Association of Urine Haptoglobin With Risk of All-Cause and Cause-Specific Mortality in Individuals With Type 2 Diabetes: A Transethnic Collaborative Work

Patients with type 2 diabetes have a higher standardized mortality ratio compared with people without diabetes (1,2). Although cardiovascular disease (CVD) is the leading cause of premature death in this population, the presence of type 2 diabetes is also associated with increased risk of mortality attributable to noncardiovascular causes such as infectious disease, cancer, and renal disease (3).

Haptoglobin is an acute-phase reactant that was initially identified as a scavenger of free hemoglobin released from red blood cells after hemolysis (4,5). In this setting, haptoglobin protects blood vessels and kidney proximal tubules from oxidative injury caused by iron overload (6). Besides hemoglobin clearance, haptoglobin is also involved in the regulation of inflammation and immune response (7). It was shown to bind and sequester proinflammatory mediator high mobility group box 1 and confer protection against polymicrobial sepsis in a mouse model (8). The increase in haptoglobin also was shown to repress expression of proinflammatory mediator tumor necrotic factor α in a cellular model (9). In alignment with its role in the regulation of inflammatory response, plasma haptoglobin level may increase two- to fivefold in the presence of inflammatory stimuli (10). Additionally, haptoglobin may stimulate proliferation and differentiation of vascular endothelial cells as a compensation for ischemia (11). Hence, it has also been considered as an angiogenic factor for vascular remodeling (12).

Noteworthy, the level of haptoglobin in biofluid has been associated with clinical outcomes in both populations with diabetes and populations without diabetes. For example, several studies have reported that a high level of urinary haptoglobin may predict progression of kidney disease in patients with type 2 diabetes (13–15). A high level of plasma haptoglobin has been associated with increased risk of mortality in patients with acute cerebral infarction or acute myocardial infarction (16,17). However, to our knowledge, the relationship between urine haptoglobin and risk of mortality has not been reported. In the current work, we aimed to study whether the level of urine haptoglobin may predict risk of mortality independent of traditional risk factors including renal function and albuminuria in patients with type 2 diabetes. Of interest, ethnicity has been shown to be a risk factor for mortality in patients with type 2 diabetes, as illustrated in the UK Prospective Diabetes Study (UKPDS) (18). Haptoglobin was previously studied in various ethnic backgrounds (13,15), but the association of this biomarker with mortality has not been scrutinized. This prompted us to adopt a transethnic approach, with two independent cohorts, one of Asian origin (Singapore) and one of European origin (France), to address this knowledge gap.

For the Singapore cohort, a total of 2,061 people with type 2 diabetes and aged >21 years were recruited consecutively from outpatient clinics in a regional hospital in Singapore between March 2004 and December 2015 (19). The inclusion criterion was type 2 diabetes diagnosed by attending physicians according to prevailing American Diabetes Association criteria. Exclusion criteria were pregnancy, clinically manifest infectious disease, cancer and autoimmune diseases with active treatment, and end-stage renal disease (defined as baseline estimated glomerular filtration rate [eGFR] <15 mL/min/1.73 m² or receipt of kidney transplantation). The study was approved by the Singapore National Healthcare Group ethics committee, and written consent has been obtained from each participant.

SURDIAGENE (SURvie, DIAbete de type 2 et GENEtique) is a single-center prospective cohort of patients with type 2 diabetes regularly visiting the Diabetes Department at Poitiers University Hospital in France. Participants with established diagnosis of type 2 diabetes for at least 2 years were enrolled consecutively from 2002 to 2012 (20). Outcome updates were performed every 2 years since 2007. Of 1,468 participants in the SURDIAGENE cohort, 1,438 with available urine samples were included in current study. The Poitiers University Hospital Ethics Committee approved the design and execution of the study. All participants have given their written informed consent.

For identification of all-cause and cause-specific mortality in the Singapore cohort, participant data were linked with the Singapore national death registry by the unique national registration identification number. Data in the death registry available for linkage were for up until July 2016, which was taken as the censor date for the current study. Death might be considered as 100% captured because it was mandatory by law to register all deaths within Singapore. Specific cause of each death was identified by the ICD code on death certificates, which were issued by authorized clinical practitioners (ICD-9-CM before 1 January 2012 and ICD-10-CM thereafter).

