.JPG?ad=18481&adview=true)
Clinical Cancer Research Vol. 11, 3038-3044, April 15, 2005
© 2005 American Association for Cancer Research
Serum Cystatin C is a Better Marker of Topotecan Clearance than Serum Creatinine
Antje Hoppe,
Sophie Séronie-Vivien,
Fabienne Thomas,
Jean-Pierre Delord,
Laurence Malard,
Pierre Canal and
Etienne Chatelut
Authors' Affiliation: EA 3035 and Department of Clinical Biology, Institut Claudius-Regaud, Toulouse, France
Requests for reprints: Etienne Chatelut, Institut Claudius-Regaud, F-31052 Toulouse, France. Phone: 33-561-42-4271; Fax: 33-561-42-4631; E-mail: chatelut{at}icr.fnclcc.fr.
 |
Abstract
|
|---|
Purpose: To evaluate plasma cystatin level as a covariate to predict topotecan pharmacokinetics. Cystatin C, a member of the cystatin superfamily of cysteine proteinase inhibitors, has been recently proposed as an alternative endogenous marker of glomerular filtration. Renal function is known as a key factor of topotecan clearance.
Experimental Design: Data were obtained from 59 patients who underwent drug monitoring for individual dosing of topotecan. Topotecan plasma concentrations versus time were analyzed using a nonlinear mixed effect model according to a two-compartment pharmacokinetic model and a first-order conditional estimation method. A proportional error model was used for residual and interpatient variabilities. Data-splitting was done randomly to create a model-building data set (44 patients) and a model validation data set (15 patients).
Results: Using the building data set, four covariates significantly decreased the objective function value and interindividual variability on topotecan clearance (CL) when tested individually: ideal body weight (IBW), serum creatinine, age, and cystatin C level. The best model was: CL (L/hour) = 20.2 [cystatin C (mg/L) / 1.06]–0.60 [IBW (kg) / 57]0.95. Prospective evaluation using the validation data set confirmed that the model based on cystatin C had a better predictive value than the models based on serum creatinine or body surface area.
Conclusion: Cystatin C is a marker of drug elimination which is superior to serum creatinine for topotecan. It deserves to be further explored as a promising covariate for drug dosing as well as selection criteria for clinical studies of drugs eliminated mainly or partially by the kidney.
Key Words: Population pharmacokinetics • individual dosing • renal function
Individual dosing of drugs mainly eliminated unchanged within the urine is made possible by assessing the renal function. Most of the methods are based on the serum creatinine level. As its rate of production is highly related to muscle mass, serum creatinine is affected by body weight, age, and sex. Consequently, the different equations and nomograms developed to estimate glomerular filtration rate from serum creatinine all integrate these covariates. The most widely used are the Cockcroft-Gault equation (1), and the equation recently proposed from the Modification of Diet in Renal Disease study (2). Cystatin C has been proposed as an alternative endogenous marker of glomerular filtration. Cystatin C is a member of the cystatin superfamily of cysteine proteinase inhibitors. It is a 120–amino acid basic protein with a molecular weight of 13 kDa described as the product of a "housekeeping gene" that is expressed in all nucleated cells. Cystatin C is supposed to meet criteria for an ideal glomerular filtration rate marker better than creatinine because (a) it is produced at a constant rate, (b) it is not secreted, and (c) it is reabsorbed by tubule epithelial cells but subsequently catabolized so that it does not return to the blood flow (for review, see ref. 3). Indeed, the reciprocal of cystatin C (1/cystatin C) is highly correlated with the comparative glomerular filtration rate reference standard (4). Moreover, despite conflicting results of individual studies (5), a recent metaanalysis showed that serum cystatin C is superior to serum creatinine as a marker of kidney function (6). We recently showed that cystatin C, a drug that is eliminated mainly by glomerular filtration, is a marker of drug elimination which is at least as good as serum creatinine in predicting carboplatin clearance.1
Topotecan (Hycamtin) is a semisynthetic analogue of camptothecin that binds to topoisomerase I-DNA complex and has shown antitumor activity in several tumor types, including ovarian cancer and small cell lung cancer. Its renal elimination accounts for 
50% of total drug disposition (7). Moreover, topotecan clearance is correlated with creatinine clearance calculated by the Cockcroft-Gault equation (8). Given the narrow therapeutic index of topotecan, a reliable method of prediction of topotecan clearance could be beneficial to its use through the design of individualized dosing regimens. Indeed, several studies have shown a close correlation between the topotecan area under the plasma concentration (AUC) versus time curve and the dose-limiting toxicity, predominantly neutropenia (see review in ref. 9). The objective of this pharmacokinetic analysis was to compare cystatin C with serum creatinine as a covariate to predict topotecan clearance using a population pharmacokinetics approach and NONMEM program. Data were obtained from 59 patients corresponding to various schedules of i.v. topotecan administration. Splitting of data was randomly done to create a model-building data set (44 patients) and a model validation data set (15 patients).
