Night work and breast cancer

E. Negri,¹ F. Bravi,² C. Fenga,³ S. Vivarelli,³ C. La Vecchia,² F.S. Violante⁴ | ¹Department of Medical and Surgical Sciences, University of Bologna; ²Department of Clinical and Community Sciences, University of Milan; ³Department of Biomedical, Dental, and Morphological and Functional Imaging Sciences, University of Messina; ⁴IRCCS University Hospital of Bologna, Italy

Introduction. In 2019, the International Agency for Research on Cancer (IARC) classified night shift work (NSW) as probably carcinogenic to humans (Group 2A), based on limited evidence in humans, sufficient evidence in experimental animals, and strong mechanistic evidence in the latter. Disruption of the circadian rhythm has been hypothesized as a possible mechanism, primarily because exposure to light during the night reduces serum melatonin levels.

Objectives. The overall objectives of the project were to analyze the association between exposure to NSW and breast cancer (BC) in two cohorts of healthcare workers (HCW); to conduct a nested case-control study within the cohort to investigate the effect of potential confounders; and to compare levels of selected aging biomarkers between NSW-exposed and non-exposed individuals.

Methods. We conducted a systematic review of NSW exposure among HCWs and the risk of BC. We established two retrospective cohorts of HCW from two hospitals in Bologna and Messina. BC cases were identified through record linkage with the Emilia-Romagna tumor registry (Bologna) or through medical records (Messina). Information on confounding factors was collected for all subjects (Messina) or for the cases and three randomly matched controls (Bologna). The hazard ratios (HRs) for BC and CI at 95% were estimated using a proportional hazards model that included age at enrollment, occupation, work duration, and the cumulative number of months with at least three-night shifts (introduced as a time-dependent variable).

Results. The systematic review included 12 studies and did not show excess risk for women exposed to NSW in general (relative risk, RR=1.05, 95% CI: 0.96-1.14). However, there was a significant excess for women exposed for over 20 years (RR=1.25, 95% CI: 1.01-1.55) based on 7 studies. Regarding the female employees of Sant’Orsola Hospital, from a database of 10,728 women with at least one service record, after excluding cases of breast or other cancer diagnosed before employment, women with less than 12 months of total service, and women with a service interruption of more than 1 year, a cohort of 8,663 women was obtained, in which 167 cases of BC and 198 cancers in other sites were identified. The median follow-up was 19.7 years. Compared to women who were never exposed (100 events out of 825×103 person-years), the adjusted HR was 0.6 (95% CI: 0.4-1.0, 31 events out of 492×103) for 1-23 months of exposure, 0.6 (95% CI: 0.3-1.1, 13 events out of 241×103) and 2.1 (1.3-3.5, 23 cases out of 190×103). Analysis of additional confounding factors and the Messina cohort is ongoing.

Conclusions. Both the literature and our data suggest that there may be an excess risk of BC in women exposed to NSW for long periods. The absence of a dose-response gradient suggests caution in interpreting this association causally.