AI-based selection of individuals for supplemental MRI in population-based breast cancer screening: the randomized ScreenTrustMRI trial

Sep 21, 2026

Uploaded by Miles

Randomized ScreenTrustMRI trial evaluates an AI scoring tool to triage people with negative screening mammograms for supplemental contrast-enhanced MRI in population-based breast cancer screening. Participants with negative mammography and a high AI score (top 6.9%) were randomly assigned to receive supplemental MRI or not. The primary endpoint targets advanced breast cancer (interval cancer or invasive component >15 mm or lymph-node–positive disease) based on 27-month follow-up. The current paper reports secondary results, focused on cancers detected by supplemental MRI, comparing efficiency with traditional breast density measures. Findings indicate nearly fourfold higher cancer detection per 1,000 MRI examinations (64 vs 16.5), with most additional cancers being invasive and frequently multifocal, supporting timely detection and favorable cost per cancer detected.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Background

Interval cancers and limitations of mammography

Rationale for supplemental MRI

Trial design

AI triage strategy and participant assignment

Endpoints and follow-up

Results focus (current paper)

Cancers detected by supplemental MRI

Efficiency vs traditional breast density measures

Implications

Cost per cancer detected and timely detection

nature medicine Article [https://doi.org/10.1038/s41591-024-03093-5](https://doi.org/10.1038/s41591-024-03093-5) AI-based selection of individuals for supplemental MRI in population-based breast cancer screening: the randomized ScreenTrustMRI trial Received: 9 January 2024 Accepted: 23 May 2024 | Published online: 8 July 2024 | | --- | | Check for updates | Mattie Salim 1,2, Yue Liu3,4, MoeinSorkhei 3,4, Dimitra Ntoula2, Theodoros Foukakis 1, Irma Fredriksson5, Yanlu Wang1, Martin Eklund 6 , Hossein Azizpour7, Kevin Smith3,4 & Fredrik Strand 1,2 Screening mammography reduces breast cancer mortality, but studies analyzing interval cancers diagnosed after negative screens have shown that many cancers are missed. Supplemental screening using magnetic resonance imaging (MRI) can reduce the number of missed cancers. However, as qualified MRI staff are lacking, the equipment is expensive to purchase and cost-effectiveness for screening may not be convincing, the utilization of MRI is currently limited. An effective method fortriaging individuals to supplemental MRI screening is therefore needed. We conducted a randomized clinical trial, ScreenTrustMRI, using a recently developed artificial intelligence (AI) tool to score each mammogram. We offered trial participation to individuals with a negative screening mammogram anda high AI score (top 6.9%). Upon agreeing to participate, individuals were assigned randomly to oneof two groups: those receiving supplemental MRI and those not receiving MRI. The primary endpoint of ScreenTrustMRI is advanced breast cancer defined as either interval cancer, invasive component larger than 15 mm or lymph node positive cancer, basedona 27-month follow-up time from the initial screening. Secondary endpoints, prespecified in the study protocol tobe reported before the primary outcome, include cancer detected by supplemental MRI, which is the focus of the current paper. Compared with traditional breast density measures used ina previous clinical trial, the current AI method was nearly four times more efficient in terms of cancers detected per 1,000 MRI examinations (64 versus 16.5). Most additional cancers detected were invasive and several were multifocal, suggesting that their detection was timely. Altogether, our results show that using anAI-based score to select a small proportion (6.9%) of individuals for supplemental MRI after negative mammography detects many missed cancers, making the cost per cancer detected comparable with screening mammography. ClinicalTrials.gov registration: NCT04832594. A full list of affiliations appears at the end of the paper. e-mail: [fredrik.strand@ki.se](fredrik.strand@ki.se) Mammography is the standard method for population-wide breast cancer screening, and randomized trials have confirmed that early cancer detection achieved through screening mammography leads to decreased breast cancer mortality1,2. However, around 30 percent of cancersin screened women are diagnosed as so-called interval cancers, becoming symptomatic after a negative screening and before the next planned screening3. Interval cancers may escape screening as they maybe fast-growing and not present on mammograms atthe time of screening, were missed by the radiologist reader or were undetectable by mammography atthe time4,5. Like other symptomatically detected cancers, interval cancers show aggressive biology and have relatively poor prognosis6,7. Contrast-enhanced MRI has been shown to be a more sensitive method for early detection of breast cancer compared with mammography8. The sensitivity of mammography is reduced for women with extremely dense breasts, while the sensitivity of MRI is unaffected9–11. In addition, women with dense breast tissue experience up toa twofold increase in breast cancer risk compared with the average woman12–17. The randomized DENSE trial conducted in the Netherlands demonstrated a 50% reduction in interval cancer for women invited to MRI co

FAQs

How did ScreenTrustMRI use AI in breast cancer screening?

ScreenTrustMRI used an AI tool to score each mammogram and identify individuals with negative screening results who had a high AI score (top 6.9%).

What were the key results for cancer detection with supplemental MRI?

Compared with traditional breast density measures, the AI approach was nearly four times more efficient, detecting 64 vs 16.5 cancers per 1,000 MRI examinations. Most additional cancers were invasive and several were multifocal, indicating timely detection.

Sep 21, 2026

Uploaded by Miles