|
Patients and Methods
|
|---|
Patients and topotecan administration. Data were obtained from 59 patients who underwent drug monitoring for individual dosing of topotecan during two separate clinical trials in the Institut Claudius-Regaud (Toulouse, France). All patients provided written informed consent to participate in the respective protocol, as approved by a regional ethical committee. The main patients' characteristics are shown in Table 1. Topotecan was administered by automatic i.v. infusion pump for 30 minutes, repeated for 5 to 15 consecutive days as monochemotherapy at daily doses ranging from 0.3 to 1.5 mg/m2.
Blood sampling and topotecan analysis. A pharmacokinetic exploration was done on day 1 of cycle 1. For 44 patients, seven blood samples were taken after the first daily infusion: immediately before, 5 minutes before the end of the 30-minute infusion, and 0.5, 1, 2, 4, and 8 hours after the end of infusion. For 15 patients, only three samples were obtained (immediately before, 5 minutes before the end of the 30-minute infusion, and 4 hours after the end of infusion). These latter two samples were selected according to a limited sampling strategy based on a previously developed Bayesian estimation method (8). The total topotecan (i.e., lactone plus ring-opened carboxylate forms) plasma concentrations were determined using a high-performance liquid chromatography method using fluorescence detection (10). Briefly, 200 µL of plasma were deproteinized with 200 µL of perchloric acid with methanol (50:50 v/v). One hundred microliters of the supernatant were injected into the high-performance liquid chromatography column (C18, 5 µm, 4 x 125 mm, ProntoSIL Eurobond, Bischoff, Germany). The mobile phase was composed of 0.012 mol/L N,N,N',N'-tetramethylethylenediamine buffer (adjusted to pH 6.0 with orthophosphoric acid) and 0.024 mol/L 1-hexan sulfonic acid in methanol (59:41 v/v). The lower limit of quantification was 0.5 ng/mL. Four seeded plasma samples with nominal values from 1.15 to 65.5 µg/L were used as quality control to validate each high-performance liquid chromatography assay: the obtained concentrations should be within ±10% of the nominal values (±15% for the lowest quality control).
Biochemical analyses. Serum creatinine and cystatin C were measured from serum samples obtained within 24 hours before topotecan pharmacokinetic study. Cystatin C was measured on a BN-ProSpec analyzer (Dade Behring, Marburg, Germany). The method was calibrated using serial dilutions of a standard made-up with urinary purified cystatin C with an assigned value of 1.51 mg/L. Intraassay and interassay coefficients of variation ranged between 2% and 3% depending on the cystatin C level and the matrix of the control sample. Serum creatinine was measured using a Konelab 60i multiparametric analyzer and corresponding kits (Thermo Electron, Helsinki, Finland) by a noncompensated kinetic Jaffe method.
Pharmacokinetic analysis. Data were analyzed using NONMEM (version V, level 1.1 running on Pentium 200 pro; ref. 11) according to a linear two-compartment pharmacokinetic model and first-order conditional estimation method. A proportional error model was used for residual and interpatient variabilities. This model has been previously selected as the best structural pharmacokinetic model for three-population pharmacokinetic analyses (8, 12, 13).
Determination of the actual individual topotecan clearance. The topotecan concentrations versus time data were composed of 59 patients. The typical value of central volume (V1) was proportional to body weight. No covariate was considered for the typical value of topotecan clearance (CL), intercompartment clearance (Q), and peripheral volume (V2). Individual pharmacokinetic variables were obtained by Bayesian estimation using the POSTHOC option. These individual clearances were considered as the actual values for evaluation of the correlations with values predicted from covariates.