In the French SURDIAGENE cohort, living status and cause of deaths were determined from patients’ hospital records, interviews with their general practitioners, and inquiry to the French national death registry. Participants were followed from baseline until death or 31 December 2015—whichever came first. Causes of death were adjudicated by an independent committee and classified according to ICD-9 or ICD-10.

For improvement of consistency on analyses of cause-specific deaths across two transethnic cohorts, categorization of causes of deaths was a priori determined according to a previous study in a Singapore cohort (21). Deaths were categorized as due to CVD, malignant tumor, infection, and renal causes in both cohorts. When death was due to a cause that was noncardiovascular, nonrenal, noninfectious, or a nonmalignant tumor, we used the term “other cause.”

Urine samples were collected after overnight fasting and stored at −80°C in both cohorts. Samples from the Singapore and SURDIAGENE participants were handled by the Singapore Khoo Teck Puat Hospital biobanking facility and the CHU Poitiers biobanking facility (CRB BB 0033-00068), respectively. Urine haptoglobin level was quantified by the same researcher using the same sandwich enzyme immunoassay kit (R&D Systems, Minneapolis, MN) for both cohorts. Undiluted urine samples were used for the assays except for samples with readings above the upper detection limit (200 ng/mL). These samples (∼20% of total) were requantified after a five-time dilution. The lowest detection limit of urine haptoglobin was 0.07 ng/mL. Readings below this level were imputed as 0.07 ng/mL. The intra-assay coefficient of variation (CV) was 2.4–3.7% and interassay CV was 5.6–7.4% in the range of 18–116 ng/mL. Urine haptoglobin concentration was normalized to 1 mg/dL urine creatinine level before data analysis. Samples used for the current studies underwent only one freeze-thaw cycle.

In the Singapore cohort, HbA_1c was measured by immunoturbidimetric method (COBAS INTEGRA 800 Analyzer; Roche, Basel, Switzerland). Creatinine, triacylglycerol, and HDL and LDL cholesterol were quantified by enzymatic methods (cobas c system; Roche Diagnostic GmbH, Mannheim, Germany). The Chronic Kidney Disease-Epidemiology Collaboration (CKD-EPI) (2009) formula was used to calculate eGFR (22). Urinary albumin was measured by solid-phase competitive chemiluminescent enzymatic immunoassay (IMMULITE; DPC, Gwynedd, U.K.). Albuminuria was expressed as albumin-to-creatinine ratio (ACR).

In the SURDIAGENE cohort, HbA_1c was measured using a high-performance liquid chromatography method (HA-8160 analyzer; Menarini, Florence, Italy). Plasma concentrations of triacylglycerol and HDL and LDL cholesterol were measured using enzymatic methods. Serum and urine creatinine and urinary albumin were measured by colorimetry and immunoturbidimetry assays (cobas analyzer; Roche Diagnostics GmbH), respectively. The eGFR was also calculated using the CKD-EPI equation.

Data analyses on association of urine haptoglobin and mortality risk were performed by two biostatisticians in Singapore and France independently, with no merged data across two cohorts. Urinary haptoglobin concentrations were ranked from lowest to highest, and the thresholds for quartile 1 (Q1)–Q2, Q2–Q3, and Q3–Q4 were defined by the values corresponding to the 25th, 50th, and 75th percentiles of the distribution in each cohort, respectively. Baseline characteristics of participants were presented across urine haptoglobin quartiles.

Differences across groups were compared by Student t test, one-way ANOVA, Mann-Whitney U test, Kruskal-Wallis test, or χ² test where appropriate. Spearman correlations were used to assess the binary relationship between urine haptoglobin and clinical variables. A stepwise linear regression model was employed to identify clinical and biochemical determinants of urine haptoglobin variance at a cross-sectional level. Natural log-transformed urine haptoglobin was the dependent variable, while clinical and biochemical variables were entered as independent variables.

