M LAW, PhD, R LIU, MBBS, FRCR, S NG, PhD, M Y LUK, MBBS, FRCR, T W LEUNG, MD, FRCR and G K H AU,

Transcription

1 The British Journal of Radiology, 82 (2009), Radiation dose measurements for personnel performing 90 Y- ibritumomab tiuxetan administration: a comparison between two injection methods for dose reduction M LAW, PhD, R LIU, MBBS, FRCR, S NG, PhD, M Y LUK, MBBS, FRCR, T W LEUNG, MD, FRCR and G K H AU, MBBS, FRCR Department of Clinical Oncology, Queen Mary Hospital, Hong Kong ABSTRACT. The purpose of this study was to directly measure, using thermoluminescent dosimeters, the radiation doses received by radiation team members performing 90 Y-ibritumomab tiuxetan administration. The occupational doses associated with two injection methods for patient administration an automatic syringe driver and an injection box were compared. The associated risks, namely cancer induction and hereditary effect, were also estimated from the results and compared with risk factors recommended by the International Commission on Radiological Protection publication 103. The results showed that the doses received by the index and thumb of the right hand and the index finger of the left hand of the radiation oncologist were significantly reduced by using the injection box method. The difference in the dose received by the medical physicist using the two methods was not statistically significant. It was observed that three pairs of latex gloves could further reduce the dose to the hands. The radiological risks of cancer induction and hereditary effect were negligible: of the order of and per 90 Y-ibritumomab tiuxetan administration, respectively, for both methods. However, the results of our study also showed that it would be possible in a busy centre for pregnant women to receive a dose of 90 Y-ibritumomab tiuxetan that exceeds the recommended annual dose limit for the surface of the abdomen. Received 2 January 2008 Revised 1 September 2008 Accepted 10 September 2008 DOI: /bjr/ The British Institute of Radiology 90 Y-ibritumomab tiuxetan (Zevalin) is a novel radioimmunotherapeutic agent for the treatment of relapsed or refractory low-grade, follicular or CD20 + -transformed non-hodgkin s lymphoma [1], and has been used in patient service in our institution since mid The use of this radiolabelled compound has raised concern about the radiation doses received by, and the associated radiological risks to, radiation team members performing routine patient administration [2]. Although there are recent reports on radiation doses to medical personnel during radionuclide therapy and routine nuclear medicine practice [3 8], the literature regarding the best method to administer Zevalin in order to comply with the as low as reasonably achievable (ALARA) principle is limited [9]. Radiation exposure measurements and an estimate of the associated radiological risks to hospital personnel involved in adminstering Zevalin to patients would be useful to ensure that occupational radiation doses are below the annual dose limit recommended by the international standard and to achieve the ALARA principle in radiation protection [10]. Radiation exposure to personnel performing radioactive unsealed source administration has generally been measured with beeper-type Geiger Muller personal pocket dosimeters. However, their bulky size makes Address correspondence to: Dr Rico Liu, Department of Clinical Oncology, Queen Mary Hospital, 102 Pokfulam Road, Hong Kong. liuky@ha.org.hk these dosimeters unsuitable for measuring radiation dose to the lens of the eye and the hands, annual dose limits for which were recommended in International Commission on Radiological Protection publication 103 (ICRP-103) [10]. Although simple to use, pocket dosimeters are not made of tissue-equivalent materials and therefore the radiation dose measured with these detectors cannot reflect the true absorbed dose in the tissue during the course of Zevalin administration. We measured the occupational dose received by radiation team members (radiation oncologist and medical physicist) using high sensitivity and tissueequivalent thermoluminescent dosimeters (TLDs), which were attached to the measurement points of interest for the quantification of occupational radiation doses received during Zevalin administration. Two methods of administration were used, namely an automatic syringe driver and an injection box constructed in-house. Doses received by the radiation oncologist and medical physicist were recorded during patient administration of Zevalin using both methods. The results were compared with the dose limits for radiation workers as recommended by ICRP [10]. Furthermore, the associated risk of cancer induction and hereditary effect from the radiation dose were estimated from the measurement results. These radiation doses were also compared with other occupational doses in the medical community to confirm that the radiation doses received by medical personnel during Zevalin administration are relatively small and The British Journal of Radiology, June

