Researchers who have spent their careers studying schizophrenia and mood disorders might be forgiven a bit of “biomarker envy.” At long last, it seems that the neurologists and neuropsychiatrists have developed some fairly sensitive and specific “lab tests” for Alzheimer’s Disease (AD). As reported in The New York Times,1,2 the most promising new test uses a radioactive dye that attaches to AD-related plaque in the brain, allowing it to be seen with a PET scan. In theory, this test could be used “…to detect the disease even before there are evident memory problems or other symptoms.”2,italics added The implications for public health are enormous. As Dr P. Murali Doraiswamy put it, this development will encourage much more testing for AD, and “…diagnosis rates, like testing rates, only go in one direction—up.”2 This candid acknowledgment ought to remind vociferous critics of psychiatry that other medical specialties, too, are grappling with the problem of “where to draw the line” between normality and pathology.
And is this near-certain increase in diagnosis of AD necessarily a good thing? My colleague, Dr Allen Frances, has spent much of the last year warning us of the dangers of “over-diagnosis,” “false positives,” and the “medicalization” of normality, in several proposed DSM-5 diagnoses—a view that has not gone unchallenged by this writer and others.3,4 Nevertheless, in this case, I wonder if our colleagues in neurology and neuropsychiatry are heading down the path of unhelpful “labeling,” if not over-diagnosis. I raise this question even while acknowledging the potential benefits of early diagnosis of AD and the conceptual advantages of unifying gradual neuropathological changes under one diagnostic “umbrella.”
To be sure, there is a difference between the situation Dr Frances describes in most of clinical psychiatry—where valid biomarkers are, with rare exception, almost universally lacking—and the situation we now face with AD. In theory, once we can agree on a set of diagnostic criteria for “probable” AD—and assuming the new test for AD pans out—we can then determine whether the new, plaque-based biomarker helps us predict onset, course, severity, and response to treatment in AD. In principle, we could compute the actual number of “false positives” for the plaque test by comparing PET findings with clinical data, such as the patient’s cognitive function, ability to carry out activities of daily living, etc.
In this new context, the term “over-diagnosis” could acquire a relatively objective—but curiously “bidirectional”—meaning. On the one hand, one might argue that a patient with AD-type symptoms (eg, memory loss, inability to calculate, etc) but a normal PET scan has been “overdiagnosed” (or misdiagnosed) if he is labeled as having AD. Conversely, one could argue that a patient with no cognitive impairment, but with a markedly abnormal PET scan, has also been “over-diagnosed,” if he is given a diagnosis of “Preclinical Alzheimer’s Disease”—and therein lies the conceptual and clinical conundrum.
Let’s take the conceptual problem first—a problem with an ancient and contentious history. As far back as classical times, there has always been a tension between those who would diagnose disease solely on the basis of “objective” pathological findings; and those, like me, who view “disease” as a condition defined by the presence of intrinsic suffering and incapacity.5 Disease, in my view, is a construct properly attributed to persons –not to minds, bodies, organs or tissues. In my diagnostic world, “no suffering and incapacity” means no dis-ease (for lack of ease is how our word “disease” originated). An otherwise healthy person with abnormal plaques on his PET scan, but with only mild memory difficulties, is not “diseased,” in my book. Perhaps we need a new term of art to describe the condition of such high-functioning “PET scan outliers.”
Draft reports from workgroups at the Alzheimer's Association International Conference on Alzheimer's Disease (AAICAD) 2010 suggest that the term “preclinical Alzheimer’s Disease” may be adopted6—but can “disease” as we ordinarily understand the term really be “pre-clinical”? And what will be the psychological effect of telling an asymptomatic patient, “You have preclinical Alzheimer’s Disease”? After all, it is not yet clear that all patients with abnormal brain amyloid deposits will go on to develop full-blown Alzheimer’s Dementia.6 I might respectfully suggest the designation, “Abnormal Amyloid Syndrome” for individuals with pathological PET findings but no clinical evidence of significant cognitive impairment. This is actually similar to the term “asymptomatic cerebral amyloidosis”—intended for research, not clinical, purposes—from the AAICAD work group.6
Now to the clinical problem: let’s say that Mr. Jones, a 70-year-old, college-educated businessman, complains to his primary care physician that “My memory is slipping.” A psychiatric exam reveals generally normal cognitive function, with some minor difficulties recalling recent information and a “borderline” Mini-mental state examination (MMSE) score of 24.7 Let’s suppose that Mr. Jones is still functioning well at home, except for occasionally misplacing his keys or forgetting an appointment. However, a PET scan ordered by the psychiatrist is fully consistent with AD. What, exactly, do we advise Mr. Jones and otherwise “normal” individuals with abnormal PET scans? And what “treatment”—if any—should we recommend?
