When Ozempic and Wegovy waiting lists first made national news in 2023, most investors assumed the problem was simple: Novo Nordisk needed to build more factories. The reality was more interesting. The bottleneck was not final fill-finish; it was upstream, at the peptide CDMO layer, where a small handful of contract manufacturers owned essentially all of the specialized capacity for building long, complex peptides at commercial scale. Two years later the peptide CDMO trade has become one of the most interesting pick-and-shovel plays in healthcare — and also one of the trickiest, because the same capex cycle that lifted Bachem and PolyPeptide could turn against them if the big pharma customers in-source aggressively.
This guide explains why peptide manufacturing is fundamentally harder than small-molecule manufacturing, which CDMOs control the capacity, and how to think about the investment thesis heading into 2026. This article is for educational purposes only and does not constitute investment advice.
Market Overview
The global peptide CDMO market is estimated at roughly $5–7 billion in 2024 and is projected to roughly double by 2030, driven almost entirely by the GLP-1 obesity buildout. That is a small market in absolute terms — smaller than many individual drug franchises — but it is one of the fastest-growing corners of pharma services. Utilization rates at the leading peptide CDMOs have been running at or near capacity since 2022, and backlog books stretch into the back half of the decade.
The supply side is structurally constrained for a reason. Building a commercial-scale solid-phase peptide synthesis (SPPS) suite for a 30+ amino acid drug like semaglutide takes three to five years from breaking ground to GMP-qualified output. The specialized reactors, purification columns, controlled substance handling, and GMP quality systems cannot be bolted on to an existing small-molecule API plant. That long build cycle is why Novo Nordisk's $16+ billion manufacturing capex plan, announced in 2023–2024, is still rolling out in phases through 2028.
Demand, meanwhile, keeps climbing. Semaglutide, tirzepatide, and the next wave (retatrutide, MariTide, VK2735, pemvidutide) all require peptide APIs. If even half the Phase 3 pipeline reaches market, global peptide API demand roughly triples.
Key Companies
The peptide CDMO space is dominated by a handful of European specialists. Most trade on Swiss or German exchanges, which makes access slightly harder for US retail investors but not impossible through IBKR, Schwab international, or ADR routes.
| Company | Ticker | Listing | Peptide Role | Market Cap (approx) |
|---|---|---|---|---|
| Bachem Holding | BACN | SIX (Zurich) | Pure-play peptide CDMO | ~$5B |
| PolyPeptide Group | PPGN | SIX (Zurich) | Peptide CDMO + own reagents | ~$1B |
| Lonza Group | LONN | SIX (Zurich) | Diversified CDMO (peptides, biologics, small mol) | ~$40B |
| Evonik Industries | EVK | Xetra (Frankfurt) | Peptides + lipid excipients + amino acids | ~$12B |
| Siegfried Holding | SFZN | SIX (Zurich) | Diversified CDMO with peptide capacity | ~$5B |
| CordenPharma (Astorg-owned) | Private | — | Peptides, lipids, sterile fill-finish | Private |
| Catalent (Novo Holdings-owned) | Private | — | Fill-finish, injectables, peptide capabilities | Private |
Bachem (BACN.SW): The Pure Play
Bachem is the closest thing to a pure-play peptide CDMO on public markets. The Swiss company was founded in 1971 and has spent five decades refining large-scale SPPS. Its Bubendorf and Sisseln sites in Switzerland and Torrance in the US were the first places most big pharma customers went when GLP-1 demand spiked. Bachem runs multi-hundred-kilogram SPPS campaigns and has disclosed a multi-year capacity expansion program — reportedly CHF 700 million+ — to roughly triple API output by the end of the decade.
Bachem's financials have been volatile. Revenue surged in 2022–2023, dipped in 2024 as customers burned down their inventories, and is re-accelerating into 2026 as new capacity comes online. Gross margins sit in the 30%+ range, with the obvious risk that a customer concentration event (loss of a major GLP-1 contract, aggressive Novo or Lilly in-sourcing) would compress them.
