Pharmacokinetics

J Clin Pharmacol 2009;49:965-972 965

The pharmacokinetics of cyclophosphamide (CYC) enantiomers were evaluated in patients with lupus nephri-tis distributed in 2 groups according to creatinine clear-ance: group 1 (90.6-144.6 mL/min/1.73 m2) and group 2 (42.8-76.4 mL/min/1.73 m2). All patients were treated with 0.75 to 1.3 g of racemic CYC as a 2-hour infusion and with 1 mg intravenous midazolam as a drug-metabolizing marker. CYC enantiomers and midazolam concentrations in plasma were measured by liquid chromatography/ tandem mass spectrometry (LC/MS/MS). The following differences (Wilcoxon test, P ≤ .05) were observed between the (S)-(–) and (R)-(+) enantiomers: AUC0-∞ 152.41 vs 129.25 µg⋅h/mL, CL 3.28 vs 3.89 L/h, Vd 31.38 vs 29.74 L, and t1/2 6.79 vs 5.56 h for group 1 and AUC0-∞ 167.20 vs 139.08 µg⋅h/mL, CL 2.99 vs 3.59 L/h, and t1/2 6.15 vs 4.99 h

for group 2. No differences (Mann test, P ≤ .05) were observed between groups 1 and 2 in the pharmacokinetic parameters of both enantiomers. No significant relation-ship was observed between midazolam clearance (2.92-16.40 mL/min⋅kg) and clearance of each CYC enantiomer. In conclusion, CYC kinetic disposition is enantioselective, resulting in higher exposures of the (S)-(–) enantiomer in lupus nephritis patients, and the pharmacokinetic param-eters of both enantiomers are not altered by the worsening of renal condition.

Keywords: Cyclophosphamide; enantiomers; lupus nephritis; pharmacokinetics; midazolam

Journal of Clinical Pharmacology, 2009;49:965-972© 2009 the American College of Clinical Pharmacology

Influence of Glomerular Filtration Rate on the Pharmacokinetics of Cyclophosphamide

Enantiomers in Patients With Lupus Nephritis

Carolina de Miranda Silva, BSC, Bruno José Dumêt Fernandes, PhD, Eduardo Antônio Donadi, MD, PhD, Lucienir Maria Silva, MD, PhD, Eduardo Barbosa Coelho, MD, PhD, Márcio Dantas, MD, PhD, Maria Paula Marques, PhD, and Vera Lucia Lanchote, PhD

From the Faculdade de Ciências Farmacêuticas de Ribeirão Preto da Universidade de São Paulo, Brazil (de Miranda Silva, Fernandes, Marques, Lanchote) and Faculdade de Medicina de Ribeirão Preto da Universidade de São Paulo, Brazil (Silva, Coelho, Dantas). Submitted for publication December 4, 2008; revised version accepted April 23, 2009. Address for correspondence: Vera Lucia Lanchote, Faculdade de Ciências Farmacêuticas de Ribeirão Preto da Universidade de São Paulo, Departamento de Análises Clínicas, Toxicológicas e Bromatológicas, Avenida do Café s/n, Campus da USP, 14040-903, Ribeirão Preto, SP, Brazil; e-mail: lanchote@fcfrp.usp.br.DOI: 10.1177/0091270009337938

Nephritis is a major cause of morbidity in sys-temic lupus erythematosus (SLE) and is pres-

ent in one half to two thirds of patients. Controlled randomized studies have shown that intravenous pulse therapy with cyclophosphamide (CYC) is effective for moderate to severe proliferative lupus nephritis.1

CYC undergoes extensive metabolism to form active (alkylating) and inactive products. Approximately 70% to 80% of the administered dose of CYC is activated by cytochrome P450 (CYP) to form 4-hydroxycyclophosphamide. Various CYP enzymes have been demonstrated to be involved in this reaction, including CYP2A6, CYP2B6, CYP3A4, CYP3A5, CYP2C8, CYP2C9, and CYP2C19, with CYP2B6 displaying the highest 4-hydroxylase activ-ity. 4-Hydroxycyclophosphamide readily diffuses into cells and spontaneously decomposes into phos-phoramide mustard, which is considered to be the ultimate metabolite responsible for the alkylating effect of CYC.2 Both the metabolites and a fraction (up to 25%) of unchanged parent compound are elimi-nated by the kidneys.3 Variations in the balance between metabolic activation and inactivation of CYC owing to drug-drug interactions, genetic factors, and

966 • J Clin Pharmacol 2009;49:965-972

DE MIRANDA SILVA ET AL

diseases make the understanding of the relationship between toxicity and exposure to CYC difficult.2

CYC is a chiral prodrug that is administered to patients in its racemic form.4 Cox et al5 reported that the (S)-(–)-CYC had twice the therapeutic index (lethal dose50/inhibition dose90) of the (R)-(+)-CYC (128.1 vs 68.9) against the ADJ/PC6 plasma cell tumor in mice. Comparison of pharmacological properties of CYC and its enantiomers in experi-ments on mice and rats revealed the most pro-nounced toxicity of (R)-(+)-CYC. (S)-(–)-CYC had the greatest antineoplastic activity in animals, but it was more toxic when compared to racemate.6 The same results were reported by Paprocka et al,7 who reported that the (S)-(–)-CYC not only exerted higher antitumor effects (Lewis lung carcinoma, 16/C mam-mary adenocarcinoma, and B16 melanoma) than (R)-(+)-CYC but revealed higher therapeutic indices as well.

