The present longitudinal data demonstrate that the APOE-ε4 risk allele impacts TL dynamics, with ε4 carriers showing significantly greater telomere loss across a 2-year window than matched non-carrier controls. Carriers of the risk allele lost, on average, 435-bp more than non-carriers. Assuming an average telomere loss in adults of 31–63 bp/year 
, the 435-bp shortening observed in APOE-ε4 carriers suggests that their lymphocytes had aged the equivalent of 7–14 years compared to non-carriers. Logistic regression data further demonstrate that the APOE-ε4 allele confers increased odds of leukocyte telomere attrition. The odds of an APOE-ε4 carrier exhibiting telomere shortening over a 2-year period were more than 6 times higher than a non-carrier, adjusting for some established risk factors and potential confounds. Further, the randomized study design provided an opportunity to examine the impact of HT on cell aging. APOE-ε4 carriers who suspended their HT regimen lost an average of 322 base pairs over the two year window, but carriers who remained on HT showed little to no TL decline. Importantly, for non-carriers there was no evidence that HT conferred a protective effect on cell aging.
Strikingly, the impact of the APOE-ε4 risk allele on accelerated cell aging was evident in high functioning, mid-life women who were free of significant medical co-morbidity. This suggests that the impact of the APOE risk allele on telomere attrition begins in the absence of, or prior to, clinical and behavioral evidence of dementia. Short leukocyte TL is an emerging biomarker of biological age and has been associated with a number of chronic and age-related disorders, including depression 
, dementia 
and early mortality 
. The present study builds on these findings by demonstrating that a major genetic risk factor for age-related cognitive decline and dementia, APOE-ε4, confers increased odds of premature, accelerated cell aging in currently high-functioning healthy individuals.
Previous cross-sectional studies indicate that APOE-ε4 status is related to TL although the nature of this relationship has been inconsistent, likely given the constraints of a cross-sectional design. Some studies report shorter TL in APOE-ε4 carriers compared to non-carriers 
, but a recent study observed the opposite association 
. These investigations provide initial evidence of a link between APOE and TL, but their cross-sectional nature limits drawing firm conclusions about the causal relationship between APOE-ε4 and accelerated cell aging. The current longitudinal findings demonstrate that APOE-ε4 carriers exhibit pronounced TL attrition over just a 2-year window and the data suggest that HT, begun at the onset of the menopausal transition, might buffer against cell aging specifically in women at risk for cognitive decline. Importantly, the putative protective actions of HT on telomere dynamics were not observed in APOE-ε4 negative women. For non-carriers, randomization off HT was associated with telomere lengthening. This pattern of telomere growth in a sub-set of individuals is not well understood but is nevertheless consistent with the few but growing number of investigations that have succesfully tracked TL change over time in the same cohort 
. In one of the first longitudinal TL studies, Epel and colleagues 
followed a cohort of 236 elderly individuals (mean age 74) over a 2.5-year period and found that 24% showed TL lengthening. The biological significance of telomere elongation over a short time period is unkown and under investigation.
A limitation of the current study is the sample size, which precluded our ability to explore the complex influence of specific HT regimens (e.g. estrogen opposed versus unopposed by progesterone) on TL dynamics. At present, no other longitudinal study has examined the relationship between HT and telomere attrition in vivo in women. Thus the current findings, which suggest a protective effect of HT for APOE-ε4 carriers, should be considered a first step which needs to be replicated in a larger cohort designed to systematically probe the impact of specific HT formulations on TL dynamics. In our sample, although it was too small to examine various types of HT, these were not distributed equally. Approximately 80% of women who suspended HT for the 2-year study were on a hormone regimen that included 17β-estradiol (versus conjugated equine estrogen), compared to 50% of women who remained on HT. Future studies must test whether the putative protective effects of HT for ε4 carriers (and the potential non-protective effects for non-carriers) are attenuated or enhanced across different hormonal regimens.
A growing body of evidence from in vitro
and in vivo
animal studies has identified a number of potential pathways by which endogenous estrogen and, subsequently, exogenous hormone therapy could modulate telomere dynamics. Estrogen stimulates hTERT gene expression and potentiates telomerase activity 
. Estrogen may also modify TL indirectly given that oxidative stress and chronic inflammatory activity are strong predictors of short TL 
and estrogen may exhibit anti-inflammatory and anti-oxidant actions 
Further studies are critical to clarify the mechanisms by which HT might decelerate telomeric aging in APOE-ε4 carriers but have no protective effect for non-carriers. One clue is that estrogen modifies the expression of the APOE gene through an estrogen receptor (ER) dependent pathway 
. Apolipoprotein-ε mRNA and protein are up-regulated in response to ERα activation and down-regulated in response to ERβ activation 
, suggesting a potential therapeutic target for differentially regulating APOE expression. Yaffe and colleagues also found evidence of a gene-hormone interaction in a longitudinal study of over 3,000 older women who underwent annual cognitive testing over a six-year follow-up. A significant interaction was found between APOE-ε4 carrier status and use of hormone therapy on long-term cognitive decline 
. Potential mechanisms through which APOE status may interact with estrogen (both endogenous and exogenous) warrant further investigation. APOE variants have different anti-oxidant profiles, with the ε4 allele showing decreased antioxidant activity relative to ε2 and ε3 alleles 
. The ε4 allele is also associated with increased accumulation of amyloid-β peptides, whose cytotoxicity is mediated in part by increased oxidative stress 
. Thus, the accelerated telomeric aging observed in ε4 carriers may be due to increased oxidative stress paired with reduced anti-oxidant capabilities conferred by the ε4 allele.
In sum, this study represents the first longitudinal demonstration of accelerated cell aging in APOE-ε4 carriers. Our finding that even high-functioning, healthy mid-life women with the ε4 allele bear markers of cellular aging suggests the need for further research to understand the potential utility of leukocyte TL as an early indicator of future dementia risk. These data provide strong evidence that the impact of the APOE risk allele on telomere attrition begins in the absence of, or prior to, clinical and behavioral evidence of dementia. Futher, the data provide initial evidence that hormone therapy, begun at the onset of the menopausal transition, might buffer against TL attrition in women at risk of cognitive decline. Importantly, terminating HT had beneficial effects for non-carriers, indicating that HT may have differential effects on cell aging across genotypic subgroups. Going forward, it will be crucial to extend these initial data to a larger epidemiological sample to systematically probe whether specific HT formulations, timing of initiation, and duration of use differentially impact cellular aging.