Evaluation of Carboplatin Pharmacokinetics in the Absence and Presence of Paclitaxel’

Vol. 2, 549-552, March 1996 Clinical Cancer Research 549

Coleman K. Obasaju, Steven W. Johnson, INTRODUCTION Andre Rogatko, Debbie Kilpatrick, Paclitaxel (Taxol) has demonstrated substantial activity as James M. Brennan, Thomas C. Hamilton, a single agent in the treatment of various solid tumors, including ovarian, breast, and lung cancers (1-6). Preclinical studies in- Robert F. Ozols, Peter J. O’Dwyer, dicated that the combination of Taxol and platinum compounds and James M. Gallo2 was superadditive in several models (1, 7, 8). The cellular basis Fox Chase Cancer Center, Philadelphia, Pennsylvania 191 1 1 for this favorable interaction has not yet been elucidated, but aspects of appropriate dosing strategies have been described in preclinical and early clinical studies. In MCF-7 breast cancer ABSTRACT and A549 lung cancer xenografts, the activity of simultaneous In a clinical trial of paclitaxel (Taxol) and carboplatin drug administration and that of the sequence Taxol preceding in combination, the severity of thrombocytopenia was less carbopiatin administration was superior to that of the reverse than would be expected with an equivalent dose of carbo- order (9). In a Phase I clinical trial of cispiatin and Taxol, more platin alone. To determine whether a pharmacokinetic in- severe neutropenia was associated with the sequence in which teraction was responsible for this observation, the effect of cisplatin was given first, and this sequence was associated with pretreatment with Taxol on the pharmacokinetics of carbo- a lower clearance of Taxol (10). The substantial clinical activity platin was examined in 1 1 patients. Each patient was ran- of the Taxol/cisplatin combination has been demonstrated in a domized to one of two treatment groups that determined the randomized trial in ovarian cancer ( 1 1 ), and studies in other order of drug treatments. The treatments were carboplatin diseases are in progress. as a 30-mm infusion alone or immediately following 175 Early clinical combination studies of Taxol with carbopla- mg/m2 Taxol administered as a 3-h i.v. infusion. The treat- tin lagged behind those with cisplatin because of concern re- ments were separated by 1 week. The carboplatin dose was garding additive myeiosuppression. In a Phase I trial in ovarian chosen to produce a target area under the concentration- cancer, in which the carboplatin dose was individualized ac- time curve (AUC) of 3.75 mg-mm/mi according to a previ- cording to the formula of Caivert et a!. (12), we found that full ously published formula (A. H. Calvert et a!., J. Clin Oncol., doses of each agent could be administered safely.3 Indeed, the 7: 1748-1756, 1989). The mean administered dose of carbo- severity of thrombocytopenia appeared to be even less than platin was 338 mg. Serial blood samples were collected over expected.3 The kinetics of both agents was characterized in that 24 h and analyzed for total and free platinum, and, in some study; although an initial impression suggested that carbopiatin patients, Taxol. The pharmacokinetics of carboplatin (i.e., AUC4 values were less than predicted, the analysis of the total clearance and volume of distribution at steady state), completed study demonstrated that the mean measured carbo- was not significantly affected by pretreatment with Taxol. platin AUC was 7.2 mg-mm/mI at a targeted AUC of 7.5 Total clearances of carbopiatin were 67.2 ± 28.8 ml/min and mg-min/ml.3 To determine more precisely whether the apparent 64.6 ± 27.9 mI/mm in the absence and presence of Taxol, alteration in platelet toxicity could be explained by a pharma- respectively (P 0.56). The AUC of free carboplatin (3.45 cokinetic interaction between Taxol and carboplatin, we de- mg-min/ml) obtained in the absence of Taxol was not signif- signed a study in which patients were treated both with carbo- icantly different from that measured in the presence of platin alone and with carboplatin following a 3-h infusion of Taxol (3.27 mg-min/ml). The AUC of carboplatin in both the Taxol. absence and presence of Taxol agreed with the projected target AUC of 3.75 mg-mm/mi. In conclusion, the applica- MATERIALS AND METHODS tion of an individualized dosing strategy is valid for the calculation of the carboplatin dose in this combination. The Study Design pharmacokinetics of carboplatin is not altered by pretreat- This trial was designed to test the effect of Taxol on ment with Taxol at a standard dose, and a pharmacokinetic carboplatin pharmacokinetics. Each patient received both car- interaction is not responsible for the altered toxicity of the combination.

