QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.105.008326v1 34/6/1041    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zandvliet, A. S. Articles by Huitema, A. D. R. Search for Related Content PubMed PubMed Citation Articles by Zandvliet, A. S. Articles by Huitema, A. D. R.

SATURABLE BINDING OF INDISULAM TO PLASMA PROTEINS AND DISTRIBUTION TO HUMAN ERYTHROCYTES

Department of Pharmacy and Pharmacology, The Netherlands Cancer Institute/Slotervaart Hospital, Amsterdam, The Netherlands (A.S.Z., J.H.B., A.D.R.H.); Eisai Ltd., London, United Kingdom (W.C.); Department of Medical Oncology, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands (J.H.M.S.); and Department of Biomedical Analysis, Section of Drug Toxicology, Utrecht University, Utrecht, The Netherlands (J.H.M.S., J.H.B.)

(Received November 14, 2005; accepted March 21, 2006)

	Abstract

Top Abstract Materials and Methods Results Discussion and Conclusion References

 
The anticancer agent indisulam has a nonlinear pharmacokinetic profile, which may be partly related to saturable binding to blood constituents. To gain insight into the complex nonlinear behavior of indisulam, we investigated binding to plasma proteins and erythrocytes. The purpose of the study was to develop a physiological model for the distribution of indisulam in blood. Concentrations of radiolabeled indisulam were measured in vitro 1) in total plasma and in ultrafiltrate to investigate plasma protein binding, 2) in erythrocytes and in plasma to investigate distribution to erythrocytes, and 3) in erythrocyte membranes to investigate nonspecific binding in erythrocytes. For in vivo assessment, 21 patients received 400 to 900 mg/m² indisulam in a 1- or 2-h infusion. Total and free concentrations in plasma and concentrations in erythrocytes were determined at multiple time points. In vitro plasma protein binding was described by a Langmuir model with a maximal binding capacity (B_max = 767 µM) and an equilibrium dissociation constant (K_D = 1.02 µM). The maximal capacity of plasma protein binding in vivo corresponded to albumin levels. The bound concentration in erythrocytes was described by a two-site model, comprising a saturable and a nonspecific binding component. The saturable component (B_max = 174 µM) may correspond to binding to carbonic anhydrase. The physiological model adequately described the nonlinear disposition of indisulam in whole blood. Indisulam was bound to plasma proteins and distributed to erythrocytes in a saturable manner. These saturable processes may be attributed to binding to albumin (in plasma) and to carbonic anhydrase (in erythrocytes).

In a phase I clinical program, indisulam was administered in four different treatment schedules: single 1-h infusion every 3 weeks (Raymond et al., 2002), daily times five 1-h infusion every 3 weeks (Punt et al., 2001), weekly times four 1-h infusion every 6 weeks (Dittrich et al., 2003), and continuous infusion over 5 days every 3 weeks (Terret et al., 2003). Phase II clinical studies are currently ongoing to evaluate the efficacy of indisulam as a single agent and in combination with standard therapies. Indisulam had moderate single-agent antitumor activity in patients with colorectal cancer, metastatic breast cancer, head and neck cancer, nonsmall cell lung cancer, renal cell cancer, or metastatic melanoma who did not respond to previous chemotherapy (Mainwaring et al., 2002; Talbot et al., 2002; Fumoleau et al., 2003; Haddad et al., 2004; Raftopoulos et al., 2004; Smyth et al., 2005).

The pharmacokinetic profile of indisulam is highly nonlinear, characterized by increasing clearance with decreasing plasma concentration. Systemic exposure (expressed as the area under the plasma concentration versus time curve) has been shown to increase disproportionately with dose in four phase I studies (Punt et al., 2001; Raymond et al., 2002; Dittrich et al., 2003; Terret et al., 2003).

A population pharmacokinetic model has been developed previously (van Kesteren et al., 2002). This empirical model comprised saturable distribution and two elimination pathways: a linear pathway and a saturable pathway. It adequately described the individual pharmacokinetic profiles for all four treatment schedules tested in phase I. It has been postulated that metabolic products of indisulam may be formed through saturable enzymatic processes, which may underlie the saturable elimination pathway (van Kesteren et al., 2002). Furthermore, it has been suggested that the saturable distribution pathway may be related to distribution to erythrocytes (van Kesteren et al., 2002). This latter hypothesis was supported by the results of an in vitro pharmacokinetic study, which demonstrated that the partition coefficient between erythrocytes and plasma of indisulam decreases with increasing incubation concentration (van den Bongard et al., 2003). However, conclusive physiological explanations for the observed nonlinear pharmacokinetic processes have not been given yet.

