- Email: [email protected]
Transfer standard for beta decay radionuclides in radiotherapy
ARTICLE IN PRESS
Applied Radiation and Isotopes 60 (2004) 519–522
Transfer standard for beta decay radionuclides in radiotherapy Klaus Thieme*, Uwe Beinlich, Eberhard Fritz AEA Technology QSA GmbH, Gieselweg 1, Braunschweig 38110, Germany
Abstract The measurement of the activity of therapeutic radiopharmaceuticals prior to the administration to patients is normally achieved via the use of radionuclide calibrators. An accurate measurement of the activity of pure beta-emitters is complex. Calibration problems can be solved by combining a primary calibration with a 90Y reference solution and a 90 Sr/90Y transfer standard with a solid source, simulating geometric effects caused by high energetic beta radiation. The recent development of a 90Sr/90Y transfer standard for this purpose is reported. r 2003 Elsevier Ltd. All rights reserved. Keywords: Ionisation chamber; Activity measurement; Syringe geometry; Calibration factors
1. Introduction An accurate measurement of activity of pure betaemitters for therapy is a problem in nuclear medicine. The use of radioactive standards traceable to a National Metrology Institute is required. Multiple misadministrations have arisen by the use of incorrect settings and calibration factors of radionuclide calibrators, also when measuring pure beta emitters such as 32P and 90Y (Tyler et al., 2002). The use of 90Y as a therapeutic radionuclide for cancer treatment is increasing as a result of the approval of conjugated monoclonal antibody therapy. In the United States the radionuclide is transferred from the nuclear pharmacies to the administration site in syringes and is generally assayed prior to use in a radionuclide calibrator. Commercial systems use welltype also called re-entrant ionization chambers detecting photon radiation. The response of such chambers to beta particles is poor. The ionization current is created by Bremsstrahlung, which depends on variables, such as the beta-particle energy material and shape of the activity containment, volume of liquid in the syringe, material and density of the container and ionization chamber walls, i.e., is highly dependent on measurement
*Corresponding author. Tel.: +49-5307-932-220; fax: +495307-932-337. E-mail address: [email protected] (K. Thieme).
geometry. The calibration is commonly performed by use of an 90Y reference solution, where aliquots are dispensed by mass into the container under investigation. Geometric dependent calibration factors must then be obtained. This handling may cause problems in pharmacies and in hospitals due to excessive exposure of the medical personnel, possible contamination, and the possibility of large measurement discrepancies. Another approach is proposed in this work. Firstly, the calibration with an 90Y reference solution is performed at a secondary standard laboratory, and then the calibration factors are transferred together with a transfer standard source. This paper describes the recent development of a 90 Sr/90Y transfer standard for the simulation of 90Y in a Becton Dickinson syringe that is traceable to a National Metrology Institute.
2. Experimental One of the most common types of syringes used in therapy with 90Y is the Becton Dickinson luer-lok type 300192. Solutions of 90Y were investigated in this syringes in the volume range from 1 ml to 10 ml. The 90 Sr/90Y activity of the transfer standard was measured in a well-type ionization chamber of a commercial radionuclide calibrator CRC15R (Capintec Inc.), with a vial/syringe sample holder CRC-2401. The 90Sr/90Y
0969-8043/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2003.11.069
ARTICLE IN PRESS K. Thieme et al. / Applied Radiation and Isotopes 60 (2004) 519–522
520
5.00
Reading normal to 2 ml
Reading normal to 2 ml
6.00
4.00 3.00 2.00 1.00 0.00 0
1
2
3
4
5
6
7
8
9
10
Volume/ml
Fig. 1. Measurement of a 10 ml Becton Dickinson syringe filled up to 9 ml volume of 90Y solution in steps of 1 ml. The reading of activity as function of filling volume is normalized to the 2 ml value.
activities were determined by direct comparison against 90 Sr/90Y reference sources calibrated at the National Institute of Standards and Technology (NIST) by microcalorimetry (Colle! and Zimmerman, 2001). Experiments with different types of well type ionization chambers, syringes and various volumes up to 10 ml have been performed previously at NIST and AEA Technology. Equal portions of volume (between 0.5 and 1 ml each) of an 90Y solution with the same activity concentration were added to syringes up to 10 ml volume. The investigated volume range corresponds to that administered to patients. The displayed values of the radionuclide calibrators are described as a ‘‘reading’’ of activity, which is arbitrary unit of activity, and may be converted by a correction term into an activity value. The reading is proportional to the volume added (see Fig. 1) and can be described by a linear function. In the following experiments the geometry was kept as close as possible unchanged. In contrast to the previous measurements *
*
the BD syringes were replaced by an aluminium cylinder of similar size and shape, and 90 Y solution by solid 90Sr/90Y sources.
