Why Pharmaceutical Companies Are Spending Billions on Peptide Manufacturing

On 20 July 2026, Samsung Biologics announced an all-cash public tender offer for Switzerland's PolyPeptide Group AG at CHF 44.31 per share — roughly CHF 1.46 billion in equity value, about $1.8 billion, and a 40% premium to PolyPeptide's undisturbed price of CHF 31.65 on 10 April. It is the largest acquisition ever made by a South Korean biopharmaceutical company. PolyPeptide's largest shareholder, holding about 55.65% of outstanding shares, has irrevocably undertaken to tender; the offer carries a two-thirds minimum acceptance threshold and is expected to close by the end of 2026.

Samsung Biologics is a mammalian-cell antibody manufacturer. It is paying a substantial premium for six cGMP sites in Belgium, France, Sweden, California and India, an innovation centre in Strasbourg, and seventy years of institutional knowledge covering more than a thousand therapeutic peptides. CEO John Rim framed it plainly: modality expansion into peptides, "including GLP-1."

The deal is not an outlier. It is the most visible datapoint in a capital cycle that has already committed well over $30 billion to peptide chemistry, and it tells you where the industry believes the constraint sits.

Why Demand for Peptide APIs Has Increased

Peptides occupy a useful middle ground. They bind targets with the selectivity of a biologic while remaining chemically synthesised, characterisable by conventional analytics, and manufacturable without a bioreactor suite. Two decades of medicinal chemistry solved their historic weakness — minutes-long plasma half-lives — through fatty-acid acylation, non-natural amino acid substitution, cyclisation and PEG-like spacers. A weekly injection became possible; then a daily oral tablet.

That unlocked the pipeline. Peptides now span metabolic disease, oncology (radioligand-targeting vectors such as those built on somatostatin analogues), rare disease and cardiology. But the demand shock is overwhelmingly metabolic.

The GLP-1 Effect

The global GLP-1 receptor agonist market is valued near $52.3 billion in 2026 and is projected to approach $97.5 billion by 2031. No prior peptide class has required this tonnage. Historic peptide drugs were dosed in micrograms to low milligrams for patient populations in the tens of thousands. GLP-1 medicines are dosed in milligrams weekly, chronically, for populations counted in tens of millions.

Two structural factors amplify this further.

• •Oral formulations. Oral semaglutide has bioavailability on the order of ~1%, meaning an oral patient-year consumes an order of magnitude more API than an injectable patient-year. Every oral GLP-1 approval is, in manufacturing terms, a multiplier on API demand rather than a substitution.
• •Loss of exclusivity. Patent expiry across Canada, China, India and Brazil is mobilising a generic supply base — Indian API houses in particular — that must build its own capacity from scratch.

How Solid-Phase Peptide Synthesis Works

Solid-phase peptide synthesis (SPPS), Bruce Merrifield's Nobel-winning method, anchors the first amino acid to an insoluble polymer resin. The chain is then extended one residue at a time through a repeating cycle: deprotect the terminal amine, couple the next protected amino acid, wash away excess reagent. Because the growing chain stays bound to the resin, clean-up between steps is a filtration rather than a separation. At the end, the peptide is cleaved from the resin and its side-chain protecting groups removed.

The elegance is also the arithmetic problem. Yield compounds multiplicatively. At 99% efficiency per cycle, a 30-residue peptide retains roughly three-quarters of theoretical yield; at 98%, roughly half. Real processes for commercial peptides frequently run hybrid routes — solid-phase assembly of fragments, then solution-phase ligation — precisely to escape that compounding.

Route choice varies even within a single drug class. Novo Nordisk's semaglutide process expresses the peptide backbone recombinantly in yeast and then chemically acylates it; Lilly's tirzepatide is a fully synthetic 39-residue peptide. Same therapeutic class, materially different factories.

Why Long and Modified Peptides Are Hard at Scale

Every failed coupling produces a deletion sequence — a molecule missing one residue, otherwise identical to the product. Every incomplete deprotection produces a truncation. Racemisation produces epimers that differ by a single stereocentre and share an exact mass. A 39-residue peptide accumulates dozens of such species, and their concentration rises with chain length.

Chemical modification compounds the problem. The lipid side chains that give GLP-1 medicines their weekly dosing make the molecules sparingly soluble and strongly surface-active. Aggregation on-resin slows coupling kinetics. Solubility limits force dilute processing, which means larger vessels for the same output.

