Good morning and good afternoon. I am excited to share our Q3 2026 results with you today. We had a strong performance yet again, and we made significant portfolio progress. We had, as mentioned, a strong performance year to date. Group sales are up 6%, driven by pharma with 7%. In Q3, in fact, we had a growth of 7% for the group in CER, and since we now have actually a stabilized currency situation, in Swiss francs, the growth was actually +8%. If you then look at US dollars, we are also at +7%. Clearly the exchange rates have stabilized. As we called out in the presentation in Q3, we had a one-time effect in Q3, which is about CHF 180 million, or 1 percentage point growth.
This is due to a next generation ERP introduction at Chugai in Japan, and this impacted the timings of the shippings. Let me just say that we had a 2% outperformance versus the consensus overall, or about CHF 360 million. The roughly CHF 180 million means that 1% of the outperformance comes from that. The other percent comes from strong performance in the portfolio. Let me also share that this effect will also then reverse in Q4 2026. Diagnostics growth was 4%, and excluding China, and you are aware of the healthcare pricing reforms there, diagnostics grew at +6%. As communicated at half year, we expect full year LOE impact of about CHF 600 million. On the pharma side, we had a very busy quarter in terms of regulatory approvals.
We had four approvals, U.S. approval for Tecentriq in stage three colon cancer, Enspryng in TED and Gazyva in INS, and in the E.U., Susvimo received approval for nAMD. We filed fenebrutinib in RMS and PPMS in U.S., which is under priority review there, and Enspryng in both the U.S. and the E.U. in MOGAD, and vamikibart in UME, also in the U.S. and the E.U. We also reported several positive phase III readouts. One is the CELESTIMO trial with Lunsumio in second line follicular lymphoma. Sefaxersen, at interim readout, showed significantly reduced proteinuria in people with IgAN, and the data will be presented tomorrow at ASN. We also look forward to sharing this data with health authorities. In diagnostics, it is worth highlighting a couple of things. One is the rapid response to two recent outbreaks.
One is Ebola in Congo, and the other one is Cyclospora outbreak in the U.S., where we quickly deployed PCR tests. In neurology, we now also received U.S. FDA approval for pTau217, which is the only marker that can be used in the primary setting for Alzheimer's. On the BD front, it was an exceptionally busy quarter. We made a series of deals, and I will come back later to that, but I just wanted to highlight two, both in the obesity space. One is with Hanmi, a potentially first in class urocortin-2 analog, which is in phase I, and also with Rona to target ALK7 with an RNAi. Both are promising emerging MOAs in the obesity space and have the potential to improve the overall body composition. Looking ahead, we have a couple of key milestones still ahead until year-end.
One is a trial of Itovebi in post-CDKi HR + metastatic breast cancer. Not mentioned on the slide, but of course, we are very much looking forward to our approvals for giredestrant this year in the adjuvant and metastatic setting. The PDUFA dates are November 30th and December 18th. In diagnostics, we also have a launch upcoming. Let me take you through the next slide. Again, just to reiterate, the growth rate is 6%, driven by pharma 7%, diagnostics grew 4%, as mentioned, this is still impacted by the healthcare pricing reforms. As you can see, the impact is getting smaller and smaller over time, so will be back to normal in the near future. Looking at currencies, we have seen a reversal of the headwinds actually in Q3, you have also seen that when you look through the slides in Q4.
In fact, in Q3, we had a positive +1% currency effect. What you see here in terms of currency effect is still really the impact mostly from Q1. Let us move to the next slide. This shows you yet another quarter of consistent strong growth. As mentioned before, in Japan, we had a one-time impact in Q3 due to the ERP system implementation. Excluding this, the growth rate for the quarter would be a strong 6%. As mentioned, this effect will reverse in Q4. Let us look at the key growth drivers. All segments are growing, let me start with oncology/hematology. We have strong Phesgo conversion from Perjeta, which continues. The HER2 franchise is going to peak this year with a strong tail. Tecentriq growth is driven by new indications.
For Alecensa, we have raised the full year outlook to low single digit growth from flat before. Polivy continued strong uptake in first line DLBCL, Columvi and Lunsumio continue to grow in later lines in both DLBCL and follicular lymphoma. Hemlibra keeps growing across the geographies, since we do not expect any competitive pressures anymore this year, we are upgrading the outlook to high single-digit growth for the full year versus what we said at half year, where we still said mid-single digit growth. Let me go to neurology. Ocrevus keeps growing strongly, there is a strong uptake of the subcut formulation. Teresa will provide more details to that. Evrysdi also had a very strong quarter and is expanding its market leadership position in the SMA. Xolair continues its strong performance in food allergy, as expected, we have now in the third quarter seen the first biosimilar launched.
Gazyva, we are very encouraged by the initial uptake in lupus nephritis in the U.S., we also look forward to the approval in SLE later this year. Vabysmo in ophthalmology continues to grow well and we also continue to gain market shares, I know Teresa will also go into that. In diagnostics, we had pathology with a strong growth and Core Lab growing despite the China healthcare price reforms, Matt surely will go into that as well. Let us look at the highlights of the portfolio, I like this slide. It is showing that we can launch up to 20 new medicines until the end of this decade, which is up one versus what we have said at half year. The fact that you see a lot of green boxes means that we have seen a lot of positive news over the last quarter.
Let me just say that we have added two molecules to the slides and had one drop off. The two that we added is Tam-Peli in lung cancer, which had very positive phase III results in second line small cell lung cancer in Chinese patients, and the data was presented at World Lung. Then the VEGF-A IL-6 DutaFab in DME. As I mentioned, this now brings us to 20 NMEs with a potential launch by 2030, and we have had great progress across most of those. Let me highlight a couple of those. First, the positive phase II interim readout for sefaxersen, which I mentioned before. Positive phase II readouts for enicepatide, petrelintide, and CT-996. Based on all this data, we have initiated several phase III studies. As mentioned before, fenebrutinib in MS and vamikibart in UME have now been filed.
For Fene, look out for the data that we are presenting at MSToronto2026 this week. Teresa will also cover that. Finally, we have increased the peak sales potential for divarasib to CHF 2.2 billion-CHF 3 billion on the back of strong Krascendo 1 I data, and we will also present this data this weekend at ESMO, so also please take a look at that. Now let us have a broader look at our pipeline. Again, a slide that I really enjoy looking at because it shows how the effect is on the late-stage pipeline as we implement the bar. In dark blue, you see all of the projects that we have initiated after the introduction of the bar. On the y-axis, you see the PTS, and you see how many projects we have now above 50%. Usually in the past, above 50% meant a very high success rate for us.
To the right, on the x-axis, you see the value. So you see an upward shift and a right-hand shift. This is exactly what we actually want to see. Now, you may ask yourself, what is the little blue dot on the left bottom corner? This is our novel antibiotic, and I am sure you understand why, on the one hand, it is important to bring this project forward and the impact it would have in case of a bacterial pandemic. But you also know that how the business model works today with antibiotics, it is hard to reflect that differently than with a dot in the bottom left corner. On the right-hand side, you see the phase III readouts this year, which has been a very strong year for us. Seven out of 8 phase III studies were positive, so it means an 88 percentage points success rate.
Last year already, we were in line with industry average in terms of phase III success rate. This year, we are actually ahead. Now, 88% is an amazing number. I do not think that this is a long-term rate, but really keeping it at around 80% would be an ideal goal for us. Now, moving to the next slide. This is another way on how we look at our portfolio. We are very rigorous at applying the bar, which means that as we move assets through the development, we constantly assess these molecules. Currently, our portfolio stands at 60 NMEs. What I can say is we have almost doubled the portfolio value since 2023. So not only do we have more meaningful projects, but I think also we have greater value in terms of potential market impact.
We do expect the number of NMEs to increase because we have a lot of things going on in research, as I will show you in a couple of seconds, but we will also continue our partnering activity. Talking about partnering, Q3 was a busy year. We closed in total six platform deals. It was a busy quarter, sorry. Q3, we closed in total six platform deals spanning different drug modalities. For example, ADCs, bispecifics, degraders, and allosteric small molecules. This also in different therapeutic areas. These collaborations will also fuel our early pipeline in the next years. Another source of innovation, clearly, to fuel our early pipeline is also the adoption of AI. My team has also shown you at Pharma Day of how we apply AI in drug discovery, and you see that on the right-hand side.
