MRI is often used in the diagnostic workup ahead of breast cancer surgery, but new data from a phase 3 trial suggest the practice is unnecessary for many patients with early-stage disease.

The findings, presented at San Antonio Breast Cancer Symposium (SABCS) 2025, show that adding preoperative breast MRI to diagnostic mammography had no impact on 5-year locoregional recurrence among patients with early-stage hormone receptor-negative (HR-negative) cancer.

Other outcomes, including distant recurrence-free survival and overall survival, were also similar whether patients received pre-op MRI or not.

The assumption has been that MRI catches disease missed by mammography and that removing any additional lesions reduces locoregional recurrences, but that has never been convincingly shown, presenter Isabelle Bedrosian, MD, breast cancer surgeon at MD Anderson Cancer Center in Houston, told Medscape Medical News. 

There is still “clearly room” for pre-op MRI, Bedrosian said, such as for women who have a high-risk germline mutations or when mammography results are ambiguous. 

But, she said, “it should not be done routinely for every new breast cancer patient.”

The findings are based on 319 patients newly diagnosed with stage I or II HR-negative breast cancer who were eligible for lumpectomy and randomized equally to either mammography alone or mammography with MRI within 30 days.

The trial focused on patients with HR-negative disease due to their higher risk for recurrence, Bedrosian noted. Most patients (72%) had stage I tumors, and 93% were clinically node negative. Just less than 20% were HER2-positive.

At a median follow-up of 61 months, 93.2% of patients in the MRI arm and 95.7% with no MRI were free of locoregional recurrence. Distant recurrence-free survival was 94.2% with MRI and 94.4% without it, whereas overall survival was 92.9% with MRI vs 91.4% without.

The results held when the analysis was limited to patients with triple-negative and HER2-positive disease, as well as the small number of women younger than 50 at diagnosis — a group often thought to benefit the most from MRI because they are more likely to have dense breast tissue.

Bedrosian pointed out that pre-op MRI often leads to more biopsies and workups that delay surgery, increase mastectomy rates, and add to patient anxiety and costs — all of which could be worth it if patients’ long-term outcomes were improved.

But the data suggest otherwise. A 2023 randomized trial from Brazil reached similar conclusions, with pre-op MRI having no impact on local recurrence, overall survival, or reoperation rates — though it did increase mastectomies by 8%.

“I’m hopeful that as these data are internalized by the breast cancer community, we will start pulling back some of the MRIs that we’re doing,” Bedrosian said. “It’s the routine, reflexive use of this modality that’s an issue here.”

Some experts discussing the results at SABCS agreed. Carlos Arteaga, MD, who moderated a press briefing, said he was not surprised by the findings.

Mammography and ultrasound — which almost all the women in this trial had, although it wasn’t mandated — are likely enough to find high-risk lesions that need to be removed, said Arteaga, breast medical oncologist at UT Southwestern Medical Center in Dallas.

Adjuvant treatment is probably sufficient to clear the small, low-risk lesions that MRI often detects, he added.

Gaorav Gupta, MD, PhD, radiation oncologist at the University of North Carolina at Chapel Hill, made similar points. “We often get false positives with MRIs, or you end up doing additional procedures for very low-risk lesions,” he said. “The evidence is not really there to say that it benefits patients.”

Not everyone shared that take, however. “I don’t see this trial as practice-changing,” said trial discussant Bruce Mann, MBBS, PhD, breast surgeon at the University of Melbourne in Melbourne, Australia.

He told Medscape that the trial was underpowered to detect a small but clinically relevant benefit from pre-op MRI. It was powered for an 8% reduction in locoregional recurrence with MRI; Mann argued that a 3% target would have been more realistic but would require many more patients.

He also noted imbalances between the arms that might have put MRI at a disadvantage. For example, MRI patients were younger and trended towards lower use of chemotherapy.

Those issues, Mann said, “make it hard to generalize about anything.”

The National Cancer Institute funded the work. The speakers had no relevant disclosures. 

M. Alexander Otto is a physician assistant with a master’s degree in medical science and a journalism degree from Newhouse. He is an award-winning medical journalist who worked for several major news outlets before joining Medscape Medical News. Alex is also an MIT Knight Science Journalism fellow. Email: aotto@mdedge.com.