A prospective study for brain metastasis imaging screening in patients with advanced HER2-positive or triple-negative breast cancer.

Introduction
Breast cancer is the most common cancer among women worldwide and the second most frequent cause of metastatic brain tumors, accounting for ∼15% of cases.1 Several molecular pathways have been implicated in promoting organotropism to the brain,2 and clinical cohort studies have demonstrated that patients with advanced human epidermal growth factor receptor 2 (HER2)-positive breast cancer and triple-negative breast cancer (TNBC) are at high risk of brain metastasis (BM).3,4 The overall incidence of BM in these breast cancer subtypes is ∼30%, and BM is an independent poor prognostic factor, increasing the risk of death 2.5- to 3-fold.5, 6, 7 The conventional treatment mainstay for BM has been local treatment modalities, including radiotherapy and surgery. However, recent clinical trials have also revealed significant intracranial activity of targeted agents in patients with HER2-positive breast cancer. Therefore systemic therapy alone or in combination with local therapies is now an effective strategy for selected patient groups.
As the expected overall survival (OS) of patients with advanced breast cancer (ABC) increases, the management strategy for BM is becoming increasingly important for patient survival and quality of life. As the extent of intracranial disease at the time of diagnosis is a well-established prognostic factor,8,9 researchers are interested in whether early detection of BM through imaging surveillance in asymptomatic patients offers survival benefits. Early detection of smaller lesions may allow for more effective and less toxic interventions, potentially resulting in improved intracranial outcomes. However, current guidelines for ABC [National Comprehensive Cancer Network (NCCN), American Society of Clinical Oncology (ASCO), and ABC-06,0710, 11, 12] do not recommend routine brain imaging for neurologically asymptomatic patients with breast cancer, in contrast to those for lung cancer or melanoma, owing to the lack of prospective clinical evidence supporting the effectiveness of early detection and intervention for BM in patients with ABC. The European Society for Medical Oncology–European Association of Neuro-Oncology (ESMO–EANO) guidelines suggest that brain magnetic resonance imaging (MRI) screening may be considered for patients with metastatic TNBC and HER2-positive breast cancer, allowing for potential subtype-oriented screening for BM.
A previous imaging study revealed similar OS between patients with symptomatic and those with occult BM.4,13 Therefore the impact of treatment for occult central nervous system (CNS) disease on survival remains questionable. Advances in local and systemic treatment options for BM may benefit patients in whom BM is detected at an early stage. Stereotactic radiotherapy (SRS) and MRI-guided navigation surgery reportedly alleviate complications of local treatment,14,15 and newer antibody–drug conjugates and brain-penetrating tyrosine kinase inhibitors exhibit significantly improved activity against BM.16,17 With the introduction of treatment modalities associated with low neurologic toxicities and better intracranial and extracranial disease control, improved survival and neurologic outcomes are expected among patients in whom BM is detected in an early stage.
Based on this clinical background, we conducted a prospective study of brain MRI screening in neurologically asymptomatic patients with advanced TNBC or HER2-positive breast cancer to evaluate the effectiveness of a serial brain MRI screening program for BM detection in these high-risk patients (NCT03617341). Here, we report the primary findings from a cohort of 112 treatment-naive patients with ABC who were serially screened from the time of ABC diagnosis to the initiation of up to third-line systemic therapy. The aim was to assess the BM detection rate and clinical outcomes after detection, and to identify patient subgroups for whom routine brain MRI screening may offer the greatest benefit.

Patients and methods

Patients
This prospective single-arm study was conducted to evaluate the clinical usefulness of brain MRI surveillance in patients with ABC (see the study scheme in Supplementary Figure S1, available at https://doi.org/10.1016/j.esmoop.2026.106892) at Yonsei Cancer Center. Eligible patients were aged ≥19 years and were diagnosed with advanced HER2-positive breast cancer or advanced TNBC according to the ASCO–College of American Pathologists (ASCO–CAP) guidelines. Patients were excluded if they had a history of prior BM, were neurologically symptomatic during the screening period, or had previously received third- or later-line systemic treatments (including endocrine therapy). Patients enrolled from October 2018 to July 2023 who had not yet received first-line systemic therapy (treatment-naive) for ABC were included in this analysis. Patients were enrolled if they met the eligibility criteria and provided informed consent. The study population was not a consecutive cohort and did not include all patients with advanced HER2-positive and TNBC in Yonsei Cancer Center during the enrollment period. The Consolidated Standards of Reporting Trials (CONSORT) diagram for patient enrollment is presented in Supplementary Figure S2, available at https://doi.org/10.1016/j.esmoop.2026.106892. This study was approved by the Institutional Review Board of Severance Hospital (approval number 4-2018-0254; approval date: 2 May 2018) and was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. Informed consent was obtained from all patients at enrollment. All procedures were carried out in compliance with relevant laws and institutional guidelines.