Covariate analyses. Data splitting was done randomly to create a model-building data set (44 patients) and a model validation data set (15 patients). The proportion of sparsely sampled patients was 11 of 44 (building data set) and 4 of 15 (validation data set). Eight covariates were tested: serum creatinine, cystatin C, WHO performance status, age, sex, body surface area (BSA, calculated according to the Dubois formula), body weight, and ideal body weight [IBW, calculated according to the Lorentz equation: IBW = height – 100 – (height – 150) / d with d = 4 if male, d = 2 if female]. The influence of each covariate on topotecan clearance was tested according to the following equation: TVCL = 
1 (covariate / mean covariate)5, where 1 is the typical value of topotecan clearance (TVCL) for a patient with the mean covariate, and 5 is the estimated influential factor for the covariate. The corresponding values of objective function value were compared with that of the model without covariate (i.e., TVCL = 1) by the 
2 test of difference. A covariate was considered as relevant if it decreased the objective function value to at least 3.84 (P < 0.05, 1 df). A stepwise backward elimination procedure was carried out. At both steps, the interindividual variability estimate was considered with the change in objective function value in order to assess the impact of each covariate.
Prospective evaluation. Covariate equations were prospectively evaluated using the model validation data set (n = 15 patients). For a patient j, the relative prediction error, pej (%) for topotecan clearance was defined as follows: pej (%) = (CLpred – CL) x 100 / CL, where CLpred is the value predicted from covariate equations for patient j, CL is the actual topotecan clearance that was obtained by the run using the data set composed of the 59 patients without covariates. Predictive performance of the formula was evaluated by computing the mean percentage error [mpe = N–1 
Nj = 1 (pej) where n is the number of patients = 15] as a measure of bias and the mean absolute percentage error (mape = N–1 Nj = 1|pej|) as an assessment of precision.
|
Results
|
|---|
Topotecan plasma concentrations ranged between 0.75 and 60.4 ng/mL (median, 9.0 ng/mL). The structural pharmacokinetic model was found to describe the data very accurately with a residual variability comprising between 10.3% and 11.4% depending on the run. The regression line between topotecan concentrations predicted according to typical values of the pharmacokinetic variables and observed concentrations as well as that between individual predicted concentrations and observed concentrations were always close to the identity line as shown for the run without covariates on topotecan clearance based on the whole data set (n = 59; Fig. 1). Mean ± 95% confidence interval (coefficient of variation for interindividual variability) pharmacokinetic parameters corresponding to the whole data set were: CL = 19.7 ± 1.7 L/hour (33%), V1 = 0.47 ± 0.08 x body weight L (43%), Q = 59.3 ± 17.8 L/hour (40%), and V2 = 41.0 ± 5.01 L (23%). Cystatin C serum levels were significantly correlated with both serum creatinine serum levels (r = 0.71, P < 0.001) and Cockcroft-Gault creatinine clearance (r = –0.57, P < 0.001).
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
|
Fig. 1. Predicted (A) or individual predicted (B) plasma topotecan concentrations versus observed concentrations. Insets, weighed residuals associated with the population predictions (WRES, A) or individual predictions (IWRES, B) versus observed concentrations.
|
|
Covariate analyses. Four covariates significantly decreased the objective function value when tested individually: IBW, serum creatinine, age, and cystatin C. The interindividual variability corresponding to the intermediate model based on these four covariates was 22.5% (versus 33.6% without covariates). Deletion of either age or serum creatinine did not significantly change either the objective function value (+0.1 or +0.6, respectively) or the interindividual variability (22.5% or 22.6%, respectively). So, the final model was: TVCL = 1 (cystatin C / 1.06)2 (IBW / 57)3. In order to compare serum creatinine to cystatin C, an intermediate model based only on serum creatinine was tested [i.e., TVCL = 1 (serum creatinine / 85.2)2 (age / 57)3 (IBW / 57)4]. Deletion of IBW was associated with a significant increase of the objective function value (+5.9), but not deletion of age (+2.1), allowing us to consider as the final model based on serum creatinine: TVCL=1 (serumcreatinine/85.2)2 (IBW/57)3.57)3.57)3. Final and alternative covariate models and their corresponding parameters are detailed in Table 2. Although sex was not a significant covariate when tested individually, sex was added to each of the two final models based on serum creatinine or cystatin C but, again, did not significantly decrease the objective function value (–2.6 in both cases). Also, IBW was replaced by body weight within the final models: objective function value increased significantly in the two cases (+6.1 for the final model based on cystatin C, and +8.0 for that based on serum creatinine) showing that IBW was a "better" covariate than body weight. Lastly, a model based on creatinine clearance calculated according to the Cockcroft-Gault equation (CrCl) was tested. The obtained model [i.e., TVCL = 20.8 ± 1.8 (CrCl/70)0.46 ± 0.23], was associated with an objective function value of 494.7 and interindividual variability of 27.6% showing a better performance than that of serum creatinine alone, but worse than that of the final model based on both serum creatinine and IBW.
