Guest Editorial by LEK Consulting
In radiopharmaceutical manufacturing, the real challenge is whether stated capacity can actually deliver reliable, end-to-end supply to patients under the tight timing constraints imposed by radioactive decay.
Headline capacity routinely overstates what can be reliably delivered to patients. Public figures often collapse fundamentally different forms of capacity into a single number, obscuring whether that capacity is regulatorily suitable, operationally reliable, geographically reachable, or compatible with the isotope in question.
Once radioactive decay and radiochemical handling requirements bind these constraints in real time, nominal capacity fragments into narrow production windows shaped by cooldown, decontamination, quality release, and transport sequencing.
This mismatch becomes especially visible as therapeutic programs transition to late-stage development and commercial supply. Manufacturing models that were adequate in early clinical phases are now being stress-tested, not because execution has deteriorated, but because operating requirements have changed. Constraints that were tolerable at low volumes become binding once programs must support reliable production at scale.
Reliability, not throughput, defines performance
Predictable execution within timing constraints matters more than theoretical capacity numbers.
Traditional pharmaceutical manufacturing is optimized around throughput. Inventory buffers absorb variability, production can be decoupled from administration, and short disruptions rarely destroy product value. Therapeutic radiopharmaceuticals operate differently. Finished product value decays continuously, quality windows compress, and delays that would be manageable in small molecules or biologics can render a batch unusable.
Performance is governed less by how much capacity exists on paper than by whether the manufacturing network can execute predictably across production, release, and delivery within narrow timing constraints. Apparent underutilization often reflects capacity reserved to preserve reliability rather than surplus availability. Networks optimized for utilization rather than reliable execution under time decay degrade rapidly once real-world variance enters the equation.
Manufacturing performance emerges from a distributed network
Geographic footprint, site proximity, and partner coordination determine performance more than infrastructure.
Radiopharmaceutical manufacturing performance cannot be evaluated at the site level in isolation. Output emerges from a distributed manufacturing network spanning isotope supply, production, quality release, logistics, and clinical scheduling, all operating under time decay. A facility may appear well equipped yet still fail to deliver if release timelines slip, transport routes are fragile, or clinical schedules shift.
Optimization therefore shifts away from equipment counts and toward network design: geographic footprint, proximity to treatment sites, redundancy across nodes, and coordination across partners. Manufacturing success increasingly reflects how the network behaves under real operating conditions rather than the theoretical capability of any single site.
Capacity requirements change as assets mature
Programs must support sustained, repeatable production and reliable delivery across a broader treatment footprint.
The gap between early-stage assumptions and late-stage realities widens as assets progress. Early development prioritizes flexibility and speed to first-in-human. Manufacturing setups often rely on small batch sizes, irregular production schedules, limited redundancy, and bespoke planning. These characteristics are appropriate early on.
Late-stage and commercial supply impose different requirements. Programs must support sustained, repeatable production and reliable delivery across a broader treatment footprint, with limited tolerance for missed runs. Scaling is driven less by larger batches than by repeatability and resilience.
Capacity that works for early trials is structurally incapable of supporting late-stage or commercial operations without redesign. When this mismatch surfaces late, options compress, and risk migrates directly to patient access.
Not all capacity is interchangeable
Isotope properties dictate facility requirements. Capacity isn’t automatically fungible across isotopes.
Radiopharmaceutical capacity is stratified by design, shaped by physical, radiochemical, regulatory, and scale constraints that do not exist in most traditional modalities.
Isotope properties dictate shielding requirements, suite geometry, contamination controls, waste handling, and quality-release sequencing, with certain isotopes introducing additional ventilation, containment, or personnel-safety requirements and decay chains that complicate quality control during release. These factors materially limit which facilities can handle which isotopes, even where infrastructure appears superficially similar.
Scale introduces a separate structural constraint. Batch sizes are bounded by radiochemistry yields, activity handling limits, operator exposure, and decay during synthesis and release. Increased demand is met less through scale-up and more through scale-out: higher batch frequency and parallelization across hot cells, suites, and sites.
This shift adds non-linear load across quality, staffing, logistics, and waste systems and is constrained by a limited pool of radiochemists, health physicists, and GMP-qualified personnel, often limiting how quickly parallel operations can be sustained.
Together, these dynamics determine which facilities can support a given asset, at what activity levels, and with what degree of reliability. Capacity is not fungible by default.
Decision lens: defining the manufacturing need before selecting the model
Manufacturing strategy decisions in radiopharma are only coherent if the underlying need is defined explicitly at the outset. The primary question is not who manufactures, but what must be delivered, to what operational standard, and when.
That definition hinges on four key considerations:
- Isotope and process scope: Which isotopes must be handled, whether single or multiple, and with what changeover, contamination, and handling constraints. These choices shape shielding, airflow, waste handling, and practical fungibility across assets.
- Delivery obligation: The level of output that must be sustained reliably, and the tolerance for missed runs or downtime. Early-stage programs may accept episodic production, while late-stage and commercial assets require repeatable delivery and, often, built-in redundancy across sites.
- Regulatory compliance level: Whether the capacity can operate in compliance with the requirements of the intended stage and market, including the applicable GMP framework (e.g., 21 CFR 212 vs. 211), inspection readiness, validation depth, and geographic licensure. Capacity that cannot sustain compliance for the required use is functionally unavailable, regardless of technical capability.
- Timing of availability: When capacity must be operational, not announced. Build-out, validation, inspection, and tech-transfer lead times mean that late availability can undermine otherwise sound manufacturing strategies.
Only once these parameters are defined does the sourcing decision follow logically: whether to build in-house, partner with specific CDMOs, or pursue a staged or hybrid approach.
When manufacturing choices are made before end-state requirements are fully defined, programs introduce significant risk that nominal capacity will not translate into reliable patient supply.
