• Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, UK. w.chan21@imperial.ac.uk.
• Division of Epidemiology, Department of Medicine, Vanderbilt Epidemiology Center, Vanderbilt-Ingram Cancer Center, Vanderbilt University Medical Center, Nashville, TN, USA.
• Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina, Greece.
• Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, UK.
• Nutrition and Metabolism Branch, International Agency for Research on Cancer, World Health Organization, Lyon, France.
• VA Palo Alto Health Care System, Palo Alto, CA, USA.
• Department of Gastroenterology, Biodonostia Health Research Institute, Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Universidad del País Vasco (UPV/EHU), San Sebastián, Spain.
• Department of Medical Oncology & Therapeutics Research, City of Hope National Medical Center, Duarte, CA, USA.
• Nantes Université, CHU Nantes, Service de Génétique Médicale, Nantes, France.
• Cancer Epidemiology Division, Cancer Council Victoria, Melbourne, VIC, Australia.

BACKGROUND: Whether blood lipids are causally associated with colorectal cancer (CRC) risk remains unclear.

METHODS: Using two-sample Mendelian randomisation (MR), our study examined the associations of genetically-predicted blood concentrations of lipids and lipoproteins (primary: LDL-C, HDL-C, triglycerides, and total cholesterol), and genetically-proxied inhibition of HMGCR, NPC1L1, and PCSK9 (which mimic therapeutic effects of LDL-lowering drugs), with risks of CRC and its subsites. Genetic associations with lipids were obtained from the Global Lipids Genetics Consortium (n = 1,320,016), while genetic associations with CRC were obtained from the largest existing CRC consortium (n = 58,221 cases and 67,694 controls). Our main analysis was a multivariable MR (MVMR) with mutual adjustments for LDL-C, HDL-C, and triglycerides. Secondary analyses, including MVMR additionally-adjusting for BMI or diabetes, were also performed.

RESULTS: Genetically-predicted LDL-C was positively associated with CRC risk in the MVMR adjusted for HDL-C and triglycerides (OR = 1.09; 95%CI 1.02-1.16 per SD increase) and additionally-adjusted for BMI (OR = 1.12; 95%CI 1.05-1.21) or diabetes (OR = 1.09; 95%CI 1.02-1.17). Associations were generally consistent across anatomical subsites. No clear evidence of association was found for other lipids, lipoproteins, or LDL-lowering drug-targets.

CONCLUSIONS: We found evidence of a weak positive association between LDL-C and CRC that did not appear to be explained by potential pleiotropic pathways such as via HDL-C, triglycerides, BMI, or diabetes.