PROJECT SUMMARY/ABSTRACT Immunosuppression is critical for preventing rejection after kidney transplantation but is often associated with a higher risk for infections. Infections leading to hospitalizations, graft failure, or mortality are more common in the kidney transplant population compared with the general population and are associated with enormous human and financial costs. Identifying approaches to reduce the risk for serious infections could improve graft and patient survival. Mycophenolate is widely used as the backbone of most immunosuppression regimens in kidney transplantation. Its incorporation into standard immunosuppression regimens has led to a reduction in the rate of rejection, but mycophenolate use is also associated with increased risk for infections with subsequent graft function loss related to infectious complications. Personalizing the dose of mycophenolate to balance the protection of graft function with risk for infections could lead to better patient and allograft survival. The goal of this proposal is to identify patients at risk for infection-related hospitalizations, infection-related graft failure, or infection-related mortality in kidney transplant recipients and to identify potential strategies to reduce this risk. In Aim 1, we will build a risk prediction model to discriminate the risk for infection-related adverse outcomes using data from the United States Renal Data System (USRDS). We will internally validate this model using USRDS data and externally validate our model using data from the Folic Acid for Vascular Outcome Reduction in Transplantation (FAVORIT) Trial. In Aim 2, we will assess the feasibility of developing an ancillary study in the ongoing PERformance and Frailty at Evaluation for Kidney Transplant (PERFEKT) study (which has 500 active participants) to understand how mycophenolate dosing (normalized to patient body surface area [BSA]) relates with infectious hospitalizations, acute kidney injury, and kidney function decline in kidney transplant recipients. This proposal will train Dr. Dinh in statistical analysis, particularly as it relates to risk prediction modeling, and will provide him with hands-on experience with prospective patient- oriented research. UCSF is well suited for the conduct of this proposal, given that UCSF is one of the largest transplant centers in the United States, performing 300-350 kidney transplants annually. Our specific aims are: Aim 1: To develop a prediction model for the risk of a composite of infection-related hospitalizations, infection- related graft failure, or infection-related death within the first three years of kidney transplantation using a USRDS cohort with >25,000 kidney transplant recipients, and to validate this model using FAVORIT. Aim 2: To assess the feasibility of developing an ancillary study to PERFEKT to understand the relationship between mycophenolate dosing (normalized to BSA), infectious and AKI events, and allograft function in kidney tra...