Contamination control and preparation of anticancer drugs

Department of Pharmacy,

National Cancer Center Hospital,

Tokyo, Japan

E-mail: [email protected]

Many anticancer drugs are mutagenic, teratogenic and/or carcinogenic. Healthcare workers who continuously handle various anticancer drugs are likely to be chronically exposed to these drugs, which could affect their health. For these reasons, it is extremely important to ensure the safety of health care workers by assessing their exposure to anticancer drugs during drug preparation and establishing appropriate methods of preparation. Exposure to anticancer drugs is most likely to occur during drug preparation. Figure 1 shows the leakage that occurred when sodium fluorescein (FL) was used as a model anticancer drug to simulate drug preparation using standard needle and syringe techniques. Guidelines recommend that appropriate preparation equipment should be used to prevent exposure to anticancer drugs,(1–3) and it has been reported that closed system drug transfer devices (CSTDs) are useful for this purpose.(4,5)

We have created two anticancer drug model preparation systems using FL and cyclophosphamide (CP) and conducted a comparative investigation of chemical contamination between conventional preparation (using needles and syringes) and preparation using CSTDs (the ChemoClave® Oncology System and BD PhaSeal™ System).(6) All manipulations were performed in a Class II biological safety cabinet (BSC), simulating actual working conditions as closely as possible.

One gram of powder containing FL at a concentration of 0.5% w/w that was adjusted using D-mannitol was sealed in a 20-ml glass vial, and this preparation was used as the test anticancer drug. The test prescription was 11mg FL in an infusion bottle containing 100ml saline. Contamination checks were performed for the work surface, side walls and glass barrier of the BSC, and the surfaces of rubber caps or joining sections of the vials, infusion bottles and needle. The scattering of the fluorescent test drug was measured in terms of the number and size of droplets under UV irradiation.

There was no contamination observed in any observed parts when the BD PhaSeal™ system was used irrespective of the preparation experience of the worker (Table 1). However, in the preparation using the ChemoClave® Oncology System, all vials, most infusion bottles and needles, but not the surface of safety cabinets, were contaminated. The size of the scattered droplets in all observed parts of conventional preparation varied from <1 to >5mm. Contamination over a wide range with droplets ≥5mm was only observed with conventional preparation; however, when the ChemoClave® Oncology System was used, most scattered droplets were 1mm or less in size in all observed parts.

For the test prescription, 800mg CP was mixed and prepared in an infusion bottle containing 100ml saline. A vial of 500mg CP was dissolved in saline to a concentration of 20mg/ml, and the solution was used for the trial. To eliminate the influence of CP on the surface of the vial, the CP vial was washed with 70% ethanol and water before use. The wipe samples were obtained from the surfaces of rubber caps and joints of the vials, infusion bottle, needle, palm sides of the left- and right-hand gloves, and working surface of the BSC. The wipe samples were analysed using liquid chromatography–electrospray tandem mass spectrometry. Furthermore, to investigate whether contamination could be eliminated by cleaning, cotton soaked well with 70% ethanol was used to wipe the surface of the infusion bottle and infusion adaptor once after the conclusion of the mixing operation, and this second wipe sample was collected and analysed.

When conventional preparation was performed using CP model vials, the maximum values of contamination were significantly higher in all observed parts than those observed when the CSTDs were used (Table 2). The maximum value on the BSC working surface, which is regarded as an indication of environmental exposure, was 514.9ng for conventional preparation. Conversely, the maximum value was 0.3ng for the ChemoClave® Oncology System and 0.5ng for the BD PhaSeal™, illustrating a significant difference between conventional preparation and preparation using CSTDs.

Furthermore, in measurements for the BSC work surface, CP was detected in 40% of measured samples for conventional preparation, whereas the levels of contamination for 90% of the measured samples were at or below the limit of detection for both devices.

Although CP was detected on the surfaces of the joining parts for all equipment when the ChemoClave® Oncology System was used, the concentration was lower than that observed with conventional preparation. However, when the BD PhaSeal™ System was used, CP was detected on the surfaces of vials, infusion bottles, and needles, in contrast to the test results of the fluorescence observation. However, the amounts measured were extremely small relative to those observed with the ChemoClave® Oncology System.

When the surface was cleaned with 70% ethanol after preparation, the median concentrations of CP on the surfaces of rubber caps and joints of the infusion bottle after wiping were 2.5ng (0.4–100.1ng) for conventional preparation, 1.1ng (0.2–254ng) for the ChemoClave® Oncology System, and 0.5ng (0.1–2.0ng) for the BD PhaSeal™ System.

