Abstract
INTRODUCTION: Age acceleration in survivors of breast cancer is a critical issue because cancer and its treatment can increase structural and numerical chromosomal aberrations, while simultaneously shortening telomere length and changing ageing phenotype. Therefore, the current study will be using machine learning architectures to accurately predict the factors that contribute to age acceleration among survivors of breast cancer. METHODS AND ANALYSIS: The Cancer Survivors' Trajectories of Ageing Research (C*STAR) is a hospital-based cross-sectional study involving multi-ethnic Malaysian survivors of breast cancer and a non-breast cancer control group, frequency-matched by age group (±5 years), sex and ethnicity. The three main stages of this study will be conducted in the predictive model development. First, a set of validated questionnaires will be used to collect the data on modifiable factors of ageing phenotypes and behavioural determinants of health. Second, 3 mL non-fasting blood samples will be collected, and lymphocytes will be isolated to determine telomere length using real-time PCR as a biomarker of age acceleration. Lastly, a machine learning architecture will be deployed to identify modifiable factors that may contribute to age acceleration in survivors of breast cancer and controls, with these factors used as input and ageing biomarkers of telomere length as output. The study outcomes may serve as guidance to enhance the quality of life of survivors of breast cancer and hinder the recurrence of cancer while ageing successfully. ETHICS AND DISSEMINATION: Ethical approval was obtained from the Research Ethics Committee, Universiti Kebangsaan Malaysia (JEP-2022-700) to carry out this study. Written informed consent will be obtained from each survivor of breast cancer and each cancer-free woman prior to participation. The results of this study will be published for future research and clinical applications.