2 M Law, R Liu, S Ng et al that the associated radiological risks are indeed negligible. The implications of the results obtained from this study for pregnant members of the radiation team are also discussed. Methods and materials Thermoluminescent dosimeters The TLDs were made of lithium fluoride (LiF: Mg, Cu, P) of the type TLD-100H (The Harshaw Chemical Company, Solon, OH), having the form of a small square wafer approximately 3.2 mm long and 0.6 mm thick. These TLDs are characterised by their small size, tissueequivalent, linear dose response between 1 msv and 10 Sv and negligible information loss at room temperature after being irradiated. The TLDs can be reset by annealing and used again to perform a new measurement. The batch of TLDs used in this study was calibrated against a 137 Cs source, which was traceable to a secondary standard, and had a known air kerma rate at the time of calibration. According to the manufacturer, the TLD-100H model is suitable for measuring the radiation dose obtained from photons from 60 Co and 137 Cs and electrons with the same TLD signal response. Bilski et al [11] studied the TLD-100H response for beta dosimetry by using 60 Co photons as a calibration source for beta exposures from a number of beta emitters, including 90 Sr/ 90 Y. Their results showed that the dose response of TLD-100H between 60 Co photons and 90 Sr/ 90 Y beta rays was within 4%. In our study, the TLD-100H was calibrated with the use of the photon energy of 137 Cs (662 kev) compared with the mean beta energy of 90 Y (939 kev) [12]. The details of calibration and signal processing for TLD-100H have been previously described [13]. The inherent variation of TLD-100H in dose response was measured to be 6%. In addition, the accuracy of measurements was affected by other factors, namely variation in the dose response of the TLDs after a series of annealing/reading cycles ( 2%) and the derived calibration factor error ( 4%) between 137 Cs and 90 Y radiations. The final results of personnel dose measurement for this study had an error of 12%. Personnel dose monitoring by TLD placement To measure the dose received by personnel, at least two TLDs were placed inside a small black plastic bag, which was then securely attached to the relevant part of the body using Micropore TM tape (3M Deutschland GmbH, Medical Markets Laboratory, Neuss, Germany) immediately before Zevalin administration. The average reading of the TLDs was taken as the dose received after being corrected for ambient noise and then applying the calibration factor. The positions of interest for both the radiation oncologist and the medical physicist were the fingers (the distal phalynx of the thumb and the index, middle and ring finger of both hands), the forehead and the chest. The personal dose equivalent, H p (d), corresponding to dose equivalent at a tissue depth (d) in millimetres (mm), has been used to describe the dose at the position of interest received by individuals subject to external exposures [10]. Thus, radiation doses at the fingers, measured with TLDs, were obtained for a skin dose of H p (0.07). The radiation dose measured with TLDs located at the forehead was intended to simulate a dose to the lens of the eye of H p (3). Radiation dose measured with TLDs at the chest was intended to simulate a whole-body dose of H p (10). Both the radiation oncologist and the medical physicist who participated in this study are right-handed. Patient schedule and Zevalin administration Because of the short physical half-life of 90 Y (64 h), radiolabelling was performed by a regional commercial radiopharmacy laboratory a few hours before patient administration. Upon arrival at our institution, the radioactivity of Zevalin was measured with a dose calibrator (AtomLab-200, Biodex, Brookhaven, NY) by the medical physicist. The patient dose was then transported to the ward for patient administration. Zevalin was prescribed in doses based on the patient s body weight and pre-treatment platelet count [14, 15]. Zevalin was administered at 14.8 MBq kg 1 in patients with platelet counts of l 1 or higher, and at 11.1 MBq kg 1 in patients with platelet counts from l 1 to l 1. The maximum Zevalin patient dose was 1184 MBq, regardless of the patient body weight. An infusion of rituximab 250 mg m 2 was given within 4 h before the administration of the Zevalin to enhance the biodistribution of the subsequent radiolabelled monoclonal antibody by blocking readily accessible CD20 sites in the peripheral blood [16]. Figure 1 shows the patient schedule on the day of administration in our institution. During handling of the patient dose containing the radionuclide 90 Y, staff were protected by syringe shields made of low atomic number Perspex. Because Zevalin was administered as a slow intravenous infusion over 