METHOD FOR DETERMINING AN INDIVIDUAL ABILITY TO RESPOND TO A STIMULUS

Abstract

An in vitro or ex vivo method for determining the ability of an individual to respond to a stimulus, based on the measurement of the expression of at least two different biomarkers, selected from different lists among three lists of biomarkers, from a blood sample of the individual, incubated with the stimulus, as well as tools allowing the implementation of this method and the use of these tools.

Claims

1. An in vitro or ex vivo method for determining an individual ability to respond to a stimulus, comprising: a) A step of incubating a blood sample of said individual with said stimulus, and b) A step of measuring the expression, from the stimulated blood sample resulting from step a), of at least two different biomarkers, selected respectively from at least two different lists, from the following lists: List S1: BST2, CCL20, CCL4, CCL8, CD209, CD3D, CD44, CD74, CD83, CLEC7A, CXCL10, CXCL2, CXCL9, DYRK2, FAM89A, HLA-DMB, HLA-DPB1, IFNG, IL1A, IRAK2, PTGS2, RARRES3, DDX58, SLAMF7, SRC, STAT2, STING, TNFA, TNFSF13B, ZBP1; List S2: ADGRE3, ARL14EP, BST2, C3, CCL2, CCL20, CCL8, CCNB1IP1, IL7R, CD209, CD3D, CD44, CD74, CD83, CDKN1A, CLEC7A, CX3CR1, CXCL10, CXCL2, CXCL9, DYRK2, FAM89A, HLA-DMB, HLA-DPB1, HLA-DRA, IFITM1, IRAK2, SLAMF7, TGFB1; List S3: 121601901-HERV0116, BST2, C3, CCL20, CCL4, CCL8, CCR1, IL7R, CD209, CD44, CD74, CD83, CLEC7A, CXCL10, CXCL9, EIF2AK4, HLA-DMB, HLA-DPA1, HLA-DPB1, HLA-DRA, IL1A, IL2, RARRES3, SLAMF7, STAT2.

2. The method according to claim 1, wherein the at least two different biomarkers are selected respectively from at least two different lists, from the following lists: List S1-1: BST2, CCL20, CCL4, CCL8, CD209, CD3D, CD44, CD83, CXCL2, DYRK2, HLA-DMB, IFNG, IL1A, IRAK2, PTGS2, RARRES3, DDX58, SRC, STAT2, STING, TNFA, TNFSF13B, ZBP1; List S2-1: ADGRE3, ARL14EP, C3, CCL2, CCNB1IP1, IL7R, CD3D, CD44, CDKN1A, CLEC7A, CX3CR1, CXCL2, DYRK2, HLA-DMB, HLA-DRA, IFITM1, IRAK2, TGFB1; List S3-1: 121601901-HERV0116, C3, CCR1, IL7R, CD44, CD74, CXCL10, CXCL9, EIF2AK4, HLA-DMB, HLA-DPA1, HLA-DPB1, HLA-DRA, IL1A, IL2, RARRES3, SLAMF7, STAT2.

3. The method according to claim 1, wherein the at least two different biomarkers are selected respectively from at least two different lists, from the following lists: - List S1-2: CCL20, CCL4, CCL8, CD209, CD44, CD83, CXCL2, IFNG, IL1A, IRAK2, PTGS2, DDX58, SRC, STING, TNFA, TNFSF13B, ZBP1; - List S2-2: ADGRE3, ARL14EP, CCL2, CCNB1IP1, IL7R, CDKN1A, CLEC7A, CX3CR1, DYRK2, IFITM1, TGFB1; List S3-2: 121601901-HERV0116, C3, CCR1, CXCL10, CXCL9, EIF2AK4, HLA-DMB, HLA-DPA1, IL2, SLAMF7.

4. The method according to claim 1, wherein the at least two different biomarkers are selected respectively from at least two different lists, from the following lists: List S1-3: IFNG, PTGS2, DDX58, SRC, STING, TNFA, TNFSF13B, ZBP1; List S2-3: ADGRE3, ARL14EP, CCL2, CNB1IP1, CDKN1A, CX3CR1, IFITM1, TGFB1; List S3-3: 121601901-HERV0116, CCR1, EIF2AK4, HLA-DPA1, IL2.