The comparative kinetic disposition of the enantiomers of CYC was investigated in 4 patients with squamous cell carcinoma of the lung treated with racemic cyclophosphamide (rac-CYC) and pure enantiomers without observation of enantioselectiv-ity.8 Holm et al9 investigated 9 patients treated with intravenous rac-CYC for various neoplastic diseases and also observed no enantioselectivity in CYC pharmacokinetics. Corlett and Chrystyn10 also did not observe differences in the kinetic disposition of CYC enantiomers administered intravenously to 6 patients. Williams et al4 investigated the enantiose-lectivity on the kinetic disposition of CYC and its metabolite, dechlorethylcyclophosphamide, in 12 cancer patients treated with intravenous rac-CYC. Authors reported enantioselectivity in dechlorethyl-cyclophosphamide formation clearance: (R)-dech-lorethylcyclophosphamide was cleared twice as fast as its enantiomer (S)-dechloroethylcyclopho-sphamide.

Patients with chronic kidney disease have altera-tions in all pharmacokinetic parameters: absorption, distribution, metabolism, and elimination. Retention of unspecified retained uremic molecules may affect hepatic enzyme activity. These enzymatic changes may result in increased or decreased hepatic metabolism.11,12 Human uremic serum contains mediators that decrease rat hepatic CYP activity and expression secondary to reduced gene expression.13

The effect of renal insufficiency on CYC disposi-tion, as an enantiomeric mixture, has been investi-gated previously. Mouridsen and Jacobsen14 investigated 5 patients with severe renal insuffi-ciency (1 due to renal amyloidosis and 4 due to

chronic pyelonephritis; creatinine clearance 1.9-9.1 mL/min) treated with racemic 14C-labeled CYC and found grossly reduced urinary elimination of the total injected radioactivity, and over 3 days, only 22% was excreted compared to 65% in patients with normal function. However, Bramwell et al15 reported no evidence of accumulation of CYC or alkylating activity in 4 myeloma patients with varying degrees of renal function impairment. Juma et al16 investi-gated 7 cancer patients with moderate to severe renal impairment and compared them with a matched group of patients with renal function. The authors reported a curvilinear relationship between creati-nine clearance and t1/2 of CYC. Patients with very low creatinine clearance values have prolonged half-lives, and the t1/2 then decreases as the creatinine clearance increases until a plateau is reached. Haubitz et al3 investigated the CYC kinetic disposi-tion in 15 patients with autoimmune diseases and impaired renal function (creatinine clearance <50 mL/min). The authors reported that the clearance of CYC was decreased in patients with reduced renal function, thereby resulting in an increase in sys-temic drug exposure.

Intravenous CYC has emerged as an effective drug, for both induction and maintenance therapy of lupus nephritis.1 However, studies to provide the most effective and least toxic regimen are necessary to establish if patients with different glomerular fil-tration rates should be treated with the same or dif-ferent doses. Because CYC is eliminated mainly by metabolism, the nonrenal clearance of numerous drugs is altered in patients with kidney disease, there is an absence of data related to the enantiose-lectivity in CYC pharmacokinetics in patients with lupus nephritis, and the higher antitumor effects and the higher therapeutic indices are related to (S)-(–)-CYC, the aim of the present study is to evalu-ate the influence of glomerular filtration rate on CYC pharmacokinetic enantiomers in patients with lupus nephritis.

MATERIALS AND METHODS

Clinical Protocol

Patients from the Division of Clinical Immunology, University Hospital, Faculty of Medicine of Ribeirão Preto, University of São Paulo, scheduled for a pulse therapy with CYC, were included in this study. The study was approved by the Research Ethics Committee of the University Hospital, Faculty of Medicine of Ribeirão Preto, University of São Paulo.

INFLUENCE OF GLOMERULAR FILTRATION RATE ON CYC ENANTIOMERS

PhARMACOkIneTICS 967

The patients were informed in detail about the study and signed a free consent form. Freedom to refuse to participate or to withdraw consent during any phase of the study was guaranteed to each patient, with no penalty regarding his or her care and treatment. The study was conducted on 18 patients with systemic lupus erythematosus, diagnosed according to the criteria established by the American College of Rheumatology.17 Adult patients (18-54 years) of both sexes, nonobese, and with cardiac functions within normal limits according to their hospital records were investigated (Table I). Patients were divided into 2 groups according to kidney impairment degree and assessed by biopsy and glomerular filtration rate (GFR), according to K/DOQI guidelines.18 The study included patients with proliferative nephritis (class III or IV) and patients with membranous nephropa-thy (class V) scheduled for a pulse therapy with CYC. GFR was evaluated using creatinine clearance. Stage 1 means kidney damage with GFR ≥90 mL/min/1.73 m2, stage 2 means kidney damage with GFR 60 to 89 mL/min/1.73 m2, and stage 3 means

kidney damage with GFR 30 to 59 mL/min/1.73 m2. Group 1 included lupus nephritis patients within class I of K/DOQI kidney impairment, and group 2 included lupus nephritis patients within classes II and III of K/DOQI kidney impairment (Table I).

