The success of transplantation rests on the confluence of multiple factors. Therapeutic drug monitoring is one such endeavor that is required for maintaining longevity of the graft. Despite advances in developing immunosuppressive agents and regimens, maintaining drug levels within an optimal range remains a significant challenge in pediatric transplantation. This is mostly attributed to isoform-specific developmental changes in biotransformation of drugs by metabolizing enzymes,1 with tacrolimus predominantly metabolized by the cytochrome P450 enzymes—CYP3A4 and CYP3A5. High variability in tacrolimus steady-state concentrations is associated with >8-fold increase risk for any rejection beyond first year of transplantation, development of de novo donor-specific antibodies, and graft loss.2,3 Although guidelines advocate tacrolimus prescribing based on genotypes for cytochrome P450 enzymes, this approach does not account for other single-nucleotide polymorphisms (SNPs), clinical factors, and organ type that influence levels of tacrolimus. The study by Min et al4 looks to circumvent this issue by ascertaining recipient factors and genotypes that have an impact on tacrolimus levels in all organ transplants. In addition, the study also integrates clinical and genetic factors to develop a prediction model for dose-adjusted tacrolimus levels. This multicenter prospective observational study recruited children aged <18 years undergoing first organ transplant (kidney, lung, heart, or liver) over the course of 4 years. In addition to pertinent demographics, tacrolimus levels at 6 pivotal time points were recorded (36–48 h post–tacrolimus initiation, 7 ± 3 d, 14 ± 3 d, 30 ± 3 d, 3 ± 1 mo, and 12 ± 3 mo posttransplant). Genome-wide association studies were done to identify SNPs associated with tacrolimus levels, with subsequent linkage disequilibrium analysis to identify redundant SNPs. A linear regression analysis was performed in the discovery and validation cohorts to ascertain if the association of genotype was influenced by age and organ type. The authors built the prediction models (linear regression and machine learning) using clinical predictors and SNPs for dose-adjusted tacrolimus levels 36–48 hours postinitiation (T1), identifying highest variability in levels immediately posttransplant and that the prediction of accurate commencement dose of tacrolimus can aid in achievement and maintenance of optimal tacrolimus levels.5 Twenty-five SNPs were associated with tacrolimus trough levels (8 significant at the genome-wide level) after adjustment for organ type, age, gender, and CYP3A4 inhibitor use, in addition to the guideline-directed SNP rs776746 in CYP3A5. Both machine learning and linear regression models showed organ type, age at transplant, rs776746, rs12333983, and rs12957142 as pertinent predictors of dose-adjusted T1 levels. The combination of clinical and SNP models was superior to either model in isolation, with a lower prediction error with the machine learning model compared with linear regression. These findings demonstrate the importance of incorporating SNPs for determining individual tacrolimus dosing because previous attempts at pharmacogenetic algorithm dosing using limited clinical factors and CYP3A5 genotype have been unsuccessful.6 Assimilating SNPs and appropriate clinical factors accounts for the complex interplay of phenotypes (organ type) and genetic variants (SNPs). The authors further elucidated this interplay by showing a significant influence of age and organ type on the association of genotype with T1 levels, with the following pertinent findings: Infants, adolescents, and heart recipients had the highest difference in T1 levels between expressors and nonexpressors of rs776746*1. Also, infants had lower dose-adjusted T1 levels—attributed to large liver size and high plasma clearance of drug. CYP3A7 is more active in infants, with transition to higher CYP3A4 activity in adults, making these enzymes predominant tacrolimus metabolizers in CYP3A5 nonexpressors. CYP3A4 expression may be enhanced in children aged 5–15 years as a result of changes in growth and sex hormones. It is important to highlight that the discovery and validation cohorts had differences in characteristics of age (4.5 versus 6.1 y; P = 0.003), race (White 63% versus 77%, Asian 20% versus 9%, mixed 9% versus 3%; P < 0.001), and organ type (kidney 29% versus 47%, heart 33% versus 25%, liver 35% versus 26%; P < 0.001). Also, the underrepresentation of patients with African ancestry led to a nonsignificance of 2 common guideline-advocated SNPs (CYP3A5*6 and CYP3A5*7) in genome-wide studies. It is reassuring, however, that genotype findings were independent of clinical confounders and demonstrated significance in the validation cohorts. This is the first genome-wide association study–based pharmacogenetic study including all types of pediatric solid organ transplant recipients, which has identified additional SNPs and clinical factors that influence tacrolimus levels. Above all, using a combination of clinical and genetic factors is superior to either alone to predict tacrolimus concentrations. It furthers the concept that genetics has application in not only the pretransplant immunological matching but also the realm of posttransplant pharmacology. This study makes a persuasive case for an individualized approach to tacrolimus dosing, with emphasis on taking into account established and newly detected SNPs that influence its levels. This has the potential to translate into not only beneficial effects with respect to allograft and patient outcomes but also system-wide applications to reduce cost of drug monitoring and hospital stay. However, for pharmacogenetics to be useful, they must be simple, practical, and economical. This seems unlikely. Currently and for the immediate future, the best way to make sure a therapeutic tacrolimus level is achieved is to start an adequate dose early for any organ and in any age group.