Peptide & oligonucleotide manufacturing

Two molecule classes are eating modern pharma's growth — GLP-1 peptides for obesity/diabetes and oligonucleotides (siRNA, antisense) for genetic medicine — and both are made by the same handful of plants. The chokepoint is not the molecule; it is qualified, GMP, large-scale solid-phase synthesis capacity, which takes years and hundreds of millions of dollars per site to build and inspect. Map plants and qualified-supplier positions, not pipeline hype.

Who owns it

The deep capacity sits with a few specialists, most of them not US-domiciled. Bachem is the scale leader, posting CHF 695.1M FY2025 sales at a 30.9% EBITDA margin and guiding to >CHF 1bn sales in 2026; its new Building K in Bubendorf was inspected by Swissmedic at end-2025 and is ramping commercial output of several tons of peptides annually. Lonza is expanding SPPS at its US sites and bought Roche's Vacaville biologics campus for $1.2bn to add domestic capacity. On the oligo side, Agilent is the cleanest US-listed pure capacity play: it is spending $725M to roughly double its Frederick, Colorado oligo lines (Trains C and D), with shipments beginning in 2026, and already makes six commercial API molecules. The privately held names — CordenPharma (Astorg-owned, which bought AmbioPharm in May 2026) and Swiss-listed PolyPeptide — round out the qualified-supplier short list but offer little or no clean US-listed exposure.

What breaks it

Three things constrain supply. First, physics and chemistry: long-chain peptides like semaglutide/tirzepatide and structurally complex oligos are made by SPPS, which is slow, reagent-intensive, and hard to scale without yield loss. Second, qualification: a sponsor cannot simply move a commercial product to a new plant — the site must be validated and FDA-inspected, so incumbents with qualified lines hold pricing power. Third, geography and policy: the BIOSECURE Act was enacted December 18, 2025, and on June 8, 2026 the DoD added WuXi AppTec to its 1260H list, reinforcing a multi-year push to move API supply out of China and into the US and Europe.

What forces the reprice

Demand is structurally above pre-2022 projections. Eli Lilly announced a $27bn US manufacturing build in 2025, doubling its domestic spend since 2020 to over $50bn, and sponsors increasingly pre-pay for dedicated capacity — Viking Therapeutics committed $150M of prepayments to CordenPharma over 2025–2028 for GLP-1 supply. The reprice catalysts are concrete: capacity coming online (Bachem Building K ramp, Agilent Trains C/D 2026 shipments), CDMO earnings showing utilization and pricing, and BIOSECURE implementation milestones that could redirect orders toward Western suppliers.

Be honest about size and earliness

The narrative is bigger than the listed revenue today. The two cleanest public ways to own this chokepoint — Bachem and Lonza — trade in the US only as thinly traded OTC ADRs (Bachem's BCHMY shows daily volume in the low thousands of shares), so US investors face liquidity and FX friction, and the primary listings sit on the SIX Swiss Exchange. Agilent is genuinely US-listed and liquid, but oligo CDMO is a single line inside a ~$1.8–1.9bn-per-quarter diversified instrument company — you are buying a fractional theme, not a pure play. The only US-listed nucleic-acid CDMO that is mostly that business, Maravai, is a cautionary tale: FY2025 revenue fell to $185.7M and the stock dropped ~68% as mRNA/CleanCap orders collapsed, and its strength is mRNA capping, not large-scale therapeutic oligo API. Critically, distinguish announced from operational capacity: groundbreakings and 2026–2030 capex plans (Bachem's $250M Vista build) are not tons shipped, and BIOSECURE's teeth are years out — OMB's company-of-concern list is not due until December 2026, with full FAR effect potentially ~three years away. This is a real, durable chokepoint with a thin set of cleanly investable handles. None of this is investment advice; it is a map of where the capacity sits and what moves it.