Patients were treated with monthly doses (750-1300 mg) of intravenous racemic CYC (Cycram, Meizler, São Paulo, SP, Brazil). Patient characteris-tics are summarized in Table I. Racemic cyclophos-phamide doses were administered as a 2-hour infusion. On one occasion, the pharmacokinetics of CYC enantiomers were evaluated, and the in vivo activity of CYP3A was also investigated using mida-zolam (MDZ; 1 mg intravenously [IV]) as a drug marker. For CYC pharmacokinetics, blood samples were collected at 15, 30, 45, and 60 minutes and 1.15, 1.3, 2, 3, 5, 8, 12, 14, 17, 20, and 24 hours after commencing the infusion. For MDZ pharmacokinet-ics, blood samples were collected at 15, 30, and 60 minutes and 2, 3, 5, and 6 hours after MDZ adminis-tration. Blood samples were transferred to tubes containing heparin (Liquemine, 5000 IU, Roche, São

Patients

Lupus

Nephritis

GFR, mL min/

1.73 m2

K/DOQI Classa

Age, y

BMI, kg/m2

Midazolam Clearance, mL/min⋅kg

Concomitant

Drugsb

Group 1 (A) Class IV 144.6 1 18.5 21.69 6.31 1, 2, 3, 4, 7, 19, 20 (B) Class IV 97.4 1 39.92 24.94 12.66 4, 5, 6, 9, 16, 20 (C) Class IV 144.2 1 24 21.06 9.15 1, 2, 3, 4, 5, 6, 7, 8, 20 (D) Class IV 96.8 1 32.92 23.01 8.95 1, 2, 3, 4, 6, 7, 10, 13, 15, 20 (E) Class IV 123.4 1 24.67 22.01 5.75 2, 4, 6, 18, 20 (F) Class III 113.4 1 25.17 19.44 2.92 1, 2, 4, 5, 6, 20 (G) Class IV 90.6 1 42.25 34.13 3.78 1, 2, 4, 6, 9, 20, 23, 24 (H) Classes IV and V 94.3 1 30.17 22.92 12.98 1, 2, 4, 5, 6, 10, 11, 12, 17, 19, 20, 22, 25 (I) Class III 99.8 1 24.92 22.15 15.13 1, 2, 4, 6, 19, 20 (J) Classes IV and V 100.3 1 36.25 21.60 16.40 1, 2, 4, 6, 7, 10, 20, 22Group 2 (K) Class IV 50.3 3 48.17 21.59 7.92 1, 2, 3, 4, 6, 9, 11, 12, 13, 20 (L) Classes IV and V 55.2 3 54 26.93 3.63 1, 2, 4, 6, 10, 14, 17, 19, 20 (M) Class IV 42.8 3 41.42 22.33 5.93 1, 2, 3, 4, 5, 6, 10, 15, 17, 19, 20 (N) Class IV 46.9 3 21.25 20.58 4.26 1, 2, 4, 6, 9, 11, 20 (O) Class IV 76.4 2 24.92 20.46 13.01 1, 6, 20 (P) Class III 65.3 2 23.08 19.06 9.03 1, 2, 4, 5, 6, 17, 19, 20, 21 (Q) Class V 60.0 2 28.5 27.49 7.63 1, 2, 4, 6, 10, 15, 19, 20, 22, 26, 27 (R) Classes IV and V 48.7 3 44.83 24.80 10.00 1, 2, 4, 12, 20, 22, 28, 29, 30

Table I Individual Data of Investigated Patients (n = 18)

BMI, body mass index; GFR, glomerular filtration rate.a. Stages of chronic kidney disease according to the K/DOQI classification.18

b. (1) Prednisone, (2) methylprednisolone, (3) acetyl salicylic acid, (4) furosemide, (5) chloroquine diphosphate, (6) ondansetron, (7) bromopride, (8) metamizol, (9) enalapril, (10) captopril, (11) hydrochlorthiazide, (12) clonidine, (13) nifedipine, (14) alendronate, (15) azathioprine, (16) metformin, (17) vitamin D3, (18) paracetamol, (19) calcium carbonate, (20) mesna, (21) thyroxine, (22) amlodipine, (23) fluoxetine, (24) amitriptyline, (25) atenolol, (26) hydroxychloroquine, (27) losartan, (28) propranolol, (29) sodium carbonate, (30) ferrous sulfate.

968 • J Clin Pharmacol 2009;49:965-972

DE MIRANDA SILVA ET AL

Paulo, SP, Brazil). Samples were centrifuged at 2000 g for 10 minutes, and the plasma samples were stored at –75°C until chromatographic analysis.