Reversible neutropenia and thrombocytopenia were the major and dose-limiting toxicities of indisulam (Punt et al., 2001; Raymond et al., 2002; Dittrich et al., 2003; Terret et al., 2003). It has clearly been demonstrated that hematological toxicity is related to the exposure to indisulam (van Kesteren et al., 2005), The nonlinear pharmacokinetics of this drug may be an important factor in the observed variability in pharmacokinetics and hematological toxicity. Therefore, it was considered of crucial importance to characterize the nonlinear properties of this drug. The objective of the current study was to develop a nondynamic physiological model to describe the in vitro distribution of indisulam in whole blood. This was identified as an important step for the full elucidation of the nonlinear pharmacokinetic profile of indisulam.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion and Conclusion References

 
Three in vitro experiments were conducted to investigate plasma protein binding (experiment 1), uptake in erythrocytes (experiment 2), and nonspecific binding to erythrocyte membranes (experiment 3). The combined results were used to define a physiological model for the distribution of indisulam in vitro in whole blood. Plasma protein binding and uptake in erythrocytes were also assessed in a clinical study. In vivo applicability of the model was evaluated using data from 21 patients.

In all in vitro experiments, carbon-14-(¹⁴C) radiolabeled indisulam (Amersham Pharmacia Biotech UK Ltd., Buckinghamshire, UK) was used, and concentrations of [¹⁴C]indisulam were measured using liquid scintillation counting. The concentration ranges used for the in vitro experiments corresponded to clinically observed plasma concentrations of indisulam.

In Vitro Plasma Protein Binding (Experiment 1). Total and free plasma concentrations of [¹⁴C]indisulam in plasma were measured in vitro at six concentration levels in triplicate. The difference between total and free concentrations corresponded to the concentration of [¹⁴C]indisulam bound to plasma proteins. The relationship between the free plasma concentration (C_free) and the protein-bound fraction (C_bound) of indisulam was examined.

Fresh whole blood samples from three male healthy volunteers were collected from an antecubital vein into heparinized tubes. Blood samples were centrifuged at 3000 rpm for 10 min at 4°C to obtain plasma. A 52 mM stock solution of [¹⁴C]indisulam in DMSO (radiochemical purity, 97.5-98.1%) was prepared. This was diluted with DMSO to obtain [¹⁴C]indisulam working solutions of 0.26, 0.52, 5.2, 13, 26, and 39 mM. Five microliters of [¹⁴C]indisulam DMSO working solutions of 0.26, 0.52, 5.2, 13, 26, 39, and 52 mM were added to 0.495 ml of human plasma (final drug concentrations: 0.26, 0.52, 5.2, 13, 26, 39, and 52 µM). For a control sample, 5 µl of [¹⁴C]indisulam DMSO solution (0.26 mM) were added to 0.495 ml of phosphate-buffered saline (PBS; pH 7.4) and mixed (final concentration 2.6 µM, which was expected to be in the range of free drug concentrations in the plasma samples). Equilibrium dialysis was performed in 1-ml modules (Yazawa Kagaku Co., Ltd., Tokyo, Japan) fitted with Seamless Cellulose Tubing (UC20-32-100; size, 20/32) with a molecular mass cut-off of 12,000 14,000 Da (Sanko Junyaku Co. Ltd., Tokyo, Japan). Each sample (0.5 ml) was equilibrated against 0.5 ml of PBS at room temperature for 24 h. Plasma and PBS were sampled from each side of the cell to determine radioactivity of [¹⁴C]indisulam in each matrix using the Liquid Scintillation Analyzer LSA-2700TR (Packard Co., Ltd., Tokyo, Japan). Fifty microliters of plasma were dissolved in 1 ml of a solution containing tissue solubilizer Soluene-350 (PerkinElmer Life and Analytical Science, Boston, MA) and 2-propanol (1:1). After storage for a day at room temperature, 15 ml of hydrophilic scintillator (Instagel; PerkinElmer Life and Analytical Sciences) containing 10% 0.5 M HCl (v/v) were added to the sample, and radioactivity was measured. To 50 µl of PBS, 15 ml of hydrophilic scintillator (ACS II; Amersham) were directly added, and radioactivity was measured.

In Vitro Distribution to Erythrocytes (Experiment 2). The uptake of indisulam in erythrocytes was studied at various incubation times and at various concentrations as described previously in detail by van den Bongard et al. (2003). Concentrations of indisulam were measured in plasma and in erythrocytes. The fraction of indisulam in erythrocytes versus the total incubation concentration was investigated. For implementation in a physiological distribution model, the relationship between the free concentration (calculated from the total plasma concentration using the data from experiment 1) and the concentration of indisulam bound to erythrocytes was established in the current analysis.