While performing conventional preparation, contamination was observed in all examined parts for tests when FL was used. This was irrespective of the number of years of preparation experience. Contamination was observed at high frequencies and concentrations even for workers with many years of preparation experience, and even when preparation was performed using standard preparation techniques according to guidelines. These results suggest that the prevention of contamination for conventional preparation methods using needles and syringes is extremely difficult. Additionally, regarding the preparation method using needles and syringes, the observation of high frequencies and concentrations of contamination with anticancer drugs on the surface of infusion bottles suggests the potential for contamination to spread from the surface of the infusion bottles to outside the BSC and administration area. The high efficiency of 70% ethanol in removing contamination on the surface of infusion bottles suggests that cleaning the surface of infusion bottles using cotton or gauze soaked with 70% ethanol is useful for eliminating anticancer drug contamination that occurs during preparation. However, it is difficult to completely remove high levels of contamination, and so, wiping has limited effectiveness.

In tests using the ChemoClave® Oncology System, both FL and CP were detected on the surface of the equipment. In this system, the pipes are closed when removing equipment joints, thereby preventing the leakage of chemicals. There is a possibility that small amounts of residual chemicals inside the pipe on the joint surface side may be dispersed when dismantling the joints. However, chemicals did not scatter to other parts, and the contamination on the BSC work surface was similar to that observed using BD PhaSeal™ System. In tests with this system, there was no scattering of chemicals on the surface of equipment and BSC for tests using FL, and the results of tests using CP revealed that very low amounts of CP were detected on the surface of equipment, illustrating that the BD PhaSeal™ System is highly effective in preventing unwanted drug exposure.

As the number of cancer patients increases, so will the number of cases of anticancer drug preparation. Although the cumulative effect of exposure to several anticancer drugs over long periods is unclear, exposure to hazardous drugs should be prevented and reduced whenever possible. In this study, and a previous study, needles and syringes used in conventional preparation methods have the potential to release drug solutions into the work environment. Because CP was detected on the equipment surface for both closed system devices, caution is required for handling of equipment when high concentration anticancer drugs are used, even if these CSTDs are used. The National Institute for Occupational Safety and Health (USA) guidelines state that that CSTDs are useful for preventing exposure, but they do not replace the BSC.

Moreover, personal protection equipment needs to be worn, and appropriate preparation methods need to be employed.(1) The two types of CSTDs used in this experiment reduced the exposure to anticancer drugs. As the structure and characteristics of these drug transfer devices differ, it is believed that the equipment to be used should be chosen according to the circumstances of the facility in which it is to be used.

• It is extremely important to ensure the safety of healthcare workers who may expose themselves to anticancer drugs during drug preparation.
• We conducted a comparative investigation of chemical contamination between conventional preparation (using needles and syringes) and preparation using closed system drug transfer devices (CSTDs).
• Contamination was observed at high frequencies and concentrations in the conventional preparation, even when workers had extensive experience and used standard techniques.
• The BD PhaSeal™ system seems to be effective in preventing unwanted drug exposure.
• It is recommended that an appropriate CSTD with structure and characteristics that suit working conditions be chosen.

1. National Institute for Occupational Safety and Health (NIOSH). NIOSH ALERT Preventing Occupational Exposure to Antineoplastic and Other Hazardous Drugs in Health Care Settings. Cincinnati, USA, NIOSH;2004:1–61.
2. American Society of Health-System Pharmacists (ASHP). ASHP guidelines on handling hazardous drugs, Am J Health Syst Pharm 2006;63:1172–93.
3. Japan Society of Hospital Pharmacists (eds). Guidelines for sterile preparation of injections and anticancer drugs. Yakuji Nippo Limited, Tokyo;2008:3–35.
4. Yoshida J et al. Use of a closed system device to reduce occupational contamination and exposure to antineoplastic drugs in the hospital work environment, Ann Occup Hyg 2009;53:153–60.
5. Connor TH et al. Effectiveness of a closed-system device in containing surface contamination with cyclophosphamide and ifosfamide in an i.v. admixture area, Am J Health Syst Pharm 2002;59:68–72.
6. Nishigaki R et al. The usefulness of a closed-system device for the mixing of injections to prevent occupational exposure to anticancer drugs. J Jpn Soc Hosp Pharm 2010;46:113–17.