3 M. A. Bookman, W. P. McGuire III, D. Kilpatrick, E. Keenan, W. M. Hogan, S. W. Johnson, P. J. O’Dwyer, E. Rowinsky, H. H. Gallion, and Received 8/1/95; revised 1 1/6/95; accepted 1 1/15/95. R. F. Ozols. Carboplatin and paxlitaxel in ovarian and peritoneal car- I Supported in part by NIH Grant CA-06972 and an appropriation from cinoma: a phase I study of the Gynecologic Oncology Group, submitted the Commonwealth of Pennsylvania. for publication. 2To whom requests for reprints should be addressed, at Fox Chase 4The abbreviations used are: AUC, area under the concentration-time Cancer Center, 7701 Burholme Avenue, Philadelphia, PA 19111. curve; GFR, glomerular filtration rate; AUC24, partial area under the Phone: (215) 728-2461; Fax: (215) 728-2741. curve to 24 h; volume of distribution of steady state.

boplatin alone and Taxol in combination with carboplatin. To Taxol Sampling minimize patient loss from the trial, the carboplatin dose was Two 8-ml blood samples were collected at the end of the split, and half administered in successive weeks. Patients re- Taxol infusion and at I 0 h from the start of infusion. Serum was

ceived the Taxol before the first or second carboplatin dose by obtained from the blood samples and stored at - 80#{176}Cuntil random assignment. The end point of the study was the corn- analyzed for Taxol by a high-pressure liquid chromatography parison of the carbopiatin AUC and other pharmacokinetic assay. parameters with or without Taxol. Measurements of Carboplatin and Taxol Eligibility Criteria Total platinum in serum and ultrafiltrate platinum in serum The eligibility criteria included patients with histologically were measured with a flameless atomic absorption spectropho- confirmed malignant tumor for which treatment with Taxol and tometry. Samples were diluted in duplicate with 0.2% nitric acid carboplatin was appropriate therapy. Patients were required to containing 0.1% (v/v) Triton X-100 and injected (20 p.1) onto a have an Eastern Cooperative Oncology Group performance sta- Perkin Elmer/Cetus model 3 100 atomic absorption spectrometer tus 0-2, adequate bone marrow function (neutrophils 2000/ equipped with an HGA 600 graphite furnace. Platinum concen-

mm3, platelet count I 00,000/mm3), renal function (creatinine trations were determined relative to a freshly prepared standard 2 mg/dl), and hepatic function (bilirubin 2 mg/dl). Pregnant curve for elemental platinum (0.2-4.0 ng). The limit of detec- or lactating women were excluded from the trial. Patients with tion was 0.01 p.g/ml (14). child-bearing potential had to implement adequate contraceptive A previously published high-pressure liquid chromatogra- measures during participation in this study. They had to be free phy method was used to quantitate Taxol in serum (15). In brief, of clinically significant infection and must have signed an a solid-phase extraction procedure was used to extract Taxol approved informed consent. from serum. Cephalomannine (final concentration 10 JaM) was used as an internal standard. Taxol was eluted from a C18 Treatment Plan Bond-Elut extraction column with acetonitrile. The acetonitrile Patients were randomized to one of two treatment groups. was evaporated, and the dried residue was reconstituted in 150 Group A. Patients received carbopiatin as a 30-mm i.v. p.1 mobile phase. The mobile phase consisted of 45% (v:v) 8 infusion at a dose calculated to produce a target AUC of 3.75 acetonitrile: water, pumped at 1 mi/mm through a Hypersil (100 mm X 4.6 mm) column. Taxol and internal standard were mg-mm/mi on day 1 . The dose was determined based on the detected at 230 nm, with a limit of quantitation of 15 ng/ml. formula of Calvert et al. (1 2): Dose (rng) = AUC (GFR + 25). This was followed on day 8 by the administration of 175 mg/rn2 Pharmacokinetic Analysis Taxol as a 3-h iv. infusion, followed immediately by carbopla- Pharmacokinetic analysis of carboplatin was conducted by tin at the same dose as on day 1 . GFR was estimated as (13): noncornpartmental analysis using the LAGRAN computer program (16). Cl1, and the terminal elimination half-life (t112) . (140 - age) weight GFR (ml/min) = . (22) in the absence and presence of Taxol were obtained using 72 (serum creatinine) standard formulas and examined for significant differences with paired t tests. A AUC24 was also obtained from the noncom- for males, and for females GFR is 0.85 times this formula. In partmental analysis and used for comparisons to the target AUC this formula, age is in years, weight in kgs, and serum creatinine of 3.75 mg-mm/mi. in mg/dl.

