Validation of urinary excretion of cyclophosphamide as ... - Springer Link

Int Arch Occup Environ Health (2008) 81:285–293 DOI 10.1007/s00420-007-0211-2

O R I G I N A L A R T I CL E

Validation of urinary excretion of cyclophosphamide as a biomarker of exposure by studying its renal clearance at high and low plasma concentrations in cancer patients Maria Hedmer · Peter Höglund · Eva Cavallin-Ståhl · Maria Albin · Bo A. G. Jönsson

Received: 10 April 2006 / Accepted: 18 May 2007 / Published online: 20 June 2007 © Springer-Verlag 2007

Abstract Objectives Cyclophosphamide (CP) is an alkylating agent classiWed as a human carcinogen. Health care workers handling this drug may be exposed during, e.g., preparation or administration. Cyclophosphamide is readily absorbed by inhalation and by dermal uptake. A biomarker, CP in urine, has frequently been used to assess the occupational exposure to CP, but has not been fully validated. The aim of this study was to investigate if the proportion of the CP dose that is excreted in urine (renal clearance) is constant over diVerent plasma drug concentrations and other pharmacokinetic parameters, e.g., urine Xow. Methods Pharmacokinetics of CP were studied in 16 breast cancer patients that were treated with postoperative adjuvant chemotherapy including CP. Plasma and urine from the patients were collected at diVerent occasions up to 12 days after the dose. Urine was collected during 4-h periods and blood was sampled at the end of each period. Analysis of CP was performed by liquid chromatography tandem mass spectrometry. The limit of detection for CP in

urine and plasma was 0.01 and 0.02 ng/ml, respectively. The precisions of the developed methods were determined to ·8%. Results The administered doses of CP in absolute amounts ranged between 800 and 2,240 mg. Mean renal clearance of CP was 8.6 (conWdence interval 6.5–10.7) ml/min and was not signiWcantly dependent of the plasma drug concentration. However, a signiWcant correlation between renal clearance and urine Xow was observed. There was a large inter-individual variation in the plasma and urine concentrations even when the same doses were given. Conclusions Cyclophosphamide in urine can be continued to be used as a biomarker to monitor occupational exposure to CP, however the inter-individual variability of excretion of CP in urine, and its dependency on urine Xow must be taken into consideration in future applications. Keywords Cyclophosphamide · Antineoplastic drug · Biological monitoring · Renal clearance · Urine Xow · Pharmacokinetics · Breast cancer patients · Risk assessment

Introduction M. Hedmer (&) · M. Albin · B. A. G. Jönsson Division of Occupational and Environmental Medicine, Department of Laboratory Medicine, Lund University Hospital, 221 85 Lund, Sweden e-mail: [email protected] P. Höglund Division of Clinical and Experimental Pharmacology, Department of Laboratory Medicine, Lund University Hospital, 221 85 Lund, Sweden E. Cavallin-Ståhl Oncology, Division V, Department of Clinical Sciences, Lund University Hospital, 221 85 Lund, Sweden

Cyclophosphamide (2-[bis(2-chloroethyl)amino]tetrahydro-2H-1,3,2-oxazaphosphorine 2-oxide; CP) is an alkylating agent that is frequently used as an antineoplastic drug. It is administered as monotherapy or in combination with other drugs to treat neoplastic and non-neoplastic diseases. Cyclophosphamide is a prodrug and must be metabolized through a group of cytochrome P450 (CYP) enzyme systems in the liver, e.g., CYP2B6, CYP3A4, CYP2C9 and CYP2A6 before active alkylating metabolites can be generated (Roy et al. 1999; Parkinson 2001). The metabolites are genotoxic due to their ability to cross-link DNA and

123

286

thereby cause DNA damages. CP is classiWed as carcinogenic to humans (IARC 1981, 1987). Cyclophosphamide is widely used around the world. Health care workers handling this drug can be occupationally exposed and CP has been detected as a surface contaminant in workplaces where antineoplastic drugs are used (Sessink et al. 1992a; McDevitt et al. 1993; Minoia et al. 1998; Connor et al. 1999; Schmaus et al. 2002; Wick et al. 2003; Hedmer et al. 2005). Amounts of CP have been detected in urine (Hirst et al. 1984; Evelo et al. 1986; Sessink et al. 1992b, 1994; Ensslin et al. 1994; Minoia et al. 1998; Turci et al. 2002; Pethran et al. 2003; Wick et al. 2003), indicating that these workers are occupationally exposed to CP. Absorption of CP occurs mainly through skin or through inhalation of a gas phase or particulates. Since there is more than one exposure route for CP and extensive personal protection is used, it is suitable to use a biomarker for the determination of exposure. Biological monitoring is also often easier to perform than ambient monitoring. Thus, CP in urine has been used as a biomarker of exposure, which allowed the internal dose of CP to be estimated (Hirst et al. 1984; Evelo et al.1986; Sessink et al. 1992b; 1994; Ensslin et al. 1994; Minoia et al. 1998; Turci et al. 2002; Pethran et al. 2003; Wick et al. 2003). The major fraction of CP in the body is eliminated by hepatic metabolism, but a small fraction is eliminated by renal excretion of unchanged drug in urine. Approximately 13% of the CP dose is excreted in urine (Sladek 1994), but individual diVerences in the excretion of CP in urine in the range 3–36% have been reported (Bagley et al. 1973; Milstedt and Jarman 1982; Bailey et al. 1991; Ren et al. 1998; Joqueviel et al. 1998). The plasma half-life of CP is approximately 5 h in humans during the Wrst days after the dose (Busse et al. 1997). The current cancer risk assessments for health care workers that are occupationally exposed to CP were made by Sessink et al. (1995) and Sorsa and Anderson (1996). In those assessments renal clearance (CLR) of CP was assumed to be independent of the plasma drug concentration of CP. However, in pharmacokinetic studies performed by Busse et al. (1997, 1999) it was shown that the mean CLR was lower during treatment with conventional doses of CP compared to high doses. This indicates that CLR of CP might be dependent on the plasma drug concentration, i.e. that the proportion of the absorbed dose excreted in urine may be lower for lower doses. If there is such dependence, biological monitoring of occupationally exposed workers using CP in urine would underestimate their absorbed doses, and thus, also the occupational risk at the low exposures. The purpose of this study was to investigate CLR of CP at diVerent plasma drug concentrations of CP and thereby validate the use of CP in urine as a biomarker. This was

