METHODS AND KITS OF ASSESSING STATUS, RISK OR PROGNOSIS OF TYPE 1 DIABETES

Abstract

The present invention relates to methods and kits of assessing status, risk or prognosis of type 1 diabetes. There is still a need for improved methods of prognosis of type 1 diabetes. The inventors have observed different alterations of iNKT and MAIT cells quantity, frequency and markers in T1D patients compared to controls and also in children with recent onset T1D compared to control children or children with established T1D. The present invention relates to a method of assessing status, risk or prognosis of type 1 diabetes in a subject comprising i) quantifying at least one population of innate-like T-cells in a blood sample obtained from the subject, ii) comparing the quantification value determined at step i) with a predetermined reference value and iii) detecting differential in the quantification value determined at step i) and the predetermined reference value is indicative of the status, risk of prognosis of type 1 diabetes.

Claims

1. A method of treating a child who has or is at risk of having type 1 diabetes (T1D), comprising: measuring in a blood sample obtained from a child, a quantity of at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 populations of Mucosal-Associated Invariant T (MAIT) cells selected from the group consisting of: MAIT CCR6+ cells, MAIT CD25+ cells, MAIT CD69+ cells; MAIT CD56+ cells, MAIT PD1+ cells, MAIT TIM3+ cells, MAIT KLRG1+ cells, MAIT BTLA+ cells, MAIT BCL2+ cells, MAIT Ki67+ cells, MAIT GzB+ cells, MAIT IFNg+ cells, MAIT TNFa+ cells, MAIT IL4+ cells, MAIT IL10+ cells, MAIT KLRG1+ cells and MAIT IL17+ cells, wherein measurements made on the blood sample of the child indicate at least one of a quantity of MAIT CCR6+ cells in the blood sample is lower a predetermined reference value determined for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT CD25+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT CD69+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT CD56+ cells in the blood sample is lower than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT PD1+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D a quantity of MAIT TIM3+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT KLRG1+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT BTLA+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT BCL2+ cells in the blood sample is lower than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT Ki67+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT GzB+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT IFNg+ cells in the blood sample is lower than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT TNFa+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT IL4+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of MAIT IL10+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, a quantity of KLRG1+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D, and a quantity of MAIT IL17+ cells in the blood sample is higher than a predetermined reference value for control children that do not have a personal or familial history of T1D or autoantibodies associated with T1D; and administering insulin to the child after measurements are made on the blood sample of the child in the measuring step.

2. The method of claim 1 further comprising measuring in the blood sample of the child at least one population of iNKT cells characterized by the presence or absence of at least one marker selected from the group consisting of CCR6, CD56, CD25, CD69, CD161, CD27, PD1, TIM3, KLRG1, BTLA, BCL2, Ki67, CD127, GzB, IFNg, IL2, TNFa, IL4, IL10 and IL17.

3. The method of claim 2 wherein the at least one population of iNKT cells is selected from the group consisting of iNKT BTLA+ cells, iNKT CCR6+ cells, iNKT CD25+ cells, iNKT CD56+ cells, iNKT CD69+ cells, iNKT CD127+ cells, iNKT CD161+ cells, iNKT PD1+ cells, iNKT CD27− cells, iNKT KLRG1+, and iNKT TIM3+ cells.

4. The method of claim 1, wherein the child is 3 to 5 years old.

5. The method of claim 1 further comprising administering to the child one or more agents that control blood glucose levels and/or increase insulin production.

6. A method treating an adult who has or is at risk of having type 1 diabetes (T1D), comprising: measuring in a blood sample obtained from the adult, a quantity of at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 populations of MAIT cells selected from the group consisting of: MAIT CD25+ cells, MAIT CD69+ cells, MAIT CD56+ cells, MAIT Ki67+ cells, MAIT IFNg+ cells and MAIT TNFa+ cells, wherein measurements made on the blood sample of the adult indicate at least one of a quantity of MAIT CD25+ cells in the blood sample is higher than a predetermined reference value for control adults that do not have TID, a quantity of MAIT CD69+ cells in the blood sample is higher than a predetermined reference value for control adults that do not have TID, a quantity of MAIT CD56+ cells in the blood sample is higher than a predetermined reference value for control adults that do not have TID, a quantity of MAIT Ki67+ cells in the blood sample is higher than a predetermined reference value for control adults that do not have TID, a quantity of MAIT IFNg+ cells in the blood sample is lower than a predetermined reference value for control adults that do not have TID, and a quantity of MAIT TNFa+ cells in the blood sample is lower than a predetermined reference value for control adults that do not have TID; and administering insulin to the adult after measurements are made on the blood sample of the adult in the measuring step.