In brief, whole blood from healthy volunteers was incubated at 37°C with [¹⁴C]indisulam at a final concentration of 518 µM(n = 6). Samples were taken immediately after the addition of [¹⁴C]indisulam, at 10 and 30 min and at 1, 2, 3, 4, 5, 6, 7, 8, and 25 h after the addition of [¹⁴C]indisulam. Another series of whole blood samples was incubated at 37°C with [¹⁴C]indisulam at concentrations of 10.4, 51.8, 130, 259, 389, and 518 µM (n = 3) during 2 h. After centrifugation, plasma (50 µl) and erythrocytes were isolated from all samples. Erythrocytes (200 µl) were washed with ice-cold isotonic PBS and were dissolved and decolorized using Solvable (1 ml; Packard, Groningen, The Netherlands), 0.1 M EDTA (100 µl), and hydrogen peroxide (500 µl). Both plasma and erythrocyte samples were mixed with 10 ml of Ultima Gold cocktail (PerkinElmer Life and Analytical Sciences), and the detection of ß radiation was performed by a liquid scintillation counter (Tri-CARB 2100 CA; PerkinElmer Life and Analytical Sciences) (van den Bongard et al., 2003).

In Vitro Binding to Erythrocyte Membranes (Experiment 3). This experiment was conducted to investigate binding of indisulam to erythrocyte membranes and to quantify the attribution of this process to total uptake in erythrocytes. For this objective, both the total concentration in erythrocytes and the membrane-bound concentration were measured. Whole blood samples from three male healthy volunteers were collected from an antecubital vein into heparinized tubes. Unlabeled indisulam (Eisai Co., Ltd., Kashima, Japan) and [¹⁴C]indisulam (9:1) were dissolved in DMSO to obtain a 259 mM stock solution (radiochemical purity 97.57-99.23%). The stock solution was diluted with DMSO to obtain working solutions of 2.6 and 26 mM. Five microliters of indisulam solution (concentration 2.6, 26, and 259 mM) were added to 4.995 ml of whole blood of each volunteer and mixed (final concentration 2.6, 26, and 259 µM). Blood was incubated at 37°C in a water bath, and samples of 2 ml were taken after 5-min and 6-h incubation. Blood samples were immediately centrifuged at 3000 rpm for 10 min at 4°C. Erythrocytes were washed twice with 4 ml of ice-cold saline and centrifuged at 2000 rpm for 3 min at 4°C.

Twenty microliters of washed erythrocytes were collected for measurement of the total erythrocyte concentration of indisulam. To 100 µl of erythrocytes, 0.8 ml of distilled water was added for hemolysis. After vigorous shaking, the suspension was centrifuged at 20,000 rpm for 20 min at 4°C (TL-100, TLA-100.3 motor; Beckman Coulter, Fullerton, CA). The pellet (erythrocyte membrane) was washed with 1 ml of distilled water and then centrifuged at 20,000 rpm for 20 min at 4°C (TL-100,TLA-100.3 motor; Beckman Coulter). The erythrocyte membranes were dissolved in 1 ml of a solution containing tissue solubilizer Soluene-350 (PerkinElmer Life and Analytical Sciences) and 2-propanol (1:1). Radioactivity was determined in triplicate samples of erythrocytes and erythrocyte membranes of each concentration and sampling time. In addition, radioactivity was measured in supernatant fractions, which were separated from the membranes. All samples were decolorized by the addition of 200 µl of 30% hydrogen peroxide. After storage for a day at room temperature, 15 ml of a hydrophilic scintillator (Instagel; PerkinElmer Life and Analytical Sciences) containing 10% 0.5 M HCl (v/v) were added, and the radioactivity was measured using the Liquid Scintillation Analyzer LSA-2700TR (Packard Co., Ltd., Tokyo, Japan).

In Vivo Blood Distribution. Because the final model should essentially describe the in vivo blood distribution of indisulam, binding to plasma protein and uptake in erythrocytes was assessed in patients. Pharmacokinetic data were obtained from a phase I dose-escalation study in Japanese patients (Yamada et al., 2005). This study was performed after a larger phase I program in Caucasian patients had been finalized. The study was reviewed by the Medical Ethics Committee of the National Cancer Center Hospital in Tokyo (Japan) and was performed in accordance with the Declaration of Helsinki (World Medical Association, 2000). All 21 eligible patients gave written informed consent before inclusion in the study. Indisulam was administered at five different dose levels [400 (n = 3), 600 (n = 3), 700 (n = 6), 800 (n = 6), and 900 (n = 3) mg/m²]. The lower doses (400 and 600 mg/m²) were administered as 1-h infusions and the higher doses as 2-h infusions. Pharmacokinetic assessment was performed according to a full sampling schedule. Total plasma concentrations and concentrations in erythrocytes of indisulam were measured in whole blood samples taken at 30 min after the start of infusion, at 0, 10, and 30 min after the end of infusion, and at 1, 2, 4, 6, 10, 24, 48, 72, 96, 120, 168, and 240 h after the end of infusion. From each patient, two additional samples were taken at the end of infusion and at 1 h after the end of infusion for measurement of total and free plasma concentrations of indisulam. Whole blood samples were centrifuged at 3000 rpm for 10 min at 5°C. Plasma samples were transferred to a glass tube (0.50 ml) and to ultrafiltration devices (0.95 ml, Centrifree; Amicon Corporation, Danvers, MA). To obtain 0.50 ml of ultrafiltrate, samples were ultracentrifuged at 1670g for 20 to 45 min. Plasma ultrafiltrate samples were transferred to a plastic vial (0.50 ml). Phosphate buffer (1.0 ml, 0.1 M, pH 6.8) was added to the plasma and plasma ultrafiltrate samples. Erythrocytes (0.50 ml) were frozen in methanol for 10 min and thawed at room temperature for 10 min. This freeze-thaw cycle was performed three times. Boric acid buffer (2.0 ml, 0.6 M, pH 6.8) was added, and the mixture was boiled for 1 min. Indisulam was extracted using ethyl acetate. After centrifugation, the organic layer was transferred into a glass tube and evaporated under nitrogen. Samples were reconstituted in 0.20 ml of CH₃CN/phosphate buffer (6.7 mM, pH 6.6) [360:640 (v/v)]. The concentration of indisulam was measured by high-pressure liquid chromatography with UV detection (Yamada et al., 2005). The method was linear between 0.052 µM and 130 µM, and the assay accuracy and precision were <15.5%. N-(3-Chloro-7-indolyl)-4-(N-methylsulfamoyl)benzenesulfonamide (ER-67771) was used as an internal standard (Owa et al., 1999).

