The first amyloid imaging ligand that used fluorine 18, florbetapir F18 (Amyvid™), is under regulatory review in the U.S. and Europe. A company spokeswoman confirmed that she expects the Food and Drug Administration to act in the first half of 2012. At the 6th Human Amyloid Imaging Conference, held 12-13 January 2012 in Miami, Florida, researchers who had gathered primarily to share breaking research news all knew this, and spent time anticipating a world where amyloid PET is entering clinical practice beyond its current narrow use in research studies and therapeutic trials. Scientists discussed how this could be done in the best interest of people who do not have dementia but may be on the way. This story distills the main points of view expressed during a conference session that set aside formal data presentation to instead host an airing of the issues to come once amyloid imaging becomes more widely available.
This conference drew primarily scientists in academia, industry, and government, as well as some physicians in private practice. Many of the academic researchers have potential conflicts of interest with companies making tracers, scanners, or AD drugs. They agreed that amyloid imaging will be useful to identify people pre-dementia who are candidates for inclusion into anti-amyloid treatment trials. This was seen as true even though work remains to be done to tighten researchers' ability to predict when an amyloid-positive person will develop symptoms and which second and third hits—ranging from cerebrovascular disease to cognitive reserve to other factors—determine the rate of this progression. Screening for trials, however, is still a research use. The real open question to many clinicians at the conference was: How will amyloid imaging occur in routine clinical care?
Chris Rowe of the University of Melbourne, Australia, began the discussion, asking first, “Why do we want amyloid imaging?” For one thing, he said, because people want to know. At least in Rowe’s catchment area, community demand for diagnostic and prognostic information is strong. Studies have shown that people cope better when they have information and early diagnosis so they can plan. For another, it is often useful in diagnosis, Rowe said. Particularly for people with MCI, predictions of progression to AD are emerging from ongoing longitudinal studies with “very high consistency.”
Rowe foresees using amyloid imaging when he believes it will affect the management of a given patient’s disease. This means he would not use it for most patients with typical AD. Cognitive tests plus an MRI with automated hippocampal volume measurement are accurate for a typical 75-year-old patient with a medical history and clear dementia. However, if a patient is 60 and still employed, Rowe would use an amyloid scan. Ditto for a patient with suspected FTLD that looks as if it might be AD instead. In FTLD cases, amyloid PET has had no false-positive scans against postmortem pathology; hence, using it to clarify a patient’s diagnosis is reliable and would affect whether that person is offered AD drugs, Rowe said. Other examples where amyloid imaging could help in day-to-day diagnosis include distinguishing semantic dementia and vascular dementia from AD where, again, only the latter would be offered a prescription of the currently approved AD drugs. In subsequent discussion, there was broad agreement that amyloid PET can help confirm a non-AD dementia diagnosis.
Rowe advised great caution when determining whether a patient’s apparent dementia might be due to depression or early AD. Both amyloid deposition and depression are sufficiently common that it is possible amyloid might occur coincidentally in a person with severe depression. If overinterpreted, an amyloid scan in a depressed, cognitively impaired person could prompt a false diagnosis of AD. In these cases, an FDG-PET can help sort out the cause of the person’s suffering.
In summary, in MCI cases, Rowe would use amyloid PET to determine if patients are suffering from prodromal AD or if their impairment is due to other causes. In people with subjective memory complaints, Rowe would use amyloid PET if cognitive tests indeed showed abnormal findings. In cognitively normal people, Rowe would use amyloid imaging only to recruit for treatment trials. In all, a significant fraction of cases are uncertain enough to warrant amyloid imaging, Rowe said.
“Imaging has produced quite an armamentarium to help diagnose AD,” said Bill Jagust of the University of California, Berkeley. “This is an extremely expensive workup for a disease that has no good drugs. Will that happen?” The answer depends on who will pay for amyloid imaging. At HAI, some physicians predicted that it would only be widely used if government or private insurance covers it, whereas physicians from other areas, such as the host city of Miami, expected significant use for a one-time scan by people who pay out of pocket. In the U.S., the rollout of amyloid imaging could become a test case for the Affordable Care Act, said Jason Karlawish, of the University of Pennsylvania, Philadelphia.
