BA25 - Leukocyte telomere dynamics, cardiovascular aging and survival in the WHI Long Life Study in the WHI Long Life Study

Investigator Names and Contact Information

Alexander Reiner (apreiner@uw.edu)

Introduction/Intent

This page provides study documentation for BA25. For description of the specimen results, see Specimen Results Description (open to public). Data sets of the specimen results are included in the existing WHI datasets located on the WHI Data on this site (sign in and a completed Data Use Agreement are required; see details on the Data site).

PI: Alex Reiner (Fred Hutchinson Cancer Research Center) Investigative team: Andrea LaCroix, Charles Kooperberg, and Cara Carty (Fred Hutchinson Cancer Research Center); Rivka Stone & Abraham Aviv (University of Medicine and Dentistry of New Jersey)

Specific Aims and Hypotheses The main goal of the proposed study is to measure LTL in WHI-LLS participants at two time points, approximately 15 years apart in order to assess LTL dynamics (LTL and its age-dependent attrition) in relation to incident CHD events and mortality. The central question we would like to answer is as follows: Can knowledge about LTL improve classification of an elderly woman’s risk of CHD and early mortality? Answering this question is consistent with the overall goal of the LLS which is to support studies of factors associated with healthy aging and examine the changing levels of intermediate markers of CVD risk on subsequent clinical events. To this end, we will use LTL measurements to test hypotheses that mechanistically link its dynamics to atherosclerotic CVD and longevity. Below we present our specific aims and hypotheses, as well as the scientific and public health ramifications of our project.

1. To investigate the prospective associations of LTL with CHD and mortality during follow-up through 2015 in the WHI Medical Record Cohort (MRC) of the WHI 2010-2015 extension using two nested case-cohort studies as follows. We propose to (a) examine associations of LTL in DNA samples collected in 1993-1998 (WHI baseline exam) with the separate outcomes of incident CHD events and mortality throughout the entire WHI follow-up period through 2015 (up to 22 years of follow-up), and (b) examine associations of LTL samples collected in 2012-2013 in the LLS subset of the MRC cohort with the combined outcome of incident CHD events and mortality through 2015 (up to 3 years of follow-up).

For this aim, we propose to use the highly efficient case-cohort sampling design in which we initially select a race-stratified random sample of MRC-eligible participants from the WHI baseline exam and subsequently select all cases of CHD events and deaths that accrue after baseline exam. Our primary goal is to assess the relationship of LTL at either time point with CHD and mortality outcomes.

Our hypothesis is that individuals with a short LTL at baseline or the LLS visit are most likely to suffer CHD events or death using time-to-event methods.

2. To investigate the association of LTL attrition between baseline exam and LLS follow-up exam (14-19 years) (a) with incident CHD events after the baseline exam (a cross-sectional analysis) and (b) with the combined outcome of incident CHD events and mortality in the WHI LLS participants from the LLS visit date through 2015 (a prospective analysis).

Based on technical considerations related to TL measurement and the average yearly rate of LTL shortening, the individual’s rate of LTL attrition can only be reliably assessed at time points separated by ≥ 10 years. The availability of high-quality DNA samples in WHI at both time points, coupled with highly accurate LTL measurements by Southern blot will allow us to compare, for the first time, both crosssectional and longitudinal LTL measurements as predictors of cardiovascular aging and mortality in elderly women.

Our hypothesis is based on the following premise: Most adults display a high degree of LTL tracking, namely individuals rarely change their relative LTL ranking (e.g., long, average, short age-adjusted LTL) over adult life. Therefore, LTL at baseline exam will be highly correlated with LTL at follow-up exam. We hypothesize that, the absolute values of age-adjusted LTLs at baseline and follow up exams, rather than the rate of LTL attrition between the two exams, will be better predictors of CHD events and mortality.

3. To investigate the associations of LTL dynamics with several cardiovascular risk factors: self-reported race, estimated individual African American (AfAm) ancestry, physical activity, cigarette smoking, C-reactive protein, plasma lipid profile, insulin resistance and total energy intake.

