EP3132270A1

EP3132270A1 - A method for diagnosing tuberculous meningitis

Info

Publication number: EP3132270A1
Application number: EP15779954.5A
Authority: EP
Inventors: Novel Njweipi Chegou; Gerhard Walzl; Anne Marceline VAN FURTH; Douwe Hendrik VISSER
Original assignee: Stellenbosch University
Current assignee: Stellenbosch University
Priority date: 2014-04-15
Filing date: 2015-04-15
Publication date: 2017-02-22
Also published as: EP3132270A4; RU2016144359A; RU2016144359A3; AP2016009538A0; WO2015159239A1; CN106461674A; WO2015159239A4

Abstract

A method for diagnosing tuberculous meningitis (TBM) is described herein. A fluid sample from a subject suspected of having TBM is tested for the presence of at least two of Interleukin-13 (IL-13), Vascular endothelial growth factor (VEGF), Cathelicidin (LL-37), IL- 7, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1a, MIP-1b, RANTES and GM-CSF. Increased levels of at least any two of these biomarkers in the sample compared to levels in subjects without TBM are indicative that the subject has TBM.

Description

A METHOD FOR DIAGNOSING TUBERCULOUS MENINGITIS

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to South African provisional patent application number 2014/02743, which is incorporated by reference herein.

FIELD OF THE INVENTION

The invention relates to a method for diagnosing tuberculous meningitis.

BACKGROUND TO THE INVENTION

One third of the world's population is currently infected with Mycobacterium (M.) tuberculosis, and each year more than 1 .5 million people die as a result of this disease [1 ]. Central nervous system (CNS) involvement accounts for approximately 1 % of all cases of TB, of which tuberculous meningitis (TBM) is the most severe form and frequently occurs during early childhood [2]. Hematogenous spread of bacilli from a primary pulmonary focus can lead to the development of a caseous granuloma, a so-called Rich focus, in the meninges, brain adjacent to the meninges or ventricular ependyma. Rupture of this Rich focus into the subarachnoid space results in the clinical features of TBM. In the majority of cases, TBM develops within a few months after the primary infection [3]. Outcome of TBM is often poor despite adequate anti-TB therapy, and early initiation of treatment is the most critical factor affecting morbidity, mortality and healthcare costs, which emphasizes the importance of early diagnosis of TBM [4]. The lack of sensitive methods for early TBM diagnosis is the most common cause for delayed diagnosis. Currently, the demonstration of acid-fast bacilli and/or culture of M. tuberculosis from cerebrospinal fluid (CSF) represent the gold standard for diagnosing TBM. However, both techniques have low sensitivities (10-20%) [5, 6]. Nucleic acid amplification techniques for the diagnosis of TBM are promising but still undergoing assessment in the pediatric population and are not yet suitable for widespread use in resource-poor countries [4]. Many attempts have been made to develop simplified, mostly antigen-detection, tests for TB but their diagnostic power remains poor. Therefore, in clinical practice the diagnosis of TBM is most often based on a combination of clinical, laboratory and radiological findings.

There is thus a need for an accurate and simple test for diagnosing TBM, especially in resource poor settings.

SUMMARY OF THE INVENTION

According to a first embodiment of the invention, there is provided a method of diagnosing tuberculous meningitis (TBM), the method comprising the steps of:

detecting the presence of or measuring the levels of at least two biomarkers selected from the group consisting of lnterleukin-13 (IL-13), Vascular endothelial growth factor (VEGF), Cathelicidin (LL-37), IL-17, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP- 1 a, MIP-1 b, RANTES and GM-CSF in a sample from a subject suspected of having TBM; and

determining whether the subject has TBM based on the detection of the biomarkers or the levels of the biomarkers in the sample.

The method may provide a positive diagnosis for TBM when level of IL-13, VEGF, LL-37, IL-12p70, IFN-Y, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b and/or RANTES in the sample is higher than typical levels of the same biomarker(s) in subjects without tuberculous meningitis, and more specifically is higher than typical levels of the same biomarker(s) in subjects with viral meningitis or bacterial meningitis.

Alternatively or in addition, a positive diagnosis for TBM may be made when the level of IL-17 and/or GM-CSF in the sample is lower than a typical level of the same biomarker in subjects without tuberculous meningitis, and more particularly is lower than typical levels of the same biomarker(s) in subjects with viral meningitis or bacterial meningitis.

The levels of at least three of the biomarkers listed above may be measured in the sample. The at least two or three biomarkers may be selected from the group consisting of IL-13, VEGF, LL-37 and IL-17, and more preferably from the group consisting of IL-13, VEGF and LL-37. The sample may be of cerebrospinal fluid or serum from the patient.

A capture agent and indicator may be used to bind to each of the biomarkers, and an indicator may be provided to indicate when binding of the capture agents and biomarkers occurs.

The capture agents may be antibodies, affybodies, ankyrin repeat proteins, armadillo repeat proteins, nucleic acid aptamers, modified nucleic acid aptamers, peptides, modified peptides, carbohydrate ligands, synthetic ligands, and synthetic polymers. The indicator may be a calorimetric, electrochemical, chromogenic, optical, fluorescent or a radio-labeled indicator.

According to a second embodiment of the invention, there is provided a device for diagnosing tuberculous meningitis (TBM), the device comprising:

a means for receiving a sample from a subject suspected of having TBM;

capture agents for binding at least two biomarkers selected from the group consisting of lnterleukin-13 (IL-13), Vascular endothelial growth factor (VEGF),

Cathelicidin (LL-37), IL-17, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b,

RANTES and GM-CSF if present in the sample; and

at least one indicator which indicates when the capture agents bind to the biomarkers.

The capture agents and indicator may be as described above. The device may include means for detecting binding of the capture agent to the biomarker, by, for example, electrical, acoustic, optical or mechanical methods.

The device may be a point-of-care device. According to a third embodiment of the inventon, there is provided a kit for diagnosing tuberculous meningitis (TBM) in a sample from a subject, the kit comprising: capture agents for binding at least two biomarkers selected from the group consisting of lnterleukin-13 (IL-13), Vascular endothelial growth factor (VEGF), Cathelicidin (LL-37), IL-17, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b, RANTES and GM-CSFif present in the sample; and

The capture agents and indicator may be as described above. The kit may include a device as described above.

