The human gut microbiome, home to a range of bacteria and various other microorganisms, is a diverse and complex feature of human homeostasis. Its specific composition varies between individuals due to pre-determined characteristics, i.e. our genetics, and modifiable factors including diet, lifestyle and probiotic consumption. The differences between individuals have highlighted its role in mediating a range of pathologies including obesity, metabolic syndrome, inflammatory disorders, and alterations in stress responses and behaviour. More recently, a relationship to cancer has been highlighted, where variations in the metabolites produced from these microorganisms is likely to have differential risk to cancer aetiology between individuals. The bacteria within the microbiome therefore has the potential for pharmaceutical intervention (through pro/prebiotics), if specific metabolites produced from these microbes are robustly associated to cancer outcomes.
Observational epidemiological studies have tried to quantify this relationship; however, they will often lack in overall certainty of causal inference and further be prone to biases, confounding and reverse causation. Mendelian randomisation (MR) has therefore been used as a tool within the field to improve the weakness pertinent to observational studies. The aim of the methodology is to utilise genetic polymorphisms, with well-characterised biological functions, as a proxy measurement for environmental exposures to provide evidence for causation. These germ line variants are independently and randomly assorted, allow for objective measurements to the degree of exposure and show a lack of association to behavioural, social or physiological factors, therefore allowing MR to be comparable to that of a randomised controlled trial (RCT) within the hierarchy of evidence – if done correctly.

Whilst MR may improve causality within this field, it is however prone to its own limitations. Specifically in the context of the microbiome, whereby, identifying variants to proxy for our trait of interest becomes increasingly more difficult. The project will therefore aim to test the relationship between the microbiome and cancer outcomes, through triangulating finding from both observational and genetic evidence. It will focus specifically on cancer outcomes that have not previously been robustly studied within this context, including (but not limited to) lung cancer.