April 19, 2012 — Florbetapir (Amyvid, Eli Lilly/Avid Radiopharmaceuticals), a new agent to detect beta-amyloid plaques in living patients with possible Alzheimer’s disease (AD), has just been approved by the US Food and Drug Administration (FDA). The question now is how this imaging option will be used in practice.
Although they are for the most part enthusiastically awaiting access to this new agent, expected to be available by June, many neurologists are also striking a cautionary note. Cost, availability, the need for expert interpretation of scans using the florbetapir tracer, and what it really means for a diagnosis of AD are a few of the concerns being raised.
Medscape Medical News polled experts in the field of AD to see how they view the approval and how they see this diagnostic tool may fit into their clinical practice.
Florbetapir is a diagnostic agent tagged with a radioisotope, fluorine-18. Used with positron emission tomography (PET), it binds to amyloid plaques in the brain. Approved by the FDA on April 9, florbetapir is 1 of 3 imaging agents in various stages of development; others are florbetaben (Bayer/Piramal Imaging SA) and flutemetamol (GE Healthcare). All are reported to detect amyloid deposition in the living brain.
Although the presence of amyloid on the scan doesn’t necessarily mean the patient has AD, a scan showing little amyloid deposition, “is inconsistent with a neuropathological diagnosis,” of AD, a press release from the company at the time of approval notes. The statement added that the safety and effectiveness of florbetapir have not been established for predicting development of dementia or other neurologic conditions, or for monitoring responses to therapies.
Still, the use of florbetapir may clear up some diagnostic dilemmas, said Sandra Black, MD, professor, Department of Medicine, Division of Neurology, University of Toronto, Ontario, Canada. “It might help in situations where you’re not quite sure what’s going on — when you don’t know whether this is aphasia due to AD or a frontal temporal dementia-type thing.”
Pedro Rosa-Neto, MD, assistant professor, neurology, neurosurgery and psychiatry, McGill University, Montreal, Quebec, Canada, felt that the new tracer could be informative in the investigation of patients with early onset or atypical presentations of dementia to rule out AD, and in cases of rapidly progressive dementia.
“Rapidly progressive Alzheimer’s disease is a highly neglected condition, frequently misdiagnosed as Creutzfeldt-Jakob disease,” Dr. Rosa-Neto noted. For these selected cases, he sees PET using florbetapir adding to the information gained from the clinical history, various magnetic resonance imaging (MRI) modalities, including fluid-attenuated inversion recovery and diffusion-weighted imaging, cerebrospinal fluid sampling, and neuropsychological and genetic evaluations.
Although the tracer could be of assistance in patients with dementia in whom it’s unclear whether amyloid is a cause of cognitive deterioration, it would not be as useful an addition in cases where all clinical, cognitive, and imaging findings strongly point to the presence of AD, said Liana Apostolova, MD, Alzheimer’s Disease Research Center, University of California, Los Angeles.
“While the test could be useful to confirm that presumption, dementia specialists can already diagnose AD under these circumstances with high sensitivity,” she said. “This is when I might say, ‘I’m pretty certain what’s going on at this point; I don’t need to subject the patient to this expensive diagnostic procedure.”
Role in Mild Cognitive Impairment?
Some clinicians agreed that there is a place for the new biomarker in patients with mild cognitive impairment (MCI). Ronald Petersen, MD, PhD, director, Alzheimer’s Disease Research Center, Mayo Clinic, Rochester, Minnesota, credited with developing the concept of MCI, said the tracer “will be helpful” to see whether amyloid can explain memory problems experienced by these patients. “It will not make the diagnosis, but it will help the clinician sort out the cause of the symptoms.”
Dr. Petersen pointed out that many patients with MCI — perhaps up to 40% — don’t have AD as the underlying cause of their cognitive problems.
For Steven T. DeKosky, MD, vice president and dean, University of Virginia School of Medicine, Charlottesville, the new biomarker “seems like a reasonably accurate detector” of whether MCI is AD in development or whether it relates to some other cause.
Other physicians said they wanted to avoid using the new tracer in cases of MCI.
“Although I think a positive amyloid imaging study would increase the likelihood that the person will decline, that is not known for certain,” said David Knopman, MD, Department of Neurology, also at Mayo Clinic Rochester, Minnesota. “So, I would definitely not do an amyloid scan routinely.”
Doctors stressed that a positive scan does not necessarily mean an asymptomatic patient will develop AD. “Many cognitively normal elderly will show amyloid deposition in their brains as the disease is believed to have an asymptomatic stage that lasts up to 20 years,” pointed out Dr. Apostolova.
“I don’t think we know enough about the prognostic factors of a positive amyloid scan,” said Adam S. Fleisher MD, director of brain imaging, Banner Alzheimer Institute, Phoenix, Arizona, during a Web-based debate on the role of neuroimaging in the diagnosis of AD earlier this year.
On the other hand, a negative scan may not mean a patient is out of the woods because there could be another explanation for their cognitive problems, such as vascular dementia.
