Medical Isotope Production Without Highly Enriched Uranium (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

8 Conversion to LEU-Based Production of Molybdenum-99: Regulatory Considerations T he objective of this chapter is to describe and discuss the important regulatory considerations for conversion of Mo-99 production from highly enriched uranium (HEU) to low enriched uranium (LEU). This chapter is also intended to support the discussion of conversion feasi- bility that appears in Chapter 10. This chapter will focus on the following three regulatory issues: 1. Physical security for HEU; 2. Drug quality and purity; and 3. Commercial sale of radiopharmaceuticals manufactured from Mo-99. The committee selected these regulatory considerations for discussion because it judged that they had the greatest potential to impact Mo-99 producers’ decisions to convert to LEU-based production. PHYSICAL SECURITY FOR HEU At the beginning of this study, the committee hypothesized that HEU- based medical isotope producers might reap substantial savings in security costs by converting to LEU-based production systems. Civilian nuclear fuel cycle facilities that handle Formula Quantities of Special Nuclear ­Materials   Special Nuclear Material is defined in Title 1 of the Atomic Energy Act of 1954. It includes plutonium, uranium-233, or uranium enriched in the isotopes uranium-233 or uranium-235. The formula quantity for HEU is a quantity greater than 5 kg. 101 

102 MEDICAL ISOTOPE PRODUCTION WITHOUT HIGHLY ENRICHED URANIUM are required to establish security plans and systems to prevent theft, diver- sion, or radiological sabotage of HEU. Security guidelines are established by the International Atomic Energy Agency (IAEA, 1999) and promulgated in part or in whole in regulations by national authorities. In the United States, for example, civilian facilities possessing formula quantities of ­special ­nuclear materials fall under the authority of the Nuclear Regulatory Commission and must meet the requirements in Title 10, Part 73 of the Code of Federal Regu- lations (10 CFR Part 73) entitled Physical Protection of Plants and ­Materials and also 10 CFR Part 74 entitled ­Material Control and Account­ing of Spe- cial Nuclear Materials. The regulations require that each facility have access controls, physical barriers, armed guards, and ­material inventory systems to secure special nuclear materials. These security systems are costly, and so the committee hypothesized that substantial cost savings might be realized by converting to LEU-based production ­ because LEU does not fall under the same formula quantity requirements. After visiting HEU and LEU production and potential production facili- ties and discussing security requirements with facility staff and national regulators, the committee concluded that the cost savings from conversion of existing HEU-based production to LEU-based production would likely be small, primarily for the following reasons: 1. Many Mo-99 producers utilize facilities that are located on multi­ purpose sites. These sites are required to have high security because they contain sensitive facilities or store HEU. For example, the Atomic Energy Canada Ltd. (AECL) Chalk River site in Ontario, Canada, has HEU spent fuel and HEU waste from the past production of Mo-99. The ANSTO site in Australia has HEU fuel onsite from a shutdown reactor. High security will be required as long as this HEU remains on site. 2. Current HEU-based producers may possess less than formula quan- tities of HEU at their facilities or are exempt from the security regulations that govern formula quantities. HEU is shipped to the target manufacturers   Small groups of committee members and staff visited major HEU-based production facilities in Canada (Atomic Energy of Canada Limited [AECL]), Belgium (Institut ­National des ­Radioéléments), and the Netherlands (Petten); LEU-based production facilities in ­Australia ­(Australian Nuclear Sci- ence and Technology Organisation [ANSTO]]) and Argentina ­ (Comisión Nactional de Energía Atómica [CNEA]); one potential domestic production facility in ­ Missouri (Missouri University Research Reactor [MURR]); and a fuel manufacturing facility in France (Compagnie pour l’ Etude et la Réalisation de Combustibles Atomiques [CERCA]). See Appendix C.   This discussion does not address the nonproliferation benefits of civilian HEU elimination, which was the primary motivation behind the Schumer Amendment (see Sidebar 1.3). See Chapter 11 for a discussion of HEU minimization efforts.   For example, 10 CFR Part 73, which regulates facilities that contain formula quantities of HEU (Category 1 facilities) does not apply to research reactor facilities in the United States (e.g., MURR) even if they possess quantities of HEU greater than formula quantities. 

