A METHOD FOR DISTINGUISHING HEALTHY INDIVIDUALS FROM INDIVIDUALS HAVING INFECTIOUS OR INFLAMMATORY CONDITIONS
20230258660 · 2023-08-17
Assignee
Inventors
Cpc classification
G01N33/92
PHYSICS
International classification
Abstract
A method of determining a health status of an individual, the method comprising the steps of assaying a biological sample from the individual for the expression of at least genes, or proteins encoded by said genes, wherein the at least three genes are selected from SAA, CRP, Hp and MxA, wherein the expression of the at least three genes, or the proteins encoded by said genes, above or below a specified threshold level correlates with the health status of the individual.
Claims
1. A method of (a) determining a health status of an individual, (b) of monitoring the effectiveness of treatment of an infectious disease in an individual with an infectious disease, or (c) of identifying an individual with an infection or an infectious disease that is suitable for treatment with a therapy for treating the infectious disease, the methods comprising the steps of assaying a biological sample from the individual for the positive expression of at least three genes, or proteins encoded by said genes, selected from SAA, CRP, Hp and MxA, and wherein the positive expression of the at least three genes, or proteins encoded by said genes, above or below a specified threshold level (a) correlates with the health status of the individual, (b) correlates with the effectiveness of treatment of the infectious disease in the individual, and (c) correlates with identifying the treatment of the infection or the infectious disease in the individual.
2. (canceled)
3. (canceled)
4. A method according to claim 1, wherein the at least three genes, or proteins encoded by said genes, are SAA and two selected from CRP, Hp and MxA.
5. A method according to claim 1, wherein all four genes, or proteins encoded by said genes are selected.
6. A method according to claim 1, wherein the at least three genes, or proteins encoded by said genes, are SAA and two selected from CRP, Hp, and MxA and wherein when the protein expression levels of SAA are less than 6 ug/ml, the protein expression levels of Hp are less than 114 mg/dl, the protein expression levels of CRP are less than 10 ug/ml, and the protein expression levels of MxA are less than 0.174 ug/ml, the individual is healthy.
7. A method according to claim 1, wherein the at least three genes, or proteins encoded by said genes, are SAA and two selected from CRP, Hp, and MxA and wherein when the protein expression levels of SAA are less than 20 ug/ml, when the protein expression levels of Hp are greater than 114 mg/dl, when the protein expression levels of CRP are greater than 5 ug/ml, and when the protein expression levels of MxA are greater than 0.174 ug/ml, the individual is experiencing an inflammatory event or is suffering from an inflammatory condition.
8. A method according to claim 1, wherein the at least three genes, or proteins encoded by said genes, are SAA and two selected from CRP, Hp, and MxA and wherein when the protein expression levels of SAA are greater than 20 ug/ml, when the protein expression levels of Hp are greater than 114 mg/dl, when the protein expression levels of CRP are greater than 10 ug/ml, and when the protein expression levels of MxA are greater than 0.174 ug/ml, the individual is experiencing an infection.
9. A method according to claim 1, wherein the at least three genes, or proteins encoded by said genes, are SAA and two selected from CRP, Hp, and MxA and wherein when the protein expression levels of SAA are about 18 ug/ml to about 21 ug/ml, when the protein expression levels of Hp are about 50 mg/dl to 220 mg/dl, when the protein expression levels of CRP are about 3 ug/ml to 29 ug/ml, and when the protein expression levels of MxA are greater than 0.174 ug/ml, the individual is at risk of onset of an autoimmune and/or an autoinflammatory condition.
10. A method according to claim 1, wherein the at least three genes, or proteins encoded by said genes, are SAA and two selected from CRP, Hp, and MxA and wherein when the protein expression levels of SAA are less than 20 ug/ml, when the protein expression levels of Hp are less than 160 mg/dl, when the protein expression levels of CRP are less than 5 ug/ml, and when the protein expression levels of MxA are less than 0.174 ug/ml, the individual is responding to treatment; and wherein when the protein expression levels of SAA are greater than 20 ug/ml, when the protein expression levels of Hp are greater than 160 mg/dl, when the protein expression levels of CRP are greater than 5 ug/ml, and when the protein expression levels of MxA are greater than 0.174 ug/ml, the individual still has the infection and is not responding to treatment.
11. A method of determining the progression of treatment of an individual with an inflammatory disease, the method comprising the step of assaying a biological sample from the individual with the inflammatory disease for the positive expression of at least two genes, or proteins encoded by said genes, selected from CRP, Hp and MxA, and negative expression of SAA, wherein the positive expression of the at least two genes, or proteins encoded by said genes, above or below a specified threshold level, and the negative expression of SAA, correlates with the effectiveness of treatment of the inflammatory disease or disorder in the individual.
12. A method according to claim 11, wherein the threshold levels for Hp protein expression levels are optionally 114 mg/dl, the threshold levels for CRP protein expression levels are 5 ug/ml, and the threshold levels for MxA are 0.174 ug/ml.
13. A method according to claim 11, wherein when the protein expression levels of Hp are less than 114 mg/dl and the protein expression levels of CRP are less than 5 ug/ml, and when the protein expression levels of MxA are less than 0.174 ug/ml, the individual is responding to treatment; and wherein when the protein expression levels of Hp are greater than 114 mg/dl, the protein expression levels of CRP are greater than 5 ug/ml, and when the protein expression levels of MxA are greater than 0.174 ug/ml, the individual is experiencing an inflammatory response and is not responding to treatment.
14. (canceled)
15. (canceled)
16. (canceled)
17. (canceled)
18. (canceled)
19. A method of treating an infection, an infectious disease, an inflammatory condition, or an autoinflammatory condition in a subject, the method comprising: receiving the results of an assay that indicates that the level of at least three genes, or proteins encoded by said genes, selected from SAA, CRP, Hp and MxA, in a sample obtained from the subject is above or below a threshold level; and administering a treatment for the disease or condition, wherein the at least three genes, or proteins encoded by said genes, are SAA, CRP and Hp.
20. The method of claim 19, wherein the treatment comprises administering a suitable antibiotic to the subject determined to have an infection; administering an anti-inflammatory drug if the subject is determined to have an inflammatory condition; or administering a suitable treatment as set out in Table 1.
21. (canceled)
22. (canceled)
23. The method of claim 1 wherein the disease or condition is selected from those set out in Table 1.
24. A method for differentiating whether an individual is healthy, has an inflammatory condition or has a viral or bacterial infection, the method comprising the steps of assaying a biological sample from the individual for the positive expression of at least three genes, or proteins encoded by said genes, selected from SAA, CRP, Hp and MxA, and wherein the positive expression of the at least three genes, or proteins encoded by said genes, above or below a specified threshold level correlates with the health status of the individual, wherein if the concentration of SAA is >20 ugf/ml, and the concentration of CRP is >10 ug/ml, the individual is determined to have an infection; wherein if the concentration of CRP>97 ug/ml or the concentration of Hp>182 mg/dl and the concentration of MxA>0.174 ug/ml, the individual has a bacterial infection or a viral infection; and wherein if the concentration of SAA<6 ug/ml or the concentration of Hp<114 mg/dl, the individual is determined to be healthy and wherein if the conditions of the above are not met the individual is determined to have inflammation.
25. (canceled)
26. A kit for determining the health status of a subject, the kit comprising a control oligonucleotide and a set of oligonucleotides for detecting SEQ ID NO: 2, 3, 5, 6, 8, 9, 11 and 12, wherein detection of at least three of SEQ ID NO: 2, 3, 5, 6, 8, 9, 11 or 12 above a threshold in the sample is indicative of the health status of the subject, and wherein the sensitivity of the assay for detecting the at least three sequences from SEQ ID NO: 2, 3, 5, 6, 8, 9, 11 or 12 is at least 80%, wherein the at least three oligonucleotides are selected from one of SEQ ID NO: 2 or 3, one of SEQ ID NO: 5 or 6, and one of SEQ ID NO: 8 or 9.
27. A kit according to claim 25, wherein the support further comprises having at least four or more oligonucleotides anchored thereon selected from one or both of SEQ ID NO: 2 or 3, one or both of SEQ ID NO: 5 or 6, and one or both of SEQ ID NO: 8 or 9.
28. A kit according to claim 25, wherein the support further comprises at least one or both oligonucleotides anchored thereon selected from SEQ ID NO: 11 and 12.
29. A method according to claim 1, wherein when a negative expression of the at least three genes encoding SAA, CRP, and Hp, or proteins encoded by said genes, are below the threshold level, the result indicates a healthy status in the subject, and the treatment, if being administered, can be stopped.
30. A method according to claim 11, wherein when a negative expression of the at least three genes encoding SAA, CRP, and Hp, or proteins encoded by said genes, are below the threshold level, the result indicates a healthy status in the subject, and the treatment, if being administered, can be stopped.
