Pregnancy After Breast Cancer: New Evidence Shows It's Safe for Women with Hormone Receptor-Positive Disease

Table of Contents

• Key Points
• Introduction: Why This Research Matters
• Study Methods: How the Research Was Conducted
• Key Findings: What the Data Showed
• No Increase in Cancer Recurrence (Disease-Free Survival)
• Better Overall Survival in Women Who Became Pregnant
• The "Healthy Mother Effect" and Other Safeguards
• The POSITIVE Trial: A Landmark Prospective Study
• The Challenge of Hormone Therapy and Fertility
• Study Limitations: What This Research Couldn't Prove
• What This Means for Patients and Their Doctors
• Frequently Asked Questions
• Source Information

Key Points

• A review of 8 studies of 3,805 patients with hormone receptor-positive early breast cancer found no increase in cancer recurrence among women who became pregnant after treatment.
• In 987 patients, disease-free survival was essentially the same for women who conceived and those who did not, with a hazard ratio of 0.96.
• In 3,504 patients, overall survival was better in the pregnancy group, but this may partly reflect the healthy mother effect rather than pregnancy itself.
• The POSITIVE trial of 518 patients found no significant difference in breast cancer-free interval after temporarily interrupting endocrine therapy to attempt pregnancy, though follow-up was short.

Introduction: Why This Research Matters

Breast cancer is the most frequent type of cancer diagnosed in women of reproductive age. Thanks to major improvements in treatment, patients who have had breast cancer now make up the largest group of young-onset cancer survivors. This means that survivorship — helping women live well after cancer — has become a crucial part of the care plan.

For young women who still hope to have children, the situation can be deeply stressful. Anticancer treatments can cause premature ovarian insufficiency (POI), meaning the ovaries stop working before age 40, and can otherwise impair fertility. These are potential long-term side effects with a major negative impact on life after cancer. In fact, relatively few women become pregnant after breast cancer, in part because of these medical barriers and in part because of fear.

Adding to the challenge, women in many countries are having their first pregnancy at older ages than previous generations. At the same time, the incidence of young-onset breast cancer is growing in many parts of the world. The result is that more and more women are being diagnosed with breast cancer before they have completed their childbearing plans. Discussing the risk of POI and available options to preserve fertility is now recognized as a key component of "oncofertility" counseling — the integration of fertility preservation into cancer care — for every woman diagnosed during her reproductive years.

Despite growing evidence in recent years that conceiving after breast cancer treatment — either naturally or through fertility preservation techniques — is safe, significant concerns have remained for one specific patient group: women with hormone receptor-positive (HR-positive) breast cancer. These are tumors that grow in response to estrogen or progesterone. The concerns are understandable. First, patients with HR-positive disease are typically advised to take endocrine (anti-hormone) therapy for 5 to 10 years after diagnosis, and pregnancy during endocrine therapy is contraindicated. Historically, patients have been told to delay pregnancy until endocrine therapy is finished, which further shrinks the already narrow window for fertility. Second, because breast cancer can be a hormone-driven tumor, and pregnancy involves very high concentrations of female hormones, there has been concern that pregnancy might increase the risk of recurrence. Third, HR-positive breast cancer carries a constant, long-term risk of recurrence — even after more than 10 years — so long follow-up is needed to fully assess the safety of pregnancy.

This systematic review and meta-analysis was designed to provide updated, high-quality evidence on whether pregnancy after treatment for HR-positive early breast cancer is safe, by pooling data from all available studies that compared survival outcomes between women who became pregnant and those who did not.

Study Methods: How the Research Was Conducted

The researchers followed the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), which are the internationally recognized standards for conducting and reporting systematic reviews. They performed a comprehensive literature search of three major medical databases — Medline, Embase, and the Cochrane Library — with no language or date restrictions, up to March 31, 2023.

The search strategy used terms combining "breast neoplasms" with "pregnancy," "pregnancies," "conception," or "gestation." Relevant articles were cross-referenced to make sure no pertinent records were missed. The study protocol was registered in advance in the PROSPERO database (registration number CRD42023394232), which promotes transparency in research.

