The recent paper by Buchhave and colleagues in the Archives of General Psychiatry is, to my knowledge, the first report of relatively long-term (about nine years) clinical follow-up of an MCI cohort with baseline CSF biomarker data. Their initial 2006 paper (Hansson et al., 2006) demonstrating the high prognostic utility of the ratios of CSF tau(s)/Aβ42 in predicting clinical “conversion” from MCI to DAT (over about five years) in these same individuals was very important to the field, and set the stage for longer-term follow-up, which they have now been able to achieve. As expected, a greater percentage of individuals “converted” from MCI to AD over nine years (53.7 percent) compared to over five years (42 percent), and the combination of tau(s) and Aβ42 measures showed a greater positive predictive value over the longer- compared to the shorter-term follow-up (91 percent versus 81 percent, respectively), supporting the notion of AD as a progressive disease with a long developmental time course. Our group (Fagan et al., 2007) and the group at the University of Washington (Li et...  Read more

The recent paper by Buchhave and colleagues in the Archives of General Psychiatry is, to my knowledge, the first report of relatively long-term (about nine years) clinical follow-up of an MCI cohort with baseline CSF biomarker data. Their initial 2006 paper (Hansson et al., 2006) demonstrating the high prognostic utility of the ratios of CSF tau(s)/Aβ42 in predicting clinical “conversion” from MCI to DAT (over about five years) in these same individuals was very important to the field, and set the stage for longer-term follow-up, which they have now been able to achieve. As expected, a greater percentage of individuals “converted” from MCI to AD over nine years (53.7 percent) compared to over five years (42 percent), and the combination of tau(s) and Aβ42 measures showed a greater positive predictive value over the longer- compared to the shorter-term follow-up (91 percent versus 81 percent, respectively), supporting the notion of AD as a progressive disease with a long developmental time course. Our group (Fagan et al., 2007) and the group at the University of Washington (Li et al., 2007) have shown that these same markers are useful for predicting (within three to five years) future MCI/dementia of the Alzheimer’s type in older individuals while they are still cognitively normal. Together, these data demonstrate that AD pathology begins to develop very early in the disease process, even prior to any cognitive symptoms. Such findings have significant ramifications for the design and evaluation of AD prevention trials of disease-modifying therapies (with the goal of preventing cognitive impairment as opposed to slowing or halting already existing deficits).

Importantly, whereas CSF Aβ42, tau and phospho-tau (p-tau) in the present study were associated with “conversion to AD” within the first five years of baseline, Aβ42 was the only marker that was associated with conversion in individuals followed for five to 10 years. These data lend further support for the proposed trajectory of such biomarker changes (Perrin et al., 2009; Jack et al, 2010; Jack et al., 2011): CSF Aβ42 dropping very early in the disease process (and plateauing) in concert with its deposition as amyloid plaques, followed by elevations in CSF p-tau and tau (and atrophy), markers of tangles and neurodegeneration, that continue as the disease progresses. The ultimate test of this hypothesis will require not only longitudinal clinical follow-up in individuals while they are still cognitively normal (preclinical/presymptomatic stage), but also assessment of biomarker changes over time within such individuals. Such studies are currently underway in both sporadic AD (e.g., The Adult Children Study [ACS], Biomarkers for Older Controls at Risk for Dementia [BIOCARD], Wisconsin Registry for Alzheimer Prevention [WRAP]), as well as genetic forms of AD (e.g., The Dominantly Inherited Alzheimer Network [DIAN], The Alzheimer’s Prevention Initiative Biomarkers Project [API-BIO]).

The data presented in the paper by Buchhave et al. are important beyond their face value. Not only do they support the proposed trajectory of specific biomarker changes very early in the course of the disease, but they also provide support for the importance for assigning an underlying etiology to the construct of MCI (Albert et al., 2011). The data demonstrate that individuals classified as being “MCI” and who have biomarkers supportive of AD (i.e., low Aβ42 and elevated tau(s) and the tau(s)/Aβ42 ratios) actually have AD, but that the clinical manifestations of the disease pathology have not yet become severe enough for some to call it “dementia.” Thus, these individuals aren’t “converting” from MCI to AD; they are not merely at risk for AD: they have AD and are progressing in their pathological and clinical course as one would expect. While this may seem like merely splitting semantic hairs, the AD lexicon needs to reflect our understanding of the protracted and progressive nature of the disease pathological processes which eventually culminate in more significant cognitive impairment (dementia). The data from the current paper have contributed nicely to this evolving concept, and I predict that longer clinical follow-up in cognitively normal individuals who exhibit AD biomarker profiles will provide support for the utility of such biomarkers to identify individuals with the disease even earlier, before the onset of any clinical symptoms (Sperling et al., 2011). Stay tuned….

References:
Albert M, DeKosky S, Dickson D, Dubois B, Feldman H, Fox N, Gamst A, Holtzman D, Jagust W, Petersen R, Snyder P, Phelps C: The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimer’s & Dem, 2011, 7:270-279. Abstract

Fagan AM, Roe CM, Xiong C, Mintun MA, Morris JC, Holtzman DM. Cerebrospinal fluid tau/β-amyloid42 ratio as a prediction of cognitive decline in nondemented older adults. Arch Neurol, 2007, 64:343-49. Abstract

Hansson O, Zetterberg H, Buchhave P, Londos E, Blennow K, Minthon L. Association between CSF biomarkers and incipient Alzheimer's disease in patients with mild cognitive impairment: A follow-up study. Lancet Neurol, 2006, 5:228-234. Abstract

Jack C, Knopman D, Jagust W, Shaw L, Aisen P, Weiner M, Petersen R, Trojanowski J. Hypothetical model of dynamic biomarkers of the Alzheimer's pathological cascade. Lancet Neurol, 2010, 9:119-28. Abstract

Jack C, Vemuri P, Wiste H, Weigand S, Aisen P, Trojanowski J, Shaw L, Bernstein M, Petersen R, Weiner M, Knopman D. Alzheimer’s Disease Neuroimaging Initiative. Evidence for ordering of Alzheimer disease biomarkers. Arch Neurol, 2011, 68:1526-35. Abstract

Li G, Sokal I, Quinn J, Leverenz J, Brodey M, Schellenberg G, Kaye J, Raskind M, Zhang J, Peskind E, Montine T. CSF tau/Aβ42 ratio for increased risk of mild cognitive impairment: A follow-up study, Neurology, 2007, 69:631-39. Abstract

Perrin RJ, Fagan AM, Holtzman DM. Multimodal techniques for diagnosis and prognosis of Alzheimer’s disease. Nature, 2009, 461:916-22. Abstract

Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, Iwatsubo T, Jack CR Jr, Kaye J, Montine TJ, Park DC, Reiman EM, Rowe CC, Siemers E, Stern Y, Yaffe K, Carrillo MC, Thies B, Morrison-Bogorad M, Wagster MV, Phelps CH. Toward defining the preclinical stages of Alzheimer's disease: Recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimer’s & Dem, 2011, 7:280-92. Abstract

View all comments by Anne Fagan