Primary outcome was all-cause mortality, and secondary outcome was cause-specific mortality. Cumulative death rate stratified by urine haptoglobin quartiles was plotted and log-rank test was applied to examine the differences across quartiles. Cox proportional hazards regression models were used to study the associations of urine haptoglobin with all-cause and cause-specific mortality. Time to death was outcome. Urine haptoglobin level was the main predictor, while selection of covariates was based on biological plausibility. Five sets of models were fitted: univariate (model 1); adjusted for age, sex, ethnicities (Singapore cohort only, Malay, Asian Indians vs. Chinese), and active smoking (model 2); adjusted for the same variables as model 2 plus diabetes duration, BMI, systolic blood pressure, HbA_1c, and HDL and LDL cholesterol (model 3); adjusted for the same variables as model 3 plus baseline eGFR and urine ACR, as they represent established key markers associated with renal outcome and mortality (model 4); and adjusted for the same variables as model 4 plus usage of insulin, renin-angiotensin system (RAS) blockers, and statins (model 5). Proportionality assumptions were examined by Schoenfeld residuals.

Harrell’s c statistic was used to assess the additive prediction value of urine haptoglobin for all-cause mortality above traditional risk factors. Given that area under ROC curve (AUC) assesses risk rankings instead of clinically important relative difference in risk magnitude, we estimated continuous net reclassification improvement (cNRI) and relative integrated discrimination improvement (rIDI) using 1,000 × bootstrap (23,24). The outcome was considered significant if the lower end of 1,000 × bootstrap estimate was >0. We included age, sex, active smoking, ethnicity (Singapore cohort only), diabetes duration, BMI, HbA_1c, systolic blood pressure, HDL cholesterol, LDL cholesterol, baseline eGFR, and urine ACR as traditional risk factors in the base model.

Data analyses were performed using STATA 14 (StataCorp, TX) or SAS 9.4 (SAS Institute, Cary, NC). A two-sided P value <0.05 was considered statistically significant.

In the Singapore cohort, mean ± SD age of participants was 58 ± 12 years. Diabetes duration was 11 ± 9 years. The proportion of Chinese, Malay, and Asian Indian ethnicity was 63%, 20%, and 17%, respectively. Characteristics of participants according to haptoglobin quartile distribution are shown in Table 1. Urine haptoglobin concentration was modestly correlated with eGFR (ρ = −0.22, P = 0.001) and moderately correlated with ACR (ρ = 0.55, P < 0.001 in the full cohort; ρ = 0.35, P < 0.01 in subpopulation with ACR ≥300 µg/mg). A stepwise linear regression model suggested that HbA_1c, LDL cholesterol, urine ACR, and Asian Indian ethnicity were the main determinants of urine haptoglobin variations. These four variables collectively explained 32.5% of urine haptoglobin variance.

Similar to observations in Singapore cohort participants, French participants with a higher level of urine haptoglobin were older and had a longer diabetes duration, higher systolic blood pressure, a significantly lower eGFR, and higher ACR (Table 1).

Urine haptoglobin concentration was modestly correlated with eGFR (ρ = −0.22, P < 0.001) and moderately correlated with ACR (ρ = 0.56, P < 0.0001 in the full cohort; ρ = 0.31, P < 0.0001 in subpopulation with ACR ≥300 µg/mg). Linear regression model showed that HbA_1c, LDL cholesterol, and urine ACR explained 7.8% of urine haptoglobin variance.

In both Singapore and SURDIAGENE cohorts, there was no significant between- sex difference in urine haptoglobin level (P = 0.27 and P = 0.96, respectively). We therefore pooled male and female participants in each cohort in the following analyses.

In the Singapore cohort, during a median follow-up of 7.5 years (interquartile range 3.5–12.8) (10,909 patient-years in total), 365 deaths were registered. Participants with a higher level of urine haptoglobin had a higher risk of all-cause mortality (log-rank test P < 0.0001) (Fig. 1).

As shown in Table 2, participants with urine haptoglobin in quartiles 2–4 had higher risk for all-cause mortality (unadjusted hazard ratio [HR] 1.47 [95% CI 1.02–2.11], 2.28 [1.62–3.21], and 4.64 [3.39–6.35], respectively) compared with those in quartile 1. The association remained significant in quartile 4 after multiple adjustments (adjusted HR 1.68 [1.15–2.45]). Complementarily, a higher level of urine haptoglobin was significantly associated with a higher risk of all-cause mortality when it was analyzed as a continuous variable. In the fully adjusted model, a 1-SD increment in natural log-transformed urine haptoglobin was associated with 1.34-fold (95% CI 1.16–1.55) increased hazards for all-cause mortality.

