SEARCH

SEARCH BY CITATION

Keywords:

  • adiponectin;
  • C-peptide;
  • paediatric;
  • residual beta cell function;
  • type 1 diabetes

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

The progression of type 1 diabetes after diagnosis is poorly understood. Our aim was to assess the relation of disease progression of juvenile-onset type 1 diabetes, determined by preserved beta cell function the first year after diagnosis, with systemic cytokine concentrations and number of autoantibodies. Juvenile patients (n = 227) had a meal-stimulated C-peptide test 1 and 6 months after diagnosis. On the basis of the C-peptide course for the duration of 1–6 months, four progression groups were defined: patients with persistently low beta cell function (‘stable-low’), rapid progressers, slow progressers and remitters. Serum concentrations of adiponectin, interleukin (IL)-1ra, inducible protein 10 (IP-10), IL-6 and glutamic acid decarboxylase (GAD), IA-2A and islet-cell antibodies (ICA) were measured at 1, 6 and 12 months. We found that adiponectin concentrations at 1 month predicted disease progression at 6 months (P = 0·04). Patients with low adiponectin had a higher probability of becoming remitters than rapid progressers, odds ratio 3·1 (1·3–7·6). At 6 and 12 months, adiponectin differed significantly between the groups, with highest concentrations among stable-low and rapid progressers patients (P = 0·03 and P = 0·006). IL-1ra, IP-10 and IL-6 did not differ between the groups at any time-point. The number of autoantibodies differed significantly between the groups at 1 month (P = 0·04), where rapid progressers had the largest number. There was no difference between the groups in human leucocyte antigen-associated risk. We define progression patterns distinguishing patients diagnosed with low beta cell function from those with rapid decline, slow decline or actual increase in beta cell function, pointing to different mechanisms of disease progression. We find that adiponectin concentration at 1 month predicts, and at 6 and 12 months associates with, distinct progression patterns.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

Type 1 diabetes mellitus is a T cell-mediated autoimmune disease [1]. Little information is available on the different patterns of type 1 diabetes progression following diagnosis, particularly in the paediatric population. In the present study, we define disease progression on the basis of changes in beta cell function as assessed by stimulated C-peptide from 1 to 6 months after clinical onset of type 1 diabetes. Stimulated C-peptide has been described as the best parameter for measuring preserved beta cell function [2]. We hypothesize that patients can have different patterns of loss in beta cell function, varying from rapid to slow, stable or even actual increases in beta cell mass. This may be partly immune-mediated, and the mechanism between stably preserved and increased C-peptide production after diagnosis may differ. Systemic cytokines, which are known to play a role in the autoimmune process leading to destruction of beta cells [1,3], could be potential biomarkers of different patterns of disease progression. In this study we investigate whether adiponectin, interleukin-1 receptor antagonist (IL-1ra), interferon (IFN)-γ, inducible protein 10 (IP-10/CXCL10) and IL-6 could be regarded as such. Adiponectin is considered to be anti-diabetic in type 2 diabetes patients [4]. It has anti-inflammatory effects and improves insulin sensitivity [5–7]. However, several studies have shown that adiponectin serum concentrations are higher in both adults and children with type 1 diabetes than in healthy controls, independent of body mass index (BMI) [8–10]. IL-1ra is the natural antagonist to IL-1β, which is thought to contribute to beta cell destruction in both type 1 and type 2 diabetes. IP-10 interacts with T cells via IFN-γ and CXCR3 to recruit T helper type 1 (Th1) cells to inflamed tissues [11]. Some studies have shown that serum level of IP-10 is significantly higher in patients with newly diagnosed type 1 diabetes compared to controls [12,13], while others dispute this [14,15]. IL-6 is a proinflammatory cytokine. It affects insulin signalling and has an insulin-independent role in glucose disposal [16].

In a recent study we investigated associations between various cytokines, including adiponectin, IL-6 and IL-1ra and endogenous beta cell function and remission status, defined as HbA1c <7·5% and <0·4 units/kg daily insulin or HbA1c <6·5% and <0·4 units/kg daily insulin [17]. The objective of the current longitudinal investigation was to describe distinct patterns of disease progression after diagnosis in patients with newly diagnosed type 1 diabetes on the basis of changes in stimulated C-peptide. We tested if the different patterns of progression correlated with the new measure of remission, insulin dose adjusted HbA1c (IDAA1c), proposed recently by the Hvidøre Study Group [18]. Furthermore, we investigated if the patterns of disease progression are confounded by parameters such as HbA1c, insulin dose, ketoacidosis at diagnosis, BMI or age, and influenced by serum concentrations of potential biomarkers such as IL-1ra, adiponectin, IP-10 and IL-6, human leucocyte antigen (HLA) genes or by the number of autoantibodies, glutamic acid decarboxylase (GAD)A, IA-2A and islet-cell antibodies (ICA).