31. A method according to claim 19, wherein when a negative expression of the at least three genes encoding SAA, CRP, and Hp, or proteins encoded by said genes, are below the threshold level, the result indicates a healthy status in the subject, and the treatment, if being administered, can be stopped.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
[0160] The invention will be more clearly understood from the following description of an embodiment thereof, given by way of example only, with reference to the accompanying drawings, in which:—
[0161]
[0162]
[0163]
[0164]
[0165]
[0166]
[0167]
[0168]
[0169]
[0170]
[0171]
[0172]
[0173]
[0174]
[0175]
[0176]
[0177]
DETAILED DESCRIPTION OF THE DRAWINGS
Materials and Methods
[0178] In this study SAA, CRP and Hp levels were retrospectively assessed in repository equine serum samples (n=162) with clinical records available. Also in this study SAA, CRP, Hp and MxA levels were retrospectively assessed in repository human serum samples (n=60) with clinical records available. Samples were included in the study if they had a healthy, infectious or inflammatory clinical record at the time of collection and divided into five groups: Healthy, Inflammation, Infection, Non-Healthy (inflammation and Infection combined) and Non-Infection (Healthy and Inflammation combined). The Kruskal Wallace test was engaged on all samples in the Healthy, Inflammatory and Infectious groups to determine differences in levels across the groups for each biomarker, where P<0.05 was considered statistically significant. Receiver operating curves were used on a subset of data across groups where data for all three biomarkers was available for each sample to identify the diagnostic ability of the biomarkers to differentiate between healthy and disease states. Diagnostic ranges were determined by Youden index (a single statistic that captures the performance of a dichotomous diagnostic test) and in Microsoft Excel using diagnostic value statistics, sensitivity, specificity, positive predicative values and negative predictive values. The biomarkers were also assessed in cases where serial sampling across a time course was available to compare performance across time.
[0179] Protein levels were determined using immunoassays. CRP and Hp levels were measured using species appropriate enzyme linked immunosorbent assays (ELISA). SAA concentrations were determined using a commercial lateral flow test. MxA levels were measured using species appropriate enzyme linked immunosorbent assays (ELISA). Statistical manipulations were performed using Microsoft Excel and MedCalc Software. Other methods can be used to detect nucleic acids, such as PCR, quantitative (or real time) PCR (qPCR), reverse transcriptase (RT) PCR (RT-PCR), RT-qPCR, Photometry, Fluorescence, lateral flow methods, and those methods described above. Such techniques can give quantitative outputs such as molar, ng/ul, ng/ml, copies/ul, and optical density (OD) readings.
C Reactive Protein (CRP)
[0180] Serum CRP protein concentrations were determined using a CRP Horse ELISA kit and performed using antibodies and standards obtained from Abcam, Cambridge, UK. All reagents were equilibrated to room temperature. Serum samples were diluted before use with 1× Diluent (supplied as 5× diluent concentrate and diluted 1/5 using deionized water). 100 μL of each standard, including zero control, and each sample was pipetted, in duplicate or dilution, into predesignated wells on a 96 well micro titer plate and covered with an adhesive plate cover. The plate was left to incubate at room temperature for thirty (30±2) minutes. The plate was then aspirated and washed four times with wash buffer (supplied as 20×concentrate and diluted 1/20 with deionized water). Following this, 100 μL CRP Horse HRP conjugate (10 μL enzyme-antibody conjugate to 990 μL of 1× diluent per test strip) was added to each well and the plate covered with an adhesive plate cover. The plate was left to incubate at room temperature for thirty (30±2) minutes in the dark. The plate was then aspirated and washed four times with wash buffer. Following this, 100 μL of TMB Substrate was added to each well, the plate covered with an adhesive plate cover and incubated in the dark at room temperature for precisely ten minutes. After ten minutes, 100 μL of Stop Solution was added to each well. Absorbance was then read at 450 nm on a microplate reader (Benchmark microplate reader, Bio-Rad, US). CRP levels were extrapolated from standard curve of CRP level (ng/mL) versus optical density at 450 nm (Excel).
C Reactive Protein (CRP) (Human Sample Analysis)
[0181] Serum CRP protein concentrations were determined using a CRP human turbi latex immunoturbidimetric assay and performed using reagents and calibrator from Spectrum Diagnostics, Obour City, Egypt. The reagents and spectrophotometer were brought to 37° C. and the reagents were shaken gently before mixing. The working reagent was prepared with 1 part of Latex Reagent and 4 parts of Buffer Reagent. The calibrator was reconstituted with distilled water, mixed gently and incubated at room temperature for 10 minutes before use. The spectrophotometer was adjusted to zero with distilled water. Following this, 500 μl working reagent and 5 μl calibrator or sample were pipetted into a cuvette and mixed. Absorbance was read at 540 nm immediately (A1) and after 2 minutes (A2) of the sample addition. The concentration of CRP was calculated using the following calculation:
(A2−A1)sample×calibrator concentration=mg/L CRP (A2−A1)calibrator
Haptoglobin (Hp)
[0182] Serum samples were diluted before use with 1× Diluent (supplied as 5× diluent concentrate and diluted 1/5 using deionized water). 100 μL of each standard, including zero control, and each sample was pipetted, in duplicate or dilution, into pre-designated wells on a 96 well micro titer plate and covered with an adhesive plate cover. The plate was left to incubate at room temperature for thirty (30±2) minutes. The plate was then aspirated and washed four times with wash buffer (supplied as 20× concentrate and diluted 1/20 with deionized water). Following this, 100 μL Haptoglobin Horse HRP conjugate (10 μL enzyme-antibody conjugate to 990 μL of 1× diluent per test strip) was added to each well and the plate covered with an adhesive plate cover. The plate was left to incubate in the dark at room temperature for thirty (30±2) minutes. The plate was then aspirated and washed four times with wash buffer. Following this, 1004 of TMB Substrate was added to each well, the plate covered with an adhesive plate cover and incubated in the dark at room temperature for precisely ten minutes. After ten minutes, 100 μL of Stop Solution was added to each well. Absorbance was then read at 450 nm on a microplate reader (Benchmark microplate reader, Bio-Rad, US). Haptoglobin levels were extrapolated from standard curve of Haptoglobin level (ng/mL) versus optical density at 450 nm (Excel).
Haptoglobin (Hp) (Human Sample Analysis)
[0183] Serum Hp protein concentrations were determined using a Hp human immunoturbidimetric assay and performed using reagents and calibrator from Kamiya Biomedical Company, Seattle, Wash., USA. All reagents and specimens were allowed to come to room temperature and mixed gently before using. The spectrophotometer was adjusted to zero with distilled water prior to sample testing. In a cuvette, 6 μl calibrator/control/sample was added to 500 μl buffer reagent and incubated at 37° C. for 5 minutes. A reading was taken at 600 nm, after which 140 μl antiserum reagent was added to the cuvette, mixed with the sample and buffer reagent and incubated at 37° C. for 5 minutes. A second reading was then taken at 600 nm. The difference between reading 1 and reading 2 was calculated and the concentration of Hp was extrapolated from standard curve of Hp level (mg/dL) versus absorbance at 600 nm (Excel).
Serum Amyloid A (SAA)
[0184] Serum SAA samples were analysed utilising a commercial lateral flow immunoassay test, Equine SAA, Stablelab. Assay comprised of immunoassay strip housed in a cartridge, mix solution for dilution and propriety EQ1 reader device for SAA quantitation. Samples were allowed to come to room temperature and allowed to mix before use. Samples required a dilution whereby 5 μL of sample (serum/plasma) was pipetted into a mix solution. 4 drops of sample (approx. 100 μL) was then added to the sample well of the lateral flow cartridge. A timer was then set for a total run time of ten minutes, following which the cartridge was inserted into associated reader device for SAA quantitation. Readout were shown in μg/ml.
Serum Amyloid a (SAA) (Human Sample Analysis)
[0185] Serum SAA samples were analysed utilising a lateral flow immunoassay test. The assay comprised of an immunoassay strip housed in a cartridge, mix solution for dilution and \handheld reader device for SAA quantitation. Samples were allowed to come to room temperature and mixed before use. Samples required a dilution whereby 5 μL of sample (serum/plasma) was pipetted into a mix solution. 4 drops of sample (approx. 100 μL) was then added to the sample well of the lateral flow cartridge. A timer was then set for a total run time of ten minutes, following which the cartridge was inserted into associated reader device for SAA quantitation. Readouts were shown in ug/ml.