Data Analysis. Data were analyzed using the NONMEM program (version V, level 1.1; Globomax LLC, Hanover, MD) (Beal et al., 1988). The first-order conditional estimation (FOCE) method with interaction between interindividual, intraindividual, and residual variability was applied. PDx-Pop (version 1.1j, release 4; GloboMax LLC) was used as an interface for data and output processing and for modeling management. The adequacy of the models was evaluated by graphical plots of observed versus model predicted concentrations and bound concentrations versus free concentrations (overlay of observed and model predicted values). Statistical discrimination between hierarchical models was based on the log-likelihood ratio test.

Results from in vitro studies were analyzed to define a model for distribution of indisulam in whole blood, comprising plasma protein binding and binding of indisulam to erythrocytes. For both of these components, models for linear binding kinetics (eq. 1) and models for saturable binding equilibriums (eq. 2) were tested.

(1)

(2)

In these equations, N is the nonspecific binding constant, B_max corresponds to the maximal binding capacity, and K_D is equilibrium dissociation constant. If the free concentration is equal to K_D, the bound concentration is half-maximal.

The physiological model was developed to describe the in vitro distribution in whole blood under equilibrium conditions. It was assumed that binding between indisulam and protein binding sites occurred instantaneously. Thus, the fractions bound and free indisulam within plasma and within erythrocytes were assumed to be in equilibrium at any moment. Free concentrations in plasma and in erythrocytes were not in equilibrium at any time, since distribution to erythrocytes was shown to be time-dependent previously. van den Bongard et al. (2003) have demonstrated that equilibrium was reached after 2 h of incubation in vitro. Therefore, only results obtained after incubation for 2 h or longer were considered in the current analysis.

Model development was conducted in a step-wise approach. Initially, a relationship between free plasma concentrations and protein bound plasma concentrations was established. Data from experiment 1 were used for this analysis. Measurement of free plasma concentrations was not performed in experiments 2 and 3. Therefore, the protein binding model was subsequently applied to total plasma concentrations that were measured in experiments 2 and 3 to calculate free plasma concentrations. In vitro, free concentrations in the liquid interior of erythrocytes were assumed to be equal to calculated free plasma concentrations. This may be reasonable because equilibrium conditions were ascertained. Finally, a model for binding of indisulam to erythrocytes was developed to describe the relationship between free and bound concentrations of indisulam in washed erythrocytes.

Clinical data were used to evaluate the in vivo applicability of the nondynamic physiological model. In vivo, steady-state conditions did not apply, and thus equilibrium conditions could not be fully guaranteed. However, interference by dynamic processes was minimized by selection of data measured after completion of indisulam infusion. Parameters describing the plasma protein binding of indisulam and its binding to erythrocytes were re-estimated with in vivo data.

	Results

Top Abstract Materials and Methods Results Discussion and Conclusion References

 
Model Development. In vitro Plasma Protein Binding. In vitro protein binding results were included in the data analysis for concentration levels 52, 130, 259, 389, and 518 µM only, because the free indisulam concentrations at incubation concentration 5.2 µM were below the lower limit of quantitation. Plasma protein binding of indisulam was high, varying from 99.6 to 99.9% in the concentration range 52 to 518 µM. The free fraction increased with increasing incubation concentration, which indicates that protein binding is saturable.