The kit may also include instructions for performing the method of diagnosis described above.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 - Flow diagram of study participants. ^a Clinical case definition of Marais et al.^{7 b} Of the 44 "no meningitis" cases, 20 patients were selected at random for analysis. CSF: cerebrospinal fluid; TBM: tuberculous meningitis.

Figure 2A - CSF and serum biomarker expression levels of TBM and non-TBM cases in CSF samples. Results are expressed as picogram/millilitre (pg/ml). Black circles (+) correspond to TBM cases (n=56); Open circles (-) correspond to non-TBM cases (n=55). Dotted lines represent the average limit of detection for the grouped cytokines. Statistical comparisons have been performed using Mann-Whitney test (^* p-value < 0.01 &≥ 0.001 ; ^** p-value < 0.001 ).

Figure 2B - CSF and serum biomarker expression levels of TBM and non-TBM cases in serum samples. Results are expressed as picogram/millilitre (pg/ml). Black circles (+) correspond to TBM cases (n=56); Open circles (-) correspond to non-TBM cases (n=55). Dotted lines represent the average limit of detection for the grouped cytokines. Statistical comparisons have been performed using Mann-Whitney test (^* p-value < 0.01 &≥ 0.001 ; ^** p-value < 0.001 ).

Figure 3A - Two-dimensional unsupervised hierarchical clustering analysis and principal component analysis of biomarkers in CSF and serum of TBM (n=56) and non-TBM (n=55) cases. The normalized values for each biomarker are depicted according to the color scale, where red and green represent expression above and below the median, respectively. The dendrogram left shows the proximity between the different biomarkers, suggesting that biomarkers within each sub-cluster probably share the same origin, common transcriptional regulation and/or common function. Significant clustering of TBM cases is seen in CSF but not in serum.

Figure 3A - Three-dimensional unsupervised hierarchical clustering analysis and principal component analysis (PCA) of biomarkers in CSF and serum of TBM and non-TBM cases. Each dot represents one subject based on values of all biomarkers studied. The percentage of variances is depicted on the 3 axes. The distance in space between each dot represents the relatedness between each individual. Visual clustering of TBM cases is seen in CSF but not in serum. Figure 4 - Biomarker expression profiles in CSF of TBM and subgroups of non-TBM cases. Unsupervised hierarchical clustering analysis of biomarker expression profiles in CSF of TBM (n=56) and subgroups of non-TBM (bacterial meningitis (n=10); viral meningitis (n=25); no meningitis (n=20)) cases. Grouped medians per subgroup were taken to illustrate the differences in biomarker expression profiles between the groups.

Figure 5 - Top canonical pathways in TBM, bacterial meningitis and viral meningitis. The top canonical pathways in tuberculous meningitis (TBM), bacterial meningitis, and viral meningitis are presented. On the y axis the -log(p-value) indicates the likelihood of an association between a specific pathway and the dataset. The threshold line represents a p-value of 0 05. Pathways associated with IL-17 signalling, hypercytokinemia, and communication between immune cells are relevant to all three types of meningitis. However, the relative abundance of markers linked to pathways associated with pathogenesis of multiple sclerosis, vitamin D receptor/retinoid X receptor activation, and CC chemokine receptor type 5 signalling in macrophages is greater in TBM compared with other types of meningitis.

DETAILED DESCRIPTION OF THE INVENTION A method for diagnosing tuberculous meningitis (TBM) is described herein. A fluid sample from a subject suspected of having TBM is tested for the presence of at least two, at least three or at least four of the following biomarkers: lnterleukin-13 (IL-13), Vascular endothelial growth factor (VEGF), Cathelicidin (LL-37), IL-17, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b, RANTES and GM-CSF. If these biomarkers in the sample are found to be over- or under-expressed compared to levels in subjects without TBM, then this is indicative that the subject has TBM.

More particularly, a positive diagnosis for TBM can be made when the level of IL-13, VEGF, LL-37, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b and/or RANTES in the sample is higher than a typical level of the same biomarker in subjects without tuberculous meningitis, and more specifically is higher than typical levels of the same biomarker(s) in subjects with viral meningitis or bacterial meningitis. Alternatively or in addition, a positive diagnosis for TBM can be made when the level of IL-17 and/or GM- CSF in the sample is lower than a typical level of the same biomarker in subjects without tuberculous meningitis, and more specifically is lower than typical levels of the same biomarker(s) in subjects with viral meningitis or bacterial meningitis.

The sample may be of cerebrospinal fluid or serum from the subject.

As used herein, the terms "comprises", "comprising", "includes", "including", "contains", "containing", and any variations thereof, are intended to cover a non-exclusive inclusion and do not exclude other elements not specifically mentioned.

In one embodiment, the at least two, at least three or at least four biomarkers are selected from the group consisting of IL-13, VEGF, LL-37 and IL-17. In another embodiment the at least two or at least three biomarkers are selected from the group consisting of IL-13, VEGF and LL-37. In yet other embodiments, the at least two biomarkers are IL-13 and VEGF, IL-13 and LL-37, or VEGF and LL-37.

A capture agent and indicator can be used to bind to each of the biomarkers, and an indicator indicates when binding of the capture agents and each of the biomarkers occurs.

The biomarkers can be detected using commercially available techniques, such as ELISA techniques or multiplex bead array technology, although it is intended that a specific point- of-care diagnostic device will be developed for performing the method, particularly for use in resource poor settings.

Cut-off or threshold values can be determined based on levels of biomarkers which are typically found in patients without tuberculous meningitis or in subjects with viral meningitis or bacterial meningitis, and the levels of the biomarkers detected in the sample can be compared to the cut-off levels when making the determination of whether or not the subject has TBM. In other words, the method will detect whether the biomarkers in the panels are under- or over-expressed relative to a subject who does not have TBM or relative to a subject who has viral meningitis or bacterial meningitis.

Antibodies, affybodies, ankyrin repeat proteins, armadillo repeat proteins, nucleic acid aptamers, modified nucleic acid aptamers, peptides, modified peptides, carbohydrate ligands, synthetic ligands or synthetic polymers can be used as the capture agents, and the indicator can be a calorimetric, electrochemical, chromogenic, optical, fluorescent or a radio-labeled indicator.