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

Subjects

The Hvidøre Remission Phase Study is a prospective, long-term observational study conducted in 18 centres representing 15 countries in Europe and Japan. Between August 1999 and December 2000, 275 patients were recruited to the study. Exclusion criteria were non-type 1 diabetes, decline of enrolment into the study by the patients or the parents or patients treated initially outside the participating departments for more than 5 days. Type 1 diabetes was diagnosed according to the World Health Organization's (WHO) criteria. The study was performed according to the Helsinki II declarations, and approved by the local ethical committee in each participating country [19]. In the present study we included 227 (83%) of the patients from the Hvidøre cohort, who had complete data on stimulated C-peptide from both 1 and 6 months after diagnosis. Of the included patients, 85% of the patients studied were white Caucasians, 116 girls and 111 boys; median age at diagnosis was 9·6 (range 0·2–16·3) and median BMI was 17·1 kg/m2 (range 12·3–33·1). Of the patients included, 20·3% presented with diabetic ketoacidosis (HCO3≤15 mmol/l and/or pH ≤7·3) at diagnosis (Table 1).

Table 1.  Clinical characteristics of patients from the four progression patterns groups.
Progression pattern groupNumberGender (male/female)Age at diagnosisBMI (kg/m2) at 1 monthKetoacidosis (%) at diagnosis
All patients227111/1169·617·120·3
(0·2–16·3)(12·3–33·1) 
Stable-low2313/108·816·338·1
(0·7–14·7)(12·6–23·0) 
Rapid progressers11758/599·116·821·4
(1·1–16·3)(12·3–33·1) 
Slow progressers5424/3010·917·412·5
(3·1–14·6)(13·0–27·6) 
Remitters3316/179·418·124·1
(0·2–15·4)(12·6–23·0) 
 HbA1c%Insulin µ/kg/24hIDAA1c
1 month6 months12 months1 month6 months12 months1 month6 months12 months
  1. Age and body mass index (BMI) are shown as median (with range, min–max). Ketoacidosis is defined as HCO3 < 15 and/or pH < 7·3. Age and ketoacidosis are determined at diagnosis while BMI is determined 1 month after diagnosis. HbA1c and insulin dose/kg are shown as median with range. IDAA1c: insulin dose adjusted HbA1c.

All patients8·97·17·70·450·530·7110·909·2310·57
(6·2–12·4)(4·9–11·6)(4·9–16·4)(0·07–2·69)(0·04–2·22)(0·05–1·94)(6·67–22·26)(5·59–16·69)(5·81–19·48)
Stable-low8·97·28·050·540·380·5611·279·3310·74
(6·2–12·4)(4·9–10·7)(4·9–11·9)(0·07–1·1)(0·14–1·52)(0·23–1·12)(6·76–15·8)(5·59–13·48)(5·81–15·47)
Rapid progressers8·17·47·60·440·560·7410·639·6010·95
(6·3–11·8)(5·0–11·6)(5·6–12·2)(0·08–2·96)(0·20–2·22)(0·19–1·94)(7·18–22·26)(6·32–16·69)(7·28–16·54)
Slow progressers9·16·97·50·450·440·6911·418·4410·07
(6·9–11·8)(5·2–11·0)(5·9–16·4)(0·14–1·14)(0·04–0·96)(0·05–1·39)(7·55–15·63)(5·81–14·86)(6·49–19·48)
Remitters8·56·47·30·570·410·6511·378·319·86
(6·5–11·9)(4·9–9·2)(5·7–10·8)(0·11–1·3)(0·07–1·00)(0·15–1·24)(7·80–15·26)(5·99–11·82)(7·52–12·96)

C-peptide

After 1, 6 and 12 months of diabetes a liquid meal challenge was utilized to stimulate endogenous C-peptide release. The test was performed in the morning after at least 8 h of fasting. The morning insulin dose was given after the test. A dose of 6 ml/kg (max 360 ml) of Boost/Sustacal (Mead Johnson, Evansville, Indiana, USA; 237 ml = 8 fl oz contains 33 g carbohydrate, 15 g protein and 6 g fat, 240 kcal total) was ingested in less than 10 min. In agreement with the Diabetes Control and Complications Trial (DCCT) protocol, capillary glucose was measured at time 0 and venous C-peptide and glucose at 90 min after the ingestion of Boost. Serum samples were labelled and frozen at −20°C until sent to the Steno Diabetes Centre, Gentofte, Denmark on dry ice for the determination of C-peptide. Samples were thawed only once for radioimmunoassay (RIA) determination. Serum C-peptide was analysed using a fluoroimmunometric assay (AutoDELFIA™ C-peptide; PerkinElmer Life and Analytical Sciences Inc., Turko, Finland). The sensitivity was below 1 pmol/l, intra-assay coefficient of variation below 6% at 20 pmol/l and recovery of standard, added to plasma before extraction, about 100% when corrected for losses inherent in the plasma extraction procedure. Due to ethical reasons the test was not performed at diagnosis.

