PROJECT SUMMARY/ABSTRACT Development of liver metastases among stage I-III colorectal cancer (CRC) patients following resection suggests the presence of clinically undetectable liver micro-metastases prior to CRC diagnosis. Non- alcoholic fatty liver disease (NAFLD) is quickly emerging as the most common liver disease worldwide with an estimated prevalence of 20-30% in the U.S. population. To date, various lines of evidence support our hypothesis of the potential role of liver fat in enhancing CRC liver metastasis; the next step is to demonstrate the utility of quantitative assessment of hepatic fat as a prognostic biomarker for stage I-III CRC patients in clinical settings. Almost all CRC patients in the U.S. get a computerized tomography (CT) scan at diagnosis, and NAFLD is a highly prevalent and treatable disease, thus, if our hypothesis is confirmed, we hope to establish a safe and cost-efficient prognostic biomarker, which is currently lacking. The recently expanded C-SCANS (CRC-Sarcopenia and Near-term Survival) cohort was derived from the Kaiser Permanente Northern California cancer registry, with ascertainment of patients with stage I-III CRC diagnosed between 2006 and 2018, aged 18−80 years, who underwent surgical resection. Inclusion criteria include that patients had baseline CT images within 4 months of diagnosis, and prior to treatment. In Aim 1, hepatic fat (quantitative assessment) will be measured in CT scans performed at time of CRC diagnosis, and prior to treatment. First, measurements will be performed manually, and then using a novel machine learning-based tool for liver segmentation, and measurement of liver and spleen attenuation. In Aims 2 and 3, we will examine (a) whether increased hepatic fat infiltration is associated with higher risk of liver metastases, recurrence, and CRC mortality, and (b) whether associations are modified by comorbidities and risk factors related to the metabolic syndrome. In Aim 4, we will examine whether hepatic fat affects liver metastases through a direct and/or indirect effect (e.g., mediated through the metabolic syndrome). Specifically, we hypothesize that the direct effect of hepatic fat is more important in explaining the relationship between hepatic fat and liver metastases than the indirect effect. Data on body composition, including skeletal muscle and total, subcutaneous and visceral fat, and hepatic fat for each patient are measured on the same CT scans, which combined with assessment of behavioral risk factors, clinical data, and metabolic plasma markers around the same time , provide an unprecedented opportunity to rigorously study these effects. Thus, this cost-efficient and innovative application will address the urgent need for more accurate prognostic biomarkers and relatively safe interventions to improve prognosis in stage I to III CRC patients. Considering the alarming rise in prevalence of NAFLD, this application has the potential to reduce morbidity and mortality in a subst...