Myxovirus Resistance Protein A (MxA)
[0186] Serum samples were diluted before use with 1× Diluent. 100 μL of each standard, including zero control, and each sample was pipetted, in duplicate or dilution, into pre-designated wells on a 96 well micro titre plate. The plate was left to incubate at room temperature for 1 hour, shaking at ca. 300 rpm on an orbital microplate shaker. The plate was then aspirated and washed three times with wash buffer (supplied as 10× concentrate and diluted 1/10 with deionized water). Following this, 100 μL Biotin Labelled Anti-MxA Antibody Solution was added to each well and left to incubate at room temperature for 1 hour, shaking at ca. 300 rpm on an orbital microplate shaker. The plate was then aspirated and washed three times with wash buffer. Following this, 100 μL of Streptavidin-HRP Conjugate was added to each well and the plate incubated at room temperature for 30 minutes, shaking at ca. 300 rpm on an orbital microplate shaker. The plate was then washed five times with wash buffer. 100 μL of Substrate Solution was added to each well, the plate covered with tin foil and incubated in the dark for 20 minutes at room temperature (no shaking carried out during this incubation). 100 μL of Stop Solution was then added to each well to stop colour development. Absorbance was then read at 450 nm on a microplate reader (Benchmark microplate reader, Bio-Rad, US). MxA levels were extrapolated from standard curve of the log of MxA level (ng/mL) versus optical density at 450 nm (Excel).
Myxovirus Resistance Protein A (MxA) (Human Sample Analysis) 100 μL of each standard, including zero control, was pipetted in duplicate into pre-designated wells on a 96 well micro titre plate. Samples were diluted before use with sample diluent. One part sample was added to 4 parts sample diluent. 100 μL HRP-conjugate reagent was added to each well, the plate covered with an adhesive strip and incubated for 60 minutes. Each well was then aspirated and washed for a total of five washes. To Wash, each well was filled with wash solution (400 μL) and then all liquid removed. After the last wash, the plate was inverted and blotted against clean paper towels to ensure complete liquid removal. 50 μL chromogen solution A and 50 μL chromogen solution B was added to each well, gently mixed and incubated for 15 minutes at 37° C. 50 μl stop solution was then added to each well and the optical density read at 450 nm using a microplate reader (Benchmark microplate reader, Bio-Rad, US). MxA levels were extrapolated from standard curve of MxA level (ng/mL) versus optical density at 450 nm (Excel).
Results
[0187] In single time point analysis levels of SAA (n=99) and CRP (n=39) there was a significant difference in biomarker values between the Healthy, Inflammation and Infection groups, p<0.000001 and p=0.010325, respectively (see
[0188] In a time course assessment for patients with a confirmed infection (n=3) (see
[0189] ROC analysis was undertaken on a subset of data for which SAA, CRP and Hp (n=24) levels were available to identify the diagnostic ability of the biomarkers to differentiate between healthy and disease states.
Distinguishing Healthy and Non-Healthy (Infection and Inflammation Combined)
[0190] The most significant area under the curve (AUC) for the Healthy and Non-Healthy Group was observed for Hp (see
TABLE-US-00004 TABLE 3 showing diagnostic performance statistics for a range of Hp cut-off values, including Youden Index associated criterion of greater than 473 ug/ml, to distinguish between healthy and non-healthy patients (infection and inflammation combined). Cut-off (ug/ml) -> 400 473 500 600 700 Sensitivity 100.0 100.0 87.5 81.3 75.0 Specificity 50.0 87.5 87.5 87.5 87.5 Pos. Predictive Value 80.0 94.1 93.3 92.9 92.3 Neg. Predictive Value 100.0 100.0 77.8 70.0 63.6 Accuracy 83.3 95.8 87.5 83.3 79.2
Distinguishing Healthy and Inflammation
[0191] The most significant AUC was observed for Hp at 0.900 and p=0.0001 with a cut off value of 473 ug/ml (
TABLE-US-00005 TABLE 4 showing diagnostic performance statistics for a range of Hp cut-off values, including Youden Index associated criterion of greater than 473 ug/ml, to distinguish between healthy and inflammation patients Cut-off (ug/ml) 473 600 796 1217 Sensitivity 100.0 70.0 20.0 20.0 Specificity 87.5 87.5 87.5 88.9 Pos. Predictive Value 90.9 87.5 66.7 66.7 Neg. Predictive Value 100.0 70.0 46.7 50.0 Accuracy 94.4 77.8 50.0 52.6
Distinguishing Infection and Non-Infection (Healthy and Inflammation)
[0192] The most significant AUC was observed for SAA when differentiating between the Infection and Non-Infection groups. AUC was 1.000 (P<0.0001) (
TABLE-US-00006 TABLE 5 showing diagnostic performance statistics for a range of SAA cut-off values, including Youden Index associated criterion of greater than 18.5 ug/ml, to distinguish between infection and non-infection patients (healthy & inflammation combined) Cut-off (ug/ml)-> 15 18.5 50 100 Sensitivity 100.0 100.0 100.0 100.0 Specificity 94.4 94.4 100.0 100.0 Pos. Predictive Value 85.7 85.7 100.0 100.0 Neg. Predictive Value 100.0 100.0 100.0 100.0 Accuracy 95.8 95.8 100.0 100.0
TABLE-US-00007 TABLE 6 showing diagnostic performance statistics for a range of Hp cut- off values, including Youden Index associated criterion of greater than 1217 ug/ml, to distinguish between infection and non-infection patients (healthy & inflammation combined) Cut-off (ug/ml).fwdarw. 400 473 500 600 700 1217 Sensitivity 100.0 100.0 100.0 100.0 100.0 100.0 Specificity 22.2 38.9 50.0 55.6 61.1 88.9 Pos. Predictive Value 30.0 35.3 40.0 42.9 46.2 75.0 Neg. Predictive Value 100.0 100.0 100.0 100.0 100.0 100.0 Accuracy 41.7 54.2 62.5 66.7 70.8 91.7
TABLE-US-00008 TABLE 7 showing diagnostic performance statistics for a range of CRP cut-off values, including Youden Index associated criterion of greater than 4.7 ug/ml, to distinguish between infection and non-infection patients (healthy and inflammation combined) Cut-off (ug/ml) 2 4.7 6 8 10 Sensitivity 100.0 100.0 80.0 60.0 40.0 Specificity 20.0 70.0 80.0 80.0 90.0 Pos. Predictive Value 38.5 62.5 66.7 60.0 66.7 Neg. Predictive Value 100.0 100.0 88.9 80.0 75.0 Accuracy 46.7 80.0 80.0 73.3 73.3
Infection and Inflammation
[0193] SAA AUC was 1.000, P<0.0001 and Youden Index criterion was >18.5 ug/ml (
TABLE-US-00009 TABLE 8 showing diagnostic performance statistics for a range of SAA cut-off values, including Youden Index associated criterion of greater than 18.5 ug/ml, to distinguish between infection and inflammation patients Cut-off (ug/ml) 15 18.5 50 100 Sensitivity 100.0 100.0 100.0 100.0 Specificity 94.4 94.4 100.0 100.0 Pos. Predictive Value 85.7 85.7 100.0 100.0 Neg. Predictive Value 100.0 100.0 100.0 100.0 Accuracy 95.8 95.8 100.0 100.0
TABLE-US-00010 TABLE 9 showing diagnostic performance statistics for a range of Hp cut-off values, including Youden Index associated criterion of greater than 796 ug/ml, to distinguish between infection and inflammation patients. Cut-off (ug/ml)-> 473 500 600 796 1217 Sensitivity 100.0 100.0 100.0 100.0 100.0 Specificity 38.9 50.0 55.6 80.0 80.0 Pos. Predictive Value 35.3 40.0 42.9 75.0 75.0 Neg. Predictive Value 100.0 100.0 100.0 100.0 100.0 Accuracy 54.2 62.5 66.7 87.5 87.5
TABLE-US-00011 TABLE 10 showing diagnostic performance statistics for a range of CRP cut- off values to distinguish between infection and inflammation patients. Cut-off (ug/ml) 2 4.7 6 8 10 Sensitivity 100.0 100.0 80.0 60.0 40.0 Specificity 20.0 70.0 80.0 80.0 90.0 Pos. Predictive Value 38.5 62.5 66.7 60.0 66.7 Neg. Predictive Value 100.0 100.0 88.9 80.0 75.0 Accuracy 46.7 80.0 80.0 73.3 73.3
Single and Combined Analysis Between Healthy, Infection and Inflammation Groups in Human Sample Sets
[0194] Threshold values were assessed for each of SAA, CRP, MxA and Hp (Table 11 below). The literature reports a wide range of normal values for Hp. The upper threshold in the literature was selected for inclusion at 220 ug/ml. MxA demonstrated the poor performance on ROC analysis for differentiating between the Healthy vs Non-Healthy, Infection vs Non-Infection and Healthy vs Inflammation and Infection vs Inflammation groups. MxA did demonstrate utility as a biomarker for differentiating between Viral and Bacterial groups (see
TABLE-US-00012 TABLE 11 Threshold values were assessed for each of SAA, CRP, MxA and Hp Cut-off values of ROC analysis SAA CRP Hp MxA Differentiation (ug/ml) (ug/ml) (mg/dL) (ng/ml) Healthy vs Non-Healthy 6 29.2 159.5 n/a Infection vs Non-Infection 10.8 50.5 193.2 n/a Healthy vs Inflammation 6 29.2 114.2 n/a Healthy vs Infection 6 29.2 159.5 n/a Infection vs Inflammation 10.8 50.5 190 n/a Bacterial vs Viral 6 97.2 182.1 174
Healthy and Inflammation
[0195] In an analysis to distinguish Healthy and Inflammation groups from each other (Table 11 and Table 12), where those with an infection were not included, CRP at a cut off value of 5 ug/ml performed with an accuracy of 65%. Analysis using a cut off value of 10 ug/ml from the literature yielded accuracy results at 52%. Sensitivity for identifying the individuals from the inflammation group (those categorised based on clinical notes which included an inflammatory condition with no infection) yielded significant false positive results. The interpretation of these results in a clinical setting may lead to a significant portion of patients being treated unnecessarily for inflammation, which may be associated with negative outcomes, e.g. in the case of steroidal treatments. This finding is surprising. The common thinking on CRP is that it is a good biomarker of inflammation. However, this finding also elucidates that CRP alone is not a good differentiator within Healthy and Inflammatory groups, and leads to false positive diagnosis for inflammation with as many as 7 of 10 healthy individuals.
