The statistic that lands hardest in Cochrane’s April 2026 review is this: 20,342 patients. That’s the combined enrollment across 17 randomized trials testing seven anti-amyloid antibodies. The programs span Biogen’s Aduhelm, Eisai’s Leqembi, Lilly’s Kisunla, and four others that never made it to market. The conclusion is blunt, “little to no difference” compared with placebo on either cognitive or functional measures. With data this large showing a trivial effect size, the signal now reaches far beyond the lab; valuation models for every amyloid-removal program suddenly look overexposed.
Quantifying triviality: below the “minimum clinically important difference”
By calling the effects “absent or trivial,” and demonstrating they fell “well below established thresholds for the minimum clinically important difference,” Cochrane reframes what “success” has meant in Alzheimer’s R&D. Leqembi and Kisunla both hit statistical significance on slowing clinical decline, gaining full FDA approvals, yet the meta-analysis argues these shifts don’t produce real-world improvement in daily function or disease level at 18 months. For payers and policy analysts, that’s a flashing sign: a drug can win the p-value battle and still lose the value war. As that idea spreads, reimbursement stress will follow.
The review also quantifies downside. Cochrane found that these antibodies “likely increase the risk of brain swelling,” referencing amyloid-related imaging abnormalities (ARIA). The clinical literature consistently connects ARIA with treatment pauses and costly MRI follow-up. Most patients show no symptoms, but the scan requirement alone inflates per-patient cost as launches scale across health systems. Hard to spin that as operationally neutral.
Seven molecules, one mechanism, billions already gone
The meta-analysis examined seven drugs: Aduhelm, Leqembi, Kisunla, bapineuzumab, crenezumab, gantenerumab, and solanezumab. Only two, Leqembi and Kisunla, made it through regulators. The other five represent more than a decade of failed R&D. The math is brutal: two successes, five dead ends, all chasing the same target. When 70% of a pathway’s assets die on the vine, investors start asking how long the hypothesis should keep drawing capital. Cochrane’s suggestion that “future research should focus on other mechanisms of action” isn’t radical, it’s relief disguised as guidance.
The 17 trials involved all used placebo-controlled designs, a methodological strength that anchors the credibility of this “trivial effect” finding. Lilly and Eisai counter that mixing approved and failed molecules muddies the statistical waters, fair point. Yet the pooled results now define the evidence set everyone will cite: regulators, investors, payers. Until something delivers clearly bigger cognitive or functional changes, the anti-amyloid economy stands on shaky ground. And maybe everyone already knows it.
Corporate defenses and commercial fallout
Lilly pointed to its Phase III TRAILBLAZER-ALZ 2 data, claiming significant slowing of decline in early Alzheimer’s. Eisai emphasized that Leqembi’s FDA review was “thorough and independent.” Their pushback isn’t just scientific, it’s defensive accounting. These two drugs were pitched as the first “disease-modifying” therapies, priced accordingly, pulling through imaging and diagnostic demand. Rebranding that as “trivial effect” threatens both the premium and the pipeline narrative beneath it. Revenue models buckle fast when that language shifts.
Meanwhile, the pipelines remain crowded with other amyloid-focused bets, especially in mid-stage trials. This isn’t just about two products, it’s the scaffolding of an entire generation of similar antibodies. When history shows five of seven molecules failed despite design tweaks, investors notice. And hesitation spreads faster than enthusiasm once data this large goes cold.
Aftershocks and what’s actually next
If agencies like CMS or European HTA bodies adopt this interpretation, coverage for anti-amyloid drugs will tighten through 2026-2027. That shift would pull future funding toward tau, neuroinflammation, and synaptic resilience programs now sitting quietly in preclinical phases. Payers could start tying payment directly to functional improvement rather than imaging markers. The arithmetic turns sobering: 20,342 patients, negligible functional gain, expensive management of risk. That combination can’t justify today’s pricing logic forever.
So maybe the Alzheimer’s pipeline in 2027 finally splinters, less amyloid, more diversified neurology bets. After thirty years of chasing the same plaque, fatigue is setting in. And that, honestly, feels overdue.