Solid sources of different length were added in line along the longitudinal axis of the aluminium cylinder to simulate the corresponding volume. For the 90Sr/90Y sources, the reading of activity was measured as a function of simulated volume. For these sources, the reading is also proportional to the volume (Fig. 2), according to a linear function. Both linear functions have a similar slope for volumes from 1 to 9 ml. From the results obtained it was concluded that when activities of pure beta-emitters are measured via Bremsstrahlung in well-type ionization chambers, and if the reading of activity is proportional to the volume in the syringe and follows a linear function both for 90Y in liquid solution and for 90Sr/90Y solid sources, then it is possible to transfer the calibration factors for the corresponding geometries from a secondary standard laboratory to the final user with a transfer standard.
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0
1
2
3
4
5
6
7
8
9
simulated Volume/ml
Fig. 2. Measurement of a 90Sr /90Y transfer standard simulating a 10 ml syringe. The active source length of 50 mm corresponds to a solution volume of 8 ml. The reading of activity as function of volume is normalized to the 2 ml value.
3. Characteristics and calibration of a transfer standard A 90Sr/90Y transfer standard (patent application pending) was recently developed by AEA Technology (Fig. 3). The transfer standard is characterized by the following features: (1) The outer dimensions of length and diameter of the source holder correspond with the syringe simulated. (2) Complete conversion of the beta emission into Bremsstrahlung is achieved. (3) It contains a solid 90Sr/90Y source. (4) The active source length of the solid 90Sr/90Y source corresponds to the height of the volume in the syringe. (5) The 90Sr/90Y transfer standard is calibrated in terms of 90Sr/90Y activity, as well as 90Y apparent activity. The radioactive source used for the 90Sr/90Y transfer standard has a diameter of 0.6470.01 mm and a length of 25.871 mm, representing a volume of 4 ml in a 10 ml BDS, which was adopted for practical reasons. Syringes are filled in pharmacies for therapeutic application commonly with 90 Y solution between 1 and 9 ml volume. The 90Sr/90Y activities are in the order of 1.2 GBq. A calibration of the transfer standard is performed in three phases: Phase 1: calibration of 90Y reference solution, Phase 2: determination of specific calibration factors in the defined geometry of the chosen ionisation chamber, Phase 3: calibration of a 90Sr/90Y transfer standard in terms of 90Y apparent activity. In Phase 1, the activity concentration of the 90Y reference solution was calibrated at the laboratory of the German Calibration Service (DKD) at AEA Technology by an efficiency tracing method of the
ARTICLE IN PRESS K. Thieme et al. / Applied Radiation and Isotopes 60 (2004) 519–522
liquid scintillation counter (Packard Tricarb3100 TR). Calibration factors for 10 ml Becton Dickinson syringes in the ionisation chamber were obtained by gravimetrically transferring aliquots of 1 ml each of an 90Y reference solution into syringes, until a total volume of 9 ml was reached. The results are shown in Fig. 4. The
521
maximum deviation of the reading between a volume of 1 and 9 ml is 4%. In Phase 3, the 90Sr/90Y transfer standard was calibrated in terms of 90Y apparent activity Aapp : The volume dependent correction terms (VDCT) are derived from the measurement of the ratios of activity A to reading R for the corresponding volume of 90Y reference solution (Table 1). VDCT ¼ A=R: The apparent activity is calculated using the equation Aapp ¼ R VDCT
Reading normal to 2 ml
1.02 1.00 0.98 0.96
Fig. 3. A 10 ml Becton Dickinson syringe and a transfer standard.
