Clinical characterization of data-driven diabetes subgroups in Mexicans using a reproducible machine learning approach
Omar Yaxmehen Bello-Chavolla, Jessica Paola Bahena-López, Arsenio Vargas-Vázquez, Neftali Eduardo Antonio-Villa, Alejandro Márquez-Salinas, Carlos A Fermín-Martínez, Rosalba Rojas, Roopa Mehta, Ivette Cruz-Bautista, Sergio Hernández-Jiménez, Ana Cristina García-Ulloa, Paloma Almeda-Valdes, Carlos Alberto Aguilar-Salinas, Metabolic Syndrome Study Group, Group of Study CAIPaDi, Olimpia Arellano-Campos, Donaji V Gómez-Velasco, Omar Yaxmehen Bello-Chavolla, César Lam-Chung, Ivette Cruz-Bautista, Marco A Melgarejo-Hernandez, Paloma Almeda-Valdés, Alexandro J Martagón, Liliana Muñoz-Hernandez, Luz E Guillén, José de Jesús Garduño-García, Ulices Alvirde, Yukiko Ono-Yoshikawa, Ricardo Choza-Romero, Leobardo Sauque-Reyna, Ma Eugenia Garay-Sevilla, Juan M Malacara-Hernandez, María Teresa Tusié-Luna, Luis Miguel Gutierrez-Robledo, Francisco J Gómez-Pérez, Rosalba Rojas, Carlos A Aguilar-Salinas, Sergio Hernández-Jiménez, Cristina García-Ulloa, Eder Patiño-Rivera, Denise Arcila-Martínez, Rodrigo Arizmendi-Rodríguez, Oswaldo Briseño-González, Humberto Del Valle-Ramírez, Arturo Flores-García, Fernanda Garnica-Carrillo, Eduardo González-Flores, Mariana Granados-Arcos, Héctor Infanzón-Talango, Victoria Landa-Anell, Claudia Lechuga-Fonseca, Arely López-Reyes, Marco Melgarejo-Hernández, Palacios-Vargas Angélica, Liliana Pérez-Peralta, Alberto Ramírez-García, David Rivera de la Parra, Sofía Ríos-Villavicencio, Francis Rojas-Torres, Marcela Ruiz-Cervantes, Sandra Sainos-Muñoz, Alejandra Sierra-Esquivel, Erendi Tinoco-Ventura, Luz Elena Urbina-Arronte, María Luisa Velasco-Pérez, Héctor Velázquez-Jurado, Andrea Villegas-Narváez, Verónica Zurita-Cortés, Enrique Graue-Hernández Aída Jiménez-Corona, Carlos A Aguilar-Salinas, Francisco J Gómez-Pérez, David Kershenobich-Stalnikowitz, Omar Yaxmehen Bello-Chavolla, Jessica Paola Bahena-López, Arsenio Vargas-Vázquez, Neftali Eduardo Antonio-Villa, Alejandro Márquez-Salinas, Carlos A Fermín-Martínez, Rosalba Rojas, Roopa Mehta, Ivette Cruz-Bautista, Sergio Hernández-Jiménez, Ana Cristina García-Ulloa, Paloma Almeda-Valdes, Carlos Alberto Aguilar-Salinas, Metabolic Syndrome Study Group, Group of Study CAIPaDi, Olimpia Arellano-Campos, Donaji V Gómez-Velasco, Omar Yaxmehen Bello-Chavolla, César Lam-Chung, Ivette Cruz-Bautista, Marco A Melgarejo-Hernandez, Paloma Almeda-Valdés, Alexandro J Martagón, Liliana Muñoz-Hernandez, Luz E Guillén, José de Jesús Garduño-García, Ulices Alvirde, Yukiko Ono-Yoshikawa, Ricardo Choza-Romero, Leobardo Sauque-Reyna, Ma Eugenia Garay-Sevilla, Juan M Malacara-Hernandez, María Teresa Tusié-Luna, Luis Miguel Gutierrez-Robledo, Francisco J Gómez-Pérez, Rosalba Rojas, Carlos A Aguilar-Salinas, Sergio Hernández-Jiménez, Cristina García-Ulloa, Eder Patiño-Rivera, Denise Arcila-Martínez, Rodrigo Arizmendi-Rodríguez, Oswaldo Briseño-González, Humberto Del Valle-Ramírez, Arturo Flores-García, Fernanda Garnica-Carrillo, Eduardo González-Flores, Mariana Granados-Arcos, Héctor Infanzón-Talango, Victoria Landa-Anell, Claudia Lechuga-Fonseca, Arely López-Reyes, Marco Melgarejo-Hernández, Palacios-Vargas Angélica, Liliana Pérez-Peralta, Alberto Ramírez-García, David Rivera de la Parra, Sofía Ríos-Villavicencio, Francis Rojas-Torres, Marcela Ruiz-Cervantes, Sandra Sainos-Muñoz, Alejandra Sierra-Esquivel, Erendi Tinoco-Ventura, Luz Elena Urbina-Arronte, María Luisa Velasco-Pérez, Héctor Velázquez-Jurado, Andrea Villegas-Narváez, Verónica Zurita-Cortés, Enrique Graue-Hernández Aída Jiménez-Corona, Carlos A Aguilar-Salinas, Francisco J Gómez-Pérez, David Kershenobich-Stalnikowitz
Abstract
Introduction: Previous reports in European populations demonstrated the existence of five data-driven adult-onset diabetes subgroups. Here, we use self-normalizing neural networks (SNNN) to improve reproducibility of these data-driven diabetes subgroups in Mexican cohorts to extend its application to more diverse settings.
