Plasma Metabolomics Profiling in Auto-Antibody-Positive Individuals Progressing to Type 1 Diabetes

Introduction

Changes in the blood metabolite levels prior to the onset of T1D have been reported in only a few studies. However, reported changes have not been replicated in independent studies and the validity of the research methods has been debated in the scientific community. Therefore, we set out to measure metabolite levels in a specific longitudinal cohort of individuals who are at high-risk of developing T1D.

Methods

Overall, 42 auto-antibody-positive participants provided a monthly blood sample with a median follow-up duration of 20 months as part of 2 related observational studies. All participants provided at least 10 samples during the study period. Further, most participants were followed for at least a year after study enrolment. 

Plasma samples were analyzed using two chromatography and mass spectrometry platforms, acquiring a comprehensive molecular profile of circulating blood metabolites and lipids. The data were pre-processed with ChromaTOF and MZmine 2. Power calculations, quality control, and post-processing of the data, as well as statistical analyses, were done using R.

Results

Overall, 674 blood samples from auto-antibody-positive participants were analyzed using the two platforms. The analyses resulted in the detection of 463 lipidomic features from the major lipid classes and 41 metabolites, including amino acids, free fatty acids, and citric acid cycle metabolites.

During the primary study, 8 participants developed T1D, and multiple samples were collected prior to onset: In total, 61 and 117 samples were analyzed from the T1D-progressors before and after the onset of T1D, respectively. Additionally, 488 samples were analyzed from the 32 at-risk participants, who did not progress to T1D. 

The measured metabolite profiles provide insights into metabolic changes that occur before, during, and after the onset of T1D. The data also provide a starting point for the INNODIA project, which aims at improving early diagnostics of T1D development.