Real-world duration of interruptions in imaging and endocrine therapy (ET) after pregnancy in early-stage estrogen receptor (ER) positive breast cancer (BC).

Abstract

11157 Background: Pre-menopausal women with early-stage ER-positive BC often desire the option for future fertility. The POSITIVE trial demonstrated that a pause in ET to conceive is safe. However, the real-world durations of these pauses, as well as how many patients resume ET, are unclear. Additionally, there are limited data about when and if women reestablish imaging surveillance after childbirth. Methods: We generated a cohort of BC patients with a pregnancy-related ICD diagnosis from the Oncoshare registry. This registry merges EMR and California Cancer Registry data for patients treated in the Stanford Health Care Alliance which includes an academic hospital, a community hospital, and a community practice network. Included patients had ≥1 pregnancy after a diagnosis of stage 0-3 ER-positive BC. Chart review confirmed pregnancy and was used to abstract treatment information. Data are presented as unadjusted percentages or medians with interquartile ranges (IQR). Among patients without bilateral mastectomies, we compared time from delivery to first imaging (MRI or mammogram) for those who breastfed vs not and those who restarted ET vs not using the Wilcoxon rank sum test. Results: Of the 317 charts reviewed, 71 were included. Most exclusions were due to no pregnancy (46%) or ER-negative disease (25%). Year of diagnosis ranged from 1994 to 2020, with the majority (79%) from 2010 to 2020. Median age at diagnosis was 33 years (IQR 30-35). The distribution of stages was 18% in situ disease, 51% stage 1, 17% stage 2, 14% stage 3. 23% (16/71) of women never started ET, with 14 attributing this decision to desiring pregnancy. Among the 55 women who started ET, 75% received tamoxifen without ovarian suppression (OS). After 2015, use of OS became more prevalent (45% of ET), coinciding with the emergence of the SOFT/TEXT data. The median time from ET onset to a pause for pregnancy was 32 months (IQR 22-53), and the median ET pause to delivery was 23 months (IQR 14-48). We recorded 90 pregnancies resulting in 85 live births. After delivery of their first child, the median follow-up was 3.8 years (IQR 2.0-6.8). 40% of patients never restarted ET. Those who did restart did so at a median of 5 months (IQR 3-11) post-delivery. Imaging surveillance resumed for 42% (27/64) at a median of 6 months post-delivery [IQR 3-11]. Time from delivery to imaging was similar between those who restarted ET and those who did not (p=0.91), and between those who breastfeed and those who did not (p=0.42). Conclusions: These data support prior work showing that fertility concerns strongly influence adjuvant therapy choice in ER-positive BC. The ET resumption rate after pregnancy was lower in our analysis compared to the POSITIVE trial (73% vs. 40%). Only 42% of women resumed surveillance imaging. This points to clear gaps between the clinical trial population and the real-world clinical setting.