Targeted Assessment of participants through Recall by Genotype for Evaluation and Translation
The overall goal of this research is to leverage genomic information to identify individuals with high genetic risk for type 2 diabetes, assess their disease progression, accumulation of co-morbidities, and healthcare utilization, and characterize their beta-cell function, insulin resistance, and pharmacogenetic responses to type 2 diabetes therapies.
Aims & Outcomes
Aims
- Characterize the physiologic responses in individuals with extreme genetic risk for type 2 diabetes through a recall-by-genotype study.
- Characterize the pharmacological responses in individuals with extreme genetic risk for type 2 diabetes.
Outcomes Measured
- Pancreatic beta-cell function and insulin resistance, measured with a mixed meal tolerance test.
- Glycemic response to sulfonylurea as a drug challenge.
The genetics behind who TARGET recruits.
TARGET’s recall-by-genotype design depends on first knowing which genetic variants carry outsized risk. These two papers — both co-authored by Dr. Leong and colleagues from this team — are the discovery and validation work behind that recruitment strategy.
Most genome-wide association studies assume genetic effects are additive — the same risk increase whether you carry one or two copies of a variant. This study instead modeled recessive effects, where risk shows up mainly in people carrying two copies, and found a locus with a substantially larger effect on type 2 diabetes risk in homozygous carriers than additive models alone would have detected. It’s exactly this kind of large, easy-to-miss genetic effect that TARGET recruits participants around.
Using up to 16 years of follow-up in a real-world, ancestrally diverse primary care network affiliated with MGH, this study asked whether a polygenic score for type 2 diabetes still adds predictive value once standard clinical risk factors are accounted for — and found that it does, with the largest added benefit for patients whose electronic health records had the least clinical risk-factor data available. That’s the same population logic TARGET uses to recruit participants at the extremes of genetic risk directly from biobank data.