123

Int Arch Occup Environ Health (2008) 81:285–293

accomplished by studying patients treated with CP. In addition, other parameters, e.g., the impact of urine Xow was investigated.

Patients and methods Patient population A total of 16 women with breast cancer participated in the study. The patients were treated with postoperative adjuvant chemotherapy. Prior to entry into the study all patients received both written and oral information about the study. The patients gave their written informed consent in accordance with the Helsinki Declaration. The study was approved by the Research Ethics Committee (LU 82–03) at Lund University (Lund, Sweden). Treatment The adjuvant chemotherapy consisted of treatment every third week. Dose calculations were based on body surface area. The chemotherapy protocol of conventional dose consisted of 600 mg/m2 of CP, 60 mg/m2 of epirubicin and 600 mg/m2 of 5-Xuorouracil. Five of the patients received escalated doses of CP consisting of 1,200 mg/m2. The chemotherapy was given intravenously as an infusion through a permanent cannula. The conventional CP dose was administrated over a 30 min period while the escalated CP dose was administrated over 60 min. In addition, betamethasone, mesna (2-mercaptoethanesulphonic acid) and tropisetron hydrochloride were administrated to some of the patients. Blood and urine collection Blood and urine samples were collected at three or four diVerent occasions up to 12 days after the chemotherapy. Blood samples were collected without using any permanent cannula due to the risk of contamination. Peripheral, venous blood was collected in 4 ml tubes (Vacuette® K2EDTA, greiner bio-one, Kremsmüster, Austria). The blood samples were centrifuged at 1,000g during 10 min to obtain plasma. The plasma was transferred to plastic test tubes with screw cap and stored at ¡20°C until analysis. At the Wrst sampling occasion an extra tube (3 ml Vacutainer LH PST™ II Plus, BD, Franklin Lakes, NJ, USA) of blood was collected for analysis of plasma creatinine as a measure of renal function. The liver function of the patients was investigated by analysis of bilirubin, conjugated bilirubin, alkaline phosphatase, glutamyl transferase, aspartate aminotransferase and alanine aminotransferase in plasma.


The patients collected urine during three or four 4-h periods. Immediately before each period the patient voided and that urine was discarded. All urine produced during each 4h period was collected in 500 ml polyethylene bottles (Kautex Textron, Bonn, Germany). Patients voided at the end of the collection period and then a blood sample was drawn. The volume of the collected urine was measured and 20 ml aliquots of urine were stored at ¡20°C until analysis. Analytical methods Chemicals Cyclophosphamide monohydrate (purity 99.5%) was purchased from Sigma-Aldrich Chemie (Schnelldorf, Germany). 2H6-labelled cyclophosphamide (CP-D6; purity 97%) was from Phychem (Bergisch Gladbach, Germany) and was used as internal standard. Ethyl acetate (HPLC 99.8%) and methanol (HPLC) were from Lab-Scan (Dublin, Ireland). Acetic acid glacial (p.a. > 99.8%) and hydrochloric acid (p.a. 37%) were from Merck (Darmstadt, Germany). Trizma base (min. 99.9%) was from Sigma Chemicals (St. Louis, MO, USA). The water was puriWed in a Milli-RQ Water PuriWer from Millipore (Billerica, MA, USA). Standard preparation Stock solutions of CP were prepared by dissolving weighed amounts in water. Several working solutions were made from these stock solutions by further dilution in water. A stock solution of CP-D6 was prepared by dissolving weighed amounts in water, which were further diluted in water. Standards up to 2,000 ng/ml were prepared. Plasma and urine samples with higher drug concentrations were diluted. Tris buVer preparation A weighed amount of 12.1 g trizma base was dissolved in water. The pH value of the solution was adjusted to eight with 4 M hydrochloric acid. The solution was then diluted to 100 ml with water. Sample preparation Aliquots of 0.5 ml of plasma were transferred into test tubes and 100 l of internal standard (1.0 g CP-D6/ml water) as well as 100 l of tris buVer were added. Five milliliter of ethyl acetate was added to the samples and they were shaken with an IKA-VIBRA-VXR (IKA Labortechnik, Staufen, Germany) for 5 min. Subsequently, they were centrifuged at 1,500g during 5 min and the organic layers were transferred to test tubes.