I do believe that this also shows you that we are frontrunner when it comes to applying AI in drug discovery. You can see on the left-hand side from LI-Go, to LO-Go, to DevGOs, this is the different phases of our early-stage research and development. A significant increase of molecules over this period of time, and we believe this will continue. As a reminder also, we have a very comprehensive AI strategy that we have rolled out a couple of years ago. This AI strategy rests on a couple of pillars. One is, we call Everyday AI, constantly training all of our employees on using AI tools to become more productive. We also monitor the usage of these AI tools in our organization, and we see that 75% of our people are very active AI users.
We feel this helps us to get more productive. On the Reshape side, this is taking the three main processes of our organization, service, marketing and sales, and R&D, and really reshaping these processes using AI to get faster, and more effective, and more productive. What we do is we map that out, we create certain use cases, which we then invest in in order to get to an acceleration. We constantly prioritize new use cases so that we are very effective in providing AI tools to accelerate the work that we are doing across these processes. The third level is around big ideas. These are the things that can really change the way we do drug discovery or diagnostics. Lab in the Loop, as you can see in the right-hand side, is one of those. Also in the diagnostic side, using AI in our products.
We have platform initiatives. One of them is deploying the AI factory. One of them is built in Europe, the other one is built in the U.S. You see that we are making extremely good progress when it comes to AI throughout our organization. Now let me talk about diagnostics. We have previously talked about the key platform launches over the last couple of years, CGM, mass spec, and sequencing. Today, I will dive deeper on the sequencing solution, which delivers unprecedented speed, and very high accuracy. It was launched end of June, and as you can see on the right-hand side, has been getting very positive feedback from customers. Now, we have a very strong launch momentum here. At DIA Day, Matt talked about the target to sell 100 instruments in the first year.
Now in the first 90 days or in the first quarter, we have actually achieved one third of that target, so I think the team is doing well here, and the feedback, I have visited customers myself, has been extraordinarily positive. Now, let us wrap up with the outlook. Here I want to reiterate our guidance that we have also provided both in diagnostics, on the Diagnostics Day, and on the Pharma Day. You see that diagnostics will continue to contribute well to our growth, not only till the end of the decade, but clearly beyond that. Based on all the positive data readouts we have had, we also have very good line of sight for growth to the end of the decade and beyond.
We feel that alone with what we have in hand, we can continue our very positive trajectory that you have seen over the last couple of years, but also that we can go into the next decades with the same kind of momentum. This is without additional positive readouts. You have seen yourself, on the late-stage slide, with the 20 NMEs with a potential launch till 2030, it is very unlikely that this will not turn into a lot of positive readouts. The amount, it is very hard for us to predict at this stage, but there are many potential shots on goals. We do believe that this will add to significant growth on top of what I just said. On top of that, we will continue to do more BD.
I do believe that the growth outlook this year, not only, but also next year and the years beyond that, is quite positive for Roche, and that there is actually no cliff situation that Roche will run into. With that, let me reiterate the guidance, mid-single digit sales growth and core EPS growth of high single digits, and we will increase dividends further. Thank you very much, and with that, I hand over to Alan.
Yeah. Thanks, Thomas. Great quarter. Hello from my side. Before I start, thanks to everybody who has really contributed here to this great sales development that we have seen, and certainly to the benefit for patients. Sales update, so I will be brief. Here is the sales bridge, with the growth of 6% in constant currencies. You see a year-to-date September left-hand side, 2025, year-to-date September 2026 on the right-hand side. When I lead you through the bridge, I think you see pharma. Then in addition, the loss of exclusivity impact of minus CHF 355 million. In total, a growth of 7% in constant rates. Let me say the CHF 355 million we think is a good number to support the CHF 600 million negative impact that we have prognosed for the full year 2026. We will have a stronger impact from Actemra in Q4.
When you look at diagnostics, very steady when it comes to growth, excluding China, 6%. I think also at half year, we had a growth of 6%. Now year to date with +6%. What you see shrinking is the impact from China. We are now at a -13% sales reduction in China compared to last year, same period. At half year, we were at -15%. When you do the math, you will see in Q3 we were at -9%. You see really that there is a clear reduction of that negative impact moving forward. When you then look at the currency impact, I think, and Thomas has alluded to it already, I think we are now at -5 percentage points when it comes to sales. At half year, we had -8 percentage points.
With the stabilization of currencies in Q3, I think there was also a positive impact on the year-to-date impact. Good, and let's look a little bit more into this. You see here the growth rates on the left-hand side in constant rates, 6.2%. That's the 6% that I've mentioned before. On the right-hand side, you see the growth rate in CHF, 1.4%. This is the 1% that we have mentioned before. In between you see the currency movements. I think it is very clear the major driver here is the U.S. dollar as before. As said, I think the impact gets smaller moving forward. Having said this, here is really our overall projection for the full year.
When you take a quick look on the left-hand side, you see that the 2026 average currency rates will remain below 2025 for both the U.S. dollar and the euro. So still a negative impact. You see overall, exchange rates movements had a negative impact on -5 percentage points on the group sales growth in the first nine months, as mentioned already. You know our calculation on the right-hand side, with the estimate for the impact for the full year 2026. If we remain and assume that all the currency rates end of September remain constant until the end of the year 2026, we expect an impact on sales of -3 percentage points, core operating profit of -5 percentage points, and Core EPS of -5 percentage points. That is a slight improvement compared to half year. We had a -4, -6 at half year.
What you see is really when you really look at the top line, you have the sales impact. You see in Q3, we had a positive impact in CHF, and we even expect a higher impact in Q4. So there's a little bit of light at the end of the tunnel, but certainly currency rates in these days are very volatile. Good, and with that, I think just to confirm the guidance again, it's my pleasure to hand over to Teresa.
Great. Thank you, Alan. Pharma sales grew by 7% at constant exchange rates, reaching CHF 36.2 billion by the end of September. All of our regions continue to deliver growth. Of note, as Thomas mentioned, Q3 sales in Japan were positively impacted by a timing effect related to the implementation of a new ERP system in the context of a business process transformation. The size of this impact is roughly about CHF 180 million at constant exchange rates. Excluding this impact, year to date, pharma sales increased at 6% at CER. As Thomas mentioned, we do expect this impact to fully reverse in Q4. Overall, pharma volumes were up by 11%. As always, all absolute values and year-over-year growth rates on this graph are presented at constant exchange rates.
Until the end of September this year, our top brands, Xolair, Hemlibra, Ocrevus, Phesgo, Vabysmo, and Polivy generated roughly $2.1 billion in new sales. I am going to share the growth dynamics for each of these brands on following slides. Why do not we just go ahead and jump in with oncology? Oncology sales increased by 1% at constant exchange rates to CHF 11.2 billion, driven by Phesgo and Tecentriq. Let us start our deep dive with breast cancer and Phesgo. The global conversion rate for Phesgo has climbed to 59%. This is an increase of 5 percentage points over what we shared at half year. It also brings us very close to achieving the previously stated milestone of at least 60% conversion rate at peak. Moving on to Kadcyla.
As we expected, we see increasing competitive pressure in the U.S. and the E.U., and that is negatively impacting the growth of Kadcyla. This is probably a good time to reiterate our HER2 franchise outlook. We continue to expect the HER2 franchise to peak this year at around CHF 9 billion at constant exchange rates 2024, followed by a steady decline through the end of the decade with a solid tail around CHF 4 billion. Half of that is Phesgo, around CHF 1 billion for Kadcyla, and the rest for Perjeta and Herceptin. Due to the increased competitive pressure, we now project Kadcyla to land a bit lower by a few hundred million. To reiterate, we do not foresee a biosimilar in the U.S. for Perjeta before 2028 and in the E.U. before 2027, nor do we foresee any sort of cliff situation around the HER2 franchise.
The Itovebi launch continues with good momentum, and we expect to see phase III INAVO121 results later this year. Last, but certainly not least, in breast cancer, giredestrant. We are eagerly awaiting the U.S. approvals for lidERA in November and evERA in December. For evERA, we are also excited about the recent NEJM publication highlighting the significantly approved PFS in ER positive breast cancer. For lidERA, we expect the single-arm sub-study in combination with abema to read out in Q4, and subsequently to be presented in an upcoming medical meeting. As a reminder, at Pharma Day, we announced an expansion of our clinical development program in adjuvant AR positive breast cancer with two new studies, a phase III IIS in collaboration with AFT and Natera for MRD guided CKi usage, and a phase I/III in collaboration with Pfizer combining giredestrant with the CDK4 atirmociclib.