Brain MRI screening and treatment for BM
Screening brain MRI was carried out before the initiation of systemic therapy in asymptomatic patients with HER2-positive ABC or TNBC (i) at baseline and (ii) at the initiation of second- and third-line systemic therapy following disease progression on prior treatment (Supplementary Figure S1, available at https://doi.org/10.1016/j.esmoop.2026.106892). If no BM was detected on baseline brain MRI, patients were monitored for the development of neurologic symptoms suggestive of brain involvement (e.g. headache, nausea, or vomiting). If such symptoms occurred during any line of chemotherapy, a brain MRI was carried out at the discretion of the treating physician. Patients diagnosed with symptomatic BM were withdrawn from the remainder of the study; however, we collected data on their subsequent treatment, including local therapies, as well as survival outcomes. If BM was not identified on baseline brain MRI screening and no BM was diagnosed during systemic therapy, a follow-up brain MRI was carried out at the initiation of second- or third-line systemic therapy after disease progression on prior treatment. For patients diagnosed with BM, treatment was determined through multidisciplinary consultation between neurosurgeons and radiation oncologists. Surgical resection was considered for patients with one or two surgically accessible and symptomatic metastases. For patients with up to four metastases measuring ≤3 cm each, SRS (e.g. gamma knife surgery or CyberKnife radiation therapy) was considered.

Study endpoints
The primary endpoint was the detection rate of BM by screening brain MRI. The secondary endpoints were as follows: (i) the incidence of brain oligometastasis (four lesions or fewer) confirmed by brain MRI; (ii) intracranial progression-free survival (PFS) after treatment for BM; (iii) OS after diagnosis of BM; (iv) the time from diagnosis to treatment for BM; and (v) cognitive impairment after BM.

Cognitive function assessment by questionnaires
In patients diagnosed with BM either by screening MRI or MRI carried out due to symptoms, a baseline assessment of cognitive function was conducted using the Korean version of the Mini-Mental State Examination (MMSE-K), before treatment for BM, such as surgical resection, stereotactic radiosurgery, or whole-brain radiotherapy (WBRT). A follow-up MMSE was carried out 6 months after completion of BM therapy.

Statistical analysis
This was a prospective observational study, and no formal statistical hypothesis testing was planned for the primary endpoint. Continuous variables were compared using Student’s t-test, and categorical variables were compared using the chi-square test. Time to BM was calculated from the date of ABC diagnosis (de novo stage IV or recurrent) to the date of BM detection on brain MRI. The cumulative incidence of BM was estimated using Kaplan–Meier curves and compared using the log-rank test. OS was defined as the time from diagnosis of BM to death from any cause and was compared using the log-rank test. Intracranial PFS was defined as the time from the first BM diagnosis to progression of intracranial disease or death from any cause. Univariate and multivariable Cox regression analyses were carried out to evaluate the impact of baseline characteristics on BM incidence. Statistical analyses were carried out using IBM SPSS Statistics version 28.0 (IBM Corp) and GraphPad Prism (GraphPad Software). Competing risk analysis was carried out using the cmprsk package in R (version 4.3.3). All P values are two-sided.