5. The method according to claim 1, wherein, in step b), the expression of at least three different biomarkers is measured, selected respectively from each of the three lists.

6. The method according to claim 1, wherein the individual is a patient in a resuscitation unit, in an intensive care unit or in a continuous care unit having received surgery or in a septic state.

7. The method according to claim 1, wherein the blood sample is a whole blood sample.

8. The method according to claim 1, wherein the stimulus comprises a molecule capable of binding at least one type of antigen-presenting cell (APC) and at least one type of adaptive immunity cell.

9. The method according to claim 1, wherein the stimulus comprises a molecule of the superantigen type, selected from the superantigens produced by staphylococcal species and the superantigens produced by streptococcal species.

10. The method according to claim 1, wherein the stimulus comprises a molecule selected from SEB (Staphylococcal Enterotoxin B) and SEA (Staphylococcal Enterotoxin A).

11. The method according to claim 1, wherein the stimulus comprises a molecule analogous to a superantigen, said molecule analogous to a superantigen being a bispecific antibody.

12. The method according to claim 1, wherein the stimulus allows direct activation of T lymphocytes.

13. The method according to claim 1, wherein the stimulus is selected from antibodies recognizing and activating a receptor on the surface of the T lymphocyte.

14. The method according to claim 1, wherein the stimulus is an anti-CD3 antibody.

15. The method according to claim 1, wherein the stimulus is of the imidazoquinoline type.

16. The method according to claim 1, wherein the stimulus is Resiquimod (R848).

17. The method according to claim 1, wherein the stimulus comprises a molecule for therapeutic purposes.

18. The method according to claim 1, wherein the expression of the biomarkers is measured at the messenger RNA (mRNA) level.

19. The method according to claim 1, wherein the expression of the biomarkers is measured by RT-PCR.

20. The method according to claim 1, wherein the expression of the biomarkers is measured by sequencing.

21. The method according to claim 1, wherein the expression of the biomarkers is measured by hybridization.

22. The method according to claim 1, wherein the expression of the biomarkers is normalized with respect to the expression of one or more housekeeping genes.

23. The method according to claim 1, wherein it comprises a step of measuring the expression, from a control blood sample without stimulation, of the same biomarkers as those measured from the stimulated blood sample.

24. The method according to claim 23, wherein it comprises a step of calculating the ratios of the expression of each biomarker in the stimulated blood sample, relative to the expression.

25. A kit comprising means for amplifying and/or detecting at least two different biomarkers, selected respectively from at least two different lists among the lists of claim 1, wherein all of the means for amplifying and/or detecting said kit allow the detection and/or amplification of at most 100 biomarkers, in total.

26. The kit according to claim 25, comprising means for amplifying and/or detecting one or more housekeeping genes.

27. The kit according to claim 25, comprising positive control means making it possible to qualify the quality of the RNA extraction, the quality of any amplification and/or hybridization method.

28. A use of: means for amplifying and/or detecting at least two different biomarkers, selected respectively from at least two different lists among the lists of claim 1, or a kit comprising such amplification and/or detection means and optionally, said kit comprises means for amplifying and/or detecting one or more housekeeping genes and/or positive control means making it possible to qualify the quality of the extraction of the RNA, the quality of any amplification and/or hybridization method, to determine an individual ability to respond to a stimulus.

Description

FIGURES

[0089] FIG. 1: The biomarkers contributing the most to the variance for the response of healthy individuals and patients in septic shock, following stimulation with SEB. (A) Principal component analysis (PCA) of the response (stimulated sample/control sample) of 10 healthy individuals (circles) and 30 patients in septic shock (triangles), following stimulation with SEB. Each individual («donor», D) is labeled by its number. The percentage of variance explained by each Principal Component (PC) axis is indicated, as well as the total variance. The position of the vector for each individual was plotted. The most important variables are represented graphically in (B) (representing 20% of the total weight of the variables for PC1 and PC2).