Now let's shift gears into lung cancer, starting with Tecentriq. The new indications continue to drive global growth led by IMforte in small cell. We just added one more indication with the U.S. approval of ATOMIC in dMMR colon cancer, and this makes Tecentriq the first approved adjuvant immunotherapy regimen in this setting, positioning it to become the new standard of care. I also want to confirm our full year outlook of low single-digit growth for Tecentriq. A quick update for Alecensa. Based on the continued strong performance, we're increasing the full year outlook from flat to low single-digit growth. To close in lung, in just a few days from now, divarasib's positive Krascendo 1 I results in second-line non-small cell will be presented in a presidential symposium at ESMO, so please do tune in for that. With that, let's move on to hematology.
We achieved growth across the entire hematology franchise, resulting in a strong 17% increase at constant exchange rates and CHF 7.1 billion in sales year to date. Hemlibra remains the key growth driver, and as you will recall, at half year, we increased our full year outlook for Hemlibra to mid-single digit. Based on continued strong performance and updates in the competitive environment, I'm happy to share that we once again increased our full year outlook, and we're now projecting that Hemlibra will go at high single digits for the full year. As a brief reminder, at Pharma Day, we presented the upcoming Hemlibra auto-injector. This device has the potential to further expand administration options and enhance patient convenience, and we expect the auto-injector to launch by the end of 2027. Next up is malignant heme.
Polivy has set a new standard of care in first-line DLBCL with continued strong uptake and market share expansion. You know we're not stopping there. The ongoing phase III SKYGLO study for Polivy plus Columvi plus R-CHOP in first-line DLBCL will read out in 2027 and has the potential to once again redefine the standard of care. Importantly, we are not adding our bispecific just onto R-CHOP as some of our competitors have done, but we are adding Columvi onto Polivy plus R-CHOP, the current standard of care in first-line DLBCL. As highlighted at Pharma Day, that phase I data was impressive, showing promising results for the combination with an ORR of 100% with 0% grade 2 CRS events. In true Roche spirit, we have raised the bar, and we are once again seeking to raise the bar on ourself for patients.
Let's also highlight the positive phase III readout of CELESTIMO for Lunsumio and second-line plus follicular. Those data reinforce Lunsumio's potential to improve clinical outcomes while offering a tolerable safety profile, simple dosing schedule, and outpatient administration. We will submit the data to health authorities and present it at an upcoming medical conference. To close, let's take a quick look at Gazyva. As I've mentioned in previous quarters, increasing competitive pressure in CLL and follicular is negatively impacting Gazyva's performance in the hematological indications. This is happening at the same time as the ongoing launch in immunology, and I will share some updates on that in an upcoming slide. In short, what we see is that the initial uptake in lupus nephritis is overcompensating for the competitive pressure across the hematological indications.
Let me repeat that we will show that sales split by therapeutic area in the future as those data sets mature. Now let's move on to neurology. The neurology franchise achieved CHF 7.6 billion in sales and a 10% growth at constant exchange rates September year to date. Ocrevus is driving the overall growth of our neurology franchise, and in turn, Ocrevus global growth is driven by the subcutaneous formulation known as Zunovo in the U.S. We now have roughly 56,000 patients globally on subcut, and that's about 12,000 more than we had just at half year. I'll get more onto the subcutaneous uptake dynamics on the next slide. But before we get there, let me confirm the overall Ocrevus outlook.
For 2026, we continue to expect high single-digit growth, and our peak sales expectations for the Ocrevus franchise remain at CHF 6 billion by 2029 in constant exchange rates at 2025. This includes CHF 2 billion in incremental sales from Ocrevus subcut, but of course, there's going to be some switching from IV to subcut on top. One more good piece of news for Ocrevus, we're excited to have received EU approval for Ocrevus in the pediatric setting for patients 10 years and older based on the positive phase III OPERETTA 2 studies. There is more exciting news for our MS franchise. We have successfully achieved U.S. filing for fenebrutinib in both RMS and PPMS. The filing used a priority review voucher, or PRV, to ensure an expedited review process and enable earlier access for patients.
Furthermore, the MSToronto2026 Medical Conference is currently underway, and we are presenting new data from the fenebrutinib RMS and PPMS studies there, some of which I will highlight on the coming slides. But first, let me share two additional neurology updates. We achieved U.S. and EU filing for Enspryng in MOGAD. The FDA granted priority review, and the PDUFA was set for the 10th of January. As a reminder, we see the MOGAD opportunity of somewhere north of $500 million. We initiated the phase III PrevenTRON trial for trontinemab in preclinical AD, and as you know, the PrevenTRON study is enrolling cognitively unimpaired individuals at a high risk of progression to symptomatic Alzheimer's and highlights our ambition to maximize the potential impact of trontinemab in AD.
Just like for the ongoing TRONTIER 1 and TRONTIER 2 studies, in early AD, we see great excitement amongst the medical community and patients for trontinemab, which is leading to strong recruitment in the PrevenTRON trial. Now, as promised, let's take a bit of a deeper dive into the MS franchise, and let's start with Ocrevus Zunovo. I already mentioned the strong global patient uptake, and you can see very clearly in the graph on the left here. This is based on volume sales across our 50 launch countries, and the adoption rate is driven both by conversion from Ocrevus IV to subcut, but also and especially by patients naive to Ocrevus, which is frankly exactly what we had intended to do.
And for the first time, I also want to share the global Ocrevus IV to subcut conversion, but especially, we can see this on the graph on the right, where you can see that we have now reached greater than 20% conversion to the subcut formulation. And we expect to continue this to climb, and in fact, today, Ocrevus subcut is the fastest-growing anti-CD20 in the market. And that is really fundamentally the story that these two graphs together tell. Subcut has allowed us to expand our MS footprint to new market segments, particularly in the community setting, where IV constraints mean that Ocrevus is simply the better-fitting option for patients, doctors, and health system needs. And all of this together allows us to continue to grow the Ocrevus market overall. And let us not forget that, again, we are not stopping here.
We are further innovating for patients with the development of our on-body injector, or OBI. The OBI delivery device has the potential to enable a convenient twice-yearly at-home self-administration. We showcased the device at Pharma Day, and we are expecting our data readouts next year. Now let us take a closer look at some of the data being presented for fenebrutinib. Fenebrutinib, as you know, is the first and only BTKI with positive phase III results in both RMS and PPMS, showing a clear positive effect for both relapsing and progressive biologies. And the additional RMS data currently being presented at MSToronto2026 are further reinforcing this profile, including acting on markers of progressive disease in the RMS population. Now, I do not want to go into too much detail here, but let me quickly cover the highlights of the data on the slide.
On the left, the NEDA-3, which is no evidence of disease activity, results show a 70% reduction of risk for fenebrutinib compared to teriflunomide and in regards to ARR, CDP, and MRI activity. In the middle, we have compared the total brain volume loss between fenebrutinib and teriflunomide patients. Not only is there a clear benefit for fenebrutinib compared to teriflunomide, but patients on fenebrutinib reached a level in line with the brain volume loss rate of healthy adults. On the right side, you see the continued improvement for two key biomarkers of progressive biology for the fenebrutinib arm, CHI3L1 for microglia and SERPINA3 for astrocytes both show continued improvement for fenebrutinib versus teriflunomide through week 96. And of note, the positive effect of the CHI3L1 biomarker underscores fenebrutinib's ability to cross the blood-brain barrier and act on microglia.
And there is one more piece of data that I want to highlight in a late-breaking abstract with the oral presentation to follow in the coming days at MSToronto2026. Fenebrutinib is showing its therapeutic effect on paramagnetic rim lesion and slowly expanding lesions known as chronic active lesions.
In short, fenebrutinib demonstrated the first prospective therapeutic effect on chronic active lesions in MS for both RMS and PPMS. In RMS, fenebrutinib reduced the volume of slowly expanding lesions by up to 30% and reduced accumulation of paramagnetic lesions by 62%, both versus teriflunomide at 96 weeks. These are exciting and meaningful results that underscore fenebrutinib's effect on chronic neural inflammation in the brain, a primary driver of disability progression in MS. So in summary, we are convinced that fenebrutinib has strong efficacy across relapsing and progressive biologies. And to further underline this, let us take a look at the next slide.
Also at MS Toronto, we presented pooled ARR data from the FENhance 1 and FENhance 2 trials in RMS. This slide really speaks for itself, so I can be brief. Fenebrutinib achieved an impressive 55.4% relative ARR reduction compared to teriflunomide. This makes fenebrutinib the first and only oral to demonstrate an ARR improvement greater than 50% versus an active comparator in RMS. That puts fenebrutinib firmly in the category of a high-efficacy treatment, as you can see here from the review article taken from Neurology. Yes, we are excited about fenebrutinib's potential in MS, and we are looking forward to bringing this truly transformative medicine to patients. Now let's move on to immunology. Our immunology franchise reached CHF 5 billion, growing 4% at constant exchange rates, with Xolair being our key growth driver.
