my personal edition > alzheimer's > news


  E-Mail this DGDispatch to a colleague

DGDispatch

Does Telomere Shortening Predict Dementia? A Critical Review: Presented at AAT

By Chris Berrie

GENEVA, SWITZERLAND -- April 24, 2006 -- Despite previous mixed reports, telomere length cannot be used to predict dementia, according to the largest longitudinal study performed in demented and normal subjects and presented here at the 9^th International Geneva/Springfield Symposium on Advances in Alzheimer Therapy (AAT).

Principal investigator Dina Zekry, MD, PhD, associate professor and head, somatic dementia unit, department of rehabilitation and geriatrics, Geneva University Hospital, Thônex, Switzerland, discussed the study findings in a presentation on April 20^th.

"The telomeres are short, repeated DNA sequences, and each time a cell makes a copy of itself, some of these repeats are lost," Dr. Zekry said. From starting levels of some 8,000 base pairs, 20 to 200 are lost during each cellular replication, she explained.

The importance of these telomeres lies in their multiple roles: they protect against degeneration, they prevent chromosome fusion, and thus cell death and neoplastic transformation, and they are essential for chromosomal stability, she indicated. "The telomere hypothesis of aging is based on the notion that telomere shortening with each cell division, and therefore with age, results in cell senescence," Dr. Zekry said.

As senescence and senescent cells contribute to aging, telomere length might be considered as a potential biomarker of aging, she added. This thus led to the "telomere hypothesis of dementia," that asks whether telomere shortening contributes to the genesis of certain age-related diseases, such as dementia, she said.

Cross-sectional studies in small cohorts of young patients and using the Mini-Mental State Examination (MMSE) for cognitive function have indicated a significant relationship between telomere length and cognitive function, Dr. Zekry said. Conversely, in a larger study in an older population, telomere length was not predictive for dementia, again according to MMSE alone, "Which is neither sensitive nor specific for detecting changes in a population of this great age," she added.

Therefore, Dr. Zekry and colleagues conducted a longitudinal study with 1-year follow-up, aiming to determine if telomere length would predict dementia in very elderly patients, conversion of mild cognitive impairment to dementia, and degree of impairment or severity of dementia.

They performed a battery of tests on subjects, including clinical and education-level evaluation, dependence scores, clinical criteria scores, and psychometric and neuropsychological assessment. Telomere length was determined in peripheral blood lymphocytes by flow cytometry.

The researchers enrolled a sample of patients (n = 449; mean age, 85.1 years; male, 26.3%) from admissions to their Geriatric Hospital over a 2-year period, with exclusion specifically of severe disorders affecting psychometric assessment. These patients were matched against 39 community-dwelling healthy controls (mean age, 75.4 years; male, 30.6%), who, upon analysis, showed a significant decrease in telomere length with age (P =.003).

Of the hospital admissions group, 205 (45.7%) were cognitively normal, 49 (10.9%) showed mild cognitive impairment (MCI), and 195 (43.4%) were classified as having dementia. Of patients in this last group, 42% showed mixed dementia, 39.5% showed Alzheimer's disease, 11% showed vascular dementia, and 7.5% showed other dementias. There were no significant differences in age and sex between these hospitalised groups.

Telomere lengths in both the dementia and nondementia patients showed no significant decline with age. Indeed, when the higher age range of 75 to 92 years was specifically considered, telomere length of these hospitalised patients was significantly greater than that of the healthy controls (P =.033); this difference was not dependent on cognitive status.

However, within the cognitively normal patients with either MCI or dementia, there were no significant differences seen in telomere length. There was also no correlation between telomere length and MMSE score. Similarly, telomere length did not differ significantly according to the aetiology of the dementias, or to the different degrees of dementia severity (according to clinical dementia rating scores [CDRs]). Multiple linear regression analysis also showed that age, sex, and cognitive status do not predict telomere length.

At the 1-year follow-up (n = 224; mean age, 84.7 years; male, 25.0%), there was again no correlation between telomere length and MMSE score. The researchers found that 13% and 18% of cognitively normal patients now showed MCI and dementia, respectively, and 9%, 43%, and 48% of the MCI patients showed a return to normal status, stable MCI, and dementia, respectively.

However, Dr. Zekry said, "There was no association between telomere length and change in cognitive status, either from normal to dementia or MCI, or from MCI to dementia."

Similarly, within the dementia patients (n = 99) at follow-up, 57.6% showed CDR stability, 40.4% CDR impairment, and 2.0% CDR improvement. However, there was again no association between telomere length and severity of dementia.

While the changes in telomere length from baseline were not associated with changes in cognitive status, Dr. Zekry noted that although 50% of the hospitalised patients showed telomere length decreases, some 40% showed telomere length increases. The cause of this telomere elongation remains unknown at present, she said.

"Thus, this large longitudinal study in old demented and normal individuals that includes a 1-year follow-up shows no evidence that telomere length can be used to predict dementia, especially in very old subjects," she concluded, while also noting that in their hospitalised patients, telomere length did not distinguish between dementia and nondementia patients, even when different types of dementia were assessed.


[Presentation title: Does Telomere Shortening Predict Dementia? A Critical Review. Abstract 105]

E-Mail this DGDispatch to a colleague

To print, use this version