The method can be used as an initial diagnostic tool whereby a positive diagnosis from this method can, if necessary, be subsequently confirmed by means of a second diagnostic method. In the interim, while waiting for the results of the second test, the subject can be started on treatment. Conversely, the method of the invention can also be used to rule out TBM, thus preventing overtreatment of non-TBM subjects.

The method of the invention can be performed using a diagnostic device which detects and indicates the presence of the biomarkers in the sample. The device has a means for receiving the sample from the subject, such as a loading or receiving area onto or into which the sample is placed. The capture agents and indicators are present in the device, and once the sample has been loaded onto or received into the device, the sample is brought into contact with the capture agents, which are allowed to bind to the biomarkers if present. The indicator will signify that binding has occurred. The device is typically a point-of-care device, such as a lateral flow device. A hand-held device is in the process of being developed which is capable of making a TBM diagnosis based on a lateral flow assay using up-converting phosphor technology. In one embodiment, the device may be a dip-stick which can be dipped into the cerebrospinal fluid or serum sample or onto which the sample can be placed, similar to many home pregnancy test kits. The capture agents are included on the dipstick, and generate a signal when they bind to the biomarkers in the panel, together with a control signal.

The invention will now be described in more detail by way of the following non-limiting examples. Examples Methods

Study population and case definition

A cohort study was conducted among children aged between three months and 13 years old, with symptoms and signs suggestive of meningitis admitted to the Tygerberg Hospital, Cape Town, South Africa during the period November 2009 to November 2012. Symptoms and signs of meningitis included one or more of the following: headache, irritability, vomiting, fever, neck stiffness, convulsions, focal neurological deficits, altered consciousness, or lethargy. All children with suspected meningitis were retrospectively classified into a TBM or a non-TBM group according to the diagnostic criteria of Marais et al. [7].

Microbiological assessment of CSF

In patients with suspected TBM, auramine-staining technique for direct fluorescence microscopy, BACTEC MGIT 960 (Becton Dickinson Diagnostic Systems, USA) culture and two commercial nucleic-acid amplification tests, the GenoType MTBDRplus (Hain Life Science, Nehren, Germany) and GeneXpert MTB/RIF (Cepheid, Sunnyvale, USA) were performed. All tests were used according to manufacturer's guidelines. Gram stain, Indian ink examination and culture on blood agar plates was done in all cases as part of routine care. A viral PCR including a Herpes virus panel (i.e. Cytomegalovirus (CMV), Human Herpes Virus-6 (HV-6), Epstein-Barr virus (EBV), Herpes Simplex Virus 1 (HSV-1 ), Herpes Simplex Virus 2 (HSV-2), and Varicella Zoster Virus (VZV)), Enterovirus and Mumps was performed in the National Health Laboratory Service of the Tygerberg Hospital when there was viral meningitis was suspected.

TBM group

TBM was classified as "Definite" when acid-fast bacilli were evident in the CSF, M. tuberculosis was cultured from the CSF, or M. tuberculosis was detected by one of the two commercial nucleic-acid amplification tests in the CSF in a patient with symptoms or signs suggestive of the disease. TBM was classified as "Probable" when patients had a diagnostic score≥12, when cerebral imaging was available, and≥10 when imaging was unavailable. TBM was classified as "Possible" when a patient had a diagnostic score of 6- 1 1 when imaging was available and 6-9 when imaging was unavailable. Diagnostic scores were based on the uniform clinical case definition of Marais et al [7]. Non-TBM group

Patients were classified as non-TBM when an alternative diagnosis was established. The non-TBM group was subdivided into 3 groups: 1 ) viral meningitis cases; 2) bacterial meningitis cases; and 3) a heterogeneous group of no meningitis cases.

Viral meningitis (VM): All children with a PCR-confirmed diagnosis of viral meningitis as well as those who had a leucocyte count of ≥ 10 x10⁶/l in the CSF in the absence of microorganisms on Gram stain or on routine culture and a clinical course consistent with viral meningitis were included in this group [18].

Bacterial meningitis (BM): All cases with a positive CSF culture, bacteria seen on Gram stain or a positive bacterial antigen latex agglutination tests. Furthermore cases with characteristic features of bacterial meningitis in the CSF (high protein, low glucose, polymorph predominance) with or without a positive blood culture for a bacterial pathogen and a clinical course consistent with bacterial meningitis were also included in this group.

No meningitis: All cases with clinical signs and symptoms of meningitis where meningitis was excluded by a normal CSF and another clear diagnosis was apparent with clinical course consistent with this diagnosis at discharge were included in this group.

Sample preparation

Blood and CSF samples were collected from all children with suspected meningitis during routine diagnostic workup. All samples were centrifuged and serum and CSF supernatant were aliquoted into sterile polypropylene micro tubes and kept at -80 °C until use. Frozen samples were quickly defrosted before use. CSF supernatant then passed through a hydrophilic Durapore® PVDF filter membrane (MultiScreen_HTs-GV plate 0.22 μηι; EMD Millipore Corporation, Billerica, MA, USA, 2013). Multiplex cytokine and chemokine analysis

The levels of host biomarkers in serum and CSF were measured by the Luminex multiplex bead array technology. Twenty-seven host markers including interleukin-1 beta ( I L-1 β), IL- 1 receptor antagonist (IL-1 RA), IL- 2, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12 (p70), IL- 13, IL-15, IL-17, Interferon gamma (IFN-γ), tumor necrosis factor alpha (TNF-a), macrophage inflammatory protein 1 alpha (MIP-1 a), macrophage inflammatory protein 1 beta (MIP-1 β), granulocyte colony stimulating factor (G-CSF), granulocyte macrophage colony stimulating factor (GM-CSF), eotaxin, fibroblast growth factor basic (FGF-basic), monocyte chemotactic protein (MCP-1 ), regulated upon activation normal T-cell expressed and secreted (RANTES), IFN-y-induced protein (IP-10), platelet derived growth factor BB (PDGF-BB) and vascular endothelial growth factor (VEGF), were evaluated in both samples types using Bio-Plex kits (Bio-Rad, Hercules, USA). Prior to analysis, serum samples were diluted 1 :4 with the sample diluent provided in the kit as recommended by the manufacturer (Bulletin#10014905; Bio-Rad). Assays were read on the Bio-Plex 200 platform (Bio-Rad) and the Bio-Plex Manager software 6.0 was used for bead acquisition and analysis. LL-37

Concentration of LL-37 in serum and CSF was assessed with an enzyme-linked immunosorbent assay (ELISA) kit (USCN Life Science Inc. Houston USA). The manufacturer's product manual was followed for this assay with serum samples diluted 1 :500. Initially the intra-assay Coefficient of Variation (CV) for LL-37 measurements was high but after introduction of 2 extra wash steps the intra-assay CV was <10%.