10 min [14, 15], it is necessary to use an automatic syringe driver or to design a radiation protection device for the syringe containing Zevalin [9]. The purpose was to enable the radiation oncologist to avoid direct contact with the syringe during the course of administration in order to minimize the radiation exposure and to perform the administration comfortably. Method A: automatic syringe driver method A commercial automatic syringe driver (Graseby, model 3500, Watford, UK) was used in this method. The syringe containing Zevalin was clamped in a fixed position in the driver (Figure 2). The flow rate was entered into the injector so that the infusion of Zevalin could be completed in 10 min. A 10 mm thick sheet of Perspex was used to cover the front surface of the driver. After Zevalin infusion, the radiation oncologist released the syringe from the automatic syringe driver. The syringe was then flushed at least twice with 0.9% normal saline. Because of residual radioactivity of Zevalin remaining in the syringe at the end of the infusion, 492 The British Journal of Radiology, June 2009

3 Occupational dose measurement in Zevalin administration Patient admission Rituximab 250 mg m _ 2 Radiolabelling Zevalin quality assurance Patient administration Patient release Time (hours on the day of administration) Figure 1. Schedule on the day of patient Zevalin administration in our institution. releasing the clamp and holding the syringe for drawing and flushing normal saline might increase the finger dose to the radiation oncologist. The medical physicist collected the empty syringe and tubing, containing a very small amount of radioactivity of 90 Y, as radioactive waste at the end of the treatment session. Method B: injection box method Figure 3 shows the injection box made in-house to perform Zevalin infusion. The injection box was an enclosure made of a Perspex sheet of 10 mm thickness. The syringe containing Zevalin was inserted into an acrylic syringe holder inside the injection box. Zevalin was administered as a slow intravenous infusion over 10 min by pushing the long piston extending outside the injection box. A long rod of a screwdriver type was used to open and close the three-way valve to draw normal saline into the syringe for flushing. In doing so, there would be no direct contact with the syringe containing residual Zevalin. Being a closed enclosure, the injection Normal saline box would minimise the b-radiation penetrating through the injector. The injection box contained no metallic components; therefore, there would be negligible bremsstrahlung radiation produced [17]. At the end of patient administration, the empty syringe and tubing were left inside the box for radiation decay in order to minimise personnel contact with any components containing very small amounts of radioactivity of 90 Y. By comparing the dose received by the radiation oncologist with that received by the medical physicist performing patient administration using both methods, the difference in personal dose could be statistically analysed to determine which method would reduce the dose significantly. One pair of gloves was worn by the radiation oncologist for all injection procedures. Additional procedure to minimise hand dose An additional procedure to further minimise the hand dose received by the medical physicist, namely wearing three pairs of latex gloves, was studied. The results were then compared with those obtained by wearing a single pair of latex gloves. During administration using the injection box (method B), TLDs were first placed at the distal phalynx of the thumb and at the index, middle and Normal saline Rod-type screw driver Syringe shield Piston To patient Figure 2. Patient administration set-up using the automatic syringe driver method (method A). A Perspex sheet of 10 mm thickness was used to cover the front surface of the automatic syringe driver for radiation protection. To patient Figure 3. Patient administration set-up using the injection box method (method B). The British Journal of Radiology, June

4 M Law, R Liu, S Ng et al ring fingers of both hands. The physicist then put on two pairs of gloves and further TLDs were placed in the same positions over the gloves on both hands. A further pair of gloves was then added. In this way, the effect of wearing one or three pairs of gloves on the average radiation dose received by both hands of the medical physicist could be compared. Hand doses were further minimised by wearing three pairs of gloves because low atomic number materials such as latex can prevent the penetration of b-radiation. The medical physicist who participated in the study reported that wearing three pairs of gloves was tolerable. Statistical method The non-parametric Mann Whitney statistical method was used to compare the difference in the occupational dose received during the two injection methods. Wilcoxon s t-test for paired samples was used to