To further investigate this observation, concentrations of protein bound indisulam (C_{plasma, bound}) were calculated by subtracting free concentrations (C_{plasma, free}) from total plasma concentrations (C_{plasma, total}). The relationship between C_{plasma, bound} and C_{plasma, free} was best described by a saturable binding model (eq. 2). The maximal plasma protein binding capacity B_{max, plasma} was estimated to be 767 µM, and binding was half-maximal at a free concentration of 1.02 µM (K_{D plasma}, Table 1). Figure 1 shows that the concentrations observed in vitro were adequately described by the model.

View larger version (11K): [in this window] [in a new window]   FIG. 1. Plot of protein bound versus free plasma concentrations of indisulam obtained from in vitro experiments (experiment 1). Both observed () and model predicted concentrations (—) are depicted.

Figure 2 represents the physiological model for the distribution of indisulam in whole blood. In vitro, free concentrations in erythrocytes (C_{erythrocytes, free}) were assumed to be equal to free plasma concentrations (C_{plasma, free}) and could be calculated from total plasma concentrations using eq. 3, which is a rearrangement of eq. 2 (with and B_max = 767 µM and K_D = 1.02 µM).

(3)

View larger version (25K): [in this window] [in a new window]   FIG. 2. Graphical representation of the physiological model of the blood distribution of indisulam. The compound is bound to plasma protein and distributed to erythrocytes. Bmax is the maximal binding capacity, KD is the equilibrium dissociation constant of the binding equilibrium, and N is the nonspecific binding constant.

View larger version (13K): [in this window] [in a new window]   FIG. 3. Concentrations of indisulam bound to erythrocytes versus free plasma concentrations in vitro. Observed values () corresponded well to individual predicted values (). Binding becomes partially saturated at higher concentrations but remains to slightly increase because of a nonspecific binding component.

(4)

View larger version (13K): [in this window] [in a new window]   FIG. 4. Observed protein bound versus free plasma concentrations of indisulam () obtained from in vivo experiments. A, the model derived from in vitro results (—) described the corresponding in vivo results reasonably well. B, inclusion of the albumin plasma concentration as a covariate for the maximal binding capacity resulted in an improved model ().

Application to in Vivo Data. In total, 350 samples were taken from 21 patients for the in vivo assessment of the blood distribution of indisulam. Samples taken during infusion (n = 21) were excluded from data analysis (see Data Analysis under Materials and Methods).

In vivo Plasma Protein Binding. The model developed with in vitro data could describe the corresponding results from the in vivo study reasonably well (Fig. 4A). The fit between observed data from the in vivo study and the model significantly improved upon the addition of the individual albumin level [ALB (µM)] as a covariate for B_{max plasma}. This important finding indicates that the albumin level of an individual patient may be predictive for the pharmacokinetic behavior of indisulam in that particular individual.

In eq. 5, the number of binding sites per albumin molecule is represented by n.

(5)

The number of binding sites was not statistically significantly different from 1, and n was consequently fixed at 1. Using in vivo results only, K_{D plasma} was re-estimated to be 0.65 µM (Table 2). This adjusted model for plasma protein binding described the in vivo observations very well (Fig. 4B).

In vivo Distribution to Erythrocytes. In 287 of 329 available samples, total plasma concentrations and concentrations in washed erythrocytes were measured. Free concentrations in plasma were calculated using eq. 5. The model for in vitro distribution to erythrocytes did not correctly describe the results from the clinical study. However, when the parameters describing the uptake in erythrocytes were re-estimated using the in vivo data, the two-site binding model proved to be adequate (Fig. 5). Table 2 lists the parameter values after re-estimation using in vivo data from 21 patients. All parameters were precisely estimated. The nonspecific binding coefficient N_erythrocytes was estimated to be higher in vivo than in vitro (70.1 versus 7.4), and the equilibrium dissociation constant K_{D erythrocytes} was lower in vivo than in vitro (0.00379 versus 0.087 µM). Other in vivo parameter estimates corresponded well to the in vitro model.

View larger version (19K): [in this window] [in a new window]   FIG. 5. Concentrations of indisulam bound to erythrocytes versus free plasma concentrations in vivo. Individual predicted values from the two-site binding model () corresponded well to the observed values ().

View larger version (14K): [in this window] [in a new window]   FIG. 6. Plasma concentration versus time profile of an individual Caucasian patient after administration of 1000 mg/m2 indisulam in a 1-h infusion. The pharmacokinetic profile is profoundly nonlinear. The concentrations at which the binding of indisulam to plasma proteins and erythrocytes are half-maximal are within the therapeutic concentration range.

	Discussion and Conclusion

Top Abstract Materials and Methods Results Discussion and Conclusion References

 
A physiological model for the distribution of indisulam in whole blood was successfully developed. We have demonstrated that indisulam binds to plasma proteins and to erythrocytes in a saturable manner, which may be attributed to binding to albumin (in plasma) and to carbonic anhydrase (in erythrocytes).