90
0
90
1
2
3
4
5 6 Volume/ml
Sr/ Y
7
8
9
10
Fig. 4. Measurement result of a 10 ml Becton Dickinson syringe filled with 90Y solution. The reading is reported as function of filling volume and is normalized to 2 ml value.
Table 1 Relative and absolute volume-dependent correction terms (VDCT) for 90Y in 10 ml Becton Dickinson syringe from measurements with a CRC15R radionuclide calibrator Active volume (ml)
Rel. correction norm. to V ¼ 4 ml
Reading MW (MBq)
VDCT
Volume corrected activity (MBq)
1 2 3 4 5 6 7 8 9
1.017 1.001 0.998 1.000 0.995 0.992 0.989 0.986 0.983
58.79 57.89 57.71 57.83 57.56 57.37 57.19 57.01 56.83
0.963 0.978 0.981 0.979 0.984 0.987 0.990 0.993 0.997
56.63 56.63 56.63 56.63 56.63 56.63 56.63 56.63 56.63
Table 2 Calibration of 90Sr/90Y transfer standards in a CRC15R radionuclide calibrator in terms of apparent 90Y activity Source no.
A Sr-90 (MBq)a
Reading Sr-90 (MBq)b
Reading Y-90 (MBq)c
Aapp Y-90 (MBq)d
LG LG LG LG
1340 1260 1250 1220
2069 1950 1938 1901
2099 1978 1966 1929
2055 1936 1925 1888
595 596 597 (NIST) 598 (NIST)
The transfer standard simulates a volume of 4 ml 90Y solution in a 10 ml Becton Dickinson syringe. a Activity measured against NIST standards. b Background corrected reading with 90Sr/90Y isotope factor. c Background corrected reading with 90Y isotope factor. d Apparent activity calculated with VDCT=0.979.
Table 3 90 Sr/90Y transfer standard. Comparison of results in terms of 90Y apparent activity in MBq measured at NIST and at AEA technology AEA data
90
Y Aapp (MBq)
NIST data
90
Y Aapp (MBq)
Source no.
90
CRC15R
CRC15R
CRC12
LG 597 LG 598
1250 1220
1925 1888
1859 1824
1886 1850
Source no.
90
LG 597 LG 598
1250 1220
Sr A (MBq)
Difference from NIST data (%) Sr A (MBq)
CRC15R 2.05 2.08
CRC12 3.53 3.54
ARTICLE IN PRESS 522
K. Thieme et al. / Applied Radiation and Isotopes 60 (2004) 519–522
and can be used by each other device of a CRC15R, to correct the readings. Four sources have been manufactured and measured under identical conditions using the corresponding 90Y dial setting. The values are multiplied with a VDCT value of 0.979 (according to 4 ml 90Y solution) to give the apparent 90Y activity (Table 2). The following relative uncertainty components are measured or estimated: ur (repeated measurements)=0.5%, ur (activity 90 Sr)=1.9%, ur (massic activity of 90Y)=0.5%, ur (calibration factor)=1.0%, ur (decay correction)=0.25%, ur (gravimetric preparation)=0.2%, ur (background correction)=0.25%, ur (geometric corrections)=2.0%, ur (chamber variability)=2.0%. The relative expanded uncertainty U(k ¼ 2) of apparent activity is calculated to 7.5%. Two of the four sources were distributed to NIST and have been investigated with different types of radionuclide calibrators (CRC12 is equipped with the same type of the well-type ionisation chamber like CRC15R). The results obtained by NIST (Zimmerman and Cessna, 2003) are compared with our values (Table 3).
4. Conclusions The NIST results are very close to the apparent activities of 90Y determined by AEA Technology. The deviation to NIST results are o4%. When the radionuclide calibrators have been calibrated with 90Sr/90Y transfer standards, the actual results encourage us to expect that activities for radiotherapy with 90Y will be assayed at hospitals in future with an uncertainty within limits of 10%. References Coll!e, R., Zimmerman, B.E., 2001. A dual-compensated cryogenic microcalorimeter for radioactivity standardisations. Appl. Radiat. Isot. 56, 223–230. Tyler, D.K., Baker, M., Woods, M.J., 2002. NPL secondary standard radionuclide calibrator. Appl. Radiat. Isot. 56, 343–347. Zimmerman, B.E., Cessna, J.T., 2003. Private communication.