Results

Baseline characteristics of patients
Among the 138 patients with advanced HER2-positive breast cancer or TNBC who were enrolled in the prospective brain MRI screening study, 112 patients who were at the initial diagnosis of ABC and had not yet undergone first-line treatment were included in the final analysis set (Supplementary Figures S1 and S2, available at https://doi.org/10.1016/j.esmoop.2026.106892). Patients enrolled before the initiation of second- (n = 24) or third-line (n = 2) treatment were excluded from the analysis. Contrast-enhanced brain MRI was carried out before the start of systemic treatment to screen for BM. The median age of the 112 patients was 52 years, and most had a good performance status [Eastern Cooperative Oncology Group (ECOG) 0, n = 102, 91.1%; Table 1]. The breast cancer subtypes were TNBC (n = 54, 48.2%), HR-negative HER2-positive (n = 35, 31.3%), and HR-positive HER2-positive (n = 23, 20.5%). The median follow-up duration was 24.9 months up to the data cut-off date of 21 September 2023. The median OS from the diagnosis of stage IV disease was 30.7 months in patients with TNBC and was not reached in patients with the HER2-positive subtype (P = 0.004, log-rank test, Supplementary Figure S3, available at https://doi.org/10.1016/j.esmoop.2026.106892). The median time from ABC diagnosis to the first, second, and third brain MRI screens was 25 days, 8.8 months, and 15.2 months, respectively. First-line systemic therapy was administered to 98.1% (53/54) of patients with TNBC and 100% (58/58) of patients with HER2-positive disease. Among patients who progressed on first-line therapy, 70.5% (31/44) of those with TNBC and 90.0% (27/30) of those with HER2-positive disease received second-line therapy. Among those who progressed on second-line therapy, 91.7% (22/24) of patients with TNBC and 81.0% (17/21) of patients with HER2-positive disease received third-line therapy.

Diagnosis of BM via brain MRI screening
The proportion of patients diagnosed with BM on brain MRI screening was 19.6% (22/112) overall (Table 2 and Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2026.106892): 24.1% (13/54) among patients with TNBC and 15.5% (9/58) among patients with HER2-positive disease [22.7% for hormone receptor (HR)-positive, HER2-positive and 11.1% for HR-negative, HER2-positive patients]. Despite MRI screening before initiation of each line of systemic treatment, 9.8% (11/112) of patients were diagnosed with symptomatic BM during treatment, and the overall cumulative incidence of BM during the study period was 29.5%. Therefore two-thirds of patients diagnosed with BM (22/33, 66.7%) were identified while asymptomatic owing to the serial screening strategy. There were 12 patients with attrition before the second brain MRI and seven before the third brain MRI (Supplementary Figure S2, available at https://doi.org/10.1016/j.esmoop.2026.106892). The reasons for attrition are described in Supplementary Table S2, available at https://doi.org/10.1016/j.esmoop.2026.106892. The median time from stage IV diagnosis to BM diagnosis was 3.3 months in patients diagnosed through asymptomatic screening and 12.9 months in those diagnosed after symptom onset. The corresponding median times are 3.6 months for TNBC and 11.1 months for HER2-positive subtypes. The cumulative detection rate of BM via MRI screening increased over time: 9.8% (11/112), 17.0% (19/112), and 19.6% (22/112) at the first-, second-, and third-line MRI studies, respectively (Figure 1). The absolute detection rates were 9.8% (11/112) at the first MRI, 18.2% (8/44) at the second MRI, and 13.0% (3/23) at the third MRI. Among the 33 patients diagnosed with BM, 51.5% (17/33) had oligometastatic disease, 45.5% (15/33) had more than four lesions, and 3.0% (1/33) exhibited leptomeningeal seeding. The detailed number, size, location, and symptoms of BM are summarized in Supplementary Table S3, available at https://doi.org/10.1016/j.esmoop.2026.106892. Patients diagnosed with symptomatic BM (n = 11) tended to have larger lesions and more numerous lesions than those diagnosed through asymptomatic MRI screening; however, statistical comparisons were not feasible because of the limited sample size. In these patients, the median time from the last negative screening MRI to the onset of neurologic symptoms and subsequent BM diagnosis was 11.2 and 11.3 months, respectively. Symptoms resolved in eight patients and worsened in two after local treatment; one patient did not receive local treatment.

