Independent effects of mental health disorders on breast cancer and their mediating factors: evidence from NHANES and two-step Mendelian randomization.

BACKGROUND

The epidemiological data regarding the correlation between mental disorders and breast cancer (BC) is still debatable. This study's objective is to clarify the connection between BC and mental disorders, such as depression, panic, and anxiety, in addition to examining the mediating role of risk factors associated with BC through genetic susceptibility.

METHODS

Cross-sectional data came from the National Health and Nutrition Examination Survey (NHANES) for the years 2015-2018, stratified by BC status. Summary statistics of genome-wide association studies (GWASs) of mostly European ancestry were used for Mendelian randomization (MR) analysis. The connections between depressive symptoms, anxious symptoms, and anxiety medication use and breast cancer were assessed using logistic regression models. The independent effects of depression, panic, and anxiety on BC were estimated through two-sample multivariable MR. Furthermore, 23 possible mediators of the connection were assessed and mediated proportions were computed using two-step MR.

RESULTS

Cross-sectional data revealed that participants taking medication for anxiety were at higher risk of BC (OR (95%CI): 2.18 (1.29-3.70), P < 0.01). Using multivariate MR to adjust for depression, anxiety, and panic, only anxiety was still strongly correlated with the occurrence of overall (β (SE): 0.053 (0.016), P < 0.05) and ER- (β (SE): 0.075 (0.025), P < 0.05) BC. After excluding 23 risk factors related to BC, no mediators were identified between anxiety and BC.

CONCLUSION

Our findings suggests that anxiety susceptibility at the genetic level is a distinct risk factor for BC, with no mediators identified in this process.