Given that, at present, we have no treatments that can halt or reverse the progression of AD neuropathology, will we not create unnecessary anxiety in thousands of otherwise “normal” individuals, by carrying out widespread PET scanning for minor memory impairment? Furthermore, what if large-scale testing reveals that 20% or 40% of the general population over age 65 years have abnormal plaques on their PET scans? Will we be doing more harm than good by telling such individuals, as one expert suggested, “You are on the Alzheimer road”?2 As for treatment, currently available medications for AD may slow the rate of decline, and delay nursing home placement—but they do not have robust clinical efficacy and do nothing to stop the inevitable down-hill slide of the illness. As Dr Doraiswamy put it, “…we ought to be cautious that we don’t stimulate all this testing before we can give people something to manage their disease. There is no point in giving them just a label.”2 Of course, if drugs are developed that actually alter the course of AD—and we are making great strides in this arena—the test in question could prove a tremendous boon. [See comments from Dr Sandra Jacobson, below]
There are no easy answers to the questions raised by this new diagnostic test for AD, but we can draw some parallels with similar conundrums in the field of psychiatry. In general, I have argued against a very fine-grained, “consequentialist” approach to psychiatric diagnosis; for example, tailoring our diagnostic criteria so as to head off anticipated patterns of over-zealous drug prescribing or pharmaceutical marketing.8,9 I generally believe that we need to “follow the science” and deal with such adverse consequences (if they occur) through intense educational efforts. For example, I support the call by my colleague, Nassir Ghaemi MD, to eliminate the “4-day rule” for the diagnosis of hypomania, based on the best available evidence.10 But I concede that my support is based, in part, on the availability of effective treatment for bipolar disorder—including one medication (lithium) that actually decreases suicide risk in bipolar patients.11
Would I support the “2-day” criterion for hypomania, even if we did not have good treatment available? Yes, probably—after all, as Dr Ghaemi reminds us, “Limitation in treatment does not justify ignorance of disease.”10 But when it comes to telling a person with no cognitive impairment and an abnormal PET scan that he is “on the Alzheimer road,” my inclination to “follow the science” comes face-to-face with the ancient admonition to “do no harm.”
Comments from Sandra A. Jacobson, MD
We are entering a new era of AD treatment involving specific therapeutic targets (for now, mostly steps in the amyloid cascade). Available data suggest that effective intervention is early intervention, certainly before function is affected, and before gross screening measures (eg, MMSE) are able to detect change, and possibly long before any symptoms. Over the last few years, AD immunotherapies—both active and passive—have progressed to the point that clinical use is probably on the near horizon. In addition, treatments like secretase inhibitors are moving forward. It looked for awhile like we would have effective treatments, but still weren't going to be very good at pre-symptomatic diagnosis. Then Pittsburgh compound B Positron emission tomography (PiB-PET) came along, and then other amyloid-binding ligands, and in a very short time, early diagnosis was revolutionized. So this is all coming to a head very quickly. It looks like we will soon be involved in a massive preventive medicine program.
We have had to deal with the issue of increased AD risk in asymptomatic patients with Apolipoprotein E (APOE) genotyping. This is a difficult area, because increased risk does not equal disease. The amyloid imaging, on the other hand, shows what is now considered a major component of the underlying pathology of AD, the amyloid burden. What needs to be sorted out is how much amyloid is “normal.” I expect that the field will have a handle on this shortly, as the cross-sectional studies are quick, and many are going on simultaneously. The longer-term follow-up of course remains to be accomplished.
Dr Jacobson is Staff Physician and Clinical Investigator at Banner Sun Health Research Institute in Sun City, AZ. She is the author (with I. Katz, MD, and R. Pies, MD) of Clinical Manual of Geriatric Psychopharmacology (Arlington VA: American Psychiatric Publishing; 2007).
1. Kolata G: New Scan May Spot Alzheimer’s. The New York Times. July 12, 2010. http://www.nytimes.com/2010/07/13/health/research/13alzh.html.
2. Kolata G: Rules Seek to Expand Diagnosis of Alzheimer’s. The New York Times. July 13, 2010. http://www.nytimes.com/2010/07/14/health/policy/14alzheimer.html.
3. Frances A. Normality Is an Endangered Species: Psychiatric Fads and Overdiagnosis. Psychiatric Times. http://www.psychiatrictimes.com/dsm-5/content/article/10168/1598676.
4. Frances A, Pies RW, Zisook S. Debate: DSM5 and the Medicalization of Grief: Two Perspectives. Psychiatric Times. http://www.psychiatrictimes.com/display/article/10168/1568760.
5. Pies R. What Should Count as a Mental Disorder in DSM-V? Psychiatric Times, April 14, 2009. Psychiatric Times. http://www.psychiatrictimes.com/display/article/10168/1402032.
6. Jeffrey S. Redefining Alzheimer's Disease: NIA and Alzheimer's Association Float New Draft Diagnostic Criteria. Medscape Today. http://www.medscape.com/viewarticle/725177.
7. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state: A practical method for grading the cognitive state of patients for the clinician.” J Psychiatr Res. 1975;12:189-198.
8. Pies R. Consequentialism in psychiatric diagnosis. (letter) Reply to Parens et al. N Engl J Med. In press.
9. Pies R. The ideal and the real: how does psychiatry escape the DSM-5 “fly-bottle”? Bulletin of the Association for Advancement of Philosophy and Psychiatry. In press.
10. Ghaemi SN. DSM 5 and Bipolar Disorder: Science versus politics. http://www.psychologytoday.com/blog/mood-swings/201004/dsm-5-and-bipolar-disorder-science-versus-politics.
11. Tondo L, Baldessarini RJ. Long-term lithium treatment in the prevention of suicidal behavior in bipolar disorder patients. Epidemiol Psichiatr Soc. 2009;18:179-183.