Two-Fifths of the Beat Was Legal Settlements and the Camlipixant Bar Moved: We Take the Other Side at 10x

Results vs. Consensus

GSK reports in Sterling on two bases: Total results under IFRS, and Core results, a non-IFRS measure that strips intangible amortisation and impairment, major restructuring, transaction-related items, and divestments and significant legal charges. The company guides on Core only. Its European coverage is written against Core in pence; the US ADR is covered in dollars per ADS, where one ADS equals two ordinary shares. Both consensus sets are shown below because they say the same thing in different units.

Q1 2026 Scorecard

The two consensus bases reconcile exactly, which is worth stating because at least two aggregators did not manage it. Turnover of £7,629m translated at the Q1 average rate of $1.35 is $10,299m, matching the $10.30bn actual on the ADS basis. Core EPS of 46.5p doubled for the ADS ratio is 93.0p, or $1.256 at the same rate, which rounds to the $1.26 quoted against the $1.16 estimate. Any write-up presenting a 46% EPS miss for this quarter has compared a per-ordinary-share figure with a per-ADS estimate. There was no EPS miss.

Year-Over-Year Comparison (Core basis, £m)

On the Total results, turnover of £7,629m produced gross profit of £5,754m against £5,579m, operating profit of £2,293m against £2,216m, and EPS of 43.2p against 39.7p. The gap between Total EPS growth of +15% CER and Core EPS growth of +9% CER is the £420m disposal gain on the Rockville manufacturing facility, partly offset by a £265m remeasurement charge on contingent consideration liabilities and the Pfizer put option, plus a Total tax rate of 14.2% against 15.9%. Total results flattered the quarter more than Core did, which is the reverse of the usual pattern and is worth remembering when the FY comparison comes around.

Quarter-Over-Quarter Comparison

The sequential comparison is seasonally distorted and should not be read as momentum. GSK's Q4 carries the year's heaviest SG&A and R&D true-ups against a turnover base that is high but not high enough to absorb them, which is why Q4 2025 printed a 19.0% Core operating margin against Q1 2026's 34.7%. The useful sequential observation is the opposite of the headline: revenue fell 11.5% and Core operating profit rose 62.2%, which is a statement about Q4's cost load rather than Q1's operating leverage. Cash generation follows the same shape, and management reiterated that the path to more than £10bn of cash generated from operations for the year is weighted, as normal, to the second half.

Revenue assessment. Five percent constant-currency growth sits at the top of the company's own +3-5% full-year range, and the mix underneath it is better than the headline. At actual rates, Specialty Medicines added £293m of turnover year on year while General Medicines shed £234m and Vaccines added £54m, netting to the group's £113m increase. Specialty alone therefore more than covered the entire General Medicines decline. That is the transition working. What it does not do is accelerate: the same arithmetic that makes Specialty's 42% share powerful today makes the remaining 58% a drag that shrinks slowly. Group growth is a mix-shift function, and the mix shifts about three points a year.

Margin assessment. The gross margin line deserves more credit than it will get this week. A 110bp constant-currency improvement on a base of 77% is a lot of absolute profit, and its drivers are Specialty mix and Shingrix regional mix, both of which are durable. The operating margin line deserves less credit than it got. Core SG&A at 26.0% of sales versus 27.4% looks like a productivity story and is roughly half a legal settlement story. Management guides SG&A to grow at a low single-digit percentage for the year, and it grew 2% underlying in Q1, so the guide is intact. But investors who marked Q1's 34.7% Core operating margin as the new run-rate will be disappointed in Q2.

EPS assessment. Below the operating line, this quarter got harder, and it is going to keep getting harder. The finance expense grew 45% at constant currency because net debt rose £1,160m in the quarter alone on the RAPT acquisition, dividends and buyback, against an exchange loss of £154m. The tax rate is guided up. The buyback offset expires at the half year. That leaves FY Core EPS growth of 7-9% CER dependent on the operating line delivering at the top of its own 7-9% range, with no help underneath. It is achievable and it is not comfortable, and the market spent Wednesday afternoon working that out.

Segment Performance

Product Group Turnover

Regional Turnover

The regional split is the cleanest read on where the quarter's fragility sits. Europe carried the group at +14% CER on a +33% Vaccines quarter, which is a Shingrix national-immunisation-programme effect that management has explicitly told us annualises from Q2. International fell 6% CER on a 17% Vaccines decline driven by channel destocking of Shingrix at the China co-promotion partner. The US, GSK's largest region at 49% of turnover, grew 6% CER with Specialty at +16% offsetting Vaccines at -2% and General Medicines at -6%. Take Europe's Vaccines line out of the group and the growth picture is considerably more pedestrian.

Individual Product Detail

Specialty Medicines: £3,226m, +14% CER

Specialty crossed 42% of group turnover this quarter, up from 39% a year ago, and every one of its three legs grew double digits at constant currency. HIV at £1,824m and +10% CER is the largest and the most strategically loaded: the release states that long-acting injectables contributed 73% of HIV growth and that US HIV sales grew 15% with long-acting injectables at 34% of total US HIV sales. Cabenuva alone contributed more than half the franchise's growth, and Apretude more than a fifth, in what the company acknowledges is "an increasingly competitive US long-acting PrEP market."

"Competitive execution is powering our portfolio transition to INSTI-led long-acting regimens, which consistently represent more than 70% of our total HIV growth and more than 1/3 of total U.S. sales."
— Deborah Waterhouse, CEO, ViiV Healthcare

Oncology's £512m at +28% CER is the fastest line in the company and the smallest, at 6.7% of turnover. Jemperli at £232m and +40% CER carried it, with Ojjaara/Omjjara at +34% and Blenrep contributing its first meaningful £23m following the US launch and second-line approvals in 19 markets. Working against it, Zejula fell 11% CER on US prior-authorisation requirements stemming from the June 2025 FDA labelling update, which is a permanent impairment of that asset rather than a phasing issue.

Respiratory, Immunology & Inflammation at £890m and +16% CER is where the launch execution question sits. Nucala grew 12% CER with COPD driving roughly half the US volume growth per management, while Exdensur, the ultra-long-acting biologic that is supposed to succeed it, delivered £11m of what the release describes as "Early commercial introductions with new patient starts in the US and channel launch inventories in Japan and Germany." The company has already redirected the Nucala sales force away from every indication other than COPD to support Exdensur, which means the transition risk is now taken and the revenue is not yet there.