REGULATORY CONSIDERATIONS 103 (e.g., AECL in Canada; CERCA in France), which manufacture and store the HEU and targets until they are needed by the isotope producers. The targets can be shipped to the reactors in less-than-formula quantities. 3. The security requirements for all nuclear facilities, including the research and test reactors that are used to irradiate targets for medical iso- tope production, were raised in many countries, including the United States, following the September 11, 2001, terrorist attacks on the United States. Consequently, the costs of security have increased even for facilities that use only LEU for medical isotope production or other purposes. However, these costs are still lower than those for facilities that store greater than formula quantities of HEU. DRUG QUALITY AND PURITY The second issue of concern to producers is the regulatory require- ments for drug quality and purity. Some producers have questioned whether Mo-99 made from LEU targets will have the same quality and consistency as that made from HEU targets. Experience to date with LEU-based produc- tion indicates that Mo-99 purity and consistency should not be an impedi­ ment to conversion. ANSTO produced Mo-99 for medical isotope use with 1.8–2.2 percent LEU targets until 2007, when it shut down its HEU-fueled r ­ eactor (High Flux Australian Reactor) and prepared to start up its LEU- fueled replacement reactor (Open Pool Australian Lightwater reactor) and produce Mo-99 using 19.75 percent LEU targets. ANSTO ­ reported to the committee that the Mo-99 produced from the 1.8–2.2 percent LEU targets, and Mo-99 produced from test batches of 19.75 percent LEU targets, had lower ­ impurities than HEU-based Mo-99 and met ­ British ­ Pharmacopeia ­imits for impurities. ANSTO was carrying out low-­activity Mo-99 produc- l tion ­trials as the present report was being finalized for ­release. A representa- tive of ANSTO reported that the quality of Mo-99 from these runs was high and equivalent to the quality of HEU-based Mo-99 it was receiving from large-scale commercial suppliers. CNEA has been producing Mo-99 using 19.75 percent LEU targets since 2002. A representative of that organization told the committee that Mo-99 purity has been consistently higher than that produced using HEU targets. Purity data for CNEA-­produced Mo-99 is pre- sented by Durán (2005).   The costs of transporting larger quantities of HEU from storage to target producers would likely be significantly higher than the costs of transporting LEU. However, such transport o ­ ccurs relatively infrequently compared to transport of targets.   There is no U.S. Pharmacopeia (USP) for Mo-99 because it is not used for diagnostic imag- ing procedures. However, there is a USP for Tc-99m.   Ian Turner, ANSTO, written communication with study director Kevin Crowley, Decem- ber 10, 2008. 

104 MEDICAL ISOTOPE PRODUCTION WITHOUT HIGHLY ENRICHED URANIUM The presence of higher concentrations of alpha emitters in LEU pro- cess streams has been cited by some producers as a potential conversion uncertainty (see Vandegrift, 2005). Irradiated LEU targets will contain higher concentrations of neptunium-239 (Np-239) and its daughter ­product ­ lutonium-239 (Pu-239) than equivalent HEU targets. However, HEU p targets contain higher concentrations of uranium-234 (U-234), which has a higher activity and shorter half-life than either U-238 or U-235. The total concentrations of alpha-emitting isotopes are not appreciably differ- ent in either target type, and both uranium and plutonium isotopes can be effectively removed during target processing. COMMERCIAL SALE OF RADIOPHARMACEUTICALS The third issue of concern to producers involves regulatory approvals for commercial sale of radiopharmaceuticals manufactured from Mo-99. This issue was brought to the committee’s attention by the Council on Radio­nuclides and Radiopharmaceuticals (CORAR)10 at the committee’s first meeting (Appendix C) and was characterized by a representative of that organization as a potentially significant barrier to conversion, especially for radiopharmaceuticals that are used in the United States. The issue was also raised by some other producers at the committee’s subsequent meetings. Because the focus of this report is Mo-99 production and use in the United States, this discussion will focus on U.S. regulatory processes. How- ever, similar processes are used by regulatory agencies in other countries. In the United States, the Food and Drug Administration (FDA) is responsible for regulating the production and use of medical isotopes. The FDA is responsible for approving the use of Tc-99m in radiolabeled compounds intended for human use, but not the production of the Mo-99 precursor. However, when the process for producing Mo-99 is changed, the FDA must approve the use of Tc-99m derived from that isotope. The approval process is ­described briefly in Sidebar 8.1. A current technetium generator producer who wanted to utilize a new source11 of Mo-99 would be required to submit a Supplemental New Drug   LEU targets contain more U-238 than HEU targets. Neutron capture by U-238 during target irradiation produces small amounts of Np-239, which decays with about a 2.3-day half-life to produce Pu-239. The half-life of U-234 is 2.45 × 105 years, versus 7.04 × 108 years for U-235 and   4.47 × 109 years for U-238. 10 CORAR is an association of North American companies involved in the manufacture and distribution of radionuclides, radiopharmaceuticals, and sealed sources for medicine and life science research. 11 For the purposes of this discussion, a “new source” includes Mo-99 obtained from a new supplier and/or from a new production process such as an LEU-based process. 