Analysis of CYC Enantiomers in Plasma

The analysis of CYC enantiomers was carried out by liquid chromatography/tandem mass spectrometry (LC/MS/MS) according to the method previously described by De Miranda Silva et al.19 Briefly, chro-matographic separation was obtained with a Chiralcel ODR column (Chiral Technologies, Inc, Exton, Pennsylvania) and LiChrospher 100 RP-18 pre- column (Merck, Darmstadt, Germany), with the mobile phase consisting of water/acetonitrile (75:25 v/v) added with 0.2% formic acid. For sample prep-aration, 0.2-mL plasma aliquots were supplemented with 25 µL antipyrine (internal standard 0.1 mg/mL methanol) and 5 mL of the ethyl acetate/chloroform mixture (75:25 v/v). After extraction for 30 minutes and centrifugation at 2000 g for 10 minutes, the organic phases were transferred to conical tubes and evaporated dry under vacuum. The residues obtained were reconstituted in 200 µL of the mobile phase and 100 µL of n-hexane. Then, 40 µL of the aqueous phase was chromatographed. The protonated ions and their respective ion products were monitored in the transition of 241→160 for CYC enantiomers and 189→104 for the internal standard. The method was linear in the concentration range of 2.5 to 25 000 hg/mL plasma for each enantiomer. The coefficients of variation obtained in the study of precision and inaccuracy were less than 15%.

Analysis of Midazolam in Plasma

The analysis of midazolam was carried out by LC/MS/MS according to the method of Jabor et al.20 Briefly, chromatographic separation was obtained with a Purospher RP-18e column with the mobile phase consisting of 10 mM acetonitrile/ammonium acetate (50:50, v/v). For sample preparation, 1-mL plasma aliquots were supplemented with 25 µL clobazam (internal standard 0.1 µg/mL methanol), 100 µL 0.1 M sodium hydroxide, and 4 mL of the toluene/isoamilic alcohol mixture (100:1, v/v). After extraction for 30 minutes and centrifugation at 2000 g for 10 minutes, the organic phases (3.5 mL) were evaporated dry. The residues were reconstituted in 50 µL of the mobile phase, and 20 µL was chromato-graphed. The protonated ions and their respective ions products were monitored in the transition of 301→259 for midazolam and 326→291 for the

internal standard. The method was linear in the con-centration range of 0.1 to 100 ng/mL. The coeffi-cients of variation obtained in the study of precision and inaccuracy were less than 15%.

Pharmacokinetic and Statistical Analysis

The enantioselective kinetic disposition of cyclo-phosphamide and midazolam was determined using WinNonlin software, Version 4.0 (Pharsight, Mountain View, California). Area under the curve for plasma concentrations versus time from 0 to infinity (AUC0-∞) was calculated by the trapezoidal method. Total clearance (CL) was obtained using the equation CL = dose/AUC0-∞. The elimination con-stant (ke) was estimated using the equation 0.693/t1/2. Distribution Volume (Vd) was obtained using the equation Vd = (dose/AUC0-∞)/ke.21

Median and mean values and 95% confidence intervals (95% CIs) were determined with the Graphpad Instat software, Version 3.0 (Graphpad Software, Inc, San Diego, California). The Wilcoxon test was employed to evaluate enantiomeric rate dif-ferences between (S)-(–)-cyclophosphamide and (R)-(+)-cyclophosphamide. The Mann-Whitney test was employed to evaluate pharmacokinetic differ-ences for each enantiomer between group 1 and group 2. P values <.05 were regarded as statistically significant.

Correlation and orthogonal regression analysis between midazolam clearance and clearance of each CYC enantiomer was performed according to the equations described by Schellens et al22 using the software GMC (Geraldo Maia Campos, Biological Research, Version 6.6, Ribeirão Preto, Brazil). P val-ues <.05 were regarded as statistically significant.

RESULTS

The plasma concentrations versus time curves for (R)-(+) and (S)-(–)-cyclophosphamide are presented in Figure 1 as means ± SEM for group 1 (n = 10) and group 2 (n = 8). The plasma concentrations of (S)-(–)-cyclophosphamide were significantly higher (P ≤ .05) than those of the (R)-(+) enantiomer.

The enantiomeric ratios (S)-(–)/(R)-(+) of the plasma concentrations of CYC 24 hours after beginning the infusion are presented in Figure 2 as means ± SEM for group 1 (n = 10) and group 2 (n = 8). A relevant obser-vation was the progressive increase of the plasma (S)-(–)/(R)-(+)-cyclophosphamide concentration ratio 6 hours after beginning the infusion.

INFLUENCE OF GLOMERULAR FILTRATION RATE ON CYC ENANTIOMERS

PhARMACOkIneTICS 969

The pharmacokinetics parameters are presented in Table II as medians (95% CI). The following dif-ferences (P ≤ .05) were observed between the (S)-(–) and (R)-(+) enantiomers: AUC0-∞ 152.41 vs 129.25 µg⋅h/mL, CL 3.28 vs 3.89 L/h, Vd 31.38 vs 29.74 L, and t1/2 6.79 vs 5.56 h for group 1 and AUC0-∞ 167.20 vs 139.08 µg⋅h/mL, CL 2.99 vs 3.59 L/h, and t1/2 6.15 vs 4.99 h for group 2. The patients of both groups presented enantioselectivity in CYC pharmacokinetics, with plasma accumula-tion of (S)-(–)-CYC.

