The Alpha Architecture Question: Is Radiopharma Betting on the Wrong Supply Model?
Byron Fitzgerald | Founder, ProGen Search | March 2026
Introduction: The Consensus That Deserves Stress-Testing
If you follow radiopharmaceutical deal flow, the pattern over the past 18 months has been unmistakable. Capital is flooding into Actinium-225. Reactor capacity. Cyclotron buildouts. Sovereign supply agreements. Long-term off-take contracts. The message from the market is clear: Ac-225 is the king of the alpha era, and the race is to lock up supply.
This consensus is not irrational. Ac-225 delivers devastating double-strand DNA breaks, possesses clinical data that has excited both Big Pharma and the FDA, and is the backbone of pivotal programmes like Fusion Pharmaceuticals' AlphaBreak trial. Novartis, AstraZeneca, and BMS have all made aggressive moves to position themselves in alpha therapy - and Ac-225 has been the default isotope at the centre of those bets.
But there is a structural question the market has not fully interrogated:
What if the supply architecture that scales best for alpha therapy is not the one receiving the most capital?
This article examines the emerging tension between the centralised Ac-225 production model and the decentralised Pb-212 generator model. It does not argue that one isotope "wins" and the other "loses." Both will likely serve the market. But the industrial logic behind each is fundamentally different - and the implications for infrastructure investment, commercial strategy, and leadership hiring are significant enough to warrant serious scrutiny.
1. The Ac-225 Model: Centralised Scale-Up
Actinium-225 has a 9.9-day half-life. That sounds manageable on paper - certainly more forgiving than some shorter-lived isotopes. But the supply model it requires is one of tightly coordinated centralised production.
The Ac-225 value chain looks like this:
Production happens at specialised reactors or high-energy cyclotrons - facilities that require enormous capital investment, years of regulatory validation, and highly specialised operational talent.
Manufacturing of the finished radiopharmaceutical occurs at centralised GMP hot-cell facilities, where the isotope is chelated to its targeting vector, quality-tested, and released.
Distribution relies on global aviation networks to move the product from manufacturing site to hospital within a window that gets tighter with every hour of transit.
Scheduling demands decay-aware orchestration - predicting demand, initiating synthesis before orders are confirmed, and coordinating QC release within compressed timelines.
Every step in this chain must work near-perfectly. A delayed flight, a scheduling misalignment, a QC hold - any of these can render a dose therapeutically useless before it reaches the patient.
As the Ac-225 market scales, this coordination burden does not shrink. It compounds. More patients means more doses, more flights, more scheduling complexity, and more failure modes. The model scales up - bigger reactors, faster logistics, tighter coordination windows.
The Investment Landscape
The capital flowing into Ac-225 infrastructure is substantial and accelerating:
Niowave signed a long-term supply agreement with Novartis in February 2026 to reinforce Novartis's alpha pipeline.
PanTera (a joint venture between IBA and SCK CEN) broke ground on a massive Actinium Production Centre in Belgium, with commercial supply expected by 2029.
Cardinal Health and TerraPower Isotopes are now producing Ac-225 on a weekly commercial basis in Indianapolis - the first entity to offer cGMP-compliant Ac-225 at true commercial scale.
NorthStar Medical Radioisotopes began routine weekly Ac-225 production in late 2025.
SpectronRx has commenced operations with validated yearly volumes.
This is real infrastructure. Real investment. Real progress. But it is all predicated on the assumption that centralised production and long-range distribution is the enduring architecture for alpha therapy.
2. The Pb-212 Model: Decentralised Scale-Out
Lead-212 operates under a completely different industrial logic.
Pb-212 itself has a short half-life - approximately 10.6 hours. As a finished patient dose, it is arguably more time-sensitive than Ac-225. So on the surface, it might appear to be the harder isotope to work with.
But the supply architecture upstream of that final dose is where the structural advantage emerges.
