Edited by: Jeannine S. McCune, Beckman Research Institute, City of Hope, United States
Reviewed by: Paolo Curatolo, University of Rome Tor Vergata, Italy; Michael Frost, Minnesota Epilepsy Group, United States
*Correspondence: Hong Xu,
This article was submitted to Obstetric and Pediatric Pharmacology, a section of the journal Frontiers in Pharmacology
†These authors have contributed equally to this work
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Sirolimus is already used in the treatment of tuberous sclerosis complex (TSC), however, with narrow therapeutic range and considerable inter- and intra-individual pharmacokinetic variability, making it hard to develop an appropriate sirolimus initial dosage regimen, especially in children with TSC. The aim of this study was to recommend the optimal sirolimus initial dosing regimen in pediatric patients with TSC. Underlying physiological and genetic factors were collected to explore the effects on clinical sirolimus concentrations by establishing a nonlinear mixed effect (NONMEM) model, and to further simulate the optimal sirolimus initial dosing regimen using Monte Carlo method in pediatric patients with TSC. The once-daily regimen and the twice-daily regimen were recommended, respectively. For once-daily regimen, the dosages of 0.10, 0.07, 0.05, 0.04, 0.03 mg/kg/day were recommended for children with weights of 5–10, 10–20, 20–30, 30–50, and 50–60 kg, respectively. For twice-daily regimen, the dosages of 0.04, 0.03, 0.02 mg/kg/day (the daily dose was divided evenly into two doses) were recommended for children with weights of 5–20, 20–40, 40–60 kg, respectively. The initial dosages of sirolimus in children with TSC were recommended for the first time.
Tuberous sclerosis complex (TSC), a genetic autosomal dominant disorder, is caused by constitutive activation of mammalian target of rapamycin complex 1 (mTORC1) due to mutations in genes coding for hamartin (
Sirolimus, an mTOR inhibitor, has shown good effects on multiple manifestations of TSC, which has been approved for treating TSC (
Pediatric patients from June 2016 to September 2019 at the Children's Hospital of Fudan University (Shanghai, China) were collected, retrospectively. Partial basic clinical dataset of some children were collected from a previous research (
Sirolimus concentrations were tested with the Emit 2000 Sirolimus Assay (Siemens Healthcare Diagnostics Inc.) with range of linear response, 3.5–30 ng/ml, whose values of inter-assay variability [coefficient of variation (CV%)] <4.0%, and values of intra-assay CV (%) <6.2%.
The blood samples used for pharmacogenomic testing came from TDM residual samples, and the analysis was measured by Admera Health (Suzhou, China) with PGxOne®160
Dataset were used to build population pharmacokinetic model using the non-linear mixed-effects modeling software, NONMEM (edition 7, ICON Development Solutions, Ellicott City, MD, USA) and a first-order conditional estimation method with interaction (FOCE-I method). The pharmacokinetic parameters included apparent oral clearance (CL/F), volume of distribution (V/F), and absorption rate constant (Ka), whose value was fixed at 0.485/h (
Equation (1) showed the inter-individual variability:
Pi was on behalf of the individual parameter value and TV(P) represented the typical individual parameter value. ηi was symmetrical distribution, which was random term with zero mean and variance omega2 (ω2).
Equation (2) showed the random residual variability:
OBSi was the observed concentration, PREi was the individual predicted concentration and ϵ1 and ϵ2 were symmetrical distribution, which was random term with zero mean and variance sigma2 (σ2).
Equation (3) showed the relation of pharmacokinetic parameters with weight:
Pi represented the i-th individual parameter, Wi represented the i-th individual weight. Wstd was the standard weight of 70 kg. Pstd was the typical individual parameter, whose weight was Wstd. index was the allometric coefficient: 0.75 for the CL/F and 1 for the V/F (
Equations (4) showed pharmacokinetic parameters and genotype:
Equations (5) and (6) showed pharmacokinetic parameters and the other continuous covariates or categorical covariates, respectively:
Pi was the individual parameter value, TV(P) was the typical individual parameter value. θ was the parameter to be estimated and Covi was the covariate of the i-th individual. Covmedian was the population median for the covariate.
