Time-to-pregnancy in patients with previous breast cancer and unexposed women: a prospective exposed-unexposed cohort study

Abstract

Background: Data on fertility after breast cancer (BC) relative to the general population are lacking. This study aimed to compare the time-to-pregnancy between women with and without prior BC seeking to become pregnant.

Methods: We conducted a prospective exposed-unexposed cohort study between March 13, 2018 and June 27, 2019, recruiting participants via the collaborative network Seintinelles. Women aged 18-43 years with a history of localized BC without relapse (exposed) were compared to women without BC (unexposed). Follow-up data were collected every six months over three years. The primary endpoint, time-to-pregnancy, was analyzed using Kaplan-Meier survival analysis with inverse probability weighting. Censoring was performed if women stopped trying to conceive for personal reasons, were lost to follow-up before conception, or completed the study without achieving pregnancy or using assisted reproductive technologies (ART).

Findings: Among 4351 women enrolled, 642 sought pregnancy during the study period (76 exposed, 566 unexposed). Among them, 50 exposed (65.8%) and 402 weighted unexposed women (weighted percentage 71.0%) became pregnant at least once. Median time-to-pregnancy was 5.0 months, 95% CI [3.0-7.0] for exposed and 3.0 months, 95% CI [2.0-5.0] for unexposed women (difference in median time-to-pregnancy: 2.0 months, 95% CI [-2.5; 5]). Two years after starting to seek pregnancy, 74.9% of weighted unexposed women and 74.1% of exposed women obtained a pregnancy. Overall, 15 exposed women (19.8%) used either ART methods (n = 7, 9.2%) or cryopreserved material (n = 8, 10.5%) and 25 unexposed women (4.4%) used ART.The median time-to-pregnancy was 3.0 months for both exposed and unexposed women (95% CI [2.0; 5.0] and [2.0; 4.0] respectively) (difference in median time-to-pregnancy: -0.0 months [-2.0; 3.0]) in women who sought pregnancy spontaneously, and 14.0 months for exposed (95% CI [6.0-27.0]) and 17.6 months for unexposed women (95% CI [17.6-30.0]) in women seeking pregnancy with ART.

Interpretation: In this study from a French collaborative research network, we found no strong evidence of a largely reduced time-to-pregnancy in exposed women seeking to become pregnant compared with unexposed women. Further research on fertility outcomes in the broader population of BC survivors is warranted.

Funding: The FEERIC study was funded by Institut National du Cancer (InCA), InCA-SHS, grant no. 2016-124, and is part of the Young Breast Cancer Project, funded by Monoprix.

Keywords: Breast cancer; Infertility; Pregnancy; Survivors; Time-to-pregnancy.

Fig. 1

Flow chart for the study cohort.

Fig. 2

Time-to-pregnancy comparison between exposed and unexposed women. All 642 women seeking pregnancy during the study period (76 exposed, 566 unexposed) are displayed in this Kaplan Meier curve. These curves depict the proportion of women attempting to get pregnant who became pregnant over time, with confidence intervals derived from 1000 bootstrap samples. The curve was adjusted by inverse probability weighting (IPW). Only the unexposed group was weighted using (IPW), while the exposed group remained unweighted. Numbers at risk are shown as unweighted (raw counts at each time point) to preserve transparency about how many participants remain under observation. The shaded areas represent 95% confidence bands estimated with 1000 bootstraps. This approach balanced the distribution of 14 confounding variables between exposed and unexposed women, making the weighted unexposed group comparable to the exposed group. Most women who attempted to get pregnant succeeded during the study period. In total, 19 exposed and 111 unexposed women were censored due to a lack of response or because they did not become pregnant before the end of the study. An additional 7 exposed and 38 unexposed women were censored because they changed their minds about trying to become pregnant. None of the 76 patients with prior BC who started trying to get pregnant after inclusion in the study experienced a relapse. Considering unweighted numbers, 417 unexposed women conceived during the study. Abbreviation: IPW: inverse probability weighting.

Fig. 3

Comparison of time-to-ART use between exposed and unexposed women. All 642 women seeking pregnancy during the study period (76 exposed, 566 unexposed) are displayed in this Kaplan Meier curve. Follow-up begins at the time women sought pregnancy. In these Kaplan–Meier, only the unexposed group was weighted using (IPW), while the exposed group remained unweighted. Numbers at risk are shown as unweighted (raw counts at each time point) to preserve transparency about how many participants remain under observation. The shaded areas represent 95% confidence bands estimated with 1000 bootstraps. Abbreviation: ART: assisted reproductive technologies.

Fig. 4

Time-to-pregnancy comparison between exposed and unexposed women. A: Time-to-pregnancy for women attempting to get pregnant naturally. Only the 608 women who sought pregnancy without resort to ART or cryopreserved material (61 exposed, 547 unexposed) are displayed in this Kaplan Meier curve; B: Time-to-pregnancy for women using ART, including cryopreserved material. Only the 34 women who sought pregnancy with resort to ART or cryopreserved material (15 exposed, 19 unexposed) are displayed in this Kaplan Meier curve. In these Kaplan–Meier, only the unexposed group was weighted using (IPW), while the exposed group remained unweighted. Numbers at risk are shown as unweighted (raw counts at each time point) to preserve transparency about how many participants remain under observation. The shaded areas represent 95% confidence bands estimated with 1000 bootstraps. Abbreviation: ART: assisted reproductive technologies.