Research design and methods: We trained SNNN and compared it with k-means clustering to classify diabetes subgroups in a multiethnic and representative population-based National Health and Nutrition Examination Survey (NHANES) datasets with all available measures (training sample: NHANES-III, n=1132; validation sample: NHANES 1999-2006, n=626). SNNN models were then applied to four Mexican cohorts (SIGMA-UIEM, n=1521; Metabolic Syndrome cohort, n=6144; ENSANUT 2016, n=614 and CAIPaDi, n=1608) to characterize diabetes subgroups in Mexicans according to treatment response, risk for chronic complications and risk factors for the incidence of each subgroup.
Results: SNNN yielded four reproducible clinical profiles (obesity related, insulin deficient, insulin resistant, age related) in NHANES and Mexican cohorts even without C-peptide measurements. We observed in a population-based survey a high prevalence of the insulin-deficient form (41.25%, 95% CI 41.02% to 41.48%), followed by obesity-related (33.60%, 95% CI 33.40% to 33.79%), age-related (14.72%, 95% CI 14.63% to 14.82%) and severe insulin-resistant groups. A significant association was found between the SLC16A11 diabetes risk variant and the obesity-related subgroup (OR 1.42, 95% CI 1.10 to 1.83, p=0.008). Among incident cases, we observed a greater incidence of mild obesity-related diabetes (n=149, 45.0%). In a diabetes outpatient clinic cohort, we observed increased 1-year risk (HR 1.59, 95% CI 1.01 to 2.51) and 2-year risk (HR 1.94, 95% CI 1.13 to 3.31) for incident retinopathy in the insulin-deficient group and decreased 2-year diabetic retinopathy risk for the obesity-related subgroup (HR 0.49, 95% CI 0.27 to 0.89).
Conclusions: Diabetes subgroup phenotypes are reproducible using SNNN; our algorithm is available as web-based tool. Application of these models allowed for better characterization of diabetes subgroups and risk factors in Mexicans that could have clinical applications.
Keywords: ethnic groups; insulin resistance; statistical models; type 2 diabetes mellitus.
Conflict of interest statement
Competing interests: JPB-L, AV-V and NEA-V are enrolled at the PECEM program of the Faculty of Medicine at UNAM. JPB-L and AV-V are supported by CONACyT.
© Author(s) (or their employer(s)) 2020. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.
Figures
References
- Ahlqvist E, Storm P, Käräjämäki A, et al. . Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables. Lancet Diabetes Endocrinol 2018;6:361–9. 10.1016/S2213-8587(18)30051-2
- Zaharia OP, Strassburger K, Strom A, et al. . Risk of diabetes-associated diseases in subgroups of patients with recent-onset diabetes: a 5-year follow-up study. Lancet Diabetes Endocrinol 2019;7:S2213-8587(19)30187-1:684–94. 10.1016/S2213-8587(19)30187-1
- Ferrannini E, Gastaldelli A, Matsuda M, et al. . Influence of ethnicity and familial diabetes on glucose tolerance and insulin action: a physiological analysis. J Clin Endocrinol Metab 2003;88:3251–7. 10.1210/jc.2002-021864
- Agbim U, Carr RM, Pickett-Blakely O, et al. . Ethnic disparities in adiposity: focus on non-alcoholic fatty liver disease, visceral, and generalized obesity. Curr Obes Rep 2019;8:243–54. 10.1007/s13679-019-00349-x
- Bancks MP, Casanova R, Gregg EW, et al. . Epidemiology of diabetes phenotypes and prevalent cardiovascular risk factors and diabetes complications in the National Health and Nutrition Examination Survey 2003-2014. Diabetes Res Clin Pract 2019;158:107915. 10.1016/j.diabres.2019.107915
- Udler MS, Kim J, von Grotthuss M, et al. . Type 2 diabetes genetic loci informed by multi-trait associations point to disease mechanisms and subtypes: a soft clustering analysis. PLoS Med 2018;15:e1002654. 10.1371/journal.pmed.1002654
- Steinley D. Stability analysis in k-means clustering. Br J Math Stat Psychol 2008;61:255–73. 10.1348/000711007X184849
- Lisboa PJG, Etchells TA, Jarman IH, et al. . Finding reproducible cluster partitions for the k-means algorithm. BMC Bioinformatics 2013;14(Suppl 1):S8. 10.1186/1471-2105-14-S1-S8
- Meilă M. Comparing clusterings—an information based distance. J Multivar Anal 2007;98:873–95. 10.1016/j.jmva.2006.11.013
- Klambauer G, Unterthiner T, Mayr A, et al. . Self-Normalizing neural networks. arXiv. 1706.02515v5 [cs.LG].