Assessment: Specialty is doing exactly what the equity story requires, and it is doing it with a quality of growth (volume, share gain, new launches) that survives scrutiny. The watch item is not the growth rate but the concentration: a franchise where one product family carries 73% of the growth of the largest leg, into a loss of exclusivity that management has now dated more precisely than before, is a franchise with a single point of failure.

Vaccines: £2,149m, +4% CER

Shingrix set a quarterly record at £1,026m and +20% CER, which is 48% of the entire Vaccines segment and the reason the segment grew at all. The composition matters more than the level. Europe grew 51% on national immunisation programme uptake and private demand; the US grew 12% "driven by inventory movements, including the launch of the new prefilled syringe." Management was pressed on the stocking question and answered it directly, describing wholesaler inventory as "pretty steady" and "very much in the typical range," and putting retail inventory at 2.4 million doses against 1.7 million a year earlier. A 41% build in the retail channel on a product whose US sales grew 12% is a real stocking contribution, volunteered rather than deflected, and it is still a pull-forward.

"Moving forward, this year, we expect tougher comparators for Europe and Japan as most large immunization programs annualize. Further penetration opportunities remain with around 11% of the eligible population immunized in our top 10 markets outside the U.S."
— Nina Mojas

Everything else in Vaccines went backwards. Meningitis fell 3% CER on Menveo delivery timing in International, partly offset by Bexsero in Europe and early Penmenvy uptake. Arexvy fell 18% CER on low out-of-season uptake and slower US market demand. Other Paediatric & Adult Vaccines, renamed this quarter from Established Vaccines, fell 9% CER on Synflorix competition in emerging markets and lower Hepatitis, Boostrix and Infanrix/Pediarix sales, partly offset by a bulk AS03 adjuvant sale. The US cumulative Shingrix immunisation rate reached 45%, up 4 points in twelve months, against roughly 11% average penetration in the ten priority ex-US markets.

Assessment: Vaccines is a one-product segment with a comp wall arriving in eight weeks. Full-year guidance for Vaccines is a low-single-digit decline to stable, and Q1 delivered +4%; the guide implies the remaining three quarters are negative. Management said so plainly, which is to their credit and does not make it less of a problem. The 11% ex-US penetration figure is the long-run bull case for Shingrix, and it is a five-year argument, not a 2026 one.

General Medicines: £2,254m, -6% CER

The declining base did what declining bases do, and slightly worse. Respiratory fell 4% CER with Trelegy, the segment's anchor at £646m, only stable at constant currency and down 4% at actual rates. The US was the problem: management attributed it to "phasing of sales volumes adversely impacted by Medicare benefit design changes and pricing unfavourability from channel mix pressures and adjustments," with Europe and International delivering the strong volume growth that offset it. Other General Medicines fell 12% CER on generic competition, lower contract manufacturing sales and phasing.

"Trelegy performance did not offset the broader portfolio decline as its growth in the U.S. was limited by increasing co-pay requirements due to Medicare redesign. These are especially pronounced in the first quarter and are expected to be less relevant in the rest of the year."
— Nina Mojas

Assessment: The Medicare Part D redesign co-pay dynamic is a genuine Q1 seasonal distortion and management is right that it eases. What does not ease is that Trelegy grew 15% CER in Q1 2025 and was flat in Q1 2026, on a US base facing structural pricing pressure. Full-year General Medicines guidance of a low-single-digit decline to stable, against a Q1 of -6%, requires a visible recovery, and management explicitly told us that recovery is second-half weighted with a tougher Q2 comparator ahead on Trelegy prior-year true-ups. This segment is where the full-year turnover guide is most likely to be tested.

Key Topics & Management Commentary

Overall Management Tone: Management was confident and specific on commercial execution, and noticeably more guarded on the pipeline than the prepared remarks implied. The prepared section read as a company that has found its rhythm under a new chief executive: dense on launch metrics, quantified on inventory, willing to name what is not working. Q&A turned harder, and the pattern of the answers was consistent: forward pipeline questions were met with process descriptions and deferrals to the July event rather than numbers, and on two occasions management volunteered a lower success bar for a pending readout than the Street was carrying. That combination, high confidence on the base business and hedged specificity on the pipeline, is what moved the stock during the call rather than on the release.

1. A Beat That Half the Street Could Not Bank

The single most consequential disclosure of the quarter was not in the release headline but in the CFO's prepared remarks, where the contribution of legal settlements to operating profit growth was quantified for the first time. The release itself frames the SG&A effect as "net favourability on legal settlements and expenses equivalent to around 4ppts impact in the quarter," and the CFO converted that into an operating-profit figure on the call.

"SG&A declined 2%, helped by positive IP settlements. On an underlying basis, SG&A grew 2%, demonstrating P&L leverage and continued productivity improvements. ... And operating profit grew 10% in the quarter, including the legal settlements, which were worth 3 percentage points."
— Julie Brown, Chief Financial Officer

Three of ten points is not half of the growth, but against a consensus that expected Core operating profit of £2.46bn, three points of £2,533m is roughly £76m, against a beat of £190m. It is therefore about 40% of the beat, and the Street's shorthand of roughly half is close enough to be directionally fair. What management did not do is say whether more settlement income is expected in the remaining quarters, which leaves modellers to assume none.

Assessment: The disclosure was voluntary, specific and immediate, which is the opposite of hiding it. But it converts an apparent 7-8% operating beat into an underlying 7% growth quarter, which is the bottom of the full-year range at the point in the year when comparators are easiest. That is the honest read, and it is why a beat printed as a sell-off.

2. Specialty Crosses 42% and the Mix Shift Becomes the Model

The structural argument for GSK is arithmetic. Specialty Medicines grew £293m year over year at actual rates while General Medicines shrank £234m; the difference plus Vaccines' £54m is the group's £113m increase. As Specialty's share moves from 39% to 42% to, on the company's own full-year guide of low-double-digit Specialty growth against low-single-digit declines elsewhere, something closer to 44% by year end, the drag from the declining base shrinks mechanically. This is the mechanism behind the 2031 outlook of more than £40bn.

