| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Departments of
1
Pharmacology and Toxicology and
3
Nephrology, University Hospital, 87042 Limoges cedex, France
2 Laboratory of Biophysics, Faculty of Pharmacy, 87025 Limoges cedex, France
4 Departments of Chest Medicine and Clinical Chemistry, Erasme University Hospital, 1070 Brussels, Belgium
aaddress correspondence to this author at: Service de Pharmacologie et Toxicologie, CHU Dupuytren, 87042 Limoges cedex, France; fax 33-555-05-61-62, e-mail marquet{at}unilim.fr
Cyclosporine (CsA) blood concentrations measured 2 h after Neoral® administration (c2) are a sensitive predictor of clinical outcome in organ transplantation, as suggested by a recent prospective clinical trial in liver transplant patients (1). c2 is now recommended as the target exposure index for the therapeutic drug monitoring (TDM) of CsA (2)(3)(4)(5)(6)(7)(8). The aim of this study was to investigate, for different types of grafts, the concentration–time relationships around c2 to evaluate the concentration error as a function of the sampling-time error and to identify the sampling-time range compatible with acceptable performance of this c2 TDM strategy.
Data obtained from three different clinical trials were studied retrospectively. Each patient gave written informed consent, and each trial was approved by a local ethics committee (9)(10)(11). The three populations were as follows:
All the patients from these three clinical trials were dosed twice daily with microemulsified CsA (Neoral). The data are summarized in Table 1
.
|
For both kidney and lung transplant patients, CsA whole-blood concentrations were measured using an enzyme-multiplied immunoassay technique (Emit; Dade-Behring Diagnostics). This method has an upper limit of quantification of 500 µg/L. In the clinical trial conducted with cardiac transplant recipients, whole-blood CsA was measured with a fluorescence polarization immunoassay (FPIA; Abbott TDx). This method has an upper limit of quantification of 1500 µg/L. For both assays, samples with CsA concentrations greater than the upper assay range were diluted 1:4 with human blank whole blood (100 µL of sample + 300 µL of blank whole blood) and then reanalyzed.
Individual pharmacokinetic profiles were fitted using nonlinear regression (NLR) to a two-compartment pharmacokinetic model where the absorption phase is described by a 
distribution (12). This pharmacokinetic model was specifically designed previously to deal with oral CsA profiles and was validated in the patient data sets analyzed here (9)(11)(13).
The consistency between observed and calculated concentrations within the first 4 h post dose was studied within each population. Seven or eight time points (i.e., 0.33, 0.66, 1, 1.5, 2, 3, and 4 h for all plus 2.5 h in heart transplant patients) were taken into account, representing a total of 140, 360, 203, and 210 concentrations for renal, heart, and lung cystic and noncystic transplant recipients, respectively. The relative differences between observed and predicted concentrations at all time points between 0 and 4 h post dose and the root mean squared error (14) were calculated using Excel (Microsoft). Regression and correlation analyses were performed using Statview (Abacus Concept).
For each patient and each profile, concentration values within ± 15 min around 120 min post dose (i.e., at 105, 110, 115, 125, 130, and 135 min) were estimated by NLR using the same pharmacokinetic model.
The relative concentration error (RCE) with respect to the concentration actually measured at 120 min (c120) post dose was then computed as follows:
 |
A total of 124 full blood CsA concentration profiles over 12 h were analyzed for the present study. We found excellent correlation over the first 4 h after dosing between observed concentrations and concentration values calculated using NLR. Correlation coefficients (r2) varied from 0.972 for cystic fibrosis transplant patients to 0.985 for renal recipients. Intercepts and slopes were not significantly different from 0 and 1, respectively. The good predictive performance resulted from the very small and nonsignificant differences between measured and calculated concentrations [mean (SD) differences, 0.6 (4.5)% 0.7 (3.9)%, 0.2 (13)%, and 0.8 (9.6)% for renal, heart, lung cystic fibrosis, and lung non-cystic fibrosis transplant patients, respectively] and good precision (root mean squared error, 1.1–5.7%).
The calculated RCE values at each studied sampling time for each profile for the five different populations are shown in Fig. 1
. For each transplant patient, whatever the type of graft, the RCE value was within ± 20% in a sampling-time interval from 110 to 130 min. A sampling-time error of ± 15 min produced a RCE >20% (up to 30%) in a few heart and lung transplant patients. As can be inferred from Fig. 1
, in many patients tmax was later than c2, which could be designated as "delayed absorption". The present data set included patients with a tmax up to 5 h. In renal transplant patients, overestimation of c2 was observed for t <120 min and underestimation for t >120 min. In lung and heart transplant patients, over- as well as underestimation of c2 could be observed regardless of the time error, depending on the patient.
|
On the basis of data obtained in renal, heart, and lung transplant patients and using a validated pharmacokinetic method, the present study shows that when the sampling-time error around 2 h post dose increases, the relative concentration error and its interindividual variability also increase significantly. Although the impact of sampling-time error differed with the type of graft, an acceptable (± 20%) estimation of the true c2 value was obtained within a time-error range of ± 10min for all 124 profiles. Interestingly, for renal transplant patients, the mean RCE was always <10% within this time range.
The c2 target values defined for CsA TDM in renal and de novo liver transplant patients have been proposed with a range of ± 20% (e.g., 1.7 mg/L ± 20% for the 0–1 month post-transplantation period in renal transplantation) (4). We compared this range with the interlaboratory CV values of the International Cyclosporin Proficiency Testing Scheme, taken as estimates of the analytical error (15). For concentrations <500 µg/L (n = 12; results for the year 2002), the mean CVs were 10% for the Emit and 6.0% for the FPIA. The interlaboratory CV obtained with a whole blood sample to which 2000 µg/L cyclosporin had been added was 7.9% for the Emit (n = 38) and 7.2% for the FPIA (n = 30).
In summary, numerous studies have promoted a Neoral monitoring strategy using CsA blood concentrations measured 2 h after drug administration, called c2, to improve the clinical benefits for transplant patients. Guidelines for c2 interpretations propose target values with a range of ± 20%. The present study shows that the accuracy of c2 monitoring is highly dependent on the correct sampling time and that a substantial difference (up to 30%) from the 2-h values (which are themselves subject to analytical inaccuracy and imprecision) can occur with a sampling-time error of 15 min. Consequently, such time errors could lead to inappropriate dose adjustment and to inadequate immunosuppression or increased risk of adverse effects. Timing errors of ± 10 min seem to be the acceptable limit for use of c2 and subsequent dose adjustment of CsA.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
2 h post dose (TE min).
) are outliers. (A), lung transplant patients with cystic fibrosis (29 profiles); (B), lung transplant patients without cystic fibrosis (30 profiles); (C), stable renal transplant patients (20 profiles); (D), heart transplant patients at week 1 post transplantation (16 profiles); (E), heart transplant patients at month 3 and year 1 post transplantation (29 profiles).