[0090] FIG. 2: Multivariate clustering analysis, following stimulation with SEB. 10 healthy individuals and 30 patients with septic shock were treated as a whole, in order to discriminate gene expression profiles. The response to stimulation with SEB revealed 3 groupings or clusters (S1; n=16, S2; n=11 and S3; n=12), by using the PAM method with correlation distance (score index=31). The dendrogram is based on the distance between the individuals of the medoid of each cluster found by the PAM method. A higher intensity of the gray level (approaching black) on the thermal map (or heatmap) indicates a higher value of the expression ratio or fold change (stimulated sample/control sample) of the biomarkers and a lower intensity of the gray level (approaching white) indicates a lower value of the expression ratio or fold change (stimulated sample/control sample) of the biomarkers. The value 10,000 Ab/c was used as the threshold for high and low mHLA-DR levels. HLA-DR: human leukocyte antigen DR.

[0091] FIG. 3: Distribution of protein TNFα secretion post-stimulation with LPS and mHLA-DR by defined clusters following stimulation with SEB. At day 3-4 after the onset of septic shock, (A) secretion of protein TNFα was measured ex vivo 24 hours post-stimulation with LPS in healthy individuals (circles) and patients in septic shock (squares), and (B) mHLA-DR was measured by flow cytometry, only in patients with septic shock (squares). Mortality (non-surviving individuals) is represented by triangles and nosocomial infections by empty squares. Defined clusters post-stimulation with SEB were obtained using the PAM method with correlation distance. **p <0.001; ***p <0.0001. SEB: staphylococcal enterotoxin B, LPS: lipopolysaccharide, mHLA-DR: monocyte human leukocyte antigen DR.

[0092] The present invention is illustrated without limitation by the following examples.

EXAMPLES

Materials and Methods

[0093] Population of Tested Individuals

[0094] The clinical study was approved by the regional ethics committee (Comite de Protection des Hommes Sud-Est II, number 11236), and registered with the French Ministry of Research (Ministere de l′Enseignement supérieur, de la Recherche et de I'Innovation; DC-2008-509) and the National Data Protection Commission (Commission Nationale de l′Informatique et des Libertés). This study was conducted on patients with septic shock admitted to the intensive care unit of Edouard Herriot Hospital (Hospices Civils de Lyon, Lyon, France) and is part of a larger study looking at immune dysfunctions related in the intensive care unit (NCT02803346).

[0095] Patients with septic shock were included prospectively. Septic shock has been defined according to the Sepsis-3 consensus of the Society for Critical Care Medicine and the European Society for Critical Care Medicine (Singer et al (2016), JAMA 315:801-10): patients requiring the administration of a vasopressor and having a measurement of the serum lactate concentration greater than 2 mmol/L in the absence of hypovolemia in a patient having an infection, or suspected of having an infection (i.e. criteria which define the onset of septic shock in a patient with sepsis). The exclusion criteria were an age below 18 years and the presence of aplasia or a known immunosuppressive disease. At admission, the collected data included demographic characteristics (age, gender) and site of primary infection; the initial severity was assessed by the simplified severity index (IGS II; range of values: 0-163) on admission. Information regarding death during ICU stay was collected, and severity 24 hours after admission was assessed by Sequential Organ Failure Assessment (SOFA) score (range of values: 0-24). Laboratory data during follow-up were also collected, including monocyte HLA-DR (mHLA-DR) values, as well as measurement of TNFα protein secretion after LPS stimulation.

[0096] At the same time, blood samples from healthy individuals (or healthy volunteers) were obtained from the national blood service (French Blood Establishment) and immediately used.

[0097] Immune Functional Tests

[0098] Incubation in TruCulture Tubes

[0099] Heparinized whole blood (1 mL) from patients in septic shock, collected on days 3-4 after onset of septic shock, or from healthy individuals, was dispensed into TruCulture tubes (Myriad Rbm, Austin, Tex., United States) prewarmed, containing medium alone («control sample») or medium with SEB (400 ng/mL). These tubes were then inserted into a dry block incubator and maintained at 37° C. for 24 hours. After incubation, the cell pellet was resuspended in 2 ml of TRI Reagent® LS (Sigma-Aldrich, Deisenhofen, Germany), vortexed for 2 minutes and allowed to stand for 10 minutes at room temperature, before storage at room temperature −80° C.

