion Mobility 91604(X).

[0732] (3) Apolipoprotein Biomarkers: Apolipoprotein B (ApoB), Lipoprotein (a)

[0733] ApoB is determined by an commercially available test being based on an immunoturbidimetric procedure that measures incre asing sample turbidity caused by the formation of insoluble immune complexes when antibody to ApoB is added to the sample. A sample containing ApoB is incubated with a buffer, and a sample blank determination is performed prior to the addition of ApoB antibody. In the presence of an appropriate antibody in excess, the ApoB concentration is measured as a function of turbidity

[0734] Lipoprotein (a) is determined by an commercially available test being based on a latex enhanced immunoturbidimetric method (Diazyme). Lp(a) in the sample binds to the specific antiLp(a) antibody, which is coated on latex particles, and causes agglutination. The degree of the turbidity caused by agglutination can be measured optically and is proportional to the amount of Lp(a) in the sample.

[0735] (4) Inflammation Biomarkers: HS CRP, Lp-PLA2 (PLAC):

[0736] hsCRPis determined by an commercially available test, with e.g. the Diazyme High Sensitive CReactive Protein (hsCRP) assay, which is based on a latex-enhanced turbidimetric immunoassay method. When an antigen-antibody reaction occurs between hs CRP in a sample and antiCRP antibody which is coated on latex particles, agglutination results. This agglutination is detected as an absorbance change (570 nm), with the magnitude of the change being proportional to the quantity of hs CRP in the sample. The actual concentration is then determined by interpolation from a calibration curve prepared from calibrators of known concentration.

[0737] Lp-PLA2 (PLAC) is determined by an commercially available test being a turbidimetric immunoassay for the quantitative determination of Lp-PLA2 (lipoprotein-associated phospholipase A2) in human plasma.

[0738] For each cardiac biomarker listed in Table 3, the direction of change in patients with cardiovascular disease relative to healthy control is listed (Table 3). This can also be calculated by ANOVA.

TABLE-US-00006 TABLE 3 Directions of change of the cardiac biomarkers indicative for cardiovascular diseases or heart failure Cardiac biomarker Direction vs. controls Total cholesterol up HDL cholesterol down Triacylglycerols up LDL-cholesterol up non-HDL cholesterol up ratio of total cholesterol to HDL-cholesterol up LDL particle number up small LDL particles up medium LDL particles up large HDL particles down ApoB up Lp(a) up hsCRP up Lp-PLA2 (PLAC) up

[0739] One or more of the above cardiac biomarkers are combined with any of the Panels 1 to 206 above (including or excluding NTP-proBNP). Combination means that values for the additional cardiac biomarker and the respective lipid metabolite biomarkers and, if taken into account, NT-proBNP are used as features of the classification model (e.g. Elastic Net). If several cardiac biomarkers are used, each biomarker is added as additional feature to the classification model. At least one of Panels 1 to 206 and one or more additional cardiac biomarkers and, optionally, NT-proBNP are used. The model is designed for an improved diagnosis of CHF including its subforms (e.g. HFrEF, HFpEF) and/or any cardiovascular disease including peripheral artery disease, coronary artery disease, atherosclerosis, cardiomyopathy, pulmonary heart disease, and other cardiac diseases and the model is fit in analogy as described above (see e.g. Example 5).

Example 8: Performance of Panel 1 Combined with NT-proBNP, HDL, and LDL

[0740] The concentrations of each of HDL cholesterol (HDLchol) and LDL cholesterol (LDLchol) were determined by respective commercially availiable assays in the respective samples referred to above. HDL cholesterol concentrations were determined turbidimetric using the Direct HDL Cholesterin fr Advia Kit from Siemens Healthcare GmbH on an Advia automated system. The Adiva automated system fom Siemens Healthcare GmbH also measured the parameters additionally necessary for the calculation of LDL cholesterol via the Friedewald equation using the respective kits from Siemens for the Advia automated system. These parameters are total cholesterol and triglycerides. Subsequently LDL cholesterol concentrations were calculated from the Friedewald equation.

[0741] The lipid metabolite biomarkers of the at least three lipid metabolite biomarkers of Panel 1+NT-pro-BNP were used as one classification model, that was then extended by the values for HDL cholesterol as well as LDL cholesterol into a second classification model in order to compare the diagnostic performance with and without consideration of HDLchol and LDLchol. These two classification models were fit on the ID Data in the same way as described in Example 5, and performance estimates (Area under the curve (AUC) values of receiver operating characteristic (ROC) analysis) were calculated on the data set referred to as VD data above for different CHF subgroups (see Table 4).

[0742] Several subgroups benefitted from the use of Panel 1 and NT-proBNP and both HDL cholesterol and LDL cholesterol (Abbreviated as 1+NT+HDLchol+LDLchol) compared to Panel 1 and NT-proBNP alone (Abbreviated as 1+NT). The respective performance values are shown in Table 4.

[0743] Since HDLchol and/or LDLchol values were missing for some subjects, only patients with complete values for both HDLchol and LDLchol were used to fit and test the model.

[0744] For comparability, the subjects lacking data for HDLchol and/or LDLchol were also excluded for calculation of the 1+NT data, so that the same set of subjects was used for all calculations in Table 4. The cut-off was determined on the ID Data with a fixed sensitivity (i.e. so that the marker panel had the same sensitivity as NT-proBNP for the comparison of HFrEF vs controls). The resulting cut-off was 0.74997 when using HDLchol and LDLchol and 0.72131 when not.

