Examining Early Life Risk Factors and Patterns of Screening for Early-Onset Colorectal Cancer

点击翻译按钮获取中文摘要

PROJECT SUMMARY/ABSTRACT In the United States (US), the incidence of early-onset colorectal cancer (EOCRC), defined as colorectal cancer (CRC) diagnosed among individuals <50 years of age, has been increasing rapidly. Compared to late- onset CRC (i.e., CRC diagnosed among individuals ≥50 years of age), EOCRC is characterized by more aggressive pathology and distinct genetic profiles, indicating that the etiology of EOCRC may be distinct. There have been a limited number of studies examining the risk factors of EOCRC, and none have focused on early life risk factors, which could be particularly relevant to early-onset diseases. Over the past few years, the American Cancer Society and the US Preventative Services Task Force reduced the recommended age of CRC screening from 50 years to 45 years, and the COVID-19 pandemic influenced the provision of preventative services across the country. These events could impact the temporal and geographic patterns of screening for EOCRC, but no study has systematically assessed such impacts in a large sample of the US population. The proposed study aims to 1) identify early-life risk factors of EOCRC and 2) evaluate spatiotemporal patterns of EOCRC screening in the US. For the first aim, we will conduct a nested case-control study within the California birth cohort and compare the birth characteristics between 1200 EOCRC cases and 60,000 control subjects matched on year of birth. In addition, we will obtain birth address matched social, behavioral, lifestyle, and environmental factors from the Centers for Disease Control and Prevention and the Environmental Protection Agency and compare these neighborhood-level factors between the 1,200 cases and 60,000 controls. Two key methodological challenges will be addressed: a) spatial autocorrelation, the tendency of adjacent geographic units to share similar attributes; b) the modifiable areal unit problem, a biasing effect that occurs when observed associations between exposures and outcomes vary based on the scale of the chosen geographic unit. For the second aim, we will assess temporal patterns in EOCRC screening and identify factors associated with potential regional variation in screening by leveraging unique claims data from Blue Cross Blue Shield (BCBS), the largest insurance provider in the US that covers one-third of all Americans. Approximately 4 million beneficiaries aged 45-49 years are enrolled in BCBS each year, allowing us a great opportunity to assess both stool-based and structural CRC screening tests. With a rigorous design and unparalleled sample sizes, the proposed study addresses research questions with clear public health significance, is very innovative, and will probably yield a high impact on our knowledge about the etiology and screening of EOCRC – a critically understudied field according to the National Cancer Institute. In addition, by bringing together a multidisciplinary team of experts in cancer epidemiology, health services research, spatial statistics, gastrointestinal oncology, and geographic information systems science, this study will contribute to the training of a promising minority predoctoral fellow with aspirations to pursue a career in cancer epidemiology research and improve the health status for all.