Autoantibodies

GADA, IA-2A and ICA were measured 1, 6 and 12 months after diagnosis. Patients without antibodies at 1 month were considered antibody-negative. The autoantibodies were detected centrally by methods described previously [19]. The results were expressed as relative units (RU).

HLA genes

Typing of the HLA-class II DRB1 locus was performed by direct sequencing of exon 2 of DRB1 according to the Immuno Histocompatibility Working Group [20]. DRB1*03/04 and DRB1*04/04 were defined as high-risk genotypes (n = 87), while DRB1*03/03 and DRB1*04/08 were considered moderate risk (n = 29). All other genotypes were classified as low risk (n = 109).

Cytokines

Blood for cytokine measurement was drawn 90 min after ingestion of the Boost test. All cytokines were detected centrally in Düsseldorf, Germany. IL-1ra was measured by multiplex-bead technology using commercially available kits (Fluorokine MAP; R&D Systems, Wiesbaden, Germany) [17], IL-6 and total adiponectin (low, middle and high molecular weight) was measured by enzyme-linked immunoassay (ELISA) [17,21], while IP-10 was measured by sandwich ELISA [22]. The samples were treated in a blinded fashion, meaning that no data for the patients were available when the serum was tested for cytokine concentrations.

The detection limits for the assays were 6·7 pg/ml for adiponectin, 13·6 pg/ml for IL-1ra, 2 pg/ml for IP-10 and 0·15 pg/ml for IL-6. Determination of cytokine concentrations lower than the detection limit were assigned a value of half the detection limit (adiponectin, n = 0; IL-1ra, n = 0; IP-10, n = 13; IL-6, n = 7). The immunoassays showed interassay variations <20% and intra-assay variations <10%.

Insulin dose adjusted HbA1c

Insulin dose adjusted HbA1c (IDAA1c) is a new surrogate measure of endogenous insulin production presented recently by Mortensen et al. on behalf of the Hvidøre Study group. IDAA1c is calculated on basis of the actual insulin dose and HbA1c as HbA1c (%) + [4 × insulin dose (units/kg per 24 h)][18].

Definition of type 1 diabetes progression patterns

The progression of type 1 diabetes was determined on basis of the change in stimulated C-peptide from 1 to 6 months after diagnosis, which could be an increase, a decrease or stable C-peptide level. When calculating the change in C-peptide, we acknowledged that minor changes due to measurement errors might occur and should be interpreted with caution. Therefore a change in C-peptide should exceed 50 pmol/l on an absolute scale, and the change should be of at least 20% between the largest and the smallest value of a relative scale. A shortcoming of this definition was that the group of patients with a stable level of C-peptide consisted of both patients with stable-low and stable-high C-peptide values. We dealt with this notion by staging patients on the basis of C-peptide values above 100 pmol/l at 1 month. Patients with values below 100 pmol/l were considered to be C-peptide-negative. Hence, four courses of C-peptide change were defined: (1) patients eliciting stable but low C-peptide production (‘stable-low’); (2) patients losing more than 20% C-peptide during the first 6 months after diagnosis (‘rapid progressers’); (3) patients losing less than 20% C-peptide (‘slow progressers’); and (4) patients with an increase of more than 20% in C-peptide production after diagnosis (‘remitters’). Five patients with C-peptide below 100 pmol/l, 1 month after diagnosis, had a considerable increase (ranging from 320 to 370 pmol/l) in C-peptide concentrations by the 6-month measurement. Despite their low C-peptide value at 1 month, we concluded that they had an increasing level of insulin production and the patients were assigned to the remitter group.

Statistical methods

Parametric and non-parametric tests of variance were used to investigate differences in clinical parameters among the C-peptide progression groups.

The autoantibodies were grouped according to the number of autoantibodies present (from none to three different islet autoantibodies); χ2 analysis was used to investigate the association of and HLA risk groups and the number of autoantibodies with the progression groups. Specific autoantibodies titres were log-transformed and regression analyses, including age and gender as dependent variables, were used to test the association with the progression groups. Simple regression analyses, including age and gender as dependent variables, were used to test for an association between IDAA1c and the progression groups.

Logistic regression was used to analyse whether cytokine concentration at 1 month could predict progression pattern group after 6 months. Only the rapid progresser, slow progresser and remitter groups were included in these analyses, because patients from the stable-low group, by definition, have stable C-peptide levels below 100 pmol/l at both 1 and 6 months. The cytokine concentrations were transformed logarithmically to obtain a normal distribution and then divided into quartiles. The analyses therefore included progression pattern groups as the dependent variable and quartiles of log cytokines, age, gender and log C-peptide measured at 1 month as the explanatory variables in an ordinal logistic regression model. Multiple regression analyses were used to investigate the relations of adiponectin, IL-1ra, IP-10 and IL-6 at 6 and 12 months with the C-peptide progression groups. The analyses included log cytokines as the dependent variable and C-peptide progression groups, age and gender as explanatory variables. Additional analyses including BMI as explanatory variable were performed for each cytokine for all the time-points. Finally, the possible influence of daily insulin dose (U/kg) on adiponectin was tested by simple regression. Associations are descriptive and were not corrected for multiple testing. Statistical analysis was performed using SAS version 9·1 (SAS Institute, Gary, NC, USA). A P-value ≤0·05 was considered significant.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