To be included, studies had to meet specific criteria:

• Be a retrospective or prospective case-control study, a cohort study, or a clinical trial
• Compare survival outcomes of premenopausal women with or without a pregnancy after prior diagnosis of hormone receptor-positive early breast cancer
• Report information on one or more outcomes of interest: overall survival (OS), disease-free survival (DFS), or breast cancer recurrences
• Provide enough data to estimate the hazard ratio (HR) with 95% confidence intervals (CIs) — the statistical measure of risk

Studies were excluded if they were case reports or case series with fewer than 10 patients, if they dealt with breast cancer diagnosed during pregnancy or within 1 year after pregnancy, or if they were ongoing studies with no published results.

From an initial pool of 13,816 records identified from databases, 6,020 duplicate records were removed, leaving 7,796 records for screening. After excluding 7,379 records by title or publication type (such as abstracts and conference presentations), 417 full-text articles were sought for retrieval. Of those, 409 were excluded for reasons such as pregnancy-associated breast cancer (198 studies), pregnancy occurring before the breast cancer diagnosis (100 studies), no hormone receptor-positive breast cancer data (78 studies), case reports or case series with fewer than 10 patients (9 studies), no survival data (9 studies), overlapping populations (8 studies), and studies that had been updated by newer publications (7 studies).

Ultimately, 8 studies met all criteria and were included in the final analysis. These 8 studies collectively included 3,805 patients with hormone receptor-positive invasive early breast cancer. Of these, 1,285 had a pregnancy after their breast cancer diagnosis, and 2,520 did not.

All 8 included studies were retrospective cohort studies — meaning they looked back in time at medical records and databases to follow the outcomes of patients who did and did not become pregnant. This is an important point: none of the studies in the meta-analysis were prospective randomized trials, which are considered the gold standard in medical research. However, all 8 studies corrected for "guarantee-time bias" (explained below), which strengthens the reliability of their findings.

Two researchers independently extracted data from each study, including first author, year of publication, study design, number of women in each cohort, survival outcomes, and matching criteria used to choose control patients. When multiple publications came from the same study population, only the most updated version was included. Quality assessment and risk of bias were evaluated using the Newcastle-Ottawa Assessment Scale (NOS), which assigns up to 9 points across three domains of bias risk for case-control or cohort studies. Studies were classified as having low, moderate, or high risk of bias based on their scores.

Statistical analysis was performed using a random-effects model (the DerSimonian and Laird method), which accounts for variability between studies. The Higgins I² index was used to measure inconsistency between studies (heterogeneity), and Egger's asymmetry test was used to assess publication bias — the tendency for positive results to be published more often than negative ones. Sensitivity analyses were also conducted to check whether any single study was driving the overall results.

Key Findings: What the Data Showed

The characteristics of the 8 included studies varied considerably. The mean age at breast cancer diagnosis ranged from 31.0 to 34.2 years in patients who later became pregnant, and from 32.0 to 40.0 years in those who did not. Studies came from around the world, including the United States, Canada, Europe, Taiwan, Scotland, and Korea.

Notably, the studies spanned different eras of cancer care. The earliest study included patients diagnosed as far back as 1981 (Scotland), while the most recent studies included patients through 2017 (Scotland) and 2015 (United States). This long time span means the results reflect real-world outcomes across decades of treatment evolution.

The median follow-up time ranged from 4.3 to 15.8 years in the pregnancy cohorts and from 3.8 to 14.7 years in the non-pregnancy cohorts. These are relatively long follow-up periods, which is especially important for hormone receptor-positive breast cancer, a type of disease known to carry a persistent risk of late recurrence.

No Increase in Cancer Recurrence (Disease-Free Survival)

Disease-free survival (DFS) refers to the length of time after treatment during which a patient survives without any signs of the cancer returning. Three studies, involving 987 patients total (286 of whom had a subsequent pregnancy), reported on this outcome.

The pooled analysis found no difference in DFS between patients who became pregnant and those who did not, with a hazard ratio (HR) of 0.96 (95% CI 0.75–1.24, P = 0.781). In plain language, this means that women who became pregnant had essentially the same risk of their cancer recurring as women who did not become pregnant.

