Title: Investigate the Role of Gut Microbiota in the Development of Type 1 Diabetes Using TEDDY Study
Project Abstract:
Gut dysbiosis in early childhood might cause or contribute to early-onset type 1 diabetes (T1D). The
Environmental Determinants of Diabetes in the Young (TEDDY) study has established a nested case–control
(NCC) microbiome study for islet autoimmunity (IA) and T1D, which provides an opportunity to study microbial
factors involved in T1D development. However, only limited, subtle microbial associations with IA or T1D have
been detected in TEDDY so far, possibly because the longitudinal microbiome data have been analyzed in cross-
sectional fashion, neglecting the heterogeneity of the disease.
We hypothesize that temporal changes in gut microbiota in various microbiome maturation phases are
associated with the development of T1D as delineated by four distinct IA phenotypes and overt T1D. We propose
to develop two novel joint modeling tools that integrate longitudinal microbiome measurements and times to
onset of IA phenotypes and T1D and to implement these methods in a secondary analysis of TEDDY microbiome
data. Our new tools will embrace unique challenges presented by TEDDY’s NCC longitudinal microbiome study.
We have three aims. 1) Identify microbial biomarkers for specific IA phenotypes. We propose a novel joint
modeling framework under the NCC study design to characterize the association between time-varying microbial
abundances and the onset of specific IA phenotypes. We will apply the proposed statistical method to TEDDY
data to pinpoint which microbial biomarkers, either taxonomically or functionally, are associated with the
occurrence of specific IA phenotypes. 2) Identify microbial biomarkers that predict the risk of progression from
IA to T1D. We will extend the statistical method to allow for the modeling of two-stage T1D progression (from
birth to the onset of IA, and from onset of IA to overt T1D) and the discovery of microbial biomarkers that predict
T1D progression using TEDDY data. 3) Evaluate and refine these statistical methods and disseminate
companion software to the community.
The proposed new methods will allow us to discover novel microbial biomarkers for each IA phenotype and T1D
in the TEDDY study, broadening our understanding of the cause of T1D and ultimately promoting the
development of new strategies to prevent or delay the disease.