"Q1 performance was strong. Sales were up 5% to more than GBP 7.6 billion. Growth was driven by Specialty Medicines, which were up 14%, with vaccines also contributing particularly through strong Shingrix sales."
— Luke Miels, Chief Executive Officer

Assessment: The mix shift is real, measurable and on schedule, and it is the reason a mid-single-digit grower deserves more than a mid-single-digit-grower multiple. It is also slow. At three points of mix shift a year, the General Medicines drag remains material through 2028, which is why the 2031 outlook is where the value sits and why management's decision to defer the portfolio detail to July mattered.

3. Shingrix Sets a Record and Pre-Warns Its Own Comp

A £1,026m Shingrix quarter at +20% CER, with Europe at +51%, is the strongest single-product print in the release, and management spent the call managing expectations down from it rather than up. The US contribution was explicitly attributed in part to inventory movement around the prefilled syringe launch, and the quantification was volunteered rather than extracted.

"In terms of stocking, we did launch the PFS to fully liquid in '26. It's easier for pharmacists. We don't factor any demand increase because of that, but it's just easier for the pharmacists to employ it. The wholesalers were pretty steady. So Q1 26.6 million doses. If you look at the end of last year, it was 0.5 million. If you look at same time last year, it was 0.4 million, so very much in the typical range. There is some increase in retail inventory. So -- and that's associated with the PFS. So it was 2.4. If you look at same time last year, it was 1.7 million doses and at the end of 2025, which is not necessarily a fair comparator of the flu season, 1.4 million."
— Luke Miels, Chief Executive Officer

On China, where Shingrix has been the largest single disappointment of the past two years, management gave a straight and unwelcome answer: patient demand is improving but reported sales will not reflect it while the co-promotion partner works down existing stock.

"So just briefly on China, the number of doses administered to patients is increasing. So that demand is improving. You will not see that in the sales numbers because that's going out from the available stock at. Probably for this year, at least majority of this year, we would not see sales numbers changing on the GSK side as that stock is being reduced."
— Nina Mojas

Assessment: Two useful things happened here. The stocking was disclosed with actual dose counts, which is more transparency than the category norm, and the China recovery was pushed out to 2027 in reported terms. Both are honest. Both mean the Vaccines line gets worse before it gets better, and the segment guide of a low-single-digit decline to stable already says so.

4. HIV: Long-Acting Injectables Are Working, and the Successor Regimen Is Not Close

The HIV franchise is 24% of group turnover and carries the largest single identifiable risk in the equity, the dolutegravir loss of exclusivity. Q1 was a good quarter for the mitigation strategy: Cabenuva +31% CER contributing more than half of total HIV growth, Apretude +44% CER contributing more than a fifth, competitor switches to Cabenuva reaching 79% in the US. The pipeline that is supposed to carry the franchise past the cliff is where the call turned uncomfortable. Management confirmed the Phase II programme for the third-generation integrase inhibitor VH184 has started, VH499's Phase II starts in the second half, Phase III begins in 2028, and approval lands at the end of 2030.

"We've already started the Phase II-A, which is the oral step we need to go through with VH184. We're expecting the VH499 Phase II to start in the second half of the year. And then basically, we would be starting the Phase III, which we are in dialogue with regulators over as well as our Phase II program in 2028, and that allows us then to generate the data that will give us an end of decade launch."
— Deborah Waterhouse, CEO, ViiV Healthcare

Asked directly whether the Phase II work could be compressed to pull that forward, the answer was unambiguous and negative.

"The FDA will not allow us or anybody else actually to skip the Phase II part of the development journey. We have to demonstrate the level of efficacy, safety as well as the kind of appropriate partner for our Q6M. So we'll go through the journey of development to demonstrate all of that, but we're in dialogue with the FDA, and there's nowhere we can skip the Phase II."
— Deborah Waterhouse, CEO, ViiV Healthcare

Assessment: The nearer-term bridge is intact and is arguably the better story: three-times-yearly Cabenuva launches in 2028, three-times-yearly Apretude has registrational data in H2 2026 and launches in H1 2027. The twice-yearly regimen, which is the one that would genuinely reset the competitive position, is now firmly a next-decade asset. Investors who had been carrying it as a 2029 event moved it out on Wednesday. That is a legitimate downgrade of a specific pipeline expectation, and it is one of the two things that broke the stock.

5. Camlipixant: Management Lowers Its Own Bar Before the Readout

The other one. Camlipixant is GSK's P2X3 antagonist for refractory chronic cough, with a pivotal readout due in the second half. On the call, the Chief Scientific Officer twice framed a 15-20% placebo-adjusted reduction at 24 weeks as the threshold for clinical significance, first unprompted in the CALM-2 discussion and then again under challenge.

"And perhaps just a reminder for everyone, here again, 15% to 20% reduction relative to placebo in Phase III at 24 weeks would be seen as significant."
— Tony Wood, Chief Scientific Officer

That is a materially lower bar than the Phase II data implied, and the parallel to a competitor's P2X3 that failed to secure approval at a similar effect size was raised directly on the call. Management's rebuttal was mechanistic rather than statistical, arguing that the failed competitor compound's problem was off-target binding at taste receptors, causing dysgeusia that unblinded the trial and constrained dose selection.

"So I think the key thing is duration of effect and also the placebo adjusted. That's the operative term. And the element with Merck's product really was the off-target effect, in fact, it's a much more promiscuously binding molecule in terms of addressing the receptors of P3X3, which are present in the taste buds. And so you get this taste dysgeusia, which unblinded the product and also limited their dose selection. So you had higher disturbance toxicity and lower efficacy plus some regulatory issues around cough monitoring."
— Luke Miels, Chief Executive Officer

Assessment: The mechanistic argument is credible and the differentiation is real. But a company that pre-frames a success threshold at half the level its own Phase II delivered is telling you something about its confidence interval, and it chose to do so twice in one hour. Volunteering a lower bar ahead of a readout is a defensible act of expectation management and an unambiguous negative for anyone modelling the asset. The stock's afternoon reflected that.

6. Bepirovirsen: The Largest Free Option in the Story

Bepirovirsen, GSK's antisense candidate for a functional cure in chronic hepatitis B, had positive Phase III headline data in the quarter, has breakthrough designation in the US with a PDUFA date of 26 October 2026, and has been accepted for priority review in China. Full B-Well 1 and 2 data come at EASL in May. Management describes the US and China as roughly two-thirds of the global commercial opportunity and is visibly more excited about it than the Street is.

