FDA Gives Lilly's Amyloid Tracer Preliminary Thumbs Down

Last November, Eli Lilly promised $300 million upfront* to snap up Avid Pharmaceuticals and its PET (and pet?) tracer florbetapir (or Amyvid) for detecting amyloid in the brain. The offer may have been for relative naught, as predicted here. And now, a newly released FDA review of the tracer's clinical data, made in anticipation of an advisory committee meeting on January 20, indicates that the agency is, likewise, sorely underwhelmed.

The first concern of the regulatory body, as previously forecasted here, is whether the amyloid tracer (or any amyloid tracer, for that matter) has any real clinical utility for diagnosing Alzheimer disease. Avid's own proposed marketing for florbetapir tepidly stipulates, "A negative florbetapir-PET scan is clinically useful in ruling out the presence of pathologically significant levels of [beta]-amyloid in the brain." But the clinical utility of a positive scan is unreliable (if not terribly confusing), because most brains (up to 80%) of cognitively normal older people will carry some amyloid burden. Consequently a positive scan (which is not just possible, but probable, in the elderly) is irrelevant in the population in which it is intended for use.

Although the FDA doesn't have problems with the scan being able to reliably detect brain amyloid (on the basis of postmortem correlations), it also has concerns about inter-rater reliability with respect to scan interpretation. The agency concludes,

Overall, FDA's major concerns with the Amyvid application relate to an apparent insufficient development of the reader training methods proposed for clinical use, including verification that these methods ensure acceptable reader-to-reader consistency in image interpretation across a patient population representative of that proposed for clinical use as well as reliability of the image interpretations with respect to the truth standards used in the phase 3 clinical trial.

PET = positron emission tomography.

* The company would offer another $500 million on the basis of florbetapir's FDA approval and other "commercial milestones," which seem unlikely at this point.

Photograph of atrophied brain from person with AD: National Institute on Alcohol Abuse and Alcoholism.