We investigated endotrophin, a profibrotic signaling molecule reflecting collagen VI formation, in serum and urine as risk marker for complications to type 2 diabetes.
Endotrophin was measured in 774 individuals with type 2 diabetes. Outcomes included a composite kidney end point, first major adverse cardiovascular event (MACE), mortality, progression of albuminuria, incident heart failure, and sight-threatening eye disease. Adjusted Cox proportional hazards models were applied.
Doubling of serum endotrophin was associated with the kidney end point (n = 49; hazard ratio 1.80 [95% CI 1.13–2.87]), first MACE (n = 66; 1.54 [1.04–2.28]), mortality (n = 156; 1.69 (1.31–2.19]), and incident heart failure (n = 42; 1.63 [1.02–2.60]). A doubling of urine endotrophin was associated with progression of albuminuria (n = 85; 1.20 [1.04–1.39]).
Serum endotrophin was a risk marker for mortality and kidney and cardiovascular complications in type 2 diabetes. Urine endotrophin was a marker for albuminuria progression.
Introduction
Hyperglycemia accelerates oxidative stress–mediated inflammation and other tissue-injuring processes (1,2), which leads to fibrosis (3,4) that compromises the function of visceral organs. Since the formation and degradation of extracellular matrix (ECM) proteins, such as collagens, are closely linked to development of fibrosis (5,6), the quantification of extracellular matrix remodeling through biomarker measurements may identify individuals with high profibrotic activity at risk for progressing to impaired organ function (7). Endotrophin is released during the synthesis of collagen type VI (7,8) and has paracrine proinflammatory and profibrotic effects (9). In this large cohort study, we investigate whether a higher level of serum or urine endotrophin is associated with risk of progression of complications to type 2 diabetes.
Research Design and Methods
Methods are described in detail in Supplementary Material. In short, 774 individuals with type 2 diabetes were recruited from Steno Diabetes Center Copenhagen from 2012 to 2016. Persons on dialysis (n = 14) were excluded. All individuals gave written informed content to the original study (StenoDot). The current study complies with the Declaration of Helsinki and was approved by The Regional Ethics Committee in The Capital Region of Denmark (H-19042436). Endotrophin levels were determined with measurement of the surrogate marker, PRO-C6, using the competitive ELISA PRO-C6 (Nordic Bioscience; Herlev, Denmark), with duplicate measurements on all samples (10). Urine endotrophin levels were normalized to urinary creatinine levels. Demographic and clinical data including retinal photo gradings and routine laboratory measurements were extracted from patient records. Information on hospital admissions, emigration, and deaths was obtained from national registers. Primary end points were 1) a composite kidney end point defined as estimated glomerular filtration rate (eGFR) decline ≥40% confirmed after minimum 1 month or unconfirmed if the measurement was the last before end-of-follow-up, development of chronic kidney disease stage 5, chronic dialysis, kidney transplantation, or kidney failure as cause of death; 2) first major adverse cardiovascular event (MACE) including cardiovascular death, nonfatal acute myocardial infarction, coronary intervention, and nonfatal stroke; and 3) mortality. Secondary end points were 1) progression in albuminuria group based on minimum one measurement, 2) incident heart failure, and 3) incident sightthreatening diabetic eye disease including proliferative retinopathy and maculopathy based on retinal photos or procedural codes. For each outcome, we excluded participants previously diagnosed with the outcome. ICD-8 and ICD-10 and procedural codes are provided in Supplementary Table 1.
Statistics
Baseline clinical characteristics were stratified by quartiles of serum and urine endotrophin levels and tested for trends across quartiles. Continuous variables are presented as means ± SD if normally distributed and otherwise as median (quartile [Q1:Q3]). Categorical data are presented as percentage. Correlation between serum and urine endotrophin was assessed with Spearman rank correlation coefficient. In the longitudinal analyses, participants were followed until first event or censoring due to emigration, death, or end-of-follow-up. Serum and urine endotrophin were log2-transformed before Cox proportional hazards models were applied. Adjustment included sex, age, diabetes duration, BMI, LDL cholesterol, smoking, HbA1c, systolic blood pressure, eGFR, and urinary albumin excretion (UAE) (except for analyses on albuminuria progression) at baseline. The proportional hazards assumption was tested with inclusion of time-dependent covariates in the Cox model. Two-sided P values <0.05 were considered statistically significant, and analyses were performed with SAS 9.4 (SAS Institute, Cary, NC).
Data and Resource Availability
Data are available on reasonable request to the corresponding author.