In a previous study, protein binding of indisulam was reported to be high (98-99%), and no saturation in the concentration range 52 to 518 µM could be demonstrated. However, in that study the precision was insufficient (standard deviation was 0.05 to 0.66% versus 0.006 to 0.027% in the current experiment) to entirely preclude saturation of protein binding (van den Bongard et al., 2003). The currently presented in vitro and in vivo studies have clearly shown that plasma protein binding of indisulam is a saturable process. The maximal binding capacity was highly dependent on individual plasma albumin levels. Therefore, albumin may be the major plasma binding protein of indisulam in the concentration range studied. The albumin level of an individual may be predictive for the pharmacokinetic profile of indisulam. Hypoalbuminemia, which is common in cancer patients (Mariani et al., 1976), may result in a reduction of albumin-bound indisulam concentrations and a consequent increase in free plasma concentrations. The clinical implications of this finding will need further investigation. In the current study, plasma protein binding in blood from cancer patients (in vivo) was lower than in blood from healthy volunteers (in vitro). This explains the bias in Fig. 4A: plasma protein binding in patients was underestimated by the in vitro model, which was based on data from healthy volunteers. The estimated 767 µM for B_max (in vitro) corresponds to an albumin level of 50.6 g/l, which is within the normal range for healthy male volunteers (34-54 g/l) (U.S. National Library of Medicine and the National Institutes of Health, 2005, http://www.nlm.nih.gov/medlineplus/ency/article/003480.htm).

In addition to binding to plasma proteins, binding to erythrocytes was also shown to be saturable. Because carbonic anhydrase is abundant in erythrocytes (Demir et al., 1997) and because indisulam is a potent inhibitor of this enzyme (Abbate et al., 2004), it was hypothesized that binding to carbonic anhydrase I and II may explain the saturable character of distribution to erythrocytes. Binding parameters describing the distribution of indisulam to erythrocytes were in agreement with the total concentration of carbonic anhydrase I and II. For the structurally related carbonic anhydrase inhibitors acetazolamide, dorzolamide, and para-iodo-benzenesulfonamide, maximal binding capacities for carbonic anhydrase I and carbonic anhydrase II in erythrocytes were 117 to 166 µM and 16.1 to 19.9 µM, respectively (total 133-186 µM) (Bayne et al., 1979; Singh and Wyeth, 1991; Hasegawa et al., 1994). The estimated 174 µM (in vitro) and 146 µM (in vivo) values for indisulam correspond well to these previously reported concentrations.

Binding of indisulam to erythrocytes was higher in vivo than in vitro. This may be due to differences in the binding affinity of indisulam related to different protein conformations in vivo and in vitro. Another explanation for the difference between the in vivo and in vitro erythrocyte binding parameters may be a potential pH gradient. Indisulam is a potent carbonic anhydrase inhibitor and can therefore cause an intracellular pH increase. Because indisulam is an acid (pK_a 7.01), a pH increase will result in a decrease of neutral indisulam molecules. Because only neutral molecules can diffuse across the erythrocyte membrane, inhibition of carbonic anhydrase could cause the ratio between C_{erythrocytes, free} and C_{plasma, free} to increase. Therefore, the estimates of N_erythrocytes and K_{D erythrocytes} for the in vivo distribution of indisulam may have been over- and underestimated, respectively. Finally, the differences in erythrocyte binding parameters between in vitro and in vivo conditions may be partially explained by the lack of steady-state conditions. Due to relatively slow redistribution of indisulam from erythrocytes back to plasma, the ratio between C_{erythrocytes, free} and C_{plasma, free} may have been higher in vivo than in the closed in vitro system. Hence, the values of N_erythrocytes and K_{D erythrocytes} may also have been over- and underestimated, respectively, to some extent because of the lack of steady-state conditions. The observed ratios of N_erythrocytes and K_{D erythrocytes} between in vitro and in vivo conditions were 0.11 and 23, respectively. Because the observed differences were large, it is likely that changes in protein conformations and/or environmental differences between in vivo and in vitro conditions can not be ruled out by a potential lack of equilibrium.

In conclusion, the blood distribution of indisulam in vivo and in vitro is highly determined by saturable processes. It is likely that these processes comprise binding to albumin (in plasma) and to carbonic anhydrase I and II (in erythrocytes). Saturable plasma protein binding may have impact on the higher ranges of the concentration-time profile of indisulam, whereas saturable erythrocyte binding may affect the terminal elimination phase. The saturable character of plasma protein binding and distribution to erythrocytes may thus provide a partial explanation for the nonlinearity in the pharmacokinetic profile. However, in vivo, indisulam is not only distributed within the blood compartment but to other body fluids and tissues as well. Furthermore, indisulam is continuously metabolized and eliminated from the body. Because distribution beyond the blood compartment and elimination are dynamic processes, it is essential to include the blood distribution in a physiological pharmacokinetic model to fully understand the exceptional nonlinear pharmacokinetic behavior of indisulam.

	Acknowledgments

 
We acknowledge Desirée van den Bongard (The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital), Sanae Yasuda, Motoharu Kakiki, Tomio Takamatsu, Yuichi Inai, Tsutomu Yoshimura (Eisai Co., Ltd.), Yasuhide Yamada, and Tomohide Tamura (The National Cancer Center Hospital, Tokyo) for contributing to this project.

	Footnotes

 
This research was supported by a grant from the Eisai network of companies. Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.

doi:10.1124/dmd.105.008326.

ABBREVIATIONS: DMSO, dimethyl sulfoxide; PBS, phosphate-buffered saline; IIV, interindividual variability.

Address correspondence to: Anthe Zandvliet, Slotervaart Hospital, Department of Pharmacy & Pharmacology, Louwesweg 6, 1066 EC Amsterdam, The Netherlands. E-mail: apaza{at}slz.nl

	References

Top Abstract Materials and Methods Results Discussion and Conclusion References

 


Abbate F, Casini A, Owa T, Scozzafava A, and Supuran CT (2004) Carbonic anhydrase inhibitors: E7070, a sulfonamide anticancer agent, potently inhibits cytosolic isozymes I and II and transmembrane, tumor-associated isozyme IX. Bioorg Med Chem Lett 14: 217-223.[CrossRef][Medline]

Bayne WF, Chu LC, and Theeuwes F (1979) Acetazolamide binding to two carbonic anhydrase isoenzymes in human erythrocytes. J Pharm Sci 68: 912-913.[Medline]

Beal SL, Boeckman AJ, and Sheiner LB (1988) NONMEM User's Guides. University of California at San Francisco, San Francisco, CA.

Demir N, Demir Y, Bakan E, and Kufrevioglu OI (1997) A different structural feature for carbonic anhydrases in human erythrocytes. Prep Biochem Biotechnol 27: 279-287.[Medline]

Dittrich C, Dumez H, Calvert H, Hanauske A, Faber M, Wanders J, Yule M, Ravic M, and Fumoleau P (2003) Phase I and pharmacokinetic study of E7070, a chloroindolyl-sulfonamide anticancer agent, administered on a weekly schedule to patients with solid tumors. Clin Cancer Res 9: 5195-5204.[Abstract/Free Full Text]

Fukuoka K, Usuda J, Iwamoto Y, Fukumoto H, Nakamura T, Yoneda T, Narita N, Saijo N, and Nishio K (2001) Mechanisms of action of the novel sulfonamide anticancer agent E7070 on cell cycle progression in human non-small cell lung cancer cells. Investig New Drugs 19: 219-227.[CrossRef][Medline]

Fumoleau P, Viens P, Cottu PH, De Grève J, Marty M, Delaloge S, Miles D, O'Reilly S, Wardley A, Schmid P, et al. (2003) A multi center phase II study of E7070, a chloroindolyl-sulfonamide anticancer agent in anthracycline and taxane pretreated breast cancer, in the 26th San Antonio Breast Cancer Symposium; 2003 Dec 3-6; San Antonio, TX. Abstract 535.

Haddad RI, Weinstein LJ, Wieczorek TJ, Bhattacharya N, Raftopoulos H, Oster MW, Zhang X, Latham VM Jr, Costello R, Faucher JL, et al. (2004) A phase II clinical and pharmacodynamic study of E7070 in patients with metastatic, recurrent, or refractory squamous cell carcinoma of the head and neck: modulation of retinoblastoma protein phosphorylation by a novel chloroindolyl sulfonamide cell cycle inhibitor. Clin Cancer Res 10: 4680-4687.[Abstract/Free Full Text]

Hasegawa T, Hara K, and Hata S (1994) Binding of dorzolamide and its metabolite, N-deethylated dorzolamide, to human erythrocytes in vitro. Drug Metab Dispos 22: 377-382.[Abstract]

Mainwaring PN, Van Cutsem E, Van Laethem JL, Propper D, Twelves C, Bridgewater J, Audhuy B, Carmichael J, Punt CJA, Cassidy J, et al. (2002) A multicentre randomised phase II study of E7070 in patients with colorectal cancer who have failed 5-fluorouracil-based chemotherapy (Abstract). Proc Am Soc Clin Oncol 21: 611.