7. The method of claim 6 further comprising measuring in the blood sample of the adult a quantity of iNKT cells characterized by the presence or absence of at least one marker selected from the group consisting of CCR6, CD56, CD25, CD69, CD161, CD27, PD1, TIM3, KLRG1, BTLA, BCL2, Ki67, CD127, GzB, IFNg, IL2, TNFa, IL4, IL10 and IL17.

8. The method of claim 6 wherein the adult is more than 18 years old.

9. The method of claim 6 further comprising administering to the adult one or more agents that control blood glucose levels and/or increase insulin production.

Description

FIGURES

[0092] FIG. 1: frequencies of MAIT cells in the peripheral blood of control children, T1D recent onset children, control adults and T1D adult patients. Patients correspond to the Table 1.***p<0.001 Mann-Whitney test.

[0093] FIG. 2: frequencies of MAIT CCR6+, MAIT CD25+ and MAIT C56+, cells in the peripheral blood of control children, T1D recent onset children, control adults and T1D adult patients. Patients correspond to the Table 1. *p<0.05, **p<0.01, ***p<0.001 Mann-Whitney test.

[0094] FIG. 3: frequencies of iNKT CCR6+ and iNKT CD25+ cells in the peripheral blood of control children, T1D recent onset children, control adults and T1D adult patients. Patients correspond to the Table 1. *p<0.05, **p<0.01.

[0095] FIG. 4: frequencies of MAIT PD1+ and iNKT PD1+ cells in the peripheral blood of control children and T1D recent onset children. Patients correspond to the Table 1. *p<0.05.

[0096] FIG. 5: frequencies of MAIT BCL2+ and MAIT KLRG1+ cells in the peripheral blood of control children and T1D recent onset children. Patients correspond to the Table 1. * p<0.05, **p<0.01.

[0097] FIG. 6: frequencies of MAIT IFNg+ and iNKT IFNg+ cells in the peripheral blood of control children, T1D recent onset children. Patients correspond to the Table 1. *p<0.05, **p<0.01.

[0098] FIG. 7: frequencies of MAIT TNFa+, MAIT IL4+, MAIT GZB+, and MAIT IL17+ cells in the peripheral blood of control children and T1D recent onset children. Patients correspond to the Table 1. *p<0.05, **p<0.01, ***p<0.001 Mann-Whitney test.

[0099] FIG. 8: principal component analysis of cytometric parameters related to MAIT and iNKT cells markers from peripheral blood of control children, T1D recent onset children and T1D children.

[0100] FIG. 9: ROC curve for the logistic regression model based on variables MAIT CCR6, MAIT CD56, MAIT CCR6+CD56+, MAIT CCR6− CD56−, MAIT CD25+PD1+ and MAIT CD25+CD56− PD1

[0101] FIG. 10A-B: Frequency and phenotype alterations of blood MAIT cells from T1D children. MAIT cells were analyzed in blood from children with recent onset T1D (n=41), children with established T1D (n=23), as compared with control children (n=22). (A) Representative staining of MAIT cells and MAIT cell frequency among T lymphocytes. (B) Representative staining and frequency of MAIT cells expressing different cell surface molecules (CCR6, CD56, CD69, CD25, PD1) and intracellular Bcl-2. P values were determined by Kruskal-Wallis test followed by the Wilcoxon rank sum test adjusted with the Holm method.

[0102] FIG. 11A-B: Functional alterations of blood MAIT cells from T1D children. MAIT cells were analyzed in blood from children with recent onset T1D (n=25) and children with established T1D (n=18), as compared with control children (n=18). (A) Representative intracellular staining of MAIT cells for cytokines and GzB after PMA/ionomycin stimulation and graphs showing the frequency of MAIT cells producing cytokines and GzB. P values were determined by Kruskal-Wallis test followed by the Wilcoxon rank sum test adjusted with the Holm method. (B) Frequencies of CD69.sup.+ and CD25.sup.+CD69.sup.+ MAIT cells from children controls (n=6) and children with recent onset T1D (n=7) after ON stimulation with 5-OP-RU at various concentrations (0 to 5 nmol/L). Blocking MR1 mAb was added when indicated. P values were determined by Mann-Whitney test.

[0103] FIG. 12A-E: Relationship between MAIT cell characteristics and disease status of patients. (A) Correlations between the age of children with recent onset T1D (n=25) at diagnosis, the frequency of MAIT cells expressing GzB after PMA/Ionomycin stimulation and the HbA1c levels. Correlations between the age and the frequency of MAIT expressing GzB are also shown for the children with established T1D and control children. P value was determined by Spearman test. (B) MAIT cell staining of surface and intracellular molecules performed on PBMC from fifteen children with recent onset T1D at the time of diagnosis (on the left) and around one year after diabetes onset (on the right). P values were determined by Signed-rank Wilcoxon Test. (C) Factorial discriminant analysis based on the expression of surface and intracellular molecules by MAIT cells from the children with recent onset T1D (n=20), the children with established T1D (n=15) and the children controls (n=18). (D) ROC Curve of the predictive model defining the T1D phenotype. (E) FACS Analysis of MAIT cell frequency among T lymphocytes and frequency of MAIT cells expressing different cell surface molecules from frozen PBMC of adult controls (n=11) and adult at risk T1D donors with at least two autoantibodies (n=11). P values were determined by Mann-Whitney test for b and e.

