Grant: CIHR Fellowship

Cirrhosis is the end stage for most liver diseases. It is caused by a build up of scar tissue and results in a gradual loss of liver function. A minority of patients are candidates for liver transplantation. As cirrhosis advances, most patients experience chronic decline, frequent complications, repeated hospitalizations and a high burden of symptoms. Nutrition and exercise have helped patients with other chronic diseases improve or maintain their overall health. This is particularly important in cirrhosis where 80% of patients have dietary issues and muscle loss contributing to physical and mental frailty. In turn, frail patients with cirrhosis have a higher risk of hospitalization and death. By taking in a liver-friendly diet and engaging in regular exercise, studies have demonstrated improvements in frailty, reduced symptom burden and higher health-related quality of life. The benefit of these treatments is not experienced by all patients however, and the reasons for this remain unclear. During my postdoctoral fellowship, I will have access to a large dataset collected across the length of a 6-month randomized controlled trial. The trial compares patients with cirrhosis receiving standard care (control group) versus those in a nutrition and exercise program (experimental group). Unique from anything else that is available, the dataset will contain a broad range of data including not only clinical features and laboratory tests but also gut bacteria (intestinal microbiome) profiles, and metabolomic profiles (small molecules) present in the urine and blood. I will use machine learning, a form of artificial intelligence to predict the response to nutrition and exercise therapy and understand the mechanisms that explain response. Findings will help medical teams and their patients determine how to personalize these treatments in the future to achieve the most benefit for each individual patient.