Results
Serum samples were available for 767 individuals, but 3 were excluded due to an inaccurate test result (coefficient of variation >25%) for serum endotrophin, and urine samples were available for 713, of whom 1 was excluded due to a coefficient of variation >25%. The cohort with serum samples included 252 (33%) females; mean ± SD age was 65 ± 12 years, diabetes duration 17 ± 8.8 years, and eGFR 76 ± 24 mL/min/1.73 m2; and median (Q1:Q3) UAE was 12.5 (5.5:73.5) mg/g or g/24 h. Median follow-up ranged from 4.1 years for the kidney end point and MACE to 6.0 years for incident heart failure. Baseline characteristics are shown in Supplementary Tables 2 and 3. With higher serum endotrophin level, there was a trend of increasing age, diabetes duration, BMI, and UAE; decreasing kidney function; and a higher proportion of individuals with complications (Supplementary Table 2). For urine endotrophin, the baseline characteristics were evenly distributed across the quartiles (Supplementary Table 3). Serum and urine endotrophin were not correlated (Spearman correlation coefficient = 0.008, P = 0.84). A higher level of serum endotrophin was significantly associated with all primary end points in both crude and adjusted analyses (Table 1). The adjusted hazard ratio (HR) per doubling of serum endotrophin was 1.80 (95% CI 1.13–2.87) for the composite kidney end point (n = 49 of 748), 1.54 (1.04–2.28) for first MACE (n = 64 of 540), and 1.69 (1.31–2.19) for mortality (n = 155 of 755). For the secondary end points, there was a significant association between a higher level of serum endotrophin and incident heart failure (n = 42 of 671) with an adjusted HR of 1.63 (1.02–2.60). Urine endotrophin level was positively associated with progression of albuminuria (n = 84 of 702; HR 1.20 [95% CI 1.04–1.39]) after adjustment (Table 1).
Associations between endotrophin and incidence of complications estimated with Cox proportional hazards model
| . | Serum . | Urine . | ||
|---|---|---|---|---|
| Crude . | Adjusted . | Crude . | Adjusted . | |
| Primary end points | ||||
| Kidney failure or ≥40% decline in kidney function (n = 49 of 756) | 2.78 (2.06–3.76) | 1.80 (1.13–2.87) | 1.19 (0.97–1.46) | 1.07 (0.86–1.31) |
| First major cardiovascular event (n = 66 of 546) | 1.78 (1.34–2.38) | 1.54 (1.04–2.28) | 1.01 (0.83–1.24) | 1.01 (0.82–1.24) |
| Mortality (n = 156 of 764) | 2.34 (1.96–2.79) | 1.69 (1.31–2.19) | 1.02 (0.90–1.16) | 1.01 (0.89–1.13) |
| Secondary end points | ||||
| Progression of albuminuria (n = 91 of 764) | 1.52 (1.18–1.97) | 1.31 (0.96–1.79) | 1.21 (1.05–1.41) | 1.20 (1.04–1.39) |
| Incident heart failure (n = 42 of 678) | 2.23 (1.60–3.10) | 1.63 (1.02–2.60) | 0.94 (0.74–1.19) | 0.95 (0.75–1.20) |
| Incident sight-threatening diabetic eye disease (n = 23 of 636) | 1.78 (1.10–2.89) | 1.74 (0.91–3.34) | 0.99 (0.72–1.36) | 0.87 (0.62–1.22) |
| . | Serum . | Urine . | ||
|---|---|---|---|---|
| Crude . | Adjusted . | Crude . | Adjusted . | |
| Primary end points | ||||
| Kidney failure or ≥40% decline in kidney function (n = 49 of 756) | 2.78 (2.06–3.76) | 1.80 (1.13–2.87) | 1.19 (0.97–1.46) | 1.07 (0.86–1.31) |
| First major cardiovascular event (n = 66 of 546) | 1.78 (1.34–2.38) | 1.54 (1.04–2.28) | 1.01 (0.83–1.24) | 1.01 (0.82–1.24) |
| Mortality (n = 156 of 764) | 2.34 (1.96–2.79) | 1.69 (1.31–2.19) | 1.02 (0.90–1.16) | 1.01 (0.89–1.13) |
| Secondary end points | ||||
| Progression of albuminuria (n = 91 of 764) | 1.52 (1.18–1.97) | 1.31 (0.96–1.79) | 1.21 (1.05–1.41) | 1.20 (1.04–1.39) |
| Incident heart failure (n = 42 of 678) | 2.23 (1.60–3.10) | 1.63 (1.02–2.60) | 0.94 (0.74–1.19) | 0.95 (0.75–1.20) |
| Incident sight-threatening diabetic eye disease (n = 23 of 636) | 1.78 (1.10–2.89) | 1.74 (0.91–3.34) | 0.99 (0.72–1.36) | 0.87 (0.62–1.22) |
Data are HR (95% CI). Estimates are given per doubling in endotrophin. Adjusted analyses included sex, baseline age, diabetes duration, BMI, systolic blood pressure, HbA1c, LDL cholesterol, smoking, eGFR, and UAE rate (not included in the model of albuminuria progression) as covariates. Kidney failure includes kidney failure as cause of death, kidney transplantation, chronic dialysis, chronic kidney disease stage 5, or sustained or observer-evaluated eGFR <15 mL/min/1.73 m2. First major cardiovascular event includes nonfatal acute myocardial infarction, coronary intervention, nonfatal stroke, or cardiovascular disease as primary cause of death. Progression of albuminuria was defined as progression from normo- to micro- or macroalbuminuria or from micro- to macroalbuminuria. Sight-threatening eye disease included proliferative retinopathy and maculopathy based on retinal photo or procedural codes including intraocular surgery, laser treatment, or anti-VEGF injections linked to a diagnosis of eye disease that is considered related to diabetes.