In the SURDIAGENE cohort, during a median follow-up of 7.2 years (interquartile range 4.6–10.8) (10,650 patient-years in total), 525 deaths were registered. Similar to observations in the Singapore cohort, participants in SURDIAGENE cohort with a higher level of urine haptoglobin had a higher risk of all-cause mortality (log-rank P < 0.0001) (Fig. 1). Of note, the log-linearity assumption for the Cox proportional hazards regression model was not met in the SURDIAGENE study. Therefore, urine haptoglobin was analyzed only as a categorical variable with quartile 1 as reference. As shown in Table 2, HR for all-cause mortality in quartile 4 versus quartile 1 was 2.67 (95% CI 2.09–3.42). The association remained statistically significant after adjustment for multiple traditional risk factors including renal function, albuminuria, and medication usage (model 5, adjusted HR 1.49 [95% CI 1.14–1.96]).

In the Singapore cohort, of 365 deaths, 158 (43%) were attributable to CVD, 62 (17%) to infection, 60 (16%) to renal causes, and 49 (13%) to malignant tumor. In univariate analysis, a higher level of urine haptoglobin was associated with an increased risk of mortality attributable to CVD, infection, and renal causes but not a malignant tumor when urine haptoglobin was analyzed as either a categorical or a continuous variable. In multivariate analysis, the association of urine haptoglobin with mortality attributable to CVD and infection remained statistically significant. A 1-SD increment in urine haptoglobin was associated with 1.26-fold (95% CI 1.01–1.56) increased hazards for CVD mortality and 1.90-fold (95% CI 1.28–2.81) increased hazards for mortality attributable to infectious diseases after adjustment for multiple clinical risk factors including eGFR and ACR. With urine haptoglobin as a categorical variable, participants in quartile 4 had a significantly higher risk for mortality attributable to infectious disease (adjusted HR 3.33 [95% CI 1.07–10.4]) in the full model. In addition, they had a significantly higher risk of CVD mortality after adjustment for demographic and cardiometabolic risk factors (HR 3.43 [95% CI 2.03–5.79]). The association did not remain statistically significant after further adjustment for eGFR, ACR, and use of medications (adjusted HR 1.73 [95% CI 0.95–3.14]) (Table 3).

In the SURDIAGENE cohort, of 525 deaths, 295 (56%) were attributable to CVD, 49 (9%) to infection, 77 (15%) to a malignant tumor, 6 (1%) to renal causes, and 98 (19%) to other causes. In univariate analysis, SURDIAGENE participants with urine haptoglobin in quartile 4 had a significantly higher risk of mortality attributable to CVD and infection but not malignant tumor. In multivariate analysis, participants with urine haptoglobin in quartile 4 had 1.64-fold (95% CI 1.13–2.38) higher hazards for CVD mortality after adjustment for multiple risk factors compared with participants in quartile 1. The association of urine haptoglobin with death attributable to infectious disease did not remain statistically significant after adjustment for traditional risk factors, although the point estimates of HR were increased in those with urine haptoglobin levels in quartiles 3 and 4 (Table 3).

In the Singapore cohort, adding urine haptoglobin onto the base model derived from clinical risk factors significantly improved cNRI (0.17, 95% CI 0.10–0.28), but there were only minimal changes in rIDI (0.003, 95% CI −0.001 to 0.009) and AUC (0.800 to 0.804).

In the SURDIAGENE cohort, adding urine haptoglobin onto the base model significantly improved both cNRI (0.23, 0.07–0.33) and rIDI (0.028, 0.003–0.065). Adding urine haptoglobin onto base model derived from traditional risk factors only modestly improved AUC (0.746 to 0.751).

In a transethnic approach with one cohort from Singapore (Asia) and one from France (Europe), we found that a higher level of urine haptoglobin was prospectively associated with an increased risk of all-cause mortality as well as mortality attributable to CVD and infection in patients with type 2 diabetes. Remarkably, the prediction of urine haptoglobin for mortality risk was consistent in both cohorts and was found on top of established risk factors including albuminuria, as suggested by the integrated discrimination improvement and net reclassification improvement analyses. However, the modest increment in AUC suggests that the additional prognostic value of urinary haptoglobin needs to be established in future studies. Altogether, our data suggest that urine haptoglobin may be a novel risk marker of mortality in patients with type 2 diabetes. To our knowledge, the current work is the first to report the relationship between urine haptoglobin and risk of mortality, although plasma haptoglobin has been associated with survival after acute myocardial infarction or cerebral infarction in patients without diabetes (16,17). Despite the large disparities in terms of race/ethnic background, baseline characteristics, and propensity for different types of diabetes complications (25), the associations of urine haptoglobin with all-cause mortality, CVD mortality, and mortality attributable to infection were consistently observed in both cohorts.