C-peptide progression patterns groups

On the basis of the change in C-peptide level from 1 to 6 months after diagnosis, four progression groups were made: (i) stable-low, (ii) rapid progressers, (iii) slow progressers and (iv) remitters. From the total number of patients included, 10·2% (n = 23) were stable-low, 51·5% (n = 117) were rapid progressers, 23·8% (n = 54) were slow progressers and 14·5% (n = 33) were remitters (Fig. 1).

image

Figure 1. Progression patterns of type 1 diabetes on the basis of the change in stimulated C-peptide from 1 to 6 months. Four progression groups were defined – stable-low: C-peptide negative patients at 1 month (<100 pmol/l; blue); rapid progressers: patients with a decrease in stimulated C-peptide from 1 to 6 months (black); slow progressers: patients with stable high stimulated C-peptide (red); and remitters: patients with increasing C-peptide from 1 to 6 months (green). (C-peptide in pmol/l).

Download figure to PowerPoint

There were no significant differences between the progression groups regarding ethnicity, gender, BMI, HbA1c, daily insulin and ketoacidosis at diagnosis. The groups differed significantly in age (P = 0·04); patients from the stable-low group were the youngest, followed by patients from the rapid progresser and remitter groups, while patients from the slow progresser group were the oldest. At 6 months there were a significant difference in HbA1c (P = 0·002) and daily insulin requirement (P = 0·001), where rapid progressers had the highest HbA1c and the highest insulin dose compared to the other groups. At 12 months, the insulin dose of rapid progresser patients was still the highest (P = 0·003), while there was no difference in HbA1c compared to the other patients (Table 1).

HLA risk genes and type 1 diabetes progression

There was no relation between HLA low-, moderate- or high-risk genes and the progression groups. All HLA risk combinations were represented in each for the progression groups (Table 2).

Table 2.  Distribution of human leucocyte antigen (HLA) risk genes among patients.
HLA genes Patients (n)Low riskModerate riskHigh risk
  1. Shown is the number of patients from the progression pattern groups with low-risk, moderate risk and high-risk HLA genes. There was no difference in HLA-associated risk between the four groups.

Stable-low9 (41%)3 (14%)10 (45%)
Rapid progresser57 (49%)13 (11%)46 (40%)
Slow progresser26 (48%)7 (13%)21 (39%)
Remitter17 (52%)6 (18%)10 (30%)

Autoantibodies and type 1 diabetes progression

There was a significant difference between the number of autoantibodies present (0, 1, 2 or 3) in patients from the four progression groups measured at 1 month (P = 0·04). Among patients from the rapid progresser group, 45·7% had three autoantibodies present, while only 30·4%, 28·3% and 33·3% of patients from the stable-low, slow progresser and remitter groups had three autoantibodies present, respectively. In contrast, 37·7% of slow progressing patients and 33·3% of remitting patients had only one autoantibody present, as opposed to 13% of the stable-low and 25% of the rapid progressing patients (Table 3).

Table 3.  Number of autoantibodies present among patients.
Autoantibodies Patients (n)0123
  1. Shown is the number of patients from the progression pattern groups with 0, 1, 2 or 3 autoantibodies present 1 month after diagnosis. There was a significant difference between number of autoantibodies present between the four groups; P = 0·04.

Stable-low2 (9%)3 (13%)11 (48%)7 (30%)
Rapid progresser15 (13%)29 (25%)19 (16%)53 (46%)
Slow progresser4 (8%)20 (38%)14 (26%)15 (28%)
Remitter3 (9%)11 (33%)8 (25%)11 (33%)

There were no differences in the number of autoantibodies present at 6 or 12 months after diagnosis. Titres of specific autoantibodies did not differ between the groups at 1, 6 or 12 months (data not shown), although there was a trend that titres of GADA were highest among patients from the stable-low and rapid progresser groups at all time-points. All three autoantibodies were present in all the four progression groups.

IDAA1c and type 1 diabetes progression

There was no association between IDAA1c and progression patterns 1 month after diagnosis, but after 6 and 12 months there was a significant association (P < 0·0001 and P = 0·04, respectively). Rapid progresser patients had the highest IDAA1c, followed by stable-low and slow progresser patients, while remitter patients had the lowest values at both 6 and 12 months (Table 1).

