Extraction Paper Chromatography Technique for the Radionuclidic

Anal. Chem. 2008, 80, 801-807

Extraction Paper Chromatography Technique for the Radionuclidic Purity Evaluation of 90Y for Clinical Use Usha Pandey,† Prem S. Dhami,‡ Poonam Jagesia,‡ Meera Venkatesh,*,† and M. R. A. Pillai§

Radiopharmaceuticals Division and Fuel Reprocessing Division, Bhabha Atomic Research Centre (BARC), Mumbai-400085, India, and Industrial Applications and Chemistry Section, Division of Physical and Chemical Sciences, IAEA, Wagramerstrasse 5, A 1400, Vienna, Austria

Targeted therapy using β- emitting radionuclides is expected to rapidly expand in the coming years. 90Y-based radiopharmaceuticals are widely used for the treatment of cancer as well as in radiation synoviorthesis.1-8 The increasing application of 90Y in nuclear medicine is due to its suitable nuclear characteristics (T1/2

64.1 h, β-max 2.28 MeV, no γ emission). Availability of large quantities of high specific activity 90Y with very high radionuclidic (RN) purity is essential for expanding the scope of targeted therapy using this important therapeutic radionuclide. High specific activity 90Y obtained from a 90Sr/90Y generator system is required for the preparation of peptide/antibody-based, 90Y-labeled radiopharmaceuticals used for receptor/antigen targeting in the tumor.9 These radiopharmaceuticals are prepared either in a central radiopharmacy or in a hospital- based radiopharmacy using 90Y in inorganic form, usually as the chloride or acetate, supplied by commercial manufacturers. In order to ensure good radiopharmacy practice, it is mandatory to perform the radiopharmaceutical quality control prior to administration to the patient. The QC test for 90Y radiopharmaceuticals or alternatively the radiochemical used for its preparation should include the RN purity evaluation to confirm that the 90Sr level is below the permissible limit. In the case of 90Y, such a quality control test to estimate the RN impurity is very important, because the parent 90Sr, which is most likely the RN impurity, has a very low maximum permissible body burden of 74 kBq (2 µCi)10 as it localizes in the skeleton. In all generator-produced radionuclides, the only potential radionuclidic contamination is the parent radionuclide, provided the parent radionuclide used is pure. Generally, RN purity estimation is performed by γ spectrometry since many radionuclides used in nuclear medicine have γ emission along with β- emission, such as 188W and 188Re.11 However, in the case of the 90Sr/90Y pair, both the parent and daughter are pure β- emitters and no γ emissions are available to permit γ analysis. In addition, the β- spectra of these two isotopes overlap to some extent. If the parent radionuclide had γ

* To whom correspondence should be addressed. E-mail: [email protected]. Fax: +91-22-25505151. † Radiopharmaceuticals Division, Bhabha Atomic Research Centre (BARC). ‡ Fuel Reprocessing Division, Bhabha Atomic Research Centre (BARC). § IAEA. (1) Davies, A. J. Oncogene 2007, 25, 3614-3628. (2) Witzig, T. E.; Molina, A.; Gordon, L. I.; Emmanouilides, C.; Schilder, R. J.; Flinn, I. W.; Darif, M.; Macklis, R.; Vo, K.; Wiseman, G. A. Cancer 2007, 109, 1804-1810. (3) Paganelli, G.; Bartolomei, M.; Grana, C.; Ferrari, M.; Rocca, P.; Chinol, M. Neurol. Res. 2006, 28, 518-522. (4) Ferrari, M.; Cremonesi, M.; Bartolomei, M.; Bodei, L.; Chinol, M.; Fiorenza, M.; Tosi, G.; Paganelli, G. J. Nucl. Med. 2006, 47, 105-112. (5) Breeman, W. A.; Kwekkeboom, D. J.; de Blois, E.; de Jong, M.; Visser, T. J.; Krenning, E. P. Anticancer Agents Med. Chem. 2007, 7, 345-357.

