PROGRESS IN ORGANIZING NATIONAL AND INTERNATIONAL COMPARISONS FOR NUCLEAR MEDICINE MEASUREMENTS

PROGRESS IN ORGANIZING NATIONAL AND INTERNATIONAL COMPARISONS FOR NUCLEAR MEDICINE MEASUREMENTS MARIA SAHAGIA, ANAMARIA CRISTINA WÄTJEN, CONSTANTIN IVAN Horia Hulubei National Institute of R & D for Physics and Nuclear Engineering, IFIN-HH, POB MG-6, 077125 Bucharest, Romania E-mail: msahagia@nipne.ro Received June 26, 2008 The paper presents the progress registered by the Radionuclide Metrology Laboratory (RML) from IFIN-HH, in improvement of the quality of radiopharmaceuticals activity measurement, and assurance of the whole traceability chain from international level to the national users. The progress in organization of comparisons for radiopharmaceuticals activity measurement is analyzed. A detailed description of two recent national comparisons, and an international one, regarding 131 I solutions, within the frame of the IAEA s CRP E2.10.05 Harmonization of quality practices for nuclear medicine radioactivity measurements is done, with the analysis of the strong and week points in measurements. The most important outcome of the paper is the analysis of the evolution in quality of comparisons since a previously reported one. At the RML level, the improvements are: demonstration of the international equivalence of primary Romanian activity standard, improvement of the secondary standard, relevance of the comparison within the IAEA frame, implementation of a quality system in standardization and in comparisons, and implication of the RML in calibrations and metrological checks. The participants reported improvements refer to growing of the awareness in quality of measurement and improvement of the measurement equipment. Reasons for future comparisons organization are presented. Key words: Radionuclide Metrology Laboratory, 131 I radiopharmaceutical solution, national and international comparisons, IAEA CRP E 2.10.05. 1. INTRODUCTION The Radionuclide Metrology Laboratory (RML), situated in the Radioisotope Department of IFIN-HH, is the owner of the primary Romanian radioactivity standard. In these circumstances it is determined to solve problems regarding the quality assurance of radiopharmaceuticals activity measurement. Besides the development of standardization methods, it has already a tradition of organizing national comparisons for the measurement of radiopharmaceuticals, Paper presented at the 12 th International Congress of the International Radiation Protection Association, IRPA12, Buenos Aires, Argentina, 19 24 October 2008. Rom. Journ. Phys., Vol. 54, Nos. 7 8, P. 619 627, Bucharest, 2009

620 Maria Sahagia, Anamaria Cristina Wätjen, Constantin Ivan 2 both in production and in use. The results of the first comparisons, for the radiopharmaceuticals containing 131 I, 57 Co, 99m Tc were already reported [1, 2]. During the period last from the first comparison, our laboratory registered a significant progress in development of absolute (primary) standardization methods for various pharmaceuticals [3, 4]. Another very important aspect is the assurance of the whole traceability chain, starting from the demonstration of international equivalence of the primary standard, by participation at key comparisons organized by the International Committee for Measures and Weights (CIPM) [5], to the national traceability of measurements [6, 7]. The secondary standard, used for the transfer of activity unit from primary standard to the secondary standardization of radiopharmaceuticals, based on a CENTRONIC IG12/20A ionization chamber, was validated by comparisons with PTB Germany [8] and improved by the use of a highest quality electrometer, type Keithley E6517A, and by the transfer of the calibration figures to the new system [9]. The quality management system, according to the SR EN ISO/IEC 17025:2005, General requirements for the competence of testing and calibration laboratories, in radioactivity measurements was implemented [10]. An extensive activity of authorized metrological checking of the radionuclide calibrators for external users, according to the metrology law, was developed [11]. RML became part of an International Atomic Energy Agency (IAEA) Coordinated Research Project, (CRP), E2.10.05 Harmonization of quality practices for nuclear medicine radioactivity measurements. A very important international result of CRP is the issue of the IAEA s Technical Report Series no. 454 [12]. According to the project program we participated also at an international comparison regarding the measurement of 131 I activity; it was registered as a supplementary key comparison, by the CIPM. The results are published [13]. This paper presents the results of the recent national and international comparisons regarding the measurement of 131 I radiopharmaceuticals. The first comparison implied two Romanian units. The second one was a large scale comparison, as a part of the CRP E2.10.05 work program. On the request of the IAEA CRP s coordinator, we organized also an international comparison with the Ankara University, Medicine Faculty, Turkey, whose results are included within the paper. The progress in organizing the comparisons, as well as the evolution of the radiopharmaceuticals activity measurement quality at the national level are analyzed. 2. FIRST 131 I COMPARISON FROM 2005 On the occasion of participation in the 131 I key comparison, in the frame of the International Reference System (SIR), a national comparison, including two significant units involved in 131 I measurements, one control laboratory and one

