Thank you very much, and good morning and good afternoon to everyone. I'm very much excited to share our Q1 2026 results with you.

Overall, very strong performance in sales in the first quarter, including some clinical and regulatory milestones that we have achieved. Q1 was strong with group sales at +6%, driven by Pharma with +7%.

Diagnostics grew +3%, despite headwinds from the China healthcare pricing reforms, which we are still experiencing in the first half year. Excluding China, diagnostics was at +5%.

In addition, a weaker and earlier flu season impacted the sales of Xofluza, and on the diagnostics side, respiratory sales in point of care, patient care and molecular — across both divisions, approximately CHF 170 million, which more or less explains the difference to consensus. On key milestones — Q1 was very busy with significant pipeline progress.

On the pharma side, we filed giredestrant in adjuvant ER-positive HER2-negative breast cancer in the U.S. using a priority review voucher, and we expect approval across two indications before the end of this year.

Zunovo was filed in the U.S. We've had a series of Gazyva filings in the U.S.

and EU across SLE and IgAN. On key readouts, we had positive FENhance 1 results in RMS — together with FENhance 2 and FENtrepid, this is the first and only BTK inhibitor with positive phase III across both RMS and PPMS, and we expect a filing in the next couple of months.

Gazyva has had four positive readouts in a row, the latest being MAJESTY in membranous nephropathy, following positive readouts in lupus nephritis, SLE, and IgAN. In obesity, we announced positive phase II results for petrelintide, showing excellent tolerability — in fact, more dropouts in placebo than the active arm — making it an ideal candidate to combine with our GLP-1/GIP CT-388, which showed at week 48 similar weight loss to a leading competitor at week 72.

On business development — we extended our collaboration with C4 Therapeutics on degrader-antibody conjugates, a promising novel platform that could overcome certain therapeutic window and resistance issues. On diagnostics, we acquired Saga Diagnostics, which provides a minimal residual disease test to monitor disease progression.

On the regulatory side, we received CE marks for new blood screening tests including hepatitis E, the Elecsys ApoE4 and Elecsys NfL for neurology. Looking ahead, we still have important phase III readouts this year and launches in diagnostics — divarasib for lung cancer, cefoxitin in IgAN, Itovebi in breast cancer, and Lunsumio in second-line follicular lymphoma.

In diagnostics, the Axelios sequencer is coming this summer. We'll also launch Elecsys pTau217, the Elecsys IGRA tuberculosis test, cobas HDV, and the ePlex gastrointestinal panel.

We have shown consistent strong growth — post-pandemic average growth of 8%, and even in 2023 and 2024 excluding pandemic sales, underlying growth of 8%. We will continue to drive this going forward.

Looking at numbers — group grew 7%, driven by Pharma at 7%. Diagnostics grew 3%, due to China healthcare pricing reforms — excluding China, Diagnostics was at +5%.

The flu season impact overall was about 1%, around CHF 170 million. By segment — in oncology, the HER2 franchise grew 2.5%, and we expect it to peak this year with a strong tail remaining.

Tecentriq returned to single-digit growth driven by new indications. Alecensa had an exceptionally strong Q1 with double-digit growth, with low single-digit growth expected for full year 2026.

In hematology, Polivy continues strong uptake in first-line DLBCL, Columvi and Lunsumio continue growing in later lines, and Hemlibra saw a very strong Q1 with market share gains. In neurology, Ocrevus keeps growing nicely with strong uptake of the subcutaneous formulation, and Evrysdi is expanding its global leadership in SMA.

In immunology, Xolair continues strong uptake in food allergy — we're expecting strong growth for full year 2026 despite the first biosimilar launch in H2. Gazyva is exciting following four positive phase III in a row.

In ophthalmology, we saw acceleration and continued market share gains. In diagnostics, strong growth in pathology and core lab despite the China pricing reforms.

On AI — we've had a long-standing collaboration with NVIDIA since 2016, specifically on DNA sequencing given the huge amount of data our sequencer produces. We've also made acquisitions like Prescient and entered strategic drug discovery collaborations.

With our recent agreement, we're expanding GPU compute infrastructure with the largest AI factory in the pharma industry. Earlier this month, we joined Anthropic, IBM, Meta and Microsoft to launch a collaborative licensing zone for AI foundational model development, called Shared AI License Foundations, or SALE — Roche Genentech is the only pharma company among the founding members.

We believe AI will have disruptive potential across our entire value chain — drug discovery, manufacturing, diagnostics, imaging. On outlook — we've made substantial progress across all therapeutic areas.

