Clinical Cancer Research Vol. 5, 2349-2358, September 1999
© 1999 American Association for Cancer Research
Pharmacokinetics and Pharmacodynamics of Lobaplatin (D-19466) in Patients with Advanced Solid Tumors, Including Patients with Impaired Renal or Liver Function
Jan Welink,
Epie Boven,
Jan B. Vermorken,
Helen E. Gall and
Wim J. F. van der Vijgh1
Department of Medical Oncology, University Hospital Vrije Universiteit, De Boelelaan 1117, 1081 HV Amsterdam, the Netherlands
 |
ABSTRACT
|
|---|
The purpose of this study was to determine the influence of impaired renal and liver function on the pharmacokinetics and pharmacodynamics of lobaplatin in cancer patients. A total of 25 patients with advanced solid tumors not amenable for standard treatment entered the study. Patients had normal organ function or an impaired liver or renal function (two levels). The starting dose of lobaplatin was 50 mg/m2 i.v. given every 3 weeks. The blood and urine of all patients were sampled for the determination of (ultrafilterable) platinum, intact lobaplatin, creatinine, and blood cell counts. No objective responses were recorded. Five patients experienced no change and received 4–10 cycles (median, 6 cycles) of lobaplatin. The extent and duration of hematological toxicity were worse in patients with impaired renal function. Thrombocytopenia was most prominent; grade 4 toxicity was observed in 15 patients in the first two cycles of treatment. The concentration-time curves of ultrafilterable platinum and intact lobaplatin revealed almost identical patterns. The elimination of ultrafilterable platinum [final half-life (t1/2 final) = 131 ± 15 min; clearance (Cl) = 125 ± 14 ml/min/1.73 m2] was much faster than that of total platinum (t1/2 final = 6.8 ± 4.3 days, Cl = 34 ± 11 ml/min/1.73 m2). No pharmacokinetic differences were observed between patients with normal organ function and those with an impaired liver function within the investigated range. An impaired renal function resulted in an increase of the t1/2 final due to a decrease of the total body Cl that resulted in a higher exposure of the body to the drug. The calculated creatinine Cl was linearly correlated with the total body clearance of ultrafilterable platinum (r = 0.91), which resulted in the dosage formula D = AUC
(1.1 ClCrU + 16), in which D represents dose, AUC represents area concentration-time curve, and ClCrU represents creatinine Cl. The thrombocyte surviving fraction correlated well with the AUC value of ultrafilterable platinum (r = 0.72). It can be concluded that the hematological toxicity and the pharmacokinetics of lobaplatin are strongly affected by renal function. The total body Cl of ultrafilterable platinum correlated well with the creatinine Cl and the thrombocyte surviving fraction. In patients with renal function, represented by a creatinine clearance 
30 ml/min/1.73 m2, the derived dosage formula will enable us to calculate the dose that is expected to lead to an acceptable extent of thrombocytopenia in a patient with a given renal function. Prospective studies with larger groups of patients are needed to prove the value of this dosage formula.
|
INTRODUCTION
|
|---|
Lobaplatin (D-19466; 1,2-diammino-methyl-cyclobutaneplatinum(II)-lactate) is a new anticancer agent and a representative of the third-generation platinum compounds. Lobaplatin consists of a nearly 50%/50% mixture of two diastereoisomers: (a) the SSS configuration (LP-D1); and (b) the RRS configuration (LP-D2) (Fig. 1)
. The compound has shown antitumor activity in human lung, gastric, testicular, and ovarian cancer xenografts, with incomplete cross-resistance to cisplatin in vitro and in vivo (1, 2, 3, 4)
. Compared with cisplatin or carboplatin, lobaplatin significantly prolonged the survival of mice bearing P388 leukemia and a cisplatin-resistant P388 variant (1)
.
Phase I and II studies with lobaplatin have demonstrated responses in relapsed ovarian cancer (5, 6, 7)
, esophageal cancer (8)
, head and neck cancer, breast cancer, and small cell lung cancer (9)
. The dose-limiting toxicity of lobaplatin is thrombocytopenia, with a nadir at approximately 2 weeks after drug administration. Leukopenia is less severe than thrombocytopenia, and the drug does not induce nephrotoxicity, neurotoxicity, or ototoxicity.
Because of the potential clinical use of lobaplatin and the limited and conflicting pharmacokinetic results (low urinary platinum excretion and short t1/2 final;2
Refs. 6
and 10
), it was decided to reestablish the pharmacokinetics of lobaplatin after an i.v. bolus injection not only in patients with normal renal and liver function but also in patients with impaired renal or liver function. The pharmacokinetics and pharmacodynamics of lobaplatin were also determined during the second and third cycle to check for any possible cumulative effects due to repetitive dosing. In addition, pharmacokinetic-pharmacodynamic relationships were established.
|
PATIENTS AND METHODS
|
|---|
Patients and Treatment Schedule.
