Mendelian Randomization in Infectious Diseases: Challenges and Opportunities
Overview
Mendelian randomization (MR) is a promising method to identify causal relationships in infectious diseases, exemplified by its role in COVID-19 drug target discovery. However, many MR studies in infection violate core assumptions, especially when using infection as an exposure, leading to biased or uninterpretable results. This review highlights these challenges and advocates for careful application of MR to address causal questions unmet by other methods.
Background
Mendelian randomization uses genetic variants as instruments to estimate causal effects of exposures on outcomes, minimizing confounding and reverse causation common in observational studies. While MR has been successful in non-infectious diseases, its application to infectious diseases has surged only recently, driven by public availability of genetic data. Infectious disease MR studies face unique challenges, particularly when infection is treated as an exposure due to complex host-pathogen interactions and genetic pleiotropy. Understanding these limitations is critical to harness MR's potential in infectious disease epidemiology.
Data Highlights
Number of PubMed results for "Mendelian randomi[z/s]ation AND infection": - Up to 2018: 48 papers - Since 2018: 747 papers
Example of MR success: - Identification of TYK2 expression as causal for severe COVID-19 led to baricitinib drug trials and approval.
Key genetic instruments: - Variants in HFE gene affecting iron status linked to pneumonia risk. - Genetic variants in blood cell traits affecting malaria susceptibility.
Key Findings
MR studies using infection as an exposure are prone to bias due to lack of strong, specific genetic instruments and complex host-pathogen interactions.
Genetic variants associated with antibody response, often in the HLA region, pose challenges due to pleiotropy and linkage disequilibrium, complicating causal inference.
Pathogen encounter is largely environmental and not strongly genetically controlled, limiting MR applicability for infection acquisition questions.
Some published MR studies have used outcome data predating pathogen emergence (e.g., COVID-19 before 2019), rendering conclusions invalid.
MR has successfully identified drug targets in infectious diseases, such as TYK2 for COVID-19, demonstrating its potential when assumptions are met.
Recent retractions highlight the importance of adhering to MR assumptions to avoid invalid conclusions in infectious disease research.
Clinical Implications
Clinicians and researchers should interpret MR findings in infectious diseases cautiously, especially when infection is modeled as an exposure, due to potential biases and assumption violations. MR can guide drug target identification and causal inference when applied appropriately, but requires careful selection of genetic instruments and consideration of pathogen biology. Awareness of these limitations will improve the design and interpretation of MR studies to inform infectious disease prevention and treatment strategies.
Conclusion
Mendelian randomization offers valuable opportunities to elucidate causal relationships in infectious diseases but faces unique challenges that can compromise validity. Rigorous adherence to MR assumptions and thoughtful study design are essential to realize its full potential in infectious disease epidemiology and translational science.
References
Davey Smith et al. 2022 -- Mendelian Randomization in Infectious Diseases: Challenges and Opportunities
