Impact of low- and high-risk operators handling irinotecan on the blood contamination of health care workers in oncology day care units

Abstract

Introduction

Health care workers handling antineoplastic drugs (ADs) are at risk of mutagenicity and adverse reproductive effects. Despite protective equipment and AD handling guidelines, AD levels are still detected in caregivers in oncology units. This study attempted to assess blood contamination by irinotecan and its metabolites in all health care workers in oncology day hospital units according to activities specific to each employment category.

Methods

The study was performed at two different hospitals: a university hospital and a comprehensive cancer centre. Forty-four participants were categorized according to their daily activity as a high-risk operator (29 nurses/ward aides and 5 cleaning staff) and a low-risk operator (7 doctors and 3 secretaries). The collected blood samples were subjected to UHPLC–MS/MS. The plasma and red blood cell (RBC) levels of irinotecan and its metabolites (SN-38; APC) were determined using a validated analytical method detection test.

Results

Two hundred sixty-four assay results were collected (132 plasma results and 132 RBC results). The comparison between low- and high-risk operator-contaminated workers was not significant (18.33% positive results in low-risk operators vs. 25.98% positive results in high-risk operators; P = 0.22). This homogeneity showed overall contamination within the unit. Positive results were obtained in 21.43% of physicians, 11.11% of secretaries, 25.86% of nurses/ward aides and 26.67% of cleaning staff. These results could be explained by the lack or failure of personal and collective protective equipment. A lack of protection and inadequate decontamination procedures can result in surface contamination.

Conclusions

This study evaluated blood contamination with irinotecan and its metabolites in health care workers from day hospital care units. Among the 24.24% of contaminations observed in care units, the difference between low- and high-risk operator contamination was not significant (P = 0.22). The impact on blood contamination found is the same between low- and high-risk caregivers. This implies that the protective precautions associated with the handling of anticancer drugs must therefore be followed by all staff, including those believed to be at low risk of exposure.

Keywords

Antineoplastic drugs blood contamination occupational exposure pharmacy personnel

Introduction

Health care workers handling antineoplastic drugs (AD) are at risk of mutagenicity and reproductive effects and developing cancer.¹ Chronic occupational exposure of health care workers can be considered.¹ Despite protective equipment and guidelines on handling ADs, AD levels are still detectable in workers’ urine samples.¹ A meta-analysis published in 2017 revealed a significant association between occupational exposure to ADs during a typical workday and increases in chromosomal aberrations in health care workers.² These findings indicate the need to limit the occupational exposure of health care workers to ADs.²

Contamination with ADs is determined at each step of the medication process,³ implying that many occupational categories are potentially exposed because chemotherapy is not restricted only to oncology units.³ Immediate and contributing causes of workers’ exposure to ADs were revealed in a 2016 study that showed that occupational exposure was likely preventable through appropriate control measures.⁴ Safety measures have been progressively improved to reduce workers’ exposure by the application in health care settings of the safe handling of cytotoxic practice guidelines by the Occupational Safety and Health Administration OSHA in 1986 and updated in 2012,⁵ the National Institutional for Occupational Safety and Health published in 2004 and updated in 2004,⁶ the International Society of Oncology Pharmacy Practitioners in 2007⁷ and more recently by the American Society of Health-System Pharmacists in 2018.⁸

However, several studies have suggested that workers are still exposed to ADs even with a decrease in surface contamination following regular environmental monitoring, and a low level of contamination has already been found.^9,10 Workers’ contamination is assessed by the biological monitoring of workers using indirect methods such as the Ames test,¹¹ but these methods are limited in terms of specificity and sensitivity.¹² Furthermore, occupational exposure is mostly assessed by urinary dosage, which has many limitations: urine samples are collected after 24 h, and the long plasma half-life of irinotecan causes urinary detection to be more uncertain and lower.¹² Many data are missing to claim or deny a risk to the exposed professional.^1–3,12