View this table:
[in this window]
[in a new window]
|
Table 2. Final and alternative covariate equations to estimate topotecan clearance obtained from the model-building data set (n = 44 patients)
|
|
Prospective evaluation and analysis of the whole data set. The final models based on cystatin C or serum creatinine obtained by analyzing the building data set were prospectively evaluated using the model validation data set (n = 15 patients). Results of bias and precision are shown in Table 3. Mean percentage error (bias), and mean absolute percentage errors (precision) were similar for the two models. Regression between actual and predicted topotecan clearance, and both 10th and 90th percentiles of percentage errors were better for the cystatin C model. Covariate analysis of the whole data set (n = 59 patients) was done. The mean coefficient and exponents (±95% confidence interval) for the cystatin C model were: CL (L/hour) = 19.7 (±1.3). (cystatin C / 1.06)–0.61 (±0.20) (IBW / 57)0.93 (±0.58); and for the serum creatinine model: CL (L/hour) = 19.8 (±1.4) (serum creatinine / 85.2)–0.51 (±0.26) (IBW / 57)1.17 (±0.57). Differences between mean exponents of covariate models obtained from the building data set or from whole data set were all <5%. For comparison with BSA, which is the covariate usually used for individual topotecan dosing, a direct proportional model between topotecan clearance and BSA was determined from the whole data set: CL (L/hour) = 19.6 (±1.7) (BSA / 1.65). The correlation between actual topotecan clearance and value calculated from these three equations obtained from the whole data set are shown in Fig. 2; corresponding percentage errors between actual and calculated values are stated in Table 4. As a means to illustrate the ability of each covariate model of initiating topotecan dosing to hit a target AUC (AUCtarget), the difference between observed AUC (AUCobs, simply equal to the dose = AUCtarget x CLpred / actual patient clearance) and AUCtarget was calculated. Distribution of the percentages of difference for the three covariate models (i.e., cystatin C, serum creatinine, and BSA models) is shown in Fig. 3.
View this table:
[in this window]
[in a new window]
|
Table 3. Prospective evaluation of the final models based on serum cystatin C or creatinine to predict topotecan clearance in 15 patients
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Comparison between actual topotecan clearance and values calculated according to the serum cystatin C, serum creatinine, and BSA models obtained from the whole data (n = 59 patients)
|
|
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
|
Fig. 3. Difference between the observed AUC corresponding to the dose individualized according to the serum cystatin C model (A), the serum creatinine model (B), or the BSA model (C), and the target AUC [(AUCobs – AUCtarget) x 100 / AUCtarget] (n = 59 patients).
