the Cancer Research UK Centre, Edinburgh
    Aberdeen Royal Infirmary, Aberdeen, Scotland
    Departments of Clinical Pharmacology
    Department of Statistics, F. Hoffmann-La Roche Ltd, Basel, Switzerland

    ABSTRACT

    PURPOSE: Clinical cases of capecitabine and other fluorouracil-based chemotherapies potentiating the effects of coumarin derivatives have been reported. This study assessed the influence of capecitabine on the pharmacokinetics (PK) and pharmacodynamics (PD) of warfarin.

    PATIENTS AND METHODS: Four patients with advanced/metastatic cancer completed the study, receiving a single oral dose of 20 mg warfarin before the start of standard capecitabine treatment (day 1), and again during the third cycle of capecitabine (day 61). PK parameters of warfarin and capecitabine and PD parameters of warfarin were assessed on days 1 and 61.

    RESULTS: During capecitabine treatment, the area under the plasma concentration time curve from 0 to infinity (AUC0-) of S-warfarin increased by 57% (90% CI, 32% to 88%) with a 51% prolongation of the elimination half-life (t1/2; 90% CI, 32% to 74%). Exposure to R-warfarin was not significantly affected. Plasma concentrations of capecitabine and its metabolites were not influenced by warfarin. During capecitabine treatment, the effect of warfarin on the baseline corrected AUC of the International Normalized Ratio (INR) increased by 2.8 times (90% CI, 1.33 to 5.70), with the maximum observed INR value almost doubling. Because of the administration of vitamin K to some patients with elevated INRs, these figures are likely to underestimate the true PD effect. Mean baseline factor VII levels dropped while on capecitabine therapy, potentially contributing to the observed PD interaction, though this effect did not reach statistical significance.

    CONCLUSION: There is a significant pharmacokinetic interaction between capecitabine and S-warfarin, resulting in exaggerated anticoagulant activity. Patients receiving warfarin anticoagulant therapy concomitantly with capecitabine should have their INR closely monitored and warfarin doses adjusted accordingly.

    INTRODUCTION

    Capecitabine (Xeloda; Roche Laboratories Inc, Nutley, NJ), is a rationally designed, orally administered prodrug of fluorouracil (FU).1 The product label states that in vitro human liver microsome assays have not suggested any significant potential for interaction between capecitabine or its metabolites and drugs which are substrates for the cytochrome P450 (CYP) isozymes 1A2, 2A6, 2C9, 2C19, 2D6, 2E1, or 3A4 (http://www.rocheusa.com/products/xeloda/pi.pdf), which would include anticoagulant coumarin derivates. In contrast, case reports in the literature and postmarketing safety surveillance of patients receiving capecitabine have provided clinical evidence for a possible interaction between capecitabine and coumarin derivatives.2-4 The primary objective of this study was to therefore investigate in detail the effect of capecitabine on the pharmacokinetics (PK) and pharmacodynamics (PD) of warfarin in a clinical setting. The risks associated with prolonged administration of therapeutic doses of warfarin to cancer patients who did not require anticoagulation could not be justified and a simple study design comparing the PK and PD profiles of warfarin after two single doses, one in the absence of capecitabine and the other during chronic exposure to capecitabine, was employed.

    PATIENTS AND METHODS

    Study Design and Patients

    This open-label PK/PD study was conducted in two clinical centers in Scotland, United Kingdom, in accordance with the principles of the Declaration of Helsinki and the International Conference on Harmonization guidelines on Good Clinical Practice. Three women and three men suffering from advanced/metastatic breast or colorectal cancer were enrolled. The study design provided for inclusion of 12 patients, but the study was terminated early in light of the consistent and conclusive PK and PD data presented in the Results section. Study-specific exclusion criteria included any evidence of bleeding disorders, known hypersensitivity to warfarin, baseline International Normalized Ratio (INR) of more than 1.4, significant abnormalities in liver function tests, significant surgery or disease that could impact on the PK of the test drugs, vegetarians or vegans, and/or systemic imidazole derivatives administered less than 4 weeks from the planned first dose of warfarin. All concomitant medications were kept to a minimum during the study. Whenever possible, any concomitant medications required were maintained at a constant dosage throughout the study period. To avoid variability in dietary vitamin K1 intake, patients were advised not to consume excessive quantities of leafy green vegetables or liver during the study, but otherwise not to deviate from their normal dietary habits.