3 Comparisons for nuclear medicine measurements 621 hospital, both of them situated in the Bucharest area, was organized. The solution was standardized absolutely, by the 4 - coincidence method, such as described in paper [5]. Three types of vials: PTB-2 ml, SIR-3.6 ml, flame sealed vials and P6-5 ml pharmaceutical vial, containing three levels of activity, situated in a low level measurement interval, were used. The uncertainty of activity, certified by the RML, was u c = 0.31% (k = 1), according to ISO, GUM, 1995 [14]. Activities were calculated at the comparison reference date: 14.11.2005 12:00 UTC. Tables 1 and 2 present the obtained results. Table 1 Results obtained at the Hospital Calibrator Type Vial type RML, certified Activity, MBq Hospital registered Activity, MBq Activity Ratio: Hospital/RML PICKER PTB 1.618 1.311 0.810 SIR 2.892 2.280 0.793 P6 4.050 3.185 0.787 Table 2 Results obtained at the Control Laboratory (CL) Calibrator Type Vial type RML, certified Activity, MBq CL registered Activity, MBq Ratio: CL/RML Capintec CRC 15 PTB 1.618 1.445 0.893 SIR 2.892 2.982 1.031 P6 4.050 4.138 1.022 From the above result one may conclude that the Picker, old calibrator, was de-calibrated, and is systematically 20% lower than the certified activity, even for the P6 geometry, for which calibration was considered, what is out of the allowed uncertainty (10% for diagnosis and 5% for therapy radiopharmaceuticals). After this exercise, the calibrator was recalibrated at the RML; a new Calibration Certificate was issued, specifying that the new certified calibration must be used in routine measurements. From the presented results one may conclude that the calibration is suitable for P6 vial, acceptable for SIR, 3.6 ml vial, but for PTB vial the result is not acceptable. The deviation is due to the geometry of vial, and to the lower value of activity. 3. SECOND NATIONAL AND INTERNATIONAL COMPARISONS WITHIN CRP E.2.10.05 The comparisons were performed in agreement with the RLM, IFIN-HH Quality System prescriptions, according to the EN ISO/IEC 17025:2005

622 Maria Sahagia, Anamaria Cristina Wätjen, Constantin Ivan 4 requirements, following the Work Procedure, Code AC-PL-LMR-07, Organization of interlaboratory comparisons on the measurement of activity of radioactive solutions and solid sources. The procedure has as reference the documents: ISO/IEC Guide 43:1997 Development and operation of proficiency testing schemes. Part 1: Proficiency testing by interlaboratory comparisons, and Part 2: Selection and use of proficiency testing schemes by laboratory accreditation bodies ; EA-2/09 EA Policy on the Accreditation of Providers of Proficiency Testing Schemes. The report of the comparison, sent to the IAEA, was in compliance with the Form: Code AC-PL-LMR-07-04, Reporting of comparison results, Annex of the Procedure AC-PL-LMR-07. 3.1. PREPARATIVE ACTIVITIES Preparative discussions were deployed with a number of 18 possible Romanian interested units: Radioisotope Department from IFIN-HH (RD), National Agency for Control of Drugs, Hospitals, and Ankara University, Turkey. The comparison was approved and surveyed by the National Commission for the Control of Nuclear Activities (CNCAN), national responsible authority. The agreement was obtained from 11 Romanian units and the foreign one. In order to promote the interest in proficiency testing among various laboratories performing measurements, RML participated at two national workshops specialized on Proficiency Testing [6, 7]. 3.2. REALISATION OF THE COMPARISON Preparation and standardization of the vials with 131 I radioactive solution A concentrated I-131 solution, with the radioactive concentration 100 MBq/mL and the chemical composition: NaI-50 g/ml+ Na 2 S 2 O 3-80 g/ml in NaOH 0.05 N, was prepared. The samples were prepared by dispensing 50 L of concentrated solution in the vials containing, each of them, 5 ml of inactive solution, with the same chemical composition. Each vial was standardized by the use of the ionization chamber, type CENTRONIC 12G/20A, calibrated with the solution absolutely standardized. The nominal, approximate, activity for a vial on the reference date, was communicated to the participants. The vials were sealed, labelled with a code number, and sent to them with authorized vehicles, or according to the export rules, accompanied by the legally required documents and two specific forms: (i) Technical Data Form, code: AC-PL-LMR-07-02, containing the chemical composition and volume of solution and the approximate activity per vial; (ii) Comparison Form, code AC-PL-LMR-07-03, to be filled by the participants after performing measurements.