Adiponectin, IL-1ra, IP-10, IL-6 and type 1 diabetes progression

Adiponectin.  There was no association between adiponectin and daily insulin dose (U/kg) at 1, 6 or 12 months after diagnosis. Logistic regression analysis of 1-month data showed that the ability of adiponectin measured at 1 month to predict which progression group a patient would be in after 6 months was significant (P = 0·05). This became even more significant when BMI was included in the analyses (P = 0·04). Patients with adiponectin concentrations within the highest quartile (ranging from 19 053·78 to 53 343·33 pg/ml) had a significantly higher probability of being rapid progressers after 6 months, while patients with adiponectin concentrations in the lowest quartile (ranging from 1898·32 to 9100·00 pg/ml) had a significantly higher probability for being remitters after 6 months. The odds ratio (OR) for patients with adiponectin within the lowest quartile for becoming a remitter was 3·1 (1·3–7·6) compared to becoming a rapid progresser. At 6 months, adiponectin concentration differed significantly between the groups (P = 0·03), which remained significant when BMI was included in the model (P = 0·02). Patients from the stable-low group had the highest amount of adiponectin among the groups, which differed significantly from the slow progresser group (P = 0·013) and the remitter group (P = 0·01) (Fig. 2a). At 12 months the adiponectin concentrations remained different between the four progression groups (P = 0·006 and P = 0·009 when BMI was included), again with highest levels in the stable-low group and lowest in slowly progressing (P = 0·02) and remitting patients (P = 0·005). Rapidly progressing patients had the second highest level of adiponectin, which also differed significantly from patients from the remitter group (P = 0·007) and slow progresser group (P = 0·04) (Fig. 2a).

image

Figure 2. Shown are the median values of adiponectin, interleukin (IL)-1ra, inducible protein 10 (IP-10) and IL-6 concentrations (pg/ml) of patients on the y-axis, which represents an arithmetic scale: stable-low (blue), rapid progressers (black), slow progressers (red) and remitters (green). Multiple regression analysis was used to analyse the cytokine levels. There was a significant difference in the serum level of adiponectin between the four progression groups at both 6 and 12 months after diagnosis; *P = 0·03 and **P = 0·006.

Download figure to PowerPoint

IL-1ra.  The same approach was used to analyse data on IL-1ra. Levels at 1 month did not reveal any predictive value of IL-1ra (P = 0·25), which was not affected by BMI. Analysis both including and excluding BMI with 6 (P = 0·21 and P = 0·11) and 12 months' data (P = 0·36 and P = 0·32) did not reveal any difference between the progression pattern groups (Fig. 2b). However, a trend of a higher IL-1ra concentration in the remitting group compared to the stable-low group at 6 months was noted (P = 0·07) (Fig. 2b).

IP-10.  Analysis of 1-month data revealed that IP-10 could not predict progression group after 6 months (P = 0·59); this was not changed by BMI. Analysis of 6- and 12-month data did not show any differences between the groups (Fig. 2c).

IL-6.  IL-6 measured at 1 month could not predict progression group 6 months (P = 0·14); this was not affected by BMI. Analyses of 6- and 12-month data showed a trend of difference between the progression groups (P = 0·09 and P = 0·08, respectively). Patients from the remitter group had the highest level at both time-points (Fig. 2d).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

We present a new strategy to define patterns of disease progression in type 1 diabetes on the basis of changes in stimulated C-peptide from 1 to 6 months after diagnosis. The rate of remission peaks 3 months after diagnosis [23]; it was speculated that delta C-peptide from 1 to 6 months would give a clearer picture of variation of disease progression than delta C-peptide from 1 to 12 months. The progression pattern groups we define are not confounded by HbA1c, BMI, ketoacidosis and gender. However, age influenced the progression patterns; stable-low and rapid progressing patients were the youngest. This is in agreement with the majority of studies, which have shown that younger age is associated with greater loss of beta cell function both before and after diagnosis [19,24,25]; however, a study did not find any association [26]. At 6 months, patients from the rapid progresser group had significantly higher HbA1c levels and insulin dose compared to the other patients. This could indicate that rapid progresser patients were more insulin-resistant.

The four progression pattern groups associate with IDAA1c at 6 and 12 months. IDAA1c is a measure for residual beta cell function and, as expected, the patients from the stable-low and rapid progressing groups displayed the highest values and patients from the slow progresser and remitter groups the lowest. An explanation for the missing association with the 1-month IDAA1c could be that HbA1c level at 1 month still reflects glycaemia before diagnosis. The relation to IDAA1c can serve as proof for the potential value of the progression pattern groups. To understand the factors and mechanisms influencing the course of loss in beta cell function following diagnosis of type 1 diabetes, it is important to have a clear definition of progression rather than simply looking at current definitions of clinical remission [27–29].

It is of great interest that HLA genotypes do not influence the progression groups. This is in agreement with previous reports, where no association between residual beta cell function and HLA genes was found [19,24,30]. However, here we show that rate of beta cell loss from 1 to 6 months is not influenced by HLA haplotypes.