We just submitted giredestrant in adjuvant breast cancer using a priority review voucher, expecting a launch this year. We also filed Gazyva in SLE in the U.S.

On key readouts in Q1 — persevERA in first-line ER-positive breast cancer failed but with a numerical benefit, with results to be presented at ASCO. More than 70% of the giredestrant opportunity is in the adjuvant setting, a much larger population, and we have another trial ongoing in the first-line setting.

Fenebrutinib has the first and only BTK inhibitor with positive phase III across RMS and PPMS, and we will file in the coming months. Petrelintide has a clean, placebo-like safety profile with double-digit weight loss, making it a good combination candidate with CT-388, and the phase II combination study will be initiated.

Looking forward, we have six pivotal phase III studies reading out including two NMEs, inavolisib and cefoxitisob. Between 2027 and 2030, we have up to 19 NMEs that could launch.

On diagnostics — several billion-dollar-plus platform opportunities, including the sequencer with clear differentiation on speed, accuracy and cost. We're also launching tests including Elecsys NfL alongside multiple sclerosis treatments and pTau217 alongside Alzheimer's treatments — combining diagnostics and pharma.

Guidance is confirmed at mid-single-digit sales growth and high single-digit core EPS growth, with a further increase in dividend in Swiss francs. With that, I'll hand over to Alan.

Alan Hippe: Yeah. Thanks, Thomas.

When we look at sales — Q1 2026 was -5% in Swiss francs, with a CHF -1.6 billion currency impact from conversion. Shifting to the +6% in CER — pharma was +842 million excluding loss of exclusivity products, with loss of exclusivity at -63 million, getting to +7% in CER.

We still expect a loss of exclusivity impact on sales of roughly CHF 1 billion for the year — Actemra and MabThera helped us in Q1, but we think they'll slow down for the rest of the year, making the CHF 1 billion still realistic. On diagnostics — +148 excluding China, representing 5% in CER.

China healthcare pricing reform was CHF -60, giving us +3% in CER, with some additional effect from respiratory. On currency — the Swiss franc has appreciated against basically all currencies.

The CER cross was +6% and the Swiss franc figure was -4.7%. The major driver is the U.S.

dollar — converting to dollars, we would have grown 9%, supporting the constant currency growth. On outlook — we always assume currency rates remain stable until year-end, which is challenging given volatility.

If currency rates at end of March remain stable, the full-year impact would be -4 percentage points on sales and -6% on core operating profit and core EPS. Last year in Q1 we had a positive currency impact, so the volatility is enormous, and we have roughly 50% of sales in U.S.

dollars. Guidance is confirmed and we're well on track.

The loss of exclusivity impact of roughly CHF 1 billion is still expected for full year 2026. Let me hand over to Teresa.

Teresa Graham: Thank you, Alan. Pharma sales grew by 7% at constant exchange rates, reaching CHF 11.5 billion in Q1, with strong performance in the U.S., international and Japan — the latter two with double-digit growth.

The EU was impacted by pricing and one-off effects for specific products. Overall pharma volumes were up 17%.

Our top brands — Xolair, Phesgo, Hemlibra, Vabysmo, Ocrevus, and Polivy — generated roughly CHF 800 million in new sales. On Xofluza — we had a strong flu season in Q1 and Q4 of 2025 and anticipated a decrease in flu-related sales in China relative to last year, which materialized with an 86% decline fully attributable to a weaker flu season.

In oncology, sales increased 3% to CHF 3.7 billion. Phesgo continues strong growth with global conversion climbing to 55%, aiming for at least 60% at peak.

Kadcyla sees increasing competitive pressure in the U.S. and EU, in line with expectations, plus negative Q1 U.S.

impact from purchasing patterns. On giredestrant — the persevERA study in first-line read out negatively, though with a numerical improvement in PFS — results will be presented at ASCO including at a separate IR event.

We successfully submitted the U.S. filing for lidERA in adjuvant ER-positive HER2-negative breast cancer using a priority review voucher, enabling approval by year-end.

We previously filed evERA in post-CDK4/6 ER-positive HER2-negative metastatic breast cancer, with the PDUFA set for December 18th. Let me address the commercial opportunity context.

Hormone receptor positive HER2-negative breast cancer accounts for about 70% of breast cancer patients — HER2 accounts for only 15%. The total SERD opportunity across all lines of therapy is somewhere between $20 billion and $30 billion.

Adjuvant ER-positive breast cancer is by far the largest piece, with three times more drug-treated patients than first-line metastatic and around five years of treatment duration. The first-line metastatic setting only represented around 10% of the overall giredestrant opportunity.

