On April 30, 2026, the FDA’s Oncologic Drugs Advisory Committee did the unglamorous thing it exists to do. It told AstraZeneca no. By a vote of 6 to 3, the outside experts concluded the company had not shown that its new breast cancer drug delivered a clinically meaningful benefit to the women who would take it. The panel had read the trial and heard the sponsor out. It looked at the centerpiece number, a near-doubling of the time before the cancer advanced, and decided the number did not mean what the company said it meant.

Four months later, on September 4, the FDA approved the drug anyway.

The drug is camizestrant, sold as Etcamah, an oral pill in a class called selective estrogen receptor degraders. The approved indication is narrow: adults with hormone receptor-positive, HER2-negative advanced breast cancer whose blood shows an emerging ESR1 mutation while they are still on standard first-line therapy. On paper the idea is elegant. Catch the resistance mutation in a blood draw before it surfaces on a scan, switch the patient off the failing hormone blocker and onto camizestrant, and buy her time. SERENA-6, the trial that tested it, was built to prove exactly that.

The design is the whole argument. SERENA-6 screened 3,256 patients, drawing their blood every two to three months, watching for the ESR1 mutation to surface. When it did, and only then, a patient became eligible. In the end 315 were randomized. Half were switched to camizestrant plus their existing therapy. The other half were kept on the aromatase inhibitor they were already taking, a drug the blood test had just flagged, at the molecular level, as beginning to fail.

That is the pivot the committee could not get past. Measured against a control arm left on a failing drug, the results were dramatic. Median progression-free survival was 16 months in the camizestrant group versus 9.2 months in the control group, a hazard ratio of 0.44, its 95 percent confidence interval running from 0.31 to 0.60, and a p-value below 0.0001. AstraZeneca led with the headline version, a 56 percent reduction in the risk of progression or death. On its own terms, the number is airtight.

MEDIAN PROGRESSION-FREE SURVIVAL (months)
Switched to camizestrant16Kept on the failing drug9.2
The comparison AstraZeneca led with. The control arm stayed on a therapy its own blood test had already flagged as failing. Source: FDA approval summary, 2026

The trouble is what the number is measured against. SERENA-6 did not test switching early against switching later. It tested switching against staying on a drug the blood had already flagged as failing. Take a patient off a treatment the moment you know it is failing, leave her counterpart on that same treatment for months longer, and of course the first patient’s cancer takes longer to show up on the next scan. You gave her a head start. The strategy’s actual claim, that catching the mutation early and switching beats waiting for the disease to progress and then switching, is the one comparison the trial never ran. There was no crossover to isolate the effect of timing. The gap between 16 and 9.2 months may measure a genuine benefit, or it may largely measure the interval the trial engineered into the control arm.

The endpoint that would settle it is overall survival, and on that the trial is silent. At the time of the analysis, the FDA notes, the overall survival data were not mature. The early trend favored camizestrant but did not reach statistical significance. Nobody yet knows whether the women who switch early live longer. They know their scans look better, sooner.

The label carries a boxed warning, the agency’s most serious, for a dangerous irregular heart rhythm when the drug is combined with certain other medications, along with precautions for an abnormally slow heart rate and for harm to a fetus. That is the price of admission for a benefit the FDA’s own advisers voted was not clearly there.

The mechanism that let the agency overrule its committee is accelerated approval, and it is doing a familiar job. It lets a drug reach the market on a surrogate measure, here progression-free survival, that is only reasonably likely to predict real benefit, on the promise that a confirmatory trial will settle the matter later. The FDA was candid about it in the approval itself: continued approval may be contingent on verification of clinical benefit in a confirmatory trial. The agency approved the drug and reserved the right to change its mind once it learns whether the drug works.

It is the same pathway that, in 2021, let the FDA wave through aducanumab, the Alzheimer’s drug, over the near-unanimous objection of its own advisory committee. That override cost the agency something. 3 of the panel’s 11 members resigned in protest, one of them calling the review a sham process. The lesson of that episode was not that outside experts are always right. In both cases the committee’s no turned out to be advisory in the fullest sense, a box checked on the way to a yes the agency was already leaning toward.

THE ADUCANUMAB PRECEDENT
3 of 11 advisers resigned
After the FDA overrode its Alzheimer's panel in 2021, 3 of its 11 members quit. Source: NBC News, 2021

Who wanted the yes is not a mystery. SERENA-6 was designed, funded, and run by AstraZeneca, which sells camizestrant and has a commercial stake in a future where every woman on first-line therapy has her blood monitored every few months for the mutation that triggers the switch to its drug. That is a large and recurring market, and it rests on a strategy the FDA’s own reviewers were unwilling to call a proven benefit.

The advisory committee voted 6 to 3 against it. The agency approved it anyway, and wrote into the approval that it does not yet know whether the drug will help the women who are about to be prescribed it live any longer.

Sources

  1. FDA – Approval summary: camizestrant for ESR1-mutated HR+/HER2- breast cancer (PFS 16 vs 9.2 months, HR 0.44; OS not mature; confirmatory requirement)
  2. FDA – Press announcement: accelerated approval of a new breast cancer treatment (Sept. 4, 2026)
  3. OncLive – FDA ODAC votes 6-3 against clinical benefit of switching to camizestrant
  4. AstraZeneca – Update on FDA advisory committee vote on camizestrant (Apr. 30, 2026)
  5. AstraZeneca – SERENA-6: camizestrant reduced risk of progression or death by 56%
  6. ASCO – SERENA-6 news release: 3,256 patients screened by ctDNA to randomize 315
  7. NBC News – Third member of FDA advisory panel resigns over aducanumab approval (2021)
  8. PharmacoEconomics & Outcomes News – FDA grants accelerated approval to controversial Alzheimer’s drug, aducanumab (2021)