compare the difference in the hand dose measurements of the medical physicist when using three pairs of gloves with those when using one pair of gloves. The software used was Statistica, version 6 (StatSoft, Tulsa, OK). Values of p # 0.05 were considered to be statistically significant. Results Table 1 shows the personal dose, normalised by the patient Zevalin radioactivity, received by the radiation oncologist during the two methods. The doses at the index fingers of both hands (p ) and the dose at the thumb of the right hand (p ) were statistically significantly different using the two methods. The two methods did not result in statistically significant different doses at the forehead and at the chest. Table 2 shows the personal dose, normalised by the patient Zevalin radioactivity, received by the medical physicist during the two methods. There was no statistically significant difference in personal dose between the two methods. The medical physicist had TLDs attached to the fingers underneath three pairs and one pair of latex gloves. It was observed that wearing three pairs of gloves could significantly reduce the left-hand dose by a factor of 1.5 and the right-hand dose by a factor of 2.7 compared with wearing a single pair of gloves (Table 3). Discussion The difference in finger dose received by the radiation oncologist using the two methods of injection could be due to the flushing procedures at the end of patient administration, in which the radiation oncologist had to dismount the syringe containing residual Zevalin from the automatic syringe driver (method A) and then to draw saline to flush the syringe. Thus, both of the radiation oncologist s hands were in close contact with the syringe. In the injection box method (method B), the radiation oncologist performed the flushing of the syringe by manipulating the long piston (Figure 3). The use of the injection box would eliminate the need for close contact between the radiation oncologist s hands and the syringe containing residual Zevalin, thus reducing hand exposure. Radiation doses at the forehead and chest of the radiation oncologist were not statistically different between the two injection methods as the distance between the radiation oncologist and the injection devices used in the two methods was about the same. The medical physicist had the role of performing radioactivity measurement and of mounting the syringe containing Zevalin into the injection device prior to patient administration, as well as assisting the radiation oncologist during infusion. After infusion, all radioactive waste was collected by the medical physicist for storage. The medical physicist also performed a radiation survey of the injection room after the patient was discharged. These procedures carried out by the medical physicist were the same regardless of injection method. This explains why no statistically significant difference in the personal dose received by the medical physicist during the two methods of injection was observed when the medical physicist wore a single pair of latex gloves. Wearing three pairs of latex gloves, the doses to the medical physicist s hands were reduced by a factor of 1.5 and 2.7 for the left and right hand, respectively less than those achieved by wearing a single pair of latex gloves (Table 3). Being right-handed, the medical physicist used his right hand to hold the syringe containing the patient dose of Zevalin, with the use of a pair of long tongs for radioactivity measurement, and to insert the syringe into the injection device. Thus, radiation protection to minimize the hand dose was achieved by wearing three pairs of latex gloves. These findings would be useful as an additional radiation safety measure for personnel manipulating highly radioactive 90 Y, such as during Zevalin labelling procedures [9]. The annual dose limits for an occupational radiation worker, as recommended by the ICRP, are 20 msv for the whole body, 150 msv for the eye lens and 500 msv for skin averaged over any 1 cm 2 area of an extremity [10]. Applied to radiation team members, the number of Zevalin administration sessions that personnel can safely perform each year is limited by the whole-body dose, as Table 1. Comparison of occupational dose levels for a radiation oncologist between the two methods of administration Method Average left-hand finger dose (msv MBq 1 ) Average right-hand finger dose (msv MBq 1 ) Average dose (msv MBq 1 ) Thumb Index Middle Ring Thumb Index Middle Ring Eye Chest A(n5 5) B(n 5 3) p-value Values of p # 0.05 were considered statistically significant. 494 The British Journal of Radiology, June 2009