With positive lidERA and evERA, we believe we've captured around 80% of the overall SERD opportunity — lidERA accounting for roughly 70% and evERA for another 10%. We have one additional phase III in first-line endocrine resistant patients, pionERA, reading out next year, representing 40% of the first-line population, where we expect 40% of patients to carry an ESR1 resistance mutation compared to just 5% in persevERA.

We see ourselves well positioned to capture a meaningful share of this market, with peak sales well north of CHF 3 billion potentially being our largest selling product. Alecensa had a very strong Q1, partly driven by buying patterns internationally.

We continue to expect low single-digit growth for the full year due to competitive pressure in the U.S. and EU.

Tecentriq growth is driven by new indications, particularly IMforte in small cell — additional indications like IMvigor011 and ATOMIC are expected to help carry growth forward. We saw a one-off buying pattern that negatively affected EU performance in Q1, but growth should gradually recover.

We confirm our full year 2026 outlook of low double-digit growth for Tecentriq. Hematology delivered strong growth of 18% at constant exchange rates, reaching CHF 2.2 billion.

Hemlibra continues strong global growth from increasing adoption in non-inhibitor patients, with U.S. performance benefiting from a base effect.

We expect low single-digit growth for the rest of 2026 due to anticipated headwinds from competitor launches later this year. We've initiated phase III trials Zebra 1 and 2 for NXT007 versus factor VIII and Hemlibra respectively, using an auto-injector from day one.

Polivy in first-line DLBCL continues steady growth, now at 39% U.S. patient share with more than 95,000 patients treated globally.

Gazyva delivered solid growth overall — in oncology, Gazyva sales in the U.S. and EU were negatively impacted by the launch of Venclexta and Acalabrutinib combination in first-line CLL increasing competitive pressure on the Venclexta/Gazyva regimen, which may be masking the immunology uptake.

Columvi and Lunsumio continue to see good uptake — we completed the U.S. filing of Lunsumio in second-line plus DLBCL, and expect the readout for the CELESTIMO study in second-line plus follicular later this year.

Neurology started the year strong with 10% growth at constant exchange rates reaching CHF 2.4 billion. Ocrevus continues good growth driven by the subcutaneous formulation Zunovo, with global patient count increasing to roughly 24,000 — up 7,000 patients from last quarter, almost 2,000 more than the increase in Q4.

Roughly 60% of Zunovo volume comes from community practices, and roughly half of patient starts are new to the Ocrevus brand. For 2026, we continue to expect high single-digit to low double-digit growth for Ocrevus overall, with peak sales expectations upgraded to CHF 9 billion by 2029 including CHF 2 billion from Ocrevus subcut.

We're pleased to report positive phase III FENhance 1 results in RMS, with full results for FENhance 1 and 2 presented at AAN a few days ago. On the rest of the neurology portfolio — Evrysdi off to a very strong start partially boosted by tender-related buying internationally, continuing to expand from the global rollout of the tablet formulation.

On ELAVIDYS in DMD — we continue to believe in the positive risk-benefit profile in the ambulatory population, and have decided to initiate a new phase III trial, ENCORE, with a placebo control arm focused on early ambulatory patients to enable regulatory resubmission in the EU and other regions. We had positive phase III Enspryng data in MOGAD presented at AAN.

On fenebrutinib — with the positive FENhance 1 readout, we now have three positive phase III trials for MS — FENhance 1 and 2 in RMS, and FENtrepid in PPMS. We're confident fenebrutinib has the potential to become the first and only high-efficacy oral treatment for both RMS and PPMS, and reaffirm peak sales potential of more than CHF 3 billion.

Fenebrutinib met the primary endpoint with significant relapse reduction — 51% in FENhance 1, 59% in FENhance 2, versus teriflunomide — translating to one relapse approximately every 17 years. Key secondary endpoints on disability progression also favored fenebrutinib.

On safety — the percentage of patients with elevated liver enzymes in FENhance 1 and 2 is lower than in FENtrepid and broadly similar to teriflunomide. There was one Hy's Law case in each of the fenebrutinib and teriflunomide arms of FENhance 1, both asymptomatic and resolved after discontinuation, with no additional cases across other fenebrutinib trials after implementing liver monitoring every two weeks during the first 20 weeks.

We did see an imbalance in fatal AEs across the fenebrutinib arms of FENhance 1 and 2 and FENtrepid, with various causes and time points, presented and discussed at AAN. Patient safety is our number one priority — there has been continuous exchange with the FDA and IDMC, and the totality of data will be submitted to regulatory authorities in the coming months.