TABLE-US-00007 TABLE 4 Performance of Panel 1 + NT-proBNP (abbreviated 1 + NT) has been compared to the performance of Panel 1 + NT-proBNP + HDL cholesterol (abbreviated HDLchol) + LDL cholesterol (abbreviated LDLchol. Area under the curve (AUC) values of receiver operating characteristic (ROC) analysis were calculated for different CHF subgroups. Calculations were carried out on the dataset referred to as VD data above, with the exception of subjects for which HDLchol or LDLchol data was missing (such subjects were excluded). UsedPanel Used Used NT-proBNP AUC Panel Panel AUC NT-proBNP NT-proBNP Subgroup estimate Sensitivity Specificity Estimate Sensitivity Specificity Used Panel CHF ASYMP 0.8713 0.5000 0.9655 0.8078 0.5698 0.8736 1 + NT CHF ASYMP 0.8743 0.5000 0.9885 0.8078 0.5698 0.8736 1 + NT + HDLchol + LDLchol DCM ASYMP 0.8790 0.5000 0.9655 0.8599 0.6667 0.8736 1 + NT DCM ASYMP 0.8799 0.5000 0.9885 0.8599 0.6667 0.8736 1 + NT + HDLchol + LDLchol HFpEF 0.8091 0.2462 0.9655 0.7059 0.3538 0.8736 1 + NT HFpEF 0.8132 0.2308 0.9885 0.7059 0.3538 0.8736 1 + NT + HDLchol + LDLchol HFpEF ASYMP 0.7850 0.2105 0.9655 0.6569 0.2632 0.8736 1 + NT HFpEF ASYMP 0.7896 0.2105 0.9885 0.6569 0.2632 0.8736 1 + NT + HDLchol + LDLchol HFpEF SYMP 0.8401 0.2963 0.9655 0.7675 0.4815 0.8736 1 + NT HFpEF SYMP 0.8435 0.2593 0.9885 0.7675 0.4815 0.8736 1 + NT + HDLchol + LDLchol HFrEF ASYMP 0.9471 0.7292 0.9655 0.9191 0.8125 0.8736 1 + NT HFrEF ASYMP 0.9474 0.7292 0.9885 0.9191 0.8125 0.8736 1 + NT + HDLchol + LDLchol

Example 9: Performance of Panel 1 Combined with NT-proBNP and HDL

[0745] The very same approach as described in Example 8 was applied with a classification model consisting of the at least three lipid metabolite biomarkers of Panel 1+ and NT-pro-BNP and HDL cholesterol (Panel: 1+NT+HDLchol). The resulting cut-off for this panel (determined on the ID Data with a fixed sensitivity as described in Example 8, above) was 0.748164136. A comparison of all three panels, 1+NT, 1+NT+HDLchol+LDLchol, and 1+NT+HDLchol, is shown in Table 5, below.

TABLE-US-00008 TABLE 5 Performance of Panel 1 + NT-proBNP (abbreviated 1 + NT) has been compared to the performance of Panel 1 + NT-proBNP + HDL cholesterol (abbreviated HDLchol) + LDL cholesterol (abbreviated LDLchol), and to Panel 1 + NT-proBNP + HDL cholesterol (abbreviated HDLchol). Area under the curve (AUC) values of receiver operating characteristic (ROC) analysis were calculated for different CHF subgroups. Calculations were carried out on the dataset referred to as VD data above, with the exception of subjects for which HDLchol or LDLchol data was missing (such subjects were excluded). Used Panel Used Used NT-proBNP AUC Panel Panel AUC NT-proBNP NT-proBNP Subgroup Estimate Sensitivity Specificity Estimate Sensitivity Specificity Used Panel CHF ASYMP 0.8713 0.5000 0.9655 0.8078 0.5698 0.8736 1 + NT CHF ASYMP 0.8743 0.5000 0.9885 0.8078 0.5698 0.8736 1 + NT + HDLchol + LDLchol CHF ASYMP 0.8745 0.5000 0.9885 0.8078 0.5698 0.8736 1 + NT + HDLchol DCM ASYMP 0.8790 0.5000 0.9655 0.8599 0.6667 0.8736 1 + NT DCM ASYMP 0.8799 0.5000 0.9885 0.8599 0.6667 0.8736 1 + NT + HDLchol + LDLchol DCM ASYMP 0.8801 0.5000 0.9885 0.8599 0.6667 0.8736 1 + NT + HDLchol HFpEF 0.8091 0.2462 0.9655 0.7059 0.3538 0.8736 1 + NT HFpEF 0.8132 0.2308 0.9885 0.7059 0.3538 0.8736 1 + NT + HDLchol + LDLchol HFpEF 0.8159 0.2154 0.9885 0.7059 0.3538 0.8736 1 + NT + HDLchol HFpEF ASYMP 0.7850 0.2105 0.9655 0.6569 0.2632 0.8736 1 + NT HFpEF ASYMP 0.7896 0.2105 0.9885 0.6569 0.2632 0.8736 1 + NT + HDLchol + LDLchol HFpEF ASYMP 0.7938 0.2105 0.9885 0.6569 0.2632 0.8736 1 + NT + HDLchol HFpEF SYMP 0.8401 0.2963 0.9655 0.7675 0.4815 0.8736 1 + NT HFpEF SYMP 0.8435 0.2593 0.9885 0.7675 0.4815 0.8736 1 + NT + HDLchol + LDLchol HFpEF SYMP 0.8442 0.2222 0.9885 0.7675 0.4815 0.8736 1 + NT + HDLchol HFrEF ASYMP 0.9471 0.7292 0.9655 0.9191 0.8125 0.8736 1 + NT HFrEF ASYMP 0.9474 0.7292 0.9885 0.9191 0.8125 0.8736 1 + NT + HDLchol + LDLchol HFrEF ASYMP 0.9467 0.7292 0.9885 0.9191 0.8125 0.8736 1 + NT + HDLchol