A total of 25 patients entered this study after they had given informed consent. Patients had recurrent or metastatic disease from solid tumor types with progression after standard treatment. WHO performance status was 
2. All patients had adequate bone marrow function, i.e. leukocyte count 4 x 109/liter, neutrophil count 2 x 109/liter, and platelet count 150 x 109/liter. The patients were stratified according to their renal and liver function to the following four groups: (a) normal renal and liver function [serum creatinine < 110 µmol/liter (1.24 mg/dl) and/or creatinine Cl > 79 ml/min/1.73 m2 and bilirubin < 20 µmol/liter (1.18 mg/dl), and no elevated level of ALAT (44 units/liter), ASAT (31 units/liter), or 
GT (50 units/liter) for males and 38 units/liter for females)]; (b) impaired renal function level I (serum creatinine = 110–165 µmol/liter (1.24–1.87 mg/dl) and/or creatinine Cl = 55–79 ml/min/1.73 m2); (c) impaired renal function level II (serum creatinine = 166–330 µmol/liter (1.88–3.73 mg/dl) and/or creatinine Cl = 30–54 ml/min/1.73 m2); and (d) liver impairment due to metastases [bilirubin = 20–50 µmol/liter (1.18–2.94 mg/dl) and/or at least one elevated enzyme level (ALAT, ASAT, or GT) up to a maximum of five times the upper limit of normal]. Patients with renal impairment were divided into two groups to obtain enough patients within a certain level of renal impairment. Each group consisted of six patients. One patient was not included in the evaluation because he had both impaired renal function and impaired liver function. The characteristics of all 25 patients are listed in Tables 1
and 2
. The study was performed according to the rules of the declaration of Helsinki and was approved by the medical ethical committee of the Hospital Vrije Universiteit.
Lobaplatin for injection was supplied by ASTA Medica AG (Frankfurt, Germany) as a sterile lyophilized powder (50 mg/vial). The drug was reconstituted with 5 ml of 0.9% NaCl. A starting dose of 50 mg/m2 lobaplatin was given as an i.v. bolus injection in 1 min every 3 weeks. The dose was reduced (by 10 mg/m2) in case of severe hematological (grade 4 for >7 days) or nonhematological toxicity (grade 3 or more, except for nausea/vomiting and alopecia) or when a treatment delay of more than 2 weeks was necessary. The patients were not hydrated before or after drug administration, and there were no restrictions on food or fluid intake.
Pretreatment and Follow-Up Studies.
Before initiation of therapy, all patients had a history and physical examination, an assessment of WHO performance status, a chest radiograph, appropriate radiological tests to determine measurable or evaluable disease, an audiogram, and routine laboratory studies that included a complete blood count with a differential leukocyte count, electrolytes, urea, creatinine, bilirubin, alkaline phosphatase, ASAT, ALAT, -GT, lactate dehydrogenase, albumin, total protein, glucose, coagulation tests, dipstick urinalysis, and creatinine clearance. An assessment of safety consisting of clinical assessment, adverse events, and a complete blood count with differential and clinical chemistry was performed weekly. Creatinine Cl and dipstick urinalysis were assessed on day 1 of the subsequent cycles, whereas the audiogram was repeated at the end of every third cycle.
Grading of toxicity was scored according to the WHO criteria. A new treatment cycle could not be started until the patient had recovered from adverse events associated with a prior treatment cycle. Recovery had occurred when the leukocyte count was 3.0 x 109/liter, the neutrophil count was 1.5 x 109/liter, and the platelet count was 150 x 109/liter. Assessment of efficacy, where applicable, was performed every second or third cycle. Response to treatment was classified according to the WHO criteria. A complete response was defined as complete disappearance of all measurable and evaluable disease over a period of at least 4 weeks. A partial remission was defined as at least a 50% decrease of the sum of the products of the largest perpendicular lesion diameters without the appearance of new lesions over a period of at least 4 weeks. No change was defined as a change from baseline within the limits of a 50% regression or a 25% increase of lesion area. Disease progression was defined as a 25% or greater increase in the size of one or more measurable or evaluable lesions.
Sampling.
During the first course, blood samples were obtained before and at 0 (just after bolus injection), 5, 15, and 30 min and 1, 2, 4, 6, 9, 11, 22, 24, 48, 72, 96, and 120 h after bolus injection. For the second and third cycle, a reduced number of blood samples were taken up to 6 h after drug administration. Samples were collected in NH4-heparin-coated tubes and centrifuged immediately for 10 min at 1,500 x g and 15°C. After removing the plasma, the erythrocyte fraction was washed with sodium phosphate buffer (pH 7.4). Four ml of the plasma were immediately transferred into a centriflo ultrafiltration membrane cone (CF 25; size exclusion limit of the membrane, Mr > 25,000; Amicon, Beverly, MA) and centrifuged for 45 min at 1,000 x g and 15°C. For the analysis of the diastereoisomers of lobaplatin, 1.5 ml of plasma ultrafiltrate was stored at -80°C in polypropylene test tubes. For the analysis of platinum in plasma, plasma ultrafiltrate and erythrocytes samples were stored at -20°C in polypropylene test tubes.
Urine was collected before and up to 48 h after the first bolus injection and during the second and third cycle (before and up to 24 h after the bolus injection). After determination of the volume, aliquots were stored at -20°C.
Analysis.
