*Purpose: This study aims to validate and apply a personalized optimal dose selection model for Tacrolimus (Tac) for kidney transplant recipients at the George Washington University Hospital. The model has been previously developed by the same team using gene markers and gut microbiome data as predictors of optimal Tac dose. This is the first model that incorporates gut microbiome information for dose selection. Previous studies have shown that certain species in the gut can turn Tac into less potent metabolites, hence requiring higher dose for treatment. If drug-microbiome interactions could be predicted for every patient, it will facilitate the selection of optimal treatments and dose which leads to fewer dose adjustments, faster achievement of therapeutic levels and fewer adverse reactions. In this study we first validate our model and then use it in a clinical trial setting.

*Methods: In a previous study we recruited 10 patients and collected their blood for gene sequencing (CYP3A5) and two stool samples for pre- and post-transplant gut microbiome metagenomics sequencing. We analyzed the relationship between the observed optimal doses of Envarsus and different gut genera. We selected Phocaeicola and Bacteroides as predictors due to their significant correlation with observed optimal Tac doses, and added these variables to a previously developed model by Jacobson et el., in which they only used gene markers for optimal Tac dose selection. Four time frames were defined for the analysis, 2≤days≤5 ,6≤days≤10, 11≤days≤30, 31≤days≤90. In each time frame, observed optimal Tac dose was used as dependent variable and defined as the mean Tac dose administered to a patient at therapeutic levels (8-12 ng/ml). Pre-transplant relative abundance of Phocaeicola was a significant predictor of optimal Tac dose with p<0.001 for days≤10 and the change in relative abundance of Bacteroides was a significant predictor with p<0.001 for 10≤days≤90. Our model proved to be superior in predicting optimal Tac dose on the training set. For the current study, we will recruit another 10 patients as validation set in order to test our model and analyze its power in predicting optimal Tac dose. One blood sample and two stool samples will be collected from each patient for gene and gut microbiome metagenomics sequencing.

*Results: The results of this validation and clinical trial are still pending. We will compare the predicted values of our model to the observe optimal Tac doses. Prediction error will be analyzed using bias and precision. For the clinical trial, physicians will use our predicted optimal dose as a guidance for dose adjustment. We will analyze time to reach therapeutic range as our outcome of interests and retrospectively compare it to a similar cohort of patients.

*Conclusions: We expect using our personalized dose selection model will lead to fewer dose adjustments, which could lead to shorter time to achieve therapeutic levels.

« Back to 2022 American Transplant Congress