Subscribe to eTOC

FDA Approves Florbetapir for Imaging Amyloid Beta for Alzheimer Disease

On April 10, the FDA approved the drug Amyvid (florbetapir F 18 Injection) for the imaging of amyloid using positron emission tomography (PET) in adults being evaluated for Alzheimer disease (AD) and other cognitive decline.

“Florbetapir F 18 is not only exciting because it is targeting a dreaded disease that affects millions, but this is only one of a handful of PET drugs that have ever been approved for use in the USA,” Val Lowe, MD, professor of radiology at Mayo Clinic in Rochester, MN, who was not involved in the approval, told Neurology Today.

Last January, because of concerns regarding the misreading of amyloid scans, the FDA Peripheral and Central Nervous System Drugs Advisory Committee unanimously agreed (16-0) that it could only approve the injection after the development of a reader training program. The manufacturer, Eli Lilly and Company, of Indianapolis, IN, complied with this stipulation, and also included a warning of possible scan misinterpretation in its prescribing information.

A negative scan, meaning little or no plaques are present, indicates a reduced likelihood of cognitive impairment as a result of AD, the FDA said, while a positive scan, or moderate to frequent plaques, can be found in patients with AD but also in those with other disorders and with normal cognitive functioning. “It's important to note that florbetapir F 18 is an adjunct to other diagnostic evaluations. A positive scan does not establish a diagnosis of Alzheimer disease, or other cognitive disorder,” according to a statement by the manufacturer.

This approval was based on three clinical studies that examined images from healthy adult patients and patients with different cognitive disorders. Based on the results of the first study, measurements of postmortem cortical amyloid burden correlated with median florbetapir F 18 scores (r=0.78; p< 0.0001). In study two, florbetapir F 18 PET showed 96 percent sensitivity and 100 percent specificity in patients who received the scan within one year of death.

In all three studies, there was an overall median sensitivity of 92 percent (ranging 69 percent to 95 percent) and specificity of 95 percent (ranging 90 percent to 100 percent) for readers trained in person (study two), and median sensitivity of 82 percent (range 69 percent to 92 percent) and specificity of 95 percent (ranging 90 percent to 95 percent) for readers trained using an electronic media-based training (study three), according to the manufacturer.

The most commonly reported adverse reactions with florbetapir in the clinical trials were headache (1.8 percent), musculoskeletal pain (0.8 percent), fatigue (0.6 percent), nausea (0.6 percent), anxiety, back pain, blood pressure increase, claustrophobia, feeling cold, insomnia, and neck pain (all 0.4 percent). The agent is manufactured for Avid Radiopharmaceuticals, Inc., a wholly owned subsidiary of Eli Lilly and Company.


Neill Graff-Radford, MD, professor of neurology at Mayo Clinic in Jacksonville, FL, said similar drugs, like the Pittsburgh Compound-B (PiB) have already made major breakthroughs in AD research, but the 20-minute half-life of PiB limits its clinical use. Florbetapir, on the other hand, has an almost two-hour half-life, which allows for its transportation to treatment centers.

Said Dr. Lowe: “The differences in terms of disease characterization will likely be insignificant between the agents.” He added that any distinctions may become more evident with time.

Florbetapir will be especially useful in cases where it is apparent that a person has some disease, but not clear which: “For example, when a person has dementia and you want to identify whether it is a frontotemporal dementia or an AD case, we often resort to expensive tests like PET scans,” Dr. Graff-Radford said. At this time, though, he described the test as “boutique.”

Dr. Lowe agreed. Since there are no successful treatments for AD, it could be hard to justify to third party payers that an amyloid imaging test is helpful or will benefit patients, he said. But the diagnosis itself may provide some comfort to the “around 15 percent of dementia patients who are difficult to diagnose,” though it may not improve their outcome.

The company's specialized training program, Dr. Graff-Radford said, should help minimize the risk of translation error in the reading of scans — especially over time. His biggest concern is that the injection will be used in cognitively normal patients, causing those with a positive scan to worry. “We don't know the natural history of cognitively normal people with amyloid plaques. We suspect that many of them will go on to develop AD, but the long-term studies aren't there yet.” There should be guidelines for the use of florbetapir in cognitively normal patients, he said.

If AD pathology begins years before diagnosis — as indicated in the Dominantly Inherited Alzheimer's Network ( — Dr. Graff-Radford said that this may provide an important window for future study using imaging agents like florbetapir to identify and delay or prevent AD-onset.