The amount of platinum in the plasma (total platinum), plasma ultrafiltrate (ultrafilterable platinum), urine, and erythrocytes was determined by flameless atomic absorption spectrophotometry. Samples of plasma, plasma ultrafiltrate, and urine were diluted 2.5 times with a solution of 0.9% sodium chloride and 0.2 M hydrochloric acid. Samples of erythrocytes (200 mg) were digested with hyamine hydroxide (500 µl) by incubation in a heating bath (55°C) for approximately 15 h. Thereafter, the mixture was diluted with 4.25 ml of hydrochloric acid solution (0.2 M). Calibration standards and quality control samples were prepared in plasma, plasma ultrafiltrate, and urine by spiking the matrix with a solution of lobaplatin in 0.9% sodium chloride and 0.2 M hydrochloric acid. The injection volume was 10 µl. For erythrocytes, the matrix was spiked with a solution of lobaplatin in water. The injection volume was 2 times 20 µl. The furnace program of the flameless atomic absorption spectrophotometry method consisted of a drying step between 100°C and 150°C, followed by ashing at 1300°C. Platinum was evaporated at 2600°C and measured with a platinum cathode lamp at a wavelength of 266 nm. The method was validated by determining the between-day accuracy and precision of the quality control samples analyzed at three concentration levels in duplicate on six different days. The within-day accuracy and precision were determined by analyzing the quality control samples in 6-fold on one day. The lower limit of quantification (inaccuracy and precision < 20%) was 0.19 µM for platinum in plasma, plasma ultrafiltrate, and urine and 0.54 µM for platinum in erythrocytes. The accuracy and precision of the analysis at the highest concentration of the calibration lines (5.42 µM) were 99.6 and 5.3%, 99.1 and 2.0%, 100.8 and 5.4%, and 102.2 and 4.1% for platinum in plasma, plasma ultrafiltrate, urine, and erythrocytes, respectively. The inaccuracy and precision of the quality control samples were less than 10%. To determine low platinum concentrations (<0.19 µM) in plasma and plasma ultrafiltrate, a larger amount of sample was injected into the graphite furnace (2 times 20 µl), and the furnace program was adjusted by increasing the time for drying and ashing of the sample in the furnace. The lower limit of quantification was 0.05 µM, and the accuracy and precision of the analysis at the highest concentration of the calibration lines (0.60 µM) was 99.5 and 3.9% and 99.5 and 1.0% for platinum in plasma and plasma ultrafiltrate, respectively.
The two diastereoisomers of lobaplatin (LP-D1 and LP-D2) in plasma ultrafiltrate were analyzed according to the high-performance liquid chromatography method described by our group (11)
. In short, lobaplatin was isolated from plasma ultrafiltrate by a solid-phase extraction procedure (C18 cartridge), separated into the individual diastereoisomers by a reverse-phase high-performance liquid chromatography column (Hypersil ODS), and detected by UV absorption at 210 nm. Calibration standards and quality control samples were freshly prepared by spiking human plasma ultrafiltrate with an aqueous solution of lobaplatin and a subsequent dilution of this sample with plasma ultrafiltrate. The lower limit of quantification was 0.071 µM for LP-D1 and 0.067 µM for LP-D2. The accuracy and precision of the analysis at the highest concentration of the calibration lines (9.100 µM for LP-D1 and 8.639 µM for LP-D2) were 104.4 and 2.0% and 103.3 and 1.9% for LP-D1 and LP-D2, respectively.
Pharmacokinetics.
Pharmacokinetic parameters were calculated by noncompartmental analysis using the pharmacokinetic software TopFit, Version 2.0 (12)
. Pharmacokinetic parameters were calculated according to standard procedures. The AUC in the noncompartment model was calculated by the linear trapezoidal rule with an extrapolation of the final log-linear part of the C-t curve to infinity: AUC = AUC0 
t + Clast/t1/2 final, in which Clast is the last detectable concentration, and t1/2 final is the final half-life. Subsequent calculations were as follows: total body Cl = D/AUC and MRT = AUMC/AUC, in which AUMC represents the area under the (first) moment curve, calculated by the formula AUMC = AUMC0 t + t x Clast/t1/2 final + Clast/(t1/2 final)2. The steady-state apparent volume of distribution is as follows: Vss = Cl x MRT. Student’s t test (two-tailed distribution and two-sample equal variance) was used for the statistical evaluation of the differences between the mean values calculated for the patient groups.
To compare the pharmacokinetics of lobaplatin with earlier data of other platinum compounds and to calculate initial half-lives, the C-t curves were also fitted with the NONLIN program TOPFIT, Version 2.0 (12)
using the biexponential equation Ct = A x e-
t + B x e-ßt and the triexponential equation Ct = A x e-t + B x e-ßt + C x e-t, respectively. The statistics used to determine the best fit included the sum of the squares of the weighted or unweighted residuals, using the F-ratio test, Akaike information criterion, Swarz criterion, and imbimbo criterion, as well as the consideration of a random distribution of the signs of the residuals (12)
.
The platinum excretion in the urine was expressed as the total amount of platinum excreted in the urine as a percentage of the dose (Ae).
To investigate the role of erythrocytes as a possible deep pharmacokinetic compartment and its consequence for the incidence of anemia, the uptake of platinum in RBCs (as a percentage of the dose) was also calculated (blood volume x hematocrit x platinum concentration in RBCs x 100%/total dose administered, in which the blood volume was estimated to be 7.7% of the body weight of the patient).
Pharmacodynamics.