On Enspryng in MOGAD — a rare demyelinating autoimmune disease with no currently approved therapies. IL-6 signaling is implicated in its pathophysiology, making Enspryng a logical fit.

We believe MOGAD represents at least CHF 500 million peak sales opportunity. In the METEOROID study presented a few days ago, Enspryng achieved a 68% risk reduction versus placebo for the primary endpoint of relapse-free patients, with response as early as eight weeks, significant reduction in annualized relapse rates, active MRI lesions and rescue therapy use, and safety comparable to placebo.

We expect to complete filing in the U.S. and EU for this indication later this year.

In immunology, the franchise grew 8% at constant exchange rates to CHF 1.5 billion. Xolair continues exceptionally strong uptake in food allergy and CSU — we expect around 20% growth for 2026 including the expected impact of the first biosimilar entering in H2.

Actemra sales declined 4% driven by biosimilar impact. On Gazyva's lupus nephritis launch — across early launch countries in the U.S.

and EU, we see positive feedback from doctors and patients on awareness, intent to treat, and treatment satisfaction. We remain confident Gazyva represents up to a CHF 2 billion opportunity across all immunology indications.

We also have the upcoming cerfactronin phase III readout in IgAN expected later this year. On Gazyva more broadly — four out of four positive phase III trials in immunology since LN results in 2024, plus the positive ALLEGORY results in SLE just presented at SLEuro.

LN is already approved with U.S. and EU launches ongoing.

SLE has been filed in the U.S. and EU, with U.S.

PDUFA set for December 4th. MN will be filed in the U.S.

and EU later this year. IgAN has been filed in the U.S., with EU filing expected later this year.

In ophthalmology, growth was 10% reaching $1.1 billion. Vabysmo started the year strong with 13% growth at constant exchange rates — a return to U.S.

growth at 4% with low double-digit volume growth and steady market share expansion. Roughly 60% of Vabysmo starts are treatment naive.

We're now seeing first signs of recovery in the branded market contraction we saw in the U.S. We received an updated FDA label for the RVO indication adding flexibility for treatment beyond six months.

In Q1 EU, we digested a negative price impact but saw good volume growth, expecting a return to growth in coming quarters. For 2026, we expect Vabysmo to accelerate compared to the 2025 growth rate of 12%.

We expect to file two new potential medicines in ophthalmology later this year, Varmecibart in UME and Enspryng in thyroid eye disease. On CVRM — together with our partner Zealand, we shared positive results of the phase II ZUPREME-1 study of petrelintide in obese patients without type 2, showing meaningful double-digit weight loss with placebo-like tolerability.

On insepotide, formerly CT-388, the positive phase II results for week 48 once-weekly insepotide showed a placebo-adjusted weight loss of 22.5% using the efficacy estimand, with a clear dose response relationship and no visible efficacy plateau at 48 weeks for the 24mg dose. We've successfully initiated two phase III trials, ENITH 1 and 2, in Q1.

For both petrelintide and carmotide, we'll present results at ADA in June including a separate IR event. For the rest of 2026 we expect additional phase II readouts for carmotide in obesity with type 2 and CT-996, our oral GLP-1, plus phase II data for emugrobart tirzepatide.

We plan to initiate a phase II combination study for insepotide and petrelintide with first patient in mid-year, and the first phase III studies for petrelintide plus CT-996. On news flow — earlier this quarter we decided to discontinue development of emugrobart in SMA and FSHD following assessment of phase II data from MANATEE and MANOEUVRE.

This has no impact on the ongoing phase II obesity trial reading out towards year-end. On our consensus gap analysis — based on post full year 2025 consensus, Roche has a biosimilar gap of CHF 6.7 billion by 2030.

Consensus expects a total of CHF 15.9 billion in additional sales by 2030 from the on-market portfolio and phase III pipeline — CHF 7.5 billion from on-market portfolio growth led by Vabysmo and Ocrevus, and CHF 8.4 billion from late-stage phase III pipeline with giredestrant, fenebrutinib and petrelintide as big growth drivers. There are still other pipeline assets not yet covered by consensus, including Enspryng in MOGAD, Enspryng in TED, and Gazyva in SLE — presenting potential upside.

Now over to Matt. Matt Sause: Congratulations, Teresa.

Good morning, good afternoon, everyone. With sales of CHF 3.3 billion, the diagnostics division grew 3% or CHF +88 million compared with 2025 at constant exchange, achieved despite the ongoing China healthcare price reform impact.

Excluding China, growth was +5%, additionally impacted by approximately CHF 40 million lower sales due to the weak Northern Hemisphere respiratory season. By customer area — Core Lab increased 4%, or +8% excluding the China policy impact.

