Twenty-two days after the Cochrane Collaboration published its meta-analysis of anti-amyloid antibodies for Alzheimer's disease — concluding that the drugs produce "absent or trivial" effects on cognitive decline and dementia severity — the public scientific fight has hardened into a methodology dispute with disclosed institutional conflicts on both sides. [1] The UK Dementia Research Institute's Bart De Strooper and Alzheimer's Research UK's Sir John Hardy have not retreated. Hardy, who first proposed the amyloid hypothesis of Alzheimer's in the 1990s, told AFP the Cochrane review is "a silly paper which should not have been published," disclosing that he has consulted for Eli Lilly, Biogen, and Eisai — the manufacturers of the two drugs (Kisunla and Leqembi) the review puts in question. [2] De Strooper has also disclosed consulting relationships with Eisai, Eli Lilly, and Johnson & Johnson, and is a founder of two spin-off biotechs. [3] The Cochrane review's authors — lead author Francesco Nonino at the IRCCS Institute of Neurological Sciences of Bologna and senior author Edo Richard at Radboud University Medical Centre — have not been working on amyloid-targeting drugs and have no comparable industry consulting income.

Tuesday's paper named the institutional pushback and the methodological critique. The press fight is not the story. The fight is between a meta-analytic methodology that pooled data across seven anti-amyloid drugs (five of which failed clinical trials and were discontinued) and a research community whose leading voices are also paid consultants to the two drug manufacturers whose products were included in the analysis. Both sides have skin in the game. What matters more is the regulatory decision the National Institute for Health and Care Excellence will make on June 10, when its committee meets to reconsider whether the National Health Service should cover Kisunla and Leqembi at any price the manufacturer offers. The NICE reconsultation closed April 28. The committee meets in five weeks. The June 10 meeting is the artifact that produces a price-and-coverage decision that will affect roughly 700,000 British patients in the early stages of Alzheimer's. The Cochrane review is the input. The reconsultation is the output.

In plain English: a major medical research group says these drugs don't really work. The leading British Alzheimer's experts say the research group is wrong, and disclose they're paid consultants to the drug companies whose products are at issue. A British government health regulator is going to decide June 10 whether the National Health Service will pay for the drugs. That decision is the one that matters, not the press fight. NICE has previously declined to recommend Kisunla and Leqembi for NHS coverage on cost-effectiveness grounds, despite the European Medicines Agency and the UK's own MHRA approving the drugs based on demonstrated clinical benefit in trials. The reconsultation is the second look at the cost-effectiveness math. The revised cost-effectiveness thresholds in NICE's draft guidance — pegged to QALY (quality-adjusted life-year) gains and a willingness-to-pay threshold of £20,000 to £30,000 per QALY — depend on what the regulator concludes about the magnitude of the benefit and the cost of treatment.

The Cochrane review's specific finding is that the absolute effects of the drugs on cognitive decline at 18 months are "absent or trivial," falling below the established thresholds for the minimum clinically important difference. [4] The senior author Edo Richard, a sceptic about anti-amyloid therapy for years, said: "The idea that removing amyloids will benefit patients was refuted by our results." [2] The pushback from De Strooper and Hardy is methodological: the review pooled 17 trials covering seven drugs, of which only one trial each studied Kisunla (donanemab, the TRAILBLAZER-ALZ 2 trial) and Leqembi (lecanemab, the Clarity AD trial). The remaining 15 trials studied drugs that failed in late-stage development and are not available to patients. "The flaw in this review is fundamental," De Strooper said. "By mixing failed drugs with the only antibodies that have actually changed clinical practice, it turns therapeutic progress into statistical noise." [3] The Recognition Health UK FAQ summarizes the magnitudes: Kisunla slowed cognitive decline by 35 percent at 18 months in the TRAILBLAZER-ALZ 2 trial; Leqembi slowed it by 27 percent in Clarity AD. [5]

Whose framing is right is the question NICE's committee has to resolve. Cochrane's argument — that pooling drugs targeting the same mechanism (amyloid-beta clearance) produces a methodologically defensible average — has weight. The argument De Strooper and Hardy make — that pooling drugs with different mechanisms (some that bind soluble protofibrils, others that target plaques on the brain surface) produces a misleading aggregate — also has weight. The Alzheimer's Society in the UK, in its April 22 response, took a middle position: "We consider anti-amyloid drugs still to be a valid avenue to explore and optimise" but acknowledged that the drugs are "not a silver bullet" and that more research is needed on combination approaches and longer-term outcomes. [6] The four-year extension data on Leqembi's clinical effect, not included in Cochrane's 18-month analysis, suggests sustained but modest benefit; this longer-term data is a piece of evidence the NICE committee will weigh.

The cost question is what produces the regulatory artifact. A 20 mL vial of Kisunla costs $695.65 in the United States. The first three doses require two vials each; subsequent doses require four vials each. Leqembi's wholesale price is $254.81 for a 200 mg vial and $637.02 for a 500 mg vial. The standard dose is 10 mg/kg of body weight; an average patient consumes about 540 mg per dose every two weeks. [7] In the United States, Medicare Part B covers both drugs. In the UK, neither is currently covered by the NHS. The NICE committee's June 10 decision turns on whether the manufacturer (Eli Lilly for Kisunla, Biogen and Eisai for Leqembi) offers a price discount that brings the cost-per-QALY below the regulator's willingness-to-pay threshold. The Cochrane review's conclusion that the drugs produce no clinically meaningful benefit, if accepted by NICE, would push the QALY math into territory where any list price would be unaffordable. The expert pushback's conclusion that the drugs produce a 25-35 percent slowing of decline, if accepted by NICE, would make the cost question more tractable.

The patient-facing reading is unsentimental. A British patient with mild cognitive impairment from early Alzheimer's, today, has access to symptomatic treatments (donepezil, rivastigmine, galantamine, memantine) on the NHS but not to disease-modifying therapy. After June 10, that patient may have access to Kisunla or Leqembi at a contracted NHS price, or may not. The American patient's situation is different: Medicare Part B coverage is in place, but with prior authorization requirements and 20 percent coinsurance after the deductible. Whether a treatment is "available" is a matter of geography, not biology. The Cochrane review and the pushback are both arguments before a regulator that will decide who gets the drug at what price. The press fight will continue. The regulator's decision will not. By mid-June, the British answer will be on the page. The American answer is already on the page. The European Medicines Agency's parallel reconsultation track sits without a date. June 10 is the day to read.

-- NORA WHITFIELD, Chicago