MRI measures of entorhinal cortex vs hippocampus in preclinical AD

Killiany et al. report that volumetric measures of the entorhinal cortex and, to a lesser extent, the hippocampus discriminate those elderly individuals in the CDR 0.5 stage of cognitive impairment who progress to greater impairment from those who remain at the CDR 0.5 stage. [1] The authors propose that MRI measures detect preclinical pathological change associated with Alzheimer disease (AD) and further suggest that AD pathology begins in the entorhinal cortex and later spreads to the hippocampus. [1] We applaud the careful work of this group, although the basis for their conclusion about the temporal spread of pathology can be challenged: because only a baseline MRI was obtained, the region in which volume loss first occurred cannot be resolved in this study.

However, we agree that medial temporal lobe structures are most vulnerable to the earliest pathological changes of AD. We find in our clinicopathologic studies that AD lesions develop initially in the entorhinal cortex and ultimately may be present in sufficient densities throughout the neocortex to meet criteria for neuropathological AD, even in the absence of dementia. [2] We also have demonstrated that substantial entorhinal and hippocampal cortical neuronal loss occurs in CDR 0.5 individuals in comparison with nondemented elderly controls. [3] Correspondingly, our MRI studies have shown that hippocampal measures of volume and shape discriminate CDR 0.5 individuals from nondemented individuals. [4]

We differ with Killiany et al. on two major issues. We use the term, "preclinical AD", to indicate the presence of disease before it is clinically recognizable. This term characterizes cases with neuropathological AD in the absence of cognitive symptoms, impairment or decline. [5] Preclinical AD thus is prior to the CDR 0.5 stage. In contrast, Killiany et al. appear to use "preclinical AD" to designate CDR 0.5 individuals who already are symptomatic by virtue of their memory complaints. At least 21 of their cases progressed in CDR severity over 3 years, suggesting that the baseline memory complaints for these individuals reflected the initial clinical expression of AD. If so, these individuals might more accurately be considered to represent very early or very mild AD rather than "preclinical AD". Postmortem studies performed by our group show that individuals with truly preclinical AD lack substantial neuronal or volume loss in the entorhinal cortex or hippocampus, [3] suggesting that volumetric measures of medial temporal lobe structures would be unlikely to detect such cases.

It also appears that Killiany et al. operationalized the diagnosis of AD as equivalent to reaching the CDR 1 stage. This approach will miss many individuals who develop diagnosable AD while still in the CDR 0.5 stage. [6, 7] Our experience indicates that pathologically verified AD can be clinically diagnosed even in CDR 0.5 individuals with a mean Mini- Mental State Examination score of 29, [7] comparable to the "questionable" group of Killiany et al. Because they apparently do not recognize these early AD cases, the entorhinal measures reported by them may simply detect AD individuals within the CDR 0.5 stage who are more severely affected than others. [1] These individuals would be most likely to progress to a greater stage of AD severity (i.e. CDR 1) over a defined period, whereas less affected CDR 0.5 individuals with AD may require longer periods to progress. Rather than the presence or absence of disease, the MRI entorhinal measures may be detecting individuals who are relatively more advanced, and hence more likely to progress, within this group of very mildly impaired individuals. Should effective disease-modifying treatments be developed for AD, we suggest that it will be important that imaging and other surrogate markers identify the even less affected individuals with AD, as well as those at the earlier stage of preclinical AD, to permit optimal benefit of the interventions.

References

1.Killiany RJ, Hyman BT, Gomez-Isla T, Moss MB, Kikinis R, Jolesz F et al. MRI measures of entorhinal cortex vs hippocampus in preclinical AD. Neurology 2002; 58:1188-1196.

2.Price JL, Morris JC. Tangles and plaques in nondemented aging and "preclinical" Alzheimer's disease. Ann Neurol 1999; 45:358-368.

3.Price JL, Ko AI, Wade MJ, Tsou SK, McKeel DW, Jr., Morris JC. Neuron number in the entorhinal cortex and CA1 in preclinical Alzheimer disease. Arch Neurol 2001; 58:1395-1402.

