Breast Cancer Patients with Brain Metastases: A Cross-Sectional Study

The prognosis of breast cancer patients with brain metastasis is poor. It was aimed to define the clinicopathological features of breast cancer patients with brain metastases and to determine the risk factors and survival outcomes associated with brain metastasis. This is a single-center, retrospective, cross-sectional study. A total number of 127 patients diagnosed with breast cancer and who developed brain metastasis between January 2011 and March 2021 were retrospectively analyzed. The survival and clinicopathological data of these patients according to 4 biological subtypes were evaluated (luminal A, luminal B, HER-2 overexpressing, and triple-negative). The median overall survival for all patients was 45.6 months. The median time from the diagnosis of breast cancer to the occurrence of brain metastasis was 29.7 months, and the median survival time after brain metastasis was 7.2 months. The time from the diagnosis of breast cancer to brain metastasis development was significantly shorter in HER-2 overexpressing and triple-negative subtypes than in luminal A and B subtypes. The median time from breast cancer diagnosis to brain metastasis was 33.5 months in luminal A, 40.6 months in luminal B, 16.8 months in HER-2 overexpressing, and 22.8 months in the triple-negative groups (p=0.003). We found the worst median survival after brain metastasis in the triple-negative group with 3.5 months. Early and close surveillance of high-risk patients may help early diagnosis of brain metastasis and may provide to perform effective treatments leading to longer overall survival times for this patient population.


Introduction
Breast cancer (BC) remains the most common cancer and the second most common cause of cancer-related death among women worldwide. Although diagnosed in an early stage, the metastatic disease occurs in up to 30% of BC patients [1]. In the course of the disease, brain metastasis (BM) has been seen in 10-15% of the metastatic BC patients [2]. Although the incidence of BM has been 5.2% for earlystage disease in a 10-year follow-up, BM occurs in more than 25% of all BC patients at some point [3]. Due to longer survivals as a result of early detection and efficient treatments such as targeted treatments, the incidence of BM is rising [4]. Advances in imaging modalities also contribute to this increase in incidence [5]. Severe cognitive, sensorial, and neurological impairments and poor prognosis are anticipated in patients with BM [6,7]. For handling poor outcomes in BM patients, it is a significant challenge to develop management strategies.
After lung cancer, BC is the second most common cause of BM, such that breast and lung cancer metastases are diagnosed more commonly than primary brain tumors [8]. In newly diagnosed BC patients, the median time for BM occurrence is 33 months, and the median survival time after BM is five months [9]. Also, BM may be the first recurrence site in some patients. At this point, the recognition of certain subsets of patients with a tendency for BM will provide us the opportunity for early detection and to take preventive measures in high-risk groups to develop specific targeted therapies that will improve the quality of life and survival outcomes of these patients [10]. e most important risk factors related to the occurrence of BM are younger age, poor performance status, tumor size, grade, nodal status, hormone receptor (HR) status, and disease burden [11]. Chemotherapy responses, survival outcomes, and sites of distant relapses are different among subtypes of BC [12]. ese molecular subtypes are based on gene expression profiles such as luminal A, luminal B, HER-2 overexpressing, and triple-negative [2]. It has been shown that these subtypes manifest differently metastasize tendencies for certain locations. While estrogen and progesterone receptor-positive tumors are inclined to metastasize to the bones, HER-2 overexpressing tumors and triple-negative tumors are more likely to metastasize visceral organs including the brain [1].
is study aims to review the clinicopathological features of BC patients with BM, determine the risk factors and survival outcomes associated with BM, and assess the relationship between treatment options and outcomes of BC patients with BM.

Materials and Methods
In this single-center cross-sectional study, a total number of 127 female BC patients who developed BM between 2011 and 2021 and followed at the department of medical oncology were retrospectively analyzed. is study was organized according to the STROBE checklist. e data were collected from patient files and hospital information system. We carefully examined the demographic data and the pathologic features of the primary tumor, including patient age at diagnosis, tumor type, histologic grade, and ER, PR, and HER-2 status. BM was determined by different imaging modalities such as magnetic resonance imaging and computed tomography. HR and HER-2 status were determined by IHC analysis. Membranous three positivity of HER-2 with IHC and two positivity of HER-2 with IHC confirmed by FISH were accepted as HER-2 positive. Negative or 1+ values of HER-2 with IHC were accepted as HER-2 negative. BC subtypes were determined according to the HR status and HER-2 expression: ER and/or PR positive and HER-2 negative (luminal A); ER and/or PR positive and either HER-2 positive or HER-2 negative with high levels of Ki-67 (luminal B); ER and PR negative and HER-2 positive (HER-2 overexpressing); and triple-negative. Overall survival (OS) was defined as the time between diagnosis of primary BC and the time for the last visit or death. Time to development of BM was defined as the time between primary BC diagnosis and the first diagnosis of BM. Survival time after BM was defined as the time between the development of BM and the time for the last visit or death. All the data obtained were recorded to the data collection form.
Before data collection, approvals from the local ethics committee and institution were obtained for this study. e Helsinki Declaration was complied with in the study. e data were analyzed via Statistical Package for the Social Sciences, version 25 (SPSS v25.0) program (IBM Company). e mean and median values of the variables were calculated by descriptive analysis. Significance was considered as a p value of ˂0.05 in a 95% confidence interval. Kruskal-Wallis and Mann-Whitney U-tests were performed for in-group and intergroup comparisons.