In the present study, we tested if the cytokines adiponectin, IL-1ra, IP-10 and IL-6 could be potential biomarkers for the four patterns of progression. We also examined if there is an association between disease progression and the presence of combinations of autoantibodies. We selected these cytokines because they had proved to be informative in earlier studies [3,17], or because of their putative involvement in the immunopathogenesis or modulation thereof in type 1 diabetes. We have presented IL-1ra, IL-6 and adiponectin data previously, where we investigated the relation between the cytokines and remission status [17]. Here we investigate whether the cytokines have any influence on the progression pattern groups and if they could be potential biomarkers for disease progression as assessed by change in stimulated C-peptide.

The vast majority of newly diagnosed patients have autoantibodies against at least one islet antigen. In this study, differences in the number of autoantibodies (ICA, IA2A and GADA) were studied in relation to disease progression. Patients with rapid progressing type 1 diabetes displayed the largest number of different islet autoantibodies shortly after diagnosis, but this difference disappeared after 6 and 12 months. Titres of all three autoantibodies were represented in all four progression groups. From our study it is difficult to conclude whether number of autoantibodies present plays a role in disease progression after diagnosis, but we propose that the presence of a large variety of autoantibodies marks a severe autoimmune condition. This is in agreement with the previous reports, that antigen spreading speeds up the progression of type 1 diabetes prior to diagnosis [31].

Our results may indicate that cytokines influence the course of type 1 diabetes. When analysing cytokine data two models were used: one adjusted for age and gender and one where BMI was included, in addition, to test for a potential effect on the level of cytokine concentration by adipose tissue. The potential effect of insulin dose/kg on adiponectin was also tested. Adiponectin measured 1 month after diagnosis could predict future disease progression. Patients from the stable-low and rapid progresser groups had significantly higher adiponectin levels 6 and 12 months after diagnosis with type 1 diabetes than patients from the slow progresser and remitter groups. This confirms the notion that type 1 diabetes patients have increased adiponectin levels compared to healthy controls subjects, and in addition indicates that patients with ‘aggressive’ type 1 diabetes (progressing more rapidly) display increased adiponectin in serum after diagnosis. The increased concentrations remain stable throughout the first year. It remains unresolved why adiponectin, which is regarded as an anti-inflammatory cytokine, is elevated in type 1 diabetes. We speculate that, similar to type 2 diabetes where adiponectin has been shown to improve insulin sensitivity, adiponectin secretion increases in patients with a rapid progression of type 1 diabetes to boost remaining insulin action. Indeed, we find that patients who tend to be the most insulin-resistant (patients with the highest HbA1c and highest insulin dose) have the highest adiponectin levels. Also, we have found previously that adiponectin correlated positively with HbA1c in type 1 diabetes patients [17]. In that study we also reported a correlation between high adiponectin serum concentrations and low C-peptide production, but that report did not take into consideration changes in C-peptide over time, and therefore does not discriminate between stable disease (slow progression) and increased beta cell function (remission) [17]. To our knowledge, this is the first time an association of adiponectin and disease progression, as assessed by the change in stimulated C-peptide, has been shown. Also this is the first time that adiponectin has been shown to be able to predict future disease progression. We investigated monomeric adiponectin, which has been described to be effective [32]. Whether high molecular weight multimers of adiponectin would add or reveal different associations is not clear and is subject to debate.

IL-1ra is an anti-inflammatory cytokine and it is conceivable that high concentrations of IL-1ra associate with preserved beta cell function in type 1 diabetes. Indeed, IL-1ra concentrations tended to differ between the four progression pattern groups, being highest in the remitter group at all time-points. Recently, we reported an association between IL-1ra and improved C-peptide 1 month after diagnosis [17], but at 1 month IL-1ra could not predict the progression pattern group after 6 months.

IP-10 is a proinflammatory cytokine that is produced by distressed beta cells, and has been shown to be important in the initial phase of a Th1-dominated autoimmune process [33]. However, in the present study we could not show any association between IP-10 and the rate of beta cell loss.

Despite being a proinflammatory cytokine, IL-6 tended to be highest among remitter patients; this is in agreement with our previous finding [17]. This is unexpected; however, it appears to be consistent.

Our results support the importance of studying different patterns of progression of type 1 diabetes in detail after diagnosis, rather than simply determining clinical remission on the basis of insulin needs. We conclude that the change in stimulated C-peptide can be used to stage different kinetics of type 1 diabetes progression, which may prove valuable for understanding factors affecting residual beta cell function and preservation after diagnosis. Also, the change in stimulated C-peptide can be used to optimize insulin treatment or potential drug interventions, where the aim is to preserve residual beta cell function. From our study we conclude that adiponectin correlates with progression of type 1 diabetes the first 12 months after diagnosis. Adiponectin appears to have a prolonged correlation, even after considerable loss of beta cell function, whereas IL-1ra may have an influence on disease activity shortly after diagnosis, where most of the patients still have a preserved beta cell mass. Finally, we propose that staging patients on the basis of stimulated C-peptide production may be a valuable tool to define biomarkers reflecting different patterns and mechanisms of disease progression, and to evaluate intervention trials, and that adiponectin measured 1 month after diagnosis could be a valuable biomarker for predicting disease progression.