Molecular lab was flat, with strong transplant growth at +15% offset by the mild respiratory season lowering testing volumes. Blood screening decreased -9% due to the ongoing conflict in the Middle East impacting some customer deliveries.

Near patient care decreased -5% driven by lower Cobas Liat sales from the mild respiratory season, partially offset by blood glucose monitoring growth of +8% from recent competitive wins. Pathology lab grew +12%, mainly driven by advanced staining growth of +9% and companion diagnostics growth of +23%.

By geography — North America grew +6%, EMEA grew +3% (impacted by the blood screening decrease from the Middle East conflict), LATAM grew +10%, and APAC declined -5% — China sales declined -14% due to the healthcare pricing reform. We expect a lessened impact of the China pricing reform in 2026, and our ambition is mid-single-digit growth for the diagnostics division this year.

On assay pipeline updates — we received CE mark for the cobas MPX test in March, addressing the approximately CHF 800 million global blood screening nucleic acid testing market and growing demand for hepatitis E screening, enabling simultaneous detection of HIV, HCV, HBV and HEV from a single sample. This strengthens our position as the only company with a full nucleic acid and serology testing solution.

We also received CE mark for our blood-based neurofilament light chain assay, supporting more informed treatment decisions for relapsing remitting MS, with age-specific NfL percentiles from a comprehensive reference data set — a differentiated approach versus other approved tests relying on a single cutoff. On pTau217 — data presented at ADPD in March across primary, secondary and tertiary care settings with 675 participants showed high diagnostic accuracy consistently across all clinical stages, positioning it as the first pTau217 test with potential approval across primary and secondary settings.

On the Exelios sequencing solution — data presented at AGBT in February from a large internal study of over 1,000 human genomes processed over 65 runs demonstrated reliably generating an average of 2.3 trillion bases of concordant duplex bases in four hours. Data from the Hartwig Medical Foundation comparing our SBX technology with a leading on-market technology for 118 tumor normal pairs demonstrated high throughput, improved batch flexibility, and 99% concordance with the leading technology.

On our Saga Diagnostics merger agreement — Pathlight's tumor-informed ultrasensitive MRD platform, currently available for early breast cancer and colorectal cancer, will be fully integrated into our Foundation Medicine organization, and we plan to leverage Exelios and our Roche Digital LightCycler PCR platform to develop a decentralized MRD solution. The MRD market is projected to grow at 31% CAGR, one of the fastest-growing diagnostics segments.

On our key launch list for diagnostics — of 11 launches, we achieved three by Q1, with good progress on the others which are on track. Now over to Bruno.

Bruno Eschli: Thank you, Matt. Upcoming IR events — Diagnostics Day on May 12th, a live event in London with Thomas opening the day with a group and strategy update, with sequencing in the center.

We have an ASCO event on June 2nd live from Chicago focused on giredestrant data and clinical future development. On June 8th we'll have an event around ADA focused on the obesity franchise, with key data on petrelintide and an update on combination development steps.

Pharma Day is scheduled for September 28th, a live event in London, including an update on our AI investments focusing on early drug development. With that, we open the Q&A.

James Gordon: First, on giredestrant — is the failure of persevERA surprising in light of lidERA? Could it be due to background CDK4/6 therapies, or prior patient experience with hormonal therapies?

Second, on obesity pricing — given low-cost GLP-1 generics already launching in places like India, how do you think about obesity pricing next decade, and could that be tougher for a drug like petrelintide? Teresa Graham: On persevERA versus lidERA — context and biology of the line of therapy really matter a lot in oncology trials.

Tumor burden is low in early breast cancer versus increasing complexity and resistance in the metastatic setting. Though persevERA didn't reach statistical significance in PFS, a numerical improvement was observed, and we believe giredestrant is active in first-line metastatic.

persevERA covered 60% of the first-line setting, and pionERA covers the remaining 40% — we're curious to see what happens there. Nothing about persevERA changes our confidence in giredestrant's importance in early breast cancer, where 70% of the opportunity sits.

On obesity pricing — this was very foreseeable; when we entered this market we assumed significant price erosion by the time we got there. Given our portfolio, clinical benefit and differentiation, we believe there continues to be a robust opportunity in obesity.

Sachin Jain: On lidERA — how should we think about launch cadence as you prepare for commercial launch? Will it need a lot of community oncologist education versus CDK4/6 AI?