|
|
|
Discussion
|
|---|
Nowadays, the topotecan dosing is done according to BSA. The recommended dose is 7.5 mg/m2 per cycle. This dosing method is associated with a large interindividual pharmacodynamic variability. Numerous patients present severe neutropenia as a consequence of a large area under the curve of plasma topotecan concentrations (14). These observations are not surprising regarding the poor correlation between topotecan clearance and BSA (12). In the summary of product characteristics, guidelines for adaptation of topotecan dosage recommend reducing the administered dose by 50% for patients with creatinine clearance ranging between 20 and 40 mL/minute, and to exclude patients with lower values (15). Only 6 out of the 59 patients of the present study had a creatinine clearance <40 mL/minute (none <20 mL/minute). Their mean topotecan clearance was 14.0 L/hour (range, 8.4-18.5 L/hour). Their individual values are shown in Fig. 2C. Most of the patients with low actual topotecan clearance had a creatinine clearance <40 mL/minute, and would receive the regular dose (7.5 mg/m2) according to these guidelines. Conversely, patients with creatinine clearance <40 mL/minute would receive a half-dose, whereas the actual topotecan clearance of some patients was not lower than the mean observed topotecan clearance (i.e., 19.7 L/hour). Thus, the dosing method based on BSA with a cutoff value of creatinine clearance for half-dosing is not satisfactory, and we previously proposed to use creatinine clearance as a continuous covariate to calculate the individual dose of topotecan (13). As previously shown (3–5), cystatin C serum levels were closely correlated with both serum creatinine and creatinine clearance in our population. Nevertheless, the results of the present pharmacokinetic analysis show that cystatin C is a better marker of topotecan clearance than both serum creatinine and creatinine clearance levels. The criteria of both stepwise backward covariate elimination procedure applied to the model-building data set and prospective evaluation showed that the model based on cystatin C was associated with a lower interindividual variability than that based on serum creatinine (Tables 2 and 3). The advantage of the cystatin C model over the serum creatinine model was modest in terms of extreme values for percentage errors between predicted and actual topotecan clearance, but was substantial if 10th and 90th percentiles were considered. These results were confirmed when models were obtained from analysis of the whole data (Table 4; Fig. 2). As shown in Fig. 3, the predictive ability of the cystatin C model to hit a target AUC was better than those of the serum creatinine and BSA models, justifying the efforts to obtain the patient cystatin C level to individualize the topotecan dose. As for serum creatinine, our results showed that cystatin C should be considered together with other covariate(s); the final model based on both cystatin C and IBW was significantly better than that based only on cystatin C. The improvement of cystatin C-based topotecan clearance prediction by also considering IBW may explain why O'Riordan et al. (who did the only pharmacokinetic study published thus far, challenging cystatin C as a marker of drug elimination), concluded that cystatin C (which was considered alone) was not a better predictor of digoxin clearance than serum creatinine was (16). The population approach we used presents the benefit over the traditional two-step approach used by O'Riordan et al., to consider several covariates together during the pharmacokinetic analysis. Topotecan is certainly not the ideal drug to extrapolate general comments regarding cystatin C and glomerular filtration rate because its renal elimination accounts only for about 50% of total drug disposition. However, these results confirmed those we recently obtained for carboplatin,1 a drug which is mainly eliminated by glomerular filtration: cystatin C is a marker of drug elimination which is at least as good as serum creatinine in predicting clearance of drug eliminated mainly or partially by the kidney. Another potential advantage of cystatin C over serum creatinine lies in the analytic field. Interassay variability of serum creatinine measurement is among the highest in clinical biochemistry (up to 25% for the low concentrations; ref. 17). It has been shown that this imprecision may impair the transferability of equations predicting the glomerular filtration rate (18, 19). For cystatin C, technical variations seem lower than those for serum creatinine, at least for the two fully automated assays available for routine use: when the same calibrators are used, the correlation line between these two assays is close to the identity line (20). Therefore, despite its higher cost compared with serum creatinine, cystatin C deserves to be further explored as a promising covariate for drug dosing as well as selection criteria for clinical multicenter studies.
|
Acknowledgments
|
|---|
We thank Dade Behring for providing the analyzer and kits for the measurement of cystatin C.
|
Footnotes
|
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: None of the authors have any conflict of interest to declare with respect to the contents of this article.
1 Thomas F, Séronie-Vivien S, Gladieff L, et al. Cystatin C as a new covariate to predict the renal elimination of drugs: application to carboplatin, submitted for publication. 
