CoChat
Sign Up
Cited by 60Chronic Kidney Disease and DiabetesMetabolomics and Mass Spectrometry StudiesDiabetes, Cardiovascular Risks, and Lipoproteins

Machine Learning Approaches Reveal Metabolic Signatures of Incident Chronic Kidney Disease in Individuals With Prediabetes and Type 2 Diabetes

Open access
Huang, Jialing, Huth, Cornelia, Covic, Marcela, Troll, Martina, Adam, Jonathan, Zukunft, Sven, Prehn, Cornelia, Wang, Li +12 more·2020·Diabetes
DOI 10.2337/db20-0586

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Early and precise identification of individuals with prediabetes and type 2 diabetes (T2D) at risk for progressing to chronic kidney disease (CKD) is essential to prevent complications of diabetes. Here, we identify and evaluate prospective metabolite biomarkers and the best set of predictors of CKD in the longitudinal, population-based Cooperative Health Research in the Region of Augsburg (KORA) cohort by targeted metabolomics and machine learning approaches. Out of 125 targeted metabolites, sphingomyelin C18:1 and phosphatidylcholine diacyl C38:0 were identified as candidate metabolite biomarkers of incident CKD specifically in hyperglycemic individuals followed during 6.5 years. Sets of predictors for incident CKD developed from 125 metabolites and 14 clinical variables showed highly stable performances in all three machine learning approaches and outperformed the currently established clinical algorithm for CKD. The two metabolites in combination with five clinical variables were identified as the best set of predictors, and their predictive performance yielded a mean area value under the receiver operating characteristic curve of 0.857. The inclusion of metabolite variables in the clinical prediction of future CKD may thus improve the risk prediction in people with prediabetes and T2D. The metabolite link with hyperglycemia-related early kidney dysfunction warrants further investigation.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.2337/db20-0586

Cites

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Huang et al. reported that a multimodal model combining two metabolites, sphingomyelin C18:1 and phosphatidylcholine diacyl C38:0, with clinical variables achieved optimal prediction of DN, yielding an AUC of 0.825, which outperformed the model based solely on established clinical predictors (AUC = 0.800).

    “Huang et al. [103] reported that a multimodal model combining two metabolites, sphingomyelin C18:1 and phosphatidylcholine diacyl C38:0, with clinical variables achieved optimal prediction of DN, yielding an AUC of 0.825, which outperformed the model based solely on established clinical predictors (AUC = 0.800).”

    From Artificial intelligence-powered prediction of diabetic complications: from clinical data to molecular omics (2026) · cites this paper as [103]

    Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

    Experimental — accuracy not yet measured

  • For CKD and diabetic nephropathy screening, studies utilized clinical data, metabolic biomarkers, genotyping data, and/or retinal images to achieve AUCs generally over 0.8.

    “For CKD39,110,111 and diabetic nephropathy112,113 screening, studies utilized clinical data, metabolic biomarkers, genotyping data, and/or retinal images to achieve AUCs generally over 0.8.”

    From Artificial intelligence in diabetes management: Advancements, opportunities, and challenges (2023) · cites this paper as [110]

  • Huang et al. reported that ML classifiers using metabolic biomarkers for CKD screening identified two metabolites in combination with five clinical variables as the best set of predictors, with a mean AUC of 0.857.

    “two metabolites in combination with five clinical variables were identified as the best set of predictors, and their predictive performance yielded a mean AUC of 0.857”

    From Artificial intelligence in diabetes management: Advancements, opportunities, and challenges (2023) · cites this paper as Huang et al.110

