NextFin News - Roche's third-quarter sales rose 7% at constant exchange rates, beating analysts' estimates, on the back of an unlikely growth engine: Xolair, a drug first approved more than two decades ago that is now growing 20% a year as food-allergy demand turns an aging blockbuster into the Swiss group's fastest-rising product.
The Basel-based company said Wednesday that Group sales grew 7% at constant exchange rates (CER) and 8% in Swiss francs in the third quarter, as the currency headwinds that dragged on the first half reversed. Roche reaffirmed its full-year outlook, expecting sales growth in the mid-single-digit range at constant exchange rates and core earnings per share in the high single-digit range, plus a further increase to its dividend in Swiss francs.
The Quarter in Numbers: Growth With a Familiar Face
Third-quarter sales came to roughly CHF 16.1 billion, up from about CHF 14.9 billion a year earlier. For the first nine months, Group sales rose 6% at constant exchange rates, or 1% in Swiss francs, to CHF 46.49 billion, as the strong franc shaved about five percentage points off the reported figure.
The Pharmaceuticals Division carried the quarter, with sales up 7% at constant exchange rates to CHF 36.21 billion for the nine-month period. Diagnostics sales grew 4% at constant exchange rates to CHF 10.28 billion, as demand for core-laboratory and pathology solutions more than offset the healthcare pricing reforms in China that have pressured the division since the start of the year. The China headwind is a multi-year story, not a single-quarter event: Chinese procurement reforms have pushed prices down across medical devices, and the division posted flat sales in the first quarter of 2025 before management set an ambition of mid-single-digit growth for 2026. The third-quarter result - mid-single-digit growth at constant exchange rates for the division - suggests the offset is working, but China remains the swing factor for diagnostics into 2027.
The five growth drivers that management has pointed to all year - Xolair, Hemlibra, Ocrevus, Phesgo and Vabysmo - generated combined sales of CHF 16.8 billion in the first nine months, up 12% at constant exchange rates. Against that, sales of products whose patents have expired, including Avastin, Herceptin, MabThera/Rituxan, Lucentis and Actemra, fell by a combined CHF 0.4 billion at constant exchange rates.
"In the third quarter, we delivered strong sales growth of +7% at constant exchange rates and +8% in Swiss francs. Most importantly, we continued to make meaningful progress for patients. We advanced another five potential new medicines into the final stage of development this year and now aim to bring up to 20 new medicines to patients by 2030." - Thomas Schinecker, Chief Executive Officer of Roche
Why an Old Drug Is Outrunning the New Ones
Xolair is the story inside the story. The medicine, an IgE-blocking antibody first approved in the early 2000s for allergic asthma, has found a second life through label expansions - first for chronic hives, then for IgE-mediated food allergies. In the first nine months of 2026, Xolair sales rose 20% to CHF 2.52 billion, all of it in the United States, making it the fastest-growing product in Roche's pharmaceutical portfolio.
The turnaround is stark. In the first nine months of 2025, Xolair sales were CHF 2.2 billion and falling 4% in Swiss francs, as the drug's original asthma franchise matured. A year later, the same molecule is growing 20%. The difference is not a new molecule - it is a new patient population. Food-allergy labeling turned Xolair from a respiratory drug with seasonal peaks into a year-round therapy for IgE-mediated food allergies, a market where avoidance is impossible and treatment is the only alternative to carrying emergency medication.
The mechanism behind the revival is straightforward and durable. Food-allergy treatment is preventive and continuous rather than episodic: patients stay on regular maintenance dosing to manage exposure risk, which converts a seasonal allergy drug into recurring revenue. That is a different demand profile from the acute-care model Xolair was built on.
The more important detail is what management said about the competition waiting at the door. On the first-quarter call, Roche executives guided to roughly 20% growth for Xolair across the full year, explicitly including the expected impact of the first biosimilar launch at the end of 2026. In other words, the 20% run-rate is not a pre-competition pop; it is the number management expects to hold even as biosimilar entry begins. The real test shifts to 2027: whether food-allergy uptake expands fast enough to absorb the price erosion that follows.