    All patients received a single oral dose of 20 mg warfarin on study day 1, 8 days before the start of capecitabine treatment. From study day 8 to study day 70 they received three cycles of 1,250 mg/m2 capecitabine orally twice daily for a period of 14 days, followed by a rest period of 7 days. On study day 61, that is, day 12 of the third cycle of capecitabine treatment, a second 20-mg dose of warfarin was administered orally. PK and PD parameters of warfarin were assessed on study days 1 and 61 at 30 minutes predose, and at 1, 2, 4, 8, 12, 24, 32, 48, 72, 96, 120, and 144 hours after dose administration. Pharmacokinetic parameters of capecitabine and of its metabolites were investigated on study days 20 and 61 at 30 minutes predose (day 20 only) and at 2, 4, and 8 hours after dose administration. On the study days of blood sampling for pharmacokinetics, both drugs were administered orally within 10 minutes after completion of a standardized breakfast with approximately 200 mL of water. For safety reasons, if the INR exceeded 3.0 at any point during the monitoring, or if it exceeded 2.5 on two consecutive measurements, an oral dose of 5 mg vitamin K was administered.

    Analytic Assay

    Plasma concentrations of capecitabine and its metabolites (5'-deoxy-5-fluorocytidine [5'-DFCR], 5'-deoxy-5-fluorouridine [5'-DFUR], FU, and -fluoro--alanine [FBAL]) were determined using standard liquid chromatography with mass-spectrometry detection (LC/MS-MS) techniques.5 Serum R- and S-warfarin concentrations were measured using the method entitled "Determination of Total R-Warfarin and S-Warfarin in Human Serum by LC/MS/MS," developed and validated at PPD Development (Madison Laboratory, Madison, WI). All original data and required supporting information for this study are archived under Project Code ABJM at PPD Development (Middleton, WI).

    All concentrations were interpolated from calibration curves ranging from 5 to 1,000 ng/mL for the warfarin analytes, 0.01 to 5 μg/mL for the capecitabine and 5'-DFCR, 0.05 to 25 μg/mL for the 5'-DFUR, 0.002 to 1 μg/mL for the FU, and 0.015 to 7.4 μg/mL for the FBAL free-base analytes. Any sample whose concentration was below the lowest point on the calibration curve was reported as below the limit of quantification. Any sample whose concentration was above the upper limit of quantitation was reanalyzed after dilution of the sample with blank human serum or plasma, as appropriate, so that the observed concentration fell within the calibrated range and then was multiplied by the aliquot factor to obtain the concentration of the original sample.

    Pharmacokinetic Parameters

    Pharmacokinetic parameters were assessed by standard noncompartmental analysis, using WinNonlin Version 3.1 (Pharsight Corporation, Mountain View, CA).6

    Pharmacodynamic Parameters

    For prothrombin time (expressed as INR), baseline corrected AUC0-144 (area under the plasma concentration time curve from 0 to 144 hours postdose) for INR (AUCcorr), baseline corrected maximum INR (Emax,corr), maximum INR (Emax), and time to reach maximum INR (tmax) were estimated. Baseline corrected AUCs were calculated by multiplying the baseline measurement by 144 hours and subtracting this from the measured AUC0-144.

    Prothrombin time was expressed as INR using the following equation:

    where ISI = International Sensitivity Index, PTpt = prothrombin time of patient, and PTref = reference prothrombin time.

    Factor VII activity was measured using an in-house, one-stage standard assay of the Hematology Laboratory of the Aberdeen Royal Infirmary (Aberdeen, Scotland, United Kingdom). The coagulation time of unknown samples was measured and the according factor VII activity was read from a standard curve, prepared from different dilutions of reference plasma (Technoclone Ltd, Vienna, Austria) in factor VII–deficient plasma (Progen Immuno Diagnostica, Heidelberg, Germany). Factor VII activity was expressed as baseline corrected AUC0-144 (AUCcorr) and baseline corrected maximum effect (Emax,corr). Baseline corrected AUCs were calculated by multiplying the baseline measurement by 144 hours and subtracting the result from the measured AUC0-144.

    Vitamin K1 concentration assessments were attempted on blood samples obtained in this study but were all below the limit of quantification of the assay employed (20 ng/mL).