(6) Kwekkeboom, D. J.; Mueller, B. J.; Paganelli, G.; Anthony, L. B.; Pauwel, S.; Kvols, L. K.; O’Dorisio, T. M.; Valkema, R.; Bodei, L.; Chinol, M.; Maecke, H. R.; Krenning, E. P. J. Nucl. Med. 2005, 46 (Suppl. 1), 62S-66S. (7) Gulec, S. A.; Mesoloras, G.; Dezarn, W. A.; McNeillie, P.; Kennedy, A. S. J. Transl. Med. 2007, 5, 15. (8) Wong, C. Y.; Savin, M..; Sherpa, K. M.; Qing, F.; Campbell, J.; Gates, V. L.; Lewandowski, R. J.; Cheng, V.; Thie, J.; Fink-Bennet, D.; Nagle, C.; Salem, R. Cancer Biother. Radiopharm. 2006, 21, 305-313. (9) Mikolajczak, R.; Parus, J. L. World J. Nucl. Med. 2005, 4, 184-190. (10) National Bureau of Standards Handbook; (Maximum permissible body burden and maximum permissible concentrations of radionuclides in air and water for occupational exposure); U.S. Government Printing Office: Washington, DC. 1963; Vol. 69, p 38. (11) Jackel, B.; Cripps, R.; Gu ¨ ntay, S.; Bruchertseifer, H. Appl. Radiat. and Isot. 2005, 63, 299-304.

Yttrium-90 used for therapy should be of very high radionuclidic (RN) purity (>99.998%) as the most probable contaminant, strontium-90, is a bone seeker with a maximum permissible body burden of 74 kBq (2 µCi) only. None of the current known methods of RN purity estimations is adequate to reliably measure the 90Sr RN impurity at such low levels. Our aim was to develop a reliable technique to accurately determine the amount of 90Sr in 90Y used for therapy. This new technique combines chelate-based extraction with paper chromatography using paper impregnated with 2-ethylhexyl, 2-ethylhexylphosphonic acid (KSM-17), which is a 90Y-specific chelator. A PC strip impregnated with KSM-17 at the point of spotting is used for chromatography. Upon development with normal saline, 90Sr moves to the solvent front leaving 90Y completely chelated and retained at the point of spotting. The activity at the solvent front (90Sr) is quantified by liquid scintillation counting, and the data are compared with the total applied activity to provide the RN purity of the test solution. The method has a sensitivity of g74 kBq (2 µCi) of 90Sr per 37 GBq (1 Ci) of 90Y. This novel, innovative, and simple technique offers a reliable solution to the unanswered problem of estimation of 90Sr content in 90Y used for cancer therapy.