Pb-212 is produced via generator systems loaded with Thorium-228 (Th-228), which has a half-life of nearly two years. This changes the logistics equation fundamentally:
Th-228 can be stockpiled. Unlike Ac-225's scarcer precursors, Th-228 is available in larger quantities as a by-product of existing nuclear processes. It can be stored for extended periods and shipped without urgent time pressure.
Generators can be deployed regionally. Small, benchtop-scale generator systems can be installed at regional manufacturing centres or radiopharmacies. These generators continuously produce Radium-224 (Ra-224, half-life ~3.7 days), from which fresh Pb-212 can be eluted on demand.
The active isotope is produced closer to where it is needed. Rather than shipping a decaying finished product across oceans, the system stockpiles stable precursors upstream and generates the short-lived therapeutic isotope at the point of manufacture.
Aviation dependency is reduced. The Th-228 precursor can travel by standard freight. The Pb-212 dose is produced regionally. The long-distance, time-critical shipment of active isotopes - the fragile link in the Ac-225 chain - is largely eliminated.
The model scales out - more generators, closer to the patient, eluted when needed. As volume increases, the coordination burden does not compound in the same way. You add nodes to the network rather than trying to push more throughput through a centralised bottleneck.
Who Is Building This Model?
The Pb-212 supply architecture is not theoretical. It is being built now:
Perspective Therapeutics operates its proprietary VMT-α-GEN generator system. Because Th-228 can be stockpiled for up to two years, Perspective secures precursor material through long-term contracts with the DOE Isotope Programme and deploys generators to its dedicated manufacturing sites. These currently support Phase 1/2a trials including VMT-α-NET, with commercial-scale generator development underway.
Orano Med recently expanded its Drug Development and Preclinical Unit in Plano, Texas, adding 11,000 square feet of dedicated R&D space and doubling GMP production capacity. This supplements their Alpha Therapy Laboratory in Indianapolis, which is focused on industrial-scale Pb-212 production. Orano's AlphaMedix programme (targeting SSTR2-positive neuroendocrine tumours with Pb-212-DOTAMTATE) is progressing through Phase 2.
AdvanCell Isotopes and Thor Medical are developing complementary Th-228 supply and generator technology platforms aimed at enabling broader decentralised production.
These are not speculative plays. They are operational programmes with secured precursor supply, dedicated manufacturing infrastructure, and active clinical trials.
3. The Structural Question: Which Model Survives at Volume?
The consensus view treats Ac-225 as the default alpha-emitter and Pb-212 as a niche alternative. That framing may be backward.
Consider what happens as alpha therapy scales from hundreds of patients to tens of thousands:
Under the Ac-225 model, every incremental patient adds coordination complexity. More centralised production means more aviation logistics, more scheduling conflicts, more decay-in-transit risk. The system gets more fragile as it gets bigger. Legacy CDMO software - ERP systems designed for traditional biologics with shelf-stable inventory - struggles to handle the predictive, rolling-forecast orchestration that a 9.9-day half-life demands.
Under the Pb-212 model, every incremental patient is served by adding capacity at the regional level. The precursor is stockpiled. The generator is local. The dose is produced on demand. The system gets more distributed as it gets bigger - and distribution, in a decay-driven market, is inherently more resilient than centralisation.
This does not mean Pb-212 is without challenges. The 10.6-hour half-life still imposes severe time constraints on manufacturing and delivery. Generator systems must be validated, maintained, and operated by qualified personnel. The regulatory framework for decentralised radiopharmaceutical production is still maturing. And crucially, the clinical data for Pb-212 programmes - while promising - is earlier-stage than some Ac-225 programmes.
But the supply architecture question is independent of clinical outcomes. Even if Ac-225 and Pb-212 prove equally effective for a given indication, the industrial model that delivers doses more reliably at scale will command a structural advantage.
4. The Investment Implication: What Gets Repriced?
If the Pb-212 generator model proves operationally cleaner than the market currently expects, the repricing could be significant.