EXAMPLE 1

[0104] We have observed different alterations of iNKT and MAIT cells quantity, frequency and markers in T1D patients compared to controls and it has also been shown that children with recent onset T1D exhibit different alterations of iNKT and MAIT cells compared to control children or children with established T1D. The alterations should suitable of assessing status, risk or prognosis of type 1 diabetes.

[0105] Early alterations of these cell populations would suggest their implication in defective immunoregulation occurring at the pre-diabetic stage. As suggested by our data in mice models (NOD mice), both cell populations could represent new opportunities to avoid deleterious immune response against beta cells and develop strategies to prevent diabetes development. Table 1 describes our current cohort. For the analyses presented below we have reanalyzed all data at the same time to ensure that all samples were studied using the same gating strategy, for increased consistency and precision.

TABLE-US-00001 Adults with Control Children with recent Control long-standing Children onset T1D Adults T1D n 22 41 29 41 Age (Yr) 9.05 (4.3-18.4)  8.3 (1.6-16)  34.9 (23.3-67.3) 38.9 (17.6-71.1) HbA1c (%) 12.1 (8.7-16.5) 8.07 (5.5-13.5)  Duration of disease 4 days (1-10)  .sup.  16.2 years (2-53)     .sup.  BMI/Z-score −0.3 (−2.2+2.4)  −0.6 (−2.7+3.7) 21.8 (18.0-27.7) 23.3 (17.5-37.2) Sex M/F 13M/9F 21M/20F 12M/17F 15M/26F

[0106] Our results reveal decreased frequencies of iNKT and MAIT cells in children with recent onset T1D (FIG. 1). A reduction of MAIT cells was also seen in adult patients with long-standing T1D compared to adult controls (FIG. 1).

[0107] Importantly, a more in depth analysis of the iNKT or MAIT biological markers showed they can be used for the discrimination between controls and patients with recent onset and patient with long-standing T1D. For example, as shown in FIG. 2, the quantification of MAIT CCR6+, MAIT CD25+ or MAIT CD56+ cells can be used for such a discrimination. MAIT cells were more often CD25-positive in patient with T1D, when compared to controls. MAIT cells were more often CCR6-negative in patient with T1D recent onset, compared to controls or adults with long-standing T1D. At last MAIT cells were more often CD56-negative in patient with T1D recent onset and can be used to discriminate it from both controls and adults with long-standing T1D (FIG. 2).

[0108] In the same way, iNKT cells were more often CD25-positive in children with recent onset T1D, compared to control children and those with established T1D (FIG. 3) whereas iNKT cells were more often CCR6-negative in adults patients with long-standing T1D compared to controls or recent onset (FIG. 3). A negative correlation between % CD25+ iNKT cells and iNKT cell frequency is also observed in the adult long-standing T1D group (r=−0.50, p=0.002**) and in the children recent onset T1D group (r=−0.35, p=0.04*).

[0109] For all of these comparisons, the groups were well matched for age. In short, the most striking finding is a decreased iNKT and MAIT cell frequency in the peripheral blood of recent onset T1D children in association with increased proportions of iNKT and MAIT cells expressing the activation marker CD25. Results have been illustrated in FIGS. 1 to 3 and summarized in Table 2. Of note, there is a positive correlation between the % of CD25+ MAIT cells and the % of CD25+ iNKT cells among the two groups of T1D patients.

TABLE-US-00002 Children Con- with T1D Con- Children trols Recent trols with T1D Test Mann-Whitney children Onset adults Follow up FRE- % MAIT +++ −−− +++ −−− QUENCY % iNKT +++ + ++ + MIGRA- MAIT +++ + +++ ++ TION CCR6+ MAIT ++ −− + ++ CD56+ iNKT + + +++ ++ CCR6+ iNKT CD56+ + + + +/− ACTI- MAIT −−− +++ −−− + VATION CD25+ iNKT CD25+ + +++ − +/−

[0110] These data lead us to propose that in T1D patients there is an abnormal activation of both iNKT and MAIT cells. We hypothesize that this could be due to increased levels of proinflammatory cytokines and/or increase levels of iNKT and MAIT cell ligands, which could be in turn linked to gut microbiota dysbiosis. Both cell types recognize bacterial ligands.

[0111] Hence, a specific analysis of differential markers expression on MAIT and iNKT cells from control children compared to recent onset T1D have been conducted. Those expression data for other molecules (PD1, KLRG1, Bcl2 . . . ,) have been summarized in Table 3 and some have been illustrated in FIGS. 4 to 7.