5 Occupational dose measurement in Zevalin administration Table 2. Comparison of occupational dose levels for a medical physicist between the two methods of administration Method Average left-hand finger dose (msv MBq 1 ) Average right-hand finger dose (msv MBq 1 ) Average dose (msv MBq 1 ) Thumb Index Middle Ring Thumb Index Middle Ring Eye Chest A(n 5 5) B(n 5 3) p-value Values of p # 0.05 were considered statistically significant. indicated by the dose at the chest (Tables 1 and 2). For radiation protection purposes, a maximum patient Zevalin radioactivity of 1184 MBq was assumed. The whole-body dose, i.e. the dose measured by the TLDs at the chest for the radiation oncologist, would be 71 msv and 36 msv per Zevalin administration using method A and B, respectively. The number of patients treated with Zevalin annually by each staff member in our institution was assumed to be 20 at most. Thus, the annual occupational whole-body dose for a radiation oncologist would be 1.42 msv and 0.72 msv using method A and B, respectively. Similarly, for the medical physicist, the whole-body dose would be 130 msv and 95 msv per Zevalin administration using method A and B, respectively, resulting in an annual dose of 2.6 msv and 1.9 msv with method A and B, respectively. Thus, the annual occupational doses of the radiation team members would be well below the annual dose limit recommended by ICRP. The annual occupational whole-body dose received by performing 20 sessions of Zevalin administration would be less than the annual whole-body dose of about 5 msv received by nuclear medicine technologists performing routine patient scanning and injection [8]. In comparison, the estimated occupational doses per case of cardiologists performing catheterization are 1.6 msv for a surface dose with no apron worn, 90 msv for a surface dose with an apron worn, 2.1 msv for the hands and 600 msv for an eye lens [18], all of which are much higher than the occupational doses measured in the current study for radiation team members performing Zevalin administration. In the case of pregnant women, ICRP recommends a dose limit of 1 msv to the surface of the woman s abdomen throughout pregnancy. Applying our results to pregnant radiation team members performing Zevalin administration would suggest that a radiation oncologist could carry out a maximum of 14 and 27 administration sessions using methods A and B, respectively. Similarly, for a pregnant medical physicist, the maximum number of administration sessions would be 7 and 10 with methods A and B, respectively. The current study shows that a pregnant radiation team member working in a centre with a significant workload could reach the dose limit. Although it is a common practice to remove a colleague from work duties that involve radiation once pregnancy has been declared, regular dose monitoring using TLDs on the abdomen is advised if a pregnant colleague has to continue with Zevalin administration. The radiation safety officer should also be consulted. The risks associated with the radiation exposure are generally regarded as deterministic and stochastic effects. The occupational doses, as measured by the current study, were far below the threshold dose of any known deterministic effect, such as a threshold dose of 2 Sv for cataractogenesis or of 2 Sv for transient erythema in a single exposure [18]. Thus, the radiation risks associated with the occupational radiation doses measured in the current study are merely stochastic determinants of cancer induction and hereditary effect, the risks of which can be quantitatively estimated from the whole-body doses we measured and by assuming risk factors of Sv 21 for cancer induction and of Sv 21 for hereditary effect [10]. For radiation protection purposes, a maximum Zevalin radioactivity of 1184 MBq has been assumed for patient administration. Thus, the risk to the radiation oncologist of cancer induction, taking the average whole-body dose as measured by the TLD at the chest, would be and per Zevalin administration for method A and B, respectively. Similarly, the risk to the medical physicist of cancer induction would be and for method A and B, respectively. The risk of hereditary effect for the radiation oncologist would be and per patient administration for method A and B, respectively. The risk of hereditary effect for the medical physicist would be and per patient administration using method A and B, respectively, (Table 4). Individual radiological doses leading to a risk of and in cancer induction and hereditary effects are regarded as negligible [18, 19]. Patient safety using the in-house injection box (method B) when a commercial alternative of an automatic syringe driver is available (method A) has to be assured. The injection box contains no electrical components, which suggests that there are no electrical safety concerns. However, quality assurance of the functioning of the injection box should be carried out before the Table 3. Comparison of hand doses for a medical physicist wearing three pairs or one pair of gloves Pair of gloves Average left-hand dose (msv MBq 1 ) Average right-hand dose (msv MBq 1 ) p-value Values of p # 0.05 were considered statistically significant. The British Journal of Radiology, June