Karlawish drew on examples from other diseases in articulating some of the ethical policy changes that arise when a disease advances from a clinical-pathological diagnosis to an "actuarial" diagnosis, where the doctor predicts future disease based on subtle clinical signs and biomarker findings. “We speak about the probability of a cardiovascular event and when to intervene to prevent it,” Karlawish said. He envisioned a day where the at-risk state of preclinical AD is as routine as cardiovascular risk calculation and preventive care. Many years of study, and successful disease-modifying treatment trials, lie between today and this future vision. In the meantime, Karlawish urged the HAI audience to respect the general public’s fear of Alzheimer’s, and to approach the transition of predictive biomarker assessments such as amyloid PET into the clinic with caution. Other scientists agreed, cautioning their colleagues to keep in mind that social consequences of a positive amyloid scan, such as problems obtaining insurance policies, should be addressed, not just a person’s ability to handle predictive health information. “We should not get swept up in our enthusiasm for the science,” said Mary Ganguli of the University of Pittsburgh Medical School, Pennsylvania.
How will amyloid imaging spread out from specialty settings? What administrative hurdles will—and should—physicians in various healthcare settings face to get their order for an amyloid scan approved? Will independent commercial clinics spring up that offer amyloid imaging and disclose results to cognitively normal people? Karlawish called on the U.S. network of federally funded Alzheimer's disease research centers to gather data on the impact of amyloid status disclosure. “As neurologists, we need to work out guidelines to protect people from adverse impact,” Karlawish said (see Part 3 of this series).
The influence of a well-meaning crop of research neurologists may be limited, cautioned Norman Foster of the University of Utah, Salt Lake City. “Right now, most people even with dementia, much less cognitively normals, never see a specialist, just their PCP. Most PCPs will not go through a complicated process of evaluating whether and how to disclose. We need clear-cut guidelines.”
As a cautionary tale of the stark difference between the careful use in differential diagnosis in specialized dementia care that Rowe had outlined and the present-day reality of primary care, Foster told of a patient who had come from a PCP with a dementia diagnosis which the patient questioned, and insisted on an amyloid test. Foster refused, assessing the patient clinically and with cognitive tests. His cognition proved to be intact and his "forgetfulness" turned out to have been due to hearing loss. A positive amyloid scan in a primary care or commercial setting could have—either incorrectly or at least prematurely—stuck this person with a diagnosis of Alzheimer’s dementia, Foster said. “Because PCPs are likely to order a scan without determining that the patient has a dementing disease, depression, or a medication-related impairment, we ought to decide what is medically advisable, not wait for society to decide whether they will pay for it,” Foster said.
Others at HAI agreed that, as with any new technology, there was potential for abuse, but emphasized the value to the patient of getting a proper differential diagnosis. “When patients get a diagnosis of MCI, it often leaves them frustrated. The percentages of progression do not satisfy them. Many would prefer to know with greater certainty whether they are dealing with AD, so that they and their families can prepare for the future. We have to factor in the benefit to the patient of a clear diagnosis,” said Gil Rabinovici of the University of California, San Francisco.
To date, an amyloid scan would clarify the diagnosis in some cases (e.g., FTD vs. AD for a person with mild dementia), while in others it would still leave the patient with essentially a risk assessment (e.g., high risk of progressing to AD dementia for an MCI patient), cautioned Bart van Berckel at the Vriije University Medical Center in Amsterdam, The Netherlands.
Finally, the 200+ scientists debated how to interpret weakly positive cases whose amyloid level hovers around the threshold of positivity. In routine clinical practice, physicians will likely use a binary read, which amounts to a thumbs up/down rather than an absolute quantitative measurement of amyloid load (see Part 4 of this series). Stephen Salloway, who generally supports the use of amyloid imaging in the clinic, noted that in some of those borderline cases, the amyloid could be a secondary pathology, not necessarily the reason for the clinical finding at hand. The growing incidence of amyloid positivity with age makes this quite possible. For these cases, a diagnostic hierarchy of how the physician weighs the individual pieces of the patient’s workup remains to be established.
Clinicians in the room easily, perhaps predictably, agreed on one thing: If cost were no limiting factor, they would prefer to see not just an amyloid scan, but also an FDG and MRI scan in many questionable cases.—Gabrielle Strobel.
- Miami: Scan and Tell? Amyloid Imaging Confronts Disclosure Dilemma
- Miami: Can the Naked Eye Tell When a Scan Is Positive?
- News Focus: 2012 Human Amyloid Imaging Conference
- Miami: When Does Amyloid Deposition Start in Familial Alzheimer’s?
- Miami: Age and Amyloid—What Has ApoE Got to Do With It?
- Miami: Longitudinal Amyloid PET Data Start Converging
- Miami: Diagnosis and Amyloid Scan Can Be at Odds
- Miami: Scientists Angle for Way to Image Tangle