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

The citation lists haven’t been fetched yet

References 55 · Cited by 60

Research works printed but not linked: 43

  1. Bikbov B, Purcell CA, Levey AS, et al.; GBD Chronic Kidney Disease Col- laboration. Global, regional, and national burden of chronic kidney disease, 1990- 2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 2020;395:709-733
  2. Webster AC, Nagler EV, Morton RL, Masson P. Chronic kidney disease. Lancet 2017;389:1238-1252
  3. Alicic RZ, Neumiller JJ, Johnson EJ, Dieter B, Tuttle KR. Sodium-glucose cotransporter 2 inhibition and diabetic kidney disease. Diabetes 2019;68:248-257
  4. Plantinga LC, Crews DC, Coresh J, et al.; CDC CKD Surveillance Team. Prevalence of chronic kidney disease in US adults with undiagnosed diabetes or prediabetes. Clin J Am Soc Nephrol 2010;5:673-682
  5. Melsom T, Schei J, Stefansson VT, et al. Prediabetes and risk of glomerular hyperfiltration and albuminuria in the general nondiabetic population: a pro- spective cohort study. Am J Kidney Dis 2016;67:841-850
  6. Markus MRP, Ittermann T, Baumeister SE, et al. Prediabetes is associated with microalbuminuria, reduced kidney function and chronic kidney disease in the general population: the KORA (Cooperative Health Research in the Augsburg Region) F4-Study. Nutr Metab Cardiovasc Dis 2018;28:234-242
  7. Li W, Wang A, Jiang J, et al. Risk of chronic kidney disease defined by decreased estimated glomerular filtration rate in individuals with different pre- diabetic phenotypes: results from a prospective cohort study in China. BMJ Open Diabetes Res Care 2020;8:e000955
  8. Ceriello A, Barkai L, Christiansen JS, et al. Diabetes as a case study of chronic disease management with a personalized approach: the role of a struc- tured feedback loop. Diabetes Res Clin Pract 2012;98:5-10
  9. Levin A, Stevens PE, Bilous RW, et al.; Kidney Disease Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013;3: 1-150
  10. Tangri N, Stevens LA, Griffith J, et al. A predictive model for progression of chronic kidney disease to kidney failure. JAMA 2011;305:1553-1559
  11. Dunkler D, Gao P, Lee SF, et al.; ONTARGET and ORIGIN Investigators. Risk prediction for early CKD in type 2 diabetes. Clin J Am Soc Nephrol 2015;10:1371-1379
  12. Floegel A, Stefan N, Yu Z, et al. Identification of serum metabolites associated with risk of type 2 diabetes using a targeted metabolomic approach. Diabetes 2013;62:639-648
  13. Wang TJ, Larson MG, Vasan RS, et al. Metabolite profiles and the risk of developing diabetes. Nat Med 2011;17:448-453
  14. Chen GC, Chai JC, Yu B, et al. Serum sphingolipids and incident diabetes in a US population with high diabetes burden: the Hispanic Community Health Study/ Study of Latinos (HCHS/SOL). Am J Clin Nutr 2020;112:57-65
  15. Carayol M, Leitzmann MF, Ferrari P, et al. Blood metabolic signatures of body mass index: a targeted metabolomics study in the EPIC cohort. J Proteome Res 2017;16:3137-3146
  16. Leal-Witt MJ, Ramon-Krauel M, Samino S, et al. Untargeted metabolomics identifies a plasma sphingolipid-related signature associated with lifestyle in- tervention in prepubertal children with obesity. Int J Obes 2018;42:72-78
  17. Razquin C, Toledo E, Clish CB, et al. Plasma lipidomic profiling and risk of type 2 diabetes in the PREDIMED trial. Diabetes Care 2018;41:2617-2624
  18. Alderete TL, Jin R, Walker DI, et al. Perfluoroalkyl substances, metabolomic profiling, and alterations in glucose homeostasis among overweight and obese Hispanic children: a proof-of-concept analysis. Environ Int 2019;126:445-453
  19. Hocher B, Adamski J. Metabolomics for clinical use and research in chronic kidney disease. Nat Rev Nephrol 2017;13:269-284
  20. Solini A, Manca ML, Penno G, Pugliese G, Cobb JE, Ferrannini E. Prediction of declining renal function and albuminuria in patients with type 2 diabetes by metabolomics. J Clin Endocrinol Metab 2016;101:696-704
  21. Herder C, Kannenberg JM, Huth C, et al. Proinflammatory cytokines predict the incidence and progression of distal sensorimotor polyneuropathy: KORA F4/ FF4 study. Diabetes Care 2017;40:569-576
  22. Chak CM, Lacruz ME, Adam J, et al. Ageing investigation using two-time- point metabolomics data from KORA and CARLA studies. Metabolites 2019;9:44
