Prostate cancer risk decreases following cessation of night shift work Manolis Kogevinas 1,2,3,4 , Ana Espinosa 1,2,3,4 , Kyriaki Papantoniou 5 , Nuria Aragonés 4,6 , Beatriz Pérez-Gómez 4,7 , Javier Burgos 8 , Inés Gómez-Acebo 4,9,10 , Javier Llorca 4,9,10 , Rosana Peiró 4,11 , Jose J. Jimenez-Moleón 4,12 , Juan Alguacil 4,13 , Adonina Tardón 4,14 , Marina Pollan 4,7 and Gemma Castaño-Vinyals 1,2,3,4 1 ISGlobal, Barcelona, Spain 2 IMIM (Hospital del Mar Medical Research Institute), Barcelona, Spain 3 Universitat Pompeu Fabra (UPF), Barcelona, Spain 4 CIBER Epidemiología y Salud Pública (CIBERESP), Madrid, Spain 5 Department of Epidemiology, Center of Public Health, Medical University of Vienna, Vienna, Austria 6 Epidemiology Section, Public Health Division, Department of Health, Madrid, Spain 7 Environmental and Cancer Epidemiology Area, National Center of Epidemiology, Carlos III Health Institute, Madrid, Spain 8 Hospital Ramon y Cajal, Madrid, Spain 9 University of Cantabria, Cantabria, Spain 10 IDIVAL, Santander, Spain 11 Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunidad valenciana (FISABIO), Valencia, Spain 12 Universidad de Granada, Granada, Spain 13 Centro de Investigación en Salud y medio Ambiente (CYSMA), Universidad de Huelva, Huelva, Spain 14 Universidad de Oviedo, Asturias, Spain Dear Sir, Night shift work has been associated with breast and prostate can- cer. A recent large pooled analysis on breast cancer that examined also time-related variables showed that risk in premenopausal women was highest in those with current or recent (2 years) night shift work, decreased in women who had left night shift work 2–20 years earlier, and was lowest at 20 years or more since last night shift work. 1 A similar pattern was observed in a large cohort study in the UK 2 that overall did not find any association between breast cancer and night shift in older women, but reported a 10% increase the first 10 years after exposure while a decrease in risk was observed following that period. A relatively similar pattern although not entirely consistent was described for night shift and CLL. 3 A Canadian case–control study on prostate cancer risk and shift work did not find differences in risk by time since last exposure. 4 We have published results from one of the largest studies on night shift work and prostate cancer using detailed occupa- tional exposure assessment in the MCC-Spain study 5 but had not evaluated patterns of risk following end of exposure. The odds ratio (OR) for ever night shift work compared to never night workers (reported in 4) was 1.14 (95%CI 0.94–1.37) and for workers with more than 28 years duration it was 1.37 (95% CI 1.05–1.81). We reanalysed the data to evaluate whether the attenuation shown for breast cancer by time since last exposure is also observed in prostate cancer. MCC-Spain is a population based case–control study enroll- ing subjects in 12 regions of Spain, 2007–2013 and examining several common tumours using the same set of population con- trols. 6 For this analysis, we used the same population as in the earlier publication. 5 Prostate cancer cases (n = 1,093) aged 27–85, were recruited in 11 hospitals in seven Spanish regions (Barcelona, Madrid, Cantabria, Valencia, Granada, Huelva and Asturias). All cases had a new histologically confirmed diagno- sis of prostate cancer from September 2008 through December 2012 and lived in the catchment area of each hospital for at least 6 months prior to diagnosis. We used rosters from pri- mary health centres around the hospitals enrolling cases to ran- domly select male controls (n = 1,387). We had recorded information on shift work for all jobs and examined detailed information on lifestyle and sociodemographic factors. A com- plete description of the study can be found elsewhere. 5 Night shift work was defined as a working schedule that involved working partly or entirely between 00:00 hr and 06:00 hr, at least three times per month. This definition included late eve- ning shifts, overnight shifts and early morning shifts. The refer- ence group consisted of men who were ever employed but had never performed night shift work for more than a year. Cumu- lative duration of night shift work was calculated as the total number of years worked at night ( P jobs(years night work/job). Subjects were allocated to exclusive categories by time since last exposure prior to diagnosis (cases) or interview (controls). The corresponding ethics committees of the participating centres and hospitals reviewed the protocol of the study. All subjects signed informed consent. The data that support the findings of our study are available from the corresponding author upon reasonable request. We used unconditional logistic regression and estimated odds ratios by time since last exposure to shift work compared to never night shift workers adjusting for age, centre and education. In an additional analysis, we also adjusted for duration of shift work, since duration was shown International Journal of Cancer IJC Int. J. Cancer: 145, 2597–2599 (2019) © 2019 UICC Letter to the Editor