Irinotecan is a potent inhibitor of topoisomerase and has been used as a first-line or second-line chemotherapeutic agent in several malignancies, particularly colorectal cancer.¹³ Irinotecan is classified as a carcinogenic agent in Group 3 of the International Agency for Research on Cancer (IARC) classification.¹⁴ Irinotecan is a good candidate to assess health care contamination for different reasons. First, it is specific to hospital contamination, in contrast to platinum compounds. Second, irinotecan has a relevant half-life for blood sampling, in contrast to more commonly used ADs such as 5-fluorouracil. Irinotecan shows a three-phase elimination profile. The mean plasma half-life of the first phase in the three-phase model is 12 min, while that of the second phase is 2.5 h, and the terminal half-life is 14.2 h.¹⁵ Following the first evaluation of plasma and red blood cell contamination with irinotecan and its metabolites in a centralized cytotoxic pharmacy unit, 22.20% of assays were positive in pharmacy staff in a study evaluating and comparing the levels of contamination in day hospitalization oncology units.^16,17

The relationship between the exposure dose and toxicity is difficult to assess. Regarding these elements, any exposure must be followed, even at ultratrace concentrations throughout professional life, justifying the need to simultaneously determine traces of chemotherapies both in the working environment and in the caregivers’ body.

This study aimed to assess caregiver and noncaregiver worker blood contamination by irinotecan and its metabolites in two different oncology day hospitalization units according to the activities specific to each employment category.

Methods

Study setting

This descriptive study was part of a larger study assessing AD contamination. The study was performed at two different hospitals: a French university hospital and a comprehensive cancer centre located in the same city. The French university hospital comprises a day hospitalization gastroenterology oncology unit of 10 beds. The cancer centre is made up of a day hospitalization oncology unit of 49 beds. The pharmaceutical unit of the French Comprehensive Cancer Centre provides on average 31,000 preparations per year with two face-to-face dual-station isolators in overpressure (SIEVE, Villeurbanne, France; date of commissioning: March 2007) dedicated to AD preparation. The centralized pharmacy unit of the French university hospital for cytotoxic drug preparations provides on average 27,000 preparations per year with brand flow-through isolators in overpressure, a single-station isolator and a dual-station isolator face to face (SIEVE, Villeurbanne, France; date of commissioning: July 2018). Neither hospital uses a closed system.

The health care workers in these oncology day hospitalization units and centralized cytotoxic pharmacy units participated after providing consent. Several professional categories were represented in these units (physicians, nurses, ward aides, secretaries and cleaning staff). Two groups were created based on the risks associated with the handling of chemotherapy. Using a rating table for daily activities, it is possible to distinguish high-risk operators (29 nurses/ward aides and 5 cleaning staff) and low-risk operators of exposure to AD (7 physicians and 3 secretaries) (Figure 1). Waste management and the administration of chemotherapies were classified as high-risk tasks, and cleaning staff and nurses were classified as high-risk operators.

Figure 1.

Risk rating table associated with AD contact by occupational categories for low- and high- risk operators.

Blood collection

This study was validated by the Committee for the Protection of Persons concerned (CPP) North-West III with the number A16-D49-VOL.30. A general presentation of the study to the team was offered, and then we collected consent from the volunteers. Collective feedback of the results to the team was presented after each investigation, followed by an individual consultation by the occupational physician.

The collection of blood samples occurred an average of 20 h after the manipulation of irinotecan. The sample time was selected to allow a good analysis of the potential systemic passage of irinotecan, considering the pharmacokinetic characteristics of irinotecan and its two metabolites: SN-38 and APC.¹⁵ The blood samples collected in a coagulation citrate tube (BD Vacutainer 9NC, reference: 367704, Plymouth, UK) were centrifuged for 10 min at 3.500 rpm in a centrifuge (CR31; Thermo Electron, Villebon sur Yvette, France) immediately after sampling and were stored at −20 °C until analysis.

Drug analysis

Drug analysis was performed independently from data collection with anonymous identification. The method was developed using total human blood kindly supplied by the French Blood Bank (Etablissement Français du Sang Bois-Guillaume, Rouen, France). The blood samples were centrifuged to separate the plasma from RBCs. For both plasma and RBCs, the first step was adding topotecan as an internal standard (IS). Topotecan, which has a structure and physicochemical characteristics similar to irinotecan, was chosen as an internal standard. Topotecan also has a structure close to the two other compounds studied: SN-38 and APC. It was difficult to obtain standards marked with stable isotopes (14C, 15N, etc.). This method uses an internal standard (topotecan) and identification criteria specified by the European Commission Decision 2002/657/EC. The method satisfies all of the requirements listed in the EMA and FDA guidelines.^18,19