PolyPeptide Group (PPGN.SW): The Turnaround
PolyPeptide Group had a rough 2023–2024. The Swiss company, which operates peptide GMP sites in Europe, the US, and India, went through a delayed capex ramp, accounting surprises, and several profit warnings that cut its share price by more than 70% from IPO highs. Into 2026 the story is a recovery and operational reset — new CEO, tightened capex discipline, and benefits from the broader GLP-1 tailwind. The stock is either a value trap or one of the highest-beta ways to play a peptide CDMO recovery, depending on how execution lands.
Lonza (LONN.SW): The Diversified Giant
Lonza is the biggest CDMO in Europe and one of the biggest in the world, with roughly 15,000 employees and services across small molecules, biologics, cell and gene therapy, and peptides. Peptides are only a slice of the business but a growing one, and Lonza's scale, balance sheet, and existing GMP systems give it an advantage when customers want capacity in a hurry. Lonza is the safer, slower way to play the theme. Its exposure to cell and gene therapy manufacturing also makes it a natural bridge stock for investors with portfolios in both peptides and gene editing.
Evonik, Siegfried, CordenPharma
Evonik Industries (EVK.DE) is a German specialty chemicals group. Its peptide business is part of its Health Care segment and benefits from Evonik's upstream amino acid and lipid excipient production — the same lipids used in mRNA vaccines and GLP-1 formulations.
Siegfried Holding (SFZN.SW) is another Swiss CDMO with growing peptide capacity and a track record of quiet, profitable execution.
CordenPharma is a major peptide CDMO owned by private equity (Astorg). It is the reason some peptide drugs you have never heard of get made at all. Not directly investable, but important for understanding capacity dynamics.
Catalent was the US-listed CDMO giant acquired in 2024 by Novo Holdings (Novo Nordisk's parent foundation) specifically to secure fill-finish and injectable capacity for Wegovy and Ozempic. The deal was partially a statement that Novo no longer trusted the open contract market to scale with it. That transaction is the most important single event in the recent peptide CDMO story.
The Science Under the Hood
Why is peptide manufacturing so much harder than making a small-molecule drug?
Solid-phase peptide synthesis (SPPS). Most therapeutic peptides are built using SPPS, a technique that earned Bruce Merrifield the 1984 Nobel Prize in Chemistry. The growing peptide chain is anchored to a polystyrene resin bead, and each amino acid is added one at a time, with protection and deprotection steps between every coupling. A 30-amino-acid peptide like semaglutide takes 30+ cycles. Each cycle must hit >99% yield just for the final product to come out at an acceptable purity. At scale, a single batch can take weeks and produce only tens of kilograms of crude peptide that then has to be purified.
Purification bottleneck. Crude SPPS output is a mixture of the target peptide and dozens of close cousins — sequences that are one amino acid short, deleted, or modified. Separating them requires preparative high-performance liquid chromatography (prep-HPLC) at scale, which is slow, uses large volumes of solvent, and depends on a small number of reversed-phase column suppliers.
Cost per gram. A small-molecule API can be made for a few dollars a kilogram. A complex therapeutic peptide can cost thousands of dollars a gram to produce. That cost profile is fine when the drug sells for $1,000+ a month, but it is why peptide drugs rarely look like generic small molecules even after patent expiry.
Alternatives. Recombinant expression in engineered microbes or yeast can sidestep SPPS for some peptides, and chemoenzymatic and hybrid approaches are advancing. Insulin has been made recombinantly since the 1980s. GLP-1 analogs are increasingly moving toward recombinant or hybrid production, and that shift is one of the biggest swing factors for the CDMO industry over the next decade.
For a deeper dive into peptides as molecules, start with natural peptides in the human body.
Investment Thesis & Risks
Bull Case
- Pick-and-shovel positioning. Every successful GLP-1, from Novo's semaglutide to Viking's VK2735, needs peptide API. CDMOs earn regardless of which molecule wins.
- Structural capacity shortage. New capacity takes 3–5 years to build; demand has been outrunning it since 2022.
- Pricing power. Customers in capacity-constrained markets pay premium take-or-pay contracts.
- Consolidation optionality. Private equity and strategic acquirers continue to take out independents (CordenPharma, Catalent). Remaining public pure plays (Bachem, PolyPeptide) are natural targets.
- Adjacency to gene therapy. Lonza and Evonik also serve the gene editing and mRNA value chain, providing a hedge.