Applied Radiation and Isotopes 60 (2004) 519–522
Transfer standard for beta decay radionuclides in radiotherapy Klaus Thieme*, Uwe Beinlich, Eberhard Fritz AEA Technology QSA GmbH, Gieselweg 1, Braunschweig 38110, Germany
Abstract The measurement of the activity of therapeutic radiopharmaceuticals prior to the administration to patients is normally achieved via the use of radionuclide calibrators. An accurate measurement of the activity of pure beta-emitters is complex. Calibration problems can be solved by combining a primary calibration with a 90Y reference solution and a 90 Sr/90Y transfer standard with a solid source, simulating geometric effects caused by high energetic beta radiation. The recent development of a 90Sr/90Y transfer standard for this purpose is reported. r 2003 Elsevier Ltd. All rights reserved. Keywords: Ionisation chamber; Activity measurement; Syringe geometry; Calibration factors
1. Introduction An accurate measurement of activity of pure betaemitters for therapy is a problem in nuclear medicine. The use of radioactive standards traceable to a National Metrology Institute is required. Multiple misadministrations have arisen by the use of incorrect settings and calibration factors of radionuclide calibrators, also when measuring pure beta emitters such as 32P and 90Y (Tyler et al., 2002). The use of 90Y as a therapeutic radionuclide for cancer treatment is increasing as a result of the approval of conjugated monoclonal antibody therapy. In the United States the radionuclide is transferred from the nuclear pharmacies to the administration site in syringes and is generally assayed prior to use in a radionuclide calibrator. Commercial systems use welltype also called re-entrant ionization chambers detecting photon radiation. The response of such chambers to beta particles is poor. The ionization current is created by Bremsstrahlung, which depends on variables, such as the beta-particle energy material and shape of the activity containment, volume of liquid in the syringe, material and density of the container and ionization chamber walls, i.e., is highly dependent on measurement
*Corresponding author. Tel.: +49-5307-932-220; fax: +495307-932-337. E-mail address: [email protected] (K. Thieme).
geometry. The calibration is commonly performed by use of an 90Y reference solution, where aliquots are dispensed by mass into the container under investigation. Geometric dependent calibration factors must then be obtained. This handling may cause problems in pharmacies and in hospitals due to excessive exposure of the medical personnel, possible contamination, and the possibility of large measurement discrepancies. Another approach is proposed in this work. Firstly, the calibration with an 90Y reference solution is performed at a secondary standard laboratory, and then the calibration factors are transferred together with a transfer standard source. This paper describes the recent development of a 90 Sr/90Y transfer standard for the simulation of 90Y in a Becton Dickinson syringe that is traceable to a National Metrology Institute.
2. Experimental One of the most common types of syringes used in therapy with 90Y is the Becton Dickinson luer-lok type 300192. Solutions of 90Y were investigated in this syringes in the volume range from 1 ml to 10 ml. The 90 Sr/90Y activity of the transfer standard was measured in a well-type ionization chamber of a commercial radionuclide calibrator CRC15R (Capintec Inc.), with a vial/syringe sample holder CRC-2401. The 90Sr/90Y
0969-8043/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2003.11.069
ARTICLE IN PRESS K. Thieme et al. / Applied Radiation and Isotopes 60 (2004) 519–522
520
5.00
Reading normal to 2 ml
Reading normal to 2 ml
6.00
4.00 3.00 2.00 1.00 0.00 0
1
2
3
4
5
6
7
8
9
10
Volume/ml
Fig. 1. Measurement of a 10 ml Becton Dickinson syringe filled up to 9 ml volume of 90Y solution in steps of 1 ml. The reading of activity as function of filling volume is normalized to the 2 ml value.
activities were determined by direct comparison against 90 Sr/90Y reference sources calibrated at the National Institute of Standards and Technology (NIST) by microcalorimetry (Colle! and Zimmerman, 2001). Experiments with different types of well type ionization chambers, syringes and various volumes up to 10 ml have been performed previously at NIST and AEA Technology. Equal portions of volume (between 0.5 and 1 ml each) of an 90Y solution with the same activity concentration were added to syringes up to 10 ml volume. The investigated volume range corresponds to that administered to patients. The displayed values of the radionuclide calibrators are described as a ‘‘reading’’ of activity, which is arbitrary unit of activity, and may be converted by a correction term into an activity value. The reading is proportional to the volume added (see Fig. 1) and can be described by a linear function. In the following experiments the geometry was kept as close as possible unchanged. In contrast to the previous measurements *
*
the BD syringes were replaced by an aluminium cylinder of similar size and shape, and 90 Y solution by solid 90Sr/90Y sources.