"A recent U.S. epidemiology study in hepatitis B patients showed that loss of surface antigen was associated with an 89% reduced risk of hepatocellular carcinoma and a 62% reduced risk in all-cause mortality."
— Tony Wood, Chief Scientific Officer

The commercial framing is deliberately narrow, which is a good sign rather than a bad one. Management described a US opportunity concentrated in about five states with high immigrant populations from Asian and African countries, patients who are already diagnosed, already insured and already motivated; and a Chinese opportunity concentrated in roughly the top 15% of accounts. On the label question, management declined to be drawn, noting only that the B-Well studies enrolled patients with surface antigen levels of 3,000 or less, that this covers about 65% of the intent-to-treat population, and that the sub-1,000 group is about 45%.

Assessment: The comparator arithmetic is the whole argument. Management put the current standard of care's functional cure rate at under 2% for lifelong nucleoside therapy and pegylated interferon at 2-4% with a heavy tolerability burden. Against a base that low, any durable functional cure rate is transformative rather than incremental. What management has not given is the number: the release and the call say only that B-Well 1 and 2 delivered a "statistically significant and clinically meaningful" increase in the rate of functional cure, with full data at EASL in May, and the Chief Scientific Officer restated the company's long-standing view that 15-20% functional cure across the population would be considered clinically meaningful without indicating where the trials landed. The efficacy figure is the single largest unknown in the position, not a settled input. If the Street is carrying a couple of hundred million for this asset, and the October PDUFA converts, the gap between ambition and consensus is the single largest source of positive surprise in the story. We are paying nothing for it at 10.4x.

7. R&D Spend Accelerates 12% and the Governance Behind It

Core R&D of £1,493m grew 12% CER and reached 19.6% of sales from 18.3%, driven by the efimosfermin and velzatinib pivotal programmes, the ADC acceleration, and the ultra-long-acting HIV treatment and PrEP work. On the Total basis, R&D grew 19% CER because of higher impairments of £172m. Full-year guidance has R&D growing ahead of sales, so this is on plan and not a surprise, but it is a real drag on the operating line that the market is not currently paying for. Asked what pipeline acceleration means operationally, the chief executive gave the most concrete answer of the call.

"So every 2 weeks, Tony, Nina and myself, Mondher and Deborah, if it's HIV with David. We look through all the clinical execution, look at what studies are on track, which are not. And then it may trigger a discussion around the protocol design, the execution on the ground. ... And then we want to gauge the level of risk we're taking. So it's really everything that you're saying. The core focus is I strongly -- and you guys know this better than me, I strongly believe the way that we're going to create value is to accelerate what we have in the late-stage portfolio and get it to patients faster in a more broad fashion if there are opportunities for life cycle management."
— Luke Miels, Chief Executive Officer

Assessment: Seven Phase III starts in 2025 and ten planned for 2026 is a genuine step-change in late-stage activity, and the governance description is specific enough to be checkable. The cost of it is visible now and the benefit is not, which is the standard shape of a pipeline investment cycle and the standard reason the market discounts it. We would rather own the company making this choice at 10x than the one deferring it at 14x.

8. The Q2 Event Is a Pipeline Update, Not a Re-Guide

GSK cancelled a planned standalone HIV investor event and folded it into a broader portfolio update at the Q2 results in July. Two separate analysts pressed on what that event actually is, and the answer both times narrowed rather than widened the expectation.

"I mean we've found in the past when we do meet the management events as stand-alone, they're useful to get granularity, but the portfolio is becoming broad enough and complex enough. We thought it would be helpful for you and shareholders for us to step you through why we're so enthusiastic and what we've been doing with our time over the last few months in terms of accelerating these assets. So that's the intent and to give, yes, just greater granularity, more depth on the data."
— Luke Miels, Chief Executive Officer

Assessment: Some portion of the shareholder base had positioned for a capital-markets-day-style reset with refreshed mid-term financial targets, and was told on the call that this is a portfolio and data event. Removing an anticipated catalyst is worth something to a stock that had run 39.8% in twelve months. It also means the 2031 outlook of more than £40bn stands unamended and unelaborated for another quarter, and the HIV number inside it stays undisclosed.

9. Capital Allocation: Net Debt Up £1.2bn, Buyback Ends at the Half Year

Net debt rose to £15,613m from £14,453m at the end of 2025, a £1,160m increase in a quarter that generated £815m of free cash flow. The bridge is clean: £1,404m net acquisition cost for RAPT Therapeutics, £643m of dividends, £326m of buyback and a £154m exchange loss, against the free cash inflow and £383m from the Rockville disposal. Leverage is 1.4x EBITDA per the CFO. Free cash flow of £815m against £697m is up 17%, helped by the $250m ViiV special dividend and the CureVac settlement, and free cash flow conversion improved to 47% from 43%.

"In the second quarter, we expect to have an outflow of $950 million for the acquisition of 35Pharma. And we are optimizing the portfolio and generating cash income to reinvest in the business, including the ViiV special dividend, the divestment of the Rockville manufacturing site and the out-licensing of linerixibat. These transactions will positively impact net debt by $1.2 billion in the first half, including linerixibat completed in Q2, yielding $400 million."
— Julie Brown, Chief Financial Officer

Assessment: The balance sheet is doing three jobs at once (buyback, dividend, business development) and doing them without strain at 1.4x. The relevant model point is that £1.7bn of the £2bn buyback is executed and the programme completes at the half year, which removes roughly a point of EPS support from H2 onwards unless it is renewed. Nobody asked whether it would be. That question belongs in July.

10. Currency and the Section 232 Settlement

Sterling strengthened against the Dollar from an average $1.26 in Q1 2025 to $1.35 in Q1 2026, which is the entire three-point gap between +5% CER and +2% AER turnover growth. The release discloses that holding 22 April closing rates of $1.35, EUR1.15 and Yen 215 for the balance of the year would cost 2 points of Sterling turnover growth and, if exchange gains and losses repeat at 2025 levels, 4 points of Sterling Core operating profit growth. Separately, the tariff position is now documented rather than assumed: a definitive agreement signed 9 April 2026 provides Section 232 relief through 20 January 2029, subject to final implementation, against a 100% proclamation on patented pharmaceuticals effective 31 July 2026. Full-year guidance is stated to be inclusive of the expected impact.

Assessment: The tariff exposure has moved from an open-ended risk to a documented three-year exemption with implementation conditions, which is a meaningful de-risking that received no airtime on the call and no credit in the price. The currency drag is the opposite: fully disclosed, fully in the price, and the reason a reaffirmed CER guide reads as a downgrade in reported terms.