REGULATORY CONSIDERATIONS 105 SIDEBAR 8.1 FDA Approval Process Technetium-99m, used in radiolabeled compounds intended for human use, is regulated by the FDA under the Food, Drug, and Cosmetic Act. The produc- tion of the Mo-99 precursor to Tc-99m is not regulated by the FDA if it is used only as a radiochemical. However, if the Mo-99 is to be used to make Tc-99m for radiolabeled compounds, its producer typically submits a Drug Master File (DMF) to the FDA; however, a DMF is not strictly required. The DMF describes the facil- ity in which the Mo-99 is made; the production process itself, including any raw materials used in production; and product test methods, specifications, stability, and release criteria. The DMF is not approved by the FDA; instead, it is used as a source of information when FDA approval is sought to sell Tc-99m radiolabeled compounds made with that producer’s Mo-99. A company seeking to sell a radiolabeled compound (e.g., a technetium generator producer) is required to submit an NDA to the FDA and pay a one- time application fee (in 2008, this fee was $1,178,000). The NDA is tied to one or more specific DMFs; the NDA for a radiolabeled compound, for example, would be tied to the DMFs for Mo-99 and any other raw materials used to make that compound. Like the DMF, the NDA describes the facilities, processes, test m ­ ethods, and specifications for producing the radiolabeled compound. The FDA must review and approve the NDA before that radiolabeled compound can be sold for human use. When a Mo-99 producer makes major changes to the process or raw m ­ aterials it uses to make that isotope, it submits an updated DMF to the FDA. Any company (e.g., a technetium generator producer) that wants to use the Mo-99 produced under this updated DMF may find it necessary to submit an sNDA to obtain FDA approval to use that isotope. There is no fee for this submission, but there is a cost to the company for preparing the sNDA (described elsewhere in this chapter). To obtain FDA approval of the sNDA, the company must demon- strate that the Mo-99 precursor and Tc-99m product derived from it meet product specifications on three full production batches of Mo-99. A single production batch for a large-scale producer can contain hundreds to thousands of 6-day curies recovered from multiple targets. An sNDA is required any time there are significant changes to the Mo-99 production process. However, if the changes to the Mo-99 production process are judged by the company to be minor, it could elect to submit a Change Being Effected (CBE) notification to the FDA instead of an sNDA. The CBE informs the FDA about the change but does not provide analytical testing data. The FDA would review the CBE and could approve it or direct the company to submit an sNDA. Application (sNDA) to the FDA (see Sidebar 8.1). The committee was told by representatives of CORAR (see also Brown, 2005) and some Mo-99 producers that a great deal of time and effort would likely be required to develop and submit an sNDA, particularly to support the three required production runs to test the new product: Protocols must be developed for 