  23. Römisch-Margl W, Prehn C, Bogumil R, Röhring C, Suhre K, Adamski J. Procedure for tissue sample preparation and metabolite extraction for high- throughput targeted metabolomics. Metabolomics 2012;8:133-142
  24. Klau S, Jurinovic V, Hornung R, Herold T, Boulesteix AL. Priority-Lasso: a simple hierarchical approach to the prediction of clinical outcome using multi- omics data. BMC Bioinformatics 2018;19:322
  25. Chang C-C, Lin C-J. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol 2011;2:1-27
  26. Liaw A, Wiener M. Classification and regression by randomForest. R News 2002;2:18-22
  27. Culp M, Johnson K, Michailides G. ada: an R package for stochastic boosting. J Stat Softw 2006;17:1-27
  28. Merscher S, Fornoni A. Podocyte pathology and nephropathy -sphingolipids in glomerular diseases. Front Endocrinol (Lausanne) 2014;5:127
  29. Mäkinen VP, Tynkkynen T, Soininen P, et al. Sphingomyelin is associated with kidney disease in type 1 diabetes (The FinnDiane Study). Metabolomics 2012; 8:369-375
  30. Liu JJ, Ghosh S, Kovalik JP, et al. Profiling of plasma metabolites suggests altered mitochondrial fuel usage and remodeling of sphingolipid metabolism in individuals with type 2 diabetes and kidney disease. Kidney Int Rep 2016;2:470-480
  31. Tofte N, Suvitaival T, Ahonen L, et al. Lipidomic analysis reveals sphingo- myelin and phosphatidylcholine species associated with renal impairment and all- cause mortality in type 1 diabetes. Sci Rep 2019;9:16398
  32. Sigruener A, Kleber ME, Heimerl S, Liebisch G, Schmitz G, Maerz W. Glycerophospholipid and sphingolipid species and mortality: the Ludwigshafen Risk and Cardiovascular Health (LURIC) study. PLoS One 2014;9:e85724 38. Floegel A, Kühn T, Sookthai D, et al. Serum metabolites and risk of myo- cardial infarction and ischemic stroke: a targeted metabolomic approach in two German prospective cohorts. Eur J Epidemiol 2018;33:55-66
  33. Jiang XC, Paultre F, Pearson TA, et al. Plasma sphingomyelin level as a risk factor for coronary artery disease. Arterioscler Thromb Vasc Biol 2000;20:2614-2618
  34. Li Z, Basterr MJ, Hailemariam TK, et al. The effect of dietary sphingolipids on plasma sphingomyelin metabolism and atherosclerosis. Biochim Biophys Acta 2005;1735:130-134
  35. Cai Q, Mukku VK, Ahmad M. Coronary artery disease in patients with chronic kidney disease: a clinical update. Curr Cardiol Rev 2013;9:331-339
  36. Miyamoto S, Hsu C-C, Hamm G, et al. Mass spectrometry imaging reveals elevated glomerular ATP/AMP in diabetes/obesity and identifies sphingomyelin as a possible mediator. EBioMedicine 2016;7:121-134
  37. Torretta E, Barbacini P, Al-Daghri NM, Gelfi C. Sphingolipids in obesity and correlated co-morbidities: the contribution of gender, age and environment. Int J Mol Sci 2019;20:5901
  38. Johnson EL, Heaver SL, Waters JL, et al. Sphingolipids produced by gut bacteria enter host metabolic pathways impacting ceramide levels. Nat Commun 2020;11:2471
  39. Czumaj A, Śledzi nski T, Carrero JJ, et al. Alterations of fatty acid profile may contribute to dyslipidemia in chronic kidney disease by influencing hepatocyte metabolism. Int J Mol Sci 2019;20:2470
  40. Sugimoto M, Wakabayashi M, Shimizu Y, et al. Imaging mass spectrometry reveals acyl-chain-and region-specific sphingolipid metabolism in the kidneys of sphingomyelin synthase 2-deficient mice. PLoS One 2016;11:e0152191 48. Li Z, Zhang H, Liu J, et al. Reducing plasma membrane sphingomyelin increases insulin sensitivity. Mol Cell Biol 2011;31:4205-4218
  41. Fan Y, Shi F, Liu J, et al. Selective reduction in the sphingomyelin content of atherogenic lipoproteins inhibits their retention in murine aortas and the subsequent development of atherosclerosis. Arterioscler Thromb Vasc Biol 2010;30:2114-2120
  42. Adachi R, Ogawa K, Matsumoto SI, et al. Discovery and characterization of selective human sphingomyelin synthase 2 inhibitors. Eur J Med Chem 2017;136: 283-293
  43. Echouffo-Tcheugui JB, Kengne AP. Risk models to predict chronic kidney disease and its progression: a systematic review. PLoS Med 2012;9:e1001344 53. Boulesteix AL, Wright MN, Hoffmann S, König IR. Statistical learning ap- proaches in the genetic epidemiology of complex diseases. Hum Genet 2020;139: 73-84

