Prediction tacrolimus blood levels based on the Bayesian method in adult kidney transplant patients

Compare the performance of multiple machine learning models in predicting tacrolimus concentration for infant patients with living donor liver transplantation

BACKGROUND

This study aims to establish multiple ML models and compare their performance in predicting tacrolimus concentration for infant patients who received LDLT within 3 months after transplantation.

METHODS

Retrospectively collected basic information and relevant biochemical indicators of included infant patients. CMIA was used to determine tacrolimus C₀ . PCR was used to determine the donors' and recipients' CYP3A5 genotypes. Multivariate stepwise regression analysis and stepwise elimination covariates were used for covariates selection. Thirteen machine learning algorithms were applied for the development of prediction models. APE, the ratio of the APE ≤3 ng ml^-1 and ideal rate (the proportion of the predicted value with a relative error of 30% or less) were used to evaluate the predictive performance of the model.

RESULTS

A total of 163 infant patients were included in this study. In the case of the optimal combination of covariates, the Ridge model had the lowest APE, 2.01 (0.85, 3.35 ng ml^-1 ). The highest ratio of the APE ≤3 ng ml^-1 was the LAR model (71.77%). And the Ridge model showed the highest ideal rate (55.05%). For the Ridge model, GRWR was the most important predictor.

CONCLUSIONS

Compared with other ML models, the Ridge model had good predictive performance and potential clinical application.

Tacrolimus Therapy in Adult Heart Transplant Recipients: Evaluation of a Bayesian Forecasting Software

BACKGROUND

Therapeutic drug monitoring is recommended to guide tacrolimus dosing because of its narrow therapeutic window and considerable pharmacokinetic variability. This study assessed tacrolimus dosing and monitoring practices in heart transplant recipients and evaluated the predictive performance of a Bayesian forecasting software using a renal transplant-derived tacrolimus model to predict tacrolimus concentrations.

METHODS

A retrospective audit of heart transplant recipients (n = 87) treated with tacrolimus was performed. Relevant data were collected from the time of transplant to discharge. The concordance of tacrolimus dosing and monitoring according to hospital guidelines was assessed. The observed and software-predicted tacrolimus concentrations (n = 931) were compared for the first 3 weeks of oral immediate-release tacrolimus (Prograf) therapy, and the predictive performance (bias and imprecision) of the software was evaluated.

RESULTS

The majority (96%) of initial oral tacrolimus doses were guideline concordant. Most initial intravenous doses (93%) were lower than the guideline recommendations. Overall, 36% of initial tacrolimus doses were administered to transplant recipients with an estimated glomerular filtration rate of <60 mL/min/1.73 m despite recommendations to delay the commencement of therapy. Of the tacrolimus concentrations collected during oral therapy (n = 1498), 25% were trough concentrations obtained at steady-state. The software displayed acceptable predictions of tacrolimus concentration from day 12 (bias: -6%; 95%confidence interval, -11.8 to 2.5; imprecision: 16%; 95% confidence interval, 8.7-24.3) of therapy.

CONCLUSIONS

Tacrolimus dosing and monitoring were discordant with the guidelines. The Bayesian forecasting software was suitable for guiding tacrolimus dosing after 11 days of therapy in heart transplant recipients. Understanding the factors contributing to the variability in tacrolimus pharmacokinetics immediately after transplant may help improve software predictions.

Improved prediction of tacrolimus concentrations early after kidney transplantation using theory-based pharmacokinetic modelling

Aims

The aim was to develop a theory-based population pharmacokinetic model of tacrolimus in adult kidney transplant recipients and to externally evaluate this model and two previous empirical models.

Methods

Data were obtained from 242 patients with 3100 tacrolimus whole blood concentrations. External evaluation was performed by examining model predictive performance using Bayesian forecasting.

Results

Pharmacokinetic disposition parameters were estimated based on tacrolimus plasma concentrations, predicted from whole blood concentrations, haematocrit and literature values for tacrolimus binding to red blood cells. Disposition parameters were allometrically scaled to fat free mass. Tacrolimus whole blood clearance/bioavailability standardized to haematocrit of 45% and fat free mass of 60 kg was estimated to be 16.1 l h⁻¹ [95% CI 12.6, 18.0 l h⁻¹]. Tacrolimus clearance was 30% higher (95% CI 13, 46%) and bioavailability 18% lower (95% CI 2, 29%) in CYP3A5 expressers compared with non-expressers. An E_max model described decreasing tacrolimus bioavailability with increasing prednisolone dose. The theory-based model was superior to the empirical models during external evaluation displaying a median prediction error of −1.2% (95% CI −3.0, 0.1%). Based on simulation, Bayesian forecasting led to 65% (95% CI 62, 68%) of patients achieving a tacrolimus average steady-state concentration within a suggested acceptable range.