It helps to understand what a hazard ratio means. A hazard ratio compares the risk of an event (like cancer recurrence or death) between two groups over time. An HR of 1.0 means no difference between the groups. An HR below 1.0 means the event is less likely in the first group, and an HR above 1.0 means it's more likely. In this case, the HR of 0.96 is very close to 1.0, confirming no harmful effect. The 95% confidence interval (0.75–1.24) also crosses 1.0, indicating the result is not statistically significant — consistent with no increased risk.

Reassuringly, the researchers found no heterogeneity in this analysis (I² = 0.0%, P = 0.506), meaning the results across the three studies were highly consistent with one another. There was also no evidence of publication bias (P = 0.406), and sensitivity analyses confirmed that no single study had significantly influenced the overall result.

Better Overall Survival in Women Who Became Pregnant

Overall survival (OS) — the length of time patients live, regardless of whether the cancer returns — was reported in six studies involving 3,504 patients, of whom 1,193 had a subsequent pregnancy.

The results here were striking. Patients in the pregnancy cohort had a statistically significant better overall survival than those in the non-pregnancy cohort, with an HR of 0.46 (95% CI 0.27–0.77, P < 0.005). This translates to a 54% relative reduction in the risk of dying among women who became pregnant after their breast cancer diagnosis.

It is important to emphasize that this does not mean pregnancy itself makes women live longer. Rather, as discussed below, this finding likely reflects a combination of the "healthy mother effect" (women who feel well enough to conceive are generally healthier) and possibly the careful medical monitoring that comes with pregnancy. Still, the key message for patients is clear: becoming pregnant did not harm survival — and it may even be a marker of better overall health.

Unlike the DFS analysis, the OS analysis showed significant heterogeneity (I² = 69.6%, P = 0.006), meaning the results varied more between studies. This is not surprising given differences in study populations, time periods, and healthcare systems. However, there was no evidence of publication bias (P = 0.259), and sensitivity analyses confirmed the findings were stable and not dependent on any single study.

Looking at the individual studies provides helpful context. In the six studies reporting OS:

• Valentini et al. 2013 (Canada, USA, Europe, Asia; patients diagnosed 1985–2010): 21 women with pregnancy vs. 66 without; assessed 15-year survival rate; high risk of bias
• Lambertini et al. 2018 (Europe; patients diagnosed before 2007): 194 vs. 492; median time from diagnosis to pregnancy 4.7 years; follow-up 9.6 years; low risk of bias
• Chuang et al. 2020 (Taiwan; 2002–2014): 87 vs. 311; median time to pregnancy 3.31 years; follow-up 4.3 vs. 3.8 years; moderate risk of bias
• Rauh-Hain et al. 2022 (USA; 2000–2015): 240 vs. 273; median time to pregnancy 2.72 years; follow-up 9.3 years; moderate risk of bias
• Anderson et al. 2022 (Scotland; 1981–2017): 102 vs. 612; median time to pregnancy 4.1 years; follow-up 15.8 vs. 14.7 years; low risk of bias
• Bae et al. 2022 (Korea; 2004–2014): 549 vs. 557; median time to pregnancy 3.3 years; follow-up 8.2 years; high risk of bias

The two other included studies — Nye et al. 2017 (USA; 2000–2010; 32 vs. 29 patients; assessed DFS; moderate risk of bias) and Lambertini et al. 2020 (global; 2000–2012; 60 vs. 180 patients; assessed DFS and pregnancy outcomes; low risk of bias) — contributed to the DFS analysis.

The "Healthy Mother Effect" and Other Safeguards

A critical issue in any research on pregnancy after cancer is something called the "healthy mother effect." The idea is simple but important: women who feel well enough to pursue pregnancy and actually conceive may be healthier overall than women who do not. Women with poorer prognoses or more aggressive disease would be less likely to attempt pregnancy, and this natural selection can make pregnancy look more favorable than it truly is.