“We will have to be very careful using amyloid imaging and making sure patients understand that a negative scan is not necessarily a cause to celebrate,” said James B. Brewer, MD, PhD, associate professor, radiology and neurosciences, Human Memory Laboratory, University of California at San Diego, who also participated in the Webinar.
Possible other causes of cognitive decline could include stroke, thyroid problems, drug interactions, chronic alcoholism, and vitamin deficiencies. Psychiatric disorders such as depression can masquerade as dementia as well.
The Alzheimer’s Association acknowledged that FDA approval of florbetapir is a “double-edged sword,” although it supports the move.
…the fact that all of the potential uses of this product are not crystal clear tempers our enthusiasm.
In a statement, the Association said that although the approval will expand the clinical and research opportunities for amyloid imaging, “the fact that all of the potential uses of this product are not crystal clear tempers our enthusiasm.” Additional research is needed to clarify the role of florbetapir-PET imaging in Alzheimer’s, it added.
Physicians weighing in agreed that amyloid measurements alone are not enough — that other imaging tests, including MRI, should also be part of the diagnostic equation. Many talked about the “multi-modal use” of this and other emerging amyloid biomarkers.
And, as Dr. Brewer pointed out, the new amyloid tracer does not measure tau, which some experts believe plays a crucial role in AD. Another agent under development by researchers at the University of California, Los Angeles, called FDDNP, images both tau and amyloid on PET.
All doctors also emphasized the importance of patient counseling before ordering an amyloid test.
“You have to have a good understanding of how that’s going to influence discussions with the patient, and your treatment and prognostic decision-making,” said Dr. Fleisher. He added that the test results could be quite anxiety provoking for a patient.
Doctors who spoke to Medscape Medical News also expressed a concern about “direct to patient” advertising and about “worried well” patients asking for this test. Many called for caution about using this new biomarker in cognitively normal patients who are anxious about their amyloid status.
The Alzheimer’s Association, too, warns of “less than scrupulous” imaging operators who make unrealistic promises to such patients about the value of florbetapir imaging. It recommends that the test be accessible only in the context of a complete evaluation of medical/neurologic status and with appropriate expert consultation.
Cost may be a significant issue in determining whom to scan using this new biomarker, doctors agreed. The price tag for the radioactive compound alone appears to be in the neighborhood of $1600, said Dr. Apostolova.
“Then there will be the added cost of getting the PET scan which in most places will be between $1000 and $1500. So I think we are looking at a cost of about $2500 to $3000 per scan here.”
Because the tracer will be costly, “we have to use it with caution and only when it’s really necessary,’ said Dr. Apostolova, “Is it going to tell me anything beyond what I already know?”
As for insurance coverage, Dr. Petersen said that it’s unclear at this time who will pay for this new agent when it becomes available in June. Dr. Knopman commented that that “to start with, no insurer will pay for it.” The agent is also not currently eligible for coverage by the Centers for Medicare and Medicaid Services (CMS) because of its specific policy on reimbursement for PET procedures.
During a press conference after approval of the agent, Stephanie Prodouz, manager of Eli Lilly Bio-Medicines Communications, said that, “Although Amyvid will not be eligible for coverage, it’s important to note that Lilly is working with a broad group of stakeholders to explore and collaborate with CMS to find a new policy for PET coverage.”
Although there is palpable excitement among some neurologists about the diagnostic possibilities of this new tracer, their enthusiasm is also tempered by the lack of available AD treatments.
“From a scientific viewpoint, the development of amyloid imaging was huge [but] the commercial availability is less so, unless it can be tied to finding better therapeutics,” said Dr. Knopman. “Until there are better therapeutics for AD, amyloid imaging does not change the landscape in clinical practice.”
Until there are better therapeutics for AD, amyloid imaging does not change the landscape in clinical practice.
But new treatments are on the horizon, according to Marwan Sabbagh, MD, director, Banner Sun Health Research institute, Sun City, Arizona. “Later this year, you will see 2 large immunotherapy studies — bapineuzumab and solanezumab — reporting their data to the world on their phase 3 trials, so the game changing elements here are perfectly timed in anticipation of these potential disease modifying agents.”
Bapineuzumab and solanezumab are both monoclonal antibodies that bind to and clear beta amyloid. Dr. Sabbagh is an investigator on both trials and serves on the steering committee for the bapineuzumab study.
Still, it’s unclear what role the amyloid plays and whether clearing it will result in any change in cognitive status. However, if safe and effective treatments become available, “that changes the whole ball game,” commented Dr. Brewer.
Dr. Apostolova agreed. “If we can find an agent that can halt the disease progression, before symptoms occur, this [florbetapir] will become the single most useful diagnostic and prognostic test out there,” she added.
For his part, Dr. DeKosky said that if an effective AD drug were available, the amyloid test could be used not necessarily to rule out AD but to determine that the cognitive impairment a patient experiences is indeed AD, so that drug could be administered.
Still, even in the absence of more effective amyloid treatments, many doctors felt that the test could nevertheless be useful. For example, said Dr. Brewer, it provides physicians with an opportunity to educate patients, help them manage risk factors and perhaps get them enrolled in a clinical trial.