Conclusion

A theory-based population pharmacokinetic model was superior to two empirical models for prediction of tacrolimus concentrations and seemed suitable for Bayesian prediction of tacrolimus doses early after kidney transplantation.

Effects of CYP3A4 and CYP3A5 polymorphisms on tacrolimus pharmacokinetics in Chinese adult renal transplant recipients: a population pharmacokinetic analysis

OBJECTIVE

Tacrolimus is used clinically for the long-term treatment of antirejection of transplanted organs in liver and kidney transplant recipients, although dose optimization is poorly managed. The aim of this study was to examine the association between tacrolimus pharmacokinetic variability and CYP3A4 and CYP3A5 genotypes by a population pharmacokinetic analysis based on routine drug monitoring data in adult renal transplant recipients.

MATERIALS AND METHODS

Trough tacrolimus concentrations were obtained from 161 adult kidney transplant recipients after transplantation. The population pharmacokinetic analysis was carried out using the nonlinear mixed-effect modeling software NONMEM version 7.2. The CYP3A4*1G and CYP3A5*3 genetic polymorphisms from the patients studied were determined by direct sequencing using a validated automated genetic analyzer.

RESULTS

A one-compartment model with first-order absorption and elimination adequately described the pharmacokinetics of tacrolimus. Covariates including CYP3A5*3 and CYP3A4*1G alleles and hematocrit were retained in the final model. The apparent clearance of tacrolimus was about two-fold higher in kidney transplant patients with higher enzymatic activity of CYP3A5*1 and CYP3A4*1G (with the CYP3A5*1/*1 or *1/*3 and CYP3A4*1/*1G or CYP3A4*1G/*1G) compared with those with lower enzymatic activity (CYP3A5*3/*3 and CYP3A4*1/*1).

CONCLUSION

This is the first study to extensively determine the effect of CYP3A4*1G and CYP3A5*3 genetic polymorphisms and hematocrit value on tacrolimus pharmacokinetics in Chinese renal transplant recipients. The findings suggest that CYP3A5*3 and CYP3A4*1G polymorphisms and hematocrit are determinant factors in the apparent clearance of tacrolimus. The initial dose design is mainly based on CYP3A5 and CYP3A4 genotypes as well as hematocrit. This result may also be useful for maintenance tacrolimus dose optimization and may help to avoid fluctuating tacrolimus levels and improve the efficacy and tolerability of tacrolimus in kidney transplant recipients.

Dosing equation for tacrolimus using genetic variants and clinical factors

AIM

To develop a dosing equation for tacrolimus, using genetic and clinical factors from a large cohort of kidney transplant recipients. Clinical factors and six genetic variants were screened for importance towards tacrolimus clearance (CL/F).

METHODS

Clinical data, tacrolimus troughs and corresponding doses were collected from 681 kidney transplant recipients in a multicentre observational study in the USA and Canada for the first 6 months post transplant. The patients were genotyped for 2,724 single nucleotide polymorphisms using a customized Affymetrix SNP chip. Clinical factors and the most important SNPs (rs776746, rs12114000, rs3734354, rs4926, rs3135506 and rs2608555) were analysed for their influence on tacrolimus CL/F.

RESULTS

The CYP3A5*1 genotype, days post transplant, age, transplant at a steroid sparing centre and calcium channel blocker (CCB) use significantly influenced tacrolimus CL/F. The final model describing CL/F (l h(-1)) was: 38.4 ×[(0.86, if days 6-10) or (0.71, if days 11-180)]×[(1.69, if CYP3A5*1/*3 genotype) or (2.00, if CYP3A5*1/*1 genotype)]× (0.70, if receiving a transplant at a steroid sparing centre) × ([age in years/50](-0.4)) × (0.94, if CCB is present). The dose to achieve the desired trough is then prospectively determined using the individuals CL/F estimate.

CONCLUSIONS

The CYP3A5*1 genotype and four clinical factors were important for tacrolimus CL/F. An individualized dose is easily determined from the predicted CL/F. This study is important towards individualization of dosing in the clinical setting and may increase the number of patients achieving the target concentration. This equation requires validation in an independent cohort of kidney transplant recipients.