11. Launch Execution: Exdensur and Blenrep Are Both Gated on Access

The two most important launches in the company are both currently constrained by reimbursement mechanics rather than demand. Exdensur is waiting on a US J-code expected in early July, with only about 20% of US commercial patients currently having access; management characterises the post-J-code position as unrestricted. Blenrep has a majority of its US use in the community setting, which management flags as the leading indicator of success for a community-ready antibody drug conjugate, and second-line approval in 19 markets including China.

"So this is a therapy area where J-code is very relevant. We -- it is going to unlock vast majority of the market at the moment, only about 20% of the commercial patients in the U.S. have access, and this is very normal in this therapy area. J-code is really a significant barrier initially and then it opens the opportunity as the J-code becomes available."
— Nina Mojas

Assessment: Access gating is the least interesting kind of launch problem because it resolves on a calendar. The July J-code is a hard, dateable catalyst on the highest-value near-term launch in the portfolio, and management has committed to giving the full picture on both assets at the Q2 results. That makes the July event more consequential than management's own framing suggested.

Guidance & Outlook

Nothing in the guidance changed, which is the point. GSK confirmed every line of the February framework at the same time as disclosing that the translation into reported Sterling had deteriorated. The CFO's phasing commentary was the substantive new information, and it was uniformly cautionary: Vaccines benefited in Q1 from US prefilled-syringe stocking and begins annualising Japanese and European public programmes from Q2; General Medicines growth is second-half weighted with a tough Q2 Trelegy comparator on prior-year true-ups; operating profit growth is predominantly second-half weighted on the phasing of productivity benefits; and Q2 comps against the RSV intellectual property settlement received in Q2 2025.

"We still expect operating profit growth to be predominantly half 2 weighted given the phasing of productivity benefits. And you'll recall, we will also be comping the RSV IP settlement received in Q2 last year."
— Julie Brown, Chief Financial Officer

Implied Q2-Q4 ramp. Q1 2025 turnover of £7,516m was 23.0% of FY 2025's £32,667m. With Q1 2026 delivering +5% CER against a full-year range of +3% to +5%, the remaining nine months must deliver roughly +2.4% to +5.0% CER. On profit, Q1 2025 Core operating profit of £2,533m was 25.9% of FY 2025's £9,783m; with Q1 at +10% CER, the balance of the year needs roughly +6.0% to +8.7% CER. Strip Q1's three points of legal settlement and the underlying quarter delivered about +7%, which sits inside that required range rather than above it. The quarter therefore banked less headroom than the headline suggests, and management's four separate Q2 cautions say the shortfall is front-loaded into the June quarter.

Street at. Consensus was modelling the upper end of the +7-9% Core EPS range before the print. The reaffirmed CER guide plus the newly quantified 4-point Sterling drag on Core operating profit growth is what moves the reported outcome toward the middle or lower end without changing a single guided number. That gap, between a maintained constant-currency range and a deteriorating reported translation, is the mechanism by which a beat becomes a de-rating.

Guidance style. The precedent argues for conservatism rather than concern. At the Q1 2025 stage GSK affirmed full-year 2025 guidance of +3-5% turnover growth, +6-8% Core operating profit growth and +6-8% Core EPS growth at constant currency. The year finished at +7%, +11% and +12%. A company that under-guided its own range by three to four points twelve months ago, and whose house style is to hold ranges through the first quarter rather than move them on a single print, is not sending a distress signal by reaffirming in April. The Q2 print on 28 July is where the year gets decided.

Analyst Q&A Highlights

What the July Investor Event Will and Will Not Contain

The scope of the deferred strategy update was the most-repeated line of questioning on the call, raised at the top of Q&A and again later in a form designed to pin management down. The first framing asked how much emphasis to expect on the mid-term 2031 outlook versus the longer term, and whether an HIV-specific 2031 number would be given. The second asked, bluntly, whether this is a comprehensive top-line and bottom-line review or a pipeline update. The one-word answer to the second is the artifact.

Q: "If I could please just go back to the Q2 business update. Just trying to understand, is this meant to be a comprehensive review, both top line and bottom line trajectory? Or is it meant to be a portfolio update around the pipeline, including ViiV, please?"
— Michael Leuchten

A: "Thanks, Michael. The latter. I mean we've found in the past when we do meet the management events as stand-alone, they're useful to get granularity, but the portfolio is becoming broad enough and complex enough. We thought it would be helpful for you and shareholders for us to step you through why we're so enthusiastic and what we've been doing with our time over the last few months in terms of accelerating these assets."
— Luke Miels, Chief Executive Officer

Assessment: Management answered honestly and immediately, and the answer removed a catalyst. Investors who had been holding for a mid-term financial reset in July now know they are getting data depth instead. The 2031 outlook stays as a single group number for at least another quarter, and the HIV component inside it remains undisclosed heading into a loss of exclusivity that is the central bear argument on the stock.

Whether the Camlipixant Phase III Bar Has Been Quietly Lowered

The sharpest exchange of the call. A questioner put together three facts management had not connected in public: that the stated Phase III significance threshold is 15-20% placebo-adjusted, that Phase II delivered roughly twice that, and that a competitor's P2X3 antagonist with a similar effect size failed to obtain approval. Management's response addressed the mechanism thoroughly and the effect-size gap not at all.

Q: "Just on camlipixant, I hear the comment on 15% to 20% benefit of Phase III being significant. But I think the Phase II was about twice as good as that. So why would it be so much lower? And isn't the 15% to 20% pretty similar to what Merck had with their P2X3 that ultimately, I know there's some differences, but ultimately, that wasn't approved by the FDA with about a 15% to 20% benefit."
— James Gordon, Barclays

A: "So I think the key thing is duration of effect and also the placebo adjusted. That's the operative term. And the element with Merck's product really was the off-target effect, in fact, it's a much more promiscuously binding molecule in terms of addressing the receptors of P3X3, which are present in the taste buds. And so you get this taste dysgeusia, which unblinded the product and also limited their dose selection. So you had higher disturbance toxicity and lower efficacy plus some regulatory issues around cough monitoring. So I think we're talking apples and pears there, but we've taken those lessons and integrated them into not only the assessment of the clinical program, and we're looking forward to getting those results."
— Luke Miels, Chief Executive Officer

Assessment: The differentiation argument on tolerability and blinding is scientifically coherent and probably correct. It is also not an answer to why the company's own success threshold sits at half its Phase II result. Management chose to restate the lower bar rather than defend the higher one, which is the behaviour of a team protecting itself against a disappointing readout. We now treat camlipixant as a low-probability option rather than a modelled asset, and we suspect a meaningful part of the Street did the same on Wednesday afternoon.