10.1021/ac701651u CCC: $40.75 Published on Web 01/04/2008

© 2008 American Chemical Society

Analytical Chemistry, Vol. 80, No. 3, February 1, 2008 801

emission, as is usual in most cases of radioactive equilibrium, estimation of RN purity can be achieved easily by γ spectroscopy. But in the case of 90Sr/90Y, β- counting will have to be done after separating them unambiguously. The presence and levels of trace amounts of 90Sr in 90Y depend on the efficiency of the separation technique used for separation of 90Y from 90Sr, the parent radionuclide, by the manufacturer. The manufacturers quote the radionuclidic impurity as e740 kBq (20.0 µCi) of 90Sr per 37 GBq (1 Ci) 90Y.12 In order to measure the above level of purity, the ability to detect 20 dpm or less of 90Sr in 1 million dpm of 90Y is required. Currently available methods involve the tedious separation of 90Sr from the 90Y solution followed by a quantification method. The manufacturers do not disclose their production or quality control methods. The product is not yet certified for medical use, and the responsibility for clinical use of 90Y lies with the hospital radiopharmacist/nuclear medicine physician involved in the preparation and use of the radiopharmaceutical. An extensive literature survey has revealed that the quality control methods adopted by researchers for determination of 90Sr breakthrough in the 90Sr/90Y generator systems do not conform with the radionuclide purity requirements. The magnitude of this challenge can be assessed by the publications on this topic during the 1955-2006 period.13-20 Although this problem should have been an important issue for the scientists involved in development of radiopharmaceuticals, it has not been addressed adequately, perhaps because of the difficulties in separating the two isotopes in an efficient manner and in counting ultralow levels of the pure β- emitter 90Sr with accuracy. The few reported procedures were found to be grossly inadequate to guarantee the stringent limits for 90Sr in the 90Y. For instance, some researchers have precipitated 90Sr in suitable forms, which was subsequently counted using liquid scintillation spectrometry,13 while some others have used paper chromatography and paper electrophoresis techniques for determining 90Sr breakthrough.14-17 Precipitation methods are generally seen to be associated with large errors and hence cannot be relied on for determination of a few counts of 90Sr in very high amounts of 90Y. The reported paper chromatography and paper electrophoresis methods based on common reagents such as saline were also tried and found to be highly nonreproducible, with poor delineation between the two isotopes. The high separation factors required for determination of even few counts of 90Sr in the 90Y milked from the generator was seen to be lacking in these methods. Another quality control procedure routinely followed in many laboratories involves the determination of 90Sr after complete decay of 90Y.18,19 This is a (12) Yttrium-90, Radiochemical yttrium chloride solution. Y-90 fact sheet. MDSNordion. http://www.nordion.net/documents/elibrary/molecular-isotopes/ Y-90/Y-90_Can.pdf. (13) Salutsky, M. L.; Kirly, H. W. Anal. Chem. 1955, 27, 567-569. (14) Doering, R. F.; Tucker, W. D.; Stang, Jr. L. G. J. Inorg. Nucl. Chem. 1960, 15, 215-221. (15) Chinol, M.; Hnatowich, J. D. J. Nucl. Med. 1987, 28, 1465-1470. (16) Ro¨sch, F.; Beyer, G. J.; Scha¨fer, G.; Poser, D. Annu. Rep. 1989 ZFK Rossendorf 1990, 711, 66-70. (17) Malja, S.; Schomacker, K.; Bylyku, E. Fifth General Conference of the Balkan Physical Union 2003. http://www.phy.bg.ac.yu/jdf/bpu5/proceedings/ Papers/SP15%20-%20008. pdf. (18) Vaughan, A. T. M.; Keeling, A.; Yankuba, S. C. S. Int. J. Appl. Radiat. Isot. 1985, 36, 803-806. (19) Dietz, L. M.; Horwitz, E. P. Appl. Radiat. Isot. 1992, 43, 1093-1101. (20) Vanura, P.; Jedinakova-Krazova Vobecka, M. J. Radioanal. Nucl. Chem. 2006, 267, 501-503.