Solid sources of different length were added in line along the longitudinal axis of the aluminium cylinder to simulate the corresponding volume. For the 90Sr/90Y sources, the reading of activity was measured as a function of simulated volume. For these sources, the reading is also proportional to the volume (Fig. 2), according to a linear function. Both linear functions have a similar slope for volumes from 1 to 9 ml. From the results obtained it was concluded that when activities of pure beta-emitters are measured via Bremsstrahlung in well-type ionization chambers, and if the reading of activity is proportional to the volume in the syringe and follows a linear function both for 90Y in liquid solution and for 90Sr/90Y solid sources, then it is possible to transfer the calibration factors for the corresponding geometries from a secondary standard laboratory to the final user with a transfer standard.
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0
1
2
3
4
5
6
7
8
9
simulated Volume/ml
Fig. 2. Measurement of a 90Sr /90Y transfer standard simulating a 10 ml syringe. The active source length of 50 mm corresponds to a solution volume of 8 ml. The reading of activity as function of volume is normalized to the 2 ml value.
3. Characteristics and calibration of a transfer standard A 90Sr/90Y transfer standard (patent application pending) was recently developed by AEA Technology (Fig. 3). The transfer standard is characterized by the following features: (1) The outer dimensions of length and diameter of the source holder correspond with the syringe simulated. (2) Complete conversion of the beta emission into Bremsstrahlung is achieved. (3) It contains a solid 90Sr/90Y source. (4) The active source length of the solid 90Sr/90Y source corresponds to the height of the volume in the syringe. (5) The 90Sr/90Y transfer standard is calibrated in terms of 90Sr/90Y activity, as well as 90Y apparent activity. The radioactive source used for the 90Sr/90Y transfer standard has a diameter of 0.6470.01 mm and a length of 25.871 mm, representing a volume of 4 ml in a 10 ml BDS, which was adopted for practical reasons. Syringes are filled in pharmacies for therapeutic application commonly with 90 Y solution between 1 and 9 ml volume. The 90Sr/90Y activities are in the order of 1.2 GBq. A calibration of the transfer standard is performed in three phases: Phase 1: calibration of 90Y reference solution, Phase 2: determination of specific calibration factors in the defined geometry of the chosen ionisation chamber, Phase 3: calibration of a 90Sr/90Y transfer standard in terms of 90Y apparent activity. In Phase 1, the activity concentration of the 90Y reference solution was calibrated at the laboratory of the German Calibration Service (DKD) at AEA Technology by an efficiency tracing method of the
ARTICLE IN PRESS K. Thieme et al. / Applied Radiation and Isotopes 60 (2004) 519–522
liquid scintillation counter (Packard Tricarb3100 TR). Calibration factors for 10 ml Becton Dickinson syringes in the ionisation chamber were obtained by gravimetrically transferring aliquots of 1 ml each of an 90Y reference solution into syringes, until a total volume of 9 ml was reached. The results are shown in Fig. 4. The
521
maximum deviation of the reading between a volume of 1 and 9 ml is 4%. In Phase 3, the 90Sr/90Y transfer standard was calibrated in terms of 90Y apparent activity Aapp : The volume dependent correction terms (VDCT) are derived from the measurement of the ratios of activity A to reading R for the corresponding volume of 90Y reference solution (Table 1). VDCT ¼ A=R: The apparent activity is calculated using the equation Aapp ¼ R VDCT
Reading normal to 2 ml
1.02 1.00 0.98 0.96
Fig. 3. A 10 ml Becton Dickinson syringe and a transfer standard.
90
0
90
1
2
3
4
5 6 Volume/ml
Sr/ Y
7
8
9
10
Fig. 4. Measurement result of a 10 ml Becton Dickinson syringe filled with 90Y solution. The reading is reported as function of filling volume and is normalized to 2 ml value.
