Isotope supply

Every radiopharmaceutical begins as an isotope made in a reactor or accelerator, and the whole field sits on a genuinely fragile production base. Roughly 95% of the world's molybdenum-99 (the parent of technetium-99m, used in about 30 million SPECT scans a year, 80% of all nuclear medicine imaging) comes from a handful of research reactors in South Africa, the Netherlands, Belgium and Australia; because Mo-99's half-life is 66 hours it cannot be stockpiled, so a reactor outage cancels scans within days (OECD-NEA, 2025 review of medical radioisotope supply).

Lutetium-177 demand rose sharply after Pluvicto's approval, and the OECD-NEA and IAEA both project that non-carrier-added Lu-177 production capacity will not keep pace with clinical demand; a key bottleneck is enriched ytterbium-176 target material, historically sourced mainly from Russia. New capacity is being built to close the gap: Canada's Darlington reactor began isotope production runs (Mo-99, Lu-177, Y-90) from 2025; India's Dhruva reactor supplies domestic Lu-177 and imports the shortfall from Russia; SHINE Technologies (Chrysalis facility, Wisconsin) is targeting over 75% of US Mo-99 demand with a $263 million DOE conditional loan commitment, and is already producing Lu-177; and Eden Radioisotopes has filed a US NRC construction permit for a new Mo-99/Lu-177 facility aiming to supply up to 50% of global Mo-99 demand.

Actinium-225 has its own, much tighter, constraint (see the alpha-therapy front): global supply of roughly 63 GBq/year against rising clinical trial demand. Ideas circulating to fix the underlying fragility include a coordinated international reserve and shared production schedule across the world's medical isotope reactors, Western enrichment capacity for ytterbium-176 so Lu-177 has more than one supplier, and a global isotope supply observatory that forecasts shortages before they hit clinics.