    Statistical Analysis

    An analysis of variance model, using a fixed effect for time and random effects for the patients, was used for assessing the effect of capecitabine on the primary pharmacokinetic and pharmacodynamic end points of AUC0- of S-warfarin and AUCcorr of INR, respectively. It was to be concluded that there was no relevant interaction if the 90% confidence limits for the geometric mean ratio of the effect of the combined treatment relative to warfarin alone were included in the reference region 0.80 to 1.25.7 Other PK and PD variables were regarded as secondary; 90% CI were calculated using the same method described for the primary parameters.

    RESULTS

    Six patients were enrolled. These patients were characterized by an age range of 43 to 63 years, a weight range of 67 to 92 kg, a body-surface area range from 1.7 to 2.2 m2, and a Karnofsky performance status range between 80 and 100. Five patients had at least one concomitant treatment during the study, none of which were CYP2C9 inducers or inhibitors. One patient (patient 1) withdrew due to an adverse event relating to capecitabine (severe hand and foot syndrome), before the second dose of warfarin was administered. Early assessment of the PK and PD data from the first four patients completing the study (patients 2, 3, 4, and 5), prompted by the consistent clinical observation of marked INR elevations in the presence of combined capecitabine and warfarin, showed that the primary statistical objectives had been reached. Consequently, to avoid exposure to the risks of excessive anticoagulation unnecessarily, the sixth patient was not given a second dose of warfarin. Further patient accrual was stopped at this point. Full PK and PD data were therefore assessable in four patients.

    Pharmacokinetics of Warfarin

    Descriptive and comparative statistics of the PK of the two enantiomers of warfarin on study days 1 and 61 revealed that higher exposure to S-warfarin was obtained with concomitant administration of warfarin plus capecitabine compared with warfarin alone (Table 1). Coadministration of capecitabine increased the AUC0- of S-warfarin by 57% (90% CI, 32% to 88%) and its elimination half-life (t1/2) by 51% (90% CI, 32% to 74%). Apparant clearance (CL/F) was decreased by 36% (90% CI, 24% to 47%). Time to reach maximum plasma concentration (tmax) was doubled, but the maximum plasma concentration (Cmax) of S-warfarin was unaffected. The 90% CI for the change in AUC0-, t1/2, and CL/F of S-warfarin when concomitantly administered with capecitabine were not within the prespecified equivalence region of 0.8 to 1.25, and thus it could not be concluded that there was no interaction.

    In contrast, comparison of the PK parameters of R-warfarin on study days 1 and 61 revealed only minor differences (Table 1), with the mean ratios of AUC0-, Cmax, CL/F, and t1/2 of R-warfarin with and without capecitabine, and their 90% CI, falling within the prespecified equivalence region.

    Pharmacokinetics of Capecitabine and Its Metabolites

    Only sparse data were available for assessing the PK of capecitabine and its metabolites, and so we did not perform a full PK profile evaluation. The plasma concentrations of capecitabine and its metabolites (5'-DFCR, 5'-DFUR, FU, and FBAL) were instead compared at selected identical time points between study days 20 and 61. We performed an exploratory comparison of the PK of capecitabine and its metabolites between day 20 and day 61 using descriptive summary statistics.

    Overall, plasma concentrations of capecitabine and its metabolites assessed at 30 minutes predose (day 20 only) and 2, 4, and 8 hours after dose administration (days 20 and 61) revealed no major differences (data not shown).

    Pharmacodynamics of Warfarin (INR and Factor VII)

    Coadministration of capecitabine and warfarin resulted in an increase of INR values with time, compared to warfarin alone (Fig 1). There was little change in the mean baseline INR between day 1 and day 61 (0.985 and 1.108, respectively; P = .1255). As shown in Table 2, capecitabine increased the primary PD parameter, INR AUCcorr, by 2.8 times, the baseline corrected maximum INR, Emax,corr, by 2.9 times, and the maximum observed value of INR, Emax, by 1.9 times. The 90% CI of AUCcorr, Emax, and Emax,corr were outside the prespecified equivalence region. In the four patients who completed the study (Patients 2, 3, 4, and 5), individual Emax INR values ranged from 0.97 to 2.00 after intake of warfarin alone, and from 1.00 to 4.40 after concomitant administration of warfarin plus capecitabine (Fig 2). Three of these four patients received vitamin K treatment as per the protocol due to INR values higher than 3.0 following their second dose of warfarin.