The pathophysiologic links between high level of urine haptoglobin and increased risk of mortality in patients with type 2 diabetes remain to be elucidated. Haptoglobin is an acute-phase protein that is elevated in response to increased free hemoglobin, inflammation, and oxidative stress (6,7,10). A high level of urine haptoglobin may reflect severe systemic oxidative stress and inflammation and therefore increased risk of mortality. On the other hand, a high level of urine haptoglobin has been associated with development and progression of diabetic kidney disease (13,14). It has been known that kidney disease is an important driver of both CVD and non-CVD mortality in patients with diabetes (26). Hence, it is reasonable to postulate that the kidney impairment may be the linkage between a high urine haptoglobin and increased risk of mortality. This hypothesis is partly supported by the substantial attenuation in the prognostic value of urine haptoglobin when eGFR and ACR were added in multivariable models.

CVD is the main cause of mortality in patients with type 2 diabetes (27). Indeed, 43% and 56% of deaths were attributable to CVD in our Asian and European participants, respectively. The association of urine haptoglobin with CVD mortality is novel but not unexpected. Similar to albuminuria, a high level of urine haptoglobin may be a biomarker of systemic microvascular dysfunction, which is a driver of excess CVD risk (15,28). In addition, increments in urine haptoglobin may suggest higher levels of oxidative stress, inflammation, and probably endothelial dysfunction (4). All of these pathophysiologic processes may eventually contribute to a higher risk of CVD mortality.

Patients with diabetes are susceptible to infectious disease (29,30). Our finding that urine haptoglobin predicts mortality due to infection is also novel. Previous studies have reported associations of haptoglobin with microvascular disease (15), impaired immunity (31), presence of inflammation response (8,12), and chronic and acute kidney disease (14,32). All of these factors may render patients with a high level of urine haptoglobin more susceptible to infectious diseases, less likely to recover from infection, and, hence, at increased risk of mortality.

The source of urine haptoglobin should also be discussed. Haptoglobin in urine may originate from leakage of plasma haptoglobin through the impaired basement membrane. However, urine haptoglobin and albuminuria are only moderately correlated (ρ = 0.55 and 0.56 in two cohorts, respectively), even in patients with macroalbuminuria (ρ = 0.35 and 0.31 in two cohorts, respectively). Also, the associations of urine haptoglobin with mortality are independent of albuminuria in both cohorts. Hence, it is unlikely that urine haptoglobin solely originates from leakage of plasma haptoglobin. An early preclinical study has shown that acute kidney injury may result in sustained activation of haptoglobin expression in proximal tubules accompanied by an increased haptoglobin excretion in urine (32). Hence, it may be reasonable to postulate that urine haptoglobin is at least partly expressed de novo in the kidney and excreted into urine. Thus, further studies are warranted to address this question.

The main strength of the current work is its transethnic design with inclusion of one Asian and one European cohort of type 2 diabetes with contrasting ethnic backgrounds. Both cohorts have adequate sample size, prospective follow-up, and relatively large numbers of deaths. Hence, our data may be considered as robust, reproducible, and probably generalizable. We have taken into account smoking, dyslipidemia, blood pressure, glycemic control, renal function, albuminuria, and use of some important medications in analyses. The association of urine haptoglobin with mortality remains strong even after we account for these traditional risk factors. Nevertheless, several important weaknesses should be mentioned. Firstly and importantly, plasma haptoglobin level and haptoglobin genotype, especially the latter, have been associated with adverse clinical outcomes in populations with diabetes (33,34). Unfortunately, data on these two variables were not available in both cohorts. Further studies are warranted to address the relationships among haptoglobin genotype, urine haptoglobin, plasma haptoglobin, and their associations with mortality risk in patients with type 2 diabetes. Secondly, as this is an observational study, residual confounding for the association between urine haptoglobin and mortality risk is inevitable. Although we have considered demographic, cardiometabolic risk factors, renal function, and albuminuria, some important determinants of mortality risk such as physical activity, dietary pattern, and socioeconomic status are not available in the current analysis. Thirdly, the association of urine haptoglobin with renal death cannot be validated because of the small number of renal deaths in the French cohort. Similarly, the multivariate analyses on mortality due to some specific causes might be underpowered due to the relatively small event numbers. Additionally, due to the nature of the study, we are unable to elucidate the pathophysiologic mechanisms underlying the association of a high level of urine haptoglobin and increased mortality risk.