This effect was a major reason why earlier studies on pregnancy after breast cancer were viewed skeptically. Over the years, however, researchers have developed statistical techniques to address this bias. One key approach involves correcting for guarantee-time bias — the phenomenon that arises when a woman must survive long enough after diagnosis to become pregnant. If you simply compare women who eventually became pregnant with women who did not, the pregnant group has a built-in "head start" because they've already survived long enough to conceive. Analyses that don't account for this can unfairly favor the pregnancy group.

All 8 studies included in this meta-analysis explicitly corrected for guarantee-time bias, which substantially strengthens the overall conclusion that pregnancy after HR-positive breast cancer is safe. The better overall survival seen in the pregnancy cohort may still partly reflect the healthy mother effect, even after statistical adjustment, but the key point is that pregnancy was not associated with any worsening of survival.

The POSITIVE Trial: A Landmark Prospective Study

The findings of this meta-analysis align with results from a groundbreaking prospective clinical trial called POSITIVE (Pregnancy Outcome and Safety of Interrupting Therapy for Women with Endocrine Responsive Breast Cancer; ClinicalTrials.gov identifier NCT02308085). This trial, first reported in 2023, tested whether it is safe for women with HR-positive early breast cancer to temporarily interrupt their adjuvant endocrine therapy to attempt pregnancy.

The POSITIVE trial enrolled 518 patients under age 42 with HR-positive early breast cancer. Patients had received at least 18–30 months of adjuvant endocrine therapy before attempting pregnancy. Early results showed that the breast cancer-free interval (BCFI) — the time before any breast cancer event — did not differ significantly from that of a control group of patients from the SOFT and TEXT trials (HR 0.81, 95% CI 0.57–1.15).

The 3-year incidence of breast cancer events was 8.9% (95% CI 6.3% to 11.6%) in the treatment-interruption group, compared with 9.2% (95% CI 7.6% to 10.8%) in the control cohort. These numbers are nearly identical, providing strong prospective support for the safety of attempting pregnancy after temporary interruption of endocrine therapy.

However, the POSITIVE trial has two important limitations that this meta-analysis helps to address. First, the trial enrolled mostly women with stage I or II disease — only 6% had stage III disease — so its results may not apply equally to all patients. Second, the median follow-up of 41 months is relatively short, which is a significant limitation for HR-positive breast cancer, where late recurrences can occur even after 10 or more years from diagnosis. By contrast, the studies in this meta-analysis had median follow-ups of 4.3 to 15.8 years in the pregnancy cohort, allowing the capture of at least some late recurrences.

The Challenge of Hormone Therapy and Fertility

One of the most complex aspects of fertility counseling for patients with HR-positive breast cancer is the management of adjuvant endocrine therapy. This therapy — typically tamoxifen, an aromatase inhibitor, or tamoxifen combined with a gonadotropin hormone-releasing hormone agonist (GnRHa) — is recommended for 5 to 10 years to reduce the risk of recurrence. Pregnancy during endocrine therapy is contraindicated because these medications pose risks to a developing baby and because the hormonal environment they create is not compatible with a healthy pregnancy.

Only about half of the included studies reported details on the use of adjuvant endocrine therapy, and only three specified the type of treatment used. In two studies, all patients on endocrine therapy received tamoxifen alone. In a third study (Lambertini et al. 2020), among the pregnancy cohort, 28.9% of patients received tamoxifen alone, 57.7% received tamoxifen combined with a GnRHa, and 7.7% received a GnRHa combined with an aromatase inhibitor.

The duration of adjuvant endocrine therapy was reported in two studies, and in both, it was significantly shorter in the pregnancy cohort than in the non-pregnancy cohort — 50 vs. 60 months (P < 0.001) in one study, and 20.9 vs. 42.3 months (P = 0.008) in the other. This makes clinical sense: women who wished to conceive understandably interrupted or completed their endocrine therapy earlier to maximize their fertility window. In one study that reported this information (Lambertini et al. 2020), the median time from breast cancer diagnosis to subsequent pregnancy was 6.3 years.

