Abstract
Background/Aim: Brain metastasis is a rare condition among patients with soft tissue sarcoma (STS), and its precise incidence remains unclear. The aim of this study was to investigate which patients should be screened for brain metastasis. Patients and Methods: We identified all patients with STS diagnosed between 2010 and 2015 in the SEER database. Incidence of brain metastasis at initial presentation and higher incidence of brain metastasis by histological subtype were investigated. In addition, risk factors for brain metastasis were examined. Results: A total of 26,676 patients were included for analysis, of whom 162 patients (0.6%) had brain metastasis. Alveolar soft part sarcoma (6.3%), malignant hemangioendothelioma (3.1%) and malignant schwannoma (2.6%) showed higher incidence of brain metastasis. Deep-rooted tumor, trunk tumor, and histological high-grade tumor tended to show higher incidence of brain metastasis. