Pharmacometric Analysis of the Effect of Furosemide on Suramin Pharmacokinetics
Stephen C. Piscitelli, Pharm.D., Alan Forrest, Pharm.D., Richard M. Lush, Ph.D., Nathan Ryan, Lloyd R. Whitfield, Ph.D., and William D. Figg, Pharm.D.
Study Objective. To characterize the effects of furosemide on the pharmacokinetics of suramin, a renally eliminated investigational antineoplastic agent.
Design. Retrospective population pharmacokinetic analysis.
Setting. Government biomedical research facility.
Patients. Twenty-six men with hormone-refractory prostate cancer and one with adrenocortical carcinoma.
Interventions. Patients received suramin by continuous or intermittent infusion with and without concomitant furosemide.
Measurements and Main Results. Optimum suramin regimens were achieved by adaptive feedback control, and pharmacokinetic data were collected both in the presence and absence of furosemide. Suramin concentrations were determined by high-performance liquid chromatography (coefficient of variation < 8%). Suramin concentrations were fit to a three- compartment linear model with six coefficients and two rate inputs, which allowed furosemide to affect suramin pharmacokinetics. Individual and population parameter estimates were determined using the iterative two- stage approach. Concomitant furosemide was associated with a median decrease in total body clearance of suramin by 36% (range 0-63%, p<0.0001). No other parameter was significantly altered, and there was no trend for change in any pharmacokinetic value with time. Suramin plasma concentrations were simulated with and without prolonged furosemide therapy in 26 patients for 12 weeks. The average suramin concentration increased by greater than 33% in 12 patients; 2 patients had a greater than 67% increase in this extreme case model.
Conclusion. Coadministration of furosemide with suramin can cause an increase in suramin concentrations; however, due to suramin’s long half- life, its rate of accumulation is very slow. Nonetheless, in individuals receiving suramin by nonadaptive control, appropriate precautions should be taken when prolonged furosemide therapy is begun.
(Pharmacotherapy 1997;17(3):431-437)
Suramin, a hexasulfonated naphthylurea, was originally synthesized in 1916 to combat a trypanosomiasis epidemic in East Africa.1 Investigation of its potential antitumor activity began after tumor regression occurred in occasional patients treated with the drug during its evaluation as a treatment for the acquired immunodeficiency syndrome.2-4 Preliminary evidence of antitumor activity was first observed in adrenocortical carcinoma5 and then in
hormone-refractory prostate cancer.6-10
Numerous mechanisms were postulated to explain the antitumor activity of suramin, such as antagonism of peptide growth factor and inhibition of glycosaminoglycan metabolism11-14; however, full understanding is elusive. Regardless of mechanism of action, prolonged exposure is argued as necessary for the drug to exert this effect. In tissue culture, suppression of human prostate cancer cell survival requires that
the cells be exposed to relatively high concen- trations (200-300 µg/ml) for 96-144 hours.13
The pharmacokinetic characteristics of suramin are unusual. The agent does not undergo metabolism, has an inordinately long elimination half-life (approximately 50 days), is highly protein bound, and its clearance appears to be by glomerular filtration.15 It is best characterized by a three-compartment model.16 Since it was thought to be renally filtered, attempts were made to find a correlation between systemic clearance and creatinine clearance. However, the analyses failed to find such a correlation, suggesting that suramin’s renal elimination may be through tubular secretion, not glomerular filtration.16 Preliminary data also suggested that probenecid affected its elimination, strengthening the hypothesis of elimination by renal secretion.
The main objective of our study was to characterize the effect of furosemide, a potent and frequently administered loop diuretic, on the pharmacokinetics of suramin. A second objective was to evaluate the clinical significance of this drug interaction by comparing the pharmacokinetics of suramin with and without furosemide in patients who began furosemide therapy while receiving suramin.
Methods
Patients
Patients from two separate suramin protocols were included in the analysis. They were eligible for these studies if they had progressive hormone-refractory prostate cancer; the exception was one who had adrenocortical carcinoma. All patients were required to have
From the Clinical Pharmacokinetics Research Laboratory, Clinical Center Pharmacy Department (Dr. Piscitelli), and the Clinical Pharmacokinetics Section, Clinical Pharmacology Branch (Drs. Lush and Figg, and Mr. Ryan), National Institutes of Health, Bethesda, Maryland; the Pharmacometric Division, Clinical Pharmacokinetics Laboratory, Millard Fillmore Hospital and State University of New York at Buffalo, Buffalo, New York (Dr. Forrest); and the Pharmacokinetics/Drug Metabolism Section, Parke- Davis Pharmaceutical Research, Ann Arbor, Michigan (Dr. Whitfield).
Funded in part by the U.S. Government, and a grant from Parke-Davis Pharmaceutical Research, Morris Plains, New Jersey.
Manuscript received October 14, 1996. Accepted pending revisions December 12, 1996. Accepted for publication in final form January 7, 1997.
Address reprint requests to William D. Figg, Pharm.D., Clinical Pharmacology Branch, Building 10, Room 5A01, 9000 Rockville Pike, National Cancer Institute, Bethesda, MD 20892.
progressive disease after their last therapeutic maneuver, and that treatment must have been completed as least 1 month before enrolling in this protocol. Each patient met the following eligibility criteria: age greater than 18 years, a histologic diagnosis of cancer confirmed by the National Cancer Institute’s (NCI’s) Laboratory of Pathology in the Clinical Center of the National Institutes of Health, life expectancy longer than 3 months, Karnofsky performance status 80% or better, and ability to make trips from home to the NCI for treatment and follow-up.
All patients signed written informed consent before enrolling, and the protocol was reviewed and approved by the institutional review board of the NCI. The conduct of these trials was monitored by the Cancer Treatment Evaluation Program, and no other forms of antitumor therapy were allowed during the study period, including radiation therapy.
Patients were excluded if they had a hemoglobin concentration less than 9 g/dl, white blood cell count less than 3.0 x 103/ul, platelet count less than 150 x103/ul, history of bleeding diathesis or coagulation disorder, or conditions requiring anticoagulation. Patients were also excluded if they had clinical or radiologic evidence of cerebral metastases, or history or clinical evidence of stroke. Those with a creatinine clearance less than 60 ml/minute were not enrolled. Abnormalities of liver function tests that could not be attributed to metastatic disease (e.g., elevated alkaline phosphatase reflecting metastatic bone disease) excluded patients, as did replacement of 50% or more of liver parenchyma with metastatic disease. Any patient who received chemotherapy, radiotherapy, or treatment with biologic response modifiers could not enroll until a minimum of 28 days had elapsed and the patient met the criteria for progressive disease.
Treatment Regimen
Suramin (Mobay Chemical Corporation, FBA Pharmaceuticals, New York, NY) was supplied in 10-ml vials containing 1 g suramin sodium USP as a sterile freeze-dried powder. Vials were reconstituted in 10 ml of sterile water. When the drug was to be delivered by continuous intravenous infusion, it was diluted to a final volume of 90 ml with 5% dextrose and placed into a portable infusion pump, and the reservoir was replaced every 24 hours. Doses were guided by adaptive control with feedback strategy using
a three-compartment, open linear model to maintain suramin plasma concentrations of 175 ug/ml. Adaptive control is a process that achieves an optimum drug regimen by individualizing therapy.17 The pharmacokinetic model uses plasma concentrations, dosing history, and patient-specific covariates to adjust the dosage throughout treatment.
Suramin was administered either by continuous or intermittent infusion, depending on the specific protocol. For intermittent infusions, the dose was diluted in 5% dextrose to a total volume of 150 ml and administered over 1 hour. The first five doses, fixed based on patient weight, were 16.1, 11.4, 9.3, 8.2, and 7.5 mg/kg, respectively. After characterization of a patient’s pharmacokinetics with the initial five doses, dosing was pharmacokinetically guided using adaptive control with feedback. Peak suramin plasma concentrations were targeted for 300 ug/ml and trough plasma concentrations for 175 ug/ml. The drug was administered only once a day and only on Monday through Friday. Patients were scheduled to receive 8 weeks of therapy.
Furosemide was not given based on protocol design, but was begun at the discretion of the physician. It was prescribed primarily for edema or evidence of congestive heart failure. It was given either intravenously or orally and for various durations (see below for details).
Suramin Assay
Plasma suramin concentrations were determined by a method described elsewhere.18 Fifty microliters of 2-naphthol [1.45 mmol/L in 75% acetonitrile/25% water vol/vol with 50 mmol/L tetrabutylammonium dihydrogen phosphate (TBAP)] and 1 ml of an organic extraction solution (75% acetonitrile/25% water vol/vol with 50 mmol/L TBAP) were added to 50 ul of the patient’s plasma. The supernatant was removed after protein precipitation and centrifugation. Fifty microliters was injected into a Waters C18 Nova-Pak cartridge 0.8 x 10.0 cm (Waters Chromatography, Milford, MA) and eluted at 1 ml/minute with a gradient mobile phase containing acetonitrile, water, TBAP 50 mmol/L, and ammonium acetate 10 mmol (pH 6.55). Two-naphthol and suramin had eluting times of 3.1 and 8.3 minutes, respectively. Both 2-naphthol and suramin were detected from ultraviolet absorbance at 238 nm. Intraassay and interassay coefficients of variation were less than
1% and 8%, respectively. The assay has a lower limit of detection of 5 µg/ml.
Pharmacokinetic Analysis
Suramin pharmacokinetics in the presence and absence of furosemide were determined using a mixed-effects population pharmacokinetic model consisting of a vector of mean population parameter values, a covariance matrix, and a model for the residual variance of the serum concentrations. The iterative two-stage (IT2S) approach to population analysis was used to obtain suramin population estimates.19 It was implemented with subroutines from the ADAPT II program.20
The IT2S approach initially involves individual fitting of each patient using M.A.P .- Bayesian estimation with initial estimates of parameter means, the population covariance matrix, and the error model obtained from the literature of by other methods such as a standard two-stage analysis. Individual pharmacokinetic parameter point estimates and asymptotic covariance matrices are obtained and used to refine the population model. This updated model is then used as priors and each data set is fit again and the population model further refined. The process is repeated until the population model reaches convergence using a maximum likelihood objective function.
The empiric variance model assumed that residual standard deviations of the observations (o) were linearly related to true values (Y): o = V1 . Y + V2, in which v1 and v2 are variance parameters. The intercept, v2, was mainly a measure of sensitivity, and the slope, v1, was closely related to the observation coefficient of variation. Initial estimates of the variance were based on assay performance. Later in the process, they were determined from the data.
A three-compartment model was selected for suramin based on previous analysis of over 300 patients studied at the NCI and Akaike’s information criterion.16, 21 Each of the six pharmacokinetic parameters used to describe the model (volumes of the central and two peripheral compartments, two distributional parameters, total clearance) could be independently displaced in either direction by the presence of furosemide. Two rate inputs were included in the model. The first, R(1), represented drug administration and the second, R(2), coded for the presence or absence of furosemide.
The majority of patients received furosemide
once/day and were coded as continuously in the presence of furosemide during this regimen. For patients receiving only a single dose or intermittent doses, the effect of furosemide was assumed to last 12 hours; thus, blood to measure concentrations that was drawn within 12 hours after a dose was coded as in the presence of furosemide. The model included two output equations, one each for the presence or absence of furosemide to ascertain differences in precision or bias in outputs.
The natural logarithm of each mean ratio (parameter with furosemide:parameter off furosemide) was calculated. Student’s t test was performed to determine if this value was statistically different from zero. To determine equivalence in parameters in the presence and absence of furosemide, two one-sided t tests were performed to calculate the probability that the true mean is between 80% and 125%.22 Bioequivalence was defined by the Food and Drug Administration as the probability that the true mean is within the range of 0.9 or higher.23
Finally, various regimens were simulated using ADAPT II for patients treated with 12 weeks of suramin with and without furosemide to determine the worst case scenario. Concentrations were compared between the furosemide and no- furosemide regimens.
Results
Patient Demographics
Twenty-six patients were evaluated (Table 1). They represent the total number of patients treated at the NCI who underwent concurrent furosemide and suramin therapy. The distribution of diseases among the study population was metastatic hormone-refractory prostate cancer in 25 and adrenocortical carcinoma in 1.
Data are reported as mean + SD. Eighteen patients received suramin by continuous infusion, and the remaining eight received it by intermittent infusion over 60 minutes (intermittent infusions did not exceed one dose/day). They all were treated using adaptive control with feedback to maintain a targeted plasma concentration for 8 weeks. The Bayesian estimation method employed to dose suramin prospectively used the Abbottbase pharmacokinetic system as described previously.16 They received 24.7 ± 9.8 doses for a total of 11,858 ± 2805 mg.
A mean (± SD) of 25.2 + 4.7 suramin plasma concentrations were obtained in patients
| Variable | Value |
|---|---|
| Age (yrs) | 64.5 ± 9.7 |
| Average body weight (kg) | 87.0 ± 13.3 |
| Ideal body weight (kg) | 66.4 ± 5.9 |
| Number of suramin doses | 24.7 ± 9.8 |
| Off Furosemide | % Change Due to Furosemide | |
|---|---|---|
| Vc (L/kg) | 0.059 (31.3) | -5.8 (8.6) |
| Vpl (L/kg) | 0.079 (50.0) | 1.6 (37.6) |
| Vp2 (L/kg) | 0.285 (43.7) | 8.1 (40.2) |
| Cldl (L/hr/kg) | 0.00846 (82.3) | 15.0 (57.0) |
| Cld2 (L/hr/kg) | 0.00185 (29.6) | 1.4 (59.1) |
| Clt (L/hr/kg) | 0.00047 | -36.2 (34.6) |
Vc = volume of distribution for central compartment; Vp1 = volume of distribution for first peripheral compartment; Vp2 = volume of distribution for second peripheral compartment; Cld1 = distributional clearance between central and first peripheral compartments; Cld2 = distributional clearance between central and second peripheral compartments; Clt = total body clearance. Data are reported as mean (CV%).
receiving continuous infusion and 24.6 + 7.4 in those receiving intermittent infusion (11.0 ± 3.5 peak, 13.6 + 4.3 trough concentrations). Of the total number of suramin plasma concentrations, 238 were taken when patients were receiving furosemide plus suramin and 397 when they were receiving only suramin.
The plasma concentration in the continuous infusion group was 179.5 ± 26.5 µg/ml. Peak and trough concentrations in the intermittent infusion group were 276.7 + 29.7 ug/ml and 159.8 + 16.8 ug/ml, respectively. Patients received their first dose of furosemide 25.5 ± 18.5 days into suramin therapy. Forty-five different furosemide dosing regimens were administered to the 26 patients; 32 regimens were given orally and 13 intravenously. Among the 45 regimens, 3 were 10 mg/dose, 24 were 20 mg/dose, 15 were 40 mg/dose, 1 was 60 mg/dose, and 2 were 80 mg/dose. The typical furosemide regimen lasted 20.0 ± 26.4 days (range 1-112 days).
Pharmacokinetic Results
The fitted mean pharmacokinetic parameters and percentage change in each parameter due to furosemide are shown in Table 2.
The pharmacokinetic model described the data well. The excellent precision and lack of bias both in the presence and absence of furosemide was shown in the plots of observed and fitted
concentrations of suramin (Figures 1 and 2). Both plots demonstrated similar estimates of the coefficient of variation and sensitivity allowing pooling of data for the final residual variance model. The final line of best fit did not differ from the line of identity: Y = 0.998 . X + 0.00006. The fitted parameter values for the residual variance model for suramin were 0.086 for v1 and 4.61 mg/L for v2.
A significant decrease in total body clearance of suramin was observed during concomitant furosemide therapy. The ratio of total body clearance on furosemide:off furosemide was 0.64, representing a 36% decrease during furosemide therapy. The probability that the true mean ratio is in the desired range for equivalency was less than the recommended 0.9. Individually, only 7 of 26 patients had log ratios for clearance within the guidelines. No other statistically significant changes in suramin pharmacokinetics were noted. The probability that the other ratios were in the desired range of 80-125% was greater than 0.90.
Simulation results comparing suramin concentrations in the presence and absence of continuous furosemide as a function of time receiving furosemide are shown in Table 3. In the worst case scenario of concomitant furosemide for 12 weeks, the simulation predicted that 46% of patients would have a greater than 33% increase in average suramin concentration. Two patients had increases of greater than 67% in their average suramin concentrations; for furosemide therapy of shorter durations, the majority of patients had increases of less than 33%.
Precision of Model Suramin concentrations with concomitant furosemide
500
observed concentration (ug/ml)
B
6
400
a
Đ
®
300
2
0
B
B
U
n = 239
200
-
r2 = 0.99
€
100
Đ
8
0
0
100
200
300
400
500
fitted concentration (ug/ml)
| % Increase | Week 1 | Week 3 | Week 6 | Week 12 |
|---|---|---|---|---|
| < 33 | 21 | 21 | 17 | 14 |
| 33-67 | 4 | 4 | 8 | 10 |
| > 67 | 1 | 1 | 1 | 2 |
Discussion
Considerable debate still surrounds the administration of suramin from standpoints of clinical efficacy1, 6-9, 14, 24-26 and the best method of administration. 16, 27-32 Much attention has focused on adaptive control with feedback to control plasma concentrations to within an estimated therapeutic range. However, several groups showed that this labor-intensive method may not be necessary in most patients.28, 30 This would hold true if individual patients do not have altered pharmacokinetics because of concurrent drug-drug interactions. Thus, the desire to characterize potential interactions fully was the basis on which we undertook this study.
Since preliminary data suggested that the systemic clearance of suramin did not correlate with creatinine clearance,16 we anticipated not finding an interaction with furosemide. Furthermore, preliminary data from our group supported the hypothesis that suramin was mainly eliminated by tubular secretion (three patients treated with and without probenecid had altered plasma concentrations), further
Precision of Model Suramin concentrations with no furosemide
500
observed concentration (ug/ml)
400
8
B
6
&
G
D
-
300
G
9
B
200
Q
a
n = 393
0
r2 = 0.99
0
100
6
U
0
0
100
200
300
400
500
fitted concentration (ug/ml)
solidifying the theory that an interaction would not be anticipated between a diuretic and suramin. However, our results indicate that concomitant furosemide therapy resulted in a 36% decrease in suramin’s systemic clearance. This suggests that suramin is primarily eliminated by secretion in the proximal tubules and undergoes competitive secretion in the presence of furosemide.
The mechanism of furosemide’s diuretic activity is inhibition of active chloride transport throughout the thicker portion of the ascending loop of Henle; also it prevents the reabsorption of sodium that passively follows chloride.33-35 Because the drug is a weak organic acid (pKa 3.9)36 and is highly bound to serum protein (> 95%),37, 38 it gains access to its site of action by secretion through the nonselective organic acid transport system at the proximal tubules. Since probenecid inhibits this system, it is not surprising that it reduces delivery of furosemide into the luminal side of the renal tubules, thereby diminishing natriuretic action.39-43 Similar to this interaction, furosemide appears to inhibit the secretion of suramin in the proximal tubules. Like furosemide, suramin is a weak organic acid that is highly protein bound and thus eliminated by secretion.
In our current trial, however, we did not evaluate the pharmacodynamic effect suramin may have on furosemide. If this interaction is by competitive inhibition in the proximal tubules, one would anticipate reduced natriuretic activity. Furthermore, numerous reports of drug-drug interactions with furosemide have been published, although most of them involved increased glomerular filtration. However, lomefloxacin clearance appears to be decreased by furosemide by competition for tubular secretion (both compounds are secreted by the renal organic anion transport system). Likewise, the clearance of cisplatin may be altered in the presences of furosemide,44 but furosemide does not affect the pharmacokinetics of mitomycin C.45
An alternative explanation for our findings is the possibility of time-dependent suramin pharmacokinetics. Others suggested that suramin clearance decreases with time.46 Using a three-compartment model, clearance decreased 30% between 8-week treatment cycles separated by an 8-week period of no drug administration. To investigate a time effect, we first examined the weighted residuals versus time in both the presence and absence of furosemide. We then fixed the coefficients of the model to 1.0 and brought the data to convergence to simulate a
picture of no furosemide. Plots of the weighted residuals versus time for this analysis were also generated. In all of these plots, there were no apparent differences or trends in residuals with time. Furthermore, we fitted the data using a model that allowed clearance to change with time. The goodness of fit for this model did not improve compared with the original model.
The potential clinical implications of concomitant suramin and furosemide are unknown. Due to the long half-life associated with suramin, it is reasonable to assume that short-term or intermittent furosemide therapy most likely would have little or minimal effect on the plasma concentrations of fixed-dose suramin therapy. We did not observe suramin toxicity since the patients had suramin dosages adjusted to maintain concentrations of 175 mg/ml. However, other centers do not use adaptive control and this interaction may be clinically significant in the absence of dosage adjustment. To address this issue further, we simulated suramin concentrations in 26 patients with no furosemide and also with continuous furosemide for 12 weeks. During continuous coadministration, the average suramin concentration increased more than 33% in 46% of patients. In two patients the increases were large enough to be associated with clinical adverse events (i.e., increased risk of developing grade III neuropathies).47
Concomitant furosemide is associated with a significant decrease in the clearance of suramin. In the setting of nonadaptive control administration where a goal is to maintain plasma concentrations at the upper limit of a therapeutic range, a significant decrease in clearance has the potential for increased toxicity,27, 29 for which appropriate precautions should be taken.
Acknowledgments
We thank Matthew Middleman, Bruce Waldrop, and Jennifer Stevens for assistance with data entry. We acknowledge the analytical assistance of Natalie McCall and Page Gernt. The clinical trials were conducted under the supervision of Eddie Reed, M.D., Charles Myers, M.D., Michael Cooper, M.D., Oliver Sartor, M.D., and Nancy Dawson, M.D.
References
1. Myers CE, Cooper M, Stein C, et al. Suramin: a novel growth factor antagonist with activity in hormone-refractory metastatic prostate cancer. J Clin Oncol 1992;10:881-9.
2. Broder S, Yarchoan R, Collins JM, et al. Effects of suramin on HTLV-III/LAV infection presenting as Kaposi’s sarcoma or AIDS-related complex: clinical pharmacology and suppression
of virus replication in vivo. Lancet 1985;2:627-30.
3. Cheson BD, Levine AM, Mildvan D, et al. Suramin therapy in AIDS and related disorders. JAMA 1987;258:1347-51.
4. Levine AM, Gill PS, Cohen J, et al. Suramin antiviral therapy in the acquired immunodeficiency syndrome. Ann Intern Med 1986;105:32-7.
5. LaRocca RV, Stein CA, Danesi R, Jamis-Dow CA, Weiss GH, Myers CE. Suramin in adrenal cancer: modulation of steroid hormone production, cytotoxicity in vitro, and clinical antitumor effect. J Clin Endocrinol Metab 1990;71:497-504.
6. Dawson N, Cooper MR, Figg WD, et al. Activity of suramin in hormone refractory prostate cancer is diminished when hydrocortisone and flutamide withdrawal are removed as potentially confounding variables. Cancer 1995;76:453-62.
7. Eisenberger MA, Fontana JA. Suramin, an active nonhormonal cytotoxic drug for treatment of prostate cancer: compelling reasons for testing in patients with hormone-refractory breast cancer. J Natl Cancer Inst 1992;84:3-5.
8. Eisenberger MA, Reyno LM, Jodrell DI. Suramin, an active drug for prostate cancer: interim observations in a phase I trial. J Clin Oncol 1993;85:611-21.
9. Bowden CJ, Figg WD, Dawson NA, et al. A phase I/II study of continuous infusion suramin in patients with hormone- refractory prostate cancer: toxicity and response. Cancer Chemother Pharmacol 1995;39:1-8.
10. LaRocca RV, Cooper MR, Uhrich M, et al. Use of suramin in treatment of prostatic carcinoma refractory to conventional hormonal manipulation. Urol Clin North Am 1991;18:123-9.
11. Horne MK III, Stein CA, LaRocca RV, Myers CE. Circulating glycosaminoglycan anticoagulants associated with suramin treatment. Blood 1988;71:273-9.
12. Hosang M. Suramin binds to platelet-derived growth factor and inhibits its biological activity. J Cell Biochem 1985;29:265-73.
13. LaRocca RV, Danesi R, Cooper MR, et al. Effect of suramin on human prostate cancer cells in vitro. J Urol 1991;145:393-8.
14. Stein CA, LaRocca RV, Thomas R, McAtee N, Myers CE. Suramin: an anticancer drug with a unique mechanism of action. J Clin Oncol 1989;7:499-508.
15. Collins JM, Klecker RW, Yarchoan R, et al. Clinical pharmacokinetics of suramin in patients with HTLV-III/LAV infection. J Clin Pharmacol 1986;26:22-6.
16. Cooper MR, Lieberman R, LaRocca RV, et al. Adaptive control with feedback strategies for suramin dosing. Clin Pharmacol Ther 1992;52:11-23.
17. Jelliffe RW, Schumitzky A, Van Guilder M, et al. Individualizing drug dosage regimens: roles of population pharmacokinetic and dynamic models, Bayesian fitting, and adaptive control. Ther Drug Monit 1993;15:380-93.
18. Supko JG, Malspeis L. A rapid isocratic HPLC assay of suramin in human plasma. J Liquid Chromatogr 1993;13:727-41.
19. Steimer JL, Mallet J, Golmard J, Boisieux J. Alternative approaches to estimation of population pharmacokinetic parameters: comparison with the nonlinear mixed effect model. Drug Metab Rev 1984;15:265-97.
20. D’Argenio DZ, Schumitzky A. ADAPT II user’s guide. Los Angeles: Biomedical Simulations Resources, University of Southern California, 1992.
21. Yamaoka K, Nakagawa J, Uno T. Application of Akaike’s information criterion (AIC) in the evaluation of linear pharmacokinetic equations. J Pharmacokinet Biopharm 1978,2:165-75.
22. Schuirmann DJ. A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability. J Pharmacokinet Biopharm 1987;15:657-80.
23. Food and Drug Administration. Statistical procedures for bioequivalence studies using a standard two treatment crossover design. Rockville, MD: Office of Generic Drugs, Division of Bioequivalence, 1992.
24. LaRocca RV, Stein CA, Danesi R, Cooper MR, Uhrich M, Myers CE. A pilot study of suramin in the treatment of metastatic renal cell carcinoma. Cancer 1991;67:1509-13.
25. LaRocca RV, Cooper MR, Stein CA, et al. A pilot study of suramin in the treatment of progressive refractory follicular lymphomas. Ann Oncol 1992;3:571-3.