5 Comparisons for nuclear medicine measurements 623 Measurements carried out and reported by the participants The participants performed the measurements, according to their routine procedures, using the calibration figure of calibrators such it is certified, and applying necessary corrections indicated in the reporting form, code AC-PL-LMR- 07-03. RML recommended the formula for the evaluation of measurement uncertainty and the 131 I half life value, T 1/2 = (8.0233 0.0019) days. Then, they filled in the form, reporting the data: type of equipment, producer and purchase date; last calibration date and performing unit; last metrological check, according to metrological law; type of corrections applied to the measurement data. Seven participants calculated and reported the measurement uncertainty. 3.3. RECEIPT AND PROCESSING OF PARTICIPANTS RESULTS The conventionally true activities for each vial, A CA were determined by the RML with an uncertainty u c1 = (0.67 0.75)%, for a coverage factor, k = 1. The reported values, A R, are those presented in the completed reporting forms, AC-PL-LMR-07-03. Both activities were calculated for the reference comparison date. The reported uncertainty, u c2, %, was given for the coverage factor (k = 1). We calculated the following quantities: (i) Ratio between the reported values, A R, and AR the conventionally true activity A CA, R ; (ii) Relative difference between A CA AR ACA results, U,%; (iii) Deviation normalized with respect to the stated ACA uncertainty, according to the EAL-P7 [15], En U, k = 2, for those k u2 2 c1 uc2 participants who calculated and reported uncertainty value; (iv) For these results, we applied an acceptability criterium, according to EAL-P7, indicating that a result is acceptable, in agreement with the reference value, in our case A CA, when the following relation is accomplished: E n < 1. Table 3 summarizes the comparison results. Participant code Equipment Table 3 Comparison results A CA, MBq A R, MBq u c2, % R U, % E n result L1a Capintec CRC-15R** 4.652 4.702 2 1.011 1.1 0.26 L1b Picker ** 4.652 4.853 4 1.043 4.3 0.53 L1c Capintec CRC-15R** 4.652 4.722 2 1.015 1.5 0.36 (continues)