The dose-limiting toxicity of lobaplatin is thrombocytopenia. Platelet counts were used to calculate the thrombocyte SF, which is defined by the relative thrombocyte nadir divided by the thrombocyte count on day 0 (just before bolus injection). Because most biological effects such as the SF of blood cells will vary exponentially with a parametric change, the following equation was used (13)
: log SF = k x AUC. From this equation, a linear relationship may be expected between logSF and the AUC.
Before treatment with lobaplatin, the creatinine Cl was calculated from the creatinine concentration in the 24 h urine sample and the serum creatinine concentration [Cl = urine creatinine (mmol/day) x 1000/1440 x serum creatinine (µmol/liter)] or from the serum creatinine concentration alone by the Cockcroft formula [Ref. 14
; Cl = K x (140 - age) x weight (kg)/serum creatinine (µmol), in which K = 1.05 for females and 1.23 for males].
|
RESULTS
|
|---|
Response.
Nineteen patients received at least two cycles of lobaplatin, and 13 of these patients showed progressive disease. The five patients for whom no change was recorded received 4–10 cycles (median, six cycles). Of interest, one patient with a local recurrence of squamous cell carcinoma in the oropharyngeal area had a clinical partial remission after two cycles that was complete after five cycles. Unfortunately, this patient died of pneumonia and bleeding from erosive gastritis after the sixth cycle of lobaplatin. An autopsy revealed the presence of a second primary tumor, colon cancer, with peritoneal spread and liver metastases, without evidence of local recurrence of oropharyngeal cancer.
Toxicity.
For the first 11 patients that entered the protocol, antiemetic treatment was not routinely given before the administration of lobaplatin. Because patients were hospitalized for their participation in the pharmacokinetic study, ondansetrone (8 mg, i.v.) could be administered immediately upon vomiting. All patients needed antiemetic treatment. Prophylaxis was given routinely in further cycles as well as in subsequent patients from cycle 1 onward. Patients usually recovered from nausea and vomiting within 24 h. In those cases where prophylaxis was not complete, ondansetrone (8 mg) or metoclopramide (10 mg) by oral route was prescribed. Other nonhematological toxicities were mild and consisted of grade 1–2 stomatitis (four patients), grade 1 alopecia (one patient), grade 1–2 paresthesias (three patients), and grade 1–2 anorexial (five patients). Eight patients experienced grade 1–2, fatigue and four patients experienced grade 3, fatigue which usually occurred in the presence of anemia.
Hematological toxicity was considerable, and thrombocytopenia was the most prominent toxicity (Table 3)
. Nadirs of blood counts were observed between days 14 and 16 after lobaplatin administration. The majority of patients experienced grade 4 thrombocytopenia, which was why we monitored blood counts around the expected nadir value every second to third day until recovery. Recovery was rapid except in patients with impaired renal function, in whom grade 4 toxicity could be present for as long as 1 week. Prophylactic platelet transfusions were given (when the platelet count decreased to <10 x 109/liter) in one of six patients with level I and five of six patients with level II impaired renal function in the first two cycles of lobaplatin treatment. In 16 patients, Hb fell below 6.0 mmol/liter (97 g/dl), which was the reason for erythrocyte transfusions. Only two patients presented with overt bleeding, one because of vaginal blood loss in the presence of pelvic disease of colon cancer (thrombocytopenia grade 1), and the other because of rectal blood loss in the presence of a local recurrence of ovarian cancer (thrombocytopenia grade 4). One patient with grade 4 leukopenia in the first cycle had a fever of unknown origin and was treated with antibiotics. In the second lobaplatin cycle, fever recurred in this patient, and Klebsiella pneumoniae was demonstrated to be the causative agent.
As can be deduced from Table 3
, in 11 of 19 patients, lobaplatin cycle 2 had to be delayed because recovery from myelosuppressive side effects was not complete. The protocol required a dose reduction of 10 mg/m2 in case of grade 4 hematological toxicity of >7 days in duration. Because of the high incidence of grade 4 thrombocytopenia for <7 days, the requirement of erythrocyte transfusions, and the delay of further treatment, it was decided from the ninth patient onward to reduce the dose of lobaplatin by 10 mg/m2 in case of grade 4 thrombocytopenia of any duration as well as in all first cycles of patients with impaired renal function (Table 2)
. Some patients required additional dose reductions upon later cycles.
Pharmacokinetics.
The semilogarithmic plots of the mean C-t curves of total platinum, ultrafilterable platinum, the diastereoisomers LP-D1 and LP-D2, and intact lobaplatin for the group of patients with normal renal and liver function are shown in Fig. 2
. Total platinum (protein-bound platinum + ultrafilterable platinum) could still be measured after 5 days, whereas ultrafilterable platinum and intact lobaplatin could only be measured for up to 11 h after the administration of lobaplatin. Due to protein binding, ultrafilterable platinum concentrations became lower than total platinum concentrations over time. After 11 h, about 60% of the circulating platinum was bound to plasma proteins. The C-t curves of the diastereoisomers LP-D1 and LP-D2 were superimposable, as were those of ultrafilterable platinum and intact lobaplatin. Similar results were obtained for the other patient groups.