287

Aliquots of 1 ml urine were transferred to test tubes and 100 l of internal standard, 100 l of tris buVer and 5 ml of ethyl acetate were added. The samples were shaken for 5 min, centrifuged at 680g during 5 min and the extracts were transferred into new test tubes. The urine samples were extracted a second time with 5 ml of ethyl acetate. Extracts from both plasma and urine samples were evaporated to dryness under a stream of nitrogen gas at room temperature. The samples were dissolved in 150 l 0.5% acetic acid, sonicated for 5 min and transferred into microvials. The samples were stored at 5°C until analysis. All samples were prepared in duplicate. Standards of CP were prepared by adding 100 l aliquots of the working solutions to 0.5 ml drug-free human plasma or 1 ml drug-free human urine, 100 l internal standard and 100 l tris buVer, followed by the sample preparation procedure. Determination by liquid chromatography–mass spectrometry The samples were analyzed by a Perkin-Elmer Series 200 liquid chromatography (LC) system with a Series 200 autosampler (Applied Biosystems, Norfolk, CT, USA). The column was a Genesis C18 (50 £ 2.1 mm) with a particle size of 4 m (Jones Chromatography, Lakewood, CO, USA). The column outlet was coupled to an API 3000 triple quadrupole mass spectrometer (Applied Biosystems/MDSSCIEX, Toronto, Canada) equipped with an electrospray ionisation (ESI) source. The LC conditions for analysis of CP were as follows: mobile phase A, 0.5% acetic acid; mobile phase B, methanol containing 0.5% acetic acid; time program, 0 min, 60% A/40% B; 6 min, 60% A/40% B; 8 min, 0% A/100% B; 8.1 min, 60% A/40% B. During 2.9 min the column was reconditioned (60% A/40% B). The Xow rate of the mobile phase was 0.2 ml/min and injection volume was 20 l. The ion spray voltage was set to 3,000 V and the temperature to 400°C. The instrument operated in positive ion mode using multiple reaction monitoring (MRM) at m/z 263.1/142.1 for CP and m/z 267.1/140.3 for CP-D6. The control fragment m/z 261.0/140.3 was used for CP. Declustring potentials for CP and CP-D6 were 40 and 44 V, respectively. Collision energies were 31 V for CP and 34 V for CP-D6, respectively. Peak-area ratios between the analyte and the internal standard were used for quantiWcation. The ratios between the concentrations found by the analyte fragment and the control fragment were not allowed to exceed 20%. If a larger deviation was observed the lowest value was reported, which occurred very rarely. To make standard curves, 1/x-weighted linear regression was used. Standard curves had correlation coeYcients >0.99 and showed good linearity. The limit of detection

123

288

(LOD) was determined as three times the amplitude of the noise.

Int Arch Occup Environ Health (2008) 81:285–293 Table 1 Demographic mean characteristics of the participating patients Parameters

Mean

Range

Pharmacokinetic and statistical methods From each patient three or four assessments of CLR were made. The amount CP excreted during a sampling interval was calculated as the product of volume and concentration of the urine. Plasma concentrations were analyzed at the end of each urine collection interval and the corresponding mid-point concentrations were calculated using the following assumed half-lives: 5 h up to 60 h from end of infusion; 10 h in the interval 60–120 h from end of infusion; and 42 h after 120 h from end of infusion. The CLR during each interval was then calculated as the amount excreted divided by the product of calculated mid-point concentration and the duration of the collection interval. The renal excretion rate of CP was calculated as the amount excreted throughout the duration of the collection period. Statistical analyses were performed using SAS (version 8.2, SAS Institute, Cary, North Carolina, USA). Mixed model analyses with CLR as a dependent variable were carried out, using subjects as random eVects and diVerent pharmacokinetic variables as covariates. The level of signiWcance was set at P < 0.05.

Results The LOD for CP in plasma and urine was 0.02 and 0.01 ng/ ml, respectively. The within-day precision of the developed methods was evaluated by analysing ten plasma and urine samples containing 1 ng of CP. The within-day precision for plasma and urine samples was determined to 4 and 2%, respectively. The between-day precisions of the methods were studied by analysis of duplicates containing 0.5, 1, 5, 50 and 100 ng/ml of CP. The duplicates were analyzed on three diVerent days within a period of a couple of weeks. For both plasma and urine methods the between-day precision were between 4–8%. Spiked plasma and urine samples containing 10 ng CP/ml were stored at room temperature during 24 and 48 h with no loss of CP. Spiked plasma and urine samples containing 10 ng CP/ml were also stored at ¡20°C during a week and a month without any losses of CP. Patient characteristics, including renal and liver tests, are presented in Table 1. The analyzed markers of hepatic and renal functions were within the range of normal, or close to normal. The administered doses of CP in absolute amounts ranged from 800 to 1,200 mg (mean 1,000 mg) for patients receiving the protocol of conventional dose and from 2,000 to 2,240 mg (mean 2,140 mg) for patients receiving the protocol of escalated dose. An overview of the CP doses for

123

Age (years)

52

38–61

Height (cm)

166

158–173

Weight (kg)

66

47–89

Body surface area (m2)

1.7

1.5–2.0 42–71

Plasma creatinine (mol/l)

56

Plasma bilirubin (mol/l)

12

6–24

Plasma bilirubin, conjugated (mol/l)

2

1–5

Plasma alkaline phosphatase (kat/l)

1.8

0.8–5.1

Plasma glutamyl transferase (kat/l)

0.9

0.3–6.3

Plasma aspartate aminotransferase (kat/l)

0.5

0.3–0.9

Plasma alanine aminotransferase (kat/l)

0.7

0.3–1.7

each patient is presented in Table 2. Also, their medications with other drugs during the study are shown. The measured parameters from each of the patients can be seen in Table 3. The Wnite time period of urine collection ranged between 175 and 330 (mean 240) min and urine volumes from 32 up to 840 ml were collected. Sampling occurred from 20 to 285 h after ended treatment of CP. The Table 2 Doses of CP that the participating patients received and other medications during the period of sample collection Patient no.