Authors
Huang, J., Huth, C., Covic, M., Troll, M., Adam, J., Zukunft, S., Prehn, C., Wang, L., Nano, J., Scheerer, M. F., Neschen, S., Kastenmüller, G., Suhre, K., Laxy, M., Schliess, F., Gieger, C., Adamski, J., Hrabe de Angelis, M., Peters, A., & Wang-Sattler, R.
Affiliations
  • Helmholtz Munich (DE)
  • Shandong University (CN)
  • Weill Cornell Medical College in Qatar (QA)
  • Profil Institute for Metabolic Research (DE)
  • National University of Singapore (SG)
  • German Center for Diabetes Research (DZD), München-Neuherberg, Germany
  • Research Unit of Molecular Epidemiology, Helmholtz Zentrum München – German Research Center for Environmental Health, Neuherberg, Germany
ORCID
Venue
Diabetes
Publisher
American Diabetes Association
Published
2020-10-05
Volume
69
Issue
12
Pages
2756-2765
ISSN
  • 0012-1797 (ISSN-L)
  • 1939-327X
DOI
10.2337/db20-0586
OpenAlex
W3092114277
Cited by
60
References
55
Citations by year
2026: 11 · 2025: 9 · 2024: 13 · 2023: 12 · 2022: 8 · 2021: 7
FWCI
4.41
Type
journal article
Language
en
Open access
Licence
www.diabetesjournals.org/content/license (vor, 46 d)
Locations
  • Diabetes (publishedVersion)
  • PubMed (publishedVersion)
  • mediaTUM – the media and publications repository of the Technical University Munich (Technical University Munich) (submittedVersion)
  • Site cant be reached (submittedVersion, cc-by)
  • PuSH - Publication Server of Helmholtz Zentrum München (submittedVersion)
Funders
  • European Union Seventh Framework Programme (HEALTH-2013-2.4.2-1/602936 (Project CarTarDis))
  • European Institute of Innovation and Technology (EIT) Health (iPDM-GO “Integrated Personalized Diabetes Management Goes Europe”)
Topics
  • Health Sciences › Medicine › Nephrology › Chronic Kidney Disease and Diabetes
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Metabolomics and Mass Spectrometry Studies
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes, Cardiovascular Risks, and Lipoproteins
Keywords
chronic kidney disease, prediabetes, type 2 diabetes, metabolic biomarkers, hyperglycemia, CKD prediction, targeted metabolomics, machine learning prediction models, KORA cohort
MeSH terms
Machine Learning, Adult, Aged, Aged, 80 and over, Blood Glucose, Diabetes Mellitus, Type 2, Humans, Middle Aged, Prediabetic State, Biomarkers, Renal Insufficiency, Chronic

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review