TL;DR
This study uses a powerful genetic approach (Mendelian Randomization) to investigate the link between fasting insulin (FI) levels and cancer risk. It found that genetically higher FI levels are associated with an increased risk of colorectal and endometrial cancers, and importantly, FI acts as a mediator in the relationship between type 2 diabetes and endometrial cancer.
What Problem Does This Paper Address?
Observational studies have long suggested a link between elevated fasting insulin (FI) levels and an increased risk of various cancers, as well as a connection between type 2 diabetes (T2DM) and cancer. However, these observational studies are prone to confounding factors (lifestyle, environment) and reverse causation (cancer influencing insulin levels). This makes it difficult to establish a definitive causal relationship. The paper aims to address this by using a genetic approach to disentangle these associations and explore whether FI plays a mediating role in the T2DM-cancer link.
Methods
The core methodology employed is Two-Sample Mendelian Randomization (TSMR). This approach leverages genetic variants (SNPs) that are known to influence a particular trait (like FI levels) as instrumental variables. Because these genetic variants are randomly assigned at conception, they are less likely to be confounded by lifestyle or environmental factors and are not subject to reverse causation.
- TSMR: The primary analysis used the inverse-variance weighted (IVW) method to assess the association between genetically determined FI levels and various cancer types.
- Sensitivity Analyses: To ensure the robustness of the findings, several sensitivity analyses were performed, including weighted-median, weighted-mode, MR-Egger, and MR-PRESSO tests. These methods help detect and account for potential pleiotropy (when a genetic variant affects multiple traits).
- Polygenic Risk Score (PRS) Analysis: To validate the TSMR findings, PRS analysis was conducted using individual-level data from the UK Biobank. This involved constructing a score based on multiple genetic variants associated with FI and then assessing its association with cancer risk.
- Multivariable Mendelian Randomization (MVMR) and MR Mediation Analysis: To investigate the mediating role of FI in the T2DM-cancer relationship, MVMR was used to adjust for both T2DM and FI genetic instruments simultaneously. MR mediation analysis then quantified the proportion of the T2DM-cancer effect that was mediated by FI.
Datasets
- Genome-Wide Association Study (GWAS) Summary-Level Data:
- Glycemic Traits (FI, fasting glucose, 2-h glucose, HbA1c): Derived from the Meta-Analyses of Glucose and Insulin-related traits Consortium (MAGIC), comprising 281,146 participants.
- Type 2 Diabetes Mellitus (T2DM): Extracted from a multi-ethnic meta-analysis of 1.4 million participants.
- Cancer Data: Summary-level genetic data for overall cancer and 14 site-specific cancers were obtained from various consortia, including the FinnGen Biobank, Breast Cancer Association Consortium (BCAC), Ovarian Cancer Association Consortium (OCAC), International Lung Cancer Consortium (ILCCO), and meta-analyses for endometrial and colorectal cancers. Only cancer types with over 400 cases were included to enhance statistical power.
- Individual-Level Data: UK Biobank data was used for the Polygenic Risk Score (PRS) analysis.
Key Insights and Ideas
- Genetic Predisposition Matters: The study strongly suggests that genetic predisposition to higher fasting insulin levels causally influences the risk of certain cancers, moving beyond correlational findings from observational studies.
- Targeted Cancer Risks: The association between high FI and cancer risk wasn't universal. The study pinpointed colorectal and endometrial cancers as the primary sites where this link is evident.
- FI as a Bridge: A crucial finding is that FI acts as a significant intermediary in the pathway from type 2 diabetes to endometrial cancer. This implies that the increased cancer risk associated with diabetes might, in part, be driven by the resulting changes in insulin levels.
- Methodological Rigor: The use of multiple MR methods and sensitivity analyses, along with PRS validation, provides a robust framework for establishing causal inference in this complex biological context.
Key Results
- Colorectal Cancer: Genetically determined high FI levels were associated with an increased risk of colorectal cancer (OR = 1.87, 95% CI: 1.23-2.84, p = 0.003).
- Endometrial Cancer: Similarly, high FI levels were linked to an increased risk of endometrial cancer (OR = 1.89, 95% CI: 1.08–3.01, p = 0.008).
- Overall Cancer and Other Sites: No significant association was found between genetically determined high FI levels and overall cancer risk or the other 12 site-specific cancers examined.
- T2DM and Endometrial Cancer Mediation: MVMR analysis showed that the association between T2DM and endometrial cancer risk was attenuated when FI was accounted for. MR mediation analysis quantified that FI mediated approximately 21.3% of the T2DM-endometrial cancer association.
- PRS Validation: The PRS analysis in UK Biobank successfully replicated the association between genetic liability to high FI levels and increased risk of colorectal and endometrial cancers.
Limitations and Future Work
- European Population Focus: The analyses were primarily conducted among European populations, which may limit the generalizability of the findings to other ethnic groups.
- Limited Power for Rare Cancers: For some site-specific cancers with fewer cases, the study may have lacked sufficient statistical power to detect weak or moderate associations.
- FI Variance Explained: The UK Biobank data did not include direct measurements of circulating FI levels, preventing an estimation of the variance in FI explained by the PRS.
- Causality Nuance: While MR is powerful for causal inference, the authors caution that claims of causality should be made with care due to the complexity of biological associations and potential limitations of genetic instruments.
- Future Research: The authors suggest that larger observational and well-designed experimental studies are needed to further explore the complex role and mechanisms of FI in cancer progression. They also highlight the need for further investigation into the associations between FI and other cancer types. The findings suggest that monitoring insulin levels and addressing subclinical insulinemia could be a strategy for colorectal and endometrial cancer prevention.
Summary updated