CP dose (mg)

1

1,200

2

990

Other medications

Flunitrazepam, ranitidine Prochlorperazine

3

1,010

4

2,208a

Lenograstim, levoXoxacin, prochlorperazine, tropisetron

Lorazepam, tropisetron

5

2,090a

Lenograstim, levoXoxacin, omeprazole, prednisolon, tropisetronb

6

2,000a

7

950

8

2,240a

Atenolol, levothyroxine – Furosemide, lenograstim, levoXoxacin, levothyroxine

9

980

Atorvastatin

10

950

Metoclopramide, tropisetron

11

800

CiproXoxacin, Xuconazole, lenograstim, nystatinb

12

2,170a

–

13

1,050

Felodipine

14

920

15

1,020

Paracetamol, prednisolone

16

1,150

–

a b

Escalated dose of CP Used naturopathic drugs

–b


289

Table 3 Measured parameters for each of the 16 participating patients Patient no.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

1st sampling period (h)a

46

47

23

22

24

24

20

92

95

51

93

68

162 167 167 165

Plasma concentration CP (ng/ml) 61

45

1,200

3,500

760

2,200

6,100

5.6

1.2

320

4.6

7.5

0.12 0.10 0.17 0.19

Urine concentration CP (ng/ml)

540 980 22,000 93,000 4,700 17,000 110,000 81

16

2,300 69

92

1.0

Urine volume (ml)

240 67

130

130

360

360

100

470 110 400

180 190 64

2nd sampling period (h)a

95

119

92

95

96

92

140 143 99

141 170 186 211 189 189

Plasma concentration CP (ng/ml) 0.76 0.80 0.38

1.1

1.2

0.83

2.2

1.4

0.54 6.7

0.61 0.34 0.09 0.04 0.11 0.16


3.8

12

9.0

7.4

30

21

1.3

6.3

0.87 1.0

1.5

2.6

Urine volume (ml)

240 510 130

260

190

200

160

310 840 470

130 400 210 75

33

71

3rd sampling period (h)a

166 167 170

140

170

144

164

188 191 171

190 212 263 236 262 285

Plasma concentration CP (ng/ml) 0.20 0.20 0.19

0.68

0.97

0.53

0.51

0.93 0.33 0.54

0.31 0.24 0.03 0.02 0.06 0.03


0.66 0.93 3.4

7.0

3.0

2.1

1.9

9.7

2.5

Urine volume (ml)

450 490 140

100

170

230

160

94 4.5

6.7

4th sampling period (h)a

1.4

46

1.8

310 390 210

3.6

1.6

1.8

3.8

330 32

150

0.29 0.65 0.62 0.44

300 400 410 180 45

260

400

259

Plasma concentration CP (ng/ml)

0.40

0.21


2.5

1.3

Urine volume (ml)

430

450

a

1.1

Time after ended treatment with CP to the blood sampling

plasma concentrations of CP ranged between 0.02 ng/ml and 6.1 g/ml and the urine concentrations between 0.29 ng/ml and 0.11 mg/ml. Plasma concentration at the same sampling time varied ninefold between individuals receiving the same doses of CP.

The calculated pharmacokinetic parameters for each patient are presented in Table 4. The calculated plasma drug concentrations of CP in the middle of the Wnite time period were between 0.02 ng/ml and 8.1 g/ml. Area under curve (AUC) and renal excretion rate of CP ranged between

Table 4 Calculated pharmacokinetic parameters for each of the 16 participating patients Patient no.

1

2

3

4

5

1st sampling period (h)a

46

47

23

22

24

Calculated plasma concentration (ng/ml) b

80

59

1,600

4,700

1,000

CLR (ml/min)

6.6

4.7

7.4

11

7.0

Urine Xow (ml/min)

0.98

0.28

0.55

0.55

1.5

2nd sampling period (h)a

95

94

119

92

Calculated plasma concentration (ng/ml)b

0.86

0.92

0.44

CLR (ml/min)

4.5

10

6

7

8

9

10

11

12

13

14

15

16

24

20

92

95

51

93

68

162

167

167

165

2,900

8,100

6.8

1.4

420

5.3

8.6

0.12

0.10

0.17

0.20

9.3

5.6

17

5.5

9.1

9.8

9.4

2.3

16

1.4

12

1.6

0.40

1.4

0.47

1.7

0.75

0.88

0.27

1.4

0.13

0.64

95

96

92

140

143

99

141

170

186

211

189

189

1.3

1.4

0.95

2.5

1.4

0.55

7.7

0.63

0.35

0.09

0.05

0.12

0.16

8.3

10

5.1

6.3

7.7

19

8.5

12

5.2

17

8.4

6.8

1.8

6.6 0.41

Urine Xow (ml/min)

