Project Summary/Abstract
In the US, type 2 diabetes (T2D) is a major cause of morbidity and mortality due to microvascular and
macrovascular complications. Nonalcoholic fatty liver disease (NAFLD) is another common co-morbidity but its
burden in patients with T2D is poorly studied. Although NAFLD affects 50-70% of T2D patients, there is neither
a screening guideline nor established treatment for NAFLD. Besides weight loss, pioglitazone – an old
diabetes drug – is the only drug recommended by the American Association for the Study of Liver Diseases for
treating NASH, the more progressive form of NAFLD, in patients with T2D. In addition, liraglutide (in phase 2
trials) and obeticholic acid (with phase 3 interim data) appear effective for improving NASH and advanced
fibrosis. At least 4 other drugs (making a total of 5) are undergoing phase 3 trials and up to 50 drugs are in
phase 2 trials. As new drugs become available in the near future, physicians and patients will face difficult
tradeoffs among long-term benefits, risks, and costs. Given the widespread nature of NAFLD in patients with
T2D, screening to identify those with NASH, who are at high-risk of advance liver disease, for early intervention
may help patients live longer with a better quality life. Yet, the lack of comparative effectiveness and cost-
effectiveness analyses of screening for NAFLD has resulted in conflicting clinical guidelines. Choosing the best
treatment and screening strategy for NAFLD is complicated and can be aided by the use of computer-based
models. Unfortunately, existing diabetes models do not include NAFLD, and NAFLD models do not account for
diabetes and its complications. The objective of this study is to develop a computer simulation model of NAFLD
in T2D to estimate its burden, and evaluate treatment and screening options for NASH. Specific aims are: 1)
Develop and validate a microsimulation model of the natural history of NAFLD, including NASH; 2) Estimate
and project the burden of NAFLD-associated liver complications in patients with T2D until 2030; 3) Assess the
cost-effectiveness of treatments for NASH in patients with T2D; and 4) Evaluate the cost-effectiveness of
screening for NASH in patients with T2D. This proposal builds upon our inter-disciplinary team’s strong
foundation of constructing decision models to assist clinical decision making and studying the epidemiology of
diabetes and NAFLD. Our model will provide a better understanding of a prevalent, important, yet
underappreciated diabetes complication, using an innovative approach to diabetes and liver disease modeling.
The study also establishes the groundwork for future evaluation of new treatment regimens and screening
modalities for patients with T2D and NAFLD, which will advance the prevention and treatment of both diseases
for millions of patients each year.
