The GLP-1 Crowd-Out

The most important fact about your ADC, radioligand or novel biologic launch in 2028 may have nothing to do with your molecule. It may be whether you can get a sterile fill-finish slot, and a completely different product class has spent the last three years buying them up.

Every injectable, however exotic its upstream, ends in the same place. A sterile fill line. That line is now the most contested square metre in pharmaceutical manufacturing, and the contest is not close.

Why fill-finish is the throat of the whole system

Upstream gets the attention. Bioreactors, conjugation suites, isotope production, viral vector lines. Each modality has its own glamorous bottleneck, and each has had capital thrown at it.

Downstream is where they all converge. A monoclonal, an ADC, a radioligand and a peptide are radically different molecules with one thing in common: at the end, somebody has to fill and finish them under sterile conditions, often into vials that then need lyophilising or into devices that need assembling.

Roughly 90% of ADCs require lyophilisation. Radiopharmaceuticals need shielded aseptic filling against a decay clock. Complex biologics need isolator lines and the qualified people to run them. These lines are slow to build, three to five years and heavily capital-intensive, and they have to be qualified by a regulator before a single commercial vial comes off them. You cannot conjure one in a quarter.

So the sterile fill line is the throat of the entire system. And the throat is being held.

What the obesity supercycle actually took

GLP-1 receptor agonists are the highest-volume injectable launch modern pharma has seen. Semaglutide, tirzepatide and the wave of candidates behind them are not a niche. They are a mass-market injectable at a scale the sterile supply chain was never sized for.

The response from the two companies at the centre of it has been to buy capacity outright. Eli Lilly has committed multiple billions to new sterile and device sites. Novo Holdings acquired Catalent in 2024, and Novo Nordisk took several of its fill-finish facilities directly, removing a large slice of merchant capacity from the open market in a single move.

The effect on everyone else is simple. A durable, multi-year claim has been placed on sterile fill-finish, on pen and auto-injector assembly, and on the isolator capacity that complex modalities depend on. Some of that is brand-new capacity that adds to the total. A great deal of it is existing capacity that used to be available to a biotech with an ADC and is now spoken for.

This is what weaponised capacity looks like

When a network can choose what to fill, the highest-volume, highest-certainty product wins the slot. That is not a moral failing. It is arithmetic.

A blockbuster GLP-1 running lines flat out, year after year, is a better customer than a 50-batch-a-year ADC programme with an uncertain launch and a complex changeover. The lyophilisation cycle, the cleaning validation between a high-potency ADC and the next product, the shielded handling a radioligand demands: all of it makes the complex modality the more expensive, lower-margin, harder customer to serve.

So capacity stops behaving like a service you can buy and starts behaving like a lever the holder controls. The question is no longer what does a fill slot cost. It is who decides whether you get one at all, and what you have to commit to in order to be served first.

That is the constraint funnel in its sharpest form. Many modalities, one narrow sterile throat, and the throat is being held by the product class that can fill it most profitably.

What it means, by seat

The shift reads differently depending on where you sit.

For drug developers, the fill-finish slot is now part of the asset, not a logistics line you sort out near approval. A slot in 2028 is a decision you make in 2026. The programmes that miss are rarely the ones with weak data. They are the ones that cleared the clinic and then found there was nowhere to fill the commercial product on time.

For investors, schedulable, secured sterile capacity belongs in diligence. Announced industry capacity overstates what is actually available to a non-GLP-1 sponsor, for the same reason that headline CDMO expansions overstate commercially schedulable supply. A clean asset with no secured downstream slot is carrying a risk that does not show up in the pipeline chart.

For CDMOs, the commercial function has become the most important one in the building. Take-or-pay terms, capacity reservation, and the discipline to hold a slice of the line open for higher-value complex work are now strategic decisions, not contract boilerplate. The business development leader who can secure and defend capacity is worth more than another sales head who can fill a funnel.

The leadership the crowd-out exposes

Two roles get harder, and more valuable, in this environment.

The first is the business development and commercial leader inside a CDMO who can structure capacity as the product. Reservation fees, multi-year commitments, the judgement to turn away easy GLP-1 volume to protect margin on complex modalities. That is a different instinct from chasing the next big single-product contract.

The second is the COO or site head who can run genuinely mixed-modality sterile operations without the high-volume product starving the complex one. Sequencing a lyophilised ADC, a shielded radioligand and a high-volume peptide through shared infrastructure, holding the changeover discipline, and keeping the regulator comfortable across all of it. That is an operating skill, and it is scarce.

What this does not cover, and where it could be wrong

A few honest boundaries.

The GLP-1 capex is also adding net new sterile capacity, and some of it will eventually spill back into the merchant market once the launch peak passes. A meaningful share of the new build is product-specific, pen and device lines that were never going to serve an ADC anyway, so the crowd-out is sharper in some line types than others. And demand forecasts for the obesity class are not guaranteed; a softening would loosen the queue.

None of that changes the near-term reality. For roughly the next 36 months, the complex modalities are price-takers on sterile time. The direction of travel is clear even if the exact timeline moves.

The takeaway

The capacity crisis the sector keeps describing as a science or a capital problem is increasingly a slot problem. In a world where every injectable ends at the same line, the scarce input is not the molecule and it is not the money. It is the fill-finish slot, and a higher-volume product got there first.

The operators who come through this will be the ones who treated downstream capacity as part of the asset, who locked their slot years ahead of the launch, and who built the commercial and operating muscle to defend it. The rest will keep optimising the parts of the process they can see, and discover too late that the binding constraint was the one part they assumed would always be there.

If you are mapping where genuine sterile capacity sits, or building the team that has to secure and run it, that is the work we do. Which networks have schedulable capacity, who has reserved what, and where the real slack is left sits behind our Weaponised Capacity report.

ProGen Search runs retained executive search and market intelligence across CDMO, ADC, radiopharma and advanced modalities. This article is provided for information only and is not investment, legal or regulatory advice. Capacity commitments, ownership and demand forecasts reflect the state of play as of Q2 2026 and may have changed since publication.