1.0

2.1

0.54

1.1

0.77

0.81

0.65

1.3

3.5

2.0

0.53

1.7

0.89

0.31

0.14

3rd sampling period (h)a

166

167

170

140

170

144

164

188

191

171

190

212

263

236

262

285


0.20

0.21

0.19

0.70

1.0

0.55

0.53

0.96

0.34

0.56

0.32

0.25

0.03

0.02

0.06

0.03

CLR (ml/min)

6.1

9.0

10

4.7

2.0

3.7

2.3

13

6.6

2.9

9.5

11

16

20

1.8

24

Urine Xow (ml/min)

1.9

2.0

0.58

0.47

0.66

0.94

0.64

1.3

1.6

0.88

1.2

1.7

1.7

0.75

0.19

2.0

4th sampling period (h)a

260

259


0.41

0.22

CLR (ml/min)

11

11

Urine Xow (ml/min)

1.8

1.9

a b

Time after ended treatment with CP Concentration in the middle of the Wnite sampling period

123

290


5.7 min·ng/ml and 2.0 min·mg/ml and 0.12 ng/min and 51 g/min, respectively. The CLR and urine Xows were between 1.4 and 24 ml/min and 0.13 and 3.5 ml/min, respectively. There was no signiWcant dependence between CLR of CP and plasma concentration of CP (P value 0.82), and the intercept (tmean CLR) was 8.6 (conWdence interval 6.5–

10.7; Fig. 1). When the plasma drug concentration values above 10 ng/ml were excluded there was still no signiWcant dependence (P value 0.42) and the intercept was 8.3 instead. We have also used other cut-oV limits within our concentration interval but still no signiWcance was obtained. There was, however, a signiWcant dependence between the CLR of CP and the urine Xow (P < 0.01; Fig. 2).

Patient 1 Patient 2 Patient 3 Patient 4 Patient 5

20

Patient 6

Renal clearance (ml/min)

Patient 7 Patient 8 Patient 9 Patient 10 Patient 11 Patient 12 Patient 13

10

Patient 14 Patient 15 Patient 16 LOD plasma method LOD urine method Cancer risk assessment

0 0,001

0,01

0,1

1

10

100

1000

10000

Plasma concentration CP (ng/ml; log scale)

Fig. 1 Plasma concentrations of CP versus CLR of CP for 16 patients. Vertical lines showing the plasma concentrations corresponding to the LOD of the analytical methods for plasma and urine are included. Also a vertical line indicating the plasma concentration corresponding to a

cancer risk of up to 400 cancer cases per million during a working period of 40 years is included. The mean CLR was used to calculate the plasma concentration from urinary levels

30

Patient 1 25

Patient 2

Renal clearance (ml/min)

Patient 3 Patient 4 20

Patient 5 Patient 6 Patient 7 Patient 8

15

Patient 9 Patient 10 Patient 11

10

Patient 12 Patient 13 Patient 14

5

Patient 15 Patient 16 Equation

0 0

1

2 Urine flow (ml/min)

3

4

Fig. 2 Urine Xow versus CLR of CP for 16 patients. The line with the equation y = 4.7 + 3.8x, showing the dependence between CLR and urine Xow, is included

123


Discussion The major Wnding of this study was that we did not Wnd that CLR was dependent on the plasma drug concentration of CP. Since the proportion of the CP dose that was excreted in urine was constant, it is possible to continue to use the levels of CP in urine as a biomarker of occupational exposure to CP. However, a dependence between urine Xow and CLR was discovered and there was a large inter-individual variation in the plasma and urine concentrations even when the same doses were given. This should be considered when using CP in urine as a biomarker of exposure. The determination of CLR at low CP plasma levels required analyses of extremely low biological concentrations of CP. Thus, highly sensitive analytical methods for determination of CP in plasma and urine were developed and validated regarding LOD, precision and stability. The LODs in this study were lower compared to previous methods for determination of CP in plasma (Sadagopan et al. 2001; Liu et al. 2004) and urine (Sottani et al. 1998; Sannolo et al. 1999; Pethran et al. 2003; Kasel et al. 2004). The precisions of our methods were satisfactory and the stability of CP in the samples was high. With these methods it is possible to perform biological monitoring of health care workers involved in drug preparation and administration of CP. Since CP is classiWed as a carcinogen, it was not ethically justiWable to study the pharmacokinetics on healthy volunteers treated with low doses of CP. Instead, patients treated with CP were monitored until the concentrations of CP in urine and plasma were in the same low range as in occupationally exposed workers. A number of diVerent cancer diseases can be treated with CP, e.g., leukemia, breast cancer, lung cancer, ovarian cancer, lymphoma, etc. Health care workers handling CP are mostly women ranging between 20 and 65 years of age. The design of the study was chosen to mimic the real conditions and the investigated subjects were women 38–61 years old. It has been reported that several drugs interact with the CYP enzyme systems in the liver and thereby alter the pharmacokinetics of CP by either induction or inhibition (Xie 2004). However, all investigated patients except one were treated with drugs not known to alter these enzyme systems. Only one patient was treated with Xuconazole, a drug that might alter the pharmacokinetics of CP and suppress the CYP genes for the enzyme CYP3A4 and CYP2C9 (Marr et al. 2004; de Jong et al. 2005). Thus, the co-administration of drugs probably did not aVect the results of the study. Liver function may inXuence the metabolism of CP. Thus, if liver function is impaired, a smaller fraction of the CP dose is metabolized and a higher fraction of CP may be eliminated in the urine. The investigated patients were