Bear Case
- In-sourcing. Novo's Catalent acquisition signaled that big pharma would rather own critical capacity than rent it. Eli Lilly's Research Triangle expansion and Concord, NC and Lebanon, IN facilities are the US version of the same trend. Every ton of in-house capacity is a ton that the CDMO layer does not sell.
- Overbuild risk. If every CDMO and every pharma client expands at once, 2028–2030 could see the industry move from shortage to glut, compressing pricing at exactly the wrong moment for levered balance sheets.
- Patent cliffs. Semaglutide loses composition patent protection in China in 2026 and in the US between 2031 and 2033 depending on formulation. Generic peptides will still need CDMO capacity, but prices (and margins) drop.
- Recombinant substitution. If more GLP-1 analogs shift from SPPS to recombinant expression, incumbent SPPS capacity becomes a stranded asset.
- Customer concentration. Any one CDMO losing a single top-three customer is a material event. Bachem and PolyPeptide have both disclosed high customer concentration historically.
- FX and Swiss franc exposure. Most listings are CHF-denominated; currency moves add another layer of volatility for non-Swiss investors.
What To Watch in 2026
- Novo Nordisk internal capacity milestones (Kalundborg expansion, Catalent site integration progress). The faster Novo self-supplies, the less work for open CDMOs.
- Eli Lilly's Lebanon and Concord site ramps. Similar in-sourcing signal.
- Bachem's capex guidance updates and utilization disclosures.
- PolyPeptide Group quarterly turnaround execution.
- Any new CDMO M&A — especially private-to-public transitions or large pharma buying a public CDMO outright.
- Shift of any major GLP-1 candidate from SPPS to recombinant expression — a credible disruption signal.
- Generic semaglutide launches in China and the resulting scramble for lower-cost peptide API.
Connection to Gene Editing
There is an underappreciated link between the peptide CDMO story and the gene editing story. The same contract manufacturers scaling peptide capacity — Lonza, Evonik, and the rest — also operate some of the most advanced cell and gene therapy suites in the world. Lonza manufactures lentiviral and AAV vectors; Evonik supplies the ionizable lipids used in lipid nanoparticle delivery of mRNA and CRISPR payloads; CordenPharma's sterile fill-finish business serves both modalities. The peptide build-out is, quietly, also a gene-therapy build-out.
The long-term strategic question is which modality wins for chronic metabolic disease. If weekly peptide injections remain the standard of care for obesity into the 2030s, CDMO demand stays strong. If in-vivo base editing or CRISPR-based one-shot approaches ever deliver durable weight control — a major "if," but not a zero — the same CDMOs that built peptide capacity will want to pivot hard into gene therapy manufacturing. The companies that have kept optionality across both modalities (Lonza most obviously) are the safest way to own the theme.
Frequently Asked Questions
Q: Why can't Novo Nordisk and Eli Lilly just make all their own peptides? A: They are trying. Both companies have announced multibillion-dollar manufacturing expansions. But new GMP peptide capacity takes years to build and qualify, and in the meantime they need CDMOs to bridge the gap. Over time, in-sourcing will grow — which is the main bear case for pure-play CDMOs.
Q: Is Bachem a better investment than PolyPeptide? A: Bachem is the higher-quality operator with a cleaner track record and tighter balance sheet. PolyPeptide is the higher-beta turnaround story. Both are exposed to the same end market. Your choice depends on risk tolerance.
Q: How can a US investor buy Swiss-listed peptide CDMOs? A: Most major US brokers (Interactive Brokers, Schwab International, Fidelity) offer direct access to SIX Swiss Exchange listings. ADR versions exist for some names but are illiquid.
Q: What is the difference between SPPS and recombinant peptide production? A: SPPS builds the peptide chemically, one amino acid at a time on a resin bead. Recombinant production engineers microbes or yeast to secrete the peptide from a DNA template. SPPS is more flexible but expensive per gram. Recombinant is cheaper at very large scale but harder for long, heavily-modified peptides.
Q: What happens to peptide CDMOs if GLP-1 demand peaks? A: Demand growth slowing does not mean demand shrinking. Even a flat GLP-1 market would still require enormous ongoing API supply. The bigger risk is overbuild — too much capacity meeting a suddenly flat market at the same moment incumbents in-source.