Group B. Patients received the two treatments in reverse RESULTS order. Patient characteristics are shown in Table 1. A total of 13 All patients were premedicated with dexarnethasone (20 patients were accrued to the study; however, only 1 1 patients mg i.v or p.o.) 12 h, 6 h, and 1 h prior to Taxol. Diphenhydra- completed the pharmacokinetic studies. mine (50 mg iv.) and cimetidine (300 mg iv.) were also given Fig. 1 illustrates a typical free carboplatin serum concen- 1 h prior to Taxol. tration-time profile in the absence and presence of Taxol. Minor differences are noted throughout the time course. Pharmacoki- Carboplatin Sampling netic parameters of carboplatin for all 1 1 patients are summa- Blood samples (8 ml) were obtained at the following times rized in Table 2. There were no significant differences in any of in relation to carboplatin administration: at 0, 15, 30 (end of the measured parameters. These values agreed with those pre- infusion), 45, 60, and 75 mm and 1.5, 2, 2.5, 4.5, 6.5, 12.5, 18.5, viously reported (17). Variability in the pharmacokinetic param- and 24.5 h. Serum was harvested from blood samples with one eters was appreciable both in the absence and presence of Taxol, half of the serum centrifuged through an ultrafree filter (Mr and further indicated by the large 95% confidence intervals cutoff 30,000; Millipore Corporation, Bedford, MA) to obtain a (Table 3). The measured AUC24 values (3.22 ± 0.762 and 3.40 serum ultrafiltrate. The remaining serum and serum ultrafiltrate ± 0.89) were slightly less than the target AUC of 3.75 rng- were stored at -80#{176}Cuntil analyzed for total platinum and free mm/mi used to determine the carboplatin dosage, and were not

platinum, respectively, by atomic absorption spectrophotome- statistically different from the target AUC (P = 0.064 and P = try. 0.27, respectively).

Table 1 Patient characteristics Table 2 Pharmacokinetic parameters of free carboplatin obtained in the absence and presence of 175 mg/m2 Taxol administered as a 3-h Patients entered/evaluable 13/11 infusion Male/female 5/8 Median age, yr (range) 65 (32-79) Carboplatin Carboplatin Paired Performance status alone and Taxol t 0 0 test 8 Mean SD Mean SD (P) 5 AUC24 (mg-min/ml) 3.40 0.89 3.22 0.76 0.44 site Primary Ci, (mI/mm) 107.5 34.5 1 12.8 36.3 0.85 Lung 6 Vd (liter) 22.0 5.9 22.1 6.7 0.99 Head and neck 3 ti,2 (h)’ 3.50 0.44 3.11 1.25 0.34 Esophagus 2 a Harmonic mean half-lives and pseudo-SD are reported and used Gastric Melanoma in the t test (22). Previous treatment data Chemotherapy 2 Table 3 Pharmacokinetic parameters for total carboplatin obtained in Radiation 2 the absence and presence of I 75 mg/m2 Taxol administered as a 3-h Both 3 infusion Neither 6 Paired Body surface area Carbopiatin Median, m2 (range) 1.67 (1.32-1.96) alone Carboplatin and Taxol t test Calculated creatinine clearance Mean SD Mean SD (P) Median, ml/min (range) 71 (33-88) Cl, (mI/mm) 67.2 28.8 64.6 27.9 0.56 (- 17.5, 14.6y’ 40#{149}- Vd. (liter) 66.1 45.5 69.0 38.5 0.78 (-25.7, 35 . l9.8r

ti,2 (h)’ 9.5 7.3 8.1 7.2 0.65 a 95% confidence interval of the difference in carbopiatin pharmacokinetic parameter. ,. I’ h Harmonic mean half-lives and pseudo-SD are reported and used -a 25- in the t test (22). C 0 20- 15 netic interactions between anticancer agents have not been 0 widely investigated. Gianni (19) has reported an inhibitory 10 effect of Taxol on doxorubicin clearance. Taxol is itself cleared U. 25% less efficiently when administered after, rather than before, 5- cisplatin (10). Our previous clinical trial suggested that the platelet toxicity of carboplatin was reduced with the concomi- 0 - -. 1 O 25 tant administration of Taxol.3 The current study was designed to minimize confounding Time (hr) variables while maintaining acceptable treatment intensity in Fig. 1 Free (ultrafiltrate) carboplatin serum concentration-time profiles these patients with advanced cancer. By administering one half in the absence (C) and presence () of Taxol (175 mg/m2 as a 3-h of the carboplatin dose in successive weeks (a sufficient wash- infusion). Data from a single patient. out interval), each patient served as his/her own control. Se- quence effects were eliminated by randomizing the order of treatments. Based on the repeated measure study design in the current investigation, and an acceptable 80% power of the t test, Similar to pharmacokinetic parameters based on free car- a patient sample size of 1 1 would detect a 25% difference in boplatin, parameters based on total carboplatin serum concen- AUC24 assuming a SD of 0.8, as observed (20). Detection of trations were comparable in the absence and presence of Taxol differences <25%, given an analagous parameter variability (Table 3). The mean Taxol plasma concentration at the end of would require a larger number of subjects, yet such differences the 3-h infusion was 6.26 ± 0.51 p.M. and at 10 h declined to may not be clinically relevant. 0.35 ± 0.97 JaM, and agreed with previously published peak The 3-h iv. infusion schedule is now the most widely used values (5.9 ± 0.9 JIM; Ref. 18). method of Taxol administration. This schedule was used in the original pilot study of Taxol/carboplatin in which we found an DISCUSSION unexpectedly low incidence of platelet toxicity.3 The end infu- Since combinations of carboplatin and Taxol are achieving sion Taxol concentrations in the current study were consistent promising clinical results in several solid tumors, we investi- with previously published data (19). Therefore, the patients and gated the potential of a pharmacokinetic drug interaction to the treatment schedule seem optimal to detect a pharmacokinetic explain an apparent alteration in clinical toxicity. Pharmacoki- interaction.

