AS738 - Blood protein biomarkers for colorectal cancer etiology, prevention, and disparity

Investigator Names and Contact Information

Introduction/Intent

Colorectal cancer (CRC) is the third most common cancer in the United States (US) and worldwide with rising incidence in younger adults^1,2. In the US, compared with White individuals, incidence/mortality of CRC is >1.2-fold higher in African American (AA) and >1.2-fold lower in Asian individuals1. Chronic inflammation, a hallmark of cancer³, serves as a dominant causal pathway for CRC^4-6, yet specific inflammatory and immune-related biomarkers driving CRC development are poorly understood. Circulating C-reactive protein and interleukin-6 are inflammatory markers that have been associated with CRC risk in some but not all studies^7-11. However, their low specificity for CRC risk^12,13 and likely differential strengths of associations by racial group¹¹ calls for a more comprehensive evaluation of the proteome for CRC-specific inflammation and immune-related biomarkers.

Emerging cost-efficient multiplexed affinity-based proteomics platform, now allow the profiling of several thousand proteins in a single biospecimen, with ~1,100 proteins involved in inflammatory/immune-related pathways. A comprehensive investigation of CRC-specific, pre-diagnostic, and causal protein biomarkers has not been undertaken; such an approach would likely lead to discovery of new drug targets for chemoprevention^14-16. Prior smaller-scale discoveries mainly in White individuals, including our own, have already shown the promise of proteomic discovery in CRC, calling for larger-scale discovery and validation. Such efforts also have the potential to develop strongly predictive early detection biomarkers that can be used to refine current risk prediction tools, providing an alternative to invasive screening tools, and potentially leading to higher screening uptake.

Our overarching hypothesis is that pre-diagnostic levels of inflammation and immune-related proteins will be associated with CRC risk and such associations may vary by racial group. To test this, we have established a consortium of 9 well-characterized prospective cohort studies with pre-diagnostic biospecimen collection. We will adopt a cost-effective approach by augmenting existing pre-diagnostic proteomic data (>70% of CRC cases) with new proteomic data generation using state-of-the-art technologies. These proteomic data in 3,955 incident CRC cases will be used to identify novel protein biomarkers for overall CRC and according to racial group. We will leverage multiple genome-wide association studies (GWAS) to determine causally associated proteins in Whites, AAs, and Asians. In a validation stage, we will quantify promising protein markers associated with CRC risk in independent cases and controls. Our specific aims:

Aim 1: To prospectively examine relationships between protein biomarkers and CRC risk.

1A. Investigate associations of pre-diagnostic levels of circulating protein biomarkers with CRC risk. In hypothesis driven analyses, we will examine the associations between ~1,100 inflammation and immune-related proteins with CRC risk. Next, we will assess associations between up to 7,000 protein biomarkers and risk of CRC. We will conduct analyses by racial group, age at diagnosis, and time to diagnosis.
1B. Examine potential causality of associations between circulating protein biomarkers and CRC risk.We will conduct a series of rigorous Mendelian randomization and colocalization analyses in 119,245 cases (88,516 White, 3,828 AA, 26,901 Asian cases) and 581,596 controls from multiple CRC GWASs to identify proteins potentially causally associated with CRC risk.

Aim 2: To validate associations in independent samples. Using pre-diagnostic blood samples from independent CRC case-control pairs we will seek to validate CRC protein biomarkers identified in Aim 1. Impact: In this first of its kind study, our multi-disciplinary investigative team, with substantial expertise in CRC epidemiology^17-29, genetics^30-33, and causal inference^19,20,34-38, will systematically examine the causal role of inflammation and immune-related proteins in CRC development. Our inclusion of diverse populations from multiple high-quality cohorts will create unprecedented and timely data to elucidate racial specific proteomic biomarkers for CRC, providing a significant step towards precision-based CRC risk assessment and prevention.