Whether the Twice-Yearly HIV Regimen Can Arrive Before the End of the Decade

A recurring line of questioning tested the credibility of the end-of-decade timeline for the twice-yearly long-acting injectable. The first attempt asked whether Phase II could be skipped or shortened; the answer was no. The follow-up was more pointed, noting that the dosing interval has not yet been achieved for the lead asset and expressing open scepticism about the schedule. The response walked through the sequence in full, which is the most useful disclosure the franchise gave all quarter.

Q: "Deborah, in your opening comments, you expect confidence in the long-acting 6-month treatment regime on the market by the end of the decade. But given that we haven't yet achieved 6 months IM dosing for VH184, I have to be honest and say we're struggling to get to that time line. Sach asked if you could accelerate it and you said no. So can you walk through the steps for development gives you the confidence in being there at the end of the decade?"
— Matthew Weston

A: "So we've got Phase II program, which has already started for VH184 because I think what people sometimes think is when we present the data at CROI, that's where we are in the process. We've already started the Phase II-A, which is the oral step we need to go through with VH184. We're expecting the VH499 Phase II to start in the second half of the year. And then basically, we would be starting the Phase III, which we are in dialogue with regulators over as well as our Phase II program in 2028, and that allows us then to generate the data that will give us an end of decade launch."
— Deborah Waterhouse, CEO, ViiV Healthcare

Assessment: The question was well-founded and the answer was substantive rather than evasive, which is the right outcome even though the content is unhelpful. A 2028 Phase III start for an end-of-2030 approval is a schedule with no slack in it, on the asset that is supposed to defend the franchise. It also means the three-times-yearly launches in 2027 and 2028 are the load-bearing part of the bridge, not the twice-yearly regimen. That reweighting is not in most models.

Competitive Threat From Once-Weekly Oral HIV Regimens

With a competitor's once-weekly oral combination approaching launch, the question was whether it compresses the long-acting injectable opportunity. Management's answer separated the market cleanly and made a specific, falsifiable claim about cannibalisation, which is the kind of answer worth recording so it can be graded later.

Q: "Just one on HIV. And obviously, we're expecting some competitor data for a once-weekly oral treatment option this year. Some of the physician feedback that we've had suggests that there'll be strong demand for that. So Deborah, just wanted to understand how you think about that from a ViiV perspective in terms of near-term impact? And then ultimately, do you think that, that is restrictive to an ultra-long-acting injectable?"
— Rajan Sharma

A: "And all the research that we've done says that the weekly orals cannibalizes the daily orals. Actually, our research shows that it doesn't impact long-acting injectables because that is a very specific patient segment where you've got people who struggle to adhere who really feel very stigmatized by taking a tablet every day or are really worried about people discovering their status as well as obviously the benefit of directly observed therapy. So I think the long-acting injectable segment won't be impacted by the once weekly, but the daily orals are most likely to be."
— Deborah Waterhouse, CEO, ViiV Healthcare

Assessment: The segmentation logic is sound and the adherence argument is the correct one. What management did not say is that the daily oral segment it concedes will be cannibalised is where Dovato lives, and Dovato at £666m is the single largest product in the HIV franchise. Defending the injectable business while conceding the oral one is a coherent position that still costs revenue. Additional caution on the competitor's own asset, on CD4 depletion at higher doses, reads as a genuine clinical concern rather than positioning, but it is not a moat.

Bepirovirsen's Commercial Ambition Against a Consensus That Carries Almost Nothing

A questioner framed the disconnect explicitly, noting that the Street carries only a couple of hundred million for the asset while management talks about it as a franchise, and asked the practical question underneath it: whether surface antigen testing, which is rarely performed today, can be made standard practice. The answer identified the diagnostic bottleneck as a solved problem waiting for a solution to exist.

Q: "In fact, there's a massive disconnect between your ambition in terms of commercial ambitions for the product and I think consensus which is only a couple of hundred million baked in. I realize we cannot go into any B-Well details, but I would like to understand how GSK thinks about getting surface antigen testing part and parcel of standard practice. KOL feedback tells us this is rarely done presently."
— Peter Verdult, BNP

A: "I mean, structurally, we've seen this in multiple disease areas. If there's no solution, there's no point looking for the problem. So many of these patients, if you look at the U.S., diagnosis rates, a prevalence of 1.2, I said before, only 500,000 are diagnosed in Europe is a similar ratio. Japan is probably the best of all of them. But our expectation is and the feedback that we've got is that once you've got the accessible treatment option because the downstream consequences of this infection are deeply unpleasant, for the individual and the health care system, we expect that testing to increase."
— Luke Miels, Chief Executive Officer

Assessment: This is the most asymmetric item in the story and the one the market gave the least attention. The launch requires a diagnostic behaviour change, which is a genuine execution risk and a slow one. But the payoff structure is unusual: management put existing nucleoside therapy's functional cure rate below 2% and pegylated interferon at 2-4% with heavy tolerability burden, against a first-in-class mechanism that has drawn breakthrough designation in the US and acceptance for priority review in China. The B-Well efficacy figure itself is not public until EASL, so the asymmetry is a structural one rather than a quantified one. A PDUFA date of 26 October gives that asymmetry a hard date inside our investment horizon, and the current multiple pays nothing for it.

Jemperli's Softer Quarter and the Path to the Peak-Sales Ambition

Oncology grew 28% at constant currency and Jemperli grew 40%, and the question was still why the quarter was soft against expectations. The follow-up asked whether the two pending rectal cancer readouts really add much given an existing tumour-agnostic label, and whether head and neck is the real step-up. Management's answer confirmed the ambition is unchanged and split it by indication, then added a commercial observation that is more revealing than the pipeline answer.

