Retinal Vascular Features and Risk of Colorectal Cancer Incidence and Mortality: A Population-Based Cohort Study

INTRODUCTION: Colorectal cancer (CRC) screening remains limited by invasiveness and suboptimal sensitivity of current methods. Retinal microvasculature, a noninvasive window into systemic vascular health, may reflect CRC-associated angiogenesis and inflammation. We investigated whether artificial intelligence (AI)-derived retinal vascular features (RVFs) are associated with CRC risk and prognosis. METHODS: This analysis included UK Biobank participants with baseline fundus imaging. Multidimensional RVFs, including tortuosity, branching angle, caliber, density, and complexity, were quantified using the Retina-based Microvascular Health Assessment System (RMHAS). Associations of RVFs with incident CRC and mortality among patients with CRC were evaluated using univariable and multivariable Cox regression models. RESULTS: Among 55,798 participants (mean age 55.47 ± 8.09 years; 55.07% female), over a median follow-up of 11.32 years, 652 incident CRC cases (including 488 colon cancers and 260 rectal cancers) and 171 post-CRC deaths were documented. After full adjustment, four RVFs showed significant associations with CRC risk: increased arterial tortuosity (hazard ratio [HR] 1.03, 95% confidence interval [CI] 1.01-1.06), wider venular minimum branching angle (HR 1.09, 95% CI 1.02-1.17), smaller arterial maximum branching coefficient (HR 0.90, 95% CI 0.82-0.99), and reduced venular angular asymmetry (HR 0.89, 95% CI 0.82-0.96). For mortality, higher venular tortuosity (HR 1.14, 95% CI 1.00-1.31) and increased arterial maximum length diameter ratio (HR 1.16, 95% CI 1.01-1.32) were associated with elevated risk, whereas venular caliber (HR 0.81, 95% CI 0.67-0.98) and venular length diameter ratio (HR 0.82-0.84) were inversely associated with mortality. Distinct RVF patterns were observed between colon and rectal cancers. CONCLUSION: Specific AI-derived RVFs were independently associated with CRC risk, its subtypes, and mortality. These findings suggest that retinal microvascular changes may reflect systemic processes related to carcinogenesis and point to the potential of retinal imaging as a noninvasive tool for CRC risk stratification.

Authors

Institutions

Publication Details

Journal
Ophthalmology and Therapy
Published
2026-06-30
DOI
https://doi.org/10.1007/s40123-026-01418-w
Primary Topic
Retinal and Optic Conditions
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Retinal Vascular Features and Risk of Colorectal Cancer Incidence and Mortality: A Population-Based Cohort Study

Danli Shi, Min‐Er Zhong, Xiaohong Yang, Tengda Huang et al.
Ophthalmology and Therapy
Retinal and Optic Conditions
article

Retinal Vascular Features and Risk of Colorectal Cancer Incidence and Mortality: A Population-Based Cohort Study

Danli Shi, Min‐Er Zhong, Xiaohong Yang, Tengda Huang, Zhishen Peng, Cong Li, Lei Liu, Yun Ren, Ying Yao, Zhenchao Du, Honghua Yu, Yong Li, Heng Li, Xue He, Yan Wang, Wenli Zhang, Jingyan Peng
article en

Abstract

INTRODUCTION: Colorectal cancer (CRC) screening remains limited by invasiveness and suboptimal sensitivity of current methods. Retinal microvasculature, a noninvasive window into systemic vascular health, may reflect CRC-associated angiogenesis and inflammation. We investigated whether artificial intelligence (AI)-derived retinal vascular features (RVFs) are associated with CRC risk and prognosis. METHODS: This analysis included UK Biobank participants with baseline fundus imaging. Multidimensional RVFs, including tortuosity, branching angle, caliber, density, and complexity, were quantified using the Retina-based Microvascular Health Assessment System (RMHAS). Associations of RVFs with incident CRC and mortality among patients with CRC were evaluated using univariable and multivariable Cox regression models. RESULTS: Among 55,798 participants (mean age 55.47 ± 8.09 years; 55.07% female), over a median follow-up of 11.32 years, 652 incident CRC cases (including 488 colon cancers and 260 rectal cancers) and 171 post-CRC deaths were documented. After full adjustment, four RVFs showed significant associations with CRC risk: increased arterial tortuosity (hazard ratio [HR] 1.03, 95% confidence interval [CI] 1.01-1.06), wider venular minimum branching angle (HR 1.09, 95% CI 1.02-1.17), smaller arterial maximum branching coefficient (HR 0.90, 95% CI 0.82-0.99), and reduced venular angular asymmetry (HR 0.89, 95% CI 0.82-0.96). For mortality, higher venular tortuosity (HR 1.14, 95% CI 1.00-1.31) and increased arterial maximum length diameter ratio (HR 1.16, 95% CI 1.01-1.32) were associated with elevated risk, whereas venular caliber (HR 0.81, 95% CI 0.67-0.98) and venular length diameter ratio (HR 0.82-0.84) were inversely associated with mortality. Distinct RVF patterns were observed between colon and rectal cancers. CONCLUSION: Specific AI-derived RVFs were independently associated with CRC risk, its subtypes, and mortality. These findings suggest that retinal microvascular changes may reflect systemic processes related to carcinogenesis and point to the potential of retinal imaging as a noninvasive tool for CRC risk stratification.

Ophthalmology and Therapy
Hong Kong Polytechnic University (HK), Chinese University of Hong Kong (HK), Shantou University (CN), Key Laboratory of Guangdong Province (CN), Guangdong Academy of Medical Sciences (CN), Guangdong Provincial People's Hospital (CN), Southern Medical University (CN)
National Natural Science Foundation of China
Good health and well-being
Openalex Percentile: Top 7%
Retinal and Optic Conditions
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.