GSK's second-quarter report on Tuesday contained two facts that look contradictory and are not. The company said it now expects to start more than 20 late-stage clinical trials in 2026, double its earlier target of 10, and launched a £1.9 billion cost-cutting program largely to fund them. In the same report, it took a £1.3 billion writedown on camlipixant, a chronic-cough drug it will now abandon after Phase 3 results, a charge that drove total operating profit down 75%.
Read together, those are not a mixed message. They are the two halves of one strategy showing up on the same day: CEO Luke Miels is betting GSK's future on running more late-stage trials faster, and running more late-stage trials faster necessarily means more expensive, visible failures. Camlipixant is the first big one. It will not be the last, and that is not a flaw in the plan. It is the plan.
The strategy: accelerate the late-stage pipeline
Miels, who took over in January, has staked his leadership on a specific idea: speed up the development of GSK's most promising drugs, add to them through disciplined bolt-on acquisitions, and get more medicines through late-stage trials and onto the market faster. The Tuesday moves are that idea made concrete. The cost program, targeting £1.9 billion in annual savings by 2029, is explicitly designed to reallocate capital toward the late-stage portfolio and R&D, and the doubling of the Phase 3 target is the output that spending is meant to buy.
The logic behind it is sound and worth stating. A large pharmaceutical company's value is its pipeline, the drugs it will sell in five and ten years, not just the ones selling now. GSK faces a specific and dated threat: its top-selling HIV medicine, dolutegravir, loses patent protection between 2028 and 2030, which will open a large hole in revenue as generics undercut it. The way you fill a patent-cliff hole is with new drugs, and the way you get new drugs is to push more candidates through the expensive final stage of testing. Accelerating the late-stage pipeline is a rational response to a clock that is already ticking.
Why acceleration guarantees more failure
Here is the part that the camlipixant writedown makes vivid, and that the celebratory framing of "20 Phase 3 trials" obscures. Late-stage drug development is where the money is spent and where drugs die most expensively.
Phase 3 trials are the large, long, costly human studies that are the last step before approval, and they are where a great many drugs that looked promising in earlier testing finally fail, because a large controlled trial is the first time a drug's real effect is measured with enough rigor to be sure. Failure at this stage is not a rare misfortune. It is the base rate. Most drugs that enter development never reach the market, and a substantial share of the failures happen in Phase 3, after enormous sums have already been spent.
So the arithmetic of Miels's strategy is unavoidable. If you double the number of late-stage trials you run, you increase both the number of drugs that could succeed and the number that will fail, and the failures are costly precisely because Phase 3 is the expensive stage. A company that commits to running more late-stage trials is committing, mathematically, to absorbing more large writedowns like camlipixant's. The £1.3 billion charge is not evidence the strategy is going wrong. It is evidence the strategy is running, because you cannot buy more shots on goal without paying for the misses, and the misses at this stage are expensive by definition.
The right way to judge the approach, then, is not by whether failures happen, they will, but by whether the successes outweigh them by enough. That is the only question that matters, and it is one no single quarter can answer.
Why camlipixant stings more than a routine miss
Two details make this particular failure sharper than the base rate alone would suggest, and honesty requires noting them. First, the size of the ambition: analysts had discussed peak sales potential for camlipixant above £2.5 billion, so this was not a minor candidate quietly shelved but a drug GSK had hoped would be a significant product. Losing a potential blockbuster is a bigger blow than losing a marginal one.
Second, the timing against the target. GSK has a high-profile goal of more than £40 billion in annual revenue by 2031, and analysts already saw a gap of several billion pounds between that ambition and their own forecasts before this failure. Every late-stage drug that dies is a candidate that was supposed to help close that gap and now cannot, which makes the acceleration strategy and the revenue target two sides of the same bet: the target is only reachable if enough of the accelerated pipeline succeeds, and each visible failure raises the pressure on the survivors.
The acquisitions are the hedge, and their own risk
There is a second element to Miels's strategy that the trial numbers do not capture, and it changes the risk profile in both directions. He has paired the internal acceleration with stepped-up dealmaking, including a record acquisition of the cancer-drug company Nuvalent in June, buying late-stage assets rather than only developing them in-house.
The logic is a hedge. If your own Phase 3 pipeline will inevitably produce failures, buying additional late-stage drugs spreads the bet across more candidates and more sources, reducing dependence on any single internal program. That is prudent. But acquisitions carry their own version of the same risk, since an acquired late-stage drug can fail its trials too, and you will have paid a premium up front for it, so a failed acquired program is a failure you bought at full price. Dealmaking widens the pipeline; it does not exempt the strategy from the base-rate problem, it just distributes it and adds acquisition risk on top.
How to read it
The clean way to hold the quarter is that GSK reported a strategy and its cost in the same breath, and understanding that they are the same thing is the whole point. The core operating results were solid, profit up 7% stripping out the charge and a beat on expectations, so the underlying business is healthy. The headline 75% profit drop is almost entirely the camlipixant writedown, which is a one-time consequence of a deliberate strategic choice rather than a sign the business is deteriorating.
The real question the report raises is not whether the strategy is failing, the writedown does not show that, but whether the acceleration will produce enough winners to justify the failures it guarantees and to close the multi-billion-pound gap to the 2031 target. That is genuinely unresolved. Running more late-stage trials is the right move against a patent cliff, and it is also a commitment to a bumpier earnings profile, with periodic large writedowns as the price of more attempts. Investors who want the upside of an accelerated pipeline have to accept the volatility of more visible failures, because the two are inseparable.
For anyone watching GSK, the useful frame is to stop reading writedowns like camlipixant's as bad news in isolation and start reading them as the cost of admission for the strategy the company has chosen. The number that will actually decide whether Miels succeeds is not how many trials fail, but how many of the twenty-plus late-stage programs now underway convert into approved, selling drugs before the end of the decade. The failures will keep coming, on schedule and by design. Whether the successes come with them is the bet, and Tuesday's report showed both sides of it at once, which is exactly what an honest report of this strategy should do.
Primary sources
- Quartz for the doubling of the 2026 Phase 3 target to more than 20 trials, the £1.3 billion camlipixant impairment, the 75% drop in total operating profit against a 7% rise in core operating profit, the £1.9 billion savings target by 2029, and Miels's comments on reallocating capital.
- Reuters via Yahoo Finance for the £1.9 billion cost program to fund late-stage studies, the record Nuvalent acquisition, and the more-than-£40 billion 2031 revenue target amid patent losses.
- UBS via Yahoo Finance for the flagged shortfall of more than £4 billion between GSK's 2031 ambition and consensus forecasts, and camlipixant's £2.5 billion-plus peak-sales ambition against a £1.1 billion forecast.