[0100] Measurement of the Expression of Biomarkers

[0101] For TruCulture cell pellet manipulation and RNA processing and detection, the protocol was followed according to Urrutia et al (2016), Cell Reports 16, 2777-2791. The cell pellets originating from the stimulations by TruCulture and preserved in the TRI Reagent° LS (Sigma-Aldrich) were thawed under shaking. Prior to processing, thawed samples were centrifuged (at 3000g for 5 minutes at 4° C.) to sediment cellular debris generated during Trizol lysis. For extraction, a modified protocol of the NucleoSpin 96 RNA tissue kit (Macherey-Nagel Gmbh&Co. KG, Düren, Germany) was followed using a vacuum system. Briefly, 600 μl of clear lysate obtained by Trizol lysis was transferred to a tube preloaded with 900 μl of 100% ethanol.

[0102] The mixture was transferred to a silica column, then washed with buffers MW1 and MW2, and RNA was eluted using 30 μL of RNase-free water. Nanostring technology was used for mRNA detection of a panel of 46 biomarkers (Table 3)—this is a hybridization-based multiplex assay characterized by the absence of an amplification; 300 ng of RNA were hybridized to the probes at 67° C. for 18 hours using a thermocycler (Biometra, Tprofesssional TRIO, Analytik Jena AG, Jena, Germany).

[0103] After removal of excess probes, samples were loaded into nCounter Prep Station (NanoString Technologies, Seattle, Wash., USA) for purification and immobilization on the inner surface of a sample cartridge for 2-3 time. The sample cartridge was then transferred and imaged on the nCounter Digital Analyzer (NanoString Technologies) where the color codes were counted and tabulated for the 46 biomarkers

TABLE-US-00003 TABLE 3 Target biomarkers used for the Nanostring ® nCounter ®, and their accession number (or chromosomal location) Accession number or Target biomarkers chromosomal location ADGRE3 NM_032571.2 ARL14EP NM_152316.1 BST2 NM_004335.2 C3 NM_000064.2 CCL2 NM_002982.3 CCL20 NM_004591.1 CCL4 NM_002984.2 CCL8 NM_005623.2 CCNB1IP1 NM_182849.2 CCR1 NM_001295.2 CD209 NM_021155.2 CD3D NM_000732.4 CD44 NM_001001392.1 CD74 NM_001025159.1 CD83 NM_004233.3 CDKN1A NM_000389.2 CLEC7A NM_197954.2 CX3CR1 NM_001337.3 CXCL10 NM_001565.1 CXCL2 NM_002089.3 CXCL9 NM_002416.1 DDX58 NM_014314.3 DYRK2 NM_003583.3 EIF2AK4 NM_001013703.2 FAM89A NM_198552.2 HLA-DMB NM_002118.3 HLA-DPA1 NM_033554.2 HLA-DPB1 NM_002121.4 HLA-DRA NM_019111.3 IFITM1 NM_003641.3 IFNG NM_000619.2 IL1A NM_000575.3 IL2 NM_000586.2 IL7R NM_002185.2 IRAK2 NM_001570.3 PTGS2 NM_000963.1 RARRES3 NM_004585.3 SLAMF7 NM_021181.3 SRC NM_005417.3 STAT2 NM_005419.2 STING NM_198282.1 TGFB1 NM_000660.3 TNFA NM_000594.2 TNFSF13B NM_006573.4 ZBP1 NM_001160419.2 121601901-HERV0116 chr12:112972627-112975754

[0104] Generation of Normalized Data

[0105] Each sample was analyzed in a separate multiplexed reaction each comprising 8 negative probes and 6 serial concentrations of positive control probes. Negative control analysis was performed to determine background for each sample. Data were imported into nSolver analysis software (version 4.0, NanoString Technologies) for quality control and data normalization.

[0106] A first standardization step using inner positive controls allows correcting the potential source of variation associated with the technical platform. To do this, we calculated for all samples the level of the average background noise as being the median +3 standard deviations of all six negative probes. Each sample below the background noise level was set to this value.