Received 10/12/04;
revised 1/18/05;
accepted 1/26/05.
|
References
|
|---|
- Cockcroft D, Gault M. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31–41.[Medline]
- Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461–70.[Abstract/Free Full Text]
- Newman DJ. Cystatin C. Ann Clin Biochem 2002;39:89–104.[CrossRef][Medline]
- Hoek FJ, Kemperman FAW, Krediet RT. A comparison between cystatin C, plasma creatinine and the Cockcroft and Gault formula for the estimation of glomerular filtration rate. Nephrol Dial Transplant 2003;18:2024–31.[Abstract/Free Full Text]
- Laterza OF, Price CP, Scott MG. Cystatin C: an improved estimator of glomerular filtration rate? Clin Chem 2002;48:699–707.[Abstract/Free Full Text]
- Dharnidharka VR, Kwon C, Stevens G. Serum cystatin C is superior to serum creatinine as a marker of kidney function: a meta-analysis. Am J Kidney Dis 2002;40:221–6.[CrossRef][Medline]
- Herben VM, Bokkel Huinink WW, Beijnen JH. Clinical pharmacokinetics of topotecan. Clin Pharmacokinet 1996;31:85–102.[Medline]
- Montazeri A, Boucaud M, Lokiec F, et al. Population pharmacokinetics of topotecan: intraindividual variability in total drug. Cancer Chemother Pharmacol 2000;46:375–81.[CrossRef][Medline]
- Gupta E, Ratain MJ. Camptothecin analogues: topotecan and irinotecan. In: Grochow LB, Ames MM, editors. A clinician's guide to chemotherapy. Pharmamacokinetics and pharmacodynamics. Baltimore: Williams & Wilkins; 1998. p. 435–57.
- Montazeri A, Culine S, Laguerre B, et al. Individual adaptive dosing of topotecan in ovarian cancer. Clin Cancer Res 2002;8:394–9.[Abstract/Free Full Text]
- Beal SL, Sheiner LB. Estimating population kinetics. Crit Rev Biomed Eng 1982;8:195–222.[Medline]
- Gallo JM, Laub PB, Rowinsky EK, Grochow LB, Baker SD. Population pharmacokinetic model for topotecan derived from phase I clinical trials. J Clin Oncol 2000;18:2459–67.[Abstract/Free Full Text]
- Léger F, Loos WJ, Fourcade J, et al. Factors affecting pharmacokinetic variability of oral topotecan: a population analysis. Br J Cancer 2004;90:343–7.[CrossRef][Medline]
- Grochow LB, Rowinsky EK, Johnson R, et al. Pharmacokinetics and pharmacodynamics of topotecan in patients with advanced cancer. Drug Metab Dispos 1992;20:706–13.[Abstract]
- Hycamtin. Summary of product characteristics. Smith-Kline Beecham plc. Brentford (UK): 2003.
- O'Riordan S, Ouldred E, Brice S, Jackson SH, Swift CG. Serum cystatin C is not a better marker of creatinine or digoxin clearance than serum creatinine. Br J Clin Pharmacol 2002;53:398–402.[CrossRef][Medline]
- Seronie-Vivien S, Galteau MM, Carlier MC, et al. Improving the interlaboratory variation for creatinine serum assay. Ann Biol Clin (Paris) 2004;62:165–75.[Medline]
- Coresh J, Eknoyan G, Levey AS. Estimating the prevalence of low glomerular filtration rate requires attention to the creatinine assay calibration. J Am Soc Nephrol 2002;13:2811–6.[Free Full Text]
- Wuyts B, Bernard D, Van den Noortgate N, et al. Reevaluation of formulas for predicting creatinine clearance in adults and children, using compensated creatinine methods. Clin Chem 2003;49:1011–4.[Free Full Text]
- Finney H, Newman DJ, Gruber W, et al. Initial evaluation of cystatin C measurement by particle-enhanced immunonephelometry on the Behring nephelometer systems (BNA, BN II). Clin Chem 1997;43:1016–22.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
A. Garcia, J. Hermida, and J. C. Tutor
Estimation of the Glomerular Filtration Rate From Serum Creatinine and Cystatin C With Regard to Therapeutic Digoxin Monitoring
J. Clin. Pharmacol.,
November 1, 2007;
47(11):
1450 - 1455.
[Full Text]
[PDF]
|
|
|
|
|
|
|
V Launay-Vacher, E Chatelut, S. Lichtman, H Wildiers, C Steer, and M Aapro
Renal insufficiency in elderly cancer patients: International Society of Geriatric Oncology clinical practice recommendations
Ann. Onc.,
August 1, 2007;
18(8):
1314 - 1321.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Patients and Methods