Statistical analysis

Statistical analyses were performed using SPSS version 20 (SPSS Inc, Chicago, IL, USA), GraphPad Prism version 4.0 (GraphPad Software, San Diego, USA) and R version 3.0.1 [19]. Contingency tables were analysed using χ2 test or Fisher's exact test. Odds ratios with 99% confidence intervals (99%CI) were calculated to measure the effect size. Median values were compared using Mann-Whitney tests, and means were compared using unpaired Student f-tests. Pairwise comparisons of non-parametric data were performed using the Kruskal-Wallis test. UHC analysis of biomarkers in CSF and serum and heat map generation, as well as PCA were done using the Qlucore Omics Explorer software (Qlucore, Lund, Sweden). Biological pathway analysis was performed using Ingenuity Pathway Analysis (Ingenuity Systems, Inc., Redwood, CA), which resulted in identification of the top canonical pathways of the markers found in the three different meningitis categories. For the diagnostic prediction model of TBM, multivariable logistic regression analysis was used. Spline analysis was done to evaluate linearity of the factors. Stepwise backward selection based on the Wald test (with p = 0.10 as cut-off) was used to select variables. The fit of the model was tested with the Hosmer-Lemeshow test for goodness of fit. A receiver operating characteristic (ROC) curve based on the predicted probabilities of the model was calculated. Area Under the Curve (AUC), sensitivity, specificity, positive and negative predictive value were calculated. Bootstrapping techniques were used for internal validation. All measurements were conducted in duplicate and the calculated mean of the measurements was used in the analyses. Laboratory personnel were blinded to the clinical information associated with the samples. In all analyses, p-values of < 0.01 were considered statistically significant unless otherwise stated.

5

Ethics

The study was conducted according to the ethical guidelines and principles of the International Declaration of Helsinki and the South African Guidelines for Good Clinical Practice. Ethical approval was obtained from the Stellenbosch University Human 10 Research Ethics Committee. The Department of Paediatrics and Child Health of the Tygerberg Children's Hospital granted approval for recruitment of study participants. Written informed consent was obtained from all patients or their caregivers.

Results

15

Baseline characteristics of studied population

One hundred and forty-six participants with suspected meningitis were eligible for analysis. In total CSF and serum samples of 56 TBM patients and 55 non-TBM patients were analysed (Figure 1 ). Patient characteristics and presenting symptoms are outlined in 20 Table 1 . Statistically significant differences between the TBM and non-TBM group were found with regard to presenting symptoms and symptom duration. Altered consciousness, focal neurological deficits and symptom duration of more than 5 days were present more frequently in the TBM-group.

Table 1 - Baseline patient characteristics (n=11 1 unless otherwise stated)

Characteristic TBM Non-TBM

N (%) N (%) OR (99%CI) p-value

Total number of patients 56 (50.5) 55 (49.5)

Age in months (Mean ± SD) 49.1 ± 39.9 56.9 ± 43.7 0.33

Sex, male 30 (53.6) 35 (63.6) 0.66 (0.24 - 1 .79) 0.28

HIV-infected (n= 55; n=47) 4 (7.3) 3 (6.4) 0.87 (0.1 1 - 6.67) 0.86

BMI, < -2SD (n=45; n=45) ^a 15 (33.3) 9 (20.0) 2.00 (0.57 - 7.04) 0.16

Race:

Black 18 (32.1 ) 9 (16.4) 1 .00

Mixed ancestry 38 (67.9) 44 (80.0) 0.43 (0.13 - 1 .43) 0.07

White ⁰ 0 (0) 2 (3.6) Presenting symptoms: ^c

Fever 44 (78.6) 48 (87.3) 0.54 (0.14 - 2.04) 0.23

Headache 22 (39.3) 33 (60.0) 0.43 (0.16 - 1 .17) 0.03

Convulsions 15 (26.8) 13 (23.6) 1 .18 (0.38 - 3.65) 0.70

Vomiting 36 (64.3) 37 (67.3) 0.88 (0.31 - 2.46) 0.74

Focal neurological deficit 24 (42.9) 7 (12.7) 5.14 (1 .47 - 1 8.00) 0.001

Irritability 22 (39.3) 18 (32.7) 1 .33 (0.48 - 3.70) 0.47

Lethargy 21 (37.5) 16 (29.1 ) 1 .46 (0.52 - 4.16) 0.35

Neck stiffness 33 (58.9) 28 (50.9) 1 .38 (0.52 - 3.71 ) 0.40

Altered consciousness 39 (69.6) 23 (41 .8) 3.19 (1 .14 - 8.92) 0.004

Symptom duration > 5 days 27 (48.2) 1 (1 .8) 50.28 (3.42 - 740.12) <0.001

TBM=tuberculous meningitis; HIV= human immunodeficiency virus.

a Standard deviations (SD) of body mass index (BMI) values were based on the World

Health Organization (WHO) Child Growth Standards.²⁰

b Due to the limited number of white patients these patients were excluded for analysis in this

5 table.

c Presenting symptoms: more than one symptom seen in most cases.

Concentrations were assessed of 28 soluble mediators in serum and CSF samples collected from meningitis suspects during diagnostic workup. Baseline concentrations of

10 each of these 28 biomarkers are illustrated in Figure 2. Statistically significantly elevated concentrations of IL-12p70, IFN-γ, VEGF, LL-37, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b and RANTES and lower concentrations of IL-17 and GM-CSF were observed in CSF of TBM patients as compared to the non-TBM group. In serum of TBM patients, significantly elevated concentrations of IL-12p70, IL-17, IFN-γ, LL-37, IL-4, IL7, IL-8, FGF-basic and G-

15 CSF and lower concentrations of MCP-1 were seen when compared to the non-TBM group. Subgroup analysis showed significant (p-value < 0.05) elevated concentrations of IL-13, VEGF and LL-37 and lower concentrations of IL-17 when CSF of TBM cases were compared to VM cases and BM cases. In serum, significant (p-value < 0.05) elevated concentrations of IL-17, LL-37, IFN-γ and FGF-basic were seen in TBM patients when

20 compared to VM and BM cases. The statistically significantly differences in biomarker expression between TBM, VM and BM cases in CSF and serum are outlined in Table 2.