The second step was precipitating the proteins. The supernatants were then diluted in water to reach an acetonitrile content compatible with solid phase extraction (SPE). Irinotecan, its metabolites and the IS were then extracted from the sample using an SPE procedure and Oasis HLB cartridges. After a drying step, the sample was dissolved in 20 μL of mobile phase before injection into a Nexera X2 UHPLC interfaced with an electrospray triple quadrupole mass spectrometer (LCMS-8030Plus; Shimadzu) used in the multiple reaction monitoring (MRM) acquisition mode after positive ionization. Two MRM transitions from the fragmentation of the [M + H] + ion were recorded for each compound. The analytical method used to detect the presence of irinotecan, the SN-38 active metabolite and the APC inactive metabolite in plasma and red blood cells (RBCs), was highly sensitive, with a limit of detection (LOD) of 2.5 pg/ml and a limit of quantification (LOQ) of 50 pg/ml, and was validated following the bioanalytical method validation guidance of the European Medicines Agency and Food and Drug Administration.^18,19 The detection levels were denoted as one when the result was <2.5 pg/ml, two when the result was between 2.5 and 50 pg/ml, and 3 when the result was >50 pg/ml.

Statistical analysis

Categorical variables were expressed as numbers and percentages and were compared with the chi-squared test The test was two-tailed, and a p-value less than 0.05 was considered statistically significant. The statistical analysis was performed using IBM®-SPSS® Statistics 22.0 software (IBM Corp. 2013. IBM SPSS Statistics for Windows, Version 22.0 Armonk).

Results

Characteristics of the participants

During this descriptive study, all of the participants worked in oncology day hospitalization. Forty-four participants representing five different occupational categories (29 nurses/ward aides, 7 physicians, 5 cleaning staff and 3 secretaries) had an M/F sex ratio of 0.16 and a median age of 39.5 years. Twenty-nine participants worked in the oncology unit of the comprehensive cancer centre, and 15 participants worked in the oncology unit of the teaching hospital. None of the participants received anticancer drugs for personal reasons nor did their family or pets.

Plasma and red blood cell contamination with irinotecan and its two major metabolites in low-risk operators

Two hundred sixty-four assay results were collected (132 plasma results and 132 RBC results) (Table 1). Thirty-nine/132 (29.55%) assays in plasma and 25/132 (18.94%) assays in RBCs were positive.

Table 1.

Plasma and red blood cell contamination with irinotecan and its two major metabolites.

Low-risk operators	Plasma			RBC
Low-risk operators	Irinotecan (pg/mL)	SN-38 (pg/mL)	APC (pg/mL)	Irinotecan (pg/mL)	SN-38 (pg/mL)	APC (pg/mL)	P
a) A low-risk operators.
Physician n = 7
Number of assays < LOD	5	6	4	6	6	6
LOD< number of assays <LOQ	2	1	1	1	1	1
Number of assays >LOQ	0	0	2	0	0	0
Median of positive assays, n (%)	6/21 (28.57%)			3/21 (14.29%)			0.45
Secretary n = 3
Number of assays < LOD	3	3	2	3	3	2
LOD <number of assays <LOQ	0	0	1	0	0	1
Number of assays >LOQ	0	0	0	0	0	0
Median of positive assays, n (%)	1/9 (11.11%)			1/9 (11.11%)			1.00
Total number low-risk operators, n (%)	7/30 (23.33%)			4/30 (13.33%)			0.51
b) A high-risk operators.
	Plasma				RBC
High-risk operators	Irinotecan (pg/mL)	SN-38 (pg/mL)	APC (pg/mL)	Irinotecan (pg/mL)	SN-38 (pg/mL)	APC (pg/mL)	P
Nurse/Ward aide n = 27 + 2
Number of assays < LOD	19	27	15	23	22	23
LOD< number of assays <LOQ	5	2	14	6	7	6
Number of assays >LOQ	5	0	0	0	0	0
Median of positive assays, n (%)	26/87 (29.89%)			19/87 (21.84%)			0.23
Cleaning staff n = 5
Number of assays < LOD	2	5	2	5	5	3
LOD< number of assays <LOQ	2	0	3	0	0	2
Number of assays > LOQ	1	0	0	0	0	0
Median of positive assays, n (%)	6/15 (40.00%)			2/15 (13.33%)			0.22
Total number high-risk operators, n (%)	32/102 (31.37%)			21/102 (20.59%)			0.08

For low-risk operators, 7/30 (23.33%) in plasma and 4/30 (13.33%) in RBC assays were positive, above the LOD. Concerning physicians, 28.57% of assays were positive in plasma and 14.29% in RBCs. Among these positive results, two assays were quantified in plasma (>50 pg/ml). The secretary pool contained the same number of positive assays in plasma and RBCs (11.11%).

