PROJECT SUMMARY/ABSTRACT
Background: The increasing incidence and prevalence of type 2 diabetes (T2D) can be mitigated by
successful prevention with behavioral changes or pharmacotherapy in individuals that are at high risk.
Identifying a specific individual’s risk for diabetes and which preventive treatment will be most effective,
however, remains a clinical dilemma. Metabolite biomarkers represent a novel approach to addressing these
clinical challenges and may also reveal mechanisms of disease development. This five-year mentored career
development proposal details a translational research training program in biomarker discovery and clinical
study implementation with a goal of identifying metabolite predictors of diabetes and preventive treatment
effect. Candidate: The applicant is a recently appointed Endocrinology faculty member at Beth Israel
Deaconess Medical Center with a long-term goal to become a physician-scientist who investigates metabolic
pathways involved in T2D development and progression. She has 2.5 years of experience in molecular
profiling in large human studies and the outlined proposal builds on this background to provide new domains of
expertise including prediction modeling, genetics, and clinical study design. Training: The applicant’s
development will occur through a blend of laboratory training, didactic courses (including an ongoing MPH),
and scientific conferences. The candidate’s mentor is a recognized leader in molecular profiling, integrative
“omics”, and cardiometabolic diseases. Her co-mentor is a leader in T2D genetics, and the members of her
advisory committee have a distinguished mentoring record and vast expertise in T2D, non-targeted
metabolomics, statistical modeling, and clinical trials. Research: Recently, the applicant identified unique
metabolite profiles associated with T2D in a large, randomized control trial that compared lifestyle changes and
metformin therapy for T2D prevention in individuals with impaired glucose regulation. These associations,
including specific phosphatidylcholines and cholesterol esters, remained after adjustments for traditional
clinical risk factors suggesting they could serve as predictors of T2D. Furthermore, these metabolites were
associated with different rates of T2D progression after lifestyle or metformin treatment and may help guide
T2D prevention decisions. The applicant proposes to further extend these findings by (Aim 1) creating
metabolite multi-marker prediction models to assess the ability of these metabolites to prospectively predict
diabetes development, (Aim 2A) integrate these findings with genetics to uncover metabolic pathways that
underly these associations and if they cause diabetes, (Aim 2B) leverage genetics to also identify completely
novel circulating small molecules associated with T2D, and (Aim 3) conduct a small prospective cohort study to
externally validate these metabolites as predictive biomarkers in a real-world clinical setting. This work will lay
the foundation for an NIH R01 application for a clinical trial to assess the predictive utility of metabolite
biomarkers for diabetes prevention and transition the candidate to research independence.