[0196] For the same groups, SAA was assessed using cut off values of 10 ug/ml (as per the literature) and 20 ug/ml (as per the data reported here) (Table 12 and Table 13). SAA performed with an accuracy of 52% for differentiating between the two groups. A sensitivity of 15% was recorded when identifying inflammation within the two groups and a specificity of 100% meaning SAA identified all healthy individuals. High specificity came at the expense of sensitivity and accuracy with 11 of 13 cases of inflammation from the Inflammation group reported as negative. In a clinical setting, such results may lead to undertreatment of the individual through withholding of suitable treatment due to the false negative report. As with CRP, SAA is not an accurate differentiator when distinguishing between Healthy and Inflammation groups. This finding is surprising as SAA and CRP are commonly discussed as being biomarkers of inflammation.
[0197] Hp has high specificity but low sensitivity at 15% meaning it could not differentiate between Healthy Individuals with Inflammation accurately.
TABLE-US-00013 TABLE 12 Performance statistics for SAA and CRP at different cut-off values to distinguish between Healthy Individuals and Individuals with Inflammation. Description (ug/ml) sensitivity specificity PPV NPV accuracy Inflammation vs Healthy, 8 100 100 45 48 SAA >= 20 Inflammation vs Healthy, 15 100 100 48 52 SAA >= 10 Inflammation vs Healthy, 92 30 63 75 65 CRP >= 5 Inflammation vs Healthy, 69 30 56 43 52 CRP >= 10 Inflammation vs Healthy, 15 100 100 48 52 Hp >= 220
TABLE-US-00014 TABLE 13 Further performance statistics for SAA and CRP at different cut-off values to distinguish between Healthy Individuals and Individuals with Inflammation. false actual Description true false true neg- pos- actual (ug/ml) positive positive negative ative itive negative Inflammation vs 1 0 10 12 13 10 Healthy, SAA >= 20 Inflammation vs 2 0 10 11 13 10 Healthy, SAA >= 10 Inflammation vs 12 7 3 1 13 10 Healthy, CRP >= 5 Inflammation vs 9 7 3 4 13 10 Healthy, CRP >= 10 Inflammation vs 2 0 10 11 13 10 Healthy, Hp >= 220
Inflammation and Infection
[0198] In an analysis to distinguish between Inflammation and Infection groups wherein Healthy were omitted from the analysis (see Table 14 and Table 15), CRP at a reference value 10 ug/ml had 80% accuracy. At this range, CRP demonstrated a very good sensitivity of 97% for identifying infection from inflammation within the two groups when healthy individuals were not included in the analysis. Specificity was not good however, and CRP incorrectly assigned a diagnosis of infection to 9 of 13 individuals from the Inflammation group who were experiencing a non-infectious inflammation. Such a diagnosis in a clinical setting could potentially lead to the administration of an inappropriate treatment to individuals who are not experiencing an infection.
[0199] In the same analysis, SAA was assessed at 10 ug/ml to distinguish between the Inflammation and Infection groups. SAA demonstrated an accuracy of 86% and specificity and sensitivity of 86% and 85% respectively. SAA accurately categorised 32 of 37 individuals from the Infection group and 11 of 13 individuals from the Inflammation group when the Healthy group was omitted. This data supports the clinical utility of SAA as a biomarker for differentiating between infection and inflammation. However, since the Healthy group were omitted from the analysis and the findings of Table 11 and Table 12 show SAA as a poor differentiator of Healthy and Inflammation groups, SAA alone is not a suitable biomarker for identifying between the three groups Healthy, Inflammation and Infection, though it does differentiate well between Inflammation and Infection or Healthy and Infection.
[0200] Hp had a specificity of 85% but low sensitivity at 24% meaning it could not differentiate between Healthy Individuals with Inflammation accurately.
TABLE-US-00015 TABLE 14 Performance statistics for SAA and CRP at different cut-off values to distinguish between Individuals with Infection and Individuals with Inflammation Description/(ug/ml) sensitivity specificity PPV NPV accuracy Infection vs Inflammation, 84 92 97 67 86 SAA>= 20 Infection vs Inflammation, 86 85 94 69 86 SAA >= 10 Infection vs Inflammation, 97 8 75 50 74 CRP>=5 Infection vs Inflammation, 97 31 80 80 80 CRP>= 10 Infection vs Inflammation, 24 85 82 28 40 Hp >= 220
TABLE-US-00016 TABLE 15 Further performance statistics for SAA and CRP at different cut-off values to distinguish between Individuals with Infection and Individuals with Inflammation actual Description true false true false actual neg- (ug/ml) positive positive negative negative positive ative Infection vs 31 1 12 6 37 13 Inflammation, SAA >= 20 Infection vs 32 2 11 5 37 13 Inflammation, SAA >= 10 Infection vs 36 12 1 1 37 13 Inflammation, CRP >= 5 Infection vs 36 9 4 1 37 13 Inflammation, CRP >= 10 Infection vs 9 2 11 28 37 13 Inflammation, Hp >= 220
Healthy Vs Non-Healthy (Infection and Inflammation)
[0201] In an analysis differentiating between Healthy and Non-Healthy groups, CRP at the cut off value of 10 ug/ml distinguished non-healthy individuals within the group with a sensitivity of 90%. However, specificity was poor meaning many healthy individuals were incorrectly assigned a non-healthy status (Table 16 and Table 17), i.e., false positives.
[0202] SAA distinguished between the Healthy and Non-Healthy group with sensitivity at 68%. There were false negative findings, and 16 of 50 non-healthy individuals were assigned a Healthy status despite being clinically non-healthy (Table 16 and Table 17), i.e., false negatives. Hp has high specificity but low sensitivity at 22% meaning it could not differentiate between Healthy Individuals with Inflammation accurately.
TABLE-US-00017 TABLE 16 Performance statistics for SAA and CRP at different cut-off values to distinguish between Healthy Individuals and Non-Healthy Individuals (Inflammation and Infection). Description/(ug/ml) sensitivity specificity PPV NPV accuracy Non-healthy vs 64 100 100 36 70 Healthy, SAA >= 20 Non-healthy vs 68 100 100 38 73 Healthy, SAA >= 10 Non-healthy vs 96 30 87 60 85 Healthy, CRP >= 5 Non-healthy vs 90 30 87 38 80 Healthy, CRP >= 10 Non-healthy vs 22 100 100 20 35 Healthy, Hp >= 220
TABLE-US-00018 TABLE 17 Further performance statistics for SAA and CRP at different cut-off values to distinguish between Healthy Individuals and Non-Healthy Individuals (Inflammation and Infection). actual actual Description true false true false pos- neg- (ug/ml) positive positive negative negative itive ative Non-healthy vs 32 0 10 18 50 10 Healthy, SAA >= 20 Non-healthy vs 34 0 10 16 50 10 Healthy, SAA >= 10 Non-healthy vs 48 7 3 2 50 10 Healthy, CRP >= 5 Non-healthy vs 45 7 3 5 50 10 Healthy, CRP >= 10 Non-healthy vs 11 0 10 39 50 10 Healthy, Hp >= 220
Infection Vs Non-Infection (Healthy and Inflammation)
[0203] CRP at the cut off value of 10 ug/ml distinguished Infection from Non-Infection groups. However, specificity was poor as previously seen in single biomarker analysis, resulting in significant false positives (Table 19), which may lead to inaccurate treatment decisions and unnecessary clinical investigations affecting the patient and economic burden. SAA performed better when distinguishing between the Infection and Non-Infection group with good sensitivity and very good specificity.