That dynamic - an older molecule extending its commercial life through new indications while a biosimilar clock ticks - is the template for how big pharma survives the patent cliff. It is also why Xolair matters more than its CHF 2.5 billion in sales suggests: it is proof that Roche can extract growth from the portfolio it already owns, not just from the pipeline it hopes to own.
The Replacement Math: Can Five Drugs Outrun Patent Loss?
The structural problem Roche is fighting is visible in the numbers. The expired-patent group lost CHF 0.4 billion at constant exchange rates in nine months. The replacement engine - the five growth drivers - added roughly CHF 1.8 billion of growth on a CHF 16.8 billion base. The spread is comfortably positive, but the base keeps shifting underneath it.
Hemlibra, the once-monthly hemophilia A injection, rose 12% to CHF 3.75 billion. Ocrevus, the multiple-sclerosis therapy, grew 7% to CHF 5.32 billion. Vabysmo, the recently launched eye drug, jumped 10% to CHF 3.20 billion. Phesgo, the breast-cancer combination, surged 18% to CHF 2.06 billion. Tecentriq, the lung-cancer immunotherapy, added a more modest 5% to CHF 2.62 billion, while the older breast-cancer franchise continued to erode: Perjeta fell 8% to CHF 2.04 billion.
The quarter also puts Roche's execution in relief against its Swiss peer. Novartis reported third-quarter results in the same window and, like Roche, leaned on a concentrated set of growth drivers - Kisqali, Kesimpta, Scemblix and Pluvicto - to offset generic competition hitting older cardiovascular and oncology products. Both companies are running the same race: grow the new portfolio fast enough that patent losses become background noise. Roche's nine-month spread - roughly CHF 1.8 billion of growth-driver gains against CHF 0.4 billion of expired-patent losses - is the cleaner version of that trade.
Regionally, the United States - still Roche's largest market at 40.2% of Group sales - grew 5% at constant exchange rates, with Xolair, Hemlibra, Ocrevus, Polivy and Tecentriq more than compensating for declines in Kadcyla and Perjeta. Europe was essentially flat, Japan surged 25% at constant exchange rates, and the International group grew 10%.
The cyclical-versus-structural call matters here. The currency swing is cyclical: in the first half, the strong franc cut reported growth to -2% in Swiss francs despite 6% growth at constant exchange rates, and the currency impact reversed in the third quarter as exchange rates stabilized. The patent-cliff pressure is structural - Avastin, Herceptin and the rest will not regain exclusivity. The Xolair-Hemlibra-Ocrevus growth is semi-structural: it rests on label expansions and new indications that extend commercial life, but it is itself exposed to biosimilar entry at the end of 2026.
The Second-Order Question: What Happens When Xolair Meets Its Biosimilar?
The first-order read of this quarter is simple: Xolair growth lifted sales above estimates. The second-order question is what the market has not fully priced. Roche's shares have traded at a patent-cliff discount for years - investors have treated the company as a slow-growth franchise whose best products are behind it. If Xolair holds its growth into 2027 despite biosimilar entry, that discount compresses, and the re-rating extends beyond Xolair to the entire pipeline multiple.
The pipeline is the hedge against the scenario where Xolair normalizes - and management is paying to make sure it is thick enough. Roche has spent heavily to bulk up the internal portfolio against the SFr5.8 billion in revenue forecast to be lost to generic and biosimilar competition by 2029. Deals include the $2.4 billion acquisition of 89bio, a move into the metabolic dysfunction-associated steatohepatitis market, and a $5.3 billion partnership with Zealand Pharma to advance the obesity candidate petrelintide. Those transactions answer the same question the third-quarter numbers raise: can acquisitions and label expansions fill the hole left by Avastin and Herceptin before the hole widens?