Unsupervised hierarchical clustering and principal component analysis in CSF and serum To evaluate the presence of a TBM-specific biomarker profile, unsupervised hierarchical 25 clustering (UHC) analysis and principal component analysis (PCA) were done. Significant clustering of TBM cases was seen in CSF samples but not in serum samples (Figure 3a). The UHC results in CSF samples were confirmed by PCA (Figure 3b). To highlight the disease-specific biomarker signature profile in CSF grouped medians of each type of meningitis (TBM, VM and BM) and the "no meningitis" group were calculated and UHC was repeated. Highly significant distinct biomarker profiles were found between the 4 subgroups (Figure 4).

5

Table 2 - Differences in biomarker expression between TBM, viral meningitis and bacterial meningitis cases in CSF and serum

TBM VM and BM p-value^a

Median [IQR] in pg/ml Median [IQR] in pg/ml

CSF

IL-13 78.07 [50.50 -104.82] VM 26.24 [17.74-64.71] 0.001

BM 29.15 [20.63-35.69] 0.003

IL-17 1.95 [0.00-6.05] VM 38.73 [3.07 - 71.34] <0.001

BM 39.81 [2.18-93.88] 0.018

VEGF 142.81 [28.05-225.66] VM 27.92 [7.92-48.73] 0.003

BM 14.46 [8.70-86.47] 0.012

LL-37 5046.41 [3297.51 -6194.68] VM 2633.04 [479.73 - 4733.16] 0.015

BM 142.46 [73.11 - 1297.42] <0.001

Serum

IL-7 23.79 [17.22-30.61] VM 17.19 [12.37-21.73] 0.017

BM 14.98 [8.25-22.77] 0.040

IL-6 32.30 [26.03- 48.35] VM 20.01 [13.60-28.74] 0.007

BM 116.79 [38.36-206.04] 0.046

IL-17 123.55 [75.21 -218.11] VM 63.73 [37.65-79.90] <0.001

BM 40.07 [22.06-86.77] 0.004

LL-37 5052.18 [1961.05- 10659.69] VM 1863.23 [1223.10-2287.02] <0.001

BM ^b 1082.34 [789.78-1873.04] 0.002

IFN-Y 478.33 [391.08 - 624.43] VM 338.87 [157.26 -416.32] <0.001

BM 222.34 [85.04-397.51] 0.005

FGF- 150.43 [94.95-222.17] VM 94.84 [66.08- 109.90] 0.001 basic BM 74.64 [64.18-111.97] 0.009 a Kruskal-Wallis with pairwise comparison, asymptotic significance 2-sided tests with p-value < 10 0.05. Only statistically significantly differences are presented. ^b LL-37 is presented in μg/ml instead of pg/ml and n=55 for TBM cases and n=9 for BM cases. VM: Viral Meningitis (n=25); BM: Bacterial Meningitis (n=25); TBM: Tuberculous Meningitis (n=56)

Ingenuity pathway analysis

15 In order to understand the biological relevance of the marker signatures found to be differentially expressed in CSF of TBM, BM and VM Ingenuity pathway analysis was performed. The top canonical pathways linked to the three different forms of meningitis are shown in Figure 5. Pathways associated with IL-17 signaling, hypercytokinemia and communication between immune cells are relevant to all three types of meningitis. This biological pathway analysis suggests that distinct signalling pathways are differentially activated in TBM, BM and VM.

Diagnostic model for TBM based on biomarker profile in CSF

As a TBM-specific biomarker signature profile was found in CSF with UHC, PCA and pathway analysis, the value of a biomarker-based diagnostic model for TBM was evaluated. All TBM cases were used for this analysis. Biomarkers that were significantly different between the TBM group and the 2 other groups of meningitis (BM and VM) were entered in the multivariable logistic regression model. Spline analysis was done to evaluate linearity of the factors. As none of the selected biomarkers showed linearity and sample size was limited, variables were dichotomized with maximum Youden-index (Maximum {(Sensitivity + Specificity) -1 }) used as cut-off values. Stepwise backward selection (p-value 0.10) based on the Wald test was used to identify variables. The final diagnostic model consists of three biomarkers, IL-13, VEGF and LL-37. A ROC curve based on the predicted probability of the model was calculated with an AUC of 0.929 (95%CI 0.877 - 0.981 ). After internal validation using bootstrapping techniques the AUC was 0.921 and regression coefficients were adjusted by a correction factor of 0.912. Sensitivity, specificity, PPV and NPV of the three biomarkers individually and in combination are outlined in Table 3. Highest specificity is obtained when all biomarkers (IL-13, VEGF and LL-37) are found upregulated with a test sensitivity of 0.52, specificity of 0.95, PPV of 0.91 and NPV of 0.66.

Table 3 - CSF biomarker-based diagnostic model for TBM

Positive biomarker B Sensitivity Specificity PPV NPV

VEGF 1 .46 1 0.60 0.72 1 .00

LL-37 2.07 1 0.66 0.75 1 .00

IL-13 2.63 0.96 0.78 0.82 0.95

VEGF & LL-37 3.53 0.93 0.86 0.87 0.92

VEGF & IL-13 4.09 0.89 0.86 0.87 0.88

IL-13 & LL-37 4.70 0.71 0.91 0.89 0.76

IL-13 & LL-37 & VEGF 6.16 0.52 0.95 0.91 0.66 Regression coefficients (B) were adjusted by a correction factor of 0.912 resulting in a corrected intercept of -3.679. Cut-off values based on maximum Youden-index (Maximum {(Sensitivity + Specificity) -1 }): VEGF: 42.92 pg/ml, LL-37: 3221 .01 pg/ml, IL-13 37.26 pg/ml. PPV: positive predictive value; NPV: negative predictive value.

A multidisciplinary group of TB experts recently proposed the essential criteria for a point- of-care test [26]. For extrapulmonary TB, the test should provide a sensitivity of 60% for probable TBM cases and a specificity of 95%. From this perspective, the diagnostic method presented herein (sensitivity 52%, specificity 95%), although approaching that performance, seems to fall short as a point-of-care test for the diagnosis of TBM. However, this method is superior in comparison to currently available diagnostic tests with even lower sensitivities.

