Tacrolimus is a key immunosuppressant to prevent transplant rejection in solid organ transplantation, but its treatment index is narrow, and there are significant differences in bioavailability between and within individuals. Insufficient dose will increase the risk of acute rejection, and excessive exposure will increase the incidence of adverse drug reactions. Therefore, it is necessary to give patients personalized and precise treatment. However, there are some limitations in individualized drug administration based solely on therapeutic drug concentration monitoring. Machine learning can learn from the existing data and automatically build the calculation model of complex relationships. It has the characteristics of high efficiency and high accuracy. Machine learning is more and more used in drug concentration prediction because of its ability to process large and complex data sets. This article reviews the application of machine learning in tacrolimus individualized and precise drug use in transplant patients.

1.Staatz C, Tett S. Clinical pharmacokinetics and pharmacodynamics of tacrolimus in solid organ transplantation[J]. Clin Pharmacokinet, 2004, 43(10): 623-653. DOI: 10.2165/00003088-200443100-00001.

2.Shuker N, van Gelder T, Hesselink DA. Intra-patient variability intacrolimus exposure: causes, consequences for clinical man agement[J]. Transplant Rev, 2015, 29(2): 78-84. DOI: 10.1016/j.trre.2015.01.002.

3.何霞, 童荣生. CYP3A4和CYP3A5基因多态性对汉族肾移植患者他克莫司血药浓度的影响[J]. 中国药师, 2013, 16(4): 497-501. [He X, Tong RS. Effect of CYP3A4 and CYP3A5 Gene Polymorphisms on Tacrolimus Plasma Concentration in Renal Transplanted Recipients in the Han Population[J]. China Pharmacist, 2013, 16(4): 497-501.] DOI: 10.3969/j.issn.1008-049X.2013.04.007.

4.Bergmann TK, Hennig S, Barraclough KA, et al. Population pharmacokinetics of tacrolimus in adult kidney transplant patients: impact of CYP3A5 genotype on starting dose[J]. Ther Drug Monit, 2014, 36(1): 62-70. DOI: 10.1097/FTD. 0b013e31829f1ab8.

5.Borobia AM, Romero I, Jimenez C, et al. Trough tacrolimus concentrations in the first week after kidney transplantation are related to acute rejection[J]. Ther Drug Monit, 2009, 31(4): 436-442. DOI: 10.1097/FTD.0b013e3181a8f02a.

6.Kuypers DR, Claes K, Evenepoel P, et al. Clinical efficacy and toxicity profile of tacrolimus and mycophenolic acid in relation to combined long-term pharmacokinetics in de novorenal allograft recipients[J]. Clin Pharmacol Ther, 2004, 75(5): 434-447. DOI: 10.1016/j.clpt.2003.12.009.

7.Deo RC. Machine learning in medicine[J]. Circulation, 2015, 132(20): 1920-1930. DOI: 10.1161/CIRCULATION AHA.115.001593.

8.Cosgun E, Limdi NA, Duarte CW. High-dimensional pharmacogenetic prediction of a continuous trait using machine learning techniques with application to warfarin dose prediction in African Americans[J]. Bioinformatics, 2011, 27(10): 1384-1389. DOI: 10.1093/bioinformatics/btr159.

9.Meng HY, Jin WL, Yan CK, et al. The application of machine learning techniques in clinical drug therapy[J]. Curr Comput Aided Drug Des, 2019, 15(2): 111-119. DOI: 10.2174/1573409914666180525124608.

10.Wallemacq P, Armstrong VW, Brunet M, et al. Opportunities to optimize tacrolimus therapy in solid organ transplantation: report of the European consensus conference[J]. Ther Drug Monit, 2009, 31(2): 139-152. DOI: 10.1097/FTD.0b013e318198d092.

11.Venkataramanan R, Shaw LM, Sarkozi L, et al. Clinical utility of monitoring tacrolimus blood concentrations in liver transplant patients[J]. J Clin Pharmacol, 2001, 41(5): 542-551. DOI: 10.1177/00912700122010429.

12.Wong KM, Shek CC, Chau KF, et al. Abbreviated tacrolimus areaunder-the-curve monitoring for renal transplant recipients[J]. Am J Kidney Dis, 2000, 35(4): 660-666. DOI: 10.1016/s0272-6386(00)70013-8.