[0112] For example, as shown in FIG. 4, the quantification of PD1+, which is implicated in cell exhaustion, can be used to discriminate control children from recent T1D onset children. iNKT and MAIT cells were more often PD1-positive in patient with recent T1D, compared to controls (FIG. 4). The same can be stated for BCl2+, KLRG1+(FIG. 5).

[0113] Cytokines and cytotoxic marker GzB, have also been identified, by the inventors, as relevant markers for the discrimination of recent T1D onset as shown in Table 3 and FIGS. 6 and 7.

TABLE-US-00003 Controls T1D Recent Test Mann-Whitney children Onset EXHAUSTION MAIT PD1+ − + MAIT KLRG1+ −−− + MAIT BCL2+ +++ ++ iNKT PD1+ +/− ++ iNKT KLRG1+ − − iNKT BCL2+ +++ ++ CYTOTOXICITY MAIT GzB+ + +++ iNKT GzB+ + + CYTOKINES Th1 MAIT IFNg+ +++ ++ MAIT TNFa+ ++ +++ iNKT IFNg+ ++ − iNKT TNFa+ ++ + Th2 MAIT IL4+ −−− +++ iNKT IL4+ +/− +/− Th17 MAIT IL17+ − + iNKT IL17+ − −

[0114] Based on those results it can be hypothesized that the decrease of MAIT or iNKT cell frequency observed in the T1D patients and more specifically in the Children with recent onset T1D could be partially explained by their strong activation, leading in turn to their exhaustion.

[0115] We also examined the correlation between HbA1c and the proportions of CD25-positive iNKT or MAIT cells in children with recent onset T1D, but there is no correlation. It appears that in this population of patients with very early diagnosis and quite high HbA1c levels, the phenotype of interest does not seem affected by metabolic impairment. The same was observed in children with established T1D and adults with long-standing disease, which tend to have lower HbA1c levels compared to the recent onset patients. Thus, biomarkers identified by the inventors can bring more information on the phenotype of interest compared to HbA1c.

[0116] Given these data, the quantification of different population of iNKT and MAIT cells was used to discriminate control children, children with T1D recent onset and children with long standing T1D. The analysis focuses on pediatric population (81 pediatric subjects), itself composed of three sub-populations: [0117] a population of control, composed of healthy subjects (22 subjects) [0118] a population of newly diagnosed subjects (41 subjects) and [0119] a population of diseased subjects (23 subjects).

[0120] With this study it was possible to refine the discrimination carried out previously on cytometric parameters and identify biological parameters which allow the best discrimination of the three sub-populations.

[0121] A principal component analysis (PCA) was carried out on the cytometric parameters obtained from MAIT and iNKT cells of those 81 pediatric subjects (FIG. 8). This PCA shows the 86 individuals in the factorial design (abscissa=Can1 and ordered=Can2), with their calculated coordinates. The axis Can1 allows to discriminate one hand sick children (negative values) and secondly controls children and recent diagnosis (positive values). The axis Can2 allows complete discrimination, children newly diagnosed located in the upper right quadrant (positive values, see: Table 5) and the child controls in the lower right quadrant (negative values). Long lasting T1D children are placed themselves between the two groups, close to zero.

TABLE-US-00004 TABLE 4 Class Means on Canonical Variables STATUT Can1 Can2 T1D childen −6.768282991 −0.620392530 T1D recent onset 1.970238152 2.224445621 control children 3.855804189 −3.018923118

[0122] Can1 and Can2 are based on two new variables also called component, each made as the linear combination of several cytometric parameters. The main factors of Can1 and Can2 are presented in Table 5.

TABLE-US-00005 TABLE 5 Raw Canonical Coefficients Label Can1 Can2 MAIT CD25+ 140.6887881 −34.7992235 MAIT CD25+ PD1+ −836.668862 −355.4912666 MAIT CD25+ PD1− −547.1229447 −195.3184607 MAIT CCR6+ CD25+ CD56− PD1+ −380.5641653 −134.9921849 iNKT CCR6−CD25+CD161+ −81.7954022 −112.6308856 MAIT CCR6+ CD25+ CD56− PD1− −62.8455747 −59.6811366 MAIT CCR6− CD25+ PD1− −163.9407468 −51.5567586 MAIT CCR6− CD25+ CD56− PD1− 193.1015985 −39.0970365 MAIT CD25+ CD56− PD1+ 495.9664932 72.7252498 MAIT CCR6+ CD25+ CD56− −133.1624776 77.4361266 iNKT CCR6−CD25+CD161+PD1− 81.8785878 112.9572917 iNKT CCR6−CD25+CD161+PD1+ 81.3180599 114.1039554 MAIT CCR6+ CD25+ PD1− 467.9625711 195.4004921 MAIT CD25+ CCR6− 601.7952535 276.8133444 MAIT CCR6+ CD25+ PD1+ 764.7454156 342.202342 MAIT CCR6− CD25+ CD56− −426.3194369 46.5641551 MAIT CD25+ CD56− PD1− 198.1796168 −16.9507561

[0123] From Table 5, MAIT and iNKT populations which are highly correlated with recent T1D onset (Can2) or long lasting T1D (Can1) can be identified. It seems that cell populations defined by specific markers combination are more relevant for such discrimination.

