Pregnancy After Interrupting Endocrine Therapy Shows No Excess Breast Cancer Risk at 6 Years

Why This Study Matters

Breast cancer is the most common cancer diagnosed in young women, and a diagnosis at a young age brings distinct medical and psychosocial burdens — among them, whether and when to pursue pregnancy after treatment. For women with hormone receptor-positive (HR+) disease, that question is complicated by two facts: pregnancy is contraindicated during adjuvant endocrine therapy (ET), and the years-long duration of that therapy, combined with age-related fertility decline, narrows the window for a future pregnancy. A longstanding fear that pregnancy itself might reactivate or accelerate disease has further discouraged patients and clinicians from considering a treatment pause, despite retrospective data suggesting otherwise.

The prospective, single-arm POSITIVE trial enrolled young women with HR+ breast cancer who wished to interrupt adjuvant ET to attempt pregnancy, tracking their outcomes against a matched external control population. Early results at a median of 41 months found no short-term increase in breast cancer risk. This updated analysis extends follow-up to roughly six years — long enough to begin addressing concerns about late recurrence, a well-recognized feature of HR+ disease.

Study Design

POSITIVE is a prospective, multicentre, multinational, investigator-initiated single-arm trial — not a randomized comparison — coordinated by the International Breast Cancer Study Group (IBCSG) with the Alliance for Clinical Trials in Oncology and the Breast International Group. Because withholding pregnancy from women who wanted it was not considered ethical, the trial compared outcomes against an external control group of 1,499 eligible patients drawn from the SOFT/TEXT trials, using bootstrap-matching to generate comparable 5-year cumulative incidence estimates and a stratified Cox model to estimate the hazard ratio.

Eligible patients were 42 years of age or younger with stage I–III HR+ breast cancer who had completed 18–30 months of adjuvant ET and wished to interrupt treatment to attempt pregnancy. The protocol allowed up to two years of ET interruption — including a mandatory three-month washout — to attempt conception, deliver, and breastfeed if desired, after which resuming ET to complete a full 5–10 years of treatment was strongly recommended. Use of assisted reproductive technology (ART) was permitted.

Patient Population

Between December 2014 and December 2019, 518 women were enrolled, of whom 516 comprised the evaluable intention-to-treat population. Median age at enrollment was 37 years (range 27–43). Seventy-five percent of participants had no prior live births, and 93% had stage I or II disease at diagnosis. Before enrollment, patients had received a median of 23.4 months of adjuvant ET; 58% had received ovarian function suppression and 62% had received chemotherapy.

Primary Endpoint Results

The trial's primary endpoint was the breast cancer-free interval (BCFI) — time from enrollment to the first invasive local, regional, or distant recurrence, or contralateral invasive breast cancer — evaluated against a prespecified safety threshold of no more than 46 events after 1,600 patient-years of follow-up. At a median follow-up of 71 months in POSITIVE and 80 months in the external control cohort, the 5-year cumulative incidence of BCFI events was 12.3% (95% CI: 9.4% to 15.2%) in POSITIVE versus 13.2% (95% CI: 11.4% to 15.1%) in controls — a difference of -0.9 percentage points (95% CI: -4.2% to 2.6%). Roughly half of all events were distant recurrences; the difference in 5-year cumulative incidence of distant recurrence-free interval events between cohorts was -2.1% (95% CI: -4.5% to 0.4%).

A sensitivity analysis restricted to HER2-negative disease (382 POSITIVE patients vs. 1,192 external-control patients) produced a similar picture: 5-year BCFI incidence of 14.1% (95% CI: 10.5% to 17.8%) in POSITIVE vs. 13.1% (95% CI: 11.0% to 15.2%) in controls, a difference of 1.0 percentage points (95% CI: -3.2% to 5.2%).

A landmark analysis of 456 patients who remained disease-free at 18 months compared the 305 (67%) who reported a pregnancy by that point with the 137 (30%) who had not. The cumulative incidence of BCFI events 42 months post-landmark was 6.7% among those who became pregnant by 18 months versus 7.6% among those who did not. A multivariable, time-dependent adjusted Cox model confirmed no evidence that pregnancy increased breast cancer event risk, with a BCFI hazard ratio (pregnant vs. not pregnant) of 0.65 (95% CI: 0.37 to 1.14).

Longer-term follow-up of the POSITIVE trial demonstrates that temporary interruption of endocrine therapy for pregnancy, including use of fertility preservation, does not increase risk of breast cancer events. This is encouraging for the increasing number of young breast cancer patients and survivors who have not completed their reproductive plans.
— Discussion, POSITIVE trial updated results, Annals of Oncology (2026)

Subgroup Analyses

Because POSITIVE has no treatment-arm comparison, its "subgroups" reflect patient and treatment characteristics within the POSITIVE cohort itself, or comparisons to the external control group, rather than randomized treatment-arm comparisons:

Where the source describes a relationship qualitatively rather than with an exact statistic, this article reports it the same way rather than inventing a number.

Safety Profile

Because POSITIVE has no comparator drug arm, "safety" here refers to two things: whether the breast cancer event rate exceeded its prespecified threshold, and the safety of pregnancy and childbirth for mothers and offspring. The prespecified stopping threshold of no more than 46 breast cancer events after 1,600 patient-years was not exceeded in either the original or the updated analysis.

No adverse-event rates for the endocrine therapies themselves are reported in this updated-results paper, since the analysis focuses on longer-term breast cancer outcomes and reproductive outcomes rather than on-treatment toxicity.

Interpretation and Broader Context

These updated results extend the reassurance offered by the original POSITIVE analysis from a median of 41 months to nearly six years of follow-up — a meaningful step given that HR+ breast cancer carries a well-documented risk of late recurrence that can persist for a decade or more. The authors note the findings are consistent with prior retrospective evidence, including a meta-analysis of more than 112,000 patients with a history of breast cancer in which those who became pregnant afterward had better disease-free survival (hazard ratio 0.66, 95% CI 0.49–0.89) and overall survival (hazard ratio 0.56, 95% CI 0.45–0.68) than those who did not.

Among 429 patients with at least two years of disease-free follow-up, 82.1% eventually resumed endocrine therapy, with cumulative resumption reaching 50% by 27 months and 73.8% by 48 months — indicating that most participants who paused treatment did return to it, consistent with the trial being designed as a temporary rather than permanent interruption.

Limitations

POSITIVE is a single-arm, non-randomized trial: outcomes are compared against an external control cohort from the separate SOFT/TEXT trials rather than a concurrently randomized comparator, so unmeasured differences between the two populations could still affect the comparison. The population itself is highly selected — patients healthy enough and motivated enough to enroll in a fertility-focused clinical trial — a caveat relevant to how broadly the findings generalize to unselected patients considering the same decision. The authors also note that continued follow-up remains warranted given the persistent risk of late recurrence in HR+ disease, since even six years may not be long enough to fully characterize recurrence risk in this population.