13.Knoop C, Thiry P, Saint-Marcoux F, et al. Tacrolimus pharmacokinetics and dose monitoring after lung transplantation for cystic fifibrosis and other conditions[J].Am J Transplant, 2005, 5(6): 1477-1482. DOI: 10.1111/j.1600-6143.2005.00870.x.

14.Woillard JB, Saint-Marcoux F, Debord J, et al. Pharmacokinetic models to assist the prescriber in choosing the best tacrolimus dose[J]. Pharmacol Res, 2018, 130: 316-321. DOI: 10.1016/j.phrs.2018.02.016.

15.James G, Witten D, Hastie T, et al. An introduction to statistical learning: with applications in R[M]. New York: Springer Text in Statistics, 2013: 1-416.

16.Hastie T. The Elements of statistical learning : data mining[J]. J Royal Statistical Society, 2004, 167(1): 192. DOI: 10.1111/j.1467-985X.2004.298_11.x.

17.薛贞霞. 支持向量机及半监督学习中若干问题的研究[D]. 西安：西安电子科技大学, 2009. [Xue ZX. Research on support vector machine and semi supervised learning[D]. Xi'an: Xi'an University of Electronic Science and Technology, 2009.]

18.傅晓华, 叶毅芳, 罗美娟,等. 人工神经网络预测肝移植术受者他克莫司血药浓度[J]. 药学学报, 2012, 47(9): 1134-1140. [Fu XH, Ye YF, Luo MJ, et al. Prediction of blood tacrolimus concentration in liver transplantation recipients by artificial neural network[J]. Acta Pharmaceutica Sinica, 2012, 47(9): 1134-1140.] DOI: 10.16438/j.0513-4870.2012.09.015.

19.Tang J, Liu R, Zhang YL, et al. Application of machine-learning models to predict tacrolimus stable dose in renal transplant recipients[J]. Sci Rep, 2017, 7: 42192. DOI: 10.1038/srep42192.

20.Thishya K, Vattam KK, Naushad SM, et al. Artificial neural network model for predicting the bioavailability of tacrolimus in patients with renal transplantation[J]. PloS One, 2018, 13(4): e0191921. DOI: 10.1371/journal.pone.0191921.

21.Woillard JB, Labriffe M, Debord J, et al. Marquet, tacrolimus exposure prediction using machine learning[J].Clin Pharmacol Ther, 2021, 110(2): 361-369. DOI: 10.1002/cpt.2123.

22.Fu Q, Jing Y, Liu Mr G, et al. Machine learning-based method for tacrolimus dose predictions in Chinese kidney transplant perioperative patients[J]. J Clin Pharm Ther, 2022, 47(5): 600-608. DOI: 10.1111/jcpt.13579.

23.Badillo S, Banfai B, Birzele F, et al. An introduction to machine learning[J]. Clin Pharmacol Ther, 2020, 107(4): 871-885. DOI: 10.1002/cpt.1796.

24.Brunet M , Gelder TV , Sberg A , et al. Therapeutic drug monitoring of tacrolimus-personalized therapy: second consensus report[J]. Ther Drug Monit, 2019, 41(3): 261-307. DOI: 10.1097/FTD.0000000000000640.

25.Campagne O, Mager DE, Tornatore KM. Population pharmacokinetics of tacrolimus in transplant recipients: what did we learn about sources of interindividual variabilities?[J]. J Clin Pharmacol, 2019, 59(3): 309-325. DOI: 10.1002/jcph.1325.

26.Van Looy S, Verplancke T, Benoit D, et al. A novel approach for prediction of tacrolimus blood concentration in liver transplantation patients in the intensive care unit through support vector regression[J]. Crit Care, 2007, 11(4): R83. DOI: 10.1186/cc6081.

27.Brooks E, Tett SE, Isbel NM, et al. Population pharmacokinetic modelling and bayesian estimation of tacrolimus exposure: is this clinically useful for dosage prediction yet?[J]. Clin Pharmacokinet, 2016, 55(11): 1295-1335. DOI: 10.1007/s40262-016-0396-1.