Earlier this year, the FDA approved an imaging agent from Eli Lilly for detecting beta-amyloid plaques in patients with cognitive impairment and who are being evaluated for Alzheimer’s disease. Known as Amyvid, this is the first diagnostic agent approved for people who undergo a positron emission tomography, or PET, scan. But the approval was not a slam dunk - the agency initially declined to approve Amyvid over concerns that scans can be accurately read.
Now, the FDA has tagged Lilly (LLY) for offering physicians what the agency calls a misleading way to interpret scan results on the Amyvid web site and in recent displays at a medical conference. To wit, Lilly offered a multi-colored image of the brain, when the Amyvid labeling information says PET scans should be displayed and reviewed using a black-and-white scale "with the maximum intensity of the scale set to the maximum intensity of all the brain pixels.”
Why does this matter? In August 10 untitled letter, the FDA writes this "presentation misleadingly suggests that Amyvid PET images can be displayed, and therefore interpreted, in color in patients with cognitive impairment who are being evaluated for Alzheimer’s Disease and other causes of cognitive decline, when this is not the case... While many PET images are displayed and reviewed in color, Amyvid scans must be displayed and reviewed using a black and white scale."
And so, the FDA has accused Lilly of misbranding its product (here is the letter). The offending image, by the way, no longer appears on the Amyvid web site (look here), but the FDA says its Office of Prescription Drug Promotion was also made aware that the colorful scans appeared in the commercial exhibit hall of the American Academy of Neurology annual meeting held in New Orleans last April.
To some, this faux pas may be seen as a black-and-white issue that reinforces concern over the FDA approval. How so? To convince the FDA to approve Amyvid, Lilly had to demonstrate that interpreting brain scans showing plaques could consistently yield an accurate diagnosis. So the drugmaker conducted reader training for interpreting results, which Lilly maintained was effective and underscored the reliability of using its diagnostic tool.
Now, though, Lilly is slapped for carelessly presenting an inappropriate way to interpret scans. As we reported previously, Lilly was cautious to note that a positive Amyvid scan does not establish a diagnosis of Alzheimer’s and that using the agent is actually an adjunct to other diagnostic evaluations. A positive scan indicating moderate to frequent amyloid plaques may be present in other types of neurological condition or older people with normal cognition.
The statement reflected controversy over reading scans. Earlier this year, Public Citizen argued Amyvid should not have been approved. Sid Wolfe, who heads Health Research Group at Public Citizen, called Amyvid inaccurate and "subject to serious inter-reader interpretation differences that has been shown to falsely diagnose Alzheimer’s Disease in patients who do not have the disease and fail to diagnose it in patients who turn out to have the disease" (back story).
Last year, the consumer group had charged that the authors of an Avymid study published in the Journal of the American Medical Association excluded data on the extent to which the scans could be intepreted accurately from one doctor to the next (back story). Lilly denied the assertions and maintained the study data was conducted and disclosed properly.
We asked Lilly for a comment and will update you accordingly. [UPDATE: A Lilly spokeswoman sends us this: "Lilly is committed to providing clinically accurate and balanced promotional materials to healthcare professionals and patients, and takes this untitled letter very seriously. Lilly took immediate steps to discontinue use of multi-colored brain images in its marketing tactics, and has reviewed all Amyvid promotional materials to ensure compliance."
"I think its also important to note that all instructional material – including material for the reader training program – shows the scans in black-and-white. The label states that Amyvid images should be read in black-and-white as do Amyvid training materials. The multi-colored images were limited in use, generally appeared on cover pages or as stylistic background on some promotional materials and were intended to help customers immediately recognize Amyvid as a PET imaging agent, rather than a structural (CT/MRI) imaging agent (PET scans of the brain are typically read in color, while structural images are read in black-and-white)."]






8 Comments
If this image were to be submitted for journal publication it would be an immediate cause for rejection since journals these days pretty much uniformly require that all such images, from Western Blots to PET scans be displayed on what's called a relative intensity scale, analogous to the grayscale on a darkroom negative. This is the only way to properly view the relative degree of anatomical changes comparing one area of the brain to another.
Good catch.
Ahhhh,Lilly has learned since 'viva zyprexa' off-label sales promotions? Thank you Ed Silverman.