26. Clark JW, Chabner BA. Suramin and prostate cancer: where do we go from here? J Clin Oncol 1995;13:2155-7.
27. Bitton R, Figg WD, Venzon DJ, et al. Pharmacologic variables associated with the development of neurologic toxicity in patients treated with suramin. J Clin Oncol 1995;13:2223-9.
28. Scher HI, Jodrell DI, Iversen JM, et al. Use of adaptive control with feedback to individualize suramin dosing. Cancer Res 1992;52:64-70.
29. Jodrell DI, Reyno LM, Sridhara R, et al. Suramin: development of a population pharmacokinetic model and its use with intermittent short infusions to control plasma drug concentration in patients with prostate cancer. J Clin Oncol 1994;12:166-75.
30. Figg WD, Stevens JA, Cooper MR. Adaptive control with feedback of suramin using intermittent infusions [letter]. J Clin Oncol 1994;12:1522-5.
31. Kobayashi K, Vokes EE, Vogelzang NJ, Janish L, Soliven B, Ratain MJ. Phase I study of suramin given by intermittent infusion without adaptive control in patients with advanced cancer. J Clin Oncol 1995;13:2196-2207.
32. Kobayashi K, Vokes EE, Stefansson K, et al. Suramin: is adaptive control necessary? [letter]. J Clin Oncol 1992;10:1984-5.
33. Burg MB. Tubular chloride transport and the mode of action of some diuretics. Kidney Int 1976;9:189-97.
34. Jacobson HR, Kikko JP. Diuretics: site and mechanism of action. Annu Rev Pharmacol Toxicol 1976;16:201-14.
35. Seely JF, Dirks JH. Site of action of diuretic drugs. Kidney Int 1977;11:1-8.
36. Kirkendall WM, Stein JH. Clinical pharmacology of furosemide and ethacrynic acid. Am J Cardiol 1968;22:162-7.
37. Bowman RH. Renal secretion of [25S] furosemide and its depression by albumin binding. Am J Physiol 1975;229:93-8.
38. Rane A, Villeneuve JP, Stone WJ, Nies AS, Wilkinson GR, Branch RA. Plasma binding and disposition of furosemide in the nephrotic syndrome and in uremia. Clin Pharmacol Ther 1978;24:199-207.
39. Hsieh YY, Hsieh BS, Lien WP, Wu TL. Probenecid interferes with the natriuretic action of furosemide. J Cardiovasc Pharmacol 1987;10:530-4.
40. Smith DE, Gee WL, Brater DC, Lin ET, Benet LZ. Preliminary evaluation of furosemide-probenecid interactions in humans. J Pharmaceut Sci 1980;69:571-5.
41. Chennavasin P, Seiwell R, Brater DC, Liang WMM. Pharmacodynamic analysis of the furosemide-probenecid interaction in man. Kidney Int 1979;16:187-95.
42. Brater DC. Effects of probenecid on furosemide response. Clin Pharmacol Ther 1978;24:548-54.
43. Homeida M, Roberts C, Branch RA. Influence of probenecid and spironolactone on furosemide kinetics and dynamics in man. Clin Pharmacol Ther 1977;22:402-9.
44. Sudoh T, Fujimura A, Shiga T, et al. Renal clearance of lomefloxacin is decreased by furosemide. Eur J Clin Pharmacol 1994;46:267-9.
45. Verweij J, Fronius SK, Stuurman M, deVries J, Pinedo HM. Absence of interaction between furosemide and mitomycin C. Cancer Chemother Pharmacol 1987;19:84-6.
46. Landau EM, Belldegrun AS, McBride JH, deKernion JB, Rosen PJ. Significant intrapatient and interpatient variability of suramin pharmacokinetics in prostate cancer. Proc Am Assoc Cancer Res 1994;35:242.
47. LaRocca RV, Meer J, Gilliatt DM, et al. Suramin-induced polyneuropathy. Neurology 1990;40:954-60.