624 Maria Sahagia, Anamaria Cristina Wätjen, Constantin Ivan 6 Participant code Equipment A CA, MBq A R, MBq u c2, % R U, % Table 3 (continued) E n result L2* Capintec CRC-15R 4.600 4.700 3 1.022 2.2 0.35 L3* Picker** 4.600 4.590 2 0.998 0.22 0.05 L4 Curiementor 3 4.599 4.521 0.983 1.7 L5 ISOMED1000, probe type 637620 4.592 4.721 5.7 1.028 2.6 0.23 L6 Picker 4.601 4.700 1.022 2.2 L7 Curiementor 2 4.582 4.980 1.087 8.7 L8 Picker 4.574 6.327 1.383 38.3 L9 Curiementor 2 4.605 3.917 0.851 14.9 L10 Curiementor 3** 4.601 4.660 4.7 1.013 1.3 0.14 L11 L11*** Picker 5.123 9.177 5.460 L12 BIODEX ATOMLAB 9.563 9.513 9.521 7.8 1.791 1.066 1.54 1.62 0.995 0.996 79.1 6.6 0.52 0.45 5.05 0.42 0.30 0.25 Not accepted * Participant in the 2005 comparison. ** Calibration performed by RML, by transfer of activity unit from the primary standard. *** Recalculated value, errors in conversion of units from Ci (provided for calibrator Picker) to MBq. 4. DISCUSSION OF RESULTS 4.1. CONCLUSIONS FROM THE RECENT COMPARISONS The analysis of the results presented in Tables 1, 2, 3, emphasizes the following aspects. The majority of laboratories use the calibration factors provided by the equipment producers. RLM carried out calibration for 5 calibrators. The metrological check is accomplished at intervals longer than one year, although the Metrology Law and Technical Metrology Instruction IML3-04 require annual check. From a total number of 15 reported values, a number of 11 results are within the difference limit from reference value < 5%, and 12 results within the limit < 10%, a proportion of 80% from participating units, being in compliance with the Pharmacopoeia and Metrology Norm NTM 3-99. Laboratory 11, reported an unacceptable result. After contacting them and correcting the conversion of units, the result is within the acceptable limit, difference 6.6%. The mean value of the ratio R = A R /A CA, calculated by considering all values is R = 1.083 (S n 1 = 22%), due to eroneous results, with differences > 10%. If the L8, L9 and L11 results are rejected from the mean calculation, the new mean ratio is

7 Comparisons for nuclear medicine measurements 625 R = 1.018 (S n 1 = 1.27%), a statistical distribution around R = 1.00 with a standard deviation in compliance with the reported uncertainties. A significant number of participants have no ideea about uncertainties in measurement. Results of participants in 2005 comparison. For participant L2, the same difference from RML value, +2.2%, is maintained. Participant L3 used in this exercise the calibration factor provided by the RML; the good maintaining and use of recalibration was confirmed. 4.2. PROGRESS REGISTERED SINCE THE 1996 COMPARISON Progress registered in our laboratory activity was materialized in: The laboratory demonstrated its international equivalence, by participating in a key and a supplementary 131 I key comparison, at the CIPM and IAEA levels. The secondary standard, ionization chamber CENTRONIC IG12/20A calibration figures were validated by participation in key comparisons and also compared with PTB standard and were confirmed, within the limit of stated uncertainties. It is now operated with a new, higher level performance electrometer, Keithley E 6517A. The comparison was part of an IAEA coordinated program, and its relevance was judged in a worldwide context; RML demonstrated also its capability to organize an international comparison. RML implemented the quality system in calibration and measurement, according to EN ISO/IEC17025 standard requirements. The comparison was performed by applying the requirements of the ISO/IEC Guide 43:1997, and EA-2/09. The interpretation of results included an acceptability criterium, defind in EAL-P7, such as it is internationally recommended for all kinds of proficiency test exercises. The RML is directly implied in carrying out metrological checks of calibrators performances and their recalibration, according to the Romanian Metrology Law, what influenced positively the quality of results obtained by some hospitals. Regarding the participants in comparison, the following progress has been registered since the previous comparison: The hospitals awarness for the necessity to participate in proficiency testes is growing. In 1996, the proportion of hospital performing measurements of the samples received for comparison was only 40%. In the present exercise,

