Transcript PowerPoint-presentation

Predictive Performance of a Myelosuppression Model for Dose Individualization;
Impact of Type and Amount of Information Provided
Johan E. Wallin, Lena E. Friberg and Mats O. Karlsson
Division of Pharmacokinetics & Drug Therapy, Uppsala University, Sweden
Background
A previously published semi-mechanistic model for myelosuppression (fig.1) has successfully been applied for several
cytotoxic drugs [1], among them etoposide. Observed
plasma drug concentration and/or neutrophil counts from the
previous cycle and a PKPD model of myelosuppression may
be used for dose individualization in order to avoid severe
neutropenia or a suboptimal tumor effect.
Aim
The aim was to evaluate the importance of PK- and PDinformation on the predictive performance, and improvement
with accumulated information with an increased number of
administered courses.
Methods
Multiple course data of etoposide plasma concentrations
and neutrophil counts were available for 44 patients from two
previously published studies [2,3]. Data was analysed using
NONMEM 6. BSV and BOV, including covariances, were
estimated. Model performance was evaluated by goodnessof-fit plots and predictive checks.
One thousand patients receiving 5 courses of therapy were
simulated from the final model. POSTHOC estimates were
obtained providing either
- only PK data
- neutrophil baseline (BASE)
- full course PD profiles
- full course PK+PD profiles
from one to four previous cycles (fig.2).
The POSTHOC estimates were used as input in a set of
models designed to find the dose expected to result in a
nadir of 1*109 cells/L. As a reference, a standard dose
reduction of 25% was used when a patient experience
severe toxicity. The resulting nadir of adjusted dose when
knowing the true parameters was used as outcome.
Fig 1. Depiction of the semi-mechanistic myelosuppression model
The average second course dose was somewhat lower with
PD-guided model based dosing compared to standard
treatment recommendations, 86 % vs. 90% of first course
dose. However with PD-guidance there was selective dose
escalation and still the number of patients experiencing
severe neutropenia was lowered by 8%. The overall number
of patients achieving a nadir in the close range to 1*109
cells/l was somewhat increased (tab.2).
Table 2. Nadir in second course of treatment after different methods of dose
adjustment
25% reduction if
no
Grade 4
adjustment
BASE
median
PK
PD
PKPD
0.994
0.987
0.741
0.927
0.957
0.911
min
0.00002
0.00002
0.00012
0.00005
max
6.221
6.221
7.146
6.582
6.532
6.384
stdev
0.960
1.011
1.100
1.024
1.050
1.034
0.75-1.25*109
19.1%
21.4%
21.8%
22.9%
22.5%
22.7%
<0.5*109 (gr4 tox)
35.8%
28.3%
27.3%
27.9%
25.7%
26.2%
0.00022 0.00027
One reason for not seeing a more marked difference
between the dosing methods is due to the relatively large
estimated BOV used for simulating data. If using smaller
BOV the difference become more apparent.
Table 1. Predictive performance using different levels of information
BASE
PK
PK+BASE
PD
PKPD
Prediction with information from one previous course
mpe
0.0048
0.026
0.039
0.042
0.049
rmse
1.25
1.37
1.17
1.09
1.08
Prediction with information from two previous courses
mpe
0.048
0.079
0.095
0.078
0.097
rmse
1.22
1.35
1.37
1.09
1.08
Prediction with information from four previous courses
mpe
0.005
0.053
0.068
0.037
0.058
rmse
1.21
1.33
1.11
1.04
1.01
Results and Discussion
There was significant BSV in CL (29%), baseline neutrophil
count (48%), drug effect slope factor (17%) and mean
neutrophil transit time (MTT; 17%). BOV in CL (30%), slope
(31%) and MTT (12%) were significant.
Predictive performance when based on PD data only was
similar to that of full PKPD information, and superior to the
use of only PK or baseline data (tab.1). Adding information
from more treatment courses improved the performance.
Fig 2. PK- and/or PD-information is used for prediction of sequential course
Conclusion
PK provided little benefit to predictive performance if PD
information was available. The ratio of BSV/BOV is of
importance in the precision of PK- or PD-guided dosing,
and the etoposide model contains rather large BOV. Modelbased dose individualization was shown to decrease the
proportion of patients with severe toxicity despite selective
dose escalation.
References
[1] Friberg LE et al. J Clin Oncol 20:4713–4721, 2002
[2] Ratain MJ, et al:. Clin Pharmacol Ther 45:226-233, 1989
[3] Ratain MJ, et al: J Clin Oncol 9:1480-1486, 1991