Historically, patients with breast cancer who wished to conceive were counseled to wait at least 2 years from the time of diagnosis. This recommendation came from older observational studies showing that the risk of recurrence is highest in the first 2 years after diagnosis, and that this is also the time needed to recover from chemotherapy-induced ovarian toxicity. For patients with HR-positive disease, the traditional advice went further: wait until completing adjuvant endocrine therapy, effectively delaying pregnancy by at least 5 years. This advice was based on a lack of data on the safety of temporarily stopping treatment to attempt pregnancy, and on the known benefits of endocrine therapy, particularly for higher-risk patients who may also receive ovarian function suppression for 5 years.

The results of this meta-analysis, together with the POSITIVE trial, support a shift in this counseling approach. Women who desire pregnancy after HR-positive breast cancer can be reasonably reassured that doing so does not appear to carry a detrimental effect on their cancer prognosis.

Study Limitations: What This Research Couldn't Prove

While this meta-analysis provides the most comprehensive evidence to date on this topic, it has limitations that are important to understand:

• All included studies were retrospective cohort studies. No randomized controlled trials were included in the meta-analysis itself. Retrospective studies can have unmeasured confounding factors — differences between groups that might influence outcomes and were not fully accounted for.
• Significant heterogeneity in the OS analysis. The I² value of 69.6% indicates substantial variation between studies in the overall survival results. This heterogeneity could stem from differences in patient populations, treatment approaches across countries and time periods, and study methodologies.
• The "healthy mother effect" cannot be fully excluded. Even with statistical corrections for guarantee-time bias, women who become pregnant after cancer may be inherently healthier than those who do not, which could partly explain the improved overall survival seen in the pregnancy group.
• Incomplete data on endocrine therapy. Only half of the studies reported details on endocrine therapy use, and only two reported its duration. The fact that pregnancy cohorts had significantly shorter endocrine therapy exposure is noteworthy, as it raises questions about whether outcomes might differ if hormonal therapy is interrupted prematurely.
• Limited data on tumor biology. In two studies, the biological characteristics of the breast cancer were not reported, and tumor subtype was instead inferred from the use of adjuvant endocrine therapy. This could introduce misclassification.
• Short follow-up in some studies. While the median follow-up in the pregnancy cohort ranged up to 15.8 years in the longest study, some studies had shorter follow-up, which could miss late recurrences typical of HR-positive disease.
• No data on fertility preservation techniques. The meta-analysis did not separately analyze outcomes for women who conceived through assisted reproductive technologies versus natural conception.

What This Means for Patients and Their Doctors

So what should a young woman with HR-positive breast cancer who dreams of having a baby take away from this research? The message is genuinely encouraging, and it reflects a meaningful shift in medical thinking.

Here are the key takeaways:

1. Pregnancy appears safe after HR-positive breast cancer. The meta-analysis found no increased risk of recurrence (DFS HR 0.96, P = 0.781) and no detriment to overall survival (OS HR 0.46, P < 0.005) in women who became pregnant after treatment.
2. Counseling should be individualized. The decision to pursue pregnancy after breast cancer should be made jointly between the patient and her oncology team, taking into account the specific characteristics of her cancer, her age, her prior treatments, and her personal values and preferences.
3. Timing matters. During active endocrine therapy, pregnancy is contraindicated. The POSITIVE trial supports the option of temporarily interrupting endocrine therapy after at least 18–30 months of treatment to attempt pregnancy, under close medical supervision, though its findings are still early (median follow-up 41 months).
4. Multidisciplinary care is essential. Fertility counseling should involve oncologists, fertility specialists, and maternal-fetal medicine experts working together. This ensures that both cancer care and reproductive goals are addressed optimally.
5. Women should not be categorically counseled against pregnancy. The authors note that despite growing evidence of safety, many patients are still being told not to attempt conception. This meta-analysis provides stronger data to support a more balanced, patient-centered conversation.

It is also worth acknowledging the emotional dimension. For many young women, fertility is not just a medical issue — it is deeply tied to identity, hopes for the future, and recovery from the trauma of cancer. A cancer diagnosis during childbearing years can feel like a double loss: the loss of health and the potential loss of the ability to have a family. Research like this helps restore options and hope.

