The Influence of Number and Timing of Pregnancies on Breast Cancer Risk for Women With BRCA1 or BRCA2 Mutations

Mary Beth Terry, Yuyan Liao, Karin Kast, Antonis C. Antoniou, Jasmine A. McDonald, Thea M. Mooij, Christoph Engel, Catherine Nogues, Bruno Buecher, Veronique Mari, Jessica Moretta-Serra, Laurence Gladieff, Elisabeth Luporsi, Daniel Barrowdale, Debra Frost, Alex Henderson, Carole Brewer, D. Gareth Evans, Diana Eccles, Jackie CookKai-ren Ong, Louise Izatt, Munaza Ahmed, Patrick J. Morrison, Charlotte J. Dommering, Jan C. Oosterwijk, Margreet G. E. M. Ausems, Mieke Kriege, Saundra S. Buys, Irene L. Andrulis, Esther M. John, Mary Daly, Michael Friedlander, Sue Anne McLachlan, Ana Osorio, Trinidad Caldes, Anna Jakubowska, Jacques Simard, Christian F. Singer, Yen Tan, Edith Olah, Marie Navratilova, Lenka Foretova, Anne-Marie Gerdes, Marie-Jose Roos-Blom, Brita Arver, Hakan Olsson, Nadine Andrieu*, EMBRACE, GENEPSO, Breast Cancer Family Registry (BCFR), HEBON, Kathleen Cuningham Foundation Consortium for Research into Familial Breast Cancer (kConFab), IBCCS, E. B. Gomez-Garcia, M. J. Blok

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


Background: Full-term pregnancy (FTP) is associated with a reduced breast cancer (BC) risk over time, but women are at increased BC risk in the immediate years following an FTP. No large prospective studies, however, have examined whether the number and timing of pregnancies are associated with BC risk for BRCA1 and BRCA2 mutation carriers.

Methods: Using weighted and time-varying Cox proportional hazards models, we investigated whether reproductive events are associated with BC risk for mutation carriers using a retrospective cohort (5707 BRCA1 and 3525 BRCA2 mutation carriers) and a prospective cohort (2276 BRCA1 and 1610 BRCA2 mutation carriers), separately for each cohort and the combined prospective and retrospective cohort.

Results: For BRCA1 mutation carriers, there was no overall association with parity compared with nulliparity (combined hazard ratio [HRc] = 0.99, 95% confidence interval [CI] = 0.83 to 1.18). Relative to being uniparous, an increased number of FTPs was associated with decreased BC risk (HRc = 0.79, 95% CI = 0.69 to 0.91; HRc =0.70, 95% CI = 0.59 to 0.82; HRc = 0.50, 95% CI = 0.40 to 0.63, for 2, 3, and >= 4 FTPs, respectively, P-trend <.0001) and increasing duration of breastfeeding was associated with decreased BC risk (combined cohort P-trend = .0003). Relative to being nulliparous, uniparous BRCA1 mutation carriers were at increased BC risk in the prospective analysis (prospective hazard ration [HRp] =1.69, 95% CI =1.09 to 2.62). For BRCA2 mutation carriers, being parous was associated with a 30% increase in BC risk (HRc =1.33, 95% CI =1.05 to 1.69), and there was no apparent decrease in risk associated with multiparity except for having at least 4 FTPs vs. 1 FTP (HRc =0.72, 95% CI = 0.54 to 0.98).

Conclusions: These findings suggest differential associations with parity between BRCA1 and BRCA2 mutation carriers with higher risk for uniparous BRCA1 carriers and parous BRCA2 carriers.

Original languageEnglish
Article number078
Number of pages14
JournalJNCI Cancer Spectrum
Issue number4
Publication statusPublished - Oct 2018



Cite this