802

Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

reliable method where 90Sr levels are estimated by liquid scintillation counting of 90Y samples that have completely decayed (∼2 months old) to quantify the 90Sr (T1/2 28.8 Y); 61 days of decay corresponds to 22.875 half-lives, which in turn corresponds to 1.16 × 10-10 fraction of 90Y remaining. This will mean that >99.99999% of 90Y would have decayed (T1/2 64.1 h). At this time, ∼50% of the dpm obtained will correspond to the 90Sr contamination in the starting material. Such a study performed on a sample of 90Y obtained from a commercial manufacturer revealed an above permissible level of 90Sr.21 Although the above-described postdecay estimation method is reliable, it cannot be used as a real-time QC technique by a hospital radiopharmacist prior to the patient administration of the radiopharmaceutical. This procedure is more for records and strictly not a quality “control” test. It will not help in intervention if the 90Sr content is found to be unacceptable. Hence, it is highly desirable to develop a reliable quality control technique for estimation of 90Sr in 90Y solution. The International Atomic Energy Agency (IAEA) has initiated a Coordinated Research Project (CRP) on “Development of Generator Technologies for Therapeutic Radionuclides” in which our group at BARC is a participant. A major problem identified during the course of the CRP was the need to develop a reliable quality control technique to estimate the radionuclidic purity of 90Y used for radionuclide therapy. This capability becomes very important when locally or in-house produced 90Sr/90Y generators are used to obtain 90Y rather than being obtained from a commercial manufacturer. Hence, the authors explored the possible ways to address this problem and have developed the reported method, extraction paper chromatography, which is a combination of solvent extraction and paper chromatography. The selectivity of chelating agents for adsorption of metals has been successfully applied in radiochemistry for separation and purification of radionuclides using liquid-liquid extraction chromatography.22 2-Ethylhexyl-2-ethylhexylphosphonic acid (KSM17), the chelating agent, was originally synthesized at the Bhabha Atomic Research Centre (BARC) to be used for the separation of 90Y since it has very high selectivity for 90Y and has absolutely no selectivity for 90Sr from acidic aqueous media.23,24 A 90Sr/90Y generator system based on the selectivity of KSM-17 for 90Y was developed by our group.25,26 In the present work, we exploited the selectivity of KSM-17 for 90Y to develop a novel paper chromatographic technique for determination of 90Sr contamination in the eluted 90Y. MATERIALS 90Sr in equilibrium with 90Y in nitrate form was obtained from the Fuel Reprocessing Division of the Bhabha Atomic Research (21) Chinol, M. Report of 2nd Research Coordination of the CRP on Development of generator technologies for therapeutic radionuclides. IAEA 2006, 6567. (22) Gatrone, R. C.; Kaplan, L.; Horwitz, E. P. Solvent Extr. Ion Exch. 1987, 5, 1075-1116. (23) Marwah, U. R. Proc. National Symposium on Organic Reagents Syntheses and use in extraction metallurgy. (ORSEUM-94), BARC, Mumbai 23. 1994. (24) Achuthan, P. V.; Dhami, P. S.; Kannan, R.; Gopalakrishnan, V.; Ramanujam, A. Sep. Sci. Technol. 2000, 35, 261-270. (25) Venkatesh, M.; Pandey, U.; Dhami, P. S.; Kannan, R.; Achuthan, P. V.; Chitnis, R. R.; Gopalakrishnan, V.; Banerjee, S.; Samuel, G.; Pillai, M. R. A.; Ramanujam, A. Radiochim. Acta 2001, 89, 413-417. (26) Dhami, P. S.; Naik, P. W.; Dudwadkar, N. L.; Kannan, R.; Achuthan, P. V.; Moorthy, A. D.; Jambunathan, U.; Munshi, S. K.; Pandey, U.; Venkatesh, M.; Dey, P. K. Sep. Sci. Technol. 2007, 42, 1107-1121.

Centre.27 Strontium-85+89 (85+89Sr) nitrate, a regular product prepared for commercial supply by Radiopharmaceuticals Division, BARC, was used as a tracer in the initial studies. 90Yttrium acetate was eluted from a two-stage KSM-17-based liquid membrane 90Sr-90Y generator.26 KSM-17 synthesized at BARC in the past was used in these experiments. All reagents used were of Analytical Reagent grade. The scintillation cocktail used for liquid scintillation studies consisted of 900 mL of dioxane, 100 g of naphthalene, and 1.2 g of PPO (2,5-diphenyloxazole). Paper chromatography strips were purchased from M/s. Whatman, UK. Semiautomatic micropipets with high accuracy (>99%) and precision (>99%), were used in all experiments to pipet radioactive solutions. γ Activity of 85+89Sr was assayed by using a NaI (Tl) scintillation counter (500-700 keV). Activity of 90Y and 90Sr was measured in an ionization chamber when used at MBq levels. A NaI (Tl) scintillation counter wherein the Bremmstrahlung radiation was counted (50-500 keV) was used in most counting experiments. A liquid scintillation counter (model, Tricarb 2100TR, Packard Instrument Co., USA) was used for the final measurement of 90Sr activity in Bq levels. METHODS Extraction Paper Chromatography (EPC) Technique. Whatman 1 chromatography paper was cut into 12 × 1 cm size. Then ∼10 µL of KSM-17 was applied at the origin of the paper chromatography strips and allowed to air-dry. These strips were used for the EPC technique. The chromatography technique involves the spotting of 5 µL of the test solution over the dried spot of KSM-17, allowing it to dry, and developing the paper in 0.9% saline solution by ascending mode. After movement of the solvent front to the top of the paper, the paper was removed, dried, and cut into 1-cm segments. The activity at different segments of the paper was measured using a NaI (Tl) solid scintillation counter in most experiments during the standardization procedure of the technique. In the final experiments where the QC of the generatoreluted product was performed, the counting of the 90Sr activity was done in a liquid scintillation counter. The migration behavior of different radiochemical species such as yttrium-90 (90Y) chloride, 90Y acetate, and 85/89Sr acetate, a mixture containing 90Sr and 90Y was studied by the EPC technique. Quality Control of 90Y Acetate. The 90Sr content of 90Y eluate was determined by preparing a 100 µL “test solution” of the eluate containing 1 mCi/mL (37 MBq/mL) by dilution of the stock 90Y solution in 0.5 M ammonium acetate. A 5 µL sample of the 90Y test solution containing ∼5 µCi (185 kBq) of 90Y was spotted on chromatography strip and developed as described earlier. The paper strip was dried, cut into 1-cm pieces, and the last three segments were counted using a LSC after placing in 10 mL of cocktail in a scintillation vial. The activity obtained in the solvent front was compared with the total spotted activity (as known from the radioactive concentration of the test solution, in terms of count rate when counted in the same LSC) to determine the amount of 90Sr present in the 90Y sample. RESULTS Migration Pattern of Different Species in the Extraction Paper Chroatography. 85+89Sr Acetate. Figure 1a illustrates the (27) Ramanujam, A.; Dhami, P. S.; Chitnis, R. R.; Achuthan, P. V.; Kannan, R.; Gopalakrishnan, V.; Balu, K. BARC report 2000/E/009, 2000.

Figure 1. Paper chromatography pattern (a) and extraction paper chromatography (b) of 85+89Sr.

chromatographic pattern of 85+89Sr acetate, which moves to the solvent front in the conventional paper chromatography. Figure 1b shows the chromatography pattern of 85+89Sr acetate in the EPC. The 85+89Sr moves toward the solvent front with an Rf ) 0.9-1.0. There is negligible activity remaining at the origin, indicating that under the conditions of the chromatographic studies there is no extraction and thereby retention of strontium by KSM-17. As seen, there is very little difference in the chromatographic pattern of 85+89Sr with or without KSM-17 impregnation at the origin although the EPC pattern is marginally better with less trailing of activity. These experiments were performed using the 85+89Sr tracer, since it was difficult to obtain pure 90Sr, which always contains 90Y in varying amounts depending upon the time elapsed after purification. Even a freshly prepared 90Sr sample will still contain traces of 90Y either due to incomplete separation or the radioactive decay. Another advantage of using the 85+89Sr tracer is that γ counting can be performed with a NaI (Tl) solid scintillation counter for estimating activity levels. 90Y Acetate. Figure 2a represents the paper chromatography pattern of 90Y acetate solution eluted from the 90Sr/90Y generator without KSM 17 impregnation. The migration pattern of 90Y acetate solution was not consistent in different exeperiments. At times, a Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

803

Figure 3. Extraction paper chromatography (EPC) pattern of 90Sr-90Y mixture.

Figure 2. Paper chromatography pattern (a) and extraction paper chromatography (b) of 90Y.

major portion remained at the point of spotting, while in other experiments, it moved to the solvent front to varying extent. The different migratory pattern may be due to the presence of multiple 90Y species in the solution depending upon the concentration of ions such as chloride and acetate. Figure 2b illustrates the migration pattern of the same 90Y acetate solution in EPC. The 90Y remains at the point of spotting and does not migrate outside the KSM-17 spot in the paper. It can be seen by comparison of the above two chromatography results that, in the presence of KSM-17, the 90Y activity is fully retained at the origin (Rf ) 0) and there is no 90Y activity moving beyond the KSM-17 impregnation spot. The migratory pattern of 90Sr and 90Y in EPC indicates that the presence of KSM-17 facilitated a clean separation between them. 90Sr and 90Y Mixture. In order to test the separation efficiency of the EPC in a mixture containing both 90Sr and 90Y, EPC was carried out in an equilibrium sample containing 90Sr and 90Y, and the results are shown in Figure 3. As expected, the 90Y remained at the origin at the KSM-17 impregnation spot and the 90Sr exclusively migrated to the solvent front. The activity is distributed almost in the ratio of 50:50 between the point of spotting and the solvent front. The separation is very clean with no detectable 804 Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

activity between the two radioactive components. These results confirm the ability of the EPC technique to clearly separate 90Sr from 90Y. 90Y Acetate/Chloride Eluate from the Generator. The 90Y obtained from a 90Sr-90Y generator26 was used in this EPC technique to evaluate this new procedure for the estimation of the RN purity of 90Y. Figure 4a shows the chromatography pattern of 90Y eluate from a 90Sr/90Y generator. The counting of the activity was performed using liquid scintillation counting by directly adding the strips to the liquid scintillation cocktail. The counts at the solvent front were negligible compared to the 90Y counts at the point of spotting (Figure 4a). However, because of the very strict limits on 90Sr in 90Y, quantification of 90Sr contamination is important and hence needs to be carefully determined. Because of the very low levels of 90Sr, use of a semilog Y-axis scale provides a better illustration. Figure 4b shows the same data redrawn in a 7-cycle semilog scale to show the presence of 90Sr in the 90Y solution. It can be seen that the solvent front has 30 counts/min, which represents the extremely low levels of 90Sr present in the 90Y sample analyzed. The point of spotting corresponding to 90Y had 1 250 000 counts/min. From the results, it can be seen that 90Sr contamination in this 90Y sample was 0.0025%. Recommended Method for Estimation of the Radionuclidic Purity of 90Y Chloride/Acetate. Based on the results above, the following protocol has been designed for the estimation of the radionuclide purity of 90Y solution. Step 1: Preparation of Test Solution. A 100 µL of 90Y acetate test solution with a radioactive concentration of 37 MBq/mL (1.0 mCi/mL) is prepared by dilution of the bulk solution with 0.5 M ammonium acetate. Please note that 90YCl3 is available in various radioactive concentrations depending upon the supplier; e.g., MDS Nordion provides 9.25 ( 1.85 GBq/mL (250 ( 50 mCi/mL). Step 2: Preparation of EPC Paper. Whatman 1 chromatography paper (12 × 1 cm) marked into 1-cm segments is taken and 10 µL of KSM 17 is applied between the second and third segments. The paper is then allowed to air-dry. Step 3: Extraction Paper Chromatography. A 5 µL sample of the test solution prepared in step 1 above is applied on the KSM17 spot on the EPC paper and allowed to air-dry. The EPC is

37 GBq (1 Ci) of 90Y12, which corresponds to 3.7 Bq or 222 dpm of 90Sr per 185 kBq (5 µCi) of the spotted solution. Hence, solvent front activity of e222 dpm corresponds to a 90Sr contamination of e740 kBq (20 µCi) per 37 GBq (1 Ci) of 90Y, and the product is consistent with the manufacturers’ specification. Lower cutoff limits can be set as deemed necessary. Step 6: Validation of the Technique. This is an optional step to be followed for validation of the EPC technique. The 5 µL aliquots of the test solution are dispensed into a liquid scintillation vial. The vial is marked with the batch number and stored for ∼60 days (∼22.5 half-life), by which time it is expected that 99.99999% of the 90Y decays and only any 90Sr contamination is present. At this point, 90Sr present will be in secular equilibrium with 90Y. The vial is then counted in the liquid scintillation counter for 60 min. The counts correspond to any 90Sr contamination and an equal amount of 90Y. If gross counting is performed to include both 90Y and 90Sr windows, the activity measured can be halved to obtain the 90Sr activity (assuming that there is not much difference in the efficiency for 90Sr and 90Y). The activity thus calculated for 90Sr should be equal to the activity in the solvent front obtained in EPC (step 4 above).