TABLE-US-00019 TABLE 18 Performance statistics for SAA and CRP combined for distinguishing between Individuals with Infection and Non-Infection (Healthy and Inflammation). Description/(ug/ml) sensitivity specificity PPV NPV accuracy Infection vs Non- 84 96 97 79 88 infection, SAA >= 20 Infection vs Non- 86 91 94 81 88 infection, SAA >= 10 Infection vs Non- 97 17 65 80 67 infection, CRP >= 5 Infection vs Non- 97 30 69 88 72 infection, CRP >= 10
TABLE-US-00020 TABLE 19 Further performance statistics for SAA and CRP combined for distinguishing between Individuals with Infection and Non-Infection (Healthy and Inflammation). Description true false true false actual actual (ug/ml) positive positive negative negative positive negative Infection vs 31 1 22 6 37 23 Non- infection, SAA >= 20 Infection vs 32 2 21 5 37 23 Non- infection, SAA >= 10 Infection vs 36 19 4 1 37 23 Non- infection, CRP >= 5 Infection vs 36 16 7 1 37 23 Non- infection, CRP >=10
Combined Healthy Vs Non-Healthy (Infection and Inflammation)
[0204] Combination of CRP and SAA to differentiate between Healthy and Non-Healthy groups shows improvement over the individual biomarker assessments with improved accuracy across all combinations (Table 20). However, while sensitivities increased, specificities were not improved. The combination allowed better detection of true positives than individual analysis, meaning non-healthy statuses are being detected more accurately and therefore the combination reduced the incidence of false negatives (see Table 21).
[0205] The clinical outcome of the combination is better ability to identify non-heathy individuals compared to individual analysis of SAA or CRP. However, there was an elevation of false positives at some combinations whereby individuals in the Healthy groups were categorised as Non-Healthy, i.e., false positives. In a clinical setting this may lead to Individuals being treated incorrectly or unnecessarily undergoing further clinical investigative work.
TABLE-US-00021 TABLE 20 Performance statistics for SAA and CRP combined for distinguishing between Healthy and Non-Healthy Individuals. Cut off Description (ug/ml) Sensitivity Specificity PPV NPV Accuracy Non-healthy (5, 10) 30 98 75 88 87 vs Healthy Non-healthy (5, 20) 30 98 75 88 87 vs Healthy Non-healthy (10, 10) 30 92 43 87 82 vs Healthy Non-healthy (10, 20) 30 92 43 87 82 vs Healthy
TABLE-US-00022 TABLE 21 Further performance statistics for SAA and CRP combined for distinguishing between Healthy and Non-Healthy Individuals. Cut off/ Description (ug/ml) Sensitivity Specificity PPV NPV Accuracy Non-healthy (5, 10) 49 7 3 1 50 vs Healthy Non-healthy (5, 20) 49 7 3 1 50 vs Healthy Non-healthy (10, 10) 46 7 3 4 50 vs Healthy Non-healthy (10, 20) 46 7 3 4 50 vs Healthy
Combined Infection Vs Non-Infection (Healthy and Inflammation)
[0206] The combination of SAA and CRP in the differentiation of Infection and Non-Infection groups shows improvement over single biomarker analysis with improved specificities for all combinations (Table 22). The improvement in specificity is associated with a reduction in false negatives for individuals in the Infection group observed in single biomarker analysis particularly for CRP.
TABLE-US-00023 TABLE 22 Performance statistics for SAA and CRP combined for distinguishing between Individuals with Infection and Non-Infection (Healthy and Inflammation). Cut off Description (ug/ml) Sensitivity Specificity PPV NPV accuracy Infection vs (5, 10) 86 91 94 81 88 Non-infection Infection vs (5, 20) 84 96 97 79 88 Non-infection Infection vs (10, 10) 86 91 94 81 88 Non-infection Infection vs (10, 20) 84 96 97 79 88 Non-infection
TABLE-US-00024 TABLE 23 Further performance statistics for SAA and CRP combined for distinguishing between Individuals with Infection and Non-Infection (Healthy and Inflammation) Cut off true false true false actual Description (ug/ml) positive positive negative negative positive Infection vs (5, 10) 10 12 35 3 13 Non-infection Infection vs (5, 20) 11 13 34 2 13 Non-infection Infection vs (10, 10) 7 12 35 6 13 Non-infection Infection vs (10, 20) 8 13 34 5 13 Non-infection
Combined Non-Infectious Inflammation Vs Other (Healthy and Infection)
[0207] Using SAA or CRP alone, it is not possible to identify individuals with non-infectious inflammation from Healthy and Infectious groups. In combination however it has been shown below to be possible to differentiate between the three possible health status outcomes (Healthy, Infection and Inflammation). This is an improvement over existing methods wherein neither CRP or SAA can accurately categorize individuals experiencing non-infectious inflammatory conditions.
TABLE-US-00025 TABLE 24 Performance statistics for SAA and CRP combined for distinguishing between Individuals with Inflammation and those with either Healthy or Infection conditions (Other) Cut off Description (ug/ml) Sensitivity Specificity PPV NPV accuracy Non-Infectious (5, 10) 77 74 45 92 75 Inflammation vs Other Non-Infectious (5, 20) 85 72 46 94 75 Inflammation vs Other Non-Infectious (10, 54 74 37 85 70 Inflammation 10) vs Other Non-Infectious (10, 62 72 38 87 70 Inflammation 20) vs Other
TABLE-US-00026 TABLE 25 Further performance statistics for SAA and CRP combined for distinguishing between Individuals with Inflammation and those with either Healthy or Infection conditions (Other) Cut off true false true false actual Description (ug/ml) positive positive negative negative positive Non- (5, 10) 10 12 35 3 13 Infectious Inflammation vs Other Non- (5, 20) 11 13 34 2 13 Infectious Inflammation vs Other Non- (10, 10) 7 12 35 6 13 Infectious Inflammation vs Other Non- (10, 20) 8 13 34 5 13 Infectious Inflammation vs Other
[0208] Analysis of SAA and CRP in combination allow differentiation between three groups: Healthy, Inflammation and Infection. This is not possible using a single biomarker and a single cut-off value. This is because there three outcomes, i.e., healthy, infection or inflammation, and a single biomarker at a single threshold can only provide two outcomes. While there is improvement over single biomarker analysis when SAA and CRP are combined, there remains a need for a more accurate combination to reduce false negative and false positive results, and correctly determine whether an individual is healthy, has an infection, or has an inflammation. One of the issues with the combination is an elevation of false positives, which in a clinical setting may lead to individuals being treated incorrectly or unnecessarily undergoing further clinical investigative work which creates a burden on the patient, the clinician and an economic burden. SAA and CRP perform well to differentiate between Infection and Non-Infection groups. However, to correctly differentiate between Healthy, Inflammation and Infection, improvements are required in distinguishing between Healthy and Non-Healthy groups and between Healthy and Inflammation groups. Improvement in differentiating the Inflammatory group from both Healthy and Infection groups would also be of significant benefit when clinical decisions are being made relating to detection, treatment, and monitoring.
Combination of SAA, CRP and Hp
[0209] A study was undertaken including the biomarker Haptoglobin to assess diagnostic improvement in differentiating the three groups (Healthy, Infection and Inflammation). Owing to the wide range of normal values reported in the literature for Hp (50-220 mg/dl), and the findings by the inventors, the upper was not suitable and so the inventors undertook a study to identify optimal cut off values. The values determined via ROC analysis for the differentiation of the following groups was 114 ug/ml (healthy vs inflammation), 159 ug/ml (healthy vs non healthy) and 193 ug/ml (infection vs non-infection), respectively.
Healthy Vs Non Healthy (Infection and Inflammation) SAA, CRP, Hp
[0210] SAA, CRP and Hp lead to an overall improvement in true and false negative rates within the Healthy and Non-Healthy groups when compared to a dual analysis of SAA and CRP and false positives appear to reduce significantly (Table 26 and Table 27).