[0107] Then, the geometric mean of the positive probes is calculated for each sample. A scale factor for a sample was a ratio of the geometric mean of the sample and the mean of all the geometric means. For each sample, all gene values are divided by the corresponding scale factor.

[0108] Finally, to normalize the differences in the amount of introduced RNA, the same method as in normalization by positive controls is used, except that geometric means were calculated for three housekeeping genes (HPRT1 (NM_000194.1), DECR1 (NM_001359.1) and TBP (NM_001172085.1)).

[0109] These genes were selected using the NormFinder method, an established approach for the identification of stable intra- and inter-group housekeeping genes, from the 6 candidate genes included in the custom gene panel. The results are expressed as an expression ratio (or «fold change»). A TruCulture tube containing SEB failed quality control and was not included in the analysis.

[0110] Measurement of mHLA-DR Expression by Flow Cytometry

[0111] The expression of HLA-DR on the surface of circulating monocytes (mHLA-DR) of patients was evaluated at days 3-4 after the onset of septic shock, on peripheral whole blood collected in EDTA tubes, by flow cytometry (NAVIOS; Beckman-Coulter, Brea, Calif., USA). The results are expressed as the number of antibodies bound per cell (Ab/C).

[0112] Protein Detection

[0113] TNFα protein in the supernatant of TruCulture tubes was quantified, for septic shock patients and healthy individuals, using the ELLA nanofluidic system (Biotechne, Minneapolis, Mich., USA), in accordance with the manufacturer instructions. The results are expressed in pg/ml.

[0114] Statistical Analysis

[0115] Results are expressed as median and interquartile ranges [IQR] for continuous variables. Parametric data were analyzed by ANOVA and non-parametric data were analyzed by Kruskal-Wallis test. Statistical analyzes were conducted using GraphPad Prism® software (version 5; GraphPad software, La Jolla, Calif., USA) and R (version 3.5.1). An adjusted p-value <0.05 was considered statistically significant. Principal component analysis (PCA) was performed using Genomics Suite 7 (Partek, St. Louis, Mo., USA).

[0116] Creation of Clusters

[0117] The data were transformed by a base logarithmic transformation 10, centered and reduced. Two distance matrices and a correlation matrix were built on the data and 10 clustering methods were launched («hierarchical», «kmeans», «diana», «fanny», «som», «model», «sota», «pam», «clara» and «agnes»). For each method, k=3 to k=18 clusters were tested. The best clustering methods were selected using 7 indices combining internal measures (connectivity, silhouette width and Dunn's indice) and stability (average proportion of nonoverlapping (APN), average distance (AD), average distance between means (ADM) and figure of merit (FOM)). The most stable method for SEB was selected: it is the PAM method using the correlation matrix (score index=31).

[0118] Results

[0119] Diversity of Response to Stimulation with SEB

[0120] In order to identify the biomarkers contributing mainly to the quantitative variation of the response to stimulation by SEB (FIG. 1A) for healthy individuals and for patients in septic shock, these were represented graphically and the weight of the biomarkers explaining the variance was obtained (Table 4). Among the largest contributors to SEB response variance (FIG. 1B) for the first component PC1 (39%), RARRES3 and STAT2 were found to be most strongly expressed by individuals on the right side of the component, while IL1A, CXCL2 and IFNG were more strongly expressed by individuals on the opposite side. Regarding the second component PC2 (19%), the variance was induced «in the lead» by an element of human endogenous retrovirus or HERV (121601901-HERV0116), but also by SLAMF7, CCL4, C3 and CXCL10.