106 MEDICAL ISOTOPE PRODUCTION WITHOUT HIGHLY ENRICHED URANIUM the production runs; generators must be prepared; the Tc-99m must be eluted; radiopharmaceuticals must be prepared and tested; all of this infor- mation must be compiled; and the sNDA must be written and submitted. The committee received presentations from industry and the FDA (see Appendix C) concerning the time, cost, and uncertainties for regulatory approvals. Perhaps the most striking aspect of the presentations was the vast difference in what industry representatives expected from the FDA—a complex, tedious, expensive, and unpredictable process—and the simple, straightforward, and readily achievable approval process described by the FDA presenter. Industry representatives provided the committee with several examples of the difficulties they have encountered in obtaining FDA approvals for new sources of Mo-99. These included a “difficult” process for obtaining approval for a backup supplier of Mo-99 who used the same production and processing protocols in an already-approved NDA (it was reported to the committee that approval took almost a year and cost more than $200,000), and another approval that took almost 2 years. A representa- tive of CORAR suggested that the FDA could require clinical trials before it would approve the use of LEU-based Mo-99. The FDA presenter told the committee that the review time for an NDA typically takes between 6 and 10 months. He also noted that engaging the FDA early during the process of developing the NDA can help ensure that the approval process runs smoothly. A consultant working for MURR who has long experience with the FDA approval process estimated it would take a minimum of about 4–6 months after submission of the necessary paper- work and cost about $84,000 to obtain approval for using Mo-99 from a new LEU-based process at the MURR reactor (MURR, 2006). A current Mo-99 producer told the committee that not all FDA approvals require long lead times. This producer obtained emergency approval of a backup Mo-99 supply in less than a week. Technetium generator producers are well acquainted with the FDA approval process and have a good understanding of its requirements. If LEU-based Mo-99 can be produced with similar chemical characteristics similar to HEU-based Mo-99—and current experience in Argentina and Australia indicates that it can—it is hard for the committee to see any rational basis for expectations of substantial delays in FDA approvals if producers submit high-quality sNDAs and work with FDA staff throughout the approval process. It is especially difficult for the committee to see how the FDA would ever require clinical trials as part of an sNDA for a new Mo-99 source. Mo-99 is a well-known isotope that can be produced with low impurities using either an HEU- or LEU-based process. Clinical trials would be a useless exercise in any case because they can be used to detect only gross adverse drug effects. 

REGULATORY CONSIDERATIONS 107 Based on information provided to the committee by industry and the FDA, it seems likely that regulatory approval for new sources of LEU- p ­ roduced Mo-99/Tc-99m would require at least 4 months and as long as 18 months depending on the quality of the application and issues raised by the FDA during the review process. The cost of the process is difficult to estimate but would likely be in the range of multiple tens to hundreds of thousands of dollars. It is important to recognize that these cost estimates represent only the direct costs for regulatory approvals. There are also likely to be ­ indirect costs for such approvals, including, for example, any opportunity costs associated with lost sales of Mo-99, technetium generators, or radio­ pharmaceutical kits as a result of the regulatory process. However, these regulatory costs are likely to be small in comparison to the physical costs of conversion. FINDINGS Three important regulatory considerations for converting Mo-99 pro- duction from HEU targets to LEU targets are described and discussed in this chapter: (1) physical security for HEU, (2) drug quality and purity, and (3) commercial sale of radiopharmaceuticals manufactured from Mo-99. On the basis of this information, the committee finds that: • Converting from HEU- to LEU-based production is unlikely to pro- duce substantial savings in security costs, including transportation security costs. • The purity of Mo-99 produced from HEU targets and LEU targets is not significantly different. Mo-99 produced from LEU targets using stan- dard production methods and practices can meet regulatory requirements for use in radiopharmaceutical production. • FDA approval for LEU-based production of Mo-99 should not be a substantial barrier to conversion. Such approvals would require at least 4 months and as long as 18 months depending on the quality of the appli­ cation and issues raised by the FDA during the review process. The cost of the process is likely to be in the range of multiple tens to hundreds of thousands of dollars. Clinical testing is unlikely to be required by the FDA for such approvals.