Results
A total number of 127 BC patients with BM were included in this study. At the time of BC diagnosis, the median age of the patients was 46.6 (range of 22-73), and the median age of the patients at the time of BM occurrence was 50.9 (range of 29-78). Considering HR status and HER-2 expression, the percentages of subtypes were 6.3% (n � 8) for luminal A, 56.7% (n � 72) for luminal B, 16.5% (n � 21) for HER-2 overexpressing, and 20.5% (n � 26) for triple-negative. According to histological subtypes, a large proportion of the patients (92.1%) had invasive ductal carcinoma (n � 117), seven patients had invasive lobular carcinoma (5.5%), and only three patients had other subtypes (2.4%).
Of the patients with BM, 33.9% (43/127) had metastatic disease at diagnosis and 6.3% (8/127)  In patients with ER and/or PR positivity (80/127), the 1year OS rate was 37.5%, and the median follow-up period after BM was 7.5 months. Of these patients, 8.8% (7/80) had only BM; 25% (20/80) had also bone metastasis, and 66.2% (53/80) had visceral metastasis among BM. We also examined patients as HER-2-positive and HER-2-negative groups regardless of their hormone status. e median duration of time from BC diagnosis to BM development and the median survival after BM were shorter in the HER-2positive group than the HER-2-negative group, and the difference was statistically significant (20.4 vs. 37.7 months; p � 0.017 and 5.3 vs. 11.1 months; p � 0.014, respectively). Of the HER-2-positive patients, 15.5% (7/45) had only BM, 26.7% (12/45) had additional bone metastasis, and 57.8% (26/45) had visceral metastasis among BM. e median time from BC diagnosis to BM was 33.5 months in the luminal A group, 40.6 months in the luminal B group, 16.8 months in the HER-2 overexpressing group, and 22.8 months in the triple-negative group. A statistically significant difference that originated from HER-2 overexpressing and triple-negative groups was observed between these groups (p � 0.003). e median survival time after the 2 e Breast Journal  e Breast Journal 3 development of BM was 10.5 months in the luminal A group, 7.6 months for the luminal B group, 11.6 months for the HER-2 overexpressing group, and 3.5 months for the triplenegative group (Figure 1). A statistically significant difference was not observed between these groups (p � 0.052). e expected survival time after BM was 18.3 months in the group that had only BM, 12 months in the group with additional bone metastasis, and 5.3 months in the group that had BM and visceral metastases. A statistically significant difference originated from visceral metastasis was found between these groups (p � 0, 003). Factors related to the development of BM and the median survival time after BM are summarized in Table 2.

Discussion
e most common metastasis sites in the clinical course of BC are bone, liver, and brain, occurring 70%, 30%, and 10-30%, respectively [13]. For many BC patients, occurrence of BM has become an important limitation of life expectancy and quality of life 8. e incidence of BM is 14% for HRpositive disease, and the median survival time after BM is 9-10 months. Also, HER-2 positivity, as seen in approximately 25% of BC patients, is related to an increased risk of BM occurrence up to 30-53% for patients with a median survival time of 11-18 months [11,14]. In this current study, the clinicopathological features of 127 BC patients with BM were evaluated retrospectively.
Both the median time from primary BC diagnosis to the occurrence of BM (29.7 months) and the median time from BM occurrence to the last contact with the patient (7.2 months) were found to be similar to [7,15]. e average follow-up time after BC diagnosis to the last contact with the patient either alive or dead was 45.6 months.
am et al. showed that aggressive pathological features are at a high risk of developing BM [10]. In particular, HER-2overexpressing and triple-negative subtypes are important risk factors for BM development [16]. Our results were consistent with the literature, for example, the BM development time was significantly shorter in HER-2-overexpressing and triple-negative subtypes compared to luminal subtypes. Ferguson et al. stated that HER-2 overexpression was significantly higher in BM patients [17]. We observed that the HER-2 positive group had a shorter median duration to BM development than the HER-2 negative group (20.4 vs. 37.7 months; p � 0.018, respectively). Similarly, the median survival after BM diagnosis was statistically shorter in the HER-2 positive group than the HER-2 negative group (5.3 vs. 11.1 months; p � 0.014, respectively).
As the first site of metastasis, BM was mostly related to the triple-negative subtype and was associated less with HRpositive breast tumors [12]. e triple-negative subtype is known to have shorter survival rates, early relapse, and early BM [18]. e median survival time after BM was shorter for the triple-negative group with 3.5 months in our current study, as reported in several studies ranging between 2.9 and 6.6 months [2,19]. Similarly, the median survival after BM for our hormone-positive patients was similar to [20]. e coadministration of anti-HER-2 therapies with chemotherapy has improved survival after BM [11]. However, in our study, it was seen that the median survival of the HER-2positive patients after BM was shorter than in the historical   e Breast Journal data, which might be due to the limited number of patients. e survival analysis of the breast cancer subtypes is shown in Figure 1. e recommended treatment for patients with BM is surgical removal when both the metastatic lesion or lesions are resectable and the systemic disease is under control [21,22]. e survival time after the development of BM was significantly longer in patients with a single metastatic lesion, and this is thought to be associated with curative interventions such as surgical removal or gamma knife therapy. e poor survival of the patients with two or more metastatic lesions might be due to the high number of visceral metastasis, lack of treatments, and the limited number of patients in this subgroup.

Conclusion
In conclusion, our study demonstrates that HER-2-positive and triple-negative patients have a potentially high risk of BM and a shorter survival trend. Similarly, we observed that patients with multiple metastases or leptomeningeal involvement had poor survivals which might be due to the lack of curative therapeutic interventions. erefore, early and close surveillance of the high-risk groups might help with the early detection of BM, and this may provide effective treatment for these patients. Also, we must be vigilant in terms of the early recognition of symptoms that occur at the time of BM occurrence.

Data Availability
e study of data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare no conflicts of interest.