Second, you've said this could be one of Roche's top drugs, perhaps in the $5 billion to $10 billion range — what factors dictate where in that range you end up, including adoption in low-risk patients, combination use with CDK4/6, and impact from other adjuvant studies like CAMBRI1 and BEACON? Third, on BTK — can you give color on the tone of FDA/IDMC discussions, particularly regarding the death imbalance and suicide-related concerns?

Teresa Graham: We expect giredestrant to have a strong launch, with quickest uptake in mid-risk population, then moving into high-risk population given difficulty tolerating CDK4/6 over time. Low-risk patients also show KOL intent to use.

We intend to resource and go after this strong out of the gate. On switch studies — they assume a switch after 3-5 years of therapy, which doesn't fully compute as the optimal approach.

With lidERA, we asked a clear scientific question and got a clear answer well-received by the KOL community. On BTK — we can't discuss ongoing regulator communications, but we've been transparent and the discussions have been productive.

On suicidality — the MS population unfortunately has a higher baseline risk of suicide; we didn't see any suicide or suicidality in our phase II trial including the long-term extension, nor in autoimmune indications studied. We're looking into this very closely and patient safety remains our number one priority.

Thomas Schinecker: On giredestrant — it has demonstrated the highest preclinical potency among SERDs. Switching studies involve a very different setting and don't necessarily translate, and the next true adjuvant switching study is years away.

On BTK — a good indicator is that on the PPMS side, all trial patients including those on placebo have been approved to move into the open label extension on fenebrutinib, which shows the benefit is seen. Peter Verdult: On obesity — given your R&D strategy of best or first-in-class assets, I understand carmotide, but petrelintide doesn't seem to be best-in-class as an amylin analog.

Is there internal debate about the roughly $2 billion commitment to a full petrelintide phase III program? Teresa Graham: The amylin class is fundamentally different from GLP-1 and serves a different role in what we see as a heterogeneous obesity market.

Petrelintide has a very different tolerability profile, important for the portion of the population that won't tolerate adverse events, who may need to lose only low teens percent of body weight rather than 20%. We believe there's a role for amylin and for petrelintide as part of a differentiated, holistic portfolio.

Thomas Schinecker: We've announced a phase II combination of petrelintide and CT-388 — for combinations, it's important to have tolerable molecules, and we see real potential between them. Sarita Kapila: Ahead of the Q4 data in DME, how should we think about competition to Vabysmo from Merck's MK-3000, given early data suggesting similar visual acuity improvement at 12 weeks?

Teresa Graham: Vabysmo has established itself as the standard of care in retinal disease, consistently reinforced by our retinal specialists. While another mechanism entering the space is interesting, they'll need to demonstrate they can cross the high bar Vabysmo has set.

We also have our own assets in this race, but people will need to see something better than Vabysmo, and as of yet, nothing in phase III has proven that. Luisa Hector: Has the de-risking of legacy pipeline assets like fenebrutinib and giredestrant changed your ambitions around business development, R&D investment or longer-term sales and margin?

Teresa Graham: Both sit in end-to-end disease areas — MS and breast cancer — where we have long-term strategic interest in leadership. It doesn't change our interest; we remain deeply interested in both and will continue hunting for opportunities.

Thomas Schinecker: From an M&A perspective, we'll continue M&A like in recent years but we're not dependent on it — we've been able to finance recent M&A from earnings in that year, unlike some other companies that must do M&A. We've had very disciplined spending and continued growth even absent M&A, without a patent cliff.

Graham Parry: First, on giredestrant — 70% of a $20 billion to $30 billion opportunity in adjuvant implies $14 billion to $20 billion. How much of that assumes CDK4/6 combo use or erosion versus essentially today's aromatase inhibitor monotherapy market aligned with lidERA?

Second, could you get a label allowing aromatase switches at launch, or just de novo patients per lidERA? Third, for Alan — thoughts on Roche's margin trajectory with this multi-billion-dollar in-house small molecule asset driving revenue growth?

Teresa Graham: We won't comment on label negotiations with the FDA. About 20-25% of early breast cancer patients currently get CDK4/6 combination therapy, with very high discontinuation rates due to adverse events.

We don't provide peak estimates at this stage, but we expect a significant share of that market. Alan Hippe: On margin — we defend or increase the margin moving forward, as we've said for this year.

We have higher burden from tech and financing to cover, but we're going to do that. Looking at the longer term, we've been pretty solid on gross margin, especially on the pharma side, roughly 80%, regardless of whether it's small molecules or other mechanisms.