4. Csernansky JG, Wang L, Joshi S, Miller JP, Gado M, Kido D et al. Early DAT is distinguished from aging by high dimensional mapping of the hippocampus. Neurology 2000; 55:1636-1643.

5. Goldman WP, Price JL, Storandt M, Grant EA, McKeel DW, Jr., Rubin EH et al. Absence of cognitive impairment or decline in preclinical Alzheimer's disease. Neurology 2001; 56:361-367.

6. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment. Arch Neurol 1999; 56:303-308.

7. Morris JC, Storandt M, Miller JP, McKeel DW, Jr., Price JL, Rubin EH et al. Mild cognitive impairment represents early-stage Alzheimer's disease. Arch Neurol 2001; 58:397-405.

The accompanying letter by Morris et al. raises three independent but interrelated issues important to those studying prodromal Alzheimer's disease (AD): (1) the nature of pathological change seen earliest in the course of those destined to develop AD, (2) whether MRI measures have been identified that parallel this pathology, and (3) the nomenclature used to describe individuals in the prodromal phase of AD.

Pathological studies have consistently reported that the parahippocampal gyrus is affected early in AD and most studies show that entorhinal (and transentorhinal) cortex layer II neurons develop tangles even before the CA1/prosubicular neurons in the hippocampus. [1. 2] Both sites are affected early in the disease process without doubt.

Quantitative measures from MRI scan now clearly show in vivo what neuropathological analyses have suggested from cross-sectional studies, i.e. that brain changes occur in specific brain regions long before overt dementia. The most commonly reported finding is that the entorhinal cortex and hippocampus show decreased volume among non-demented individuals with cognitive problems. [3. 5] Moreover, recent longitudinal data demonstrate a greater change over time in the entorhinal cortex than in the hippocampus during the prodromal phase of AD. [4] consistent with the pathological findings.

The nomenclature used to describe individuals in the prodromal phase of AD is varied. In addition, the populations of non-demented individuals with cognitive problems that have been studied differ in their range of impairment. We use the term 'preclinical' to refer to individuals who do not yet meet clinical criteria for probable AD. These individuals clearly have clinical expression of AD, even though they are not demented. In our recent MRI study, there were 21 such 'questionable' subjects who subsequently progressed to overt dementia. We used the Sum of Boxes from the CDR scale to quantify their degree of functional difficulty at baseline. Of the 21 'questionable' subjects (mean Sum of Boxes = 1.8), one had a Sum of Boxes of 3.0, while all the others had a Sum of Boxes of 2.5 or lower. None had a box score of 3.5 or higher. Thus, our evaluation of these individuals indicated that they were not demented at the time the baseline MRI scans was obtained. Clearly, it will be important for groups studying such individuals to reach a consensus on terminology and methodology in order to facilitate examination of individuals with prodromal AD.

References

1. Braak H, Braak E. Neuropathological staging of Alzheimer-related changes. Acta Neuropathol (Berl). 1991;82:239-259.

2. Arriagada PV, Growdon JH, Hedley-Whyte ET, Hyman BT. Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer's disease. Neurology, 1992; 42:631-639.

3. Dickerson BC, Goncharova I, Sullivan MP, Forchetti C, Wilson RS, Bennett DA, Beckett LA, deToledo-Morrell L. MRI-derived entorhinal and hippocampal atrophy in incipient and very mild Alzheimer's disease. Neurobiol Aging, 2001;22:747-754.

4. Du A, Schuff N, Zhu X-P, Jagust W, Chui H, Weiner M. Larger rates of atrophy of entorhinal cortex than of hippocampus in Alzheimer's disease. 8th Scientific Meeting of the International Society of Magnetic Resonance in Medicine; Honolulu, 2002.

5. Xu Y, Jack CR Jr, O'Brien PC, Kokmen E, Smith GE, Ivnik RJ, Boeve BF,Tangalos RG, Petersen RC. Usefulness of MRI measures of entorhinal cortex versus hippocampus in AD. Neurology, 2000;54:1760-1777.