Changes of objective function value (OFV) was calculated using covariate inclusions and a decrease in the OFV >3.84 (
The final model was estimated by individual plots, distribution of weighted residuals for model (density
The influence of sirolimus initial doses on the probability to achieve the target concentration (5–15 ng/ml) (
A total of 15 children with TSC, included seven boys and eight girls and whose ages were from 1.08 to 13.95 years old. The demographic data of patients was shown in
Demographic data of patients.
Characteristic | Mean ± SD | Median (range) |
---|---|---|
7/8 | / | |
6.16 ± 2.80 | 5.75 (1.08–13.95) | |
23.83 ± 8.88 | 22.00 (10.00–50.00) | |
45.59 ± 2.29 | 46.10 (40.30–48.30) | |
8.18 ± 6.15 | 7.00 (1.00–28.00) | |
23.31 ± 6.38 | 24.00 (10.00–37.00) | |
31.97 ± 10.41 | 29.00 (16.00–59.00) | |
4.71 ± 1.27 | 4.50 (1.80–8.20) | |
71.81 ± 4.75 | 70.60 (64.70–81.10) | |
3.47 ± 2.51 | 2.60 (1.30–11.80) | |
2.01 ± 1.02 | 1.70 (0.40–4.80) | |
5.91 ± 3.65 | 5.00 (1.00–17.70) | |
37.86 ± 2.88 | 37.75 (30.80–45.30) | |
126.44 ± 10.15 | 128.50 (107.00–147.00) | |
27.59 ± 1.35 | 27.45 (25.10–31.00) | |
334.15 ± 13.01 | 332.00 (309.00–364.00) |
Pharmacogenetics analysis and Hardy–Weinberg equilibrium.
Gene | Variation | Genotype | Frequency | % | P value |
---|---|---|---|---|---|
rs1045642 | A/A | 1 | 6.67 | 0.7913 | |
A/G | 5 | 33.33 | |||
G/G | 9 | 60.00 | |||
rs1751034 | C/C | 1 | 6.67 | 0.1013 | |
C/T | 2 | 13.33 | |||
T/T | 12 | 80.00 | |||
rs757110 | A/A | 3 | 20.00 | 0.7821 | |
A/C | 8 | 53.33 | |||
C/C | 4 | 26.67 | |||
rs2231142 | G/G | 8 | 53.33 | 0.9299 | |
G/T | 6 | 40.00 | |||
T/T | 1 | 6.67 | |||
*1/*1 | 12 | 80.00 | – | ||
*3/c.820-6326A>C | 2 | 13.33 | |||
*1/c.820-6326A>C | 1 | 6.67 | |||
*1/*1 | 5 | 33.33 | – | ||
*1/*2 | 5 | 33.33 | |||
*2/*2 | 3 | 20.00 | |||
*1/*3 | 1 | 6.67 | |||
*1/*17 | 1 | 6.67 | |||
*1A/*1A | 9 | 60.00 | – | ||
*1A/*1G | 6 | 40.00 | |||
*1/*3 | 7 | 46.67 | – | ||
*3/*3 | 8 | 53.33 | |||
*1/*1 | 6 | 40.00 | – | ||
*1/*3 | 8 | 53.33 | |||
*3/*3 | 1 | 6.67 | |||
*1/*1 | 9 | 60.00 | – | ||
*1/*6 | 2 | 13.33 | |||
*6/*6 | 1 | 6.67 | |||
*28/*80 | 2 | 13.33 | |||
*28/*28/*80 | 1 | 6.67 | |||
rs1042597 | C/C | 3 | 20.00 | 0.1213 | |
C/G | 4 | 26.67 | |||
G/G | 8 | 53.33 | |||
rs1902023 | A/A | 7 | 46.67 | 0.6985 | |
A/C | 6 | 40.00 | |||
C/C | 2 | 13.33 |
Pearson chi-squared test.
