Machine learning-based method for tacrolimus dose predictions in Chinese kidney transplant perioperative patients.

WHAT IS KNOWN AND OBJECTIVES: Tacrolimus (TAC), a first-line immunosuppressant in solid-organ transplant, has a narrow therapeutic window and large inter-individual variability, which affects its use in clinical practice. Successful predictions using machine learning algorithms have been reported in several fields. However, a comparison of 10 machine learning model-based TAC pharmacogenetic and pharmacokinetic dosing algorithms for kidney transplant perioperative patients of Chinese descent has not been reported. The objective of this study was to screen and establish an appropriate machine learning method to predict the individualized dosages of TAC for perioperative kidney transplant patients. METHODS: The records of 2551 patients were collected from three transplant centres, 80% of which were randomly selected as a 'derivation cohort' to develop the dose prediction algorithm, while the remaining 20% constituted a 'validation cohort' to validate the final algorithm selected. Important features were screened according to our previously established population pharmacokinetic model of tacrolimus. The performances of the algorithms were evaluated and compared using R-squared and the mean percentage in the remaining 20% of patients. RESULTS AND DISCUSSION: This study identified several factors influencing TAC dosage, including CYP3A5 rs776746, CYP3A4 rs4646437, haematocrit, Wuzhi capsules, TAC daily dose, age, height, weight, post-operative time, nifedipine and the medication history of the patient. According to our results, among the 10 machine learning models, the extra trees regressor (ETR) algorithm showed the best performance in the training set (R-squared: 1, mean percentage within 20%: 100%) and test set (R-squared: 0.85, mean percentage within 20%: 92.77%) of the derivation cohort. The ETR model successfully predicted the ideal TAC dosage in 97.73% of patients, especially in the intermediate dosage range (>5 mg/day to <8 mg/day), whereby the ideal TAC dosage could be successfully predicted in 99% of the patients. WHAT IS NEW AND CONCLUSION: The results indicated that the ETR algorithm, which was chosen to establish the dose prediction model, performed better than the other nine machine learning models. This study is the first to establish ETR algorithms to predict TAC dosage. This study will further promote the individualized medication of TAC in kidney transplant patients in the future, which has great significance in ensuring the safety and effectiveness of drug use.

Population pharmacokinetic analysis and dosing guidelines for tacrolimus co-administration with Wuzhi capsule in Chinese renal transplant recipients.

WHAT IS KNOWN AND OBJECTIVES: Tacrolimus (TAC) is a first-line immunosuppressant which is used to prevent transplant rejection after solid organ transplantation (SOT). However, it has a narrow therapeutic index and high individual variability in pharmacokinetics (PK) and pharmacogenomics (PG). It has been reported that the metabolism of TAC can be affected by genetic factors, leading to different rates of metabolism in different subjects. Wuzhi Capsule (WZC) is a commonly used TAC-sparing agent in Chinese SOT to reduce TAC dosing due to its inhibitory effect on TAC metabolism by enzymes of the CYP3A subfamily. The aims of this study were to assess the effect of TAC+WZC co-administration and genetic polymorphism on the pharmacokinetics of TAC, by using a population pharmacokinetic (PPK) model. A dosing guideline for individualized TAC dosing is proposed based on the PPK study. METHODS: The medical records of 165 adult patients with kidney transplant and their 824 TAC concentrations from two kidney transplantation centres were reviewed. The genotypes of four single-nucleotide polymorphisms (SNPs) in CYP3A5*3 and ABCB1 (rs1128503, rs2032582 and rs1045642) were tested by MASSARRAY. A PPK model was constructed by nonlinear mixed effect model (NONMEM® , Version 7.3). Finally, Monte Carlo simulations were employed to design initial dosing regimens based on the final model. RESULTS AND DISCUSSION: The one-compartmental PPK model with first-order absorption and elimination of TAC was established in kidney transplant recipients (KTRs). CYP3A5*3 had significant impact on the PPK model. The haematocrit (HCT), postoperative time (POD) and CYP3A5*3 genotypes had a significant influence on TAC clearance when combined with WZC. The model was expressed as 23.4 × (HCT/0.3)-0.729  × 0.837 (combination with WZC) × e-0.0875(POD/12.6) ×1.18 (CYP3A5 expressors). For patients carrying the CYP3A5*3/*3 allele and with 30% HCT, the required TAC dose to achieve target trough concentrations of 10-15 ng/ml was 4 mg twice daily (q12h). For patients with the CYP3A5*3/*3 allele, the required dose was 3 mg TAC q12h when combined with WZC, and for patients with the CYP3A5*1/*1 or *1/*3 allele, the required dose was 4 mg of TAC q12h when co-administered with WZC. WHAT IS NEW AND CONCLUSION: Wuzhi Capsule co-administration and CYP3A5 variants affect the PK of TAC Dosing guidelines are made based on the PPK model to allow individualized administration of TAC, especially when co-administered with WZC.

CD-PLLD co-delivering docetaxel and MMP-9 siRNA plasmid for nasopharyngeal carcinoma therapy in vivo.

The co-delivery of a drug and a target gene has become a primary strategy in cancer therapy. Based on our previous study, a synthesized star­shaped co­polymer consisting of ß­cyclodextrin (CD) and a poly(L­lysine) dendron (PLLD) was used to co-deliver docetaxel (DOC) and matrix metalloproteinase 9 (MMP­9) small interfering RNA, via CD­PLLD/DOC/MMP­9 complexes, into mice implanted with HNE­1 human nasopharyngeal carcinoma (NPC) tumor cells in vivo. Unlike the commonly used amphiphilic co­polymer micelles, the obtained CD derivative may be used directly for a combined delivery of nucleic acid and hydrophobic DOC without a complicated micellization process. In vivo assays demonstrated that CD­PLLD/DOC/MMP­9 inhibited HNE­1 tumor growth and decreased proliferating cell nuclear antigen expression levels, indicating a potential strategy for NPC therapy. In addition, the distribution of DOC and MMP­9 was investigated; CD­PLLD/DOC/MMP­9 complexes were phagocytized in reticuloendothelial systems, including the liver and spleen, which requires further study. Furthermore, the complexes did not cross the blood­brain barrier due to their large molecular size, suggesting they may be relatively safe. Additionally, the complexes mediated increased DOC concentrations with prolonged blood circulation and EGFP expression in HNE­1 tumors. These results suggest the future potential application of CD-PLLD/DOC/MMP-9 for NPC therapy.