Unlike any other theory in drug development, the amyloid hypothesis of Alzheimer’s disease has maintained an almost cult-like following, despite two decades littered with a long line of late-stage failures and not a single success. But the storied history holds valuable lessons that, if heeded, could improve the odds of the next wave of therapies, both for AD and other neurodegenerative diseases.
Since 2002, amyloid-lowering compounds have been tested in at least 30 failed Phase II or Phase III trials, many of which were explained away by bad trial designs, lack of biomarkers or the need for better compounds (see Table: “Full of Fail”).
One issue that’s clouded the picture is that the disease is defined by the presence of amyloid deposits in the brain, making a circular argument out of questions about the peptide’s role.
Still, multiple lines of genetic data converge on β amyloid as the primary driver of AD, creating a case for the target that some in the field find air tight.
The assumption has been that targeting the disease-causing agent will translate into a disease-modifying therapy. One of the biggest lessons is that this logic can no longer be taken for granted.
This is particularly true in neurodegenerative diseases, because they lie undetected for decades, recruiting additional pathological mechanisms along the way. After symptoms have started, it may be too late to target the initial trigger.
Amyloid failures also call into question the viability of targeting protein aggregates, or at least any single aggregation-prone protein, a common approach in neurodegenerative diseases.
The silver lining is that the amyloid work has led to better tools for selecting AD patient