Conclusions
In this cohort of 774 unselected individuals with type 2 diabetes, we demonstrated an independent association between serum level of the collagen type VI formation marker endotrophin and hazard of kidney events, first MACE, mortality, and incident heart failure. For urine endotrophin level, we showed an association with albuminuria progression. Higher level of serum endotrophin was previously associated with mortality, cardiovascular events, and kidney function decline in individuals with type 2 diabetes and microalbuminuria (11) and in individuals with type 1 diabetes and with increased arterial stiffness, kidney function decline, development of kidney failure, and mortality but not with cardiovascular events or heart failure (12,13). The robust results across diabetes cohorts are promising for the application of serum endotrophin as a risk marker for these weighty complications. Endotrophin measured in urine appears as a less sensitive risk marker. Higher level of urine endotrophin is associated with 1-year progression of chronic kidney disease (14) but contradictorily associated with a lower risk of kidney function decline in the previously mentioned type 1 diabetes cohort (13). The clear association between serum endotrophin level and serious adverse outcomes for individuals with diabetes is promising in the search for biomarkers for risk stratification for rapid kidney function decline, cardiovascular disease, and other diabetes complications. This could eventually ease the process of prioritizing protective treatments between individuals. Endotrophin may be an interesting target to combat the burdensome consequences of fibrogenesis in visceral organs. Alternatively, a reduction in serum endotrophin may be a biomarker for other successful interventions targeting the kidney and cardiovascular complications in diabetes. The study is based on a large, unselected, and well-characterized cohort. The extensive follow-up data cover all hospital contacts, laboratory reports, and detailed retinal characterizations, enabling comprehensive end point definitions. The fact that this type 2 diabetes cohort was unselected is important for broader clinical relevance and possible future biomarker implementation. For generalization, it is important to emphasize that this study was carried out in a secondary care setting; hence, this population does not represent individuals with a very low risk of diabetes complications.
Higher level of serum endotrophin, reflecting collagen type VI formation and thus fibrosis, was associated with higher risk of kidney failure or kidney function decline, first MACE, incident heart failure, and mortality, whereas urinary level was only a marker for albuminuria progression.
This article contains supplementary material online at https://doi.org/10.2337/figshare.20063303.
Article Information
Acknowledgments. The authors thank the participants and the laboratory technicians who collected the biobank material in the StenoDot trial at Steno Diabetes Center Copenhagen.
Funding. This study was supported by the Danish Research Foundation and the Innovation Fund Denmark (0172-00270B).
Duality of Interest. P.F.R. and T.W.H. have shares in Novo Nordisk A/S. P.R. has received research support and personal fees from AstraZeneca and Novo Nordisk and personal fees from Astellas, Boehringer Ingelheim, Eli Lilly, Gilead, Mundipharma, Sanofi, and Vifor. A.L.M., F.G., M.A.K., and D.G.K.R. are full-time employees at Nordic Bioscience. Nordic Bioscience is a privately owned, small- to medium-sized enterprise partly focused on the development of biomarkers. F.G., M.A.K., and D.G.K.R. hold stocks in Nordic Bioscience. The patent for the ELISA used to measure endotrophin levels is owned by Nordic Bioscience. The funders provided support in the form of salaries for authors A.L.M., F.G., M.A.K., and D.G.K.R. A.L.M., F.G., M.A.K., and D.G.K.R., who are employees of Nordic Bioscience, performed the laboratory analysis and revised the manuscript. No other potential conflicts of interest relevant to this article were reported.
All fees to P.R. are given to Steno Diabetes Center Copenhagen. None of the authors received fees, bonuses or other benefits for the work described in the article. The funders did not play any additional role in the study design, data collection, analysis, decision to publish, or preparation of the manuscript.
Author Contributions. N.H.T., A.L.M., P.F.R., T.W.H., F.G., D.G.K.R., and P.R. contributed to conceptualization and methodology. A.L.M. performed sample analysis. N.H.T. contributed with cohort database development. P.F.R., T.W.H., F.G., M.A.K., D.G.K.R., and P.R. contributed to study supervision. N.H.T. analyzed the data. N.H.T. wrote the original draft of the manuscript. All authors read and agreed to the published version of the manuscript. N.H.T. and P.R. are the guarantors of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