- Dennis JM, Shields BM, Henley WE, et al. . Disease progression and treatment response in data-driven subgroups of type 2 diabetes compared with models based on simple clinical features: an analysis using clinical trial data. Lancet Diabetes Endocrinol 2019;7:442–51. 10.1016/S2213-8587(19)30087-7
- Zou X, Zhou X, Zhu Z, et al. . Novel subgroups of patients with adult-onset diabetes in Chinese and US populations. Lancet Diabetes Endocrinol 2019;7:9–11. 10.1016/S2213-8587(18)30316-4
- Bello-Chavolla OY, Antonio-Villa NE, Vargas-Vázquez A, et al. . Metabolic score for visceral fat (METS-VF), a novel estimator of intra-abdominal fat content and cardio-metabolic health. Clin Nutr 2020;39:S0261-5614(19)30294-8. 10.1016/j.clnu.2019.07.012
- Bello-Chavolla OY, Almeda-Valdes P, Gomez-Velasco D, et al. . METS-IR, a novel score to evaluate insulin sensitivity, is predictive of visceral adiposity and incident type 2 diabetes. Eur J Endocrinol 2018;178:533–44. 10.1530/EJE-17-0883
- Arellano-Campos O, Gómez-Velasco DV, Bello-Chavolla OY, et al. . Development and validation of a predictive model for incident type 2 diabetes in middle-aged Mexican adults: the metabolic syndrome cohort. BMC Endocr Disord 2019;19:41. 10.1186/s12902-019-0361-8
- Basto-Abreu A, Barrientos-Gutiérrez T, Rojas-Martínez R, et al. . [Prevalence of diabetes and poor glycemic control in Mexico: results from Ensanut 2016.]. Salud Publica Mex 2020;62:50–9. 10.21149/10752
- Bedogni G, Bellentani S, Miglioli L, et al. . The fatty liver index: a simple and accurate predictor of hepatic steatosis in the general population. BMC Gastroenterol 2006;6:33. 10.1186/1471-230X-6-33
- SIGMA Type 2 Diabetes Consortium, Williams AL, Jacobs SBR, et al. . Sequence variants in SLC16A11 are a common risk factor for type 2 diabetes in Mexico. Nature 2014;506:97–101. 10.1038/nature12828
- Almeda-Valdes P, Gómez Velasco DV, Arellano Campos O, et al. . The SLC16A11 risk haplotype is associated with decreased insulin action, higher transaminases and large-size adipocytes. Eur J Endocrinol 2019;180:99–107. 10.1530/EJE-18-0677
- Hernández-Jiménez S, García-Ulloa AC, Bello-Chavolla OY, et al. . Long-Term effectiveness of a type 2 diabetes comprehensive care program. The CAIPaDi model. Diabetes Res Clin Pract 2019;151:128–37. 10.1016/j.diabres.2019.04.009
- Gubbi S, Hamet P, Tremblay J, et al. . Artificial intelligence and machine learning in endocrinology and metabolism: the dawn of a new era. Front Endocrinol 2019;10:185. 10.3389/fendo.2019.00185
- Bello-Chavolla OY, Rojas-Martinez R, Aguilar-Salinas CA, et al. . Epidemiology of diabetes mellitus in Mexico. Nutr Rev 2017;75:4–12. 10.1093/nutrit/nuw030
- Herrington WG, Alegre-Díaz J, Wade R, et al. . Effect of diabetes duration and glycaemic control on 14-year cause-specific mortality in Mexican adults: a blood-based prospective cohort study. Lancet Diabetes Endocrinol 2018;6:455–63. 10.1016/S2213-8587(18)30050-0
- Shyr ZA, Wang Z, York NW, et al. . The role of membrane excitability in pancreatic β-cell glucotoxicity. Sci Rep 2019;9:6952. 10.1038/s41598-019-43452-8
- Zimmet PZ, Tuomi T, Mackay IR, et al. . Latent autoimmune diabetes mellitus in adults (LADA): the role of antibodies to glutamic acid decarboxylase in diagnosis and prediction of insulin dependency. Diabet Med 1994;11:299–303. 10.1111/j.1464-5491.1994.tb00275.x
- Delitala AP, Pes GM, Fanciulli G, et al. . Organ-specific antibodies in LADA patients for the prediction of insulin dependence. Endocr Res 2016;41:207–12. 10.3109/07435800.2015.1136934
- Segovia-Gamboa NC, Rodríguez-Arellano ME, Muñoz-Solís A, et al. . High prevalence of humoral autoimmunity in first-degree relatives of Mexican type 1 diabetes patients. Acta Diabetol 2018;55:1275–82. 10.1007/s00592-018-1241-9
Source: PubMed