Plasma and red blood cell contamination with irinotecan and its two major metabolites in high-risk operators

For high-risk operators, 32/102 (31.37%) plasma assays and 21/102 (20.59%) RBCs assays were positive above the LOD (Table 1). Concerning the nurse/ward aide staff, 29.89% of plasma assays and 21.84% RBCs assays were positive. Five positive assays were quantified (>50 pg/ml, including one of 76 pg/ml). For cleaning staff, 40.00% and 13.33% of assays were positive in plasma and RBCs, respectively.

Proportion of positive blood contamination results among low- and high-risk operators

Figure 2 shows that for low-risk operators, 21.43% of physicians had positive results (in either plasma or RBCs), and 11.11% of secretaries had positive results. Considering high-risk operators, 25.86% of nurses/ward aides had positive results, and 26.67% of cleaning staff had positive results. Low-risk operators had 18.33% positive results, and high-risk operators had 25.98% positive results (P = 0.22).

Figure 2.

Proportion of positive blood contamination results among low- and high- risk operators.

Discussion

To investigate blood health care occupational contamination with irinotecan, it was essential to follow the entire process from preparation to elimination. The first stage of preparation in the compounding unit has already been studied.¹⁶ This study in a centralized cytotoxic pharmacy unit of a university hospital showed blood contamination of 22.22% of the pharmacy staff.¹⁶ Next, irinotecan administration and waste management were evaluated in the day hospitalization unit and revealed irinotecan contamination of 24.24% of staff. Notably, these results were similar between the care unit and the compounding unit, although caregivers in the care unit did not directly handle irinotecan, and individual and collective protective measures were reduced compared with those of the compounding unit. Conversely, the compounding unit handled ADs more regularly but deployed more effective protective measures appropriate to the risk involved, including isolators. These similar results could be explained by personal protective equipment (PPE) being deployed according to the perceived risk.

Among the 24.24% of contaminations observed in care units, the comparison between low- and high-risk operator contamination was not significant (P = 0.22) (Figure 2). This unexpected homogeneity showed overall contamination within the unit, warranting reassessment of the means to protect both high- and low-risk operators. However, distinguishing low- and high-risk operators is necessary because their daily practices are not the same and the causes of contamination are very different.

We demonstrated a contamination rate of 25.98% in high-risk operators (Figure 2). Among these high-risk operators, cleaning staff were included among the high-risk operators because waste management is one of the tasks most at risk of contamination (Figure 1).³ Cleaning staff were the most contaminated, with 40% having positive results in the plasma. These results could be due to cleaning procedures that could be neglected or inadequate. Cleaning staff among occupational categories are often the least considered if we consider, for instance, their inclusion in studies on exposure and contamination by ADs. They are often not included in training sessions, limiting their opportunities for learning about the risks involved in handling chemotherapy.²⁰ An assessment of AD contamination levels before and after cleaning showed that current cleaning practices might not be effective in removing residual traces of drugs from surfaces that would be at risk of contamination.²¹ Thus, standard decontamination procedures should be reconsidered.²²

Considering nurses/ward aides, the level of contamination was worrisome (25.86%) (Figure 2). Nurses were included among the high-risk operators because chemotherapy administration is considered a high-risk task (Figure 1).^3,23 Nurses set up chemotherapy and perform repositioning of the equipment during administration. There is thus a repetition of contact with the chemotherapy and a risk of chronic occupational exposure. The maximum concentration observed in this study was found in a nurse at 76 pg/ml in plasma. Some nurses are not sufficiently aware of the risks associated with occupational exposure to ADs.²⁴ Safe handling practices are not always followed by nurses who administer ADs, although guidelines have existed for some time.²⁵

Notably, nurses’ practices in a day hospital unit differed from practices in a specialized cancer unit. The diverse activities performed in the day hospital unit seemed to be associated with an increased risk of contamination. Conversely, a specialized oncology unit has a main activity related to AD administration, likely explaining their lower blood contamination. In the cancer centre, the staff receive continuous and in-depth training in oncology.

