Background: Breast cancer (BC) incidence in Africa is projected to double in less than two decades. While several risk factors have been identified, known modifiable lifestyle risk factors are estimated to explain approximately 30% of BC. Recent studies suggest that stressful life events may influence BC risk, but the evidence is limited. In particular, the biological mechanisms linking stress to BC risk and survival are not well understood. Metabolomics could help to identify novel biomarkers of stress associated with BC risk and survival that could inform cancer control strategies and reduce BC incidence. However, no previous study has examined the relationship between metabolic signatures of stress and BC risk and survival.

Objectives: We aim to 1) investigate the association between self-reported exposure to stress and BC risk and survival, 2) identify metabolic signatures related to stress and 3) assess whether these signatures are associated with BC risk and survival in Black African women.

Methods: We used data from a case–control study conducted in Soweto, South Africa, with available LC–MS untargeted metabolomics data from serum samples of 400 BC cases and 400 population-based matched controls. All data, including stress-related factors, were collected using validated questionnaires. Nearly 6,000 molecular features were detected from metabolomics analyses. Multivariable conditional logistic regression and Cox proportional hazard models controlling for known BC risk and prognostic factors were used to assess the association between self-reported exposure to stress at different life stage and BC risk and survival, respectively. Latent class mixed models were applied to identify stress?trajectory groups across the life course. Lasso regression is being used to relate metabolomics data to self-reported stress in order to identify metabolic signatures of stress and subsequently investigate their associations with BC risk and survival.

Results: High self-reported lifetime exposure to stress was associated with an increased risk of BC (OR=1.49, 95% CI=1.02–2.16; high vs. low; P-trend=0.04). This association was particularly stronger for stress exposure during ages 26–35 (OR=1.51, 95% CI=1.00–2.30; P-trend=0.03), ages 19–25 (OR=1.84, 95% CI=0.98–3.49; P-trend=0.02), and in the five years preceding recruitment (OR=1.62, 95% CI=1.04–2.51; P-trend=0.07). High lifetime cumulative stress exposure was also associated with increased mortality among cases.
Our trajectory analyses showed that women with consistently high stress levels—defined as chronic stress—had a higher risk of BC compared to women with persistently low stress levels. In our metabolomics analyses, higher levels of cortisol, a stress-related metabolite, were associated with an increased risk of BC. Analyses of metabolic signatures of stress are ongoing, and preliminary results will be presented at the conference.

Conclusions/Implications: These findings suggest that high lifetime stress exposure is associated with increased BC risk and mortality in Black African women and that metabolic pathways related to cortisol may play a role in BC risk. Our results highlight the importance of considering psychosocial stress as a modifiable factor in BC prevention and prognosis. Ongoing analyses of metabolic signatures of stress will help clarify the complex association between stress and BC risk and underlying biological mechanisms.