We now have more than 145,000 patients treated with Xolair for food allergies since launch. That is an increase to 15,000 patients since half year. As you saw in Q3, and as expected, we saw the first U.S. biosimilar in launch for Xolair. Let me briefly comment on the Q3 performance and the outlook for Xolair. Q3 was impacted by some buying pattern, and we continue to expect around 20% growth at full year. Actemra sales declined by 10% year to date, driven by biosimilar impact, as was mentioned earlier. Then moving on to Gazyva, we see continued strong launch uptake and patient share gains in lupus nephritis across early launch countries. By now, we've treated about 5,000 patients with Gazyva for lupus nephritis globally. As mentioned on the hematology side, the uptake in lupus nephritis is more than compensating for the erosion in hematology.
One more good piece of news this quarter for Gazyva, we received U.S. approval for INS, and looking forward, we are anticipating U.S. approvals in both SLE and MN for Q4. As Thomas mentioned, one additional highlight of this quarter was certainly the positive phase III interim data for sefaxersen in IgAN. Sefaxersen is the first antisense oligonucleotide evaluated in this disease, and the results show that it has best-in-class potential for selective complement suppression with a clinically meaningful reduction in 24-hour proteinuria. The positive UPCR data will be presented tomorrow at the ASN conference, and we're also planning to share that data globally with health authorities. Of course, the study is still ongoing to determine long-term functional eGFR outcomes. With that, let's go on to ophthalmology. Ophthalmology sales reached CHF 3.3 billion year to date, growing 9% at constant exchange rates.
Vabysmo is the key driver here, and we continue to see global growth in market share gains. By now, Vabysmo has established itself as the preferred first-line treatment, including in the U.S. We have also reached another milestone with an impressive 1.5 million patients treated globally. One more update for Vabysmo. Earlier this quarter at EURETINA, we shared updated real-world data from the VOYAGER study, which now includes more than 5,000 patients. These results reinforce Vabysmo's robust effectiveness and durability across previously untreated and treated eyes in AMD, DME, and RVO. We are happy to see the strong clinical results for Vabysmo once again replicated in real-world clinical practice. Let me also reiterate that we expect low double-digit growth for Vabysmo in 2026.
As previously stated, we consider consensus peak sales expectations for Vabysmo of around CHF 6 billion as very reasonable, and we are confident that we are on track to achieve this number. Beyond Vabysmo, we had quite a bit of additional news flow in ophthalmology this quarter. Enspryng just achieved its U.S. approval in TED just last week. Susvimo achieved EU approval in AMD, and we filed vamikibart in UME in both the U.S. and the EU with a PDUFA set for July 23. We moved our bispecific PD-1/VEGF dutafab into phase III in DME. Of note, this includes a head-to-head trial against Vabysmo, which is frankly something that we are increasingly hearing from ophthalmologists. They want to see trials that use the current standard of care as the comparator, not historical standards of care. Next up, I will cover our CVRM pipeline.
As you heard at Pharma Day, we continue to develop our CVRM assets at pace. There are quite a few positive readouts and recently announced phase III trials to cover. Let's start with enicepatide and petrelintide, the foundational building blocks of our CVRM pipeline. I am going to stay a bit more high level here as we have a more detailed slide on both molecules right after this. Let's just cover the Q3 highlights here. Both enicepatide and petrelintide delivered positive phase II data in obesity with type two diabetes. For petrelintide, we also presented the full ZUPREME-1 data at EASD and initiated the phase III program consisting of ZUPREME- 3, ZUPREME-4, and ZUPREME-5. More on the next slide, but first let's look at the rest of the pipeline, starting with CT-996, our oral GLP-1. First of all, we have a new name to unroll, tifaglipron.
At Pharma Day, we announced positive phase II data for tifeglipron in obesity, and based on these positive results, we are moving the asset to phase III, with trials to be initiated in the first half of 2027. As Thomas mentioned, we also added two new molecules to the CVRM pipeline with partnerships both in obesity. First, a UCN2 analog in phase I via our partnership with Hanmi Pharmaceuticals. Of note, UCN2 has the potential to simultaneously promote weight loss and preserve lean body mass. Second, an ALK7 targeting RNAi via our Rona Therapeutics partnership. This asset is preclinical, and we think it is a very exciting new approach in the obesity field, directly targeting adipose tissue biology, unlike the incretin-based therapies, which target the CNS. I am looking forward to sharing more on these molecules in the future. But now let's take a closer look at enicepatide and petrelintide.
You all know that we at Roche have the ambition to become a top three player in obesity. The two foundational enemies that we believe will get us there are enicepatide and petrelintide, both as monotherapies and in combination. Let me highlight not only why we believe in the potential of these assets, but also the significant progress we have made in the last few months. Let's start with enicepatide. Enicepatide is intentionally designed with a biased GLP-1/GIP agonism, which we believe is key to the strong phase II results we have seen so far. In obese overweight patients without type 2 diabetes, enicepatide did show robust weight loss of up to 22.7% at 48 weeks, with no sign of plateau. Importantly, more than a quarter of treated patients achieved a weight loss of 30% or more at 48 weeks.
Recently, we shared that in obese and overweight patients with type 2 diabetes, a weight loss of 15.5% at 48 weeks was achieved. Again, there were no signals of weight loss plateau after that 48-week mark. Even more impressive, however, was the HbA1c reduction in this population. At 48 weeks, enicepatide achieved reductions of up to 2.65%, highlighting its potential for best-in-disease glycemic control. On the back of these strong phase II results, we have started a comprehensive phase III program. Two trials in obesity are already ongoing, and four trials in type 2 diabetes and glycemic control are to be initiated in the first half of next year. This includes a head-to-head trial against tirzepatide, underscoring our conviction in enicepatide's best-in-disease potential. Additionally, we are expecting to initiate a CVOT and HFrEF trial in the first half of next year.
Now let us turn our attention to PETRI. Together with our partner Zealand, we believe that this long-acting amylin has the potential to redefine the weight management experience, in particular by delivering double-digit weight loss and improved tolerability compared to current treatment options. Positive phase II results from ZUPREME-1 and 2 studies were announced this year and gave us the confidence to move to phase III. To pick just one highlight, the recent ZUPREME-2 results show that discontinuations due to GI AEs are very similar between PETRI and the placebo arm. 1.9% of patients on petrelintide discontinued treatment due to GI AEs, and 1.7% did so on placebo. This while achieving statistically significant and clinically meaningful weight loss in HbA1c reduction. We are moving at pace with PETRI as well.
We initiated all three phase III studies in obesity this quarter, with each of them already having achieved FPI. One of the great things about having both enicepatide and petrelintide in our portfolio is that we can combine them, and so we are doing just that in our upcoming Zenergy phase II trial. Combining these two molecules with their respective strong clinical profiles has the potential to address unmet needs for patients beyond the scope of monotherapy, in particular for patients needing greater weight loss and/or better glycemic control. That phase II trial is expected to initiate later this year. Taken all together, you can see why we are excited to have two strong and combinable building blocks in our CVRM pipeline, and how they support our ambition to become a top three player in obesity.
With that, let us move to my final three slides, few slides. Here we have the 2026 Key Pharma news flow. A lot of green check marks, as Thomas noticed earlier, which we love to see, and all of the updates I have already covered on previous slides. To close things out, let us take a quick look at 2027. This is our expected news flow for next year, which we presented to you for the first time at Pharma Day just a few weeks ago. When you look at the table, one thing is immediately obvious, and that is that 2027 is a year with many key regulatory milestones for us. Our regulatory and launch teams will be exceptionally busy in the months to come to ensure our launch success for our key NMEs online extensions.
However, let's not forget about the upcoming clinical readouts, which include phase III data for two high-profile NMEs, NXT007 in hemophilia A, and afimkibart in ulcerative colitis. I'm very much looking forward to next year and to updating you on all the progress that we will achieve across pharma for patients. With that, let me hand it over to Matt.
Thanks, Teresa, and congrats on the strong quarter. With that, let's get started on diagnostics. As you heard from Thomas and Alan, with sales of CHF 10.3 billion, the diagnostics division grew by +4% at constant exchange or +CHF 388 million compared to September 2025. As you also heard earlier, this growth was achieved despite the ongoing impact of healthcare pricing reforms in China. Excluding the effect of China, the growth of the business was +6%, as Alan mentioned. Sales were additionally impacted by CHF 70 million loss of sales due to the weaker respiratory season in the northern hemisphere. Despite that, very steady progress and development from the diagnostics division. Looking forward into next year, we expect to return to our constant ambition of mid to high single-digit sales growth.