Disclosure

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

On behalf of all authors no conflicts of interest are to be disclosed.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

We thank Novo Nordisk for support throughout this study, with special thanks to Ralf W. Ackermann and Julie S. Hansen. We are grateful to the technicians Britta Drangsfeldt and Susanne Kjelberg at Steno Diabetes Centre for their assistance.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix
  • 1
    Atkinson MA, Eisenbarth GS. Type 1 diabetes: new perspectives on disease pathogenesis and treatment. Lancet 2001; 358:2219.
  • 2
    The Diabetes Control and Complications Trial (DCCT) Research Group. Effects of age, duration and treatmentof insulin-dependent diabetes mellitus on residual beta-cell function: observations during eligibility testing for the Diabetes Control and Complications Trial (DCCT). J Clin Endocrinol Metab 1987; 65:306.
  • 3
    Schloot NC, Hanifi-Moghaddam P, Abenhus-Andersen N et al. Association of immune mediators at diagnosis of Type 1 diabetes with later clinical remission. Diabet Med 2007; 24:51220.
  • 4
    Spranger J, Kroke A, Mohlig M et al. Adiponectin and protection against type 2 diabetes mellitus. Lancet 2003; 361:2268.
  • 5
    Kubota N, Terauchi Y, Yamauchi T et al. Disruption of adiponectin causes insulin resistance and neointimal formation. J Biol Chem 2002; 277:258636.
  • 6
    Huerta MG. Adiponectin and leptin: potential tools in the differential diagnosis of pediatric diabetes? Rev Endocr Metab Disord 2006; 7:18796.
  • 7
    Berg AH, Combs TP, Du X, Brownlee M, Scherer PE. The adipocyte-secreted protein Acrp30 enhances hepatic insulin action. Nat Med 2001; 7:94753.
  • 8
    Imagawa A, Funahashi T, Nakamura T et al. Elevated serum concentration of adipose-derived factor, adiponectin, in patients with type 1 diabetes. Diabetes Care 2002; 25:16656.
  • 9
    Frystyk J, Tarnow L, Hansen TK, Parving HH, Flyvbjerg A. Increased serum adiponectin levels in type 1 diabetic patients with microvascular complications. Diabetologia 2005; 48:191118.
  • 10
    Celi F, Bini V, Papi F et al. Circulating adipocytokines in non-diabetic and Type 1 diabetic children: relationship to insulin therapy, glycaemic control and pubertal development. Diabet Med 2006; 23:6605.
  • 11
    Rotondi M, Chiovato L, Romagnani S, Serio M, Romagnani P. Role of chemokines in endocrine autoimmune diseases. Endocr Rev 2007; 28:492520.
  • 12
    Nicoletti F, Conget I, Di MM et al. Serum concentrations of the interferon-gamma-inducible chemokine IP-10/CXCL10 are augmented in both newly diagnosed Type I diabetes mellitus patients and subjects at risk of developing the disease. Diabetologia 2002; 45:110710.
  • 13
    Shimada A, Morimoto J, Kodama K et al. Elevated serum IP-10 levels observed in type 1 diabetes. Diabetes Care 2001; 24:51015.
  • 14
    Rotondi M, Romagnani P, Brozzetti A et al. Serum concentrations of the interferon-alpha-inducible chemokine IP-10/CXCL10 are augmented in both newly-diagnosed Type I diabetes mellitus patients and subjects at risk of developing the disease. Diabetologia 45:11071110. Diabetologia 2003; 46:1020–1.
  • 15
    Rotondi M, Lazzeri E, Romagnani P, Serio M. Role for interferon-gamma inducible chemokines in endocrine autoimmunity: an expanding field. J Endocrinol Invest 2003; 26:17780.
  • 16
    Carey AL, Steinberg GR, Macaulay SL et al. Interleukin-6 increases insulin-stimulated glucose disposal in humans and glucose uptake and fatty acid oxidation in vitro via AMP-activated protein kinase. Diabetes 2006; 55:268897.
  • 17
    Pfleger C, Mortensen HB, Hansen L et al. Association of IL-1ra and adiponectin with C-peptide and remission in patients with type 1 diabetes. Diabetes 2008; 57:92937.
  • 18
    Mortensen HB, Hougaard P, Swift P et al. New definition for the partial remission period in children and adolescents with type 1 diabetes. Diabetes Care 2009; 32:138490.
  • 19
    Mortensen HB, Swift PG, Holl RW et al. Multinational study in children and adolescents with newly diagnosed type 1 diabetes: association of age, ketoacidosis, HLA status, and autoantibodies on residual beta-cell function and glycemic control 12 months after diagnosis. Pediatr Diabetes 2008.
  • 20
    Marcel GJ. Genomic analysis of the human MHC. DNA based typing for HLA alleles and linked polymorphisms. Tilanus IHWG technical manual. Distributed by the International Histocompatibility Working Group, IHWG Press c/o Fred Hutchinson Cancer Research Center, Seattle, WA 98109-1024, USA. 2001: ISBN number: 0-945278-02-02-0.
  • 21
    Muller S, Martin S, Koenig W et al. Impaired glucose tolerance is associated with increased serum concentrations of interleukin 6 and co-regulated acute-phase proteins but not TNF-alpha or its receptors. Diabetologia 2002; 45:80512.
  • 22
    Hanifi-Moghaddam P, Schloot NC, Kappler S, Seissler J, Kolb H. An association of autoantibody status and serum cytokine levels in type 1 diabetes. Diabetes 2003; 52:113742.
  • 23
    Martin S, Pawlowski B, Greulich B, Ziegler AG, Mandrup-Poulsen T, Mahon J. Natural course of remission in IDDM during 1st yr after diagnosis. Diabetes Care 1992; 15:6674.
  • 24
    Bonfanti R, Bognetti E, Meschi F et al. Residual beta-cell function and spontaneous clinical remission in type 1 diabetes mellitus: the role of puberty. Acta Diabetol 1998; 35:915.
  • 25
    Wallensteen M, Dahlquist G, Persson B et al. Factors influencing the magnitude, duration, and rate of fall of B-cell function in type 1 (insulin-dependent) diabetic children followed for two years from their clinical diagnosis. Diabetologia 1988; 31:6649.
  • 26
    Greenbaum CJ, Mandrup-Poulsen T, McGee PF et al. Mixed-meal tolerance test versus glucagon stimulation test for the assessment of beta-cell function in therapeutic trials in type 1 diabetes. Diabetes Care 2008; 31:196671.
  • 27
    Muhammad BJ, Swift PG, Raymond NT, Botha JL. Partial remission phase of diabetes in children younger than age 10 years. Arch Dis Child 1999; 80:3679.
  • 28
    Couper J, Donaghue K. Phases of diabetes. Pediatr Diabetes 2007; 8:447.
  • 29
    Sochett EB, Daneman D, Clarson C, Ehrlich RM. Factors affecting and patterns of residual insulin secretion during the first year of type 1 (insulin-dependent) diabetes mellitus in children. Diabetologia 1987; 30:4539.
  • 30
    Schiffrin A, Suissa S, Poussier P, Guttmann R, Weitzner G. Prospective study of predictors of beta-cell survival in type I diabetes. Diabetes 1988; 37:9205.
  • 31
    Von HM, Sanda S, Herold K. Type 1 diabetes as a relapsing–remitting disease? Nat Rev Immunol 2007; 7:98894.
  • 32
    Yamauchi T, Kamon J, Minokoshi Y et al. Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med 2002; 8:128895.
  • 33
    Christen U, McGavern DB, Luster AD, Von Herrath MG, Oldstone MB. Among CXCR3 chemokines, IFN-gamma-inducible protein of 10 kDa (CXC chemokine ligand (CXCL) 10) but not monokine induced by IFN-gamma (CXCL9) imprints a pattern for the subsequent development of autoimmune disease. J Immunol 2003; 171:683845.