    The factor VII activity over time after combined treatment with warfarin and capecitabine compared with that after warfarin alone is shown in Figure 3. The day 61/day 1 geometric mean ratios of baseline corrected AUC (hours*%activity) and baseline corrected Emax (maximum decrease from baseline in % activity) were 0.923 (90% CI, 0.369 to 2.311) and 0.726 (90% CI, 0.468 to 1.128), respectively. There was a notable difference in mean baseline factor VII activity between day 1 and day 61 (85.5% and 57.3%, respectively), but this difference did not reach statistical significance (P = .1204).

    DISCUSSION

    Warfarin occurs as a pair of enantiomers that are extensively and differently metabolized by human CYP isozymes.8 R-warfarin is primarily metabolized by CYP1A2 and CYP3A4, while S-warfarin, the more active enantiomer, is metabolized almost exclusively by CYP2C9. The isozymes CYP2C9 and CYP3A4 are also involved in the metabolism of the anticoagulants phenprocoumon and acenocoumarol.9,10 Clinically relevant thromboses are relatively common within the oncology setting, in association with either the underlying malignancy or its treatment.11 Therapeutic or prophylactic anticoagulation with coumarin derivatives is therefore a frequent backdrop to anticancer intervention with chemotherapeutic agents.12,13

    Capecitabine is an oral fluoropyrimidine carbamate used widely in the treatment of metastatic breast cancer and colorectal cancer, with more than 500,000 patients treated since the mid-1990s. Capecitabine is absorbed rapidly, largely unchanged, from the gastrointestinal tract before being sequentially metabolized via 5'-DFCR and 5'-DFUR to FU. The final metabolic step in the cytotoxic production cascade occurs preferentially at the site of the tumor through the enzyme thymidine phosphorylase, leading to the demonstration of higher concentrations of FU in tumor versus normal tissue.14 Postmarketing safety surveillance for capecitabine and case reports in the literature2-4 have provided evidence for a possible interaction between capecitabine and various coumarin derivatives, particularly warfarin. In addition, retrospective case series documenting interactions between warfarin and FU suggest the potential for a more general fluoropyrimidine:warfarin class effect.13,15-17

    The effects of capecitabine on the pharmacokinetics and pharmacodynamics of warfarin in patients with advanced/metastatic solid tumor were formally investigated in this study. When capecitabine and warfarin were coadministered, the mean AUC0- of S-warfarin increased by 57% (90% CI, 32% to 88%), with a 51% (90% CI, 32% to 74%) prolongation of the t1/2 and corresponding decrease in the CL/F of S-warfarin by 36% (90% CI, 47% to 24%), compared with when warfarin was administered alone (Table 1, Fig 1). This is the first documentation of a fluoropyrimidine drug affecting the pharmacokinetics of warfarin in man. Changes in the pharmacokinetics of R-warfarin during coadministration of capecitabine, in comparison with when warfarin was administered alone, fell within the statistically prespecified equivalent region, implying an enantiomer-specific effect (Table 1). No statistically significant effects of warfarin on the pharmacokinetics of capecitabine were discernable.

    These data suggest that in humans capecitabine or its metabolites may be interacting with CYP2C9, but that they are less likely to be having any major effect on CYP3A4 or CYP1A2. As a significant proportion of R-warfarin elimination may also occur through noncytochrome-mediated reduction reactions, it is, however, impossible to rule out a capecitabine-related effect on CYP3A4 or CYP1A2 masked by a compensatory increase in R-warfarin metabolism through such non-CYP pathways.8

    Two lines of evidence pertain to the potential mechanisms underlying the apparent capecitabine-CYP interaction noted in this study. First is the the lack of direct inhibition of CYP2C9 by capecitabine noted in human donor liver microsome assays, reported in the capecitabine product label (http://www.rocheusa.com/products/xeloda/pi.pdf), and second is the long time delay before evidence of interaction reported in some clinical cases. Although in vitro liver microsome assays for investigating drug interactions may be influenced by the concentrations of inhibitors and probe substrates used, as well as by the exact methodology employed, both sets of data are consistent with enzyme downregulation, rather than direct enzyme inhibition, representing the underlying mechanism.2 A less likely alternative would be that a slowly accumulating metabolite, one that is not generated in sufficient amounts in the in vitro studies, is responsible.