Now let me walk you through the results by our customer areas. Sales in the Core Lab increased at +5%, impacted by the previously mentioned policy impacts in China. Excluding this effect, sales in the Core Lab grew at +9%. Sales in the molecular lab grew at +3%, with strong growth in transplant at +17%. However, this was partially offset by the previously mentioned mild respiratory season that lowered testing volumes. The same effect in our near patient care business, which decreased at -5%. This was driven by a -21% decline in our cobas liat sales, but was partially offset by growth in our blood glucose monitoring business at +1%, following competitive wins. Sales in our pathology lab grew strongly at +10%, driven by advanced staining growth of +7% and companion diagnostics growth of +21%.
Now let's take a geographic view of our performance. In North America, the business grew at +9%. In EMEA, the business grew at +3%. LATAM grew strongly at +12%. In APAC, the business declined at -4%. As previously mentioned, sales growth was impacted by the healthcare pricing reform in China. As a result of this, China sales declined at -13%. Excluding this effect, APAC grew at +6%. You heard a little bit about our groundbreaking AXELIOS sequencing launch. Now I'd like to transition to our portfolio updates, starting with the AXELIOS sequencing solution, where as you heard, we've experienced strong launch momentum, achieving approximately a third of our instrument placements that we're targeting for the full year, within the first quarter of launch.
I would emphasize that in order to continue this momentum, it is critical to demonstrate the capability of AXELIOS across the most commonly used applications, while at the same time building out the ecosystem of multi-omic tools and partners to support their use. I am going to walk you through the slide from left to right. Starting on the left, I would like to highlight data from UCSF. This was presented this week in Montreal at the American Society of Human Genetics conference. This data illustrates the capability of AXELIOS for single-cell RNA sequencing in core labs and some of its key advantages. UCSF demonstrated a high correlation between AXELIOS and existing sequencing by synthesis workflows across a broad spectrum of cell populations, as you can see on the chart. What I would call out is that AXELIOS delivered approximately the same number of reads.
However, these reads were approximately three times longer, consisting of 300 base pairs as opposed to 90 base pairs with the comparator. This enabled deeper isoform resolution. Finally, UCSF demonstrated the suitability of AXELIOS for its core lab operations across different applications, including 10x, single cell, and bulk RNA sequencing. If you move to the center of the slide, the second set of data, which was presented by GeneDx, again at ASHG, highlights the complementary capabilities of our SBX Duplex high accuracy and SBX Simplex single read, longer read modes for whole genome sequencing in rare disease. Today, GeneDx uses frontline short read sequencing combined with additional confirmatory testing for certain variants. AXELIOS has the potential to consolidate these workflows onto just a single platform by combining our high accuracy shorter reads using our duplex mode together with the longer read capability of our simplex mode.
Our SBX Duplex mode showed concordance with existing frontline testing for all assessed variants. Furthermore, the longer reads generated by SBX Simplex showed the potential for direct readthrough of expanded repeats that would normally require additional confirmatory methods. Finally, on the far right, I would like to highlight our collaboration with Twist Bioscience to develop a target enrichment kit for AXELIOS. This collaboration will deliver workflows for different sample types, including cell-free DNA and FFPE tissue samples. Together with our previous announced collaborations such as 10x Genomics and Olink, this will further expand our SBX ecosystem and enable a broader range of applications and workflows. Overall, these examples further demonstrate the capability of AXELIOS across all relevant sequencing applications, together with expanding our ecosystem of partnerships.
Now I would like to move to our molecular diagnostics infectious disease portfolio and start with the CE mark for our cobas HDV assay, which was announced on July 14. Hepatitis D affects nearly 12 million people globally and severely affects people who are already co-infected with HBV virus. Despite being the most severe form of viral hepatitis, testing still relies largely on manual methods or lab-developed tests, and there is currently no fully automated molecular IVD solution available for hepatitis D. With cobas HDV, we are introducing the first high throughput automated molecular IVD assay for detection and quantification of hepatitis D virus. This will support the identification of infected patients, as well as monitor their response to treatment. This is becoming increasingly important as novel therapies for hepatitis D become available.
With this launch, we further strengthen our leading infectious disease portfolio, specifically in the highly competitive and valuable hepatitis segment, and expand our molecular testing offering in an area of growing clinical relevance. Now I'd like to continue with molecular diagnostics and pick up on something that Thomas mentioned earlier. I'd like to highlight our rapid outbreak response capability to address public health challenges on a global basis. Within our TIB Molbiol organization, we maintain a library of over 15,000 test pre-designs across a wide range of pathogens and more than 3,000 positive control materials. Combined with our strong expertise in oligonucleotide design, we can rapidly develop primer and probe combinations for new pathogens. These tests can be rapidly deployed on our LightCycler and X800 systems to provide prompt response to emerging public health needs.
Given our global installed base, this is something we can scale very rapidly. A recent example is the Ebola Bundibugyo outbreak in the Democratic Republic of Congo, where we developed a PCR test within just six days of the virus genome being published. I would call out that commonly used PCR tests that were available at the time of the outbreak were not able to detect the Bundibugyo viral clade, leading to delays in diagnosis. Our test is able to accurately detect this viral clade, and we're actively working to support the response to the outbreak in the DRC. A second example is the recent U.S. outbreak of cyclosporiasis. In August, we announced two new PCR RUO tests, one specifically targeting Cyclospora and another detecting the parasite as part of a multiplex gastrointestinal parasite panel. This provided a key resource to support outbreak surveillance.
These examples highlight our leadership in molecular diagnostics and underscore our commitment to support global public health response to emerging infectious diseases. To conclude, I'd like to report progress on our key launch list for the diagnostics division. For the 11 launches here, we've achieved seven by Q3, and we're making good progress towards the other. I would just call out some of the tremendous launches we've had this year, such as our pTau217, our Elecsys Neurofilament light chain, and our Elecsys TB IGRA assay, all of which have tremendous potential for the growth of our division in the future, as well as the ones highlighted by Thomas. With that, thank you very much for your attention, and I'll hand it over to Bruno for the Q&A.
Thank you, Matt. With that, we will start our Q&A session. First questions come from James Gordon from Barclays. James, please.
Hello, James Gordon from Barclays. Thanks for taking the questions. One question on hem oncology, which was Genmab and AbbVie recently announced strong headline data from the EPCORE DLBCL-2, first-line DLBCL study, and they had a strong benefit for Epkinly on R-CHOP, and it looked like a clean side effect profile. The question was, does that raise the bar for what you hope to see for Columvi plus Polivy in SKYGLO? If SKYGLO delivers on your expectations, how do you think first-line patients will be treated in DLBCL? Is it going to be a segmentation by the severity of someone's disease or something like that, balancing side effects and efficacy? The second question was on neurology, and the slide on ARR benefits in RMS was interesting, and I saw the strong benefits that fenebrutinib has got.
But then also, how do you think about ARR versus CDP benefits, so disability versus relapse rates? Because different agents have different profiles there, it seems. More broadly in MS, $9 billion peak sales for Ocrevus, but what is that assuming for the BTKs, and do you still think you can do that even if the BTKs are very good there?
Great. Let's start with Epkinly and where we think second-line DLBCL would go. I would say, I think we found that data very encouraging for a variety of different reasons. First of all, let's remember that we've already seen very strong phase II data for our own combination in this setting. We do believe that that phase II data will translate into phase III. Although Epkinly is chemo-free, let's just remember it does actually carry quite a bit of toxicity, including higher rates and severity of hematological toxicity and CRS, compared to what we've seen in both SUNMO and STARGLO. In addition, it carries a much more cumbersome dosing regimen. I think we have both a combination that we believe is going to be very efficacious and much more tolerable, and frankly easier to give from a community physician perspective.
I think if anything, the data that we saw actually just raises our confidence in what we're likely to see with our own trials. When it comes to fenebrutinib, let's start with your last question first, which is we think there is actually still plenty of room for BTKs in MS. We still have 45% of patients today that are on a low-efficacy therapy. Those patients are very ripe for moving over to a BTK. So we continue to believe that we have that $9 billion opportunity for Ocrevus between Subcut, and let's not forget, we also have OBI coming as well. There's still a lot of opportunity for us with Ocrevus in this space. Then we believe fenebrutinib will ultimately just crack up a whole new market segment, very much like Subcut did for us with the community physicians for Ocrevus.