[0124] Moreover, the analysis of the correlation of each variable with the factor axis, show that the most relevant variables on Can2 (for control vs recent onset discrimination) % MAIT (−22%), iNKT CD25+(19.1%), MAIT CCR6− PD1+(19.2%), iNKT CD25+PD1+(18.8%).

[0125] Such multifactorial analysis, based on several cytometric parameters of MAIT and iNKT cells, allow an efficient discrimination of the three pediatric populations. The representation of individuals in the factorial design (FIG. 8) is very relevant. Variable % MAIT (highest correlation with Can2 axis, and p-value=0.0003), allows to discriminate at best newly diagnosed patients and control patients; healthy individuals tend to have a higher % of MAIT individuals recently diagnosed.

[0126] Hence it seems that the study of MAIT or iNKT cells and more particularly the study of CD25, CCR6 or PD1 subpopulation is highly relevant for early diagnosis of T1D.

[0127] In order to propose a tool for the risk evaluation or early diagnosis of T1D patient, the inventors have constructed a logistic regression model based of cytometric parameters and status of patients (use of PROC LOGISTIC (SAS 9.3). A method of backward selection was used. Starting from 55 parameters used in the Discriminant Analysis, the selection method backward leads to a model that includes an intercept and variable MAIT CCR6, MAIT CD56, MAIT CCR6+CD56+, MAIT CCR6− CD56−, MAIT CD25+PD1+ and MAIT CD25+CD56− PD1+, all significant at 5% (Table 6).

TABLE-US-00006 TABLE 6 Wald Cytometric parameter DF Chi-SquarePr> ChiSq MAIT CCR6 1 5.4091 0.0200 MAIT CD56 1 5.1571 0.0232 MAIT CCR6+ CD56+ 1 5.1552 0.0232 MAIT CCR6− CD56− 1 5.4440 0.0196 MAIT CD25+ PD1+ 1 5.8562 0.0155 MAIT CD25+ CD56− PD1+ 1 5.9922 0.0144

[0128] The ROC curve, plotted in FIG. 9, assessed the performance of the model to classify subjects based on their status (recent/healthy). The area under the curve (AUC, Area Under Curve) is 0.9788 or very close to the maximum value of 1. This curve shows the sensitivity (proportion of diagnoses classified diagnosed ordinate) versus 1−specificity (proportion of control subjects classified diagnosed).

[0129] The model confusion matrix (for a median threshold s=0.5) is presented in Table 7.4 of 53 subjects (7.5%) were misclassified: 2 and 2 control subjects newly diagnosed subjects.

TABLE-US-00007 TABLE 7 STATUT Frequency Predicted Statut Row Pct T1D recent onset control children Total T1D recent onset 31 2 33 93.94 6.06 control children 2 18 20 10.00 90.00 Total 33 20 53 Frequency Missinq = 10

[0130] Furthermore, from the model coefficients (Table 8), it is possible to diagnose a new patient, according to this data.

TABLE-US-00008 TABLE 8 Variable Coefficient Value Intercept α −11040.61 MAIT CCR6 β.sub.1 110.368 MAIT CD56 β.sub.2 105.802 MAIT CCR6+ CD56+ β.sub.3 −105.708 MAIT CCR6− CD56− β.sub.4 111.147 MAIT CD25+ PD1+ β.sub.5 −96.8976 MAIT CD25+ CD56− PD1+ β.sub.6 154.676

[0131] Such diagnosis can be made using the following logistic regression based formula, with 71 the probability of a subject to be diagnosed sick:

[00001]$\mathrm{logit}\left(\pi \right)=\log \left(\frac{\pi }{1-\pi }\right)=\alpha +{\beta }_1\times \mathrm{MAIT}\mathrm{CCR}6+{\beta }_2\times \mathrm{MAIT}\mathrm{CD}56+{\beta }_3\times \mathrm{MAIT}\mathrm{CCR}{6}_{+}\mathrm{CD}{56}_{+}+{\beta }_4\times \mathrm{MAIT}\mathrm{CCR}{6}_{-}\mathrm{CD}{56}_{-}+{\beta }_5\times \mathrm{MAIT}\mathrm{CD}{25}_{+}\mathrm{PD}{1}_{+}+{\beta }_6\times \mathrm{MAIT}\mathrm{CD}{25}_{+}\mathrm{CD}{56}_{-}\mathrm{PD}{1}_{+}$

[0132] Noting X′β the previous result

[00002]$\pi =\frac{\exp \left(X^{\prime }\beta \right)}{1+\exp \left(X^{\prime }\beta \right)}$

[0133] For example, two hypothetical patients were defined (Table 9). Given their respective values for the values of the model, the patient would be diagnosed patient A (π≥0.6), and the patient would be diagnosed healthy B (π<0.6).