In conclusion, the level of urine haptoglobin predicts risk of mortality independent of albuminuria and other traditional risk factors, consistently, in a transethnic approach with Asian and European participants. Our data suggest that urine haptoglobin may potentially be a novel biomarker for mortality risk in patients with type 2 diabetes, opening new research avenues to understand the pathophysiology behind this finding.

Acknowledgments. The authors thank Singapore National Registry of Disease Office for identification of deaths by database linkages (S17-S0035). All patients included and followed in the SURDIAGENE study are warmly thanked for their participation in this research. Their general practitioners are acknowledged for their help in collecting clinical information. The authors thank Sonia Brishoual (Biological Resources Center, BRC BB-0033-00068, Poitiers, France) for biochemical assays and Alexandre Pavy and Marie-Claire Pasquier (Information Technology Department, CHU Poitiers, Poitiers, France) for data management. The staff of the Diabetes Department at CHU Poitiers in France are acknowledged for their help with data collection and monitoring.

Funding. This work was supported by Khoo Teck Puat Hospital STAR II grants (grants 17202 and 18203). The SURDIAGENE cohort was supported by grants from the French Ministry of Health (PHRC-Poitiers 2004, PHRC-IR 2008), Association Française des Diabétiques (Research Grant 2003), Groupement pour l’Etude des Maladies Métaboliques et Systémiques (GEMMS Poitiers, France), and Société Francophone du Diabète (Bourse de recherche 2017 [to P.-J.S.]).

The funders listed above had no roles in study design, data collection, analysis, writing, or decision to publish.

Duality of Interest. S.H. has served as a consultant or on advisory panels for AstraZeneca/Bristol-Myers Squibb/Valbiotis; has received honorarium or speaking fees from AstraZeneca/Bristol-Myers Squibb, Abbott, Boehringer Ingelheim, Eli Lilly, Janssen, Merck Sharp & Dohme, Novartis, Novo Nordisk, Sanofi, Servier, and Takeda; has received research grants from Abbott, Dinno Santé, and Takeda; and has received travel grants from Janssen, AstraZeneca/Bristol-Myers Squibb, Merck Sharp & Dohme, and Sanofi. No other potential conflicts of interest relevant to this article were reported.

The analysis and interpretation of the data were carried out without the participation of any of these organizations and companies.

Author Contributions. J.-J.L. and S.H. designed the Singapore and SURDIAGENE studies, respectively. S.L., P.-J.S., E.G., R.W.M.C., and R.L.G. acquired and researched data. J.-J.L. drafted the manuscript. J.-J.L., S.L., P.-J.S., E.G., R.W.M.C., R.L.G., S.H., and S.C.L. critically revised the manuscript, contributed important intellectual knowledge, and approved publication of the work. S.H. and S.C.L. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Center Organization. Recruiting physicians: Samy Hadjadj (coordinator), Frédérique Duengler, Louis Labbé, Aurélie Miot, Xavier Piguel, Florence Torremocha, Pierre-Jean Saulnier, Richard Maréchaud. Gérard Mauco (Department of Biochemistry, CHU Poitiers) and Thierry Hauet (INSERM U1082, CHU Poitiers) are acknowledged for helping with biochemical assays.

Adjudication Procedure. Case inquiry: Samy Hadjadj (coordinator), Sonia Brishoual, Céline Divoy, Cécile Demer, Aurélie Miot, Xavier Piguel, Florence Torremocha, Nathalie Fauvergue, Séverin Cabasson, Pierre-Jean Saulnier, Philippe Sosner, Ahmed Amine Kasmi. Local coordination: Stéphanie Ragot (coordinator and biostatistician). Independent adjudication committee of SURDIAGENE: Jean Michel Halimi (Chairman, Tours, France), Gregory Ducrocq (Paris, France), Ronan Roussel (Paris, France), Pierre Llatty (Poitiers, France), Vincent Rigalleau (Bordeaux, France), Charlotte Hulin (Poitiers, France), David Montaigne (Lille, France), Philippe Zaoui (Grenoble, France).