291

treated with CP, epirubicin and 5-Xuorouracil one or several times before they participated in this study. These chemotherapy treatments may have an eVect on liver and renal function. From the result of the investigated markers of liver and renal function it was observed that some of the markers were slightly outside the normal range in a few of the patients. However, this small deviation was considered not to have an impact on the result. Blood samples were collected at the end of the sampling time period. It was not practicable to collect blood in the middle of the sampling period since the patients were investigated in their homes. Therefore, the plasma drug concentration in the middle of the sampling period was calculated. In the literature it has also been reported that there is a variability in the CP pharmacokinetics, both inter-individual and intra-individual variations of the CP pharmacokinetics has been described (Moore et al. 1991; Sladek 1994; Chen et al. 1995; Busse et al. 1997; Batey et al. 2002). We observed both inter- and intra-individual variations of CLR and urine Xow in this study. According to Xie (2004) the variation in CP pharmacokinetics may be a result of both environmental and genetic inXuences on the expression and activity of CP-metabolizing CYP enzymes. Previous pharmacokinetic studies of CP have reported mean CLR’s during the Wrst 24 h after treatment to be between 15 and 20 ml/min in patients treated with conventional doses of CP (Chen et al. 1995; Busse et al. 1997; Ren et al. 1998; Busse et al. 1999). During chemotherapy the patients may have received intensive hydration, and this might have eVected the CLR due to increased urine Xow and thereby an increased excretion of CP in urine (Chen et al. 1995). In a study made by Busse et al. (1999) high urine Xows, approximately 2–6 ml/min, were reported during 24 h after the dose of CP for patients treated with conventional doses. In this study CLR seems to be lower, probably due to the fact that the patient were studied between 20 and 285 h after the CP dose and were not intensively hydrated. The urine Xow was in the range of 0.13–3.5 ml/min (mean 1.1 ml/min) and values around 1 ml/min were considered to be normal. Urine Xow can only have a substantial eVect on CLR if a drug is mostly reabsorbed (Rowland and Tozer 1995). CLR of CP is thought to be predominantly the result of glomerular Wltration and tubular reabsorption since CLR of CP is much lower than the product of the fraction unbound and glomerular Wltration rate. However, it does not seem necessary to adjust for dilution in association with collection of spot urine samples with, e.g., creatinine in urine, since this substance also co-variates with the urine Xow. The calculated elimination half-life of 5 h for CP up to 60 h after treatment corresponds to the mean elimination half-life reported by Busse et al. (1997). However, after the

123

292

60 h, in this study, the elimination half-lives were considerably longer. Mean elimination half-lives were calculated instead of individual half-lives since we had only a few observations per subject. A biomarker of exposure should ideally give a measure of the internal dose of the substance that has an eVect in the body. In the case of CP it is the metabolite phosphamide mustard (PAM) that has antineoplastic activity. In a study by Joqueviel et al. (1998) urine from patients treated with CP was collected during one and two days after treatment. One day after treatment CP was primarily excreted in urine as unmetabolized CP (mean 16%), followed by carboxyphosphamide (CXCP; mean 10%) and dechloroethylcyclophosphamide (DCCP; mean 3%). Two days after the dose the excretion of CXCP dominated the urine (mean 22%), followed by CP (mean 15%) and DCCP (mean 5%). The mean excretion of PAM and PAM degradation products in urine one and two days after the dose was 0.3 and 0.6%, respectively. It is therefore most suitable to use CP in urine to assess the short-term exposure to CP. Thulin et al. (1996) developed a method for analysis of hemoglobin adducts of nornitrogen mustard, another metabolite of CP, but it is uncertain if this method is sensitive enough to measure occupational exposure to CP. In this study large individual variations in plasma and urine concentrations of CP can be seen although the same doses of CP were administered. Thus, the biomarker CP in urine cannot estimate the internal dose of CP in exposed workers with exactness since the doses either might be underestimated or overestimated. However, the biomarker reXects all exposure routes (dermal uptake, inhalation and ingestion). The urine concentration makes it possible to calculate the plasma concentration and since the distribution volume of CP is known (approximately 0.7 l/kg) the amount of CP in the body can be approximated (Sladek 1994). Also, the use of CP in urine as a biomarker at group level should be more accurate. From Fig. 1 it is evident that CP in urine is a more sensitive biomarker than CP in plasma, and would therefore be preferable for the monitoring of occupational exposure. The lowest concentration of CP that was measured in patient urine in this study was 0.29 ng/ml. With a urine production of 1–2 l per day this urine concentration corresponds to an excreted CP amount of 0.29–0.58 g CP per day. The cancer risk assessment for health care worker by Sessink et al. (1995) was based on extrapolation from both animal and patient data and much more detailed than the assessment by Sorsa and Anderson (1996). Sessink et al. estimated that a mean daily excretion of 0.18 g CP would give up to 400 cancer cases per million during a working period of 40 years. Thus, CLR in this study was investigated at urine and plasma concentrations similar to those of Sessink et al. A daily urinary excretion of 0.18 g CP