Bruno Eschli: It's worth noting we're well-entrenched commercially in these areas — we have a big breast cancer sales force and a multiple sclerosis sales force, so these launches land right in our sweet spot from an S&M perspective. James Quigley: First, on Vabysmo — what's the current bottleneck for U.S.

acceleration despite increased funding toward late last year? Has that started lifting through the quarter?

Second, on CT-388 — to what extent is there flexible dosing in the new phase III trials, and was semaglutide considered as a reference or active comparator? What's the progress on combination development with petrelintide?

Teresa Graham: On Vabysmo — patient assistance foundations are not part of our commercial strategy and we don't factor them into sales planning. What we're hearing is physicians have gotten savvier about where to use patient assistance dollars, directing them to patients who truly need them — that's separate from anything on our commercial side.

On CT-388 — the first two trials are standard against placebo because that's currently required for regulatory approval of a new molecule in this space. We'll look at other things going forward and more to come on plans, including the fixed dose combination.

Bruno Eschli: On the combination development — first patient in is expected mid-year, and we'll go into more trial design detail at the upcoming ADA IR event. Simon Baker: First, can you share physician feedback following the AAN presentations of fenebrutinib, given some recent drugs have seen a disconnect between negative market and positive physician impressions?

Second, on the AI factory — when and how will we be able to assess the impact of your significant AI investments? Teresa Graham: KOL feedback at AAN was extraordinarily positive — people are encouraged by the data and reassured by the safety profile, particularly the relative understandability of liver enzymes and lack of a pattern with other adverse events.

45% of MS patients today are on orals or low-efficacy therapies — extremely low-hanging fruit for fenebrutinib acting on both progressive and relapsing disease. Thomas Schinecker: On AI, you'll get an update at Pharma Day with details on how many pipeline molecules came through AI tools and how we'll track that going forward.

Our teams are now utilizing AI constantly in research. Bruno Eschli: Stay tuned for Pharma Day with Aviv presenting our AI strategy and tracking approach.

It's early days — basically all new NMEs now have an AI component, and these molecules are entering the clinic in the next couple of quarters, but it will take one to three years to collect enough data on impact to success rates or new target identification. Justin Smith: First, on Alzheimer's — could the pTau217 test launch expand the biologics market?

Second, on Polivy — given upcoming competition, are the sequential U.S. share gains sustainable?

Matt Sause: On pTau217 — one of the biggest challenges to Alzheimer's diagnosis is specialist availability, often taking over two years. A primary care claim with a blood-based biomarker has the potential to greatly expand access, if it achieves 90% negative and positive predictive value with 90% sensitivity and specificity for guideline inclusion.

Teresa Graham: From the pharma standpoint, we're extremely excited to have these assays in the market, as proper diagnosis has been a limiting factor for uptake. On Polivy — yes, share gains are sustainable, and we'll be releasing additional combination data.

Thomas Schinecker: We also have our own first-line development program ongoing with STARGLO. Recent study starts have basically all built on the Polivy regimen, showing it's been widely perceived as standard of care first line — a roughly CHF 2 billion opportunity we're tracking nicely towards.

Richard Vosser: First, on Hemlibra — you mentioned strong growth and switches back from ALTUVIIIO; how sustainable is that, and how will you defend against a competitor launch later this year? Second, on Gazyva — what's the drag from CLL competition on the rest of the year as we think about the lupus rollout?

Teresa Graham: We're seeing patients well-controlled on Hemlibra try ALTUVIIIO and then switch back — around 30% of those who switched have come back, with no slowdown in that trend. ALTUVIIIO's growth is really coming from other factor VIII products.

We expect increased competition but are improving the patient experience with a new injection kit and auto-injector in development, and we're the only company brave enough to go head to head with Hemlibra via NXT007. On Gazyva — the oncology drag is probably too soon to tell precisely; it's tamping down what we see as initial uptake in immunology, but it's early days since launch in Q4.

Bruno Eschli: One more comment on Hemlibra — it doesn't cause inhibitors, which has also been driving the switch back. Matthew Weston: First, on giredestrant — what proportion of ESR1 mutant patients already know their status, given the importance for the late-line population?

Second, given MFN pricing pressure, how are discussions with European governments going about accepting higher prices, and will giredestrant roll out rapidly ex-U.S.? Teresa Graham: Testing in breast cancer is quite high, particularly in mature markets, so most patients would know their status, supporting a strong launch.

Thomas Schinecker: On U.S. pricing — we'll be on the market towards the end of the year, when we set the price.

We're already in discussions with European politicians, who understand pricing will change for new introductions — it's not a steep one-step increase since already-marketed medicines aren't repriced. There will likely be more differentiation between medicines with strong data and those without, and giredestrant has very strong data plus engaged patient groups, putting us in a beneficial position to get appropriate pricing in Europe too.