When we look at fenebrutinib, again, I think when you look at ARR versus CDP, in RMS, fenebrutinib set a new efficacy bar. We have just never seen medicine with efficacy like what we've seen with fenebrutinib, 50% reductions in ARR versus teriflunomide, 70% reductions in brain regions. A clinically meaningful and nominally significant improvements in cCDP12. We are definitely in that high efficacy category. We believe that fenebrutinib really does represent that best-in-class BTK based on the non-covalent binding, the differentiated clinical efficacy that we've seen in phase III, and that dosing really is optimized to reach the concentrations in the blood and CNS that we believe will be clinically meaningful in terms of immune modulation. Net-net, I think Feni has set a very high efficacy bar. We see benefits in both progression and disability.
We think it is the one to beat from an efficacy perspective.
James, did we answer all your questions?
Yes, you did. Well, I've got lots more questions, but I'll get back in the queue. Thank you.
Okay.
Very polite of you.
Then let's move on. Next questions go to Matthew Weston from UBS.
Thanks very much, Bruno. Two from me, please. The first one, quite simple. Teresa, when should we expect to see first overall survival data from lidERA? Obviously, there are a number of clinicians who are looking at the strength of clinical efficacy data when they look at guidelines and how to integrate giredestrant into the treatment algorithm, and OS is clearly going to make a huge difference to some of the more skeptical docs. The second question, just reflecting on the Pharma Day, it felt like you were more enthusiastic on TL1A combos than perhaps monotherapy. Maybe I misinterpreted, but if not, when should we think about the real commercial contribution of afimkibart actually having an impact on Roche?
Yep. So great questions, Matt. Thank you. So with lidERA, the next OS cutoff is next year, and we anxiously await that just like everyone else. So, more to come. In terms of the TL1A combos, I think we are still very encouraged by or very excited by what we plan to see with monotherapy data with TL1A. Obviously, that's the data that we had with the phase IIs, which were very impressive. So I would expect afimkibart to make a meaningful contribution off the bat as we would expect with any new launch in a disease area of this size. I think our excitement around the combos comes from the fact that, as with any immunological disease, these diseases are very heterogeneous and complex. So the reality is, in order to get the maximal amount of efficacy benefit, a combination, rational combination, likely makes sense.
I think those are things that we will continue to pursue. I wouldn't say I am unexcited about the mono opportunity. I think the data that we saw in phase II was quite compelling and would represent a significant step forward for these patients.
Could I just push, Bruno, if I'm allowed?
Yes, please.
The OS data, can you give us any indication during the year, first half, second half? I'd love a month or a week, but-
It is more towards
lidERA OS.
the first half.
Yeah, it is more towards the second half.
More towards the first half.
Yeah.
Perfect. Many thanks indeed.
Yeah, we haven't been more specific than that. Yep.
No, that's great. First half's perfect. Thank you.
Yep. Then let's go on. Next question go to Richard Vosser. Richard?
Hi. Yeah, thanks, Bruno. Couple of questions, please. Firstly, just going back to Ocrevus, we've seen competitive pressures from Briumvi. Just when you think those might abate a little bit, and also, on Zunovo, it does seem at the moment that it's more switching, which is obviously good for the LOE, but really not adding incremental potential. When can we see that actually adding? Secondly, just giredestrant. We get a lot of discussion and questions around the use of sampling and patient assistance programs in oncology. We do not normally see this, so perhaps you could give us some color, the types of patients, how long this will be used for, why it is being used for, the reasons behind it. I think that would help everybody. Thanks very much.
Yeah. Absolutely. These are all very good questions. Where to start? Let's just take them in order. With Briumvi, I think the reality is Briumvi is getting inroads because it is using very aggressive contracting strategies. We have talked about this now for the last couple of quarters. This is something that we believe ultimately is going to be manageable, or is ultimately manageable, and it is probably ultimately not super sustainable. If you look at their share of market versus their share of value. These are aggressive strategies. I think we do believe that ultimately we will be able to handle the Briumvi impact. From a Zunovo standpoint, actually 50% of starts of Zunovo are naive to Ocrevus. We are seeing a significant uptake in the naive population.
This is exactly, as I mentioned before, what we really had hoped to see with Zunovo. Ocrevus was very heavily penetrated into our academic centers, into urban centers, where IV capacity is very readily available. But when you look into more rural settings or more community-based treatment where IV capacity is more limited, this is a perfect place for Ocrevus Zunovo, and that is where we saw a lot of our initial uptake, and that is where a lot of those new naive patients come from. I will say, as you do see switching in some of the bigger institutions, patients love the convenience of the 10-minute injection. I think you really see a nice pickup on both sides. That is really what you want both for Ocrevus. As I mentioned, it is the fastest growing anti-CD20 therapy out there right at the moment.
We are quite pleased with what we are seeing with Ocrevus Zunovo. It might surprise you to learn that I am not going to reveal all of my launch plans to you on this call, but what I can tell you is that we are very well prepared for ensuring that patients have the ability to access giredestrant out of the gate. That does in fact mean that we will do things maybe that you have seen more similarly or more familiarly in the immunology settings. We do anticipate creating an access hub program where patients will enroll and where we will be able to give them assistance as we do in all of our therapeutic areas with maintaining reimbursement. We will have a bridge program. I will not go into specifics on that.
What I can tell you is that when giredestrant launches, what you can expect from us in terms of how we will report. Of course you will see sales. What we will also be able to give you are the total number of patients who are enrolled in the hub program, and we will provide that to you on a quarterly basis. All patients in the hub program will have received a prescription. It will start to give you a sense of how many patients are actually actively either getting or in the process of getting giredestrant. Hopefully that gives you a little bit of a sense of what you can expect in terms of the metrics that we will be using more publicly to measure the launch. Just to reiterate, we are expecting that kind of hockey stick-like uptake.
We all know that this is a chronic therapy. These patients stay on therapy for a very long time, five years plus. We would expect to see more of a hockey stick-like uptake. Then again, just to reiterate what I said at half year as well, consensus is about CHF 500 million-CHF 600 million for giredestrant next year, and we think that that is right in line.
Perfect. Thank you very much, guys.
Richard, answered all your questions?
Yeah. Thank you.
Okay. Next questions go to Graham Parry from Citi. Graham, please.
Great. Thanks for taking the questions. Follow-ups there on giredestrant as well. On the coverage programs, just a couple of points or questions there. One, do you expect that to cover Medicare patients as well as commercial? What sort of percentage of patients do you think will be covered by insurance at launch, and how do you expect that to progress through 2027, or is that going to be more of a step-up in 2028? Do you need the overall survival data for EMA filing? On the Pharma Day, and just now, you suggested the consensus CHF 500 million-CHF 600 million is in line with your internal expectations, but does that include just naive patients, or are you including switch patients who are already on aromatase inhibitors or have stopped aromatase inhibitor CDK4/6 due to tolerability issues? Thank you.
Yeah. Maybe just to mix it up, we will take them in the reverse order this time around. In terms of who are the patients that we expect to come on initially, I think it will be a mix. Obviously, the biggest sort of bolus of patients you might expect are patients who have dropped off or are looking to drop off their CDK 4/6s due to tolerability. It is clearly a fairly large group of patients, and we would expect that they are anxiously awaiting the giredestrant launch. Then you have naive patients as well who would clearly be very appropriate to start from a monotherapy perspective as they begin their journey. Then over time, and as our data sets mature, I think obviously you are also going to get those patients who are in need of adjunctive therapy to their CDK 4/6s.
Net, I think out of the gate, there is actually a significant opportunity within these patients, and as Bruno reminded me before we started today, we have put a ton of epi data in the appendix of this slide deck. Anything and everything you would ever want to know about the epidemiology of hormone receptor-positive breast cancer is all in the appendix. Please take a look at that. Also for Pharma Day, the Pharma Day appendix is where it is. There is a ton of great data, so please do take a look. In terms of overall survival, we do intend to start our conversations with EMA next year.
In keeping with our longstanding commitment to make sure that we go to markets with the filing packages that we believe have the highest likelihood of success to achieve sustainable reimbursement, we would hope to have OS in that package when we begin to talk to our European regulators, which we would expect to do next year. In terms of giredestrant, I am not going to get into a ton of details for competitive reasons on exactly what the bridge program will look like. Clearly with any new medication, there is always payer challenges, and you can rest assured that we have been thinking very strategically about how to ensure that we are able to help as many patients as possible get on therapy as quickly as possible out of the gate.
I will not say more on that, but know that we have put a lot of care and thought into our patient assistance programs for giredestrant at launch.
I think maybe, Parry, what we can add for the bridging programs, I think they are set up for the first six to nine months
Yes
in this phase when there is still some blocks in place until we have resolved these issues. Any additional questions? All your questions are answered?