Q: "Maybe you could talk a little bit about the softer Q1 relative to expectations after a pretty strong run of results. Particularly interested in how endometrial and rectal outlook is shaping up. I mean you do have AZUR-1 and 2 pending, but they're in the MSI-H setting and you already have a tumor-agnostic MSI-H label. So I would imagine they're going to have pretty limited impact. Is it going to be JADE in head and neck that really catalyze the next step up?"
— Emmanuel Papadakis

A: "And I'd add, Emmanuel, we've still got a lot to do operationally in the U.S. in terms of endometrial. If you look at the stats, about 60% of oncologists just used Keytruda despite the overall survival. So we've got plenty of area to target those individuals, and we do have their market research that if a physician can recite the survival benefit, they're a lot more likely, obviously, to use Jemperli."
— Luke Miels, Chief Executive Officer

Assessment: A candid admission that 60% of prescribers in the lead indication still default to the incumbent, four years after approval and with a demonstrated overall survival benefit, is a commercial execution gap rather than a clinical one. That cuts both ways: it explains the soft quarter and it quantifies the remaining headroom. The peak ambition management reaffirmed splits roughly evenly between endometrial and the colorectal plus head and neck opportunity, and the rectal readout later this year is the near-term proof point.

Shingrix US Stocking and the Quality of the Record Quarter

A three-part question tested the durability of the record Shingrix print, asked whether China is recovering, and probed the expanding dementia programme. The stocking answer was the most quantified disclosure of the entire call, and it did not fully support the demand framing in the prepared remarks.

Q: "A question on Shingrix with 3 brief parts. First, can you quantify the magnitude of U.S. inventory stocking? Second, are things improving in China? And thirdly, it seems like you're all in on dementia starting a 34,000-patient trial after being cautious for a long time. Is that how to read it?"
— Steve Scala

A: "So Q1 26.6 million doses. If you look at the end of last year, it was 0.5 million. If you look at same time last year, it was 0.4 million, so very much in the typical range. There is some increase in retail inventory. So -- and that's associated with the PFS. So it was 2.4. If you look at same time last year, it was 1.7 million doses and at the end of 2025, which is not necessarily a fair comparator of the flu season, 1.4 million. So stocking has a component there, but we're also seeing reasonable underlying demand following the strategy of focusing on comorbid."
— Luke Miels, Chief Executive Officer

Assessment: Retail inventory of 2.4 million doses against 1.7 million a year earlier is a 41% increase in the channel on a product whose US sales grew 12%. Management's own framing, that stocking "has a component there" alongside reasonable underlying demand, is the fair characterisation. Volunteering the dose counts rather than deflecting was the right call and it lets modellers do their own arithmetic, which is what a company confident in the underlying demand does. The China answer, that reported sales will not move for most of 2026 while the partner destocks, was the genuinely negative half of the exchange.

What They're NOT Saying

1. The absolute value of the legal settlements: the release gives an SG&A effect of "around 4ppts" and the CFO gave three points of operating profit growth, but no Sterling figure and, more importantly, no statement about whether further settlement income is expected in the remaining three quarters. Modellers are left to assume zero, which is prudent and may be wrong in either direction.
2. Any framing of Q2 beyond a list of headwinds: management named four separate Q2 pressures (Shingrix European and Japanese programme annualisation, the Trelegy prior-year true-up comparator, the RSV intellectual property settlement comp, and second-half-weighted productivity benefits) without offering a single quantification or a Q2 range. That is a lot of caution with no anchor.
3. The HIV contribution to the 2031 outlook: asked directly whether a 2031 HIV number would be provided, management folded it into the July portfolio event and said only that they "want people to look at the total business in aggregate." A company confident in its post-loss-of-exclusivity HIV trajectory would give the number.
4. The bepirovirsen label: "I'm not going to get into the details of the label." Given that the sub-1,000 surface antigen group is 45% of the intent-to-treat population and the sub-3,000 group is 65%, the label breadth is the single largest determinant of the asset's commercial value, and it is six months from being known.
5. Any Blenrep or Exdensur revenue trajectory: both were deferred with an explicit promise of "the full picture" at Q2. Exdensur delivered £11m and Blenrep £23m against a portfolio that badly needs both to scale.
6. Why the camlipixant Phase III bar sits at half the Phase II result: the mechanistic differentiation from the failed competitor compound was addressed at length; the effect-size question that prompted it was not addressed at all.
7. Whether the buyback will be renewed: £1.7bn of £2bn is executed and the programme completes at the half year. That removes roughly a point of Core EPS support from H2. Nobody asked and management did not volunteer.
8. The £265m contingent consideration and put-option remeasurement charge: disclosed in the release, absent from the call. The Pfizer put option was fully derecognised at 31 March as Pfizer exited its ViiV shareholding, which is a structural change to one of the group's most complex ownership arrangements, and it received no management commentary whatsoever.

Market Reaction

• Pre-print setup: the ADR closed at $54.47 on 28 April, up 11.1% year to date against the S&P 500's 4.3%, up 39.8% over trailing twelve months and up 0.4% over the trailing thirty days. The 52-week closing range entering the print was $36.20 to $61.18. The London line closed at 2028p. Positioning was long and the stock had already been rewarded for the story.
• Release-day session: the ADR opened at $50.00, a gap of 8.2%, traded a range of $49.99 to $52.13, and closed at $51.40, down 5.6% or $3.07. Volume was 11.2 million shares against a 30-day average of 4.1 million, a 2.8x multiple. The close was 2.8% above the intraday low.
• Primary listing: GSK.L closed at 1918p, down 5.4% from 2028p, on 14.9 million shares. The London line opened at 2050p, up 1.1% on the release alone.
• Peer and index cross-check: the S&P 500 was unchanged on the session. Large-cap pharma comparators were flat to modestly lower, with the largest peer decline at 1.8% and two comparators closing higher. The move was idiosyncratic to GSK.

The most informative fact about the reaction is its sequence. The results were released at 07:00 BST, an hour before the London open, and the London line opened up 1.1%. The market read the print, priced it positively at the opening auction, and then reversed through the session, with the 12 noon BST results call sitting in the middle of the decline. Pre-open indications were not uniformly positive, with wire coverage marking the shares down around 3% shortly after the release crossed and the ADR quoted lower in US pre-market trade, so the release did some of the damage. But the opening print was up and the bulk of the decline came after it.

Three things changed between the open and the close, and each of them came on the call. Management quantified the legal settlement contribution at three of the ten points of operating profit growth, converting a 7.7% operating beat into a roughly 7% underlying growth quarter. Management confirmed the July event is a pipeline and portfolio update rather than a comprehensive financial re-guide, removing a catalyst a portion of the shareholder base had positioned for. And management twice restated a camlipixant Phase III success bar at 15-20% placebo-adjusted, roughly half what Phase II delivered, ahead of a second-half readout. Add the four separate Q2 headwinds the CFO listed without quantifying, and a stock that entered the print up 39.8% over twelve months had ample reason to give back a day.

