Contrast-enhanced mammography sized breast cancer best but did not cut reoperations

Design

Randomised controlled trial, two Swedish breast centres

Population

415 women with early breast cancer eligible for primary surgery

Primary outcome

Treatment change of at least 10% due to CEM

Effect

8.4% (95% CI 5.3–13.1); reoperation 7.9% vs 10% (P = .39)

PROCEM randomised 415 women with early breast cancer suitable for primary surgery, at two Swedish breast centres between 2020 and 2024, to preoperative contrast-enhanced mammography or standard work-up with mammography, tomosynthesis and ultrasound. The primary outcome was a change in treatment in at least 10%.

Contrast mammography changed treatment in 8.4% (95% CI 5.3–13.1), so the trial missed its target. Mastectomy (17% vs 18%) and reoperation (7.9% vs 10%) did not differ significantly. Against histopathology, contrast mammography estimated malignant extent most closely (bias 3.5 mm) compared with mammography or tomosynthesis (4.8 mm) and ultrasound (10.7 mm), in nine of ten subgroups; DCIS was the exception.

The practical reading: contrast mammography is the most accurate of the available tools for sizing invasive cancer, and a reasonable alternative to breast MRI where MRI is scarce. But routine use for every patient did not reduce reoperations or mastectomies, so selective use — dense breasts, lobular histology, discordant imaging — is better supported than blanket staging.

• Consider contrast-enhanced mammography for preoperative sizing where conventional imaging is discordant or limited
• Prioritise dense breasts and lobular histology when choosing whom to stage
• Do not rely on contrast mammography to size DCIS; use calcification extent
• Biopsy additional findings before they change the surgical plan
• Do not expect routine use to lower reoperation rates on its own

Why it matters

Separates better measurement from better outcomes, which is the question that justifies routine staging.

Don't overread it

More accurate sizing did not translate into fewer reoperations or mastectomies in this trial.

The statistics, in plain English

A treatment change of 8.4% with a confidence interval of 5.3–13.1% includes the 10% target, but the point estimate fell short, so the trial is formally negative. The size differences between modalities are a few millimetres; they matter most at margins and in multifocal disease.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.