TABLE-US-00009 TABLE 9 Sujet Variable Valeur X′β π Pa- MAIT CCR6 95 1.042 0.73932 tient MAIT CD56 10 A MAIT CCR6+ CD56+ 10 MAIT CCR6− CD56− 5 MAIT CD25+ PD1+ 0 MAIT CD25+ CD56− PD1+ 0 Pa- MAIT CCR6 95 −331.459 <0.0001 tient MAIT CD56 20 B MAIT CCR6+ CD56+ 20 MAIT CCR6− CD56− 2 MAIT CD25+ PD1+ 0 MAIT CD25+ CD56− PD1+ 0

[0134] Such model have also been used on a data set which has not been used for the regression model construction (Table 10).

TABLE-US-00010 TABLE 10 Statut Variables Values X′β π T1D recent onset MAIT CCR6+ 97.7 8.8087808 0.99985061 MAIT CD56+ 17.7 MAIT CCR6+ CD56+ 17.5 MAIT CCR6− CD56− 2.16 MAIT CD25+ PD1+ 0.062 MAIT CD25+ CD56− PD1+ 0.062 T1D recent onset MAIT CCR6+ 94.5 367.497568 1 MAIT CD56+ 11 MAIT CCR6+ CD56+ 11 MAIT CCR6− CD56− 5.48 MAIT CD25+ PD1+ 8.22 MAIT CD25+ CD56− PD1+ 7.53 control MAIT CCR6+ 98.6 −13.94301 8.8029E−07 MAIT CD56+ 40.1 MAIT CCR6+ CD56+ 40 MAIT CCR6− CD56− 1.17 MAIT CD25+ PD1+ 0 MAIT CD25+ CD56− PD1+ 0 control MAIT CCR6+ 92.9 −2.9174316 0.05129855 MAIT CD56+ 33 MAIT CCR6+ CD56+ 32.2 MAIT CCR6− CD56− 6.23 MAIT CD25+ PD1+ 0.076 MAIT CD25+ CD56− PD1+ 0.076

[0135] Such results confirmed the effectiveness of the quantification of MAIT or iNKT cells and more particularly their subpopulation for the follow up of T1D pathology and more particularly it early diagnosis.

[0136] To conclude, dysregulation of iNKT and MAIT cells may occur time before diabetes diagnosis, during the prediabetes period and might modify the balance between the activating and inhibiting ligands of MAIT and iNKT cells in such a way that both cell types could be activated before the multiple seroconversion state.

EXAMPLE 2

[0137] Material and Methods

[0138] Human samples. Peripheral blood samples were obtained from control children and from T1D children admitted in the Pediatric Endocrinology department of Necker hospital, Paris, France at T1D onset (i.e. within 10 days from first insulin injection), or with established disease. None of the control children had a personal or familial history of T1D or autoantibodies associated with T1D. Non-inclusion clinical following parameters contain: infection during the admission and associated others autoimmune disease. The Ethics Committee (Comité de protection des personnes (CPP) Ile-de-France) approved the clinical investigations and written informed consent was obtained from all the parents. For the Milan cohort, peripheral blood from healthy control subjects and patients at T1D onset (i.e., within 10 days from first insulin injection) were collected. The study was approved by the San Raffaele Hospital Ethic Committee (protocol: DRI-003). At risk subjects were enrolled in the Type 1 Diabetes TrialNet Pathway to Prevention Trial (TNO1 trial, former TrialNet Natural History Study). The overall objective of this study is to perform baseline and repeated assessments over time of the immunologic and metabolic status of individuals who are at risk for T1D (first/second degree relatives of patients with T1D). The study was approved by the San Raffaele Hospital Ethics Committee (protocol: NHPROTOCOL32803 TN01). Our local study was approved by the TrialNet Ancillary Studies Subcommittee. All analyses were performed blinded and all subjects included in this study signed the informed consent prior to blood donation.

[0139] Cell preparations. Human peripheral blood mono-nucleated cells (PBMC) of patients from Necker Hospital were isolated from fresh blood samples by Ficoll-Paque (Leucosep) or samples from San Raffaele hospital were defrosted in RPMI with 10% FCS (Fetal Cow Serum).