Steve Scala: First, why does Roche still have confidence in pionERA after persevERA's failure? Second, why isn't Roche running a cardiovascular risk reduction study of its IL-6 assets like competitors?

Third, Thomas, why are you opening Diagnostics Day this year when you haven't before? Teresa Graham: persevERA showed a numerical improvement despite missing statistical significance, and we believe giredestrant is active in first-line.

pionERA is studied in a different, ESR1-mutation-enriched population — 40% versus 20-25% in the general first-line setting — designed specifically to uncover that signal. On IL-6 cardiovascular — don't assume we haven't looked.

Thomas Schinecker: On Diagnostics Day — I previously opened it because I was CEO of Diagnostics; this is the first time I'll open it without holding that role. Bruno and I agreed I'd alternate between Pharma Day and Diagnostics Day going forward, giving group updates each time.

Bruno Eschli: On cardiovascular — we're focusing on other modalities too, such as NLRP3 as one of our key targets, and we've also looked into IL-6 in that context. Michael Leuchten: First, on the FDA's increased disclosure triggering petitions for label changes, including for Ocrevus — can you talk through the process and timeline, and whether this is likely to happen more broadly?

Second, clarifying Vabysmo U.S. "return to growth" — sequential revenues were 3% in Q1 versus 7% in Q4 — what variable gives you confidence in that?

Teresa Graham: On Vabysmo — we saw a 4% U.S. increase but a more significant volume increase, in a highly contracted market where numbers can look different.

We're also seeing unbranded Avastin use trickle down, supporting confidence the branded market is coming back. On the FDA letter — I assume you're referring to the British Medical Journal letter on PPMS.

Ocrevus in PPMS has been used in 450,000 patients with 1.4 million patient years of exposure, approved in 130 countries — an approval not being questioned by KOLs, patients or the FDA. This was reinforced by the OHAN study released in September last year, not mentioned in that letter.

We have incredible confidence in Ocrevus's benefit and safety in PPMS. Bruno Eschli: On Vabysmo share — if we zoom into the branded segment excluding biosimilars, we keep gaining share in all indications globally, with a very constant run rate of share gains quarter-over-quarter.

Teresa Graham: Every time we poll KOLs, Vabysmo is seen hands down as the most efficacious drug in their armamentarium. Rajesh Kumar: First, on oncology pipeline and capital allocation — your oncology portfolio looks light on pipeline; how are you thinking about augmenting it?

Second, on obesity trials — given patient churn issues once patients realize they're on placebo, would you consider active control trials for registration purposes? Third, on AI — where in the financials will we see the benefit — R&D intensity, SG&A intensity, over the next 3-5 years?

Thomas Schinecker: On capital allocation — we look across our five therapeutic areas for opportunities that fit scientifically as potentially best or first-in-class, combined with appropriate financials so we don't overpay. We continue evolving our oncology pipeline through CDKs and ADCs while remaining disciplined, because there are always more opportunities than money — overpaying for one deal means missing a better opportunity elsewhere for less.

We assess everything that crosses the table. On AI — we believe it will impact us across all cost lines, already used in manufacturing and reshaping processes that used to take months down to weeks or days.

The biggest disruptive opportunity is in changing how fast we do drug development and how many molecules we can screen virtually, feeding data back in a "Lab-in-the-Loop" approach using proprietary models. We've trained 100,000 people on "Everyday AI" tools with strong productivity feedback.

Alan Hippe: We're investing heavily in data infrastructure with a huge ERP program, facilitating AI use moving forward by having all data centralized in the cloud. Teresa Graham: On obesity trial design — placebo-controlled trials are still mandated by the FDA for regulatory approval, which is why we run them.

We have a broad suite of patient retention tools and strategies for the CT-388 studies, but for competitive reasons won't elaborate further. Alan Hippe: On the third sub-question about whether our next M&A deal will be in oncology — honestly, I don't know.

It takes two to tango, and we don't know who our next partner will be. Thomas Schinecker: We're looking across all therapeutic areas.

Bruno Eschli: With that, we close the Q&A session. I hand back to Thomas for final words.

Thomas Schinecker: Thank you very much for attending today's call. We have continued strong sales momentum, and recent readouts give us much more confidence in the short to midterm growth outlook.

We have the opportunity to launch up to 19 NMEs between 2027 and 2030. We continue to work on M&A to increase that number, and many of these assets carry significant value, giving us confidence in the longer-term growth outlook of the company.

All I can say is you can count on us, and we will deliver.