References

1 . World Health Organization (WHO). Global Tuberculosis Control: WHO report 201 1 . http://www.who.int/tb/publications/global_report/201 1 /gtbr1 1_main.pdf (April 15th 2012, date last accessed)

2. Wolzak NK, Cooke ML, Orth H, van Toorn R. The changing profile of pediatric meningitis at a referral centre in Cape Town, South Africa. Journal of tropical pediatrics. 2012 Dec;58(6):491 - 5. PubMed PMID: 22791086.

3. Wallgren A. The time-table of tuberculosis. Tubercle. 1948 Nov;29(1 1 ):245-51 . PubMed PMID:

18101320.

4. Thwaites GE, van Toorn R, Schoeman J. Tuberculous meningitis: more questions, still too few answers. Lancet neurology. 2013 Oct;12(10):999-101 0. PubMed PMID: 23972913.

5. van Well GT, Paes BF, Terwee CB, Springer P, Roord JJ, Donald PR, et al. Twenty years of pediatric tuberculous meningitis: a retrospective cohort study in the western cape of South Africa. Pediatrics. 2009;123(1 ):e1 -8. PubMed PMID: 19367678.

6. Thwaites G, Chau TT, Mai NT, Drobniewski F, McAdam K, Farrar J. Tuberculous meningitis.

Journal of neurology, neurosurgery, and psychiatry. 2000 Mar;68(3):289-99. PubMed PMID: 10675209. Pubmed Central PMCID: 1 736815.

7. Marais S, Thwaites G, Schoeman JF, Torok ME, Misra UK, Prasad K, et al. Tuberculous meningitis: a uniform case definition for use in clinical research. The Lancet infectious diseases. 2010 Nov;1 0(1 1 ):803-12. PubMed PMID: 20822958.

8. Nagesh Babu G, Kumar A, Kalita J, Misra UK. Proinflammatory cytokine levels in the serum and cerebrospinal fluid of tuberculous meningitis patients. Neuroscience letters. 2008 May 2;436(1 ):48-51 . PubMed PMID: 1 8359564.

9. Frankenstein Z, Alon U, Cohen IR. The immune-body cytokine network defines a social architecture of cell interactions. Biology direct. 2006;1 :32. PubMed PMID: 17062134. Pubmed Central PMCID: 1636025. Kashyap RS, Deshpande PS, Ramteke SR, Panchbhai MS, Purohit HJ, Taori GM, et al. Changes in cerebrospinal fluid cytokine expression in tuberculous meningitis patients with treatment. Neuroimmunomodulation. 2010;17(5):333-9. PubMed PMID: 20407285.

Misra UK, Kalita J, Srivastava R, Nair PP, Mishra MK, Basu A. A study of cytokines in tuberculous meningitis: clinical and MRI correlation. Neuroscience letters. 201 0 Oct 8;483(1 ):6- 10. PubMed PMID: 20691761 .

Patel VB, Singh R, Connolly C, Kasprowicz V, Ndung'u T, Dheda K. Comparative utility of cytokine levels and quantitative RD-1 -specific T cell responses for rapid immunodiagnosis of tuberculous meningitis. Journal of clinical microbiology. 201 1 Nov;49(1 1 ):3971 -6. PubMed PMID: 21 880971 . Pubmed Central PMCID: 3209083.

Yilmaz E, Gurgoze MK, llhan N, Dogan Y, Aydinoglu H. lnterleukin-8 levels in children with bacterial, tuberculous and aseptic meningitis. Indian journal of pediatrics. 2002 Mar;69(3):219- 21 . PubMed PMID: 12003295.

Simmons CP, Thwaites GE, Quyen NT, Torok E, Hoang DM, Chau TT, et al. Pretreatment intracerebral and peripheral blood immune responses in Vietnamese adults with tuberculous meningitis: diagnostic value and relationship to disease severity and outcome. Journal of immunology. 2006 Feb 1 ;1 76(3):2007-14. PubMed PMID: 16424233.

Donald PR, Schoeman JF, Beyers N, Nel ED, Carlini SM, Olsen KD, et al. Concentrations of interferon gamma, tumor necrosis factor alpha, and interleukin-1 beta in the cerebrospinal fluid of children treated for tuberculous meningitis. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 1995 Oct;21 (4):924-9. PubMed PMID: 8645841 .

Ceyhan M, Kanra G, Ecevit Z, Secmeer G, Erdem G, Akan O, et al. Tumor necrosis factor- alpha and interleukin-1 beta levels in children with bacterial, tuberculous and aseptic meningitis. The Turkish journal of pediatrics. 1997 Apr-Jun;39(2):177-84. PubMed PMID: 9223915.

Mastroianni CM, Lancella L, Mengoni F, Lichtner M, Santopadre P, D'Agostino C, et al. Chemokine profiles in the cerebrospinal fluid (CSF) during the course of pyogenic and tuberculous meningitis. Clinical and experimental immunology. 1998 Nov;1 14(2):210-4. PubMed PMID: 9822278. Pubmed Central PMCID: 1905098.

Michos AG, Syriopoulou VP, Hadjichristodoulou C, Daikos GL, Lagona E, Douridas P, et al. Aseptic meningitis in children: analysis of 506 cases. PloS one. 2007;2(7):e674. PubMed PMID: 17668054. Pubmed Central PMCID: 1933255.

R Core Team (2013) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-proiect.org/

www.who.int/childqrowth/standards

Szczucinski A, Losy J. Chemokines and chemokine receptors in multiple sclerosis. Potential targets for new therapies. Acta neurologica Scandinavica. 2007 Mar;1 15(3):137-46. PubMed PMID: 17295707. Talat N, Perry S, Parsonnet J, Dawood G, Hussain R. Vitamin d deficiency and tuberculosis progression. Emerging infectious diseases. 2010 May;16(5):853-5. PubMed PMID: 20409383. Pubmed Central PMCID: 2954005.

Shapira Y, Agmon-Levin N, Shoenfeld Y. Mycobacterium tuberculosis, autoimmunity, and vitamin D. Clinical reviews in allergy & immunology. 2010 Apr;38(2-3):1 69-77. PubMed PMID: 19543859.