Appendix

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Disclosure
  8. Acknowledgements
  9. References
  10. Appendix

Members of the Hvidøre Study Group on Childhood Diabetes who have contributed to the Remission Phase Study:

Henk-Jan Aanstoot MD, PhD, Erasmus University Medical Centre, Rotterdam, the Netherlands

Carine de Beaufort MD, PhD, Clinique Pédiatrique, Luxembourg

Francesco Chiarelli, Professor, MD, Clinica Pediatrica, Chieti, Italy

Knut Dahl-Jørgensen, Professor, MD, Dr Medical. SCI and Hilde Bjørndalen Göthner MD, Ullevål University Hospital, Department of Paediatrics, Oslo, Norway

Thomas Danne MD, Charité, Campus Virchow- Klinikum, Berlin, Germany

Patrick Garandeau MD, Unité D'endocrinologie Diabetologie Infantile, Institut Saint Pierre, France

Stephen A. Greenek MD, University of Dundee, Scotland

Reinhard W. Holl MD, University of Ulm, Germany

Mirjana Kocova, Professor, MD, Pediatric Clinic-Skopje, Republic of Macedonia

Pedro Martul MD, PhD, Endocrinologia Pediatrica Hospital De Cruces, Spain

Nobuo Matsuura MD, Kitasato University School of Medicine, Japan

Henrik B. Mortensen MD, Dr Med. SCI, Department of Pediatrics, Glostrup University Hospital, Denmark

Kenneth J. Robertson MD, Royal Hospital for Sick Children, Yorkhill, Glasgow, Scotland

Eugen J. Schoenle MD, University Children's Hospital, Zurich, Switzerland

Peter Swift MD, Leicester Royal Infirmary Children's Hospital, Leicester, UK

Rosa Maria Tsou MD, Paediatric Department Oporto, Portugal

Maurizio Vanelli MD, Paediatrics, University of Parma, Italy

Jan Åman MD, PhD, Örebro Medical Centre Hospital, Department of Paediatrics, Sweden