Mariani G, Strober W, Keiser H, and Waldmann TA (1976) Pathophysiology of hypoalbuminemia associated with carcinoid tumor. Cancer 38: 854-860.[Medline]

Oda Y, Owa T, Sato T, Boucher B, Daniels S, Yamanaka H, Shinohara Y, Yokoi A, Kuromitsu J, and Nagasu T (2003) Quantitative chemical proteomics for identifying candidate drug targets. Anal Chem 75: 2159-2165.[Medline]

Owa T, Yoshino H, Okauchi T, Yoshimatsu K, Ozawa Y, Sugi NH, Nagasu T, Koyanagi N, and Kitoh K (1999) Discovery of novel antitumor sulfonamides targeting G1 phase of the cell cycle. J Med Chem 42: 3789-3799.[CrossRef][Medline]

Ozawa Y, Sugi NH, Nagasu T, Owa T, Watanabe T, Koyanagi N, Yoshino H, Kitoh K, and Yoshimatsu K (2001) E7070, a novel sulphonamide agent with potent antitumour activity in vitro and in vivo. Eur J Cancer 37: 2275-2282.[CrossRef][Medline]

Punt CJA, Fumoleau P, van de Walle B, Faber MN, Ravic M, and Campone M (2001) Phase I and pharmacokinetic study of E7070, a novel sulfonamide, given at a daily times five schedule in patients with solid tumors. A study by the EORTC-early clinical studies group (ECSG). Ann Oncol 12: 1289-1293.[Abstract/Free Full Text]

Raftopoulos H, Escudier B, Renshaw G, Lara PN, Dutcher J, Nanus D, Petrylak DP, and Sacris L (2004) A phase II multicenter study of the cyclin-dependent kinase inhibitor indisulam in patients with inoperable and/or metastatic renal cell carcinoma (RCC) (Abstract). J Clin Oncol 22 (Suppl): 4629.

Raymond E, ten Bokkel Huinink WW, Taieb J, Beijnen JH, Faivre S, Wanders J, Ravic M, Fumoleau P, Armand JP, and Schellens JHM (2002) Phase I and pharmacokinetic study of E7070, a novel chloroindolyl sulfonamide cell-cycle inhibitor, administered as a one-hour infusion every three weeks in patients with advanced cancer. J Clin Oncol 20: 3508-3521.[Abstract/Free Full Text]

Singh J and Wyeth P (1991) The enzyme-inhibitor approach to cell-selective labelling-I. Sulphonamide inhibitors of carbonic anhydrase as carriers for red cell labelling: in vitro uptake of pIBS by human red blood cells. Int J Rad Appl Instrum A 42: 251-259.[Medline]

Smyth JF, Aamdal S, Awada A, Dittrich C, Caponigro F, Schoffski P, Gore M, Lesimple T, Djurasinovic N, Baron B, et al. (2005) Phase II study of E7070 in patients with metastatic melanoma. Ann Oncol 16: 158-161.[Abstract/Free Full Text]

Talbot DC, Norbury C, Slade M, von Pawel J, Bosquee L, Ellis PA, Gatzemeier U, and Ravic M (2002) A phase II and pharmacodynamic study of E7070 in patients with non-small cell lung cancer (NSCLC) who have failed platinum-based chemotherapy (Abstract). Proc Am Soc Clin Oncol 21: 1306.

Terret C, Zanetta S, Roche H, Schellens JHM, Faber MN, Wanders J, Ravic M, and Droz JP (2003) Phase I clinical and pharmacokinetic study of E7070, a novel sulfonamide given as a 5-day continuous infusion repeated every 3 weeks in patients with solid tumours. A study by the EORTC Early Clinical Study Group (ECSG). Eur J Cancer 39: 1097-1104.[CrossRef][Medline]

van den Bongard HJ, Pluim D, Waardenburg RC, Ravic M, Beijnen JH, and Schellens JHM (2003) In vitro pharmacokinetic study of the novel anticancer agent E7070: red blood cell and plasma protein binding in human blood. Anticancer Drugs 14: 405-410.[CrossRef][Medline]

van Kesteren C, Mathot RAA, Raymond E, Armand JP, Dittrich C, Dumez H, Roche H, Droz JP, Punt CJA, Ravic M, et al. (2002) Population pharmacokinetics of the novel anticancer agent E7070 during four phase I studies: model building and validation. J Clin Oncol 20: 4065-4073.[Abstract/Free Full Text]

van Kesteren C, Zandvliet AS, Karlsson MO, Mathot RAA, Punt CJA, Armand JP, Raymond E, Huitema ADR, Dittrich C, Dumez H, et al. (2005) Semi-physiological model describing the hematological toxicity of the anti-cancer agent indisulam. Investig New Drugs 23: 225-234.[CrossRef][Medline]

World Medical Association (2000) World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA 284: 3043-3045.[Free Full Text]

Yamada Y, Yamamoto N, Shimoyama T, Horiike A, Fujisaka Y, Takayama K, Sakamoto T, Nishioka Y, Yasuda S, and Tamura T (2005) Phase I pharmacokinetic and pharmacogenomic study of E7070 administered once every 21 days. Cancer Sci 96: 721-728.[CrossRef][Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.105.008326v1 34/6/1041    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zandvliet, A. S. Articles by Huitema, A. D. R. Search for Related Content PubMed PubMed Citation Articles by Zandvliet, A. S. Articles by Huitema, A. D. R.