TABLE-US-00004 TABLE 4 Weights of the biomarkers responsible for the greatest variance of the first component (PC1) and the second component (PC2) for stimulation by SEB in the two populations. For each component, the biomarkers were ranked, from the highest weight (in absolute value) to the lowest weight (in absolute value). PC1 Weight PC2 Weight IL1A −0.2167 121601901- 0.2687 HERV0116 RARRES3 0.2137 SLAMF7 0.2608 IFNG −0.2097 CCL4 0.2454 STAT2 0.2065 CXCL10 0.2445 CXCL2 −0.2039 C3 0.2345 CCL20 −0.2006 CD74 0.2344 CD209 −0.1965 HLA-DRA 0.2285 PTGS2 −0.1964 CXCL9 0.2239 ZBP1 0.1940 HLA-DPA1 0.2208 CD83 −0.1866 ADGRE3 −0.2049 CDKN1A −0.1833 CD44 0.1968 DDX58 0.1811 HLA-DPB1 0.1942 HLA-DMB 0.1798 HLA-DMB 0.1858 CX3CR1 0.1782 TNFSF13B 0.1805 BST2 0.1774 IRAK2 0.1726 IL2 −0.1721 SRC 0.1637 TNFA −0.1717 TNFA 0.1574 IRAK2 −0.1706 BST2 0.1489 HLA-DPB1 0.1654 STING 0.1447 CCL2 −0.1623 FAM89A 0.1350 HLA-DRA 0.1513 RARRES3 0.1221 HLA-DPA1 0.1491 CD83 0.1181 CD74 0.1462 DDX58 0.1069 IFITM1 0.1449 CCL8 0.1066 DYRK2 0.1415 ZBP1 0.1058 TNFSF13B 0.1365 CCR1 0.1048 IL7R 0.1332 CDKN1A 0.0962 CCL8 −0.1327 DYRK2 −0.0959 SRC −0.1300 IL2 0.0903 CD44 −0.1234 CLEC7A −0.0890 ARL14EP 0.1176 CX3CR1 0.0823 STING −0.1162 ARL14EP 0.0783 CCL4 −0.1116 IL7R −0.0770 CCNB1IP1 0.1049 STAT2 0.0715 CLEC7A 0.1015 IL1A 0.0574 EIF2AK4 0.0968 IFITM1 0.0561 CXCL9 −0.0958 CCL2 0.0463 C3 −0.0589 CCL20 0.0461 CD3D 0.0583 CD3D 0.0453 SLAMF7 −0.0505 PTGS2 −0.0444 CCR1 −0.0371 EIF2AK4 0.0353 TGFB1 −0.0277 IFNG 0.0338 CXCL10 −0.0211 TGFB1 0.0228 121601901- −0.0125 CD209 0.0188 HERV0116 FAM89A 0.0090 CCNB1IP1 0.0119 ADGRE3 0.0008 CXCL2 −0.0084

[0121] Immune Functional Test as a Stratification Tool for Sepsis Patients

[0122] By taking into account the two populations (healthy individuals and patients), we carried out an unsupervised classification (clustering) with the entire molecular panel in order to identify the gene motifs. Healthy individuals were clustered together after SEB stimulation, showing great homogeneity in their immune response. During SEB stimulation, 6 patients were grouped with healthy individuals (n=16, cluster S1) and the others were separated into 2 groups of almost equal number (n=11 for cluster S2 and n=12 for the cluster S3; FIG. 2). The donor composition of each cluster is presented in Table 5.

TABLE-US-00005 TABLE 5 Individual composition (per donor) of the clusters obtained after stimulation with SEB. Healthy individuals appear in italics, non-survivors in bold, and those having developed a nosocomial infection are underlined. D: Donor Cluster S1 Cluster S2 Cluster S3 D4; D5; D8; D9; D10; D1; D11; D14; D26; D2; D3; D6; D12; D25; D13; D15; D16; D17; D29; D31; D32; D37; D27; D28; D30; D33; D18; D19; D20; D21; D38 ;D40; D41 D34; D35; D39 D22; D23; D24

[0123] A bivariate analysis was then carried out between the clusters and the biological or clinical parameters.

[0124] For SEB stimulation, a statistically significant result was found for mHLA-DR (adjusted p=0.0131) as well as for TNFα protein secretion after LPS stimulation (adjusted p≤0.0001; Table 6).

[0125] As expected due to the classification with healthy individuals, the 6 patients in cluster S1 present the highest median for mHLA-DR (10938 Ab/C, IQR:[9456-14642]and present a concentration of highest TNFα protein after LPS stimulation (3799 pg/mL, IQR:[2067.2-5401.2]).

[0126] By comparing the results of the clusters S1 and S2, the only significant difference is the median concentration of TNFα protein after stimulation by LPS (p<0.0001). The cluster S2 presents the lowest median TNFα protein level among the 3 clusters.