TABLE-US-00027 TABLE 26 Performance statistics for SAA, CRP and Hp combined for distinguishing between Healthy and Non-Healthy Individuals. Cut off (ug/ml) or Positive Description (mg/dl) type Sensitivity Specificity PPV NPV accuracy Healthy vs (5, 10, 114) Healthy 60 94 67 92 88 Non-healthy Healthy vs (5, 10, 159) Healthy 70 84 47 93 82 Non-healthy Healthy vs (5, 10, 220) Healthy 100 72 42 100 77 Non-healthy Healthy vs (5, 20, 114) Healthy 60 94 67 92 88 Non-healthy Healthy vs (5, 20, 159) Healthy 70 84 47 93 82 Non-healthy Healthy vs (5, 20, 220) Healthy 100 70 40 100 75 Non-healthy Healthy vs (10, 10, 114) Healthy 60 88 50 92 83 Non-healthy Healthy vs (10, 10, 159) Healthy 70 84 47 93 82 Non-healthy Healthy vs (10, 10, 220) Healthy 100 72 42 100 77 Non-healthy Healthy vs (10, 20, 114) Healthy 60 88 50 92 83 Non-healthy Healthy vs (10, 20, 159) Healthy 70 84 47 93 82 Non-healthy Healthy vs (10, 20, 220) Healthy 100 70 40 100 75 Non-healthy
TABLE-US-00028 TABLE 27 Further performance statistics for SAA, CRP and Hp combined for distinguishing between Healthy and Non-Healthy Individuals. Cut off (ug/ml) or positive true false true false actual actual Description (mg/dl) type +ve +ve −ve −ve +ve −ve Healthy vs (5, 10, 114) Healthy 6 3 47 4 10 50 Non-healthy Healthy vs (5, 10, 159) Healthy 7 8 42 3 10 50 Non-healthy Healthy vs (5, 10, 220) Healthy 10 14 36 0 10 50 Non-healthy Healthy vs (5, 20, 114) Healthy 6 3 47 4 10 50 Non-healthy Healthy vs (5, 20, 159) Healthy 7 8 42 3 10 50 Non-healthy Healthy vs (5, 20, 220) Healthy 10 15 35 0 10 50 Non-healthy Healthy vs (10, 10, 114) Healthy 6 6 44 4 10 50 Non-healthy Healthy vs (10, 10, 159) Healthy 7 8 42 3 10 50 Non-healthy Healthy vs (10, 10, 220) Healthy 10 14 36 0 10 50 Non-healthy Healthy vs (10, 20, 114) Healthy 6 6 44 4 10 50 Non-healthy Healthy vs (10, 20, 159) Healthy 7 8 42 3 10 50 Non-healthy Healthy vs (10, 20, 220) Healthy 10 15 35 0 10 50 Non-healthy
Infection Vs Non Infection (Healthy and Inflammation):SAA, CRP, Hp
[0211] SAA, CRP and Hp demonstrated a significant improvement in the detection of true positives and a reduction in the false negative rate compared to SAA and CRP alone or combined.
TABLE-US-00029 TABLE 28 Performance statistics for SAA, CRP and Hp combined for distinguishing between Infection and Non-Infection Cut off (ug/ml) or Sensi- Spec- Description (mg/dl) tivity ificity PPV NPV accuracy Infection vs (5, 20, 114) 84 96 97 79 88 Non-infection Infection vs (5, 10, 114) 62 86 91 94 81 Non-infection Infection vs (5, 20, 159) 84 96 97 79 88 Non-infection Infection vs (5, 20, 220) 84 96 97 79 88 Non-infection Infection vs (10, 20, 114) 84 96 97 79 88 Non-infection Infection vs (10, 10, 114) 86 91 94 81 88 Non-infection Infection vs (10, 20, 159) 84 96 97 79 88 Non-infection Infection vs (10, 20, 220) 84 96 97 79 88 Non-infection
TABLE-US-00030 TABLE 29 Further performance statistics for SAA, CRP and Hp combined for distinguishing between Infection and Non-Infection Cut off (ug/ml) or positive true false true false actual Description (mg/dl) type +ve +ve −ve −ve +ve Infection vs (5, 20, 114) 31 1 22 6 37 23 Non-infection (5, 10, 114) 32 2 21 5 37 23 Infection vs (5, 20, 159) 31 1 22 6 37 23 Non-infection Infection vs (5, 20, 220) 31 1 22 6 37 23 Non-infection Infection vs (10, 10, 114) 32 2 21 5 37 32 Non-infection Infection vs (10, 20, 114) 31 1 22 6 37 23 Non-infection Infection vs (10, 20, 159) 31 1 22 6 37 23 Non-infection Infection vs (10, 20, 220) 31 1 22 6 37 23 Non-infection
Non-Infectious Inflammation Vs Other (Healthy and Infection): SAA, CRP, Hp
[0212] The combination of Hp surprisingly improved the differentiation of the Inflammation group within the three groups, Improved sensitivity and specificity and overall accuracy were observed compared to CRP and SAA. The false negative rate was maintained and there was a decrease in false positives.
TABLE-US-00031 TABLE 30 Performance statistics for SAA, CRP and Hp combined for distinguishing between Non-Infectious Inflammation and Other (Healthy and Infection) Cut off (ug/ml) or Sensi- Spec- accu- Description (mg/dl) tivity ificity PPV NPV racy Non-Infectious (5, 10, 114) 77 85 59 93 83 Inflammation vs Other Non-Infectious (5, 10, 159) 46 89 55 86 80 Inflammation vs Other Non-Infectious (5, 10, 220) 15 100 100 81 82 Inflammation vs Other Non-Infectious (5, 20, 114) 85 83 58 95 83 Inflammation vs Other Non-Infectious (5, 20, 159) 54 87 54 87 80 Inflammation vs Other Non-Infectious (5, 20, 220) 15 98 67 81 80 Inflammation vs Other Non-Infectious (10, 10, 114) 54 85 50 87 78 Inflammation vs Other Non-Infectious (10, 10, 159) 46 89 55 86 80 Inflammation vs Other Non-Infectious (10, 10, 220) 15 100 100 81 82 Inflammation vs Other Non-Infectious (10, 20, 114) 62 83 50 89 78 Inflammation vs Other Non-Infectious (10, 20, 159) 54 87 54 87 80 Inflammation vs Other Non-Infectious (10, 20, 220) 15 98 67 81 80 Inflammation vs Other
TABLE-US-00032 TABLE 31 Further performance statistics for SAA, CRP and Hp combined for distinguishing between Non-Infectious Inflammation and Other (Healthy and Infection) Cut off (ug/ml) or positive true false false actual Description (mg/dl) type +ve +ve true −ve −ve +ve Non-Infectious (5, 10, 114) 10 7 40 3 13 47 Inflammation vs Other Non-Infectious (5, 10, 159) 6 5 42 7 13 47 Inflammation vs Other Non-Infectious (5, 10, 220) 2 0 47 11 13 47 Inflammation vs Other Non-Infectious (5, 20, 114) 11 8 39 2 13 47 Inflammation vs Other Non-Infectious (5, 20, 159) 7 6 41 6 13 47 Inflammation vs Other Non-Infectious (5, 20, 220) 2 1 46 11 13 47 Inflammation vs Other Non-Infectious (10, 10, 114) 7 7 40 6 13 47 Inflammation vs Other Non-Infectious (10, 10, 159) 6 5 42 7 13 47 Inflammation vs Other Non-Infectious (10, 10, 220) 2 0 47 11 13 47 Inflammation vs Other Non-Infectious (10, 20, 114) 8 8 39 5 13 47 Inflammation vs Other Non-Infectious (10, 20, 159) 7 6 41 6 13 47 Inflammation vs Other Non-Infectious 10, 20, 220) 2 1 46 11 13 47 Inflammation vs Other
Combination of SAA, CRP, Hp and MxA were Assessed Using ROC Analysis to Determine Cut Off Values of Interest.