As well, said Dr. Apostolova, the test results may spur patients to make necessary work-related arrangements or put their personal affairs in order. “Knowledge is power,” she said.
According to Eli Lilly, the information on scans using the new tracer correlates highly with what is seen on autopsy. In a study that used the majority interpretation of 5 readers, there were 96% sensitivity and 100% specificity in patients who underwent scanning within a year of death.
Dr. Black called this autopsy correlation research “elegant.” “It showed that indeed they were really picking up the amyloid” prior to death in these patients, she said.
The clinical studies also showed that the new diagnostic tracer appears to be very safe. The most common adverse reactions reported in these trials were headache (1.8%), musculoskeletal pain (0.8%), fatigue (0.6%), nausea (0.6%), anxiety (0.4%), back pain (0.4%), increased blood pressure (0.4%), claustrophobia (0.4%), feeling cold (0.4%), insomnia (0.4%), and neck pain (0.4%).
The studies are outlined in more detail in the product prescribing information.
But although the new imaging agent is safe and promises to be helpful diagnostically, training of interpreters will be key. Florbetapir was approved only with the proviso that Eli Lilly improve education initiatives for readers of the images. In January 2011, the FDA’s Peripheral and Central Nervous System Drugs Advisory Committee voted 13 to 3 against recommending approval for the tracer, but in a second unofficial vote, the committee voted 16 to 0 in favor of approval if the company would agree to structured training for those reading the scans.
“There was a fear that radiologists would not know how to interpret the scans, as well as a concern about what would be an appropriate use of this agent,” said Dr. Black.
After working with the FDA and nuclear medicine experts, Eli Lilly developed both an online and an in-person training program for physicians. According to the company, images should be interpreted only by readers who have successfully completed a training course. The company cautions that errors may occur in the estimation of plaque density during image interpretation.
“This is a new kid on the block; it’s a new technique” that requires training, commented Dr. Apostolova, who carries out imaging research. “It’s not the case that one day, all of a sudden, one’s brain’s tissue becomes full of amyloid. It’s a slow protracted continuous build up.”
Patients will fall on a continuum, with some having minimal amyloid binding and others moderate or severe amounts, she added. “The last ones would be the easiest to call, but there will be many intermediate cases where we have to decide on a threshold that we will be using to call a scan positive versus calling it negative.”
The lack of familiarity in reading scans could pose “a big problem” for clinicians since “it’s not straightforward,” commented Dr. Petersen.
Dr. Knopman agreed, saying he’s “very concerned” about the scan interpretation issues. “I am dubious that radiologists who don’t do a lot of them will misinterpret them,” he said. “Even after nuclear medicine physicians have taken the course provided by Lilly, if they have low instances of contact with the test, they are likely to make mistakes in interpretation.”
He added that he’s particularly worried about the risk for overdiagnosis of Alzheimer’s, especially when the amyloid imaging is not interpreted in light of the clinical history.
“For example, up to 30% of cognitively normal people over age 70 years are expected to have a ‘positive’ scan,” he said. “I’m worried that a normalperson will hear that their amyloid scan shows ‘Alzheimer’s disease’ and will conclude that they have the disease and are at imminent risk of losing their memories. In fact, an abnormal amyloid scan should be interpreted as showing amyloidosis, not Alzheimer’s.”
Dr. Black said her “hunch” is that florbetapir will be more likely to be used in specialty memory clinics that have access to experts who can properly interpret the scans than in general family practices.
One of the advantages of the new agent is its versatility. Florbetapir has a half-life of almost 2 hours, compared to a half-life of 20 minutes for the currently available amyloid imaging agent, carbon 11–labeled Pittsburgh compound B (C-PiB).
The discovery of PiB was truly ground-breaking and set the stage for the current series of amyloid tracers now hitting the market, said Dr. DeKosky, who led the clinical trials of PiB.
“PiB was 7 years ahead of all these others,” he said. “We were able to find a way to non-invasively detect amyloid in living people harmlessly, and then several people found a way to develop different compounds. They are all to be congratulated, but the first one is always the one for which there’s romance,” he says wryly.
But because even florbetapir still loses over half of its radioactivity every 2 hours, it must be distributed in a timely fashion from a radiopharmacy to an imaging center. To address this problem, Siemens PETNET Solutions, a subsidiary of Siemens Medical Solutions USA Inc, announced last week a manufacturing and distribution agreement with Eli Lilly.
Beginning in June, Siemens will supply florbetapir to imaging centers in limited US markets. By the end of the year, the company anticipates having 25 manufacturing centers and co-located radiopharmacies offering the compound.
Dr. Apostolova pointed out that to administer florbetapir, a center must also have a PET scanner. In addition to the manufacturing and distribution of florbetapir, Siemens announced its new Biograph mCT PET/CT (computed tomography) scanner, and related software.
Many neurologists said they have faith that the distribution network used by Lilly/Avid will make the agent widely available to clinicians.
“Now we have PET scanners everywhere, which we didn’t have maybe 7 or 8 years ago,” Dr. Black noted. “And now we have an 18-fluorine compound and there are others that should be emerging soon.”
This widening and promising landscape, she said, “is very exciting.”