STUDY QUESTIONIs gonadotrophin stimulation as part of IVF associated with an increased risk of relapse in breast cancer?SUMMARY ANSWERControlled ovarian stimulation (COS) in connection with IVF in women with previous breast cancer was not associated with an increased risk of breast cancer relapse.WHAT IS KNOWN ALREADYBreast cancer is the most common malignancy among women worldwide and the leading cause of cancer death among females. The use of COS with gonadotrophins with subsequent cryopreservation of oocytes or embryos in order to enhance the chances of pregnancy after cancer treatment is the current most established fertility preservation method for women with breast cancer. To date, there are only a few small retrospective hospital-based controlled studies evaluating the risk of breast cancer relapse in patients undergoing fertility preservation with or without COS, showing no evident risk of relapse in breast cancer after the use of gonadotoxic agents.STUDY DESIGN, SIZE, DURATIONThis was a retrospective, population-based cohort study comprising 5857 women with previous breast cancer of whom 337 were exposed to COS. Exposure (COS) and outcomes (relapse and death) were identified for all patients from 2005 to 2014 by assessing the National Quality Register for Assisted Reproduction, the Swedish Medical Birth Register, the National Patient Register, the Swedish Prescribed Drug Register, the Swedish Cause of Death Register, the National Breast Cancer Register and the Swedish Cancer Register. Matching according to set criteria was possible for 334 women, who constituted the control group. A total of 274 women had undergone IVF after completing breast cancer treatment and 63 women had undergone COS for fertility preservation at the time of breast cancer diagnosis.PARTICIPANTS/MATERIALS, SETTING, METHODSWomen aged 20–44 years previously diagnosed with breast cancer and exposed to COS were matched for age at breast cancer diagnosis ±5 years, tumour size and lymph node involvement with a non-exposed control group, including women with known T- and N-stages. In a subsequent analysis, the matched cohort was assessed by also including women with unknown T- and N-stages. A secondary analysis comprised the entire non-matched cohort, including all women with known T- and N-stages. Also here, a subsequent analysis included women with missing data for T- and N-stages. The risk of relapse in breast cancer was estimated as crude hazard ratios (HRs) and 95% CI using Cox proportional hazards models in the primary and secondary analyses where T- and N-stages were known: otherwise the risks of relapse were only given descriptively.MAIN RESULTS AND THE ROLE OF CHANCEIn the primary matched analysis, relapse occurred in 20 of 126 women exposed to COS (15.9%) compared with 39 of 126 (31.0%) in the control cohort (HR = 0.70; 95% CI 0.39–1.45; P = 0.22). In the subsequent analysis, also including women with unknown T- and N-stages, relapse occurred in 27 of 337 (8.0%) women having undergone COS compared with 71/334 (21.3%) among the non-exposed. In the secondary adjusted analysis, relapse occurred in 20 of 126 (15.9%) exposed women and in 918 of 3729 (24.6%) non-exposed women (HR = 0.81; 95% CI 0.49–1.33; P = 0.70). In the subsequent analysis, including unknown T- and N-stages, relapse occurred in 27 of 337 (8.0%) women in the exposed group and 1176 of 5520 (21.3%) in the non-exposed cohort.LIMITATIONS, REASONS FOR CAUTIONA substantial degree of missing data on important prognostic variables was a limitation, particularly when analysing the total cohort. Furthermore, data on confounding factors, such as BMI, were not completely covered. Another limitation was that a pre-specified variable for relapse was not in use for the majority of the National Breast Cancer Register. Furthermore, the follow-up time from available register data (2005–2014) is rather short. Finally, we cannot be sure whether the prognostic information from receptor status, showing a lower incidence in the exposed group, is representative. Information on T- and N-stages was missing in more than half of the patients.WIDER IMPLICATIONS OF THE FINDINGSIn this large, retrospective, matched cohort study, we found no increased risk of relapse in breast cancer among women who had been exposed to gonadotrophins as part of IVF. This is reassuring but might be confounded by the selection of a group of women with a more favourable prognosis than those not undergoing IVF. The present study strengthens previous findings by being large, national and register based. Its results are applicable to women undergoing fertility preservation as well as to those undergoing regular IVF treatment.STUDY FUNDING/COMPETING INTEREST(S)Supported in part by grants from the Swedish state under the agreement between the Swedish government and the county councils the ALF-agreement (ALFGBG-720291), The Assar Gabrielsson Fund (FB 15-20), The Breast Cancer Fund and the Swedish Association of Local authorities and Regions, SKR. There are no conflicts of interest to declare.TRIAL REGISTRATIONN/A