Visser DH, Schoeman JF, AM VANF. Seasonal variation in the incidence rate of tuberculous meningitis is associated with sunshine hours. Epidemiology and infection. 2012 May 31 :1 -4. PubMed PMID: 22647556.

Flores LL, Steingart KR, Dendukuri N, Schiller I, Minion J, Pai M, et al. Systematic review and meta-analysis of antigen detection tests for the diagnosis of tuberculosis. Clinical and vaccine immunology : CVI. 201 1 Oct;18(10):1 616-27. PubMed PMID: 21832100. Pubmed Central PMCID: 3187027.

Lemaire JF, Casenghi M. New diagnostics for tuberculosis: fulfilling patient needs first. Journal of the International AIDS Society. 2010;13:40. PubMed PMID: 20973946. Pubmed Central PMCID: 2987791 .

Harris J, De Haro SA, Master SS, Keane J, Roberts EA, Delgado M, et al. T helper 2 cytokines inhibit autophagic control of intracellular Mycobacterium tuberculosis. Immunity. 2007 Sep;27(3):505-17. PubMed PMID: 1 7892853.

Van Dyken SJ, Locksley RM. lnterleukin-4- and interleukin-13-mediated alternatively activated macrophages: roles in homeostasis and disease. Annual review of immunology. 2013;31 :317- 43. PubMed PMID: 23298208. Pubmed Central PMCID: 3606684.

Djoba Siawaya JF, Beyers N, van Helden P, Walzl G. Differential cytokine secretion and early treatment response in patients with pulmonary tuberculosis. Clinical and experimental immunology. 2009 Apr;156(1 ):69-77. PubMed PMID: 191 96252. Pubmed Central PMCID: 2673743.

Schrumpf JA, van Sterkenburg MA, Verhoosel RM, Zuyderduyn S, Hiemstra PS. Interleukin 13 exposure enhances vitamin D-mediated expression of the human cathelicidin antimicrobial peptide 18/LL-37 in bronchial epithelial cells. Infection and immunity. 2012 Dec;80(12):4485- 94. PubMed PMID: 23045480. Pubmed Central PMCID: 3497402.

Yancopoulos GD, Davis S, Gale NW, Rudge JS, Wiegand SJ, Holash J. Vascular-specific growth factors and blood vessel formation. Nature. 2000 Sep 14;407(6801 ):242-8. PubMed PMID: 1 1 001067.

van Bruggen N, Thibodeaux H, Palmer JT, Lee WP, Fu L, Cairns B, et al. VEGF antagonism reduces edema formation and tissue damage after ischemia/reperfusion injury in the mouse brain. The Journal of clinical investigation. 1999 Dec;104(1 1 ):1 613-20. PubMed PMID: 10587525. Pubmed Central PMCID: 409867.

van der Flier M, Stockhammer G, Vonk GJ, Nikkels PG, van Diemen-Steenvoorde RA, van der Vlist GJ, et al. Vascular endothelial growth factor in bacterial meningitis: detection in cerebrospinal fluid and localization in postmortem brain. The Journal of infectious diseases. 2001 Jan 1 ;183(1 ):149-53. PubMed PMID: 1 1 106541 .

Matsuyama W, Hashiguchi T, Umehara F, Matsuura E, Kawabata M, Arimura K, et al. Expression of vascular endothelial growth factor in tuberculous meningitis. Journal of the neurological sciences. 2001 May 1 ;186(1 -2):75-9. PubMed PMID: 1 1412875.

van der Flier M, Hoppenreijs S, van Rensburg AJ, Ruyken M, Kolk AH, Springer P, et al. Vascular endothelial growth factor and blood-brain barrier disruption in tuberculous meningitis. The Pediatric infectious disease journal. 2004 Jul ;23(7):608-13. PubMed PMID: 15247597. Husain N, Awasthi S, Haris M, Gupta RK, Husain M. Vascular endothelial growth factor as a marker of disease activity in neurotuberculosis. The Journal of infection. 2008 Feb;56(2):1 14-9. PubMed PMID: 181 58186.

De Y, Chen Q, Schmidt AP, Anderson GM, Wang JM, Wooters J, et al. LL-37, the neutrophil granule- and epithelial cell-derived cathelicidin, utilizes formyl peptide receptor-like 1 (FPRL1 ) as a receptor to chemoattract human peripheral blood neutrophils, monocytes, and T cells. The Journal of experimental medicine. 2000 Oct 2;1 92(7) :1069-74. PubMed PMID: 1 1015447. Pubmed Central PMCID: 2193321 .

Coffelt SB, Marini FC, Watson K, Zwezdaryk KJ, Dembinski JL, LaMarca HL, et al. The proinflammatory peptide LL-37 promotes ovarian tumor progression through recruitment of multipotent mesenchymal stromal cells. Proceedings of the National Academy of Sciences of the United States of America. 2009 Mar 1 0;106(10):3806-1 1 . PubMed PMID: 19234121 . Pubmed Central PMCID: 2656161 .

Yuk JM, Shin DM, Lee HM, Yang CS, Jin HS, Kim KK, et al. Vitamin D3 induces autophagy in human monocytes/macrophages via cathelicidin. Cell host & microbe. 2009 Sep 17;6(3):231 - 43. PubMed PMID: 19748465.

Sonawane A, Santos JC, Mishra BB, Jena P, Progida C, Sorensen OE, et al. Cathelicidin is involved in the intracellular killing of mycobacteria in macrophages. Cellular microbiology. 201 1 Oct;13(10):1 601 -17. PubMed PMID: 21790937.

van der Does AM, Beekhuizen H, Ravensbergen B, Vos T, Ottenhoff TH, van Dissel JT, et al. LL-37 directs macrophage differentiation toward macrophages with a proinflammatory signature. Journal of immunology. 2010 Aug 1 ;185(3):1442-9. PubMed PMID: 20610648. Liu PT, Stenger S, Li H, Wenzel L, Tan BH, Krutzik SR, et al. Toll-like receptor triggering of a vitamin D-mediated human antimicrobial response. Science. 2006 Mar 24;31 1 (5768):1770-3. PubMed PMID: 16497887.

Merres J, Hoss J, Albrecht LJ, Kress E, Soehnlein O, Jansen S, et al. Role of the Cathelicidin- Related Antimicrobial Peptide in Inflammation and Mortality in a Mouse Model of Bacterial Meningitis. Journal of innate immunity. 2013 Aug 22. PubMed PMID: 23969854.