View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
|
Fig. 2. Semilogarithmic plot of the mean C-t curves of platinum in plasma and plasma ultrafiltrate and intact lobaplatin (LP-D1 and LP-D2) in plasma ultrafiltrate in the patient group with normal liver and renal function (n = 6); A, up to 30 h; B, up to 120 h.
|
|
The influence of renal and liver function on pharmacokinetics is shown in Fig. 3
by the C-t curves of total platinum and ultrafilterable platinum for the four patient groups. Impaired liver function within the investigated range did not affect the C-t curves of total platinum or ultrafilterable platinum. In contrast, the concentration of these platinum species became higher when the renal function became lower. Table 4
gives a summary of the pharmacokinetic parameters of platinum in plasma and plasma ultrafiltrate and the pharmacokinetic data of the diastereoisomers LP-D1 and LP-D2 and intact lobaplatin (LP-D1 + LP-D2) in plasma ultrafiltrate. All parameters have been determined by noncompartmental analysis of the C-t data. For the patient group with impaired renal function level II, only four patients could be evaluated for intact lobaplatin because of interfering endogenous substances in the chromatograms. Table 5
shows the cumulative urinary platinum excretion up to 48 h after the bolus injection of lobaplatin.
View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
|
Fig. 3. Semilogarithmic plot of the mean C-t curves of total platinum (A) and ultrafilterable platinum (B) in patients with normal liver and renal function (n = 6), with impaired liver function (n = 6), and with impaired renal function level I (n = 6) and level II (n = 6).
|
|
View this table:
[in this window]
[in a new window]
|
Table 4 Mean pharmacokinetic data of lobaplatin after i.v. bolus injection calculated by noncompartmental analysis
|
|
View this table:
[in this window]
[in a new window]
|
Table 5 Cumulative amount of platinum excreted in the urine (Ae) after i.v. bolus injection of lobaplatin, expressed as a percentage of the dose (% D)
|
|
The t1/2 final of total platinum in plasma was 6.8 ± 4.3 days, measured from 1–5 days after administration. No difference was observed in the t1/2 final of total platinum between the different patient groups. Also, no difference was detected between the values of the pharmacokinetic parameters of the patient group with normal liver and renal function and the group with an impaired liver function. An increase in the AUC/D was measured in the patient groups with impaired renal function level I (P = 0.03), which was more pronounced for the patients with level II renal impairment (P = 0.02). This increase corresponded with a decrease in the total body Cl (P = 0.02), Vss (P = 0.07), and Ae(0–24 h) (P = 0.3) and was also more pronounced for the patients with level II renal impairment (P = 0.004, P = 0.01, and P = 0.001, respectively). No statistically significant differences were observed between the pharmacokinetic data of patients with impaired renal function level I and level II (P = 0.1).
For ultrafilterable platinum, the t1/2 final was much shorter (131 ± 15 min, measured from 2–11 h) than that of total platinum. No differences were observed between the pharmacokinetics of the patient group with normal liver and renal function and the group with impaired liver function. An impaired renal function level I, however, resulted in an increase of the t1/2 final (P 0.001) that corresponded with an increase in the AUC/D (P 0.001) and MRT (P 0.001) and a decrease in Cl (P 0.001). This was more pronounced for the patients with level II renal impairment. Between patients with renal impairment level I and level II, a statistically significant difference was observed for the t1/2 final (P = 0.004), AUC/D (P = 0.007), MRT (P = 0.005), and Cl (P = 0.01).
No differences were seen between the values of the diastereoisomers LP-D1 and LP-D2 and between the values for intact lobaplatin and ultrafilterable platinum, as shown in Fig. 2
. Changes in these values due to differences in renal function between the groups were comparable to those observed for ultrafilterable platinum.
Most of the platinum was excreted in the urine within the first 6 h after drug administration (Table 5)
. For patients with impaired renal function, the decrease in platinum excretion is explained by the difference in the first 6 h.
To check for changes in the pharmacokinetics between subsequent cycles, values of AUC(0–6 h) normalized to a dose of 50 mg/m2 were calculated. No differences were observed for the pharmacokinetics between the subsequent cycles. AUC(0–6 h) decreased linearly when the dose was reduced in a second or third cycle on the basis of grade 4 thrombocytopenia in the preceding cycle.
The best NONLIN fits of the plasma C-t curves of total platinum and ultrafilterable platinum for the patient group with normal renal and liver function were obtained with a triexponential function and a biexponential function, respectively. The resulting values of the pharmacokinetic parameters were similar to the values obtained with the noncompartmental analysis. The mean initial half-lives obtained were 9.2 ± 6.8 and 119 ± 36 min for t1/2 and t1/2ß of total platinum, respectively, and 14.6 ± 3.1 min for t1/2 of ultrafilterable platinum.
Platinum was taken up in erythrocytes. In patients with normal renal and liver function, maximum levels (Cmax) of 2.1 ± 0.4 µM were reached at about 3 h (Tmax, 3.2 ± 0.7 h) after bolus injection. At that time, ultrafilterable platinum levels in plasma were still about 1.5 times higher than those in erythrocytes. At the time of the maximum concentration, about 1.7% of the dose was present in erythrocytes. The t1/2 final of platinum present in erythrocytes was 11 ± 7 days as measured over days 1–5. Comparable values were obtained for the patients with impaired liver function. In patients with impaired renal function, Cmax increased nonsignificantly to 2.5 ± 0.5 µM (representing 2.2% of the dose at that moment) for level I and to 2.8 ± 0.8 µM (representing 2.1% of the dose at that moment) for level II. Tmax increased significantly [level I, 4.8 ± 1.3 h (P = 0.03); level II, 4.6 ± 1.1 h (P = 0.03)]. The t1/2 final of platinum in erythrocytes did not change in relation to renal function.