Ultimately, this meta-analysis of 3,805 patients with HR-positive early breast cancer adds to the growing body of evidence that pregnancy after breast cancer can be considered safe. While it does not prove that pregnancy improves survival — and the healthy mother effect likely contributes to the better survival rates seen in the pregnancy group — it provides strong reassurance that pregnancy does not worsen outcomes. For young women with HR-positive breast cancer who wish to become mothers, the message from this research is hopeful: your dreams of having a family do not have to be abandoned.

Frequently Asked Questions

I have hormone receptor-positive breast cancer. Is it safe to get pregnant after treatment?

A review of 8 studies of 3,805 patients with hormone receptor-positive early breast cancer found no increase in cancer recurrence among women who became pregnant afterward. Disease-free survival was essentially the same as in women who did not conceive. Overall survival was better in the pregnancy group, though this may partly reflect the healthy mother effect.

What does a hazard ratio of 0.96 for disease-free survival mean?

A hazard ratio compares the risk of an event, such as cancer recurrence, between two groups over time. A value of 1.0 means no difference. In this analysis of 987 patients, the hazard ratio for disease-free survival was 0.96, very close to 1.0, meaning women who became pregnant had essentially the same recurrence risk as those who did not.

Does pregnancy after breast cancer improve my survival?

In this analysis of 3,504 patients, women who became pregnant had better overall survival, with a hazard ratio of 0.46. However, this does not mean pregnancy itself makes women live longer. It likely reflects the healthy mother effect, where women well enough to conceive are generally healthier, and possibly closer medical monitoring during pregnancy.

Can I pause hormone therapy to try to conceive?

Pregnancy during endocrine therapy is contraindicated. The POSITIVE trial enrolled 518 patients under 42 with hormone receptor-positive early breast cancer who had received at least 18 to 30 months of endocrine therapy, then interrupted it to attempt pregnancy. Early results showed no significant difference in breast cancer-free interval versus a control group, though follow-up was short.

How long after diagnosis did women in these studies become pregnant?

Timing varied across the 8 studies. Median time from diagnosis to pregnancy ranged from about 2.7 to 6.3 years. Historically, patients were advised to wait at least 2 years, and those with hormone receptor-positive disease were often told to wait until completing 5 to 10 years of endocrine therapy. This research supports a shift in that counseling.

What are the limitations of this research on pregnancy after breast cancer?

All 8 included studies were retrospective, so unmeasured differences between groups may exist. The overall survival analysis showed significant variation between studies. Some studies had short follow-up, which could miss late recurrences typical of hormone receptor-positive disease. Data on endocrine therapy and fertility preservation techniques were incomplete, and the healthy mother effect cannot be fully excluded.

I have hormone receptor-positive breast cancer and want to have a baby — when should I get a second opinion about pregnancy after treatment?

A second opinion is reasonable when you are weighing pregnancy against endocrine therapy timing. Pooled data from 3,805 patients with hormone receptor-positive early breast cancer found no increase in recurrence (hazard ratio 0.96) and no worse overall survival in women who conceived afterward. The POSITIVE trial supports temporarily interrupting endocrine therapy after 18–30 months to attempt pregnancy under close supervision. Because timing, tumor characteristics, and prior treatments vary, an independent review can help confirm whether that option fits your situation. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original article title: Safety of pregnancy after breast cancer in young women with hormone receptor-positive disease

Authors: L. Arecco, E. Blondeaux, M. Bruzzone, M.M. Latocca, E. Mariamidze, S. Begijanashvili, E. Sokolovic, G. Gentile, G. Scavone, S. Ottonello, A. Boutros, I. Vaz-Luis, C. Saura, R.A. Anderson, I. Demeestere, H.A. Azim Jr, E. de Azambuja, F.A. Peccatori, L. Del Mastro, A.H. Partridge, & M. Lambertini

Journal: ESMO Open, Volume 8, Issue 6, 2023 (article 102031)

Publication date: Available online 23 October 2023

DOI: 10.1016/j.esmoop.2023.102031

Study registration: PROSPERO number CRD42023394232

This patient-friendly article is based on peer-reviewed research. The original study was published as an open-access article under the CC BY-NC-ND license by the European Society for Medical Oncology. Patients who have questions about pregnancy after breast cancer should consult their oncology and fertility care teams for personalized medical advice.