    Since the final step of the activation cascade of capecitabine generates FU, it is tempting to speculate that the noted clinical interaction between FU and warfarin may be occurring through similar mechanisms. However, there is no direct evidence to confirm this. Similar delays in the onset of clinically relevant pharmacodynamic interactions have been noted for FU though, and FU administration does modulate the constitutive expression of certain CYP isozymes in a whole rat model.15,18 Retrospective studies of the effects of FU and capecitabine on INR in man have failed to show any predisposing factors for FU:warfarin interactions; in particular, no obligate links to the presence or absence of liver metastases or to overt hepatic dysfunction were discernable.2,13 The demonstration of a clear pharmacokinetic interaction between capecitabine and S-warfarin in the current study, with patients acting as their own controls, fully supports these earlier observations.

    In the presence of warfarin coadministered with capecitabine, the INR values were increased (by 2.8 times in baseline corrected AUC, by 1.9 times in Emax, and by 2.9 times in Emax,corr), compared with when warfarin was administered alone (Table 2, Fig 2). Administration of oral vitamin K to three of four patients was indicated following the coadministration of capecitabine and warfarin because their INR exceeded a value of 3.0, therefore the increases in these INR parameters reported are likely to represent underestimates of the true effect.

    In addition to the clear pharmacokinetic explanation, we attempted to investigate whether a direct pharmacodynamic interaction between capecitabine and warfarin involving vitamin K and/or levels of active clotting factors could have contributed to the observed INR changes.

    It has previously been suggested that FU, and by extension, capecitabine-induced mucositis within the gastrointestinal tract may limit the intake and/or the absorption of vitamin K, increasing the sensitivity of patients to vitamin K antagonists, such as warfarin.17,19 Unfortunately, all blood samples tested in this study, including those at baseline, were below the limit of quantification of the vitamin K1 assay employed. Nevertheless, an effect on vitamin K intake/absorption is unlikely to be a major contributing factor, considering that in this study and in other studies no clear links between the severity of fluoropyrimidine toxicity in the gastrointestinal tract and the degree of INR elevation in the presence of warfarin have been noted.13 Similarly, patients with elevated INRs in the presence of warfarin who continued on fluoropyrimidines, both in this study (data not shown) and in studies involving FU,13,15 rapidly responded to orally administered vitamin K.

    There was no discernable difference in the extent to which warfarin lowered factor VII activity levels in the presence or absence of capecitabine (Fig 3) after baseline correction, but there was a noticeable drop in the mean baseline factor VII levels. This drop in baseline factor VII activity following capecitabine therapy did not reach statistical significance in this small study, nor was it associated with elevation of the baseline INR, but it is consistent with other recent observations of the effect of capecitabine on the activity levels of certain clotting factors in man including factors II, VII, and X.20 Larger studies with capecitabine and other fluoropyrimidines will be required in the future to fully investigate the extent to which direct effects on clotting factor activity levels contribute toward INR changes in any noted fluoropyrimidine:warfarin interaction.

    In conclusion, this study demonstrates a clear pharmacokinetic interaction between capecitabine and S-warfarin, resulting in exaggerated anticoagulant activity. In addition, there was some suggestion of a direct pharmacodynamic effect of capecitabine on factor VII activity levels that may have contributed to the observed INR effects. These data provide further justification for the advice that patients receiving concomitant oral warfarin anticoagulant therapy and capecitabine should have their INR monitored closely throughout treatment and their warfarin dose adjusted accordingly. This is particularly relevant to patients receiving long-term treatment whose warfarin dose may have been considered to be stable before commencing capecitabine. In the event of severe or uncontrollable warfarin interaction, if anticoagulation is still indicated, noncoumarin-based approaches, for example, with low molecular weight heparin, should be considered as alternatives.

    Authors' Disclosures of Potential Conflicts of Interest

    The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Bruno Reigner, Hoffmann-La Roche; Susan Grange, Hoffmann-La Roche; Markus Abt, Hoffmann-La Roche; Erhard Weidekamm, Hoffmann-La Roche. Consultant/Advisory Role: Jim Cassidy, Roche; Duncan Jodrell, Roche. Honoraria: Jim Cassidy, Roche; Duncan Jodrell, Roche. Research Funding: Duncan Jodrell, Roche. Other Remuneration: Duncan Jodrell, Roche. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.

    Acknowledgment

    We thank John Curran and Anne Young of the CRUK Pharmacology and Drug Development Group Clinical Trials Unit, Edinburgh, Scotland.

    NOTES

    Authors' disclosures of potential conflicts of interest are found at the end of this article.

    REFERENCES

    Pentheroudakis G, Twelves C: The rational development of capecitabine from the laboratory to the clinic. Anticancer Res 22:3589-3596, 2002

    Copur MS, Ledakis P, Bolton M, et al: An adverse interaction between warfarin and capecitabine: A case report and review of the literature. Clin Colorectal Cancer 1:182-184, 2001

    Buyck HC, Buckley N, Leslie MD, et al: Capecitabine-induced potentiation of warfarin. Clin Oncol (R Coll Radiol) 15:297, 2003 (letter)

    Benedetti F: Important Prescribing Information: Potential Xeloda Interaction with Coumarin Derivatives. http://www.fda.gov/medwatch/safety/1999/xeloda.htm

    Reigner B, Blesch K, Weidekamm E: Clinical Pharmacokinetics of capecitabine. Clin Pharmacokinet 40:85-104, 2001

    Gibaldi M, Perrier D: Pharmacokinetics. New York, NY, Marcel Dekker, 1982, pp 409-417

    Schuirmann DJ: A comparison of the two one-sided tests procedure and the power approach for assessing the bioequivalence of average bioavailability. J Pharmacokinet Biopharm 15:657-680, 1987

    Wittowsky AK: Warfarin and other coumarin derivatives: Pharmacokinetics, pharmacodynamics, and drug interactions. Seminars in Vascular Medicine 3:221-230, 2003

    O'Reilly RA, Aggeler PM, Leong LS: Studies on the coumarin anticoagulant drugs: The pharmacodynamics of warfarin in man. J Clin Invest 42:1542-1551, 1963

    Holford N: Clinical pharmacokinetics and pharmacodynamics of warfarin. Clin Pharmacokinet 11:483-504, 1986

    Baron JA, Gridley G, Weiderpass E, et al: Venous thromboembolism and cancer. Lancet 351:1077-1080, 1998

    Mousa SA: Anticoagulants in thrombosis and cancer: The missing link. Semin Thromb Hemost 28:45-52, 2002

    Masci G, Magagnoli M, Zucali PA, et al: Minidose warfarin prophylaxis for catheter-associated thrombosis in cancer patients: Can it be safely administered with fluorouracil-based chemotherapy J Clin Oncol 21:736-739, 2003

    Schuller J, Cassidy J, Dumont E, et al: Preferential activation of capecitabine in tumor following oral administration to colorectal cancer patients. Cancer Chemother Pharmacol 45:291-297, 2000

    Magagnoli M, Masci G, Carnaghi C, et al: Minidose warfarin is associated with a high incidence of International Normalized Ratio elevation during chemotherapy with FOLFOX regimen. Ann Oncol 14:959-960, 2003

    Kolesar JM, Johnson CL, Freeberg BL, et al: Warfarin-5-FU interaction: A consecutive case series. Pharmacotherapy 19:1445-1449, 1999

    Brown MC: An adverse interaction between warfarin and 5-fluorouracil: Case report and review of the literature. Chemotherapy 45:392-395, 1999

    Afsar A, Lee C, Riddick DS: Modulation of expression of constitutive rat hepatic cytochrome P450 isozymes by 5-fluouracil. Can J Physiol Pharmacol 74:150-156, 1996

    Brown MC: Multisite mucous membrane bleeding due to a possible interaction between warfarin and 5-fluoruracil. Pharmacotherapy 17:631-633, 1997

    Tomasevic Z, Jelic S, Filipovic-Ljeskovic I, et al: Could capecitabine related hyperbilirubinaemia be a consequence of haemolysis Proc Am Soc Clin Onc 21:88b, 2002 (abstr 2166)

查询更多Pharmacokinetics相关信息在本站>>

《临床肿瘤学医学期刊》2005年7月第23卷第7期