[0213] SAA, CRP, MxA and HP were combined using ROC analysis cut-off values of interest as shown in Table 11 and in
Healthy Vs Non Healthy (Infection and Inflammation): Hp
[0214]
TABLE-US-00033 TABLE 32 Performance statistics for SAA, CRP and Hp combined for distinguishing between Healthy and Non-Healthy groups. Cut off (ug/ml) or Sensi- Spec- Description (mg/dl) tivity ificity PPV NPV Accuracy Healthy (5, 6, 114) 40 94 57 89 85 vs Non-healthy Healthy (5, 6, 159) 50 86 42 90 80 vs Non-healthy Healthy (10, 6, 114) 40 88 40 88 80 vs Non-healthy Healthy (10, 6, 159) 50 86 42 90 80 vs Non-healthy Healthy (29, 6, 114) 50 82 36 89 77 vs Non-healthy Healthy (29, 6, 159) 60 80 38 91 77 vs Non-healthy Healthy (29, 10, 114) 70 80 41 93 78 vs Non-healthy Healthy (29, 10, 159) 80 76 40 95 77 vs Non-healthy Healthy (29, 20, 114) 70 80 41 93 78 vs Non-healthy Healthy (29, 20, 159) 80 76 40 95 77 vs Non-healthy
TABLE-US-00034 TABLE 33 Further performance statistics for SAA, CRP and Hp combined for distinguishing between Healthy and Non-Healthy groups. Cut off (ug/ml) or true false true false actual actual Description (mg/dl) +ve +ve −ve −ve +ve −ve Healthy vs (5, 6, 114) 4 3 47 6 10 50 Non-healthy Healthy vs (5, 6, 159) 5 7 43 5 10 50 Non-healthy Healthy vs (10, 6, 114) 4 6 44 6 10 50 Non-healthy Healthy vs (10, 6, 159) 5 7 43 5 10 50 Non-healthy Healthy vs (29, 6, 114) 5 9 41 5 10 50 Non-healthy Healthy vs (29, 6, 159) 6 10 40 4 10 50 Non-healthy Healthy vs (29, 10, 114) 7 10 40 3 10 50 Non-healthy Healthy vs (29, 10, 159) 8 12 38 2 10 50 Non-healthy Healthy vs (29, 20, 114) 7 10 40 3 10 50 Non-healthy Healthy vs (29, 20, 159) 8 12 38 2 10 50 Non-healthy
Infection Vs Non-Infection (Healthy and Inflammation): Hp
[0215]
TABLE-US-00035 TABLE 34 Performance statistics for SAA, CRP and Hp combined for distinguishing between Infection and Non-Infection groups Cut off (ug/ml) or Sensi- Spec- Description (mg/dl) tivity ificity PPV NPV accuracy Infection vs (29, 20, 114) 84 96 97 79 88 Non-infection Infection vs (29, 20, 159) 84 96 97 79 88 Non-infection
TABLE-US-00036 TABLE 35 Further performance statistics for SAA, CRP and Hp combined for distinguishing between Infection and Non-Infection groups Cut off (ug/ml) or false true actual actual Description (mg/dl) +ve −ve false −ve +ve −ve Infection vs (29, 20, 114) 1 22 6 37 23 Non-infection Infection vs (29, 20, 159) 1 22 6 37 23 Non-infection Non-Infectious Inflammation vs Other (Healthy and Infection): Hp
TABLE-US-00037 TABLE 36 Performance statistics for SAA, CRP and Hp combined for distinguishing between Individuals with Inflammation and those with either Healthy or Infection conditions (Other) Cut off (ug/ml) or Sensi- Spec- Description (mg/dl) tivity ificity PPV NPV Accuracy Non-Infectious (5, 6, 114) 46 87 50 85 78 Inflammation vs Other Non-Infectious (5, 6, 159) 15 89 29 79 73 Inflammation vs Other Non-Infectious (10, 6, 114) 23 87 33 80 73 Inflammation vs Other Non-Infectious (10, 6, 159) 15 89 29 79 73 Inflammation vs Other Non-Infectious (29, 6, 114) 8 91 20 78 73 Inflammation vs Other Non-Infectious (29, 6, 159) 0 94 0 77 73 Inflammation vs Other Non-Infectious (29, 10, 114) 31 89 44 82 77 Inflammation vs Other Non-Infectious (29, 10, 159) 23 94 50 81 78 Inflammation vs Other Non-Infectious (29, 20, 114) 38 87 45 84 77 Inflammation vs Other Non-Infectious (29, 20, 159) 31 91 50 83 78 Inflammation vs Other
TABLE-US-00038 TABLE 37 Further performance statistics for SAA, CRP and Hp combined for distinguishing between Individuals with Inflammation and those with either Healthy or Infection conditions (Other) Cut off (ug/ml) or true false true false actual actual Description (mg/dl) +ve +ve −ve −ve +ve −ve Non-Infectious (5, 6, 114) 6 6 41 7 13 47 Inflammation vs Other Non-Infectious (5, 6, 159) 2 5 42 11 13 47 Inflammation vs Other Non-Infectious (5, 10, 114) 3 6 41 10 13 47 Inflammation vs Other Non-Infectious (5, 10, 159) 2 5 42 11 13 47 Inflammation vs Other Non-Infectious (5, 20, 114) 1 4 43 12 13 47 Inflammation vs Other Non-Infectious (5, 20, 159) 0 3 44 13 13 47 Inflammation vs Other Non-Infectious (10, 6, 114) 4 5 42 9 13 47 Inflammation vs Other Non-Infectious (10, 6, 159) 3 3 44 10 13 47 Inflammation vs Other Non-Infectious (10, 10, 114) 5 6 41 8 13 47 Inflammation vs Other Non-Infectious (10, 10, 159) 4 4 43 9 13 47 Inflammation vs Other
Differentiation Between Healthy, Inflammation, and Viral Vs Bacterial Groups
[0216] A companion study was carried out to assess the performance of the biomarkers Hp, SAA, MxA and CRP in differentiating between viral and bacterial infection based on the studies presented above. CRP, MxA and Hp emerged as potential candidates for differentiating between viral and bacterial infection with all three demonstrating diagnostic capability with P<0.001 in ROC analysis (see Table 38 and Table 39).
TABLE-US-00039 TABLE 38 Performance statistics for SAA, CRP, Hp and MxA combined for distinguishing between Healthy, Inflammation, Bacterial and Viral groups simultaneously. positive Sensi- Description type tivity Specificity PPV NPV Accuracy Infection vs Infection 81 96 97 76 87 Non-infection Bacterial vs Bacterial 100 100 100 100 100 Viral Bacterial vs Bacterial 88 100 100 96 97 Non-bacterial Viral vs Non- Viral 76 97 94 88 90 viral Healthy vs Healthy 100 76 45 100 80 Non-healthy Inflammation Inflamma- 38 96 71 85 83 vs Non- tion inflammation Note that the cut off values used for SAA, CRP, CRP, Hp, MxA, SAA, Hp were 20 ug/ml, 10 ug/ml, 97 ug/ml, 182 mg/dl, 0.174 ug/ml, 6 ug/ml, and 114 mg/dl, respectively.
TABLE-US-00040 TABLE 39 Further Performance statistics for SAA, CRP Hp and MxA combined for distinguishing between Healthy, Inflammation, Bacterial and Viral groups simultaneously. true false true false actual actual_ Description +ve type +ve +ve −ve −ve +ve ve Infection vs Infection 30 1 22 7 37 23 Non- infection Bacterial vs Bacterial 14 0 16 0 14 16 Viral Bacterial vs Bacterial 14 0 44 2 16 44 Non- bacterial Viral vs Non- Viral 16 1 38 5 21 39 viral Healthy vs Healthy 10 12 38 0 10 50 Non-healthy Inflammation Inflammation 5 2 45 8 13 47 vs Non- inflammation Note that the cut off values used for SAA, CRP, CRP, Hp, MxA, SAA, Hp were 20 ug/ml, 10 ug/ml, 97 ug/ml, 182 mg/dl, 0.174 ug/ml, 6 ug/ml, and 114 mg/dl, respectively.
DISCUSSION
[0217] The main objective of this study is to evaluate serum amyloid A (SAA), C reactive protein (CRP), haptoglobin (Hp) and myxovirus resistance protein A (MxA) to assess their usefulness in determining an individual's health status by distinguishing between healthy, infectious, and inflammatory conditions. It is also an objective to use the expression levels of these proteins (or the genes encoding said proteins) to differentiate between viral and bacterial infection.
[0218] The evaluation of CRP, SAA, Hp, and MxA in this study demonstrated a number of surprising findings. SAA, CRP and Hp and MxA are all elevated in the presence of infection. However, only Hp is consistently elevated in the presence of non-infectious inflammation. Elevated SAA and CRP are not associated with healthy controls or non-infectious inflammatory disease but are positively correlated with the presence of infection. This is a departure from previous findings in the literature which report SAA and CRP as being non-specific markers associated with inflammatory processes. Elevated Hp is positively correlated with both inflammatory conditions and infectious conditions. In time course analysis, SAA and CRP are earlier, more dynamic indicators of the onset of infection and can be used to monitor disease progression and response to therapy.
[0219] These findings are meaningful and present the possibility of a unique method for the differentiation of healthy, inflammatory and infectious groups by combining the biomarker levels at particular reference points. This could be extended to biomarker levels at particular reference points over particular periods of time. SAA is shown here to be a sensitive and specific marker of infection which can classify groups into infection vs non-infectious, thereby identifying infectious cases which may be candidates for antimicrobial treatment. However, SAA is not a good indicator of healthy or inflammatory groups by itself, and so those patients in inflammatory groups who have an inflammatory disease, which may progress to a more aggressive or infectious condition, or those with sterile inflammatory conditions, are not classified/detected when SAA is used, alone and so are at risk of being either untreated or mistreated. This poses a significant problem as in the absence of a method to definitively differentiate healthy, inflammatory, and infectious conditions, clinicians may feel the need to treat a patient to avoid the risk associated with withholding treatment.
[0220] The data promotes SAA as a definitive marker of infection which could be used to differentiate/identify infection from inflammatory conditions. The data also strongly indicates that SAA in combination with Hp can be used to both detect infection and differentiate non-infectious inflammation from a healthy status.
[0221] Though the biomarkers are useful as single biomarkers for detecting groups i.e. individual groups, alone they lack the required sensitivity and specificity to differentiate between all groups in the study. However, when considered together it is possible to correctly identify the groups to a high degree of accuracy. The combination of the biomarkers as described here is also useful to monitor response to therapy in both inflammatory and infectious cases, whereby a decrease in the relevant biomarkers is associated with an improved health status. The method may also be engaged to assess the efficacy of anti-inflammatory or anti-microbial therapies. It may also be engaged as an ongoing screen in chronic inflammatory cases.
