Polysocial risk score, lifestyle, genetic factors and risk of incident lung cancer

Abstract

Objectives

Lung cancer (LC) is the most frequently diagnosed cancer globally; however, the role of social risk factors in its development is not well understood. While previous studies have identified various lifestyle and genetic factors contributing to LC, the interplay between these elements and broader social determinants of health is still unclear. In this study, we aimed to construct a polysocial risk score (PsRS) that captures the multifaceted nature of social environment risk exposure and its relationship with incident LC, and to explore whether the effect of PsRS is influenced by lifestyle behaviours and heritable risk.

Study design

Cohort study.

Methods

In the UK Biobank cohort, 349,553 participants without previous a cancer diagnosis were recruited. For PsRS construction, 12 social determinants of health were calculated across three domains consistently associated with incident LC. Cox models were used to estimate the association between PsRS and incident LC. Healthy lifestyle and LC genetic risk scores were constructed to evaluate whether lifestyle behaviours and genetic susceptibility modified the effect of PsRS on LC incidence. Mediation analysis was used to estimate whether a healthy lifestyle mediates the effect of PsRS' on LC incidence.

Results

Compared with participants with low PsRS (≤3), the fully adjusted hazard ratio (HR) (95 % CI) for high PsRS (≥7) in developing LC was 2.75 (2.43–3.12). We observed an additive interaction between PsRS and lifestyle. The proportion of mediation effect of lifestyle in the association between PsRS and LC was 6.41 % (95 % CI: 5.74–7.08 %). Individuals with high PsRS and genetic risk had a 4.63-fold higher risk of incident LC.

Conclusion

A high PsRS is associated with a higher risk of LC, and lifestyle influences this association. High heritable susceptibility and unfavourable social vulnerability may synergistically contribute to higher LC incidence.