The consequences are physical and economic. SPPS is famously solvent-intensive; commonly cited estimates place waste generation in the thousands of kilograms per kilogram of peptide, with solvent recovery now a first-order design constraint rather than a sustainability footnote. Raw materials — protected amino acids, coupling reagents, resin — typically represent 60-70% of cost of goods, and protected amino acid supply is itself concentrated. A single large-scale peptide facility can exceed $500 million in capital cost and take five to seven years from decision to qualified commercial output.

Purification Is the Real Bottleneck

Ask any peptide CDMO where the constraint binds and the answer is chromatography. Crude SPPS material must be resolved to typically >98% purity by preparative reversed-phase HPLC, against impurities engineered by the process itself to resemble the product. Deletion sequences and epimers co-elute. Purification can triple total production time and is where yield is genuinely lost.

Scaling chromatography is not like scaling a reactor. Throughput is governed by column loading, cycle time, resin lifetime and solvent volume, all of which scale unfavourably. This is why Axplora committed €50 million at Mourenx, France specifically for large-scale HPLC and continuous chromatography, and why continuous and simulated-moving-bed approaches have moved from conference posters to capital plans.

API Versus Sterile Fill-Finish

These are two different industries sharing a supply chain.

A company can be long API and short fill-finish, or the reverse — and the binding constraint has shifted between them repeatedly since 2022. Novo Holdings' $11 billion purchase of three Catalent sites was a fill-finish transaction. Novo Nordisk's $4.1 billion second Clayton, North Carolina facility, completing between 2027 and 2029, is fill-finish. Lonza's ~CHF 500 million Stein plant, operational in H2 2027, is fill-finish. Lilly's $6 billion Huntsville, Alabama plant — breaking ground in 2026 for completion in 2032 — is API. Samsung Biologics, which already owns enormous sterile capacity, bought the chemistry it lacked.

Why CDMO Capacity Became Strategically Valuable

Capacity is now a competitive moat with a five-to-seven-year replacement time. You cannot buy your way out of a shortage inside a product cycle, and the 2022-2023 GLP-1 shortages demonstrated what unmet demand costs in market share, compounding-pharmacy exposure and political attention.

Hence the acquisitions and the concrete.

North American peptide capex hit a five-year high in 2026, driven as much by supply-chain geography as by volume: concentration of peptide API in Asia-Pacific is now treated as a strategic risk by both sponsors and governments.

The originators have moved in parallel rather than relying solely on contract supply. Lilly's Alabama plant sits inside more than $50 billion of US capital commitments announced since 2020, alongside roughly $9 billion at Lebanon, Indiana and a doubled $2 billion at Concord, North Carolina. Novo Nordisk took the acquisition route with Catalent. India's generic manufacturers are building a third pole of supply aimed at post-exclusivity markets. Three different strategies, one conclusion: chemistry capacity is the asset.

Buying an operating CDMO converts a seven-year build into a closing date. That is what Samsung Biologics paid a 40% premium for.

Will Manufacturing Capacity Become a Bottleneck?

Partly — and selectively. For the current injectable GLP-1 franchises, the acute shortage phase has passed; announced capacity broadly tracks announced demand. The risks sit elsewhere.

• •Oral formulations are the largest single swing factor. Broad oral GLP-1 adoption could absorb capacity faster than it is being built.
• •Generic entry across the 2026-2031 exclusivity cliff will add demand from manufacturers with no existing peptide footprint.
• •Purification and specialty inputs — protected amino acids, chromatography resin, solvent recovery — could bind before reactor volume does.
• •Smaller developers face crowding out. Capacity contracted years ahead by metabolic-disease sponsors is capacity a Phase II oncology peptide cannot access at any reasonable price.

The more likely outcome is not a universal shortage but a two-tier market: abundant, competitively priced capacity for high-volume commercial peptides, and scarce, expensive, long-lead capacity for everything else.