Notably, low-risk operators had a contamination rate of 18.33% in plasma and RBCs. The tests were positive in 21.43% of the physicians. Physicians’ activities are at a reduced risk compared with administration or waste management because their contact with chemotherapy is reduced. Their main risks of contamination are caused by surface contamination in the care units.²⁶ Physicians’ professional interactions with these high-risk operators could affect these results. For example, a nurse who has handled chemotherapy could dissipate hazardous drugs by spreading them throughout the surfaces in the unit.²³

Interestingly, secretaries who were not involved in chemotherapy had 11.11% positive results (Figure 2). The reason could be due to surface contamination in the care units and also that they do not wear protective equipment.²⁶ The dermal route was shown to be the main route of contamination.²³

In our study, the contamination of both high- and low-risk health workers suggests the need to regularly monitor their exposure threshold through blood tests. Thus, caregivers could be alerted quickly to problematic findings so that corrective actions can be taken, such as chemical decontamination.²² This prevention would help to reduce the mutagenic risks for health care workers.²⁰ Adverse effects caused by ADs are heterogeneous according to profession.²⁷ Additionally, no health care workers should be excluded from the training sessions, which are essential to obtain knowledge about the risks and good practices. Awareness of the risk will allow the staff to be at a lower risk of contamination.

This work has some limitations. First, our bicentric study should be shared with other regions and countries for further development. Our observational study has a long-term goal of improving the health of caregivers working in contact with ADs. Thus, blood contamination should be assessed on a larger scale to gain more insight. This work brings together a general day hospitalization unit and a specialized cancer unit to collect contamination data. The diverse risk activities of nurses in a day hospitalization unit differ from the practices in a specialized cancer unit.

Conclusion

This work evaluated blood contamination with irinotecan and its metabolites in health care workers from two day hospital units at two different centres. This contamination could be explained by the lack or failure of PPE and CPE, unfollowed procedures or unassimilated training. These failures can cause surface contamination, thus exposing health care workers who do not handle ADs to contamination. These defects can be explained by the poor perception and knowledge of the risks associated with the handling of ADs.

The assessment of blood contamination should be extended to home nurses. Their handling practices, resources, perception and knowledge available would differ from the hospital environment.

Footnotes

Acknowledgements

The authors are grateful to all of the participants involved in the present article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

H. Benoist

R. Delépée

G. Saint-Lorant

References

Ursini

Omodeo Salè

Fresegna

, et al. Antineoplastic drug occupational exposure: a new integrated approach to evaluate exposure and early genotoxic and cytotoxic effects by no-invasive buccal micronucleus cytome assay biomarker. Toxicol Lett 2019; 316: 20–26.

Roussel

Witt

Shaw

, et al. Meta-analysis of chromosomal aberrations as a biomarker of exposure in healthcare workers occupationally exposed to antineoplastic drugs. Mutat Res 2019; 781: 207–217.

Hon

Teschke

Chua

, et al. Occupational exposure to antineoplastic drugs: identification of job categories potentially exposed throughout the hospital medication system. Saf Health Work 2011; 2: 273–281.

Hon

Abusitta

. Causes of health care Workers’ exposure to antineoplastic drugs: an exploratory study. Can J Hosp Pharm 2016; 69: 216–223.

Occupational Safety and Health Administration . Controlling occupational exposure to hazardous drugs. Occupational safety and health administration. Am J Health Syst Pharm 1996 15; 53:1669–1685.

National Institute for Occupational Safety and Health (NIOSH). Preventing Occupational Exposure to Antineoplastic and Other Hazardous Drugs in Health Care Settings. https://www.cdc.gov/niosh/docs/2004-165/pdfs/2004-165.pdf?id = 10.26616/NIOSHPUB2004165, (2004, accessed 2 December 2020).

International Society of Oncology Pharmacy Practitioners Standards Committee. ISOPP Standards of practice. Safe handling of cytotoxics. J Oncol Pharm Practice 2007; 13 :1–81.

