A recently approved imaging agent to detect beta-amyloid deposits in brains of patients is changing the diagnostic landscape for dementia, despite lingering practice management issues and uncertainty over what the scan results mean.
Florbetapir (Amyvid, Eli Lilly) contains the radioisotope fluorine-18 that binds to beta-amyloid plaque. Beta amyloid is believed to be a cause of Alzheimer’s disease and is thought to begin accumulating in the brain a decade or more before onset of clinical impairment. Florbetapir PET results provide sensitivity and specificity in the 83 to 90 percent range.
“The florbetapir scan is a new tool in our arsenal for diagnosis for individuals with early state memory changes,” says Marc E. Agronin, MD, medical director for mental health and clinical research at Miami Jewish Health Systems in Miami.
A negative scan in a patient with cognitive impairment means at the time of the scan the patient does not have the proteins associated with Alzheimer’s, and physicians need to look for another type of dementia to explain the degeneration of cognitive function. A negative scan does not mean the patient can’t get Alzheimer’s or some other form of dementia later.
A positive scan means the patient has the protein associated with Alzheimer’s and is considered at much higher risk for Alzheimer’s.
“Florbetapir is not a metabolic study, like FDG, rather it’s pathologic,” says Mark S. Schechter, MD, director of PET imaging and nuclear medicine, Imaging Healthcare Specialists in San Diego. Florbetapir enables imaging of amyloid plaque that heretofore was discoverable only at autopsy or biopsy.
Although florbetapir signals a huge leap toward more accurate diagnoses non-invasively, the presence of amyloid does not confirm an Alzheimer’s diagnosis, says Mark Brody, MD, president, Brain Matters Research in Delray Beach, Fla.
Cognitively normal people can have amyloid in their brains, he says. A positive scan could raise a false red flag with these individuals. Plus, the florbetapir scan does not differentiate between Alzheimer’s pathology and pathology associated with other forms of dementia, such as Dementia with Lewy bodies or Parkinson’s.
However, a negative amyloid scan provides good precision that the cognitive decline is not caused by Alzheimer’s.
The key, according to Brody, is to put florbetapir imaging in clinical context. The procedure should not be done in isolation, but as a diagnostic adjunct.
Desert Medical Imaging in Indian Wells, Calif., was among the first to integrate florbetapir into its practice. Neurologist Christopher Hancock, MD, says early adoption was motivated by Phase III clinical trials attempting to identify treatment agents to retard or remove amyloid plaque.
“Florbetapir imaging is a component of our dementia work-up to help narrow differential diagnosis, which includes standard MRI brain studies, neuroquant volumetric analysis and normal pressure hydrocephalus spinal fluid quantification,” he says.
Impact on practice
Practice management uncertainties with florbetapir are deterring some providers from early adoption. The most obvious reason to wait is lack of reimbursement. Currently, neither the Centers for Medicare & Medicaid Services nor private payers reimburse for florbetapir scans.
“Once coverage begins, we’ll likely see more people asking for it or have a greater willingness to have it done,” Agronin says.
The Penn Memory Center at the University of Pennsylvania (UPenn) in Philadelphia postponed florbetapir imaging, partly because it wasn’t reimbursed.
“We’re not using florbetapir because it is not covered by insurance and we’re not going to charge patients,” says David A. Wolk, MD, neurologist at UPenn.
A florbetapir scan is relatively expensive. The wholesale price of a single dose of florbetapir costs approximately $1,600, not including scanner time and staff overhead. Some practices charge patients $3,000 or more.
Clinical issues also factored into UPenn’s decision. “Until we have a medicine that’s directed specifically at Alzheimer’s and treats the disease pathology, there will be hesitation about what you are getting out of this scan,” Wolk says.
Because there is no proven treatment if results are positive, Medicare is asking how the results of a florbetapir scan are going to change disease management.
The impact may be somewhat circuitous. More Alzheimer’s patients identified using florbetapir scans translates into more patients recruited for clinical trials. This may, in turn, hasten discovery of a disease-modifying treatment, says Brody.
Until reimbursement is settled, which is anticipated toward the end of 2013, the procedure is out of pocket for patients. Eli Lilly primes the imaging pump by providing vouchers for up to three free scans if imaging and memory centers agree to share diagnostic information with them.
The imaging provider RadNet adopted florbetapir PET despite lack of reimbursement.
“We wanted to get our 30 PET sites on board early to learn how to read scans correctly and learn how to manage florbetapir operational issues and patient scheduling, so if reimbursement comes, we’re ready to go,” says Judith G. Rose, MD, RadNet’s medical director of PET/CT.
RadNet developed a florbetapir-related marketing plan to promote diagnostic options to clinicians who handle dementia cases. “We started with our referral base of Alzheimer’s specialists and physicians currently referring dementia patients to us for F18-FDG screening,” Rose says.
Lack of a reader training program for radiologists interpreting florbetapir scans delayed FDA approval of florbetapir one year. The agency wanted Lilly to establish a program to ensure accuracy and consistency in reading and interpreting the new scans. Lilly responded by establishing a six-hour online interactive training program, and recommends in its package insert that physicians who interpret florbetapir scans take the course.
RadNet went a step further. Since florbetapir scans are binary, either positive or negative, Rose implemented a policy of primary and secondary reads for each scan. “Both radiologists must agree or the case comes to me to break the tie,” she says.
RadNet also set up a special central server so every radiologist who interprets florbetapir scans can look at every other case in the system to learn from colleague experiences. Without this, each florbetapir reader would see only the three studies performed at his or her site with the Lilly vouchers. Now, radiologists at each RadNet site push all three voucher studies to the central server so all RadNet radiologists have access to all 60 studies, instead of just the three performed locally.
Rose and other adopters of florbetapir expect early efforts to pay off as amyloid-related dementia therapies reach Phase III clinical trials.
According to the National Institutes of Health, as of 2012, more than 400 pharmaceutical Alzheimer’s treatments were being investigated in 1,012 clinical trials worldwide, approximately one-quarter of which are in Phase III trials.
Reduction of beta-amyloid levels in the brain is a common target of research compounds. University of California Los Angeles researchers, for instance, announced in November 2012, they had demonstrated for the first time that a compound called CLR01 safely crossed the blood-brain barrier and removed existing amyloid-beta and tau aggregates in a living Alzheimer’s mouse model.
As early adopters wrestle with clinical, fiscal and operational challenges, trials could help close the circle. Florbetapir is in a position to play an important role in future studies that address treatment challenges associated with cognitive decline once an amyloid-targeting therapy reaches Phase III trials and beyond.