Abstract 5165: Molecular features reflecting biological and functional ages in elderly colorectal cancer patients

Aging is a physiological process that is accompanied by functional decline over time. Also, aging is the single biggest risk factor for developing cancer and limitation of active cancer treatment. The number of elderly patients with colorectal cancer (CRC) has increased, however, the knowledge of the clinical outcomes of colorectal cancer in elderly patients are limited. Elderly CRC patients are a very heterogeneous population because the aging process caused complex co-morbidities and chronological age cannot predict the biological and functional changes. In this study, we analyzed cDNA microarray data using integration of receiver operating characteristic (ROC) curve and protein-protein interaction (PPI) network information, demonstrating the aging effect of cancer based on large scale data. We identified molecular features that discriminate 120 elderly (70 ≥age) and 31 young (40 ≤ age) CRC patients. Elderly group had distinct molecular features compared with the young group, and we suggest that these features may underlie the different cancer characteristics of colorectal cancer based on aging. First, we quantified the molecular features of total 15815 probes using ROC analysis. The ROC representing the performance of enolase superfamily member 1 (ENOSF1, rTS) corresponded to an area under curve (AUC) of 0.73 in the training set and 0.71 in the test set. Also the 82 differentially expressed genes (DEGs) over 2 fold with high significance (p