No, that's great. Thanks, Bruno.
Then let's move on. We move to Steve Scala from Cowen. Steve.
Thank you so much. Two questions. First for Alan. There are several different 340B reform proposals in the U.S. Should one of them be passed, what is the range of possible outcomes for Roche? For instance, could it be a material upside for Roche? Secondly, regarding the PD-1/ VEGF target, since Roche's Pharma Day, one of your peers that had been firmly on the sidelines has completed a big investment. Roche is now one of a dwindling group of major oncology companies not involved in this target. Is this Roche's final word on the topic, or is Roche's view in transition? Thank you.
You want to?
Why don't you take the first one? I take the second one.
I can take the 340B topic. So obviously, none of the 340B proposed rules are yet final. We are looking at them closely and considering what any of them might mean for our business. What I can tell you right now, based on what we know, is we would not expect any of them to be material, and they seem very mitigatable from our perspective.
Yeah, and to your second question, we are definitely not on the sideline. We have actually internal programs which look at bispecifics and trispecifics in this space that also includes PD-1/VEGF. Steve, does this answer all your questions?
It does, Bruno. Thank you.
Then we move on, and next one is Luisa Hector from Berenberg. Luisa.
Thank you, Bruno. I have a couple. I will go for giredestrant again. So maybe just whether you could comment, we are estimating that around half of the eligible patients might be Medicare. I do not know if you can give any color around that. Then the second question really connects to the outlook for R&D spend, but also business development, because we can see the high level of success that you have had this year. So I wonder if that changes your appetite for deals or the focus of the deals may be now shifting more early stage since you have such a full late-stage pipeline, and therefore, what does that mean as we think about R&D spend into 2027? Thank you.
Yeah, I can take the second question, and you take the first. I can go with the second.
R&D?
No, let me just quickly-
Yeah, he's going to take R&D.
give an outlook on R&D spend also for next year and also in terms of what you can expect. First of all, we do expect that the growth that we have, we can continue our good growth trajectory into next year. I think what we have already committed is that we will at least keep margins stable. I can say that we will definitely increase margins also next year. Of course, this means that also we will have more money flowing into R&D because we have so many trials ongoing. I just want to reassure you, we will increase margins next year. That's the first thing that I can say. On the BD side, I completely agree with you. We have a very full late-stage pipeline. The pressure that maybe other companies feel, we don't feel that pressure.
That doesn't mean that we're not open to late-stage opportunities, but they have to make sense. They have to make sense financially. They have to make sense from a scientific perspective. And clearly what you have seen, we've done more early-stage deals also to make sure that we have a full pipeline once we get beyond all those different medicines that will read out. I think we are in a good position here. As mentioned before, we have sufficient M&A firepower if we need it or if we want to spend it, but we're not in the same pressure situation. Certainly, we will not overspend, and we'll stay disciplined.
Absolutely. In terms of the insurance breakdown, we have not disclosed that to date. What I can tell you is that we are actively looking at all types of insurance and again, how we can help ensure that patients get on therapy as quickly as possible.
Mm-hmm. Luisa, did these answers help?
That's great. Thank you.
Yeah. Then we go on. The next one I need you is Trung Nguyen from RBC Capital. Trung, you are unmuted.
Apologies. Thanks for taking my questions. I have two, if that's okay. Just the first one, can you just share us your expectations for the giredestrant label ahead of that November 30 PDUFA? Do you have any color from the FDA on safety concerns? It does look like the data safety is quite clean, but we're just curious if there's any expectations for any warnings, particularly around bradycardia signals we saw in lidERA. Could that translate to any monitoring requirements? On Xolair, you reiterated the 20% full-year growth target. We understand the buying patterns in Q3, but on our calculations, you need close to 40% growth for Q4 to achieve that target. Just what gives you confidence in that trajectory? In Q1, you indicated you shouldn't expect any additional Xolair biosimilars beyond the first launch for at least two years. Does that guidance still hold? Thank you.
Yeah. So in terms of the giredestrant label, we're obviously in active discussions with the FDA right now, so I'm not going to comment on what we believe will be in the label other than to say, you should anticipate it would be broadly in line with what we saw in lidERA. Again, many of the safety events that, like QT prolongation, et cetera, that you've seen with other SERDs, you just simply don't see with giredestrant. Again, I think you've seen the data, you've seen the trial design, and we should extrapolate a label. In terms of Xolair, so we do believe that the buying pattern had an impact in Q3.
We obviously do now have the first biosimilar that is in market for Xolair, and that will have an impact as well, but we are still quite confident in the around 20% growth that we have been messaging for full year. In terms of additional biosimilars, we expect that there could be one or two additional biosimilars coming in U.S. That is possible, but that is contingent on U.S. approval. Right now we are staying with the outlook that we had provided earlier, which is again, around 20% in 2026 and flat in 2027.
Maybe trying quickly to add, I think we had two very strong quarters beginning of the year, a +26% and +28%, so now the third quarter is a +8%, so I think it is doable. The other thing is I think the two additional potential biosimilars to come next year, they were always back on our mind. We will have to see how this plays out, but currently it stands as Teresa.
I think we can say, Bruno, we are at +20% year to date.
Yeah.
I think we need another Q4 with 20% here.
Yeah. Exactly.
to get to where we want to be.
Exactly.
Yeah.
Yeah. Any other questions?
That's great.
No?
Thanks very much.
Okay. Then we go on. Next one would be Michael Leuchten from Jefferies. Michael, please.
Thanks, Bruno. Could I please ask about the NOVERRA study and the CDK4/6 combo study that Teresa flagged? They seem to be single-arm studies. Just wondering why, and how you would be able to get those on label if and when they read out. Then clarification please on Vabysmo, Teresa. Helpful to confirm the growth outlook for the end of this year, low double digit. Dynamics into next year, I think previously you said, there is no structural change to the growth profile for Vabysmo. Does this still stand for 2027 if the U.S. growth momentum doesn't come back and we see a little bit of a sequential flatlining in Europe?
Yeah. So just to confirm with NOVERRA, those trials are not meant to be label-enabling. They are meant to generate safety in some experience data. But they are not anticipated or expected to be label-enabling. When it comes to Vabysmo, I think we have been pretty consistent in where we believe the market for Vabysmo is heading into the future. I think the U.S. has re-baselined. We are expecting to remain in that sort of low single digit growth for Vabysmo in terms of sales. It is probably important to note though, that the volume growth in the U.S. is much higher. There is a contracting effect that kind of takes effect there. Then when you look ex-U.S., there is actually significant growth for Vabysmo and sort of more opportunity for growth and penetration to come outside of the U.S.
I think we remain very confident in that CHF 6 billion consensus peak sales number and believe we are well on track to achieve that.
Michael, did this answer all your questions?
Yes, thank you.
Then, next questions go to James Quigley from Goldman Sachs. James.
Great. Thanks, Bruno. Thanks so much for my questions. This is James Quigley from Goldman Sachs. I've got two, please. One follow-up on Xolair. Can you give us a rough idea of the split between the immunology and the food allergy indications in terms of sales or proportion or whatever you can give us? Just trying to think about biosimilars as well in terms of what point could the biosimilar start to hit the food allergy indication? Are there any other exclusivities or protection that you have for that indication? The second one is on Tam-Peli. What are your plans here for indication expansion? Some of the competitors in China are looking at significantly expanding into a broad set of cancers. Given the speed of movement out in China, how quickly can you move to establish an early advantage across B7-H3 tumors?
What are the gating factors that you need to overcome here as well? Thank you.
Yeah. Great. Let's start with Tam-Peli. I think we shared at Pharma Day that we have great excitement around Tam-Peli. Obviously the global phase III trial in second line is being initiated by the end of the year. The details of that study design, including the comparator that we'll use, will be announced in due course. Then we are actively looking at accelerating clinical development with global trials across multiple indications, non-small cell, small cell, and additional tumor types that have that B7-H3 expression, pancreatic, breast, HCC. We do believe that the data we've seen with Tam-Peli, both in terms of efficacy, tolerability, is very compelling. We have plans for a broad development program for that molecule. Stay tuned and you will definitely hear more. In terms of the breakdown for allergy and food allergy, we have not previously shared that.
I think we're seeing great growth actually, also in our chronic urticaria indication as well. Xolair overall, it continues to grow both in allergy and in food allergy.
I think maybe as a hint, I think since we launched food allergy, all the growth we have seen since then basically was food allergy.
Food allergy.