Nadirs of thrombocytes were reached between 14 and 16 days after the administration of lobaplatin. Therefore, the SF of the thrombocytes was calculated in each patient on day 14 or close to this time. A linear relationship was found between log SF and the AUC of ultrafilterable platinum. The correlation coefficient was 0.72 (Fig. 4
, cycle 1). In all patients, preceding cycles of lobaplatin resulted in cumulative thrombocytopenia (lowering of the lines in Fig. 4
, cycles 2 and 3).
The values in Table 4
show that the total body Cl of ultrafilterable platinum (ClUFPt) was dependent on the renal function. When ClUFPt was plotted against the creatinine Cl, as calculated from the creatinine concentration in the 24 h urine samples (ClCrU), a good linear relationship (correlation coefficient, 0.91) was obtained [ClUFPt = 1.1 ClCrU + 16 (Fig. 5)
]. Substituting ClUFPt with D/AUC resulting in the following formula for the dose (Dg mg): D = AUC (1.1 ClCrU + 16). A comparable result was obtained when the creatinine Cl was calculated by the Cockcroft formula [correlation coefficient, 0.85; dosage formula, D = AUC (1.5 ClCrC + 3)]. These formulas can be used to calculate a tolerable dose for each individual patient with a creatinine Cl of 30 ml/min/1.73 m2 by selecting the target AUC that belongs to a desired degree of thrombocytopenia.
|
DISCUSSION
|
|---|
Antineoplastic drugs generally have a narrow therapeutic window. However, the plasma levels of a certain drug at which toxicity or therapeutic effects occur may vary from patient to patient. Therefore, cancer chemotherapy may be optimized by individualizing a patient’s therapy using individual calculations. To achieve this goal, it is necessary to establish pharmacokinetic-pharmacodynamic relationships for each anticancer drug. This will allow the calculation of the optimal dose for a given patient to obtain a defined acceptable level of toxicity and the highest individual therapeutic effect. It is much more complicated to establish a relationship between pharmacokinetics and treatment response than it is to establish a relationship between pharmacokinetics and side effects. Often the degree of toxicity is related to the AUC or Css of a drug (15)
. The first onset for a pharmacokinetic-toxicity relationship can be obtained during a Phase I study, after which an extension or a confirmation of the relationship can be obtained during the following Phase II or, if desired, phase III studies. Using such a relationship, it becomes possible to administer the highest possible dose to obtain the highest possible therapeutic effect in an individual patient. As soon as a pharmacokinetic-therapeutic effect relationship can be established, and it appears that an optimal therapeutic effect can be obtained at a lower dose level, individual therapeutic windows are obtained. In the present study, we established pharmacokinetic-pharmacodynamic relationships for lobaplatin, leading to a dosage formula that allows the individualization of the lobaplatin dose.
The pharmacokinetic results showed that the C-t curves of total platinum (protein bound platinum + ultrafilterable platinum), ultrafilterable platinum, and intact lobaplatin were similar during the first few hours after an i.v. bolus injection of lobaplatin. Thereafter, the concentrations for ultrafilterable platinum and intact lobaplatin were lower than those of total platinum, indicating that protein binding only takes place slowly. This is in agreement with our in vitro data (obtained according to Ref. 16
), which revealed that only 6% of lobaplatin was bound to human plasma proteins for after 2 h of incubation. After 8 and 24 h, this percentage increased to 16% and 33%, respectively. This also means that the centrifugation time needed to obtain plasma ultrafiltrate did not influence the concentration of ultrafilterable platinum. Ultrafilterable platinum was detectable in the circulation for 11 h in patients with normal liver and renal function and for up to 96 h in patients with impaired renal function. During this longer period of time, the composition of ultrafilterable platinum may also be changed. Total platinum concentrations measured after these periods represent lobaplatin that has reacted with plasma proteins. These protein-bound platinum species are eliminated from the body with slow elimination rates, which correspond to the turnover rates of the plasma proteins. This explains why the t1/2 final of total platinum is less sensitive for changes in renal function than the free platinum species.
The final half-lives of ultrafilterable platinum were similar for the patients with normal liver and renal function and the patients with impaired liver function. An increase was detected in patients with impaired renal function, which was more pronounced for patients with level II impairment. These differences are in agreement with the amounts of platinum excreted in the urine, i.e., no difference in Ae(0–24 h) was seen between patients with normal liver and renal function and patients with impaired liver function, but a decrease in Ae(0–24 h) was seen in patients with impaired renal function, which was more pronounced for patients with level II impairment. The high Ae values indicate that urinary platinum excretion is the major route of elimination. Therefore, impaired renal function will have a major impact on the elimination of platinum from the body, which corresponds to a lower total body Cl and higher AUC and MRT values.