Power

Coyne

. ASHP Guidelines on handling hazardous drugs. Am J Health Syst Pharm 2018; 75: 1996–2031.

Palamini

Floutier

Gagné

, et al. Evaluation of decontamination efficacy of four antineoplastics (ifosfamide, 5-fluorouracil, irinotecan, and methotrexate) after deliberate contamination. J Occup Environ Hyg 2021; 18: 43–50.

10.

Ziegler

Mason

Baxter

. Occupational exposure to cytotoxic drugs in two UK oncology wards. Occup Environ Med 2002; 59: 608–612.

11.

Ames

Mccann

Yamasaki

. Methods for detecting carcinogens and mutagens with the Salmonella/ mammalian-microsome mutagenicity test Mutat Res 1975; 31: 347–364.

12.

Izzo

Charlier

Bloise

, et al. A UHPLC-MS/MS-based method for the simultaneous monitoring of eight antiblastic drugs in plasma and urine of exposed healthcare workers. J Pharm Biomed Anal 2018; 154: 245–251.

13.

Vanhoefer

Harstrick

Achterrath

, et al. Irinotecan in the treatment of colorectal cancer: clinical overview. J Clin Oncol 2001; 19: 1501–1518.

14.

International Agency for Research on Cancer (IARC), Monographs on the evaluation of Carcinogenic Risk to Humans. Complete List of Agents evaluated and their classification, https://monographs.iarc.who.int/wp-content/uploads/2018/06/mono99.pdf, 2010. (2010, accessed 2 December 2020)

15.

Chabot

. Clinical pharmacokinetics of irinotecan. Clin Pharmacokinet 1997; 33: 245–259.

16.

Benoist

Breuil

Le Neindre

, et al.

Does equipment change impact blood contamination with irinotecan and its two major metabolites in a centralized cytotoxic pharmacy unit?

J Oncol Pharm Pract 2020; 26: 1823–1828.

17.

Bechet

Benoist

Beau

, et al. Blood contamination of the pharmaceutical staff by irinotecan and its two major metabolites inside and outside a compounding unit. J Oncol Pharm Pract 2021; 4: 10781552211012059.

18.

Guideline-bioanalytical-method-validation_en.pdf [Internet]. [cited 20 mars 2021]. https://www.ema.europa.eu/en/documents/scientific-guideline/guideline-bioanalytical-methodvalidation_en.pdf.

19.

Booth

Arnold

DeSilva

, et al. Workshop report: crystal city V--quantitative bioanalytical method validation and implementation: the 2013 revised FDA guidance. AAPS J 2015; 17: 277–288.

20.

Kieffer

Verhaeghe

Lagrassa

, et al. Preventing the contamination of hospital personnel by cytotoxic agents: evaluation and training of the para-professional healthcare workers in oncology units. Eur J Cancer Care 2015; 24: 404–410.

21.

Chu

Hon

Danyluk

, et al. Pilot assessment of the antineoplastic drug contamination levels in British columbian hospitals pre- and post-cleaning. J Oncol Pharm Pract 2012; 18: 46–51.

22.

Simon

Odou

Decaudin

, et al. Chemical decontamination of hazardous drugs: a comparison of solution performances. Ann Work Expo Health 2020; 64: 114–124.

23.

Fransman

Vermeulen

Kromhout

. Dermal exposure to cyclophosphamide in hospitals during preparation, nursing and cleaning activities. Int Arch Occup Environ Health 2005; 78: 403–412.

24.

Hon

Teschke

Shen

. Health care Workers’ knowledge, perceptions, and behaviors regarding antineoplastic drugs: survey from British Columbia, Canada. J Occup Environ Hyg 2015; 12: 669–677.

25.

Kim

Lee

Jung

, et al. Korean Nurses’ adherence to safety guidelines for chemotherapy administration. Eur J Oncol Nurs 2019; 40: 98–103.

26.

Palamini

Gagné

Caron

, et al. Cross-sectional evaluation of surface contamination with 9 antineoplastic drugs in 93 Canadian healthcare centers: 2019 results. J Oncol Pharm Pract 2020; 26: 1921–1930.

27.

Combes

Barre

Duche

, et al. In vitro binding and partitioning of irinotecan (CPT-11) and its metabolite, SN-38, in human blood. Invest New Drugs 2000; 18: 1–5.