[0127] By comparing the cluster S1 to S3, there is a significant difference for the two parameters (p <0.001), the cluster S3 presents an intermediate median level of TNFα protein concentration after stimulation by LPS between the 3 clusters, while the median levels of mHLA-DR are the lowest (FIG. 3).

[0128] Moreover, we can observe that among the 20 (out of 30) patients who suffered io from at least one comorbidity, 10 (50%) belonged to S3, representing 83.3% of the cluster.

[0129] Similarly, among the 5 non-surviving patients, the four who died before day 28 (80%) belong to cluster S2, representing 36% of this cluster, while the fifth, who died late in hospital, belongs to cluster S3 (Table 6). It should be noted that the only patient who developed a nosocomial infection is in cluster S2.

TABLE-US-00006 TABLE 6 Bivariate analyzes between clusters S1, S2 and S3 during SEB stimulation for clinical and biological parameters. 6 parameters are represented when statistical analyzes were performed between clusters S1 (n = 16 or n = 6 when there was no information available for healthy individuals), S2 (n = 11) and S3 (n = 12) defined using the PAM method with correlation distance. The p-value adjusted for multiple tests was output. The presence of comorbidities was affirmative when at least one comorbidity was present in the patient: chronic lung disease, heart failure, myocardial infarction, ulcer, diabetes, renal failure or malignant solid tumor. Cluster S1 Cluster S2 Cluster S3 Adjusted (n = 16) (n = 11) (n = 12) p-value Status 0.0002 Healthy individual, n(%) 10 (62.5) 0 (0) 0 (0) Patients, n(%) 6 (37.5) 11 (100) 12 (100) Comorbidities* 0.2348 no, n(%) 1 (16.7) 6 (54.5) 2 (16.7) yes, n(%) 5 (83.3) 5 (45.5) 10 (83.3) CCI* median, [IQR] 2 [1.2-4.2] 1 [0-1.5] 2 [1-5] 0.1593 SOFA* median (day 1), [IQR] 7.5 [6.2-8] 8 [6.5-10.5] 8.5 [8-10] 0.6383 Mortality* 0.2416 no, n(%) 6 (100) 7 (63.6) 11 (91.7) yes, n(%) 0 (0) 4 (36.4) 1 (8.3) mHLA-DR* median 10938 7301 3839.5 0.0131 (day 3-4) (Ab/C), [IQR] [9456-14642] [4653-11673] [3444-6250] Median TNFα secretion, post- 3799 282.7 700.8 0.0001 stimulation with [2067.2-5401.2] [122.2-861.8] [457.8-913.3] LPS (pg/mL), [IQR] SOFA: sequential organ failure assessment CCI: Charlson Comorbidity Index HLA-DR: human leukocyte antigen DR Ab/C: antibodies bound per cell TNFα: tumor necrosis factor alpha LPS: lipopolysaccharide IQR: interquartile range *: parameters measured exclusively for patients in septic shock

[0130] The developed immune functional test has therefore made it possible to demonstrate that, if the immune response of healthy individuals is homogeneous, the immune response of patients with septic shock is heterogeneous, and the heterogeneity of the response lies in the adaptive arm of immunity. Patients grouped in cluster S1, with healthy individuals, have a more «normal»/«healthy» immune profile, unlike other patients. A priori, these patients would not require any particular vigilance and a standard of care would be sufficient. Patients in the cluster S2 correspond to «severe» patients, characterized by a high mortality rate. These patients, whose immunity appears to be strongly impaired and who present a greater probability of mortality, could advantageously benefit from more «aggressive» and/or earlier therapeutic interventions. Finally, the third group (patients of the cluster S3) corresponds to patients with an intermediate to severe phenotype, who may show a degree of immune recovery. Thus, these patients whose immunity seems to be recoverable could be the subject of personalized treatments (e.g. IL-7, interferon γ). Thus, these results show that the immune functional test developed in the context of this invention makes it possible to obtain a stratification of patients, which the reference markers (or gold standard) commonly accepted by the scientific community, such as mHLA- DR or even TNF-α.