The recovery of 2.8% off the intraday low into the close is worth noting. The opening gap of 8.2% was a positioning flush at a price the market did not defend; the 5.6% close is the considered number. That distinction matters for anyone treating the sell-off as an entry point, because the entry is at 1918p and not at the low.

Street Perspective

Debate: Is the Earnings Quality Complaint a Real Problem or an Accounting Artifact?

Bull view: the settlement benefit is disclosed, quantified and non-recurring by definition, and stripping it still leaves roughly 7% constant-currency operating profit growth on a business guiding to 7-9% for the year. Gross margin expansion of 110bps at constant currency is the structural line and it is entirely clean.

Bear view: a company that needs a settlement to clear consensus in its easiest comparative quarter, ahead of four separate second-quarter headwinds it declined to quantify, is a company whose full-year range is at risk from the top down. Underlying growth at the bottom of the guided range in Q1 is not a comfortable place to start.

Our take: the bulls have the arithmetic. Q1 2025 Core operating profit was 25.9% of the full year, so the balance of 2026 needs roughly 6.0% to 8.7% constant-currency growth to land in the range. Underlying Q1 delivered about 7%, which is inside that band rather than below it. The quarter banked less headroom than the headline implied and it did not create a hole. The complaint is fair as a description of optics and overstated as a description of risk.

Debate: Does the 2031 Outlook Survive the Dolutegravir Loss of Exclusivity?

Bull view: the long-acting injectable transition is already carrying 73% of HIV growth, competitor switches to Cabenuva reached 79% in the US this quarter, and two three-times-yearly launches arrive in 2027 and 2028, well inside the window. The franchise is being rebuilt on injectables before the oral cliff arrives.

Bear view: the twice-yearly regimen, the asset that would genuinely reset the competitive position, cannot enter Phase III before 2028 and does not launch before end-2030. Meanwhile a competitor's once-weekly oral is about to cannibalise exactly the daily-oral segment where Dovato, the franchise's largest product at £666m, lives. Management has also declined to disclose the HIV component of the 2031 number.

Our take: the bears have the better of this one on timing, and the market repriced it correctly on Wednesday. The bridge to 2031 rests on the three-times-yearly launches, not the twice-yearly regimen, and that is a narrower bridge than the story implied a week ago. The refusal to disclose the HIV 2031 number, when asked, is the tell. This is the pillar most likely to break, and it is the reason our conviction is moderate rather than high.

Debate: Is 10x a De-Rating Opportunity or a Value Trap?

Bull view: a reaffirmed high-single-digit Core EPS algorithm, a 3.65% covered yield, leverage at 1.4x, a fastest-growing 42% of revenue compounding at 14%, and a documented three-year tariff exemption, all for 10.4x forward earnings. The de-rating happened on optics rather than fundamentals.

Bear view: the discount has persisted for years and nothing structural has changed to close it. Below the operating line, the finance expense grew 45%, the tax rate is guided up, and the buyback that has been absorbing both completes at the half year. The declining base is still 30% of revenue. Cheap is what this asset is, not a temporary condition.

Our take: the bear case correctly describes the past and understates what changed in the last eighteen months. Oncology at +43% in 2025 and +28% in Q1, ten pivotal starts planned for 2026 against seven in 2025, and a first-in-class hepatitis B candidate with an October PDUFA are not the characteristics of a perpetual value trap. The re-rating requires the pipeline to convert, and there are four pivotal readouts and a PDUFA before year-end to test it. We are willing to underwrite that at 10.4x with a 3.65% yield paying us to wait.

Debate: Is Bepirovirsen a Free Option or a Distraction?

Bull view: current standard of care delivers functional cure in under 2% of patients on lifelong therapy, or 2-4% with pegylated interferon and severe tolerability burden. Any durable improvement on that comparator is category-defining, the Phase III readout was positive on a statistically significant and clinically meaningful basis, regulators have granted breakthrough designation in the US and priority review in China, and consensus reportedly carries only a couple of hundred million.

Bear view: the launch depends on making surface antigen testing routine when it is currently almost never performed, in a disease where most patients are undiagnosed. The label may be restricted to a low-antigen subgroup representing 45% of the trial population. China, roughly a third of the opportunity, has a pricing environment and an entrenched interferon habit that management admits it is still "reflecting on very actively."

Our take: both are right, and that is what makes it an option rather than a forecast. We model nothing for bepirovirsen and treat the 26 October PDUFA as an unpriced binary. The bear objections are real but they are launch-curve objections, not thesis objections: they shape how fast the revenue arrives rather than whether the asset works. At a multiple that pays nothing for it, an asymmetric outcome with a hard date inside twelve months is exactly the kind of thing a 10x entry is supposed to buy.

Model Update & Valuation Framework

This is our initiation on GSK, so the table below sets our estimates against the company's guidance rather than against a prior model.

What would change our mind. A camlipixant miss is already largely discounted and we carry it at zero. The genuinely damaging outcomes are a bepirovirsen label restricted to the sub-1,000 surface antigen group, which would validate the bear framing of a niche asset; a Q2 print showing General Medicines still running at -6% with no second-half recovery visible, which would put the turnover guide at risk from the bottom up; or any slippage in the three-times-yearly HIV PrEP timeline, which is now the load-bearing element of the loss-of-exclusivity bridge. Any two of those together would take us to Hold.

Thesis Scorecard Post-Earnings

This is our first published view on GSK, so the pillars below establish the thesis rather than grade a standing one. Each is scored against what this quarter's print and call actually revealed.

Overall: thesis established. Three of four bull pillars confirmed on hard evidence in the quarter; the HIV pillar is neutral on a timeline that lengthened. Both of the major bear points were confirmed, and the loss-of-exclusivity point escalated rather than merely persisted. That is not a comfortable scorecard, and it is priced: the stock fell 5.6% on the day these facts emerged, into a multiple that assumes the bear case is broadly correct.

Action: Initiate at Outperform with moderate conviction. Buy the de-rating rather than chase strength, size for a position that has to survive a soft Q2 print on 28 July, and treat the October PDUFA as the asymmetry we are being paid to wait for. The 3.65% yield is the carry on the wait.