626 Maria Sahagia, Anamaria Cristina Wätjen, Constantin Ivan 8 the number of respondents was 63%, all of them measuring and reporting results. This interest must be improved in the future. The RD and some hospitals succeeded in purchasing last generation calibrators, what is reflected in the quality of results. In 1996, the proportion of hospitals measuring activities within the required uncertainty limit 10%, was only 66%, while during this comparison the proportion raised at 80%. Some human errors still persist in measurement, although some calibrators proved to be well calibrated during the metrological check. The overall result, less than 100% of measurements being in compliance with the legal requirements, is a serious reason for recommending future new comparison exercises. 5. CONCLUSIONS The paper presents the results obtained by the Radionuclide Metrology Laboratory from IFIN-HH, Romania, in organization of two recent 131 I national and one international, comparisons, within the frame of the IAEA s CRP-E 2.10.05 and analyses them. Many equipments had the benefit of the RML calibration Ten results from fifteen provided an agreement better than 2.6% with the conventionally true value of activity The progress registerred by the RML and by the participants, by comparison with the previous comparisons, is underlined. Acknowledgement. The authors thank all the participants for measuring the samples and reporting the results. They are also indebted to the IAEA for the support in accomplishment of this work (CRP E 2.10.05, Contract ROM 12921/2004). REFERENCES 1. M. Sahagia, A. C. Razdolescu, E. L. Grigorescu, C. Ivan, A.Luca, National comparison of 131 I solution measurements in Romanian nuclear medicine units, IRPA9 Congress, Vienna, Proc. Vol 3, pp. 497 499 (1996). 2. M. Sahagia, E. L. Grigorescu, A. C. Razdolescu, C. Ivan, A. Luca, Quality assurance in the measurement of the activity of radiopharmaceuticals, 2 nd Balkan Congress of Oncology, Izmir, Turkey, 1998. 3. M. Sahagia, A. C. Razdolescu, E. L. Grigorescu, C. Ivan, A. Luca, Measurement of the activity of the radiopharmaceuticals used in therapy, Conf. IRPA EUROPE, Paris, 2006, paper P 113, pp 1 6, electronic publication, Ed. IRPA, http://www.irpa2006europe.com. 4. M. Sahagia, Standardization of 99m Tc, Appl. Radiat. Isot. 64, 1234 1237(2006). 5. M. Sahagia, A. C. Razdolescu, C. Ivan, A. Luca, Assurance of the traceability chain for 131 I measurement, Appl. Radiat. Isot. 66, 4, 539 544 (2008). 6. M. Sahagia, M. J. Woods, The National Dissemination of International Measurements, International Workshop Nuclear Proficiency Testing, AIP Conf. Proc 1036, pp. 12 25, 2008. ISBN 978-0-7354-0560-8.

9 Comparisons for nuclear medicine measurements 627 7. M. Sahagia, A. C.Razdolescu, E. L. Grigorescu, A. Luca, C. Ivan, From international equivalence to national traceability in radionuclide metrology, 1 st International Proficiency Testing Conference, Proc.pp. 253 261, 2007, ISBN 978-973-8132-65-8. 8. E. L. Grigorescu, A. Luca, M. Sahagia, A. C. Razdolescu, C. Ivan, 4 Gamma ionization chamber method for secondary standardization of radioactive solutions, Rom. Journal of Physics, Vol. 48, 91 95 (2003). 9. M. Sahagia, A. C. Razdolescu, Quality assurance in nuclear medicine radioactivity measurements, Rom. Rep. Phys., Vol. 59, 4, 1119 1126 (2007). 10. M. Sahagia, A. C. Razdolescu, A. Luca, C. Ivan, Importance of the primary radioactivity standard laboratory and implementation of its quality management, American Institute of Physics, AIP Conf. Proc. 899, 523 524 (2006). 11. A. C. Razdolescu, M. Sahagia, A. Luca, S. Bercea, C. Dumitrescu, H. Schrader, Results obtained in the metrological certification of a commercially available radionuclide calibrator, Appl. Radiat. Isot. 56, 957 958 (2002). 12. Contributors to drafting and review: M. Dondi, C. P. Herbst, A. Iwahara, M. Morengo, S. Mather, J. P. Norenberg, V. Olsovcova, P. Oropesa Verdicia, P. Otiz-Lopez, L. Joseph, M. Ghafoori, J. A. Onto, N. Ramamoorthy, M. Sahagia, K. R. Shortt, S. Tastan, P. Vincze, M. J. Woods, B. E. Zimmerman, Quality Assurance for Radioactivity Measurements in Nuclear Medicine, International Atomic Energy Agency (IAEA) Vienna, Austria Technical Report Series 454 (TRS 454), pp. 1 96, 2006, ISBN 92-0-105306-1, ISSN 0074-1914. 13. B. E. Zimmerman, A. Megzifene, K. R. Short, Establishing measurement traceability for national laboratories: Results of an IAEA comparison of 131 I, Appl. Radiat. Isot, 66, 6 7, 954 959 (2008). 14. ISO/IEC/OIML/BIPM, 1995. Guide to the Expression of Uncertainty in Measurement. International Standard Organisation, Geneva, Switzerland, 1995. 15. EAL-P7, Inter-laboratory Comparisons of the European Cooperation for Accreditation of Laboratories, 1996.