The target carboplatin AUC of 3.75 mg-mm/mi, calculated 6. Chang, A., Kim, K., Glick, J., Anderson, T., Karp., D., and Johnson, using the Calvert formula (12), was achieved when carboplatin D. Phase II study of Taxol in patients with stage IV non-small cell lung was given alone (measured AUC 3.40 ± 0.89 mg-min/rnl). The cancer: the Eastern Cooperative Oncology Group (ECOG) results. Proc. Am. Soc. Clin. Oncol., 11: 293, 1992. Calvert formula was originally derived from measurements of 7. Rowinsky, E. K., Cazenave, L. A., and Donehower, R. C. A novel 51Cr EDTA clearances to estimate the GFR, whereas in our investigational antimicrotubule agent. J. Natl. Cancer Inst., 82: 1247- investigation GFR was based on the Cockroft-Gault formula 1259, 1990. ( 13). Although ‘Cr EDTA clearances provide an accurate mea- 8. William, C. R. Taxol-based combination chemotherapy and other in surement of GFR, the serum creatinine-based approximation vivo preclinical antitumor studies. Monogr. J. Natl. Cancer Inst.. 15: served here as an adequate estimate of GFR for dose selection. 47-53, 1993. It is reported however that following carbopiatin treatment, 9. Clark. J. W., Santos-Moore, A. S., and Choy, H. Sequencing of Taxol serum creatinine may not predict GFR ( 1 2). Therefore, this and carboplatinum therapy. Proc. Am. Assoc. Cancer Res., 36: 298, conclusion may not be valid for subsequent drug cycles. The 1995. implication of this observation is for carboplatin dose modifi- 10. Rowinsky, E. K., Gilbert, M. R., McGuire, W. P., Noe, D. A., cation. A dose reduction of carboplatin in response to excessive Grochow, L. B., Forastiere, A. A., Ettinger, D. S., Lubejko, B. G., and Clark, B. Sequences of Taxol and cisplatin: a phase I and pharmacologic toxicity in the first cycle should probably be based on a fixed study. J. Clin. Oncol., 9: 1692-1703, 1991. reduction of the first cycle dose, rather than a reapplication of I 1. McGuire, W. P., Hoskins, W. J., Brady, M. F., Kucera, P. R., Look, the Caivert formula using serum creatinine estimates of GFR. K. Y.. Partridge, E. E., and Davidson, M. A phase II trial comparing In the 1 1 patients studied here, the AUC of carboplatin was cisplatin/cytoxan (PC) and cispiatin/Taxol (PT) in advanced ovarian similar whether or not the dose had been preceded by 175 cancer (AOC). Proc. Am. Soc. Clin. Oncol., 12: 255, 1993. mg/rn2 Taxol (3.22 ± 0.76 versus 3.40 ± 0.89 mg-mm/mI). 12. Calvert, A. H., Newell, D. R., Gumbrell, L. A., O’Reilly, S., Additional pharmacokinetic parameters, including Ci,, and Burnell, M., Boxall, F. E., Siddik, Z. H., Judson, I. R., Gore, M. E., and terminal half-life, were similarly unaffected by Taxol pretreat- Wiltshaw, E. Carboplatin dosage: prospective evaluation of a simple ment. We may reasonably conclude therefore that Taxol has no formula based on renal function. J. Clin. Oncol., 7: 1748-1756, 1989. influence on the disposition of carboplatin at this dose level. 13. Cockcroft, D. W., and Gault, M. H. Prediction of creatinine clear- Indeed, a clearance interaction would be unlikely since Taxol is ance from serum creatinine. Nephron, 16: 31-41, 1976. primarily eliminated by metabolism, whereas carboplatin’s ma- 14. Schilder, R. J., LaCreta, F. P., Perez, R. P., Johnson, S. W., Brennan, J. M., Rogatko, A., Nash, S., McAleer, C., Hamilton, T. C., jor route of elimination is renal excretion (17, 21). Unlike Roby, D., Young, R. C., Ozols, R. F., and O’Dwyer, P. J. Phase I and cisplatin, plasma