Table 1 Relative and absolute volume-dependent correction terms (VDCT) for 90Y in 10 ml Becton Dickinson syringe from measurements with a CRC15R radionuclide calibrator Active volume (ml)
Rel. correction norm. to V ¼ 4 ml
Reading MW (MBq)
VDCT
Volume corrected activity (MBq)
1 2 3 4 5 6 7 8 9
1.017 1.001 0.998 1.000 0.995 0.992 0.989 0.986 0.983
58.79 57.89 57.71 57.83 57.56 57.37 57.19 57.01 56.83
0.963 0.978 0.981 0.979 0.984 0.987 0.990 0.993 0.997
56.63 56.63 56.63 56.63 56.63 56.63 56.63 56.63 56.63
Table 2 Calibration of 90Sr/90Y transfer standards in a CRC15R radionuclide calibrator in terms of apparent 90Y activity Source no.
A Sr-90 (MBq)a
Reading Sr-90 (MBq)b
Reading Y-90 (MBq)c
Aapp Y-90 (MBq)d
LG LG LG LG
1340 1260 1250 1220
2069 1950 1938 1901
2099 1978 1966 1929
2055 1936 1925 1888
595 596 597 (NIST) 598 (NIST)
The transfer standard simulates a volume of 4 ml 90Y solution in a 10 ml Becton Dickinson syringe. a Activity measured against NIST standards. b Background corrected reading with 90Sr/90Y isotope factor. c Background corrected reading with 90Y isotope factor. d Apparent activity calculated with VDCT=0.979.
Table 3 90 Sr/90Y transfer standard. Comparison of results in terms of 90Y apparent activity in MBq measured at NIST and at AEA technology AEA data
90
Y Aapp (MBq)
NIST data
90
Y Aapp (MBq)
Source no.
90
CRC15R
CRC15R
CRC12
LG 597 LG 598
1250 1220
1925 1888
1859 1824
1886 1850
Source no.
90
LG 597 LG 598
1250 1220
Sr A (MBq)
Difference from NIST data (%) Sr A (MBq)
CRC15R 2.05 2.08
CRC12 3.53 3.54
ARTICLE IN PRESS 522
K. Thieme et al. / Applied Radiation and Isotopes 60 (2004) 519–522
and can be used by each other device of a CRC15R, to correct the readings. Four sources have been manufactured and measured under identical conditions using the corresponding 90Y dial setting. The values are multiplied with a VDCT value of 0.979 (according to 4 ml 90Y solution) to give the apparent 90Y activity (Table 2). The following relative uncertainty components are measured or estimated: ur (repeated measurements)=0.5%, ur (activity 90 Sr)=1.9%, ur (massic activity of 90Y)=0.5%, ur (calibration factor)=1.0%, ur (decay correction)=0.25%, ur (gravimetric preparation)=0.2%, ur (background correction)=0.25%, ur (geometric corrections)=2.0%, ur (chamber variability)=2.0%. The relative expanded uncertainty U(k ¼ 2) of apparent activity is calculated to 7.5%. Two of the four sources were distributed to NIST and have been investigated with different types of radionuclide calibrators (CRC12 is equipped with the same type of the well-type ionisation chamber like CRC15R). The results obtained by NIST (Zimmerman and Cessna, 2003) are compared with our values (Table 3).
4. Conclusions The NIST results are very close to the apparent activities of 90Y determined by AEA Technology. The deviation to NIST results are o4%. When the radionuclide calibrators have been calibrated with 90Sr/90Y transfer standards, the actual results encourage us to expect that activities for radiotherapy with 90Y will be assayed at hospitals in future with an uncertainty within limits of 10%. References Coll!e, R., Zimmerman, B.E., 2001. A dual-compensated cryogenic microcalorimeter for radioactivity standardisations. Appl. Radiat. Isot. 56, 223–230. Tyler, D.K., Baker, M., Woods, M.J., 2002. NPL secondary standard radionuclide calibrator. Appl. Radiat. Isot. 56, 343–347. Zimmerman, B.E., Cessna, J.T., 2003. Private communication.