Market Implications

• •CDMO valuations have repriced on scarcity, not growth. A 40% premium for a capacity-constrained asset implies buyers are paying for time-to-market, not near-term EBITDA. Expect remaining independent peptide CDMOs to trade at strategic premiums.
• •Consolidation should continue. Samsung Biologics is now a multi-modality platform; Lonza, WuXi, Thermo Fisher and Catalent's successors face the same modality-gap logic. Independent mid-cap peptide specialists are a shrinking set.
• •The picks-and-shovels layer is under-owned. Protected amino acids, chromatography resin, large-format columns, solvent recovery systems and isolator-based aseptic equipment all sit upstream of the announced capex and may be tighter than the headline facilities.
• •Manufacturing has become a disclosed competitive variable. Capex commitments, site qualification timelines and fill-finish redundancy now belong in the same analytical frame as trial readouts.
• •Geographic diversification carries a cost. Onshoring to the US and Europe raises unit costs relative to Asian supply; whether payers absorb that or margins do is an open question for the generic wave.

Key Takeaways

• •The deal. Samsung Biologics' ~$1.8 billion tender offer for PolyPeptide, announced 20 July 2026 and expected to close by year-end, buys peptide chemistry it could not build fast enough — six cGMP sites and seventy years of process knowledge.
• •The demand shock. GLP-1 medicines turned peptide manufacturing from a milligram business into a metric-tonne business, with the class projected to grow from ~$52 billion in 2026 to ~$97 billion by 2031.
• •The chemistry. SPPS builds peptides one residue at a time on resin. Yield compounds multiplicatively, so long, lipidated sequences are structurally difficult and waste-intensive to make.
• •The bottleneck. Purification, not synthesis, is the binding constraint, because process-generated impurities closely resemble the product.
• •Two capacity pools. API and sterile fill-finish are distinct, and recent multi-billion-dollar deals have targeted both as the constraint moved between them.
• •Time is the moat. Capacity takes five to seven years to build, which is why acquisition has become the preferred route.
• •The forward risk. A two-tier market — adequate capacity for blockbuster peptides, scarcity for smaller programmes.

Frequently Asked Questions

Why is Samsung Biologics buying a peptide company?

Samsung Biologics' expertise is mammalian-cell biologics and sterile fill-finish. Peptides are chemically synthesised, requiring different plant, chemistry and regulatory know-how. Acquiring PolyPeptide adds that modality immediately rather than over a five-to-seven-year build, and brings manufacturing sites across the US, Europe and India.

Are peptides regulated as biologics or as small molecules?

Classification varies by jurisdiction and molecule. In the US, most synthetic peptides of 40 amino acids or fewer are regulated as drugs rather than biologics, which is why abbreviated generic pathways exist for products such as liraglutide. Recombinantly produced peptides may be treated differently. The distinction materially affects the route to market for generic entrants.

Why is purification harder than synthesis?

SPPS generates impurities that are chemically near-identical to the product — sequences missing one amino acid, or differing by a single stereocentre with the same molecular mass. Separating them requires preparative HPLC operating close to its resolution limit, and the process loses real yield doing so.

Does the same factory make the API and the finished injection?

Usually not. API manufacture is a chemical operation; filling sterile vials, cartridges and autoinjector pens is an aseptic operation under different GMP requirements. Most products cross at least one company boundary between the two, which is why both types of capacity have been acquired aggressively.

Will there be another GLP-1 shortage?

The acute 2022-2023 injectable shortage has largely resolved. The credible future pressure points are oral formulations, which consume far more API per patient, and generic entry adding manufacturers without existing capacity. Watch purification throughput and protected amino acid supply rather than headline reactor volume.

What does this mean for non-metabolic peptide drugs?

It is a headwind. Capacity contracted years in advance by large metabolic sponsors is unavailable to smaller programmes, and pricing for what remains reflects scarcity. Securing manufacturing early has become a genuine strategic risk item for peptide biotechs.

Sources and Further Reading

• •Samsung Biologics — offer to acquire PolyPeptide
• •PolyPeptide Group — tender offer announcement
• •DCAT Value Chain Insights — CDMO/CMO key moves in 2026
• •CordenPharma — peptide platform investment
• •Pharma Manufacturing — Bachem capacity expansion
• •FiercePharma — Lilly's $6bn Alabama API facility
• •C&EN — Making Peptides At Large Scale

Disclaimer: This article is an industry and market analysis provided for educational and informational purposes only. It is not investment advice, financial advice, or medical advice, and it contains no laboratory or synthesis protocols. Figures reflect company announcements and public reporting as of August 2026 and may change. All products supplied are strictly for in-vitro laboratory and research use, not for human consumption.