No differences were observed between the pharmacokinetics of ultrafilterable platinum and intact lobaplatin (LP-D1 plus LP-D2). This means that all free platinum exists as intact lobaplatin, indicating that no metabolites were formed up to the lowest measured concentration of intact lobaplatin (0.14 µM). Also, the pharmacokinetics of the two diastereoisomers, LP-D1 and LP-D2, were similar, indicating that the difference in the molecular configuration has no influence on the pharmacokinetics.
From the pharmacokinetic data obtained during successive courses in which the dose was lowered in some patients, it appeared that the AUC(0–6 h) was linearly related with the dose within the considered dose range of 20–50 mg/m2. No differences were found between the pharmacokinetic parameters of successive courses. This corresponds with the lack of influence of lobaplatin on the renal function, which remained constant during a cycle and between cycles.
Previous pharmacokinetic data on lobaplatin (6
, 10)
correspond with our findings, except for the t1/2 final and the urinary platinum excretion. Gietema et al. (6)
have measured shorter final half-lives for total platinum and ultrafilterable platinum, although lobaplatin was also given as an i.v. bolus injection (30–50 mg/m2). The difference can be explained by the shorter sampling period of their pharmacokinetic study (<11 h). The urinary platinum excretion over the first 24 h in that study was 73 ± 4% of the administered dose, which is in accordance with our findings (70 ± 10% D). In contrast, Mross et al. (10)
have measured only 26 ± 8% of the dose excreted in the urine during the first 48 h after the administration of lobaplatin as an i.v. bolus injection (20–50 mg/m2). No explanation can be given for this difference. Also a shorter t1/2 final for total platinum (1.4 ± 1.4 days) was calculated that might be due to blood sampling for only 2 days instead of 5 days in our study.
Similarities are observed between the pharmacokinetics of lobaplatin and carboplatin (17)
. For carboplatin, the urinary excretion was 77 ± 5% D over the first 24 h after administration of the dose, the t1/2 final of total platinum was 5.8 ± 1.6 days, and the t1/2 final of ultrafilterable platinum was 120 ± 11 min. The total body Cl of ultrafilterable platinum was 107 ± 1.9 ml/min/1.73 m2. These similarities can be explained by the very limited protein binding of both compounds (18)
. The differences between the pharmacokinetics of lobaplatin and cisplatin (19
, 20)
are attributed to a higher extent of plasma and tissue protein binding. The t1/2 final of total platinum (5.4 ± 1 days) after the administration of lobaplatin was comparable with that of cisplatin, which can be explained by the turnover of the same plasma proteins to which both compounds bind.
The percentage of the dose taken up by erythrocytes at the time of the maximal platinum concentration was higher after the administration of lobaplatin (1.7% D) than after cisplatin (1.2% D) or carboplatin (0.4% D; Ref. 18
), indicating that lobaplatin has the highest affinity for erythrocytes. Because of the observed t1/2 final of lobaplatin in erythrocytes, these cells have to be considered as a deep pharmacokinetic compartment. However, the limited number of patients prevents a statement whether a relationship exists between platinum uptake and the incidence of anemia.
The observed relationship between the thrombocyte SFs and the calculated AUC values of ultrafilterable platinum is in accordance with the results obtained by Mross et al. (10)
. Such a relationship means that the level of thrombocytopenia caused by exposure to a certain AUC can be calculated. We found a linear relationship between the total body Cl of ultrafilterable platinum and the creatinine Cl. Because ultrafilterable platinum exclusively represents intact lobaplatin (Fig. 2)
, this relationship led to the lobaplatin dosage formula D = AUC (1.1 ClCrU + 16), with D expressed in mg, AUC expressed in mg · min/ml, and ClCrU expressed in ml/min. This formula, which is very similar to that of Calvert et al. (21)
derived for carboplatin, allows the calculation of the dose of lobaplatin to obtain a target AUC in a patient with a known renal function. This AUC can be targeted either to prevent myelosuppression or to obtain a desired therapeutic effect. If, for example, for a certain patient, a SF of 30% will lead to an acceptable thrombocytopenia, the AUC should not exceed 25 µM · h (see Fig. 4
, cycle 1), which for the first cycle corresponds to an AUC value of 0.60 mg · min/ml to be used in the dosage formula. With this target AUC, the maximum tolerated dose of single-agent lobaplatin can then be calculated for each individual patient with a given renal function. Also, when lobaplatin is used in combination therapies, and no pharmacokinetic interaction exists between lobaplatin and the other drugs, the dose of lobaplatin can be calculated on the basis of the target AUC by means of the dosage formula. However, the target AUC determined by the desired antitumor effect or toxicity will depend on the drug combination used. For each individual combination therapy, new pharmacokinetic-pharmacodynamic relationships have then to be established.
It can be concluded that prospective studies with larger groups of patients are needed to prove the value of the presented dosage formula and the value of the target AUC in relation to the derived pharmacokinetic-pharmacodynamic relationship. Additional pharmacokinetic-pharmacodynamic relationships also needed to be developed for combination therapies.
|
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.
1 To whom requests for reprints should be addressed. 
2 The abbreviations used are: t1/2 final, final half-life; SF, surviving fraction; C-t, concentration-time; Cl, clearance; MRT, mean residence time; ALAT, alanine aminotransferase; ASAT, aspartate aminotransferase; -GT, -glutamyltransferase; AUC, area under the concentration curve; Hb, hemoglobin; AUC/D, normalized AUC; % D, percentage of the dose.