The pipeline is the hedge against the scenario where Xolair normalizes. In the third quarter alone, Roche won US approval for Tecentriq in early-stage colon cancer, Gazyva for a childhood kidney condition and Enspryng in thyroid eye disease. The FDA granted Priority Review to fenebrutinib for both relapsing and primary progressive multiple sclerosis and to giredestrant in early-stage and advanced breast cancer, with management saying it is "fully prepared to launch later this year, pending approval." In diagnostics, the Elecsys pTau217 blood test became the first FDA-cleared single-biomarker test to support both rule-in and rule-out assessment of Alzheimer's amyloid pathology.
There is a valuation layer to this that the headline sales beat does not capture. A company trading on a patent-cliff narrative is assigned a lower multiple not because its current earnings are weak - Roche's core earnings guidance remains in the high single digits - but because investors discount the durability of those earnings. Every quarter in which the growth-driver basket expands at double-digit rates while expired products erode at single-digit rates is evidence that the durability assumption is wrong. That is why the second-order effect of an Xolair beat is not the CHF 2.5 billion in sales; it is the possibility that the market stops pricing Roche as if its best years are behind it.
The expectation gap is the crux. Consensus has priced Roche as a patent-cliff story with mid-single-digit growth. The third-quarter print says the replacement portfolio is already firing at double-digit rates on the growth-driver basket. The question for 2027 is whether the pipeline approvals land on schedule and whether Xolair's food-allergy base keeps expanding after biosimilars arrive.
The Bear Case, and What Would Break It
The strongest argument against the optimistic read is that Xolair's 20% is a one-time label-expansion pop rather than a durable run-rate. Once food-allergy penetration saturates, the argument goes, growth normalizes to mid-single digits - precisely as biosimilars arrive. At the same time, Tecentriq's 5% growth trails the portfolio's fastest growers, the breast-cancer franchise keeps eroding, and China's diagnostics pricing reforms have not finished working through the system.
The answer is diversification. Even if Xolair normalizes, the growth engine spans five molecules across five different disease areas - hemophilia, multiple sclerosis, ophthalmology, breast cancer and allergy. No single product's slowdown collapses the bridge, and the 2026 guidance already internalizes biosimilar entry.
The falsifying signal is specific: if Xolair growth in 2027 falls below roughly 10% and the top-five growth-driver basket decelerates below about 8% at constant exchange rates, the replacement thesis is broken and the patent-cliff discount deserves to widen again.
What Comes Next
Short term, the market reacted positively: Roche's Swiss-listed shares rose 2.58% to 389.90 francs on Wednesday, closing near the 52-week high of 391.80 francs, while the German-listed depositary receipts gained 2.16%.
Medium term, the confirmed guidance - mid-single-digit sales growth and high-single-digit core earnings per share growth at constant exchange rates - signals that management sees the replacement engine on track through year-end. Investors will watch the fourth-quarter Xolair trajectory for the first read on biosimilar impact.
Long term, the structural question is whether the 2030 pipeline target of up to 20 new medicines can keep the growth engine running after the current drivers mature. Three scenarios frame the path: a base case of Xolair near 20% in 2026 with gentle deceleration in 2027 and on-schedule approvals; an upside case where fenebrutinib and giredestrant approvals exceed expectations and food-allergy uptake stays above 20%, triggering a multiple re-rating; and a downside case where biosimilar erosion accelerates, China diagnostics worsen, or pipeline decisions slip, forcing a guidance cut.
The signals to watch are the fourth-quarter Xolair print against 2027 guidance, the FDA decision dates for fenebrutinib and giredestrant, and the growth rate in China diagnostics. If Xolair holds above 10% growth into 2027 while the top-five basket stays above 8% at constant exchange rates, the patent-cliff narrative that has capped Roche's valuation for years will have lost its foundation.
Roche's growth story has quietly shifted from "what comes after the patents expire" to "the old blockbuster is outgrowing the new ones" - and the next two quarters will show whether that reversal is a bridge to the future or a final extension of the past.
The data-cutoff for this report is the Roche third-quarter 2026 sales release published in Basel on 22 October 2026, with market levels as of that day's close.
Explore more exclusive insights at nextfin.ai.