I think whereas the other indications were more flattish, but I think that's how you can calculate it back. Maybe one add on. I think, James, if you go to the Pharma Day presentation, I think there you have a full list of what we announced with respect to Tam-Peli. I think we have a phase III global study in small cell lung cancer to be initiated later this year in Q4. We have, I think then a global study in non-small cell lung cancer to initiate first half 2027. We are looking into prostate cancer. We are looking into breast cancer. I think this is already what we have communicated. Any other questions?
Just on Xolair, whether the biosimilars can attack the food allergy straightaway or if there's any other protections.
They can.
Oh, yeah, it fully depends on what label they receive. I think with allergy in particular, we would expect some stickiness for continuing patients. I think this is an area that will be interesting to watch as the biosimilars come into play, because again, you're dealing with allergic patients, and conservatism in that population may be higher.
Got it. Thank you.
Frankly, you're also getting, particularly in food allergy, a lot of young patients. So then you also have to convince parents to be comfortable with that as well.
And biosimilars do not have data in that setting.
Correct. They do not.
Very good. Then next questions go to Justin Smith from Bernstein. Justin, please.
Thank you, Bruno. Just a couple on the zilebesiran, if I can. KOLs we have spoken to have expressed a bit of concern about whether or not there would be an antidote. Then second one, just could you talk a little bit about the challenges of managing the placebo arm, given these patients are receiving frequent treatment so infrequently?
And I'm sorry, for what molecule?
zilebesiran.
Oh, zilebesiran. That makes more sense now. This is something that we actually looked at when we began to design the trials, but ultimately, when you looked at the effect that we were having, we didn't believe that it was necessary or was going to make sense. I think we're very comfortable in the efficacy and safety profile that we've seen from zilebesiran through very large phase II trial populations. I think we're very encouraged by what we've seen so far. Frankly, the rapidity with which these trials have enrolled, I think speak overarchingly to the excitement for something that has this profile, particularly given the difficulty in controlling these patients over time.
Thank you.
Okay. Next questions go to Peter Verdult from BNP Paribas. Peter.
Yeah, thanks, Bruno. Peter Verdult here, BNP Paribas. Two, please, one on outlook and one on obesity strategy. Just Thomas, I totally hear you on Roche not having a patent cliff to deal with per se, but I would like to kick the tires and gauge your early thoughts into next year and whether you would push back on the idea that top-line growth decelerates from mid to high single digits enjoyed in the last three years to something closer to low single digits.
I am saying that not to try and be difficult, but I get that giredestrant and fenebrutinib will be good launches, but you do have growth pressures or slowdowns on a number of products. I will not list them out. You know what they are. We have discussed many of them on the call. That just seems to be pointing to 2027 perhaps being a lower growth year.
I realize today is not an official guidance platform for next year, but I would appreciate some early thoughts given what you said in your opening remarks. Secondly, for Thomas or Teresa on obesity strategy, given we now see self-pay 50% of U.S. injectable obesity market, 90% of the oral market, would Roche have any interest in developing a consumer Rx model longer term? Will you build a portfolio around your obesity pipeline to include aesthetics and/or men's/women's wellness products? The question on outlook and obesity strategy. Thank you.
Yeah, so I can take the first question, and I can only say that we are confident that we can maintain a similar momentum that we have. What I also have said previously is that, in terms of margins, we always said that margins will be at least stable. What I can also commit to, even though that we will invest more in R&D, and this investment in R&D that we need on top of all of the efficiency gains that we are putting in place, on top of that, we need those investments for all of the clinical trials and all the exciting new molecules that we have. Despite that, we will expand margins next year.
In terms of the obesity strategy, I think we see the evolution of the market just as you do. Cash pay is certainly going to be a growing part of how monotherapy obesity molecules or medicines are paid for, and we would intend to compete appropriately.
Peter, did this help?
As always, yes. Thank you, Bruno.
Okay. Then next question's with Urban Fritzsche from ZKB. Urban, you are unmuted. Urban, you are unmuted. You're muted. You're still muted. We can't hear you.
Sorry. It was my fault. Okay.
It was. You're now unmuted. Urban? Okay, let's maybe quickly go on, and then we give it a second try later on. Next questions would go to Morgan Stanley, Thibaud Boutherin. Thibaud?
Yes. Thank you. Just a follow-up on Tam-Peli. Based on the data we've seen at WCLC, the efficacy looks competitive against other B7-H3, and potentially more favorable on the safety profile regarding the ILDs. Is this the key point of differentiation in your view, based on the data we've seen so far? Then the second question, just on Xolair for next year. Have you already made price consideration in the U.S. given the entrance of biosimilar? Do you expect to have a very meaningful price impact next year? In that context, can you give us an indication on the direction of revenue for Xolair next year, if you have any chance of maintaining this function or growing this function next year? Thank you.
Okay. For Tam-Peli, we believe that it's both safety and efficacy that are going to be differentiating for this molecule. I think we've seen it have what we believe to be best in class data on both. So I think we're excited about it from both ends. I'm sorry, I didn't quite hear your question. Xolair and guidance or outlook for next year? Is that-
Yeah. It was just the outlook for Xolair in 2027. If you already took a price cut, if you're expecting to take a meaningful price cut next year and-
Sure
just if you can give us an indication on the direction of sales for Xolair next year.
Sure. I think I mentioned previously that we would expect Xolair to be flattish next year.
I think, Thibaud, the first biosimilar has launched with a 15% price cut as you would expect. Nothing exceptional there.
Yeah.
I think we are always, as we have been before, not driving the prices downwards.
Yeah. We are not driving the race to the bottom.
Yeah. Did this answer your questions?
Yes. Thank you.
Yeah. Then we have a final analyst in the queue, which is Rajesh Kumar from HSBC. Rajesh, please.
Thank you very much for taking my question. Appreciate you expect to sustain current level of growth with operational gearing. Clearly, the Street seems to be getting you down to 4%-5% top-line growth and a bit of margin pressure. I can see that you have an expectation of flat Xolair versus Street modeling a 9%-10% decline in 2027. What are the other assumptions? Maybe efficiency on the cost side on AI or R&D phasing or trial phasing costs that we might be missing when we are modeling 2027, that we are getting a lower top-line growth and a weaker operational gearing?
Yeah. Let me just comment on that. I can just reiterate that we do believe we have good momentum, and this momentum will carry us not only through the end of the decade but into the next decade. Second, please also keep in mind the other revenue line. We have a number of income that we get from molecules that have been licensed out, that will give us a certain boost. That's why I made the comment that I made. Even though that we stay disciplined in our investment in R&D and M&A, we know that we also have to invest in the launches. We have to invest in all the trials. We know that we have a certain boost that will help us on the margin.
Okay
We have even told the markets that we will grow the other revenue line from 3% of sales to 5% of sales until the end of this decade, until 2030.
Just so that we are clear, are you making a specific assumption about orforglipron royalties that the market might be getting wrong? Or that's a bit more bullish than what the street is modeling?
There are multiple molecules that have been licensed out that will add to growth in that line. If you just look at the P&L and you assume whatever growth on the sales line, the reality is that any additional growth that will come in that will add to the margin. I think what I said is more or less independent of how exactly successful that will be. There is just a certain tailwind that we get through that.
Thank you very much.
Okay. Did this help, Rajesh?
Yeah. Perfect. Thank you.
Okay. Then I think we have one final question. Steve came back. Steve, the final question goes to you.
Oh, thank you so much. So there will be some important competitor BTK data at MS Toronto tomorrow. What specifics in that data would represent a clear win for fenebrutinib?
Do you want to answer or should I answer?
That's a good question because
Did you want to answer that? Go ahead.
I can answer that. I think that fenebrutinib put in quite a high bar in terms of efficacy, so we have not seen any data on the efficacy side of that other company. So we're going to be very curious to see how their efficacy data looks.
Okay. I think with that, we will close the call. Thomas, over to you.
Yeah. Thank you very much, Bruno, and a huge thanks to the team. As we mentioned, I think we have good sales growth this year and good sales momentum. The team and the organization is very disciplined. Also, we have good line of sight for the next couple of years. We have a number of NME launches. I think that Teresa also showed on her slides how many approvals we will get in the next months. This will already carry us to the end of the decade. On top, we have up to 20 NMEs that we could launch, and there are many more readouts that are going to happen. Overall, I think we've made significant progress on our pipeline.
Also when you look at phase III success rate, if you look at pipeline value, et cetera, it also shows that as an organization, we've been quite disciplined on cost. We have rearranged and shift money within our organization to make sure that we can fund a lot of these activities. Of course, we are very disciplined also on BD and also in the implementation of AI. Overall very determined, focused, and we'll continue to execute, and we will deliver.