123


corresponds to a plasma concentration of 0.015 ng/ml (Fig. 1). An increase of up to 400 cancer cases per million in occupationally exposed personnel may be regarded as a low risk of cancer, but it cannot be neglected. However, it should be remembered that the risk assessment by Sessink et al. (1995) was based on exposure to CP only but in reality handling of antineoplastic drugs often involves several drugs with similar mechanisms of action, i.e. alkylating properties, and critical eVects (cancer, reproductive eVects). In summary, CLR of CP was not dependent on the plasma drug concentration and it will therefore be possible to continue to use CP in urine as a biomarker of occupational exposure to CP. With the developed analytical methods it will be possible to perform biological monitoring of CP in occupationally exposed workers. However, both the large inter-individual variability of excretion of unchanged CP in urine and the Wnding that the excretion of CP is dependent on urine Xow has to be considered in future applications where CP in urine is used as a biomarker of short-term exposure to CP to estimate the absorbed dose of CP. Acknowledgments We would like to thank Ms Eva Assarsson, Ms Else-Marie Åkerberg and Ms Inger Bensryd for their excellent medical skills and Ms Gertrud Wohlfart for her excellent and skilful laboratory assistance. Financial support for this study was obtained from the Swedish Council for Work Life and Social Research, the Swedish Research Council, the County Councils of Southern Sweden and the Medical Faculty at Lund University in Lund, Sweden.

References Bagley CM Jr, Bostick FW, DeVita VT Jr (1973) Clinical pharmacology of cyclophosphamide. Cancer Res 33:226–233 Bailey H, Mulcahy RT, Tutsch KD, Rozental JM, Alberti D, Arzoomanian RZ, Tombes MB, Trump DL, Wilding G (1991) A phase I study of SR-2508 and cyclophosphamide administered by intravenous injection. Cancer Res 51:1099–1104 Batey MA, Wright JG, Azzabi A, Newell DR, Lind MJ, Calvert AH, Boddy AV (2002) Population pharmacokinetics of adjuvant cyclophosphamide, methotrexate and 5-Xuorouracil (CMF). Eur J Cancer 38:1081–1089 Busse D, Busch FW, Bohnenstengel F, Eichelbaum M, Fischer P, Opalinska J, Schumacher K, Schweizer E, Kroemer HK (1997) Dose escalation of cyclophosphamide in patients with breast cancer: consequences for pharmacokinetics and metabolism. J Clin Oncol 15:1885–1896 Busse D, Busch FW, Schweizer E, Bohnenstengel F, Eichelbaum M, Fischer P, Schumacher K, Aulitzky WE, Kroemer HK (1999) Fractionated administration of high-dose cyclophosphamide: inXuence on dose-dependent changes in pharmacokinetics and metabolism. Cancer Chemother Pharmacol 43:263–268 Chen TL, Passos-Coelho JL, Noe DA, Kennedy MJ, Black KC, Colvin OM, Grochow LB (1995) Nonlinear pharmacokinetics of cyclophosphamide in patients with metastatic breast cancer receiving high-dose chemotherapy followed by autologous bone marrow transplantation. Cancer Res 55:810–816

Int Arch Occup Environ Health (2008) 81:285–293 Connor TH, Anderson RW, Sessink PJ, BroadWeld L, Power LA (1999) Surface contamination with antineoplastic agents in six cancer treatment centers in Canada and the United States. Am J Health Syst Pharm 56:1427–1432 de Jong ME, Huitema ADR, van Dam SM, Rodenhuis S, Beijnen JH (2005) EVects of co-medicated drugs on cyclophosphamide bioactivation in human liver microsomes. Anticancer Drugs 16:331–336 Ensslin AS, Stoll Y, Pethran A, Pfaller A, Rommelt H, Fruhmann G (1994) Biological monitoring of cyclophosphamide and ifosfamide in urine of hospital personnel occupationally exposed to cytostatic drugs. Occup Environ Med 51:229–233 Evelo CT, Bos RP, Peters JG, Henderson PT (1986) Urinary cyclophosphamide assay as a method for biological monitoring of occupational exposure to cyclophosphamide. Int Arch Occup Environ Health 58:151–155 Hedmer M, Georgiadi A, Rämme Bremberg E, Jönsson BAG, Eksborg S (2005) Surface contamination of cyclophosphamide packaging and surface contamination with antineoplastic drugs in a hospital pharmacy in Sweden. Ann Occup Hyg 49:629–637 Hirst M, Tse S, Mills DG, Levin L, White DF (1984) Occupational exposure to cyclophosphamide. Lancet 1:186–188 IARC (1981) Monographs on the evaluation of the carcinogenic risk of chemicals to humans, vol 26. Some antineoplastic and immunosuppressive agents. International Agency for Research on Cancer, Lyon IARC (1987) Monographs on the evaluation of the carcinogenic risk to humans, suppl 7. Overall evaluation of carcinogenicity: an updating of IARC monographs, vol 1–42. International Agency for Research on Cancer, Lyon Joqueviel C, Martino R, Gilard V, Malet-Martino M, Canal P, Niemeyer U (1998) Urinary excretion of cyclophosphamide in humans, determined by phosphorus-31 nuclear magnetic resonance spectroscopy. Drug Metab Dispos 26:418–428 Kasel D, Jetter A, HarlWnger S, Gebhardt W, Fuhr U (2004) QuantiWcation of cyclophosphamide and its metabolites in urine using liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom 18:1472–1478 Liu JJ, Kestell P, Findlay M, Riley G, Ackland S, Simpson A, Isaacs R, McKeage MJ (2004) Application of liquid chromatography– mass spectrometry to monitoring plasma cyclophosphamide levels in phase I trial cancer patients. Clin Exp Pharmacol Physiol 31:677–682 Marr KA, Leisenring W, Crippa F, Slattery JT, Corey L, Boeckh M McDonald GB (2004) Cyclophosphamide metabolism is aVected by azole antifungals. Blood 103:1557–1559 McDevitt JJ, Lees PS, McDiarmid MA (1993) Exposure of hospital pharmacists and nurses to antineoplastic agents. J Occup Med 35:57–60 Milsted RA, Jarman M (1982) Metabolism of high doses of cyclophosphamide. Cancer Chemother Pharmacol 8:311–313 Minoia C, Turci R, Sottani C, Schiavi A, Perbellini L, Angeleri S, Draicchio F, Apostoli P (1998) Application of high performance liquid chromatography/tandem mass spectrometry in the environmental and biological monitoring of health care personnel occupationally exposed to cyclophosphamide and ifosfamide. Rapid Commun Mass Spectrom 12:1485–1493 Moore MJ (1991) Clinical pharmacokinetics of cyclophosphamide. Clin Pharmacokinet 20:194–208 Parkinson A (2001) Biotransformation of xenobiotics. In: Klaassen CD (ed) Casarett and Doull’s toxicology: the basic science of poisons 6th edn. McGraw-Hill, New York Pethran A, Schierl R, HauV K, Grimm CH, Boos KS, Nowak D (2003) Uptake of antineoplastic agents in pharmacy and hospital personnel. Part I: monitoring of urinary concentrations. Int Arch Occup Environ Health 76:5–10