Brandenburg LO, Varoga D, Nicolaeva N, Leib SL, Wilms H, Podschun R, et al. Role of glial cells in the functional expression of LL-37/rat cathelin-related antimicrobial peptide in meningitis. Journal of neuropathology and experimental neurology. 2008 Nov;67(1 1 ):1041 -54. PubMed PMID: 18957897.

Claims

CLAIMS:

1 . A method of diagnosing tuberculous meningitis (TBM), the method comprising the steps of:

measuring the levels of at least two biomarkers selected from the group consisting of lnterleukin-13 (IL-13), Vascular endothelial growth factor (VEGF), Cathelicidin (LL-37), IL-17, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-

1 b, RANTES and GM-CSF in a sample from a subject suspected of having TBM; and

determining whether the subject has TBM based on the levels of the biomarkers in the sample.

2. A method according to claim 1 , wherein a positive diagnosis for TBM is made when the level of IL-13, VEGF, LL-37, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b and/or RANTES in the sample is higher than a typical level of the same biomarker in subjects without tuberculous meningitis.

3. A method according to claim 1 , wherein a positive diagnosis for TBM is made when the level of IL-17 and/or GM-CSF in the sample is lower than a typical level of the same biomarker in subjects without tuberculous meningitis.

4. A method according to claim 1 , wherein a positive diagnosis for TBM is made when the level of IL-13, VEGF, LL-37, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-1 b and/or RANTES in the sample is higher than a typical level of the same biomarker in subjects with viral meningitis or bacterial meningitis.

5. A method according to claim 1 , wherein a positive diagnosis for TBM is made when the level of IL-17 and/or GM-CSF in the sample is lower than a typical level of the same biomarker in subjects with viral meningitis or bacterial meningitis.

6. A method according to any one of claims 1 to 5, wherein levels of at least three of the biomarkers are measured in the sample.

7. A method according to any one of claims 1 to 6, wherein the at least two or three biomarkers are selected from the group consisting of IL-13, VEGF, LL-37 and IL-17.

8. A method according to any one of claims 1 to 7, wherein the at least two or three biomarkers are selected from the group consisting of IL-13, VEGF and LL-37.

9. A method according to any one of claims 1 to 8, wherein the sample is cerebrospinal 5 fluid.

10. A method according to any one of claims 1 to 8, wherein the sample is serum.

1 1 . A method according to any one of claims 1 to 10, wherein capture agents are used o to bind each of the biomarkers.

12. A method according to claim 1 1 , wherein the capture agents are selected from the group consisting of antibodies, affybodies, ankyrin repeat proteins, armadillo repeat proteins, nucleic acid aptamers, modified nucleic acid aptamers, peptides, modified5 peptides, carbohydrate ligands, synthetic ligands and synthetic polymers.

13. A method according to either of claims 1 1 or 12, wherein one or more indicators are provided to indicate when binding of each of the capture agents and biomarkers occurs.

0

14. A method according to claim 13, wherein the indicator is selected from the group consisting of a calorimetric, electrochemical, chromogenic, optical, fluorescent and a radio-labeled indicator. 5

15. A device for diagnosing tuberculous meningitis (TBM), the device comprising:

a means for receiving a sample from a subject suspected of having TBM; capture agents for binding at least two biomarkers selected from the group consisting of lnterleukin-13 (IL-13), Vascular endothelial growth factor (VEGF), Cathelicidin (LL-37), IL-17, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-0 1 b, RANTES and GM-CSF if present in the sample; and

16. A device according to claim 15, which includes capture agents for binding at least5 three of the biomarkers.

17. A device according to either of claims 15 or 16, wherein the at least two or three biomarkers are selected from the group consisting of IL-13, VEGF, LL-37 and IL-17.

18. A device according to any one of claims 15 to 17, wherein the at least two or three biomarkers are selected from the group consisting of IL-13, VEGF and LL-37.

19. A device according to any one of claims 15 to 18, wherein the capture agents are selected from the group consisting of antibodies, affybodies, ankyrin repeat proteins, armadillo repeat proteins, nucleic acid aptamers, modified nucleic acid aptamers, peptides, modified peptides, carbohydrate ligands, synthetic ligands and synthetic polymers.

20. A device according to any one of claims 15 to 19, wherein the indicator indicates binding of the capture agent to the biomarker by electrical, acoustic, optical or mechanical methods.

21 . A device according to any of the claims 15 to 20, which includes measuring means for measuring the levels of the detected biomarkers.

22. A device according to any one of claims 15 to 21 , which further includes amplifying means for increasing the sensitivity of the detection of the biomarkers.

23. A device according to any one of claims 15 to 22, which is a hand-held point-of-care device.

24. A kit for diagnosing tuberculous meningitis (TBM) in a sample from a subject according to the method of any one of claims 1 to 14, the kit comprising:

capture agents for binding at least two biomarkers selected from the group consisting of lnterleukin-13 (IL-13), Vascular endothelial growth factor (VEGF), Cathelicidin (LL-37), IL-17, IL-12p70, IFN-γ, IL-6, IL-10, IL-13, IP-10, MIP-1 a, MIP-

1 b, RANTES and GM-CSF if present in the sample; and

at least one indicator which indicates when the capture agent binds to the biomarkers.

25. A kit according to claim 24, wherein levels of at least three of the biomarkers are measured in the sample.

26. A kit according to either of claims 24 or 25, wherein the at least two or three biomarkers are selected from the group consisting of IL-13, VEGF, LL-37 and IL-17.

27. A kit according to any one of claims 24 to 26, wherein the at least two biomarkers are selected from the group consisting of IL-13, VEGF and LL-37.

28. A kit according to any one of claims 24 to 27, which further includes a device of any one of claims 15 to 23.

29. A kit according to any one of claims 24 to 28, wherein the capture agents are selected from the group consisting of antibodies, affybodies, ankyrin repeat proteins, armadillo repeat proteins, nucleic acid aptamers, modified nucleic acid aptamers, peptides, modified peptides, carbohydrate ligands, synthetic ligands and synthetic polymers.

30. A kit according to any one of claims 24 to 29, which further includes instructions for performing the method of any one of claims 1 to 14.