The final population pharmacokinetics model was as follow:
Model evaluation.
Parameter estimates of final model and bootstrap validation.
Parameter | Estimate | SE (%) | Bootstrap | Bias (%) | |
---|---|---|---|---|---|
Median | 95% Confidence interval | ||||
6.48 | 33.2 | 6.65 | [2.67, 13.60] | 2.62 | |
124 | 69.8 | 133 | [27, 2126] | 7.26 | |
0.485 (fixed) | – | – | – | – | |
0.066 | 70.9 | 0.050 | [0.003, 0.159] | −24.24 | |
0.003 | 60.4 | 0.003 | [0.003, 0.178] | 0 | |
0.312 | 27.2 | 0.306 | [0.003, 0.415] | −1.92 | |
1.249 | 54.8 | 1.153 | [0.010, 2.332] | −7.69 |
95% confidential interval was displayed as the 2.5th, 97.5th percentile of bootstrap estimates. CL/F, apparent oral clearance (L/h); V/F, apparent volume of distribution (L); Ka, absorption rate constant (h−1); ωCL/F, inter-individual variability of CL/F; ωV/F, inter-individual variability of V/F; σ1, residual variability, proportional error; σ2, residual variability, additive error; Bias, prediction error, Bias = (Median-Estimate)/Estimate×100%.
Simulation of sirolimus concentrations at different initial dosages once a day.
Simulation of sirolimus concentrations at different initial dosages which were split evenly into two doses a day.
Initial dose recommendation of sirolimus in pediatric tuberous sclerosis complex.
Once a day | Split evenly into two doses a day | ||
---|---|---|---|
Dosage (mg/kg/day) | Body weight (kg) | Dosage (mg/kg/day) | |
0.10 | 5–20 | 0.04 | |
0.07 | 20–40 | 0.03 | |
0.05 | 40–60 | 0.02 | |
0.04 | |||
0.03 |
Sirolimus, also known as rapamycin, is a macrolide antibiotic immunosuppressant, which has been widely used for liver transplantation (
Fortunately, the combination of population pharmacokinetics and Monte Carlo simulation can successfully predict and recommend the best initial dosing regimen, the method has been widely used in clinical practice (
In our final model, body weight was included as a covariate in models that affected sirolimus clearance. As many studies have reported, there was a non-linear relationship between drug clearance and body weight in pediatric patients, and it may be well described with allometric scaling using a coefficient of 0.75 for clearance and 1 for volume (
Based on a once-daily regimen or a twice-daily regimen, we simulated two initial dosing regimens. The probability of reaching the target concentrations (5–15 ng/ml) (
There were limitations in the present study, because of the low incidence of the TSC in children, the collection of patients was extremely difficult, which was also the objective reason for our small number of patients. In addition, the study is its retrospective nature and will be verified in future prospective studies.
We established a population pharmacokinetic model of sirolimus in pediatric patients with TSC and the initial dosages of sirolimus in children with TSC were recommended for the first time. Large-scale pediatric TSC population needs to be validated.
The datasets generated for this study are available on request to the corresponding authors.
Z-PL and HX conceived and designed the study. D-DW, and XC collected the data. D-DW and XC built the model and evaluated the data. D-DW wrote the manuscript. XC reviewed and edited the manuscript. All authors contributed to the article and approved the submitted version.
This work was supported by Clinical Pharmacy Key Specialty Construction Project of Shanghai (No. YZ2017/5). Important Weak Subject Construction Project of Shanghai (No. 2016ZB0305). Scientific research project of Science and Technology Commission of Shanghai Municipality (No. 18DZ1910604). The China Scholarship Council (No. 201906100164).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.