6 M Law, R Liu, S Ng et al Table 4. Tabulation of the radiological risk per Zevalin administration comparing the radiation oncologist with the medical physicist Method Radiation oncologist Medical physicist Cancer induction ( ) Hereditary effect ( ) Cancer induction ( ) Hereditary effect ( ) A B Maximum patient Zevalin radioactivity of 1184 MBq was assumed. patient administration session by simulating patient injection procedures using normal saline in the syringe (Figure 3). Conclusions The widespread interest in using Zevalin for the treatment of relapsed or refractory low-grade, follicular or CD20 + transformed non-hodgkin s lymphoma has drawn attention to the issue of radiation protection for staff. The current study aimed to reduce the radiation dose received by team members performing Zevalin infusion by using the automatic syringe driver method and an injection box made in-house. We have shown that the finger dose received by the radiation oncologist can be reduced using the injection box method, during which there is no close contact between staff and the syringe containing residual Zevalin during drawing and flushing normal saline procedures. At the end of Zevalin infusion, the empty syringe and tubing, containing a very small amount of 90 Y radioactivity, were left inside the injection box as a radioactive waste storage container. The medical physicist spent less time managing the radioactive waste and, therefore, personal exposure was minimized more with the injection box method than with the automatic syringe driver method. According to the ALARA principle of radiation protection, using the injection box method and wearing three pairs of latex gloves on both hands would significantly reduce the occupational dose for radiation team members. References 1. Oyen WJG, Bodel L, Giammarile F, Maecke HR, Tennall J, Luster M, et al. Targeted therapy in nuclear medicine current status and future prospects. Ann Oncol 2007;18: Zhu X. Radiation safety consideration with yttrium 90 ibritumomab tiuxetan (Zevalin). Semin Nucl Med 2004;1: Lancelot S, Guillet B, Sigrist S, Bourrelly M, Waultier S, Mundler O, et al. Exposure of medical personnel to radiation during radionuclide therapy practices. Nucl Med Comm 2008;29: Wrzesien M, Olszewski J, Jankowski J. Hand exposure to ionizing radiation of nuclear medicine workers. Radiat Prot Dosimetry 2008;130: Vanhavere F, Berus D, Buls N, Covens P. The use of extremity dosemeters in a hospital environment. Radiat Prot Dosimetry 2006;118: Pant GS, Sharma SK, Rath GK. Finger doses for staff handling radiopharmaceuticals in nuclear medicine. J Nucl Med Technol 2006;34: Lindner O, Busch F, Burchert W. Performance of a device to minimise radiation dose to the hands during radioactive syringe calibration. Eur J Nucl Med Mol Imaging 2003;30: Lundberg TM, Gray PJ, Bartlett ML. Measuring and minimizing the radiation dose to nuclear medicine technologists. J Nucl Med Technol 2002;30: Cremonesi M, Ferrati M, Paganelli G, Rossi A, Chino M, Bartolomei M, et al. Radiation protection in radionuclide therapies with 90Y conjugates: risks and safety. Eur J Nucl Med Mol Imaging 2006;33: International Commission on Radiological Protection. Recommendations of International Commission on Radiological Protection, Publication 103. USA: Elsevier, Bilski P, Budzanowski M, Olko P, Christensen P. Properties of different thin-layer LiF:Mg, Cu, P TL detectors for beta dosimetry. Rad Prot Dosimetry 1996;66: Delacroix D, Guerre JP, Leblanc P, Hickman C. Radionuclide and radiation protection data handbook 2002, 2nd edition. Radiat Prot Dosimetry 2002;98: Law M, Cheng KC, Wu PM, Ho WY, Chow LWC. Patient effective dose from sentinel lymph node lymphoscintigraphy in breast cancer: a study using a female humanoid phantom and thermoluminescent dosemeters. Br J Radiol 2003;76: EANM procedures guidelines for radio-immunotherapy for B-cell lymphoma with a 90Y-radiolabelled ibritumomab tiuxetan (Zevalin). Eur J Nucl Med Mol Imaging 2007;34: Wagner HN, Wiseman GA, Marcus CS, Nabi HA, Nagle CE, Fink-Bennett DM, et al. Administration guidelines for radioimmunotherapy of non-hodgkin s lymphoma with 90Y-labeled anti-cd20 monoclonal antibody. J Nucl Med 2002;43: Wiseman G, Gordon L, Leigh B, Witzig T, Emmanouilides C, Czuczman M, et al. Safety and efficacy of the Zevalin radioimmunotherapy regimen are not diminished by extending the time interval between rituximab infusion and Zevalin injection. Blood 2000;96:251b. 17. Zanzonico PB, Binkert BL, Goldsmith SJ. Bremsstrahlung radiation exposure from pure b-ray emitters. J Nucl Med 1999;40: Hall EJ, Giaccia AJ. Radiobiology for the radiologists, 6th edition. Philadelphia, PA: Lippincott Williams & Wilkins, Wall BF, Kendall GM, Edwards AA, Bouffler S, Muirhead CR, Meara JR. What are the risks from medical X-rays and other low dose radiation? Br J Radiol 2006;79: The British Journal of Radiology, June 2009