Data-driven subgroups of newly diagnosed type 2 diabetes and the relationship with cardiovascular diseases at genetic and clinical levels in Chinese adults

Weihao Wang; Tong Jia; Yiying Liu; Hongrong Deng; Zihao Chen; Jing Wang; Zhaoxu Geng; Ran Wei; Jingtao Qiao; Yanhua Ma; Xun Jiang; Wen Xu; Jian Shao; Kaixin Zhou; Ying Li; Qi Pan; Wenying Yang; Jianping Weng; Lixin Guo

doi:10.1016/j.dsx.2023.102850

Data-driven subgroups of newly diagnosed type 2 diabetes and the relationship with cardiovascular diseases at genetic and clinical levels in Chinese adults

Diabetes Metab Syndr. 2023 Sep;17(9):102850. doi: 10.1016/j.dsx.2023.102850. Epub 2023 Aug 30.

Authors

Weihao Wang¹, Tong Jia², Yiying Liu³, Hongrong Deng⁴, Zihao Chen⁵, Jing Wang⁶, Zhaoxu Geng⁶, Ran Wei¹, Jingtao Qiao¹, Yanhua Ma¹, Xun Jiang¹, Wen Xu⁴, Jian Shao⁷, Kaixin Zhou⁸, Ying Li⁵, Qi Pan⁹, Wenying Yang¹⁰, Jianping Weng¹¹, Lixin Guo¹²

Affiliations

¹ Department of Endocrinology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, PR China.
² Institute for Artificial Intelligence, Peking University, Beijing, China.
³ Institute of Biophysics, Chinese Academy of Sciences, Beijing, China; College of Life Sciences, University of Chinese Academy of Sciences, China.
⁴ Department of Endocrinology and Metabolism, Guangdong Provincial Key Laboratory of Diabetology, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China.
⁵ School of Software and Microelectronics, Peking University, Beijing, China.
⁶ Institute of Biophysics, Chinese Academy of Sciences, Beijing, China.
⁷ No. 9 XingDaoHuanBei Road, Guangzhou International Bio Island, Guangzhou, 510005, Guangdong Province, China.
⁸ The Fifth People's Hospital of Chongqing, Chongqing, China.
⁹ Department of Endocrinology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, PR China. Electronic address: panqi621@126.com.
¹⁰ China-Japan Friendship Hospital, Beijing, China. Electronic address: ywying_1010@126.com.
¹¹ Department of Endocrinology and Metabolism, Guangdong Provincial Key Laboratory of Diabetology, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China. Electronic address: wjianp@mail.sysu.edu.cn.
¹² Department of Endocrinology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, PR China. Electronic address: glx1218@163.com.

PMID: 37683311
DOI: 10.1016/j.dsx.2023.102850

Abstract

Background: To subgroup Chinese patients with newly diagnosed type 2 diabetes (T2D) by K-means cluster analysis on clinical indicators, and to explore whether these subgroups represent different genetic features and calculated cardiovascular risks.

Methods: The K-means clustering analysis was performed on two cohorts (n = 590 and 392), both consisting of Chinese participants with newly diagnosed T2D. To assess genetic risks, multiple polygenic risk scores (PRSs) and mitochondrial DNA copy numbers (mtDNA-CN) were calculated for all participants. Furthermore, Framingham risk scores (FRS) of cardiovascular diseases in two cohorts were also calculated to verify the genetic risks.

Results: Four clusters were identified including the mild age-related diabetes (MARD)(35.08%), mild obesity-related diabetes (MOD) (34.41%), severe autoimmune diabetes (SAID) 19.15%, and severe insulin-resistant diabetes (SIRD) 11.36% subgroups in the MARCH (metformin, and acarbose in Chinese patients as the initial hypoglycemic treatment) cohort. There was a significant difference in PRS for cardiovascular diseases (CVD) across four subgroups in the MARCH cohort (p < 0.05). Compared with the SIDD and SIRD subgroups, patients in the MOD subgroup had a relatively lower PRS for CVD (p < 0.05) in the MARCH cohort. Females had a higher PRS compared to males, with no significant difference in FRS across the four clusters. The MOD subgroup had a significantly lower FRS which was consistent with the results of PRS. Similar results of PRS and FRS were also replicated in the CONFIDENCE (comparison of glycemic control and b-cell function among newly diagnosed patients with type 2 diabetes treated with exenatide, insulin or pioglitazone) cohort.

Conclusion: There are different CVD risks in diabetic subgroups based on clinical and genetic evidence which may promote precision medicine.

Keywords: Cardiovascular risks; Diabetes subgroups; Machine learning; Polygenic risk score.