BM risk and the number of organ metastatic sites
The median time from ABC diagnosis to BM diagnosis was 25.1 months. The cumulative incidence of BM was 14.3% and 20.5% at 6 and 12 months after the diagnosis of stage IV disease, respectively (Figure 2A). The estimated probability of BM diagnosis at 12 months using the one-minus Kaplan–Meier method was 23.4% (corresponding to a 76.6% 12-month BM-free survival; Figure 2B and Supplementary Figure S4A, available at https://doi.org/10.1016/j.esmoop.2026.106892). A competing-risk analysis accounting for death as a competing event was additionally carried out, and the difference between the one-minus Kaplan–Meier method and cumulative incidence function curves was modest in our cohort (Supplementary Figure S4B, available at https://doi.org/10.1016/j.esmoop.2026.106892). The rate of BM diagnosis did not differ according to cancer subtype (TNBC versus HER2-positive breast cancer, Supplementary Figure S5A, available at https://doi.org/10.1016/j.esmoop.2026.106892). Patients with metastatic involvement of three or more non-CNS organ sites had a higher risk of BM diagnosis than those with involvement of less than three organ sites (Supplementary Figure S5B, available at https://doi.org/10.1016/j.esmoop.2026.106892; P < 0.001, log-rank test). Multivariable Cox regression analysis using baseline clinical parameters revealed that metastatic involvement of three or more non-CNS organ sites at baseline was an independent predictive factor for BM diagnosis (hazard ratio: 3.38, 95% confidence interval: 1.55-7.34, P = 0.002; Table 3). Recurrent disease status also conferred a higher risk of BM than de novo stage IV disease (hazard ratio 3.06, 95% confidence interval 1.34-7.03, P = 0.008). Among patients with metastatic involvement of three or more non-CNS organ sites (n = 26), the incidence of BM was 46.2% during the study period, compared with 29.5% among patients with less than three non-CNS organ sites (n = 86). In this subgroup, 38.5% (10/26) of patients were diagnosed with BM via brain MRI screening, whereas the corresponding rate was 14.0% (12/86) among those with less than three non-CNS organ sites (Supplementary Tables S4 and S5, available at https://doi.org/10.1016/j.esmoop.2026.106892). Notably, 83.3% (10/12) of BM diagnoses in patients with three or more non-CNS organ sites were identified through brain MRI screening (Supplementary Table S5, available at https://doi.org/10.1016/j.esmoop.2026.106892).

BM treatment
SRS in the form of gamma knife surgery was the most common initial local treatment modality for patients diagnosed with BM (22/33, 66.7%; Supplementary Table S6, available at https://doi.org/10.1016/j.esmoop.2026.106892). WBRT was carried out in 9/33 (27.3%) patients, and the remaining 2 (6.1%) patients did not receive any treatment for BM. Among patients who underwent local treatment for BM (n = 31), WBRT was more commonly used in patients with TNBC (6/15, 40.0%) than in those with HER2-positive breast cancer (3/16, 18.8%). None of the patients underwent surgery as the initial treatment for BM. After the initial gamma knife surgery, three HER2-positive subtype patients (two symptomatically and one through asymptomatic screening) subsequently underwent surgical resection for BM at 15.4, 17.5, and 20.1 months after the first BM diagnosis, respectively. The median OS following a diagnosis of BM (n = 33) was 23.3 months (Supplementary Figure S6A, available at https://doi.org/10.1016/j.esmoop.2026.106892). OS after BM diagnosis did not significantly differ according to the mode of detection (Supplementary Figure S6B, available at https://doi.org/10.1016/j.esmoop.2026.106892). Patients with TNBC had a shorter OS (11.9 months) than those with HER2-positive breast cancer (27.4 months; Supplementary Figure S7, available at https://doi.org/10.1016/j.esmoop.2026.106892). The median intracranial PFS was 6.5 months for all patients with BM (n = 33; Supplementary Figure S8A, available at https://doi.org/10.1016/j.esmoop.2026.106892). Intracranial PFS did not differ between patients with screening-detected (n = 22) and those with interval symptomatic detection (n = 11; P = 0.182; Supplementary Figure S8B, available at https://doi.org/10.1016/j.esmoop.2026.106892). Intracranial PFS also did not differ according to subtype (TNBC versus HER2+, P = 0.182; Supplementary Figure S8C, available at https://doi.org/10.1016/j.esmoop.2026.106892).

Systemic treatment
The systemic treatment regimens received as first-, second-, and third-line therapy are summarized in Supplementary Table S7, available at https://doi.org/10.1016/j.esmoop.2026.106892. For patients diagnosed with BM on the first screening MRI at the time of stage IV disease diagnosis (n = 11), systemic therapy was initiated concurrently with the diagnosis of BM. Among patients diagnosed with BM on the second (n = 8) or third (n = 3) screening MRI, which were carried out at the time of systemic disease progression, the systemic treatment regimen was changed in all patients (n = 11). Among patients diagnosed with symptomatic BM during the interval period (n = 11), systemic therapy was maintained in two patients, both of whom were receiving an trastuzumab + pertuzumab + docetaxel regimen. By contrast, systemic therapy was modified in six patients following the diagnosis of BM, while three patients did not receive further systemic treatment due to clinical deterioration.