Received 3/18/99;
revised 6/ 7/99;
accepted 6/ 9/99.
|
REFERENCES
|
|---|
-
Voegeli R., Schumacher W., Engel J., Respondek J., Hilgard P. D-19466, a new cyclobutane-platinum complex with antitumor activity. J. Cancer Res. Clin. Oncol., 116: 439-442, 1990.[Medline]
-
Harstrick A., Bokemeyer C., Scharnofske M., Hapke G., Reille D., Schmoll H. J. Preclinical activity of a new platinum analogue, lobaplatin, in cisplatin-sensitive and -resistant human testicular, ovarian, and gastric carcinoma cell lines. Cancer Chemother. Pharmacol., 33: 43-47, 1993.[Medline]
-
Fiebig H. H., Hens H., Mross K. Phase I clinical trial of lobaplatin (D-19466) after intravenous bolus injection. Onkologie, 17: 142-148, 1994.
-
Meijer C., Mulder N. H., Timmer-Bosscha H., Sluiter W. J., Meersma G. J., de Vries E. G. E. Relationship of cellular glutathione to the cytotoxicity and resistance of seven platinum compounds. Cancer Res., 52: 6885-6889, 1992.[Abstract/Free Full Text]
-
Gietema J. A., Guchelaar H. J., de Vries E. G. E., Aulenbacher P., Sleijfer D. T., Mulder N. H. A Phase I study of lobaplatin (D-19466) administered by 72 h continous infusion. Anticancer Drugs, 4: 51-55, 1993.[Medline]
-
Gietema J. A., Veldhuis G. J., Guchelaar H. J., Willemse P. H., Uges D. R., Cats A., Boonstra H., van der Graaf W. T., Sleijfer D. T., de Vries E. G. E. Phase II and pharmacokinetic study of lobaplatin in patients with relapsed ovarian cancer. Br. J. Cancer, 71: 1302-1307, 1995.[Medline]
-
Kavanagh J., Edwards C. L., Freedman R. S., Finnegan M. B., Balat O., Tresukosol D., Burk K., Loechner S., Hord M., Franklin J. L. A trial of lobaplatin (D-19466) in platinum-resistant ovarian cancer. Gynecol. Oncol., 58: 106-109, 1995.[Medline]
-
Köhne-Wömpner C. H., Harstrick A., Bachmann P. Single agent lobaplatin is active in patients with oesophageal squamous cell carcinoma: a Phase II evaluation. Ann. Oncol., 5: 8 1994.[Free Full Text]
-
Fiebig H. H. Phase I and II studies with lobaplatin. 7th International Symposium on Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy, 176 Free University Amsterdam 1995.
-
Mross K., Meyberg F., Fiebig H. H. Pharmacokinetic and pharmacodynamic study with lobaplatin (D-19466), a new platinum complex, after bolus administration. Onkologie, 15: 139-146, 1992.
-
Welink J., Pechstein B., Van der Vijgh W. J. F. Determination of the two diastereoisomers of lobaplatin (D-19466) in plasma ultrafiltrate of cancer patients with a normal or an impaired kidney or liver function by high-performance liquid chromatography with ultraviolet detection. J. Chromatogr. Biomed. Appl., 675: 107-111, 1996.
-
Heinzel G., Woloszcak R., Thomann P. TopFit 2.0. Pharmacokinetic and Pharmacodynamic Data Analysis System Gustav Fischer Verlag/VCH Publishers New York 1993.
-
Ratain M. J., Schilsky R. L., Conley B. A. Pharmacodynamics in cancer therapy. J. Clin. Oncol., 10: 1739-1753, 1990.
-
Cockcroft D. W., Gault M. H. Prediction of creatinine clearance from serum creatinine. Nephron, 16: 31-41, 1976.[Medline]
-
Workman P., Graham M. A. Pharmacokinetics and Cancer Chemotherapy. , 17: 58-61, CSHL Press Plainview, NY 1993.
-
Van der Vijgh W. J. F., Klein I. Protein binding of five platinum compounds. Cancer Chemother. Pharmacol., 18: 129-132, 1986.[Medline]
-
Elferink F., Van der Vijgh W. J. F., Klein I. Pharmacokinetics of carboplatin after iadministration. Cancer Treat. Rep., 71: 1231-1237, 1987.[Medline]
-
Van der Vijgh W. J. F. Clinical pharmacokinetics of carboplatin. Clin. Pharmacokin., 21: 242-261, 1991.[Medline]
-
Vermorken J. B., Van der Vijgh W. J. F., Klein I. Pharmacokinetics of free and total platinum species after short-term infusion of cisplatin. Cancer Treat. Rep., 68: 505-513, 1984.[Medline]
-
Vermorken J. B., Van der Vijgh W. J. F., Klein I. Pharmacokinetics of free and total platinum species after rapid and prolonged infusions of cisplatin. Clin. Pharmacol. Ther., 39: 136-144, 1986.[Medline]
-
Calvert A. H., Newell D. R., Gumbrell L. A., O’Reilly S., Burnell M., Boxal F. E., Siddik Z. H., Judson I. R., Gore M. E., Wiltsaw E. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J. Clin. Oncol., 11: 1748-1756, 1989.
Top
ABSTRACT
INTRODUCTION