A detailed review of what is known about LTL and CVD risk factors is provided in Sections II.E-G below. Briefly, recent studies have suggested that, despite a greater burden of CVD risk factors, AfAms have longer LTL than whites. In addition, lifestyle factors impact age-dependent LTL attrition by mechanisms that are likely mediated through inflammation/oxidative stress. The availability of self-reported and objectively measured physical activity levels, sedentary time, total energy intake, smoking history, and CVD biomarker data (HOMA-IR, CRP, triglycerides, total cholesterol, LDL, and HDL) at both time points will allow us to correlate changes in LTL with atherosclerotic CVD risk over time.

Our hypothesis comprises two major elements. The first is based on the findings that AfAms have less predilection to atherosclerosis than whites and that after the 8th decade AfAms also display less CHD mortality than whites (the mortality cross-over phenomenon). We posit that the longer LTL in AfAms explains in part these two, presumably related, observations. The second element is based on findings that: (a) sedentary lifestyle and smoking are associated with a shorter LTL; and (b) LTL is inversely correlated with CRP (and other inflammatory indices) and positively with HDL. It follows that sedentary lifestyle, smoking and persistently high CRP and low HDL (at baseline and follow-up exams) will be associated with a faster rate of LTL attrition. That said, in and of themselves these parameters will exact only a relatively small effect on the individual’s overall LTL ranking at follow-up LLS exam compared with the baseline exam.

4. Among women in the MRC cohort who participated in the WHI Hormone Therapy Trial, investigate the effect of randomization to hormone therapy (HT) on LTL attrition.

The WHI placebo-controlled clinical trial of HT provides the unique opportunity to examine the potential impact of HT on LTL attrition, given that the promoter of the catalytic subunit of telomerase, which is referred to as the telomerase reverse transcriptase (TERT), has an estrogen response element. This is the only trial setting, to our knowledge, in which the effects of long term, randomized HT exposure on LTL attrition can be examined.

Our hypothesis is that women treated with HT will display a slower rate of LTL attrition, since activation of TERT entails increased telomerase activity, which attenuates the rate of LTL shortening. We further hypothesize that although HT might attenuate the rate of LTL shortening, such an effect would exert a minor influence on the overall ranking of the individual woman with respect to LTL.

5. To examine, using existing WHI genotype data: (a) the relative contribution of common and rare genetic variants to both baseline LTL and LTL attrition; and (b) the association of LTL-regulating genes (loci) with incident CHD events and mortality.

Genome-wide association studies (GWAS) of LTL, performed in cohorts comprising mainly whites, have deciphered LTL-regulating genes that provide mechanistic insights into the potential roles of LTL dynamics, and by inference HSC-TL dynamics, in CVD. Emerging data suggest that a subset of these genes might be longevity genes. However, little is known about LTL-regulating genes in AfAms. The role of genetic factors on rate of LTL attrition in adults has never been examined. The availability of dense genome-wide genotype and imputed genotype data on common and rare variants available through WHI-GARNET, WHI-SHARe, WHIMS+, and WHISP provide a rich resource for studying the genetic determinants of LTL dynamics in a bi-racial sample of women. By applying a novel statistical genetic modeling approach that allows genetic (co)variance to be estimated from SNP data on unrelated individuals, we will estimate the genetic and environmental contributions separately for baseline and change in LTL across a 15 year period in older women.

Our hypothesis is that in both whites and AfAms LTL-regulating genes belong to two major categories: genes that are engaged in telomere maintenance in all cells (intrinsic regulators) and genes that modify the pace of HSC replication (extrinsic regulators). We note, however, that the main impact of the intrinsic regulators might be during intra-uterine growth and that of the extrinsic regulators is likely to be early after birth, during extra-uterine growth, when the rate of HSC replication and LTL attrition is considerably higher than during adult life. That said, these genes leave a lasting impact on LTL throughout the individual’s life span and a few might be longevity genes.