protein binding of carboplatin is low, and its pharmacokinetic study of ormaplatin administered on a day I and day 8 displacement from protein binding sites by another drug would schedule. Cancer Res., 54: 709-717, 1994. not significantly alter unbound drug concentrations (17). The 15. Jamis-dow, C. A., Klecker, R. W., Sarosy, 0., Reed, E., and Collins, reported interaction of cisplatin and Taxol (10) may perhaps J. M. Steady-state plasma concentrations and effects of Taxol for a 250 result from such differences. However, for carboplatin/Taxol, it mg/m2 dose in combination with granulocyte-colony stimulating factor is clear that an apparent clinical interaction does not have a in patients with ovarian cancer. Cancer Chemother. Pharmacol., 33: pharmacokinetic basis. 48-52, 1993. 16. Rocci, M. L., and Jusko, W. J. LAGRAN program for area and REFERENCES moments in pharmacokinetic analysis. Comput. Programs Biomed., 16: 203-216, 1983. I . Rowinsky, E. K., and Donehower, R. C. Paclitaxel (Taxol). N. Engl. J. Med., 332: 1004-1012, 1995. 17. Wagstaft. A. J., Ward, A., Benfield, P., and Heel, R. C. Carboplatin. 2. McGuire, W. P., Rowinsky, E. K., Rosenshein, N. B., Grumbine, Drugs, 37.- 162-190, 1989. F. C., Ettinger, D. S., Armstrong, D. K., and Donehower, R. C. Taxol, 18. Rowinsky, E. K.. Cazenave, L. A., and Donehower, R. C. Taxol: a a unique antineoplastic agent with significant activity in advanced novel investigational antimicrotubule agent. J. Natl. Cancer Inst., 82: ovarian epithelial neoplasms. Ann. Intern Med., 111: 273-279, 1989. 1247-1259, 1990. 3. Thigpen, T., Blessing, J., Bali, H., Hummel, S., and Barret, R. Phase 19. Gianni, L. Use of Taxol with other anticancer agents. Proceedings II trial of Taxol as second-line therapy for ovarian cancer: a gynecologic of the European Organization for Research on Treatment of Cancer oncology group study. Proc. Am. Soc. Clin. Oncol., 9: 156, 1990. Early Drug Development Meeting, p. 54. Corfu, Greece, 1995. 4. Holmes, F. A., Walters, R. S., Therigult, R. L., Forman, A. D., 20. SOLO Power Analysis, J. L. Hintze, BMDP Statistical Software, Newton, L. K., Rabner, M. N., Buzdar, A. U., Frye, D. K., and Horto- Los Angeles, CA. 1991. bagyi, G. N. Phase II trial of Taxol. An active drug in metastatic breast 21. Gianni, L., Kearns, C. M., Giani, A., Capri, G., Vigano, L., Lo- cancer. J. Natl. Cancer Inst., 83: 1797-1805, 1991. catelli, A., Bonadonna, G., and Egorin, M. J. Nonlinear pharmacokinet- 5. Murphy, W. K., Winn, R. S., Fossella, R. F., Shin, D. M., Hynes, H. E., Gross, H. M., Davila, E.. Leimert, J. T., Dhingra, H. M., Raber, ics and metabolism of paclitaxel and its pharmacokinetic/pharmacody- M. N., Krakoff, I. H., and Hong. W. K. A Phase II study of Taxol in namic relationships in humans. J. Clin. Oncol., 13: 180-190, 1995. patients with non-small cell lung cancer. Proc. Am. Soc. Clin. Oncol., 22. Lam, F. C., Hung, C. T., and Perrier, D. Estimation of variance for 11.-294, 1991. harmonic mean half-lives. J. Pharm. Sci., 74: 229-231, 1985.

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 1996 American Association for Cancer Research. Evaluation of carboplatin pharmacokinetics in the absence and presence of paclitaxel.

C K Obasaju, S W Johnson, A Rogatko, et al.

Clin Cancer Res 1996;2:549-552.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/2/3/549

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/2/3/549. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.