No differences were observed between group 1 and group 2 in the pharmacokinetics parameters of both CYC enantiomers.

The patients investigated presented IV midazo-lam clearance of 2.92 to 16.40 mL/min⋅kg (Table I). No differences were observed in midazolam clear-ance between group 1 and group 2. There was no correlation between midazolam clearance and clear-ance of each CYC enantiomer when all patients were put together or when patients were evaluated as group 1 and group 2.

DISCUSSION

The presence of lupus nephritis in the patients included in this investigation was confirmed by biopsy and clinical and laboratory tests. Patients with creatinine clearance within the normal range (90.6-144.6 mL/min/1.73 m2) were included in group 1, whereas patients with creatinine clearance ranging from 30 to 89 mL/min/1.73 m2

(42.8-76.4 mL/min/1.73 m2) were included in group 2 and were characterized as stages II and III of chronic kidney disease.18 All patients studied had a clinical indication for the use of rac-CYC, with the prescribed dose ranging from 750 to 1300 mg.

The patients also received other medications dur-ing the study that could not be withdrawn because of ethical reasons (Table I). Ondansetron, which was administered to all patients except for patient 1, is an inducer of CYC metabolism. Gilbert et al23 observed a 17% reduction in the area under the plasma concentration versus time curve of CYC in patients with breast cancer. Prednisone, which was administered to most patients, is also considered an inducer of CYC metabolism. Faber et al24 reported an increase in the plasma concentrations of CYC metabolites in patients treated with prednisone for 12 days. However, the kinetic disposition of CYC was evaluated only in patients with preserved renal function.

Jarman et al8 reported the absence of enantioselec-tivity in the binding of CYC to plasma proteins and in the renal excretion of the unchanged drug in 4 patients with lung cancer. Corlett and Chrystyn10 also observed no enantioselectivity in the pharma-cokinetics of CYC in 6 patients treated with the racemic drug. However, Williams et al4 reported enantioselectivity in 12 cancer patients on the for-mation clearance of the metabolite dechloroethylcy-clophosphamide, with preferential metabolism of (R)-(+)-CYC (0.25 vs 0.14 L/h).

0 5 10 15 20 250

2

4

6

8

10

12

14

16

(S)–(–)–CYC Group 1

(R)–(+)–CYC Group 1

(S)–(–)–CYC Group 2

(R)–(+)–CYC Group 2

CY

C p

lasm

a co

nce

ntr

atio

n (

ug

/mL

)

Time (h)

Figure 1. Pharmacokinetic profiles of (S)-(–)-CYC and (R)-(+)-CYC in patients with lupus nephritis from group 1 (n = 10) and group 2 (n = 8). Data are expressed as mean ± SE.

0 5 10 15 20 25

0,70,80,91,01,11,21,31,41,51,61,71,81,92,02,12,22,3

Group 1

Group 2

(S)-

(-)-

CY

C/(

R)-

(+)-

CY

C p

lasm

a ra

tio

s

Time (h)

Figure 2. Plasma concentrations (S)-(–)/(R)-(+)-CYC ratios in patients with lupus nephritis from group 1 (n = 10) and group 2 (n = 8). Data are expressed as mean ± SE.

970 • J Clin Pharmacol 2009;49:965-972

DE MIRANDA SILVA ET AL

The plasma concentration versus time curves of the (S)-(–)-CYC and (R)-(+)-CYC enantiomers obtained for patients of groups 1 and 2 are shown in Figure 1. Despite the associated administration of different medications (Table I), interpatient variation in the pharmacokinetic parameters was low, as dem-onstrated by the coefficients of variation of 31% and 27% for (S)-(–)-CYC and (R)-(+)-CYC, respectively, obtained for group 1 patients. In group 2, coeffi-cients of variation of 25% and 35% were observed.

The pharmacokinetics of CYC were enantioselec-tive in the patients with lupus nephritis. Patients of group 1 (152.41 vs 129.25 µg⋅h/mL) and group 2 (167.20 vs 139.08 µg⋅h/mL) presented larger areas under the plasma concentration versus time curve of the (S)-(–)-CYC enantiomer. It should be emphasized that the (S)-(–)-CYC enantiomer is considered the active one (eutomer) by most investigators.5-7

The enantiomer ratios of the plasma concentra-tions of (S)-(–)/(R)-(+)-CYC increased as a function of time of drug administration, with the highest ratio being observed between 6 to 20 hours (Figure 2). For patients of group 1, ratios close to 1 were observed immediately after beginning the infusion and close to 1.5 at 24 hours after beginning the infusion. For patients of group 2, values close to 1 were observed immediately after beginning the infusion and close to 2.5 after 24 hours. These

results suggest that (R)-(+)-CYC is the enantiomer preferentially metabolized in patients with lupus nephritis.