Figure 4. Extraction paper chromatography of 90Y (a) and the same data redrawn in semilog (b ).

developed in ascending manner by inserting the paper in a chromatography jar containing 0.9% saline. Care is taken to ensure that the solvent level is below the 1-cm mark of the paper. After the solvent has moved to the top, the paper is removed, cut into 1-cm segments, and three segments of the solvent front are inserted in a liquid scintillation vial containing 10 mL of cocktail. The samples are counted for 60 min. Please note that the counting needs to be done immediately after completion of the experiment. The counts in the 90Sr fraction will keep increasing on storage due to the growth of 90Y. In such cases, 90Sr will be overestimated. Alternatively, the growth can be calculated depending upon the elapsed time after separation and a correction applied. Step 4: Calculation of Results. Using the counter efficiency, the activity at the solvent front is calculated. The test solution contained 37 MBq/mL (1.0 mCi/mL), and hence, the original activity used in EPC was 1.85 × 105 Bq (5 µCi). The percent RN impurity is calculated from the results. For example, if 6000 counts were obtained in the solvent front for a 60 min counting, this corresponds to 100 counts/min, and if the efficiency of the counter is 90%, the activity is 111 dpm or 1.85 Bq. The radionuclidic impurity in this case is 0.001%, and the RN purity of the product is 99.999%. The solution thus contains 370 kBq (10 µCi) of 90Sr in 37 GBq (1 Ci) of 90Y. Step 5: Setting 90Sr Impurity Limit for 90Y QC. The limit quoted by the commercial manufacturers of 90Y is 740 kBq (20 µCi) per

DISCUSSION 90Y, an important therapeutic radionuclide, is available from 90Sr/90Y generator systems. Several types of 90Sr/90Y generator systems have been reported in the literature for elution of 90Y. The extremely high toxicity of 90Sr limits its levels in the 90Y used and necessitates its absolute quantification before administering the 90Y radiopharmaceutical to the patient. Although a few methods have been reported by researchers for determination of 90Sr breakthrough in the 90Y eluates, none of these have the absolute specificity for either 90Y or 90Sr, necessary to detect very low amounts of 90Sr in 90Y eluates. While methods based on complete decay of 90Y are reliable, they are not very useful for a hospital radiopharmacy; other methods such as precipitation, paper chromatography or paper electrophoresis are not adequately sensitive or selective. A method based on high separation factor for 90Y from 90Sr would be necessary to achieve the desired sensitivity since it is expected to measure 90Sr at (95% is generally acceptable for most QC requirements. Hence, the chromatography techniques are designed to handle counts in the range of ∼50 000-100 000 counts/ min, and a radiochemical species present at ∼5% can be measured with an error less than ∼1.4%, which is adequate. However, in order to count 90Sr with acceptable accuracy at a level of 10-4%, the amount of 90Sr activity detected has to be high, which in turn requires severalfold higher 90Y activity to be applied for testing. It would not only be difficult to handle high amounts of radioactivity for QC purposes, but the errors in 90Y measured will also increase due to excessive count rate and dead time losses. We have developed an innovative approach for quantification of the activity of the two fractions. The solution to be analyzed is prepared with a known radioactive concentration of 90Y, which is easily accomplished since the supplier quotes the radioactive concentration on the product specification sheet. Dilution of the product in ammonium acetate provides a solution in acetate form for chromatography and the desired radioactive concentration. The 100-µL sample of a test solution of 37 MBq/mL (1 mCi/mL) is made, and 5 µL (185 kBq, 5 µCi) of this is used for spotting. Hence, the total activity spotted in the EPC is predecided, 185 kBq (5 µCi) in this case. The quantification of 90Sr activity is performed by liquid scintillation counting to use the advantage of high efficiency (>90%) and low background