293 Ren S, Kalhorn TF, McDonald GB, Anasetti C, Appelbaum FR, Slattery JT (1998) Pharmacokinetics of cyclophosphamide and its metabolites in bone marrow transplantation patients. Clin Pharmacol Ther 64:289–301 Rowland M, Tozer TN (1995) Clinical pharmacokinetics. Concepts and applications. Lippincot, Williams & Wilkins, Philadelphia Roy P, Yo LJ, Crespi CL, Waxman DJ (1999) Development of a substrate-activity based approach to identify the major human liver P-450 catalysts of cyclophosphamide and ifosfamide activation based on cDNA-expressed activities and liver microsomal P-450 proWles. Drug Metab Dispos 27:655–666 Sadagopan N, Cohen L, Roberts B, Collard W, Omer C (2001) Liquid chromatography-tandem mass spectrometric quantiWcation of cyclophosphamide and its hydroxy metabolite in plasma and tissue for determination of tissue distribution. J Chromatogr B Biomed Sci Appl 759:277–284 Sannolo N, Miraglia N, Biglietto M, Acampora A, Malorni A (1999) Determination of cyclophosphamide and ifosphamide in urine at trace levels by gas chromatography/tandem mass spectrometry. J Mass Spectrom 34:845–849 Schmaus G, Schierl R, Funck S (2002) Monitoring surface contamination by antineoplastic drugs using gas chromatography–mass spectrometry and voltammetry. Am J Health Syst Pharm 59:956– 961 Sessink PJ, Anzion RB, Van den Broek PH, Bos RP (1992a) Detection of contamination with antineoplastic agents in a hospital pharmacy department. Pharm Weekbl Sci 14:16–22 Sessink PJ, Boer KA, Scheefhals AP, Anzion RB, Bos RP (1992b) Occupational exposure to antineoplastic agents at several departments in a hospital. Environmental contamination and excretion of cyclophosphamide and ifosfamide in urine of exposed workers. Int Arch Occup Environ Health 64:105–112 Sessink PJ, Van de Kerkhof MC, Anzion RB, Noordhoek J, Bos RP (1994) Environmental contamination and assessment of exposure to antineoplastic agents by determination of cyclophosphamide in urine of exposed pharmacy technicians: is skin absorption an important exposure route? Arch Environ Health 49:165–169 Sessink PJ, Kroese ED, van Kranen HJ, Bos RP (1995) Cancer risk assessment for health care workers occupationally exposed to cyclophosphamide. Int Arch Occup Environ Health 67:317–323 Sladek NE (1994) Metabolism and pharmacokinetic behavior of cyclophosphamide and related oxazaphosphorines. In: Powis G (ed) Anticancer drugs: reactive metabolism and drug interactions. Pergamon Press, Oxford Sorsa M, Anderson D (1996) Monitoring of occupational exposure to cytostatic anticancer agents. Mutat Res 355:253–61 Sottani C, Turci R, Perbellini L, Minoia C (1998) Liquid–liquid extraction procedure for trace determination of cyclophosphamide in human urine by high-performance liquid chromatography tandem mass spectrometry. Rapid Commun Mass Spectrom 12:1063– 1068 Thulin H, Zorcec V, Segerbäck D, Sundwall A, Törnqvist M (1996) Oxazolidonylethyl adducts to hemoglobin and DNA following nornitrogen mustard exposure. Chem Biol Interact 99:263–275 Turci R, Sottani C, Ronchi A, Minoia C (2002) Biological monitoring of hospital personnel occupationally exposed to antineoplastic agents. Toxicol Lett 134:57–64 Wick C, Slawson MH, Jorgenson JA, Tyler LS (2003) Using a closedsystem protective device to reduce personnel exposure to antineoplastic agents. Am J Health Syst Pharm 60:2314–2320 Xie HJ (2004) Pharmacogenetic and pharmacokinetic studies of cyclophosphamide. In cell, animal and human. Karolinska Institutet at Karolinska University Hospital, Stockholm

123