[0140] Flow cytometry and antibodies. Cells were stained in PBS containing 5% FCS and 0.1% azide. For human PBMC the following antibodies were used: CD3 (OKT3), CD4 (OKT4), Vα7.2 (3C10), CD161 (HP-3G10), CCR6 (G034E3), CD56 (HCD56), CD69 (FN50), Bcl-2 (100), IFN-γ (4S B3), IL17A (BL168), TNF-α (MAb11) mAbs from BioLegend; CD8 (SK1), PD1 (MIH4), CD25 (M-A251), IL4 (8D4-8), granzyme B (GB11) mAbs from BD Biosciences; CD3 (REA), CD161 (REA) mAbs from Miltenyi and CD4 (RPA-T4) mAb from eBiosciences. Data acquisition was performed using BD Biosciences LSRFortessa cytometer or FACS ARIA III cytometer for cells from patients from Necker hospital and Beckman Coulter Gallios for cells from patients from San Raffaele hospital.

[0141] In vitro cell stimulation. For ligand stimulation of human MAIT cells, 5-OP-RU solution was obtained after incubating 1 molar equivalent of 5-ARU with 2 molar equivalent of methylglyoxal (Sigma-Aldrich). Hela cells and PBMC from children controls and children with recent onset T1D were plated at a final concentration of 500 000 cells/mL each on 24-well plate, in RPMI supplemented with 10% FCS from children controls and children with recent onset T1D were plated at a final concentration of 106 cells/mL in RPMI supplemented with 10% FCS. PBMC were stimulated ON with various concentration of 5-OP-RU (0-5 nmol/L). Blocking MR1 (26.5 mAb) from Biolegend was added when indicated at 10 μg/mL. MAIT cell activation was analyzed by flow cytometry.

[0142] Intra-cellular staining. For human Bcl-2 staining, after surface staining lymphocytes were resuspended in fixation/permeabilization buffer (Foxp3 staining kit from eBioscience) and incubated at 4° C. in the dark then, cells were washed with PERM Wash buffer (eBioscience) and labeled with appropriate mAbs. For cytokine and granzyme B analysis of human MAIT cells, PBMC obtained from fresh samples, were analyzed after stimulation for 6 h at 37° C. in RPMI medium supplemented with 10% FCS with PMA (25 ng/ml) and ionomycin (1 μg/ml), in the presence of Brefeldin A (10 μg/ml).

[0143] Statistical analysis. For human studies, statistical tests between two groups were performed using two-tailed Mann-Whitney test and signed-rank Wilcoxon test with Graph Pad Prism. The Kruskal-Wallis test followed by the Wilcoxon rank sum test adjusted with the Holm method and the Spearman correlation test was applied for all the correlation analysis with Software R. A factorial discriminant analysis was performed using the XLSTAT 2016 Software. A logistic regression model was fitted with the PROC CANDISC software (SAS version p. 3) then a backward elimination procedure was applied. Prognostic validity of the model was evaluated by the receiver operating characteristic (ROC) curve analysis and measured using the area under the ROC curve (AUC). Statistical analyses were performed using the GraphPad Prism software version 5.00.288 and the R software version 3.2.3.

[0144] Results

[0145] Alteration of Blood MAIT Cell Frequency and Phenotype in Children with Recent Onset T1D

[0146] We first began the investigation of MAIT cells in T1D by analyzing MAIT cell frequency and phenotype in fresh peripheral blood samples from children with recent onset T1D and children with established T1D as compared to age-matched control children (data not shown). MAIT cells can be identified in human blood as CD4.sup.− T lymphocyte expressing Vα7.2 TCRα gene segment and CD161.sup.high (FIG. 10a). MAIT cell frequency and number was decreased (3-fold) in the blood of recent onset T1D children whereas no significant difference was observed in children with established disease as compared to control children (FIG. 10a and data not shown). Decreased frequency was observed in both CD8+ and double negative (DN) MAIT cell subsets (data not shown). Of note there was no difference in the frequencies of conventional CD4 and CD8 T cells, and of Vα7.2.sup.+CD161.sup.− T cells between the three children populations confirming that the decrease of MAIT cell frequency at the onset of T1D was not consecutive of changes in other T cell populations nor to down-regulation of the CD161 marker (data not shown). Analysis of MAIT cell phenotype showed a decreased frequency of MAIT cells expressing tissue recruitment/adhesion molecules (CCR6, CD56) at the onset of the disease, an increased frequency of MAIT cells expressing the activation/exhaustion markers CD25 and PD1, and a decreased frequency of MAIT cells expressing the anti-apoptotic molecule Bcl-2 (FIG. 10b and data not shown). Multi-parametric analysis of MAIT cells in the children with established T1D highlighted the intermediate phenotype of MAIT cells between those from recent onset T1D and control children (data not shown). Interestingly in recent onset children the frequency of MAIT cells expressing migratory CCR6.sup.+ or anti-apoptotic Bcl-2 molecules were positively correlated with the frequency of MAIT cells, whereas MAIT cell CD25 expression was negatively correlated with MAIT cell frequency (data not shown). These data suggest that decreased blood MAIT cell frequency could reflect their migration to inflamed tissues and/or their death by apoptosis subsequent to their activation.

