Jiuzhou Pharma Q3 2026: Revenue Returns to Growth at RMB 1.36B, but RMB 5B in Peptide Backlog Is Still Bottlenecked by Capacity

9M26 revenue reached RMB 3.94B (-5.38%) and net profit RMB 617M (-17.50%), yet Q3 alone grew 5.58% to RMB 1.36B with net profit up 3.41% to RMB 229M. New small-molecule CDMO projects rose over 25% YoY and TIDES added 20+ projects with a backlog above RMB 5B, while group-wide capacity utilisation sits at just ~70%.

On 22 October, Jiuzhou Pharmaceutical (603456.SH) reported third-quarter 2026 results. For the first nine months revenue was RMB 3.937B, down 5.38% year on year; net profit attributable to shareholders was RMB 617M, down 17.50%; and net profit excluding non-recurring items was RMB 615M, down 17.54%. In Q3 alone revenue reached RMB 1.361B (+5.58%), net profit RMB 229M (+3.41%) and recurring net profit RMB 229M (+4.02%) — after sequential improvement in Q2, the company is officially back to year-on-year growth. The nine-month recurring net margin was 15.63%.

How to read the numbers: the cumulative figure still falls, the inflection is quarterly

Split across three quarters, the picture is clearer. The cumulative decline stems mainly from pricing and product-mix pressure in generic APIs plus the delivery timing of certain legacy CDMO projects, while single-quarter growth with profit expanding in line with revenue suggests the drag from legacy project adjustments is fading and new projects are starting to contribute. Management attributed the slight decline in Q3 gross margin to customer delivery mix, calling it normal quarterly volatility with the full-year level expected to hold — a claim that needs to be tested against the Q4 delivery mix before it can be read as a trend.

Capacity utilisation at 70%, but the sites tell two different stories

In an investor call with 52 institutions after the results, the most valuable disclosure was on capacity: group CDMO utilisation is around 70%, but the sites are sharply divergent. The small-molecule workshops at Ruibo Suzhou are fully booked and running above 80%, serving as the core plant for late-phase and commercial orders from multinationals; Zhejiang Ruibo is seeing utilisation climb steadily as Phase III projects enter process performance qualification (PPQ).

That ten-point gap between 70% group-wide and 80%-plus at the flagship plant is exactly where second-tier CDMOs stand in 2026: not a shortage of orders, but orders concentrated in a few sites and a few stages. It also explains why the same report can show new small-molecule CDMO projects up more than 25% while revenue still contracts — 18 to 36 months separate a signed project from PPQ and from commercial batch revenue.

Peptides: RMB 5B backlog, and the constraint is capacity, not demand

TIDES is the fastest-growing block. It added more than 20 projects in the first nine months, spanning IND through ND stages, concentrated in GLP-1, Parkinson's disease and autoimmune indications. Solid-phase peptide capacity was expanded 300% year on year in H1, yet management is explicit that growth remains constrained by capacity scale; the company is accelerating construction of commercial peptide lines, with backlog value above RMB 5B.

Capex tells the same story: the 2026 fixed-asset investment plan is capped at RMB 1B, with 2027 expected at a similar level, directed at the Ruibo Suzhou small-molecule and peptide commercial lines, oligonucleotide capabilities, a new conjugation (DC) line at the Taizhou site and renovation of legacy workshops at Zhejiang Ruibo. The Ruibo Suzhou commercial small-molecule line is scheduled to start production in H2 2027, with the Taizhou DC line planned for 2027.

Put RMB 1B a year in context: WuXi AppTec guides 2026 capex at RMB 7.5-8.5B, with RMB 10.1B committed to its Changzhou site alone, while Asymchem and Porton operate in the multi-billion range. Jiuzhou's intensity means it cannot build everything — doing two of peptides, conjugation and oligonucleotides really well is already a stretch. That is a structural constraint on second-tier CDMOs in 2026, not a question of execution.

Customer mix: Novartis remains the anchor, overseas front-ends fill in

Novartis is still the core strategic customer. The Novartis product cycle maps directly onto Jiuzhou's order book: Entresto still contributed USD 2.486B in sales as a cash-flow anchor despite generic competition, Kisqali took over as the new core at USD 3.211B (+51% YoY), and Fabhalta grew 96%. On the funnel side, by the end of H1 the company had 1,224 Phase I/II projects, 99 Phase III and 44 commercial, up 66, 5 and 5 respectively from end-2025; the key-account team won 11 new projects in H1.

Overseas localisation is starting to pay. In the first nine months Jiuzhou Europe delivered 8 projects; Ruibo US booked 42 orders and added 17 new customers; Japan and Korea contributed 28 new orders. The Suzhou pilot plant hosts more than 30 active programmes, at least five in late-stage clinical development across oncology and autoimmune disease. The model is a technical BD approach — overseas front-ends in the US, Kamakura and Konstanz take on development services, with Chinese sites scaling and delivering from kilogram quantities upward — competing on process understanding rather than price.

External variable: how brokers read Announcement No. 48

On 16 October, MOFCOM and MOST opened consultation on a revised catalogue of technologies prohibited or restricted from export (Announcement No. 48 of 2026). Five biopharma technologies are involved, three of which — proprietary cell and gene therapy process platforms, siRNA drug process platforms and engineered pathogen gain-of-function techniques — are proposed for the prohibited list. Sinolink Securities judges the impact limited: mainstream licence-out deals do not transfer technology ownership, and CDMO value lies in outsourced services with customer IP typically held offshore.

For a structure like Jiuzhou's — Chinese capacity serving overseas customers — the near-term effect is sentiment. The medium-term question is whether its oligonucleotide build-out lands inside the 'services' framework or brushes against 'technology transfer'. That is the question worth asking in the next few investor calls.

Editorial: three paths to margin repair, and Jiuzhou picked the hardest

Line up three peers and Q3 2026 shows three entirely different margin curves. Asymchem grew Q3 revenue 24.04% and net profit 81.49%, but emerging businesses ran at a 30.95% gross margin against 47.97% for small molecules — trading margin for scale. Samsung Biologics saw its operating margin fall from 50.4% to 42.9% as Plant 5 PPQ costs and the first Rockville consolidation landed — a classic ramp-up tax, trading near-term margin for capacity position through 2032. Lonza anchored mid-term CORE EBITDA margin in the high thirties at its Capital Markets Day, the structural margin of a pure-play CDMO.

Jiuzhou is a fourth path, and the hardest: protect profitability in legacy small molecules (management says profit, not revenue, is the priority) while betting on peptides, conjugation and oligonucleotides converting capacity into revenue in 2027-2028. That requires two things to hold simultaneously — no cash-flow stall in the base business, and no delay in validating and delivering new capacity.

The valuation anchor for this kind of company therefore should not be backlog value but the conversion rate from orders to revenue. Two verifiable inputs: the PPQ schedule of new projects, and utilisation at the flagship sites. How much of that RMB 5B peptide backlog converts, and how fast, matters far more than the headline number — especially with group utilisation at 70% and dedicated peptide lines not yet built.

The risk is equally clear: sector-wide capacity expansion means a healthy order book is not the same as durable undersupply. If demand growth lags capacity release, falling utilisation and new depreciation will compress margins together — a particularly dangerous combination for a second-tier CDMO spending a tenth of the leader's capex while betting on three new modalities at once.