The earliest maps of a new land, when viewed beside contemporary maps, often have the look of a child's drawing. The borders are rough, whole regions are missing, others misrepresented. If people followed such a map, they might get terribly lost. And yet, some map is better than none at all.
This week, at the Alzheimer's Association International Conference in Vancouver, researchers from the Mayo Clinic reported results of their study of potential biomarkers for preclinical Alzheimer's disease. "Biomarker," a relatively new term in medicine, is a kind of shorthand for any of the various hints that clue us in to the complex ways in which a disease develops. In Alzheimer's disease research, the most studied biomarkers include MRI and PET scan images of the brain.
Their results suggest that among cognitively normal older adults there is a potentially large population who occupy an uncharted territory. Their biomarker results are neither normal, nor clearly abnormal. Dr. David Knopman, the investigator who presented the results, admitted that he and his colleagues "hadn't expected to encounter this result."
One of the biomarkers in question comes in the form of MRI images that show parts of the brain that are atrophied -- areas in which the death of brain cells has "shrunk" the brain. The other set of biomarkers involves two kinds of PET images -- one that measures brain metabolism and the other measures amyloid plaques, which are dense deposits of protein believed to be the "signature" of Alzheimer's disease.
Researchers have embraced these biomarkers because, although they have not discovered what causes this complex disease that develops over a lifetime, they are convinced that discovering its biomarkers is the key to transforming the diagnosis. They hope these advances mean we may one day go from relying on a physician's history and physical exam to determine whether a patient has dementia to a continuum that a physician can diagnose even before a person is ill simply by measuring the presence of so-called "biomarker signatures."
The Mayo Clinic researchers focused on the proposed biomarkers of preclinical Alzheimer's disease, a developing concept that describes individuals who are outwardly normal, but who have the amyloid plaques seen in Alzheimer's disease. Researchers are guessing that, if followed over time, these people will progress to mild cognitive impairment -- notable memory loss with retained ability to engage in day-to-day tasks -- and, ultimately, dementia. Their goal was to use the MRI and PET biomarker results to classify their subjects into the proposed stages of preclinical Alzheimer's disease, follow them for at least a year, and then restage them.
Preclinical Alzheimer's disease is truly at the uncharted frontier between what is a normal versus a diseased brain, a frontier that entices us to explore it. Discovering its biomarkers promises to reveal an early diagnosis and interventions to prevent cognitive impairment. But exploring it is not without peril. If we discover that biomarkers behave in ways that do not fit with what we expected, then we risk labeling otherwise healthy older adults with something we do not understand.
The Mayo Clinic researchers looked at a group of cognitively normal adults aged 70 and older. About half of them lacked any of the biomarkers they were looking for. They excluded these cases sorted their remaining subjects into three groups. The first group, which they called "stage 1," had evidence of amyloid plaques when given a PET scan. The second group, "stage 2," had these plaques, plus signs of brain atrophy. The third group, called "stage 3," showed the same plaques and brain degeneration seen in stage 2, but they also shows signs of cognitive changes.
The researchers' assumption was that the stages of preclinical Alzheimer's disease follow a simple stage-by-stage model, meaning, a person moves in a straight line from one stage to the next, beginning when amyloid plaques start to form, progressing to neurodegeneration as seen on MRI or PET scan, and then to subtle cognitive changes.
And that is what they discovered. Their subjects fit into one of the stages and, over time, they moved stage-by-stage as predicted. Over time, nearly half developed cognitive problems that were severe enough a physician diagnosed them with mild cognitive impairment, a condition widely believed to represent an intermediate state between normal aging and dementia.
But the researchers also discovered unexpected territory. As many as one quarter of their subjects did not fit onto their staging map. Instead, they had a mixed biomarker profile. Their brains were not "normal" but they were not clearly "abnormal." These older adults had MRI or metabolic PET scan biomarkers that showed neurodegeneration of an Alzheimer's pattern, but they did not have the pathologic hallmark of Alzheimer's disease. Namely, a PET scan that detects amyloid was entirely normal, a finding that researchers described as "remarkable."
What do these people have? Are they diseased? Or is this a variation of normal aging? The investigators labeled these people with MRI or PET scans seen in persons with Alzheimer's disease but without amyloid plaques as having "sNAP," shorthand for "Suspected nonAmyloid Pathway," a name that reveals that we know more about what disease these people do not have than we know about what disease, if any, they do have.
They did not have other common causes of neurodegeneration, including brain vascular disease and Parkinsonism. They were less likely than the subjects who fit into one of the three stages of preclinical Alzheimer's disease to have an APOE4 gene, one of most widely recognized genetic risk factors for developing Alzheimer's disease dementia. Over time, they were no more likely than persons who had no biomarkers to experience cognitive decline.
Persons with this curious mix of some but not all of the biomarkers of Alzheimer's disease warn us that if we rush to diagnose Alzheimer's disease as early as possible, we could leave as many as one-quarter of our patients in a kind of diagnostic limbo. They also challenge the very foundation of the amyloid hypothesis of Alzheimer's disease. How can someone have an MRI or metabolic PET scan that looks like Alzheimer's disease, and yet not have amyloid plaques on PET imaging?
This result recalls other anomalous research results that were in fact the key to launching scientific revolutions, such how the mysterious black-body radiation inaugurated quantum physics. To ignore persons with "Snap" or otherwise brush them under the rug is to miss an opportunity. They may well be the secret to figuring out not only what Alzheimer's disease is, but what a normal brain is as well.
Dr. Jason Karlawish is professor of Medicine, Medical Ethics and Health Policy at the University of Pennsylvania's Perelman School of Medicine.