Clinical impact of routine MRI screening for brain metastasis in patients with breast cancer: A meta-analysis.

Le résumé fourni par la source

e14031 Background: Brain metastases (BM) are a frequent and clinically significant complication of metastatic breast cancer (BC), often associated with substantial morbidity and poor prognosis. While brain MRI is routinely performed in symptomatic patients, the clinical value of routine MRI screening in asymptomatic patients remains uncertain and is not uniformly recommended by current guidelines. Clarifying the detection yield of routine MRI screening is essential to inform surveillance strategies and future recommendations. Methods: A systematic search of PubMed, Scopus, and Embase, supplemented by ASCO, SABCS, and SNO conference proceedings, was conducted through January 21, 2026. Studies evaluating routine brain MRI screening in patients with BC without known BM or clinically evident CNS disease at baseline were included. Observational studies and clinical trials with extractable detection yield data were eligible. The primary outcome was asymptomatic BM detection yield, defined as the proportion of screened patients with previously unknown, asymptomatic BM detected on MRI. The secondary outcome was total BM detection yield. Pooled estimates were calculated using random-effects meta-analysis of proportions with logit transformation. Heterogeneity was assessed using I², τ², and Cochran Q. Sensitivity analyses included leave-one-out analyses. Results: Of 744 records identified, 7 studies comprising 653 patients met the criteria for inclusion (5 observational [71.4%], 2 clinical trials [28.6%]). Median age was 51.5 years (IQR 42.5–53.7). Cohorts included patients with advanced (n=372) or any stage disease (n=281). All studies excluded patients with known BM or clinically evident CNS disease at baseline. Median follow-up was 16.0 months (IQR, 9.0–20.05). Routine MRI screening detected previously unknown asymptomatic BM in a pooled 14.9% of patients (95% CI, 8.4%–25.1%), with substantial heterogeneity (I²=84.9%, τ²=0.52; Q=33.1, df=5, p<0.001). The pooled total BM detection yield was 13.6% (95% CI, 7.1%–24.5%), with high heterogeneity (I²=85.2%, τ²=0.67; Q=33.9, df=5, p<0.001). Leave-one-out analyses demonstrated robust pooled estimates, with asymptomatic detection yields ranging from 12.1% to 18.6%, indicating no single study disproportionately influenced results. Conclusions: Routine brain MRI screening in patients with BC without known CNS disease identifies previously unrecognized asymptomatic BM, albeit with variable detection yields. These findings highlight the potential clinical relevance of routine MRI screening in selected populations and support the need for prospective, risk-adapted studies to define its optimal role. Meta-analytic results. Metric Asymptomatic BM detection yield Total BM detection yield Pooled detection yield 0.149 0.136 95% confidence interval 0.084–0.251 0.071–0.245 p value (Q) <0.001 <0.001 Heterogeneity (I², %) 84.9 85.2

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.