Cognitive function evaluation by MMSE
MMSE scores were collected for 16 patients (12 screening-detected BM and 4 with interval symptomatic detection). MMSE scores were obtained at the time of BM diagnosis in all 16 patients and at 6 months after BM diagnosis in 12 patients. The MMSE score did not differ between the time of BM diagnosis (29.58 ± 0.79) and 6 months after BM diagnosis and brain radiotherapy (29.58 ± 0.90). Cognitive function in patients with screening-detected BM did not differ from that in patients diagnosed with interval symptomatic BM.

Discussion
As the treatment landscape for breast cancer continues to evolve with the development of novel systemic therapies, researchers have shown increasing clinical interest in the detection of and management strategies for BM.18 In this prospective study, we conducted brain MRI screening at ABC diagnosis and at the initiation of each line of systemic therapy (from first to third line) in neurologically asymptomatic patients with advanced HER2-positive breast cancer or TNBC. The study screening protocol resulted in the detection of asymptomatic BM in 19.6% of patients, and the cumulative incidence increased with each subsequent line of therapy (9.8% at ABC diagnosis, 17.0% at initiation of second-line therapy, and 19.6% at initiation of third-line therapy). Notably, two-thirds of BM diagnoses were made in asymptomatic patients through this screening program. This study is one of the recently conducted prospective trials, together with the study by Ahmed et al.,19 evaluating brain MRI screening for asymptomatic BM in patients with ABC. Our study demonstrated the potential clinical utility of baseline and serial brain MRI screening in patients with newly diagnosed ABC. These results support the integration of routine brain MRI screening into clinical practice for patients with advanced HER2-positive breast cancer or TNBC.
BM remains a critical challenge in patients with ABC and has been consistently associated with poor clinical outcomes in large-scale registry studies, including the SystHERs6 and registHER studies.7 Previous studies have reported an estimated cumulative risk of BM in patients with ABC of up to 25%, and this risk is especially pronounced in those with HER2-positive or TNBC subtypes.6,7,20,21 The reported median time to BM diagnosis after ABC diagnosis is 2-3 years.20 Despite this substantial risk, current clinical guidelines do not recommend routine brain MRI screening in patients with ABC, largely owing to earlier studies that revealed no survival difference between patients with occult BM and those with symptomatic BM.4,13 However, those studies were retrospective in nature, subject to selection bias, and did not focus on high-risk subtypes (HER2-positive breast cancer and TNBC). Therefore the actual rate of BM detection via screening MRI, the impact of BM diagnosis on clinical outcomes, and the effectiveness of brain MRI screening were previously unclear. Accordingly, the practice patterns regarding brain MRI screening in asymptomatic patients vary widely among physicians, ranging from routine screening in asymptomatic patients to symptom-directed brain imaging.22,23 Our prospective screening study addressed this critical knowledge gap by demonstrating that a brain MRI screening program can yield a high detection rate of asymptomatic BM. These findings provide supportive evidence for considering early brain MRI screening strategies in high-risk ABC populations. However, this study was not designed to directly evaluate survival benefits, such as intracranial PFS or OS, compared with symptom-directed imaging in a randomized manner. Therefore further prospective interventional studies are warranted to establish definitive evidence for BM screening.
The BM detection rate of 19.6% among patients with HER2-positive ABC or advanced TNBC in this study is consistent with previous reports.21 In the SystHERs study, the BM diagnosis rate was 8.9% at the initial ABC diagnosis and 21.7% thereafter. The BM incidence observed in our study is also comparable to that in patients with non-small-cell lung cancer, for whom brain MRI screening is mandatory (10%-15% at diagnosis).24 Previous studies have indicated that younger age, recurrent ABC rather than de novo stage IV ABC, HR-negative status, and a higher disease burden are associated with an increased risk of CNS metastasis.6,7 Our study also suggested that the number of metastatic non-CNS organ sites is a significant risk factor for BM diagnosis in patients with ABC (hazard ratio 3.38). Brain MRI screening was effective in these patients, with an overall BM detection rate of 38.5%; 83.3% of such BM were diagnosed via brain MRI screening. However, this result should be interpreted with caution and validated in future studies, given that this was a single-center study with a limited number of BM events (n = 33).