A large proportion of current alpha-therapy infrastructure investment is predicated on the Ac-225 centralised model being the enduring architecture. Reactor expansions, cyclotron buildouts, and specialised CDMO capacity are all being sized and funded on that assumption.
Should the market discover - through clinical trial logistics, commercial-scale dose delivery, or regulatory evolution - that the decentralised generator model offers a more reliable path to volume, several things follow:
Centralised Ac-225 CDMO capacity may face utilisation pressure earlier than expected.
Vertically integrated companies with regionalised pharmacy networks (such as Telix, which has acquired both upstream cyclotron capability via ARTMS and downstream distribution via its RLS radiopharmacy network) may hold a structural advantage.
Pure-play centralised alpha CDMOs face the risk that their physical infrastructure is optimised for a logistics model that the market outgrows.
To be clear: Ac-225 is not going away. It has clinical momentum, established supply relationships, and Big Pharma commitment. But the market may be underpricing how much of the alpha therapy future belongs to decentralised architectures - and overpricing how much can be reliably delivered through centralised ones.
The most likely outcome is a split market: Ac-225 dominant in high-volume academic and hospital settings with established centralised supply chains, and Pb-212 capturing an expanding share in regional networks, community oncology, and geographies where aviation logistics are unreliable.
The question is whether the current capital allocation reflects that split - or whether it is still pricing Ac-225 as the obvious endgame.
5. The Talent Implication: A Different Kind of Hire
The supply architecture question has a direct and underappreciated consequence for leadership hiring.
If the market moves toward more decentralised, generator-enabled alpha production, the talent profile changes:
Centralised model demands site heads, MSAT engineers, and QC leaders who can operate large-scale hot-cell facilities under the dual regulatory burden of FDA aseptic standards and NRC radiation containment. These are exceptionally scarce profiles - the "Tri-Brid" operators who understand aseptic manufacturing, radiation safety, and GMP simultaneously. Current time-to-fill for these roles sits at 9-12 months.
Decentralised model still requires these profiles at the manufacturing level, but also creates demand for a broader, distributed network of Authorised Nuclear Pharmacists, regional operations leaders, and generator maintenance specialists. The talent geography shifts from a handful of centralised facilities to a wider footprint of regional sites.
In either scenario, the talent constraint remains severe. The radiopharmaceutical sector faces 19.4% vacancy rates for CT technologists and 12.6% for nuclear medicine technologists at the clinical level. The manufacturing layer is even more constrained.
But the shape of the talent challenge differs by model. Centralised Ac-225 operations concentrate the talent problem into a small number of ultra-specialised roles at mega-facilities. Decentralised Pb-212 operations distribute the problem across a larger number of moderately specialised roles at regional sites.
Companies and investors building alpha-therapy strategies should be stress-testing their workforce plans against both architectures - not just the one they are currently funding.
6. Conclusion: The Architecture Matters as Much as the Isotope
The radiopharmaceutical sector is in the middle of its most consequential infrastructure buildout since the commercialisation of Lutathera and Pluvicto. Billions are being deployed. Strategic positions are being locked in.
The question this article raises is not whether alpha therapy will succeed. It will. The clinical data is too compelling, the unmet need too large, and the capital commitment too deep for the sector to stall.
The question is whether the market has correctly identified which supply architecture will dominate at scale - and whether the current capital allocation reflects that reality or an assumption that deserves more scrutiny.
The next moat in alpha therapy may not be who controls the most isotope.
It may be who builds the least fragile delivery model.
Byron Fitzgerald is the Founder of ProGen Search, a specialist executive search and strategic advisory firm focused on biopharmaceutical manufacturing, CDMO strategy, and leadership capital. For more on how supply architecture shifts are reshaping leadership requirements in radiopharma, contact byron.fitzgerald@progensearch.com or visit progensearch.com.