[0222] CRP demonstrated a good ability to correctly identify inflammation within combined Inflammation and Healthy groups. However, it has poor specificity, misclassifying many healthy samples as inflammation in the analysis. SAA demonstrated a poor ability to identify inflammation within the two groups at just 15% specificity. Specificity was 100%, meaning that it correctly categorised all healthy individuals but also misclassified a significant number of individuals with inflammation as healthy. In an analysis where the biomarkers were assessed for their ability to distinguish between non-infectious inflammation and infection, when the Infection and Inflammation groups were combined, CRP was demonstrated to be good at correctly identifying the presence of infection. However, specificity was poor, and it could not correctly categorise individuals in the Inflammation group leading to false positives. SAA showed good performance when identifying infection and differentiating within the Infection and Inflammation groups combined. However, the results from Table 12 herein SAA was unable to accurately differentiate between Healthy and Inflammation groups means that the Inflammation group, if presented alongside Healthy individuals as is the normal course of events in clinical settings (e.g., non-inflammatory conditions and disease screening), may not be correctly diagnosed as having an inflammation and may be misclassified as apparently healthy individuals.
[0223] The outcome of a dichotomous assessment, if applied to a clinical setting, may have serious health consequences for individuals being assessed, since it is not possible to achieve three outcomes using this assessment. Individuals presenting for health status assessment are generally healthy (non-infectious) or non-healthy (infectious or non-infectious), where non-healthy is either infection or inflammation and non-infectious is either healthy or inflammation. Infection can further be differentiated into Viral and Bacterial. The study presented here further assessed the biomarkers individually and then combined the biomarkers for each of the three main groups on this basis, i.e., Healthy vs Non-Healthy (where the Non-Healthy group is a combination of both Inflammation and Infection groups) and Infectious vs Non-Infectious (where the Non-Infectious group is a combination of the Healthy and Inflammatory group) and Inflammation vs Other (a combination of Healthy and Infectious groups).
[0224] Combining SAA and CRP to determine the health status of an individual allows for differentiation between the Healthy, Inflammation and Infection groups, which is a significant improvement over existing methods, where neither CRP nor SAA can accurately identify individuals within the Inflammation group. Inflammation can now be distinguished from Healthy and Infection Groups in a health status assessment, whereas previously this was not possible as testers could not use the biomarkers CRP and SAA individually.
Summary of CRP, SAA, HP Combination Findings
[0225] The combination of SAA, CRP and HP provide a new and improved method for the identification and differentiation of the three main groups that make up a health status assessment, i.e., Healthy, Inflammation, and Infection. Haptoglobin is primarily associated with hemolytic anemia. It has a wide range of normal values and to date it has been difficult to determine appropriate cut off values in the literature. Haptoglobin is not reported as a major human acute phase protein when compared to SAA and CRP and appears to have received limited interest. These studies show, however, that Hp, as part of a three outcome analysis using SAA and CRP, improves differentiation between the groups which ultimately leads to better identification of health status and more accurate selection of treatment for Individuals.
[0226] In practical use, when SAA (and/or CRP) values are below the threshold levels reported here, the presence of an elevated Hp is a positive indicator for inflammatory disease without an infectious process. This is a significant improvement upon existing methods of differentiating infectious and inflammatory conditions.
[0227] A testing method sequence is proposed for assessing the efficacy of the biomarkers together when differentiating between the three groups (healthy, inflammatory, and infectious) (see Table 40).
[0228] This method could be further developed through the addition of MxA into the sequence to further improve one or more of the parameters of sensitivity, specificity, positive predictive value and negative predictive value.
[0229] CRP is the biomarker most frequently associated with the identification of bacterial infection. CRP on its own however cannot be used to distinguish between the multiple groups associated with a health status assessment. The sensitivity and specificity and overall accuracy for CRP as a biomarker for bacterial infection varies in the literature. Studies carried out by the inventors identified a cut off value of 97 ug/ml associated with a sensitivity of 75% and a specificity of 100% for identifying individuals within the Bacterial Infection Group. Similar reasoning may be applied to SAA and HP. SAA, though a good biomarker for differentiating between infection and non-infection, demonstrated no ability to differentiate between viral and bacterial groups as seen in
[0230] The inventors developed a combination of biomarkers, namely at least two, three or all from SAA, CRP, HP and MxA, and cut off values to differentiate between Healthy, Inflammation, and Infection groups, and further between Viral groups, simultaneously. The preferred combination determined is as follows: SAA, CRP, CRP, HP, MxA, SAA, Hp at the following thresholds 20 ug/ml, 10 ug/ml, 97 ug/ml, 182 mg/dL, 0.174 ug/ml, 6 ug/ml, 114 mg/dL, with the following logic: [0231] if (‘SAA’>20) and (‘CRP’>10), then the determination is an Infection; [0232] if (‘CRP’>97) or ((‘Hp’>182) and (‘MxA’>0.174)), then the determination ise Bacterial Infection; [0233] otherwise determine Viral Infection; [0234] else of (‘SAA’<6) or (‘Hp’<114) determine Healthy; [0235] otherwise return “Inflammation”
[0236] The combination of the biomarkers according to the cut off values herein demonstrate a very significant and surprising improvement over current methods that are unable to accurately differentiate simultaneously between Healthy, Inflammatory and Infection groups including Bacterial and Viral groups. The logical parameters above demonstrate the excellent sensitivity and specificity with significant improvement over CRP for the identification of Bacterial Infection Groups wherein the combination was able to detect all groups simultaneously with an accuracy ranging from 83% to 100% relating to the group identified with average accuracy for the method.
[0237] Put another way, in one combination there is SAA, CRP and Hp. In the other combo MxA is added to differentiate between viral and bacterial groups, for example. If the logic for Mxa and CRP at the 97 ug/ml cut off is removed (see text/comments on Table 38 and 39) you get “Infection” instead of “Viral or “Bacterial”.
[0238] The health status assessment of subjects can be carried out simultaneously or sequentially on all three or four biomarkers to determine the health status of the subject. If an infection status is identified in the subject, the assessment can be carried out in a single and/or in multiple step algorithm logic analysis, e.g., determine infection followed by differentiation of viral and bacterial infection. Additional biomarkers of viral or bacterial infection may also be assessed, such as procalcitonin, calprotectin, biomarkers associated with the coagulation cascade, and interleukin 1, 6, 8, and 11. If an infection (bacterial or viral) or an inflammation status is determined in the sample from the subject, then the subject is selected for appropriate treatment, as detailed in, for example Table 1. Subjects selected for treatment can be further monitored using the same method to monitor changes in health status include severity of the ailment and a return to healthy status.
[0239] The analysis of the sample from the subject may be interpreted by eye through the use of colorimetric methods and/or through electronic devices where algorithm logic can be employed according to the preferred combination herein to determine health status.
[0240] The study has found that SAA, CRP and Hp when combined can accurately distinguish between three groups in a health status assessment. The combination of biomarkers may be interpreted simultaneously, and the study has shown greater diagnostic accuracy compared to single or dual biomarker evaluations. The study has also found that SAA, CRP and Hp when in combination can accurately distinguish an outcome method of health assessment as detailed in the steps above, providing for a further outcome by differentiating viral and bacterial individuals accurate within a group that also comprises of healthy and inflammation groups.
[0241] These studies surprisingly found that despite all the biomarkers assessed being reportedly pro-inflammatory markers, they do not all behave in the same way in response to inflammation and are elevated or at normal levels based on the type of inflammation involved. The Applicant also presents a method of combining the biomarkers at particular reference levels to achieve greater levels of accuracy than currently available methods. This method can differentiate between healthy and inflammatory patients as well as inflammatory and infectious patients potentially leading to a reduction in the number of non-infectious cases treated unnecessarily with antimicrobials and an increase in the number of infections identified and treated correctly. This method also identifies inflammatory conditions as distinct from infectious conditions, and so patients with inflammatory symptoms without infection can also be treated and/or monitoring appropriately. Furthermore, the study shows that the biomarkers may be engaged together in a system whereby infection and/or inflammation can be detected early and corresponding treatment efficacy can be measured using the biomarkers over hours and/or days.
[0242] The identification of a biomarker or a panel of biomarkers which provides additional information on whether a condition is infectious or inflammatory in nature would present a number of positive outcomes, including a foreseeable reduction in unnecessary use of the already limited arsenal of antimicrobials available, rapid intervention, better clinical decision making and improved health management at home and in a clinical setting through monitoring regimes.
[0243] SAA, Hp, CRP, and MxA have been studied here as potential diagnostic and prognostic indicators that provide superior differentiation than currently available methods and which will aid clinicians in the treatment decision making process.
[0244] In the specification the terms “comprise, comprises, comprised and comprising” or any variation thereof and the terms “include, includes, included and including” or any variation thereof are considered to be totally interchangeable and they should all be afforded the widest possible interpretation and vice versa.
[0245] The invention is not limited to the embodiments hereinbefore described but may be varied in both construction and detail.