This training grant is being sought by a PhD student entering the dissertation phase of her degree in Epidemiology. The candidate aspires to become a leading independent scientist in the field of aging research. She will devote 100% of her time to this research for the duration of her PhD. Her hands-on training during her dissertation phase, coupled with carefully selected courses, are designed to make her expert in the use of advanced statistical methods to define associations between individual and neighborhood factors, genetic variants, and telomere length. This dual training approach will prepare her for future studies investigating molecular markers associated with aging and the role neighborhood environment and genetics play in health disparities and aging. Throughout this award, she will be mentored by senior experts in behavior, geospatial analyses, aging outcomes, genetics, molecular markers and biostatistics. The process of cellular aging, as measured by telomere length, likely occurs through the complex interactions of multiple factors over the lifespan. The goal of the research proposed here is to better explain the complex, multifactorial nature of cellular aging using a surrogate biomarker, telomere length. We will undertake a multilevel statistical approach that includes identifying individual-level (Aim 1) and neighborhood- level factors (Aim 2) that affect telomere length and the relationship between telomere genetic variants and telomere length (Aim 3), in an existing multi-ethnic longitudinal cohort of older men. Associations have been reported between telomere length and genetic variants, individual risk behaviors (smoking, body mass index) and sociodemographic variables (an individual's race, sex, social status). However, neighborhood level factors (social isolation, structural decline, violence, etc) represent an important consideration for aging studies since they have been associated with age-related disease and are particularly relevant for older adults, who may live in unfavorable neighborhoods for long time periods. This is the first study to account for the collective effects of multiple factors on telomere length. Using linear regression and backward regression, the applicant will model baseline telomere length (defined by T/S ratio) as a continuous variable, with individual level measures, and select neighborhood factors entered separately as independent variables, using a p-value of 0.20 as the threshold for inclusion. Analyses will be repeated stratifying by race. Clustering and multicollinearity will be considered when assessing the effect of neighborhood variables on telomere length. In order to test the effects of genetic haplotypes in the neighborhood context, we will use multivariable analysis to model telomere length and a vector of covariates identified in Aims 1 and 2. Studies of this nature are critical if genes or individual risk factors affect telomere length more profoundly in certain neighborhood settings. This study will build the foundations of a conceptual model that can be used as the basis for future studies on aging.
The association between genetic variants and telomere length may depend on individual and neighborhood risk factors. Using an interdisciplinary, multi-level statistical approach, this research will better explain the complex, multifactorial nature of cellular aging(measured here by telomere length). This study will help identify health disparities in aging and can serve as a conceptual model for future aging research.
|Lynch, Shannon M; Mitra, Nandita; Ravichandran, Krithika et al. (2017) Telomere Length and Neighborhood Circumstances: Evaluating Biological Response to Unfavorable Exposures. Cancer Epidemiol Biomarkers Prev 26:553-560|
|Lynch, Shannon M; Peek, M K; Mitra, Nandita et al. (2016) Race, Ethnicity, Psychosocial Factors, and Telomere Length in a Multicenter Setting. PLoS One 11:e0146723|
|Zhang, Yaqing; Morris 4th, John P; Yan, Wei et al. (2013) Canonical wnt signaling is required for pancreatic carcinogenesis. Cancer Res 73:4909-22|
|Lynch, Shannon M; Rebbeck, Timothy R (2013) Bridging the gap between biologic, individual, and macroenvironmental factors in cancer: a multilevel approach. Cancer Epidemiol Biomarkers Prev 22:485-95|