Immunosuppressive treatment is a critical factor in the success of any kidney transplant surgery. Immunosuppressive medications such as Tacrolimus (Tac) have a narrow therapeutic range. High dosage of Tac reduces the risk of acute rejection, but it also increases the odds of adverse events and nephrotoxicity (Schiff et al., 2007). On the other hand, low dosage will increase the risk of acute rejection. Currently, transplant physicians select the initial dosage of immunosuppressive medication based on a patient’s Body Mass Index (BMI), i.e., 0.10-0.20 mg/kg per day. Then they adjust it on a daily basis by monitoring the trough concentration of Tac until it reaches 8-12 ng/ml for the first 3 months and 5-10 ng/ml for the 3-6 months post-transplant period (Jusko et al., 1995). This activity continues until patients reach a stable therapeutic level, usually within 5-6 days. Trial-and-error has been the most common practice to determine the optimal Tac dosage (Provenzani et al., 2013). This rule of thumb “one-size-fits-all” approach does not serve the unique needs of patients, particularly African American patients. Physicians desperately prefer to apply a personalized dosage for each patient at the start of immunosuppressive therapy to reduce the number of adjustments. Thus far, an optimal therapeutic dose selection approach has not been practiced (Davis et. al, 2018). Achieving an early immunosuppressive therapeutic level is proven to significantly decrease the risk of acute rejection by 58% by day 2 after transplantation (Schiff et. al, 2007). Studies have also shown that a lack of drug management and immunosuppressive medication modulation results in sever infectious conditions for kidney transplant recipients (Shih et al., 2014).

​

The need for a comprehensive individualized drug management plan for kidney transplant recipients is significant and utilization of a personalized plan will significantly improve transplant outcomes for all patients. Improved transplant outcomes such as lower rejection rates, shorter hospital stays, and higher organ utilization all translate into reducing the healthcare cost of post-transplant kidney care. An immediate benefit of this project will be for patients who belong to minority groups, especially African Americans.  Since most African Americans express the homozygote alleles of CYP3A5*1 gene, they tend to have higher metabolism, hence they require higher dosage of Tac.  But the current method of administering Tac does not accommodate pharmacogenomics factors. African American patients historically experience worse transplant outcomes compared to other patients (Taber et al., 2017). Practicing personalized kidney care in centers with high portion of African American patients will drastically improve the outcome of kidney transplantation for such populations. Transplant centers currently do not have the capacity to individualize immunosuppressive dosage because the cost of gene sequencing is extremely high and there are simply too many complex factors that need to be considered to select an individualized dosage. This project will develop a supervised Machine Learning algorithm which can accommodate the problem of high dimensionality based on gene variants and the gut microbiome composition to select an optimum Tac dosage.

Previous attempts to personalize immunosuppressive medication for kidney patients have been limited, yet they are deemed successful. A study by Thervet et. al conducted a prospective study to investigate whether using personalized initial Tac dosage through pharmacogenetic testing leads to improved outcomes. Results showed that a higher percentage of the group which received personalized dosage (based on the CYU3A5 genotype) reached therapeutic levels by day 3 (43.2% vs. 29.1%; P = 0.03). They also required fewer dose modifications.

 

Previous studies have been limited in three aspects: 1) they only accounted for few gene markers; 2) they did not explore the role of gut microbiome in the metabolization of immunosuppressive medication; and 3) the personalized dosage was limited to only one medication, not the whole cocktail of medications that kidney transplant patients consume.

This project will improve our knowledge of the field by including 21 genes to the genotyping of kidney patients, and by analyzing the effect of gut microbiome on immunosuppressive dosage, and considering simultaneous medication cocktail selection. These elements will be packaged in an algorithm to produce a personalized prescription for each patient before the transplant surgery begins. We believe if enough transplant centers apply to this comprehensive personalized medicine package, it will change the practice of immunosuppressive medication treatment. Instead of relying on a “one-size-fits-all” approach, transplant centers can apply a personalized dosage and increase their focus on transplant outcomes rather than numerous dose adjustments. If transplant centers adopt this method, they will know the optimum immunosuppressive medication dosage of patients before performing the transplant operation, while decreasing the odds of experiencing incidents of post-transplant complications. We believe the benefits of applying this service will be readily noticeable in shorter hospital stays and higher rate of organ utilization due to lower incidents of graft failure.  Such benefits will lead to significant savings and higher successful kidney graft cases which will improve the status of any transplant center.