The impact of early detection of and intervention for BM in patients with ABC was uncertain in previous studies, probably because previous BM treatment modalities were less effective than current approaches. Recent clinical advancements in therapeutic modalities have reduced their neurotoxicity and enhanced efficacy. A landmark trial (NCCTG N0574) revealed that SRS is less toxic than WBRT and yields a similar survival outcome, establishing SRS as the treatment of choice for patients with one to three BMs.15,25 Regarding WBRT, hippocampal avoidance combined with memantine can alleviate cognitive dysfunction in patients with low-risk BM.26 Therefore outcomes of local treatment for BM have improved substantially in recent years. In our study, the median OS after BM diagnosis was favorable (23.3 months), and cognitive function, as assessed by MMSE, was preserved between baseline and 6 months after BM diagnosis.
In addition, systemic treatment has now become a feasible option for patients with previously untreated or stable BM. Unlike monoclonal antibody agents, trastuzumab deruxtecan has demonstrated robust activity against BM in the TUXEDO, DESTINY-Breast03, and DESTINY-Breast12 trials.17,27,28 Moreover, the HER2CLIMB trial showed significant intracranial efficacy with tucatinib-based combination treatment.16 Therefore multiple systemic treatment strategies (e.g. systemic therapy alone for active BM and continuation of systemic drug maintenance beyond progression) are now available for HER2-positive breast cancer. With these advances in BM management, not all asymptomatic patients with incidentally detected BM require immediate SRS or WBRT; some may be appropriately managed with systemic therapy. In light of these developments, the potential clinical superiority of early screening and intervention for BM, compared with symptom-driven diagnosis and treatment, warrants evaluation in a prospective interventional study.
We also note that brain-active systemic treatment options remain limited for patients with TNBC. In our study, patients with TNBC were more likely to receive WBRT than those with HER2-positive disease (40.0% versus 18.8%) because of multiple BMs or leptomeningeal seeding. This difference may reflect distinct biologic and clinical characteristics, as well as disparities in available treatment options, between the two subtypes. These findings suggest that the MRI screening platform proposed in this study may not sufficiently reduce the use of neurotoxic WBRT in patients with TNBC, underscoring the need for further advances in both BM screening strategies and brain-effective systemic therapies for this population.
A major strength of our analysis was the prospective enrollment of patients at a standardized timepoint (before first-line treatment) and systematic data collection with long-term follow-up MRI surveillance. This design enabled a more accurate estimation of BM incidence and treatment outcomes while minimizing selection bias. However, several limitations should be acknowledged. This study was limited by its small sample size, lack of randomized comparison between screening and symptom-based detection strategies, and single-institution setting. In addition, the attrition rate (19 of 112, 17.0%, patients) represents a practical challenge to implementing repeated brain MRI surveillance in patients with ABC. Most attrition events were attributable to patient refusal of brain MRI or transfer to another institution (14 of 19, 73.7%, patients), whereas clinical deterioration accounted for 4.5% (5 of 112 patients) of the overall cohort. These findings identify patient refusal and clinical deterioration as important real-world barriers to BM screening and underscore the need for more accessible and patient-friendly approaches to early BM detection, particularly for patients with limited tolerance for MRI or declining clinical status.
Recently, Ahmed et al. reported a phase II trial evaluating brain MRI surveillance at baseline and at 6 months in patients with stage IV breast cancer.19 However, they enrolled across a wide range of treatment lines (first to eighth), which limits the generalizability of their findings. Furthermore, the study included all breast cancer subtypes, including HR-positive/HER2-negative disease, further broadening the study population. By contrast, our study focused specifically on high-risk subtypes (only those with HER2-positive breast cancer and TNBC), and implemented BM screening beginning at the time of ABC diagnosis. Jerzak et al. have also reported interim results of a randomized study comparing routine BM screening with symptom-directed brain imaging; final results are awaited.29
In conclusion, our prospective brain MRI screening program detected BM in 19.6% of patients with HER2-positive ABC or advanced TNBC, with approximately two-thirds of cases (22/33) identified at an asymptomatic stage. These findings suggest that serial brain MRI surveillance may facilitate earlier detection of BM in high-risk populations. Further prospective clinical studies are warranted to determine the clinical utility and survival benefit of such a screening strategy. As therapeutic modalities for breast cancer-related BM continue to evolve, early detection and timely intervention may represent an effective approach to improving patient outcomes.