Approved Research

Artificial neural network-based integration of genomic, metabolomic, environmental and clinical information to predict the risk of cerebrovascular-specific diseases in patients with colorectal cancer

Lay summary

Colorectal cancer (CRC) is one of the most prevalent diseases and the second leading cause of death worldwide. With the advance in diagnosis and therapy of CRC, patient death due to this cancer reduces greatly, putting mortality from other non-cancer diseases an unignorable concern for this population at present. Among them, cerebrovascular-specific disease (CVSD) constitutes a major factor contributing to the mortality in CRC patients. However, the current evidence concerning the CVSD occurrence in CRC is limited. A recent Surveillance, Epidemiology, and End Results (SEER)-based cohort study from our group indicated an increased CVSD mortality in CRC and also found several clinical parameters linked to this outcome. However, due to the incomplete patient information in SEER database, our study only addressed the relationship between some clinicopathological factors and CVSD events in CRC, which may not provide adequately accurate information on this correlation. As a continuation of our previous work, the present project will further analyze the genomic, metabolomic and environmental predispositions to CVSD in CRC patients by using the more homogeneous, comprehensive, and complete data in the UK Biobank database. More importantly, we also seek to develop and validate an artificial neural network-based model to predict the risk of CVSD in this population. The estimated project duration is three years. We believe that our project results will provide helpful information in uncovering the potential mechanism of CVSD development in CRC and can also be useful in guiding risk stratification and preventive intervention optimization of CVSD in the patients.