[0147] Alteration of Blood MAIT Cell Function in Children with Recent Onset T1D

[0148] Cytokine and GzB production by fresh blood MAIT cells was analyzed after PMA-ionomycin stimulation. MAIT cells from children with recent onset T1D produced less IFN-γ, whereas their production of TNF-α, IL-4, and GzB was increased as compared with MAIT cells from control children (FIG. 11a and data not shown). Of note, among these effector molecules only the frequency of GzB correlated with the frequency of MAIT cells, the higher GzB production was observed in patients with the lower frequency of MAIT cells (data not shown). Multi-parametric analysis of cytokines and GzB production by MAIT cells also showed an intermediate status of blood MAIT cells from children with established T1D, between those from control children and recent onset T1D children, as already observed for MAIT cell surface phenotype (data not shown). We next analyzed the ability of MAIT cells to respond to specific TCR activation by the ligand 5-OP-RU. Upon stimulation, MAIT cells from control children up-regulated CD69 and CD25 activation markers. Addition of blocking MR1 mAb confirms that this activation was TCR-dependent. Interestingly, MAIT cell activation was significantly reduced in children with recent onset T1D (FIG. 11b). Together, these results highlight functional alteration of MAIT cells in children with recent onset T1D.

[0149] Association Between MAIT Cell Alterations and Clinical Characteristics

[0150] We next investigated potential links between phenotype and functional alterations of MAIT cells and clinical characteristics of children with recent onset T1D (data not shown). Interestingly, the frequency of GzB.sup.+ MAIT cells was negatively associated (r=−0.71, P<0.0001) with children's age at diagnosis (FIG. 12a), which is in agreement with the current view that T1D is more aggressive in the youngest children. We speculate that production of GzB by MAIT cells, reflecting their cytotoxic potential, could be involved in the physiopathology of T1D. Other MAIT cell parameters, such as their frequency, CCR6 and Bcl-2 expression, were also associated with the age at diagnosis (data not shown). No significant correlations between MAIT cell parameters and age were observed in controls and children with established T1D (data not shown). Production of GzB by MAIT cells inversely correlated with HbA1c level at the onset of the disease but not in children with established T1D (FIG. 12a). Indeed, a more aggressive disease associated with sustained MAIT cell abnormalities suggests a shorter time of hyperglycemia before the onset thereby lower levels of HbA1c.

[0151] To further explore the link between MAIT cell parameters and clinical characteristics, 15 of the children with recent onset T1D were further analyzed one year later. Although MAIT cell frequency among T cells remain to a similar level after one year of insulin, both CCR6+ and Bcl-2.sup.+ MAIT cell frequencies significantly increased. Conversely the frequencies of CD25+, PD1+, IL-17A.sup.+, and to some extent GzB.sup.+, MAIT cells decreased to levels observed in the control children (FIG. 12b). This longitudinal analysis strengthened the data obtained in the transversal analysis showing that MAIT cell phenotype was more similar to controls in the children under insulin therapy than at disease onset.

[0152] MAIT Cells as a New Biomarker of T1D

[0153] Canonical analysis was performed to compare MAIT cell alterations in the three groups of children (controls, recent onset and established T1D). This analysis revealed that MAIT cell parameters were sufficient to discriminate the three groups of children analyzed (FIG. 12c). Moreover a statistical regression analysis identified four surface markers that define a predictive model for the diagnosis of the disease tested on the ROC curve (FIG. 12d).

[0154] Importantly, we confirmed in another cohort of children from Milano that frequency and phenotype alteration of MAIT cell was observed in recent onset T1D as compared to age-matched control children (data not shown). However technical difficulties impacting CD56 analysis on these frozen cells did not allow to applying the predictive model.

[0155] Finally, to test whether MAIT cell alterations could be detected before the onset of diagnosis, we characterized MAIT cells in adults at risk to develop T1D defined as direct relatives of T1D patients with at least two positive autoantibodies (data not shown). As compared to the control group, there were increased frequencies of CD25.sup.+ and PD1.sup.+ MAIT cells and a trend toward a lower CCR6.sup.+ MAIT cell frequency, even though the number of individuals analyzed were limited (FIG. 12e). Altogether our data in patients suggest that MAIT cells represent a new biomarker in T1D and they could play a role in the pathogenesis of T1D. Therefore we investigated MAIT cells in mouse models, which allow analysis in tissues at different stages of disease development and could be manipulated to determine whether MAIT cells are involved in T1D physiopathology.

REFERENCES

[0156] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.