September 23, 2026

@michaelokun

My take on the negative ceperognastat trial is that it may ultimately teach us as much about how we develop Alzheimer’s therapies as about tau itself. Ceperognastat was designed to inhibit OGA and thereby alter tau biology, but in people with early symptomatic Alzheimer’s it failed to slow cognitive and functional decline; importantly, the higher dose was associated with greater, not less, clinical progression and more serious adverse events. What makes this trial especially instructive is the disconnect between what happened to the biomarkers and what happened to people: despite signals of less tau accumulation in some regions and substantially less brain-volume loss, patients did not improve, and those receiving the higher dose actually did worse. The accompanying editorial makes what I think is the critical point: biomarkers are enormously valuable for demonstrating target engagement and understanding mechanism, but until they are rigorously validated as surrogates, we cannot confuse moving a biomarker in the right direction with helping a person. The high-dose findings also remind us that OGA and O-GlcNAc biology extend well beyond tau, so manipulating a seemingly attractive pathway can have unintended biological consequences. I would not interpret this as evidence that tau is the wrong target; tau burden remains closely linked to neurodegeneration and clinical decline, and several next-generation tau approaches remain under study. Rather, the lesson is that the target, the drug, the dose, the timing and, above all, the clinical outcome all matter. At the end of the day, the patient, not the biomarker, must be the sun. https://cutt.ly/ayvQFfZA