Transcript PowerPoint-presentation

Population Pharmacokinetics of Tacrolimus in Paediatric Haematopoietic Stem Cell
Transplantation and Implementation in a Dose Adaptation Tool
Johan E. Wallin1, Lena E. Friberg1, Anders Fasth2 and Christine E. Staatz3
1. Div. Pharmacokinetics & Drug Therapy, Uppsala University, Sweden
2. Dept.Pediatrics, University of Gothenburg, Gothenburg, Sweden
3. School of Pharmacy, University of Queensland, Brisbane, Australia.
Background
Tacrolimus is a valuable immunosuppressant option to
ciclosporin in the prevention of graft-versus-host disease (GVHD)
following allogeneic hematopoietic stem cell transplantation
(SCT). Limited data is available on the pharmacokinetics of
tacrolimus in paediatric hematopoietic stem cell recipients.
Aim
The aim of this study was to characterise the population
pharmacokinetics of tacrolimus in paediatric SCT recipients in the
first year post-transplant; to identify factors that may explain
pharmacokinetic variability; and to develop an Excel-based
dosing macro, utilising the model and Bayesian forecasting
techniques, to assist with tacrolimus dosing in new paediatric
SCT recipients.
Methods
Data were collected retrospectively from the medical records of
22 children transplanted between 1997 and 2007 at the Queen
Silvia’s Children’s Hospital in Gothenburg, Sweden. Population
pharmacokinetic analysis was performed using NONMEM
version 6. Tacrolimus distribution volume (V) and clearance (CL)
were allometrically scaled to lean body weight.
A previously suggested model with a time-dependent increase in
tacrolimus apparent clearance (CL/F) developed for liver
[1]
transplant patients was compared to a simpler one- and twocompartment model. A step-wise covariate search was performed
where covariates screened for influence on the pharmacokinetic
parameters were liver function tests (AST, ALT, GGT, ALP),
bilirubin, albumin, creatinine clearance (Schwartz formula), sex,
age and post-operative day (POD).
Table 1. Final model parameters
Typical value
RSE
Inter individual
variability
0.124 L//h/kg0.75
11%
57%
28%
0.0044
35%
V
8.97 L/kg
20%
104%
37%5
F
12.5%
12
61%
33%
F-POD
-0.00186
6%
Prop. error
0.353
32%
18%
62%
Time dep. prop.error
0.298
25%
2.3 days
44%
CL
CL – CrCL
Half life time dep. Error
RSE
F decreased with time post-transplant, to only 4.5% after a year.
The estimated bioavailability was approximately half that
reported in adult SCT recipients, whereas CL and V values were
similar to adult values[2]. A time dependent error model was
chosen, where the error component decline during the first days
of therapy. Model goodness of fit plots can be seen in Fig.1.
A macro was constructed in MS Excel to select an initial
tacrolimus dose based on the final covariate model (Fig.2).
Measured drug concentration can be used in a Bayesian function
to estimate individual parameters for dose adaptation to
maximize the likelihood of achieving target concentrations at the
time of transplantation. The population model indicated the need
for a 50% higher initial dose than the current for the typical
individual to achieve the target concentration at the time of
transplantation, alternatively starting the treatment 24 hours
earlier.
Figure 2. Macro with covariate guided initial dose and Bayesian adaptation
to achieve the target concentration (12-15 ng/ml)
Figure 1. Goodness of fit-plots for the final model
Observations vs. Individual predictions (Run 2164)
Conditional weighted residuals vs. Time (Run 2164)
424
401
4
419
424
425
401 421
424 408
426
419
413
421
402
418
424
417
3
414
403
402
log Observations
424412
424
405
420
417
414
405
408
426
414
419
410421
426
417 415
402
410
2
409
420 409
402
426 425
424
417
1
410
410
424
418
402 410
424
415
408
421
416
404
0
1
417
2
425403
402
424
410
426
0
426
413
412
421
418411
408
411
410 403
424
419
419
404
425
420
403
405 418
414 418
414
402
402
417
403
424
409
420
414
421
417
405
411
412
405
419
415
418
418
415
419
410
408
426413
410
420
425
421
412 421
418
420
Conditional weighted residuals
402
-2
421
424 419426
415
412
421
408
413
420
419
421
404
2
3
log Individual predictions
10^1.0
10^1.5
10^2.0
10^2.5
10^3.0
420
403
420
404
416
409
401
409
418
404
10^3.5
10^4.0
log(Time)
Results and Discussion
All subjects received tacrolimus initially as a continuous
intravenous infusion at 0.03 mg/kg/day, starting two days before
transplantation. Patients were converted to oral tacrolimus
therapy two to three weeks after transplantation, divided into
twice daily administration. A one-compartment model with first
order absorption and elimination was considered sufficient to
describe the data.
In the final model, mean CL was 0.124 L/h/kg0.75, V was 8.97
L/kg and bioavailability (F) was 12.5% (Tab.1). Inter-individual
variability in CL, V and F was 57%, 104% and 61%, respectively.
Covariate analysis suggested tacrolimus CL positively correlated
with CrCL. Typical value of CL varied from 0.082-0.216 L/h/kg0.75
at the minimum and maximum CrCL values.
Conclusion
Routine therapeutic drug monitoring can be implemented in our
dosage adaptation tool to assist with tacrolimus dosing in new
paediatric SCT recipients. In future work, sample times for
maximum information on an individual level will be evaluated
using optimal design analysis.
References
1.
Antignac, M., et al., Population pharmacokinetics of tacrolimus in full liver transplant patients:
modelling of the post-operative clearance. Eur J Clin Pharmacol, 2005. 61(5-6): p. 409-16.
2.
Jacobson, P., et al., Factors affecting the pharmacokinetics of tacrolimus (FK506) in hematopoietic
cell transplant (HCT) patients. Bone Marrow Transplant, 2001. 28(8): p. 753-8.