Total clearance was higher for the (R)-(+)-CYC enantiomer than for (S)-(–)-CYC in both investigated groups (group 1: 3.89 vs 3.28 L/h; group 2: 3.59 vs 2.99 L/h; Table II). In contrast, the CL values reported by Williams et al4 for cancer patients treated with 2100 mg/m2 rac-CYC and by Corlett and Chrystyn10 for patients treated with 300 to 750 mg/m2 rac-CYC did not indicate enantioselectivity. Total clearance of the (R)-(+)-CYC and (S)-(–)-CYC enantiomers was 6.9 L/h and 7.2 L/h, respectively, in the study by Williams et al4 and 0.049 and 0.048 L/h⋅kg in the study by Corlett and Chrystyn.10 The higher total clearance reported by Williams et al4 for both CYC enantiomers (6.9 L/h and 7.2 L/h) when compared to the present investigation (2.99-3.89 L/h) was proba-bly due to the fact that the patients investigated by these authors included children and adolescents in addition to adult patients (9 of 12).

The data shown in Table II also indicate a shorter elimination half-life for the (R)-(+)-CYC enantiomer (group 1: 5.56 vs 6.79 h; group 2: 4.99 vs 6.15 h). Williams et al4 reported elimination half-lives of 5.8 and 5.7 hours for (R)-(+)-CYC and (S)-(–)-CYC, respectively, whereas Corlett and Chrystyn10 obtained values of 6.82 and 7.13 hours.

Table II Pharmacokinetic Disposition of Cyclophosphamide Enantiomers in Patients With Lupus Nephritis Treated With 750 to 1300 mg of rac-CYC

Group 1 (n = 10) Group 2 (n = 8)

(S)-(–)-CYC t1/2, h 6.79 (5.72-7.49)† 6.15 (5.18-7.19)† Kel, h–1 0.10 (0.09- 0.13)† 0.11 (0.09-0.13)† AUC, h⋅µg/mLa 152.41 (130.58-169.86)† 167.20 (145.48-186.64)† Vd, L 31.38 (26.07-39.61)† 27.08 (24.24-29.43) Vd, L/kg 0.51 (0.42-0.63)† 0.47 (0.45-0.51) CL, L/h 3.28 (2.95-3.92)† 2.99 (2.66-3.49)† CL, L/h⋅kg 0.055 (0.046-0.064)† 0.055 (0.030-0.068)†(R)-(+)-CYC t1/2, h 5.56 (4.42-6.51) 4.99 (4.04-6.24) Kel, h–1 0.12 (0.10- 0.17) 0.13 (0.10-0.18) AUC, h⋅µg/mLa 129.25 (108.03-148.62) 139.08 (97.84-157.19) Vd, L 29.74 (24.92-28.25) 27.55 (18.48-44.72) Vd, L/kg 0.48 (0.40-0.61) 0.5 (0.36-0.75) CL, L/h 3.89 (3.35-4.85) 3.59 (2.97-5.63) CL, L/h⋅kg 0.062 (0.052-0.079) 0.070 (0.053-0.101)

Data are expressed as median and 95% confidence interval.a. AUC corrected for 1000 mg rac-CYC dose.†P < .05 Wilcoxon test, (S)-(–)-CYC versus (R)-(+)-CYC.

INFLUENCE OF GLOMERULAR FILTRATION RATE ON CYC ENANTIOMERS

PhARMACOkIneTICS 971

The volume of distribution of CYC was enantiose-lective only in patients of group 1 (0.51 vs 0.48 L/kg for the (S)-(–) and (R)-(+) enantiomers, respectively; Table II). The enantioselectivity in the volume of dis-tribution observed cannot be attributed to plasma protein binding because Jarman et al8 reported poor, nonenantioselective binding of CYC to plasma pro-teins (35% for (R)-(+)-CYC and 36% for (S)-(–)-CYC).

Many of the components and mediators of host defense and inflammatory responses have the ability to interact with and alter the levels and activities of CYP-based drug-metabolizing enzymes. Although the majority of examples reported involve a down-regulation of the enzyme, there are a number of examples in which a specific CYP form is induced during such responses. The modulation of CYP dur-ing host defense is a multifactorial and widespread consequence of inflammation and has the potential to modify drug disposition in human therapeutics and to cause adverse drug responses.25

CYC enantioselectivity in patients with lupus nephritis could be supported by the existence of enantioselectivity toward CYP enzymes. Although there are no data regarding if the metabolism of CYC enantiomers is catalyzed by different enzymes or by the same enzymes in a different extension, an inhibition/induction of CYP expression due to inflammatory disease could lead to an alteration in clearance values between the enantiomers and to differences in plasma enantiomeric ratios.

The pharmacokinetic parameters obtained for the 2 CYC enantiomers did not differ between groups 1 and 2 (Table II), inferring that stages II and III of chronic kidney disease (group 2), when compared to stage I (group 1), do not modify the pharmacokinetics of the CYC enantiomers in patients with lupus nephritis. Thus, adjustments of the CYC dose are not necessary for patients with lupus nephritis in stage II or III of chronic kidney disease. In contrast, Haubitz et al3 reported a reduction in CYC clearance administered as the enantiomer mixture to patients with autoim-mune diseases and reduced renal function. The authors obtained total clearance of 4.7, 3.4, and 2.8 L/h, respectively, for patients with normal creatinine clearance, as well as creatinine clearance ranging from 25 to 50 mL/min and 10 to 24 mL/min. However, patients with normal kidney function and breast can-cer but without autoimmune diseases served as a reference in the study published by Haubitz et al.3

Various CYP isoenzymes are involved in the bio-activation of CYC in humans. CYP2B6 accounts for approximately 45% of 4-OH-CYC formation, CYP3A4 for approximately 25%, and CYP2C9 for

approximately 12%. Other isoenzymes such as CYP2C19, CYP2A6, and CYP2C8 contribute to a lesser extent. CYC can also be directly detoxicated by the oxidation of its side chain, leading to the forma-tion of the inactive metabolite 2-dechloroethylcyclo-phosphamide. This reaction results in the formation of equimolar amounts of chloroacetaldehyde and is mainly mediated by CYP3A4, CYP3A7, and CYP3A5 and, to a lesser extent, CYP2B6.26 Although CYP3A is not the major isoenzyme involved in CYC metabo-lism, it presents a wide interpatient variation in its activity as a result of a combination of genetic and nongenetic factors such as hormone and health sta-tus and the impact of environmental stimuli.27 Because of the high interindividual variability in CYC pharmacokinetics,2 we evaluated the CYP3A activity to clarify one of many possible mechanisms that could account for the high interindividual vari-ability in the pharmacokinetics of CYC.

Midazolam total clearance varied by 5-fold among the investigated patients (2.92-16.40 mL/min⋅kg; Table I). Our finding is consistent with the variation determined by other investigators who have used IV midazolam as a phenotyping probe. Chen et al28 reported clearance values of IV midazolam from 5.8 to 11.5 mL/min⋅kg in 37 women and from 5.5 to 9.5 mL/min⋅kg in 29 men. The same values were reported by Kharasch et al29 in the investigation of 99 healthy volunteers (7.6 ± 2.2 mL/min⋅kg, mean and SD). No statistically significant differences in midazolam clearance were noted between group 1 and group 2. No significant relationship was observed between midazolam clearance and clearance of each CYC enantiomer (Table III), suggesting that CYP3A phe-notyping does not contribute to CYC enantiomer dose individualization.

y = a + bx; a = y – bx; b = Syy – Sxx + [(Syy – Sxx)

2 + 4S2xy]½

2Sxy

Table III Orthogonal Regression Equations and Correlation Coefficients for Clearances Between

Midazolam and CYC Enantiomers Evaluated in 18 Patients With Lupus Nephritis

Midazolam Clearance

(S)-(–)-CYC clearance –4.1867 + 853069x r = 0.2467(R)-(+)-CYC clearance –152 261 + 222.2901x r = 0.0465

from Schellens et al.22 r = correlation coefficient (P > .05).

972 • J Clin Pharmacol 2009;49:965-972

DE MIRANDA SILVA ET AL

In conclusion, the kinetic disposition of CYC is enantioselective, with a higher exposure to (S)-(–)-CYC in patients with lupus nephritis, and enantiose-lectivity is not altered by worsening of renal function. However, the impact of higher exposure to (S)-(–)-CYC versus (R)-(+)-CYC on the clinical efficacy of lupus nephritis treatment remains to be determined.

Financial disclosure: The authors are grateful to Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) for financial support and for granting research fellowships and to Conselho Nacional de Desenvolvimento Científico and Tecnológico (CNPq) for granting research fellowships.

REFERENCES

1. Boumpas DT, Sidiropoulos P, Bertsias G. Optimum therapeutic approaches for lupus nephritis: what therapy and for whom? Nat Clin Pract Rheumatol. 2005;1:22-30.2. De Jonge ME, Huitema ADR, Rodenhuis S, Beijnen JH. Clinical pharmacokinetics of cyclophosphamide. Clin Pharmacokinet. 2005;44:1135-1164.3. Haubitz M, Bohnenstengel F, Brunkhorst R, Schwab M, Hofmann U, Busse D. Cyclophosphamide pharmacokinetics and dose requirements in patients with renal insufficiency. Kidney Int. 2002;61:1495-1501.4. Williams ML, Wainer IW, Granvil CP, Gehrcke B, Bernstein ML, Ducharme M. Pharmacokinetics of (R)- and (S)-cyclophosphamide and their dechloroethylated metabolites in cancer patients. Chirality. 1999;11:301-308.5. Cox PJ, Farmer PB, Jarman M, Jones M. Observations on the dif-ferential metabolism and biological activity of the optical isomers of cyclophosphamide. Biochem Pharmacol. 1976;25:993-996.6. Kleinrok Z, Chmielewska B, Czuczwar JS, et al. Comparison of pharmacological properties of cyclophosphamide and its enantiomers. Arch Immunol Ther Exp. 1986;34:263-273.7. Paprocka M, Kusnierczyk H, Budzynski W, Rak J, Radzikowski C. Comparative studies on biological activity of /+/R and /–/S enantiomers of cyclophosphamide and ifosfamide: I. Antitumour affect of cyclophosphamide and ifosfamide enantiomers. Arch Immunol Ther Exp. 1986;34:275-284.8. Jarman M, Milsted R, Smyth JF, Kinas RW, Pankiewicz K, Stec WJ. Comparative metabolism of 2-[Bis(2-chloroethyl)amino]tetra-hydro-2H-1,3,2-oxazaphosphorine-2-oxide (cyclophosphamide) and its enantiomers in humans. Cancer Res. 1979;39:2762-2767.9. Holm KA, Kindberg CG, Stobaugh JF, Slavik M, Riley CM. Stereoselective pharmacokinetics and metabolism of the enantiom-ers of cyclophosphamide. Biochem Pharmacol. 1990;39:1375-1384.10. Corlett SA, Chrystyn H. High-performance liquid chromato-graphic determination of the enantiomers of cyclophosphamide in serum. J Chromatogr B Biomed Appl. 1996;682:337-342.11. Churchwell MD, Mueller BA. Selected pharmacokinetic issues in patients with chronic kidney disease. Blood Purif. 2007;25:133-138.12. Nolin TD, Frye RF, Matzke GR. Hepatic drug metabolism and transport in patients with kidney disease. Am J Kidney Dis. 2003;42:906-925.

13. Michaud J, Dubé P, Naud J, et al. Effects of serum from patients with chronic renal failure on rat hepatic cytochrome P450. Br J Pharmacol. 2005;144:1067-1077.14. Mouridsen HT, Jacobsen E. Pharmacokinetics of cyclophosph-amide in renal failure. Acta Pharmacol Toxicol (Copenh). 1975;36: 409-414.15. Bramwell V, Calvert RT, Edwards G, Scarffe H, Crowther D. The disposition of cyclophosphamide in a group of myeloma patients. Cancer Chemother Pharmacol. 1979;3:253-259.16. Juma FD, Rogers HJ, Trounce JR. Effect of renal insufficiency on the pharmacokinetics of cyclophosphamide and some of its metabolites. Eur J Clin Pharmacol. 1981;19:443-451.17. Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythema-tosus. Arthritis Rheum. 1997;40:1725.18. US National Kidney Foundation. NKF K/DOQI GUIDELINES: K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease: Evaluation, Classification, and Stratification. 2002. http://www .kidney.org/professionals/KDOQI/guidelines_ckd/p4_class_g1.htm. Accessed November 10, 2008.19. De Miranda Silva C, Dumet Fernandes BJ, Marques Pereira MP, et al. Determination of cyclophosphamide enantiomers in plasma by LC-MS/MS: application to pharmacokinetics in breast cancer and lupus nephritis patients. Chirality. 2008;21:383-389.20. Jabor VA, Coelho EB, Santos NAG, Bonato PS, Lanchote VL. A highly sensitive LC-MS-MS assay for analysis of midazolam and its major metabolite in human plasma: applications to drug metabolism. J Chromatogr B Analyt Technol Biomed Life Sci. 2005;822:27-32.21. Gabrielsson J, Weiner D. Pharmacokinetic and Pharma-codynamic Data Analysis: Concepts & Applications. Stockholm: Swedish Pharmaceutical Press; 2000.22. Schellens JHM, Van Der Wart JHF, Danhof M, Van Der Velde EA, Breimer DD. Relationship between the metabolism of antipy-rine, hexobarbitone and theophylline in man as assessed by a ‘cocktail’ approach. Br J Clin Pharmacol. 1988;26:373-384.23. Gilbert CJ, Petros WP, Vredenburg J, et al. Pharmacokinetic interaction between ondansetron and cyclophosphamide during high-dose chemotherapy for breast cancer. Cancer Chemother Pharmacol. 1998;42:497-503.24. Faber OK, Mouridsen HT, Skovsted L. The biotransformation of cyclophosphamide in man: influence of prednisone. Acta Pharmacol Toxicol (Copenh). 1974;35:195-200.25. Renton KW. Alteration of drug biotransformation and elimina-tion during infection and inflammation. Pharmacol Ther. 2001;92: 147-163.26. Zhang J, Tian Q, Yung Chan S, et al. Metabolism and transport of oxazaphosphorines and clinical implications. Drug Metab Rev. 2005;37:611-703.27. Wojnowski L, Kamdem LK. Clinical implications of CYP3A poly-morphisms. Expert Opin Drug Metab Toxicol. 2006;2:171-182.28. Chen M, Ma L, Drusano GL, Bertino JS Jr, Nafziger AN. Sex differences in CYP3A activity using intravenous and oral midazo-lam. Clin Pharmacol Ther. 2006;80:531-538.29. Kharasch ED, Walker A, Isoherranen N, et al. Influence of CYP3A5 genotype on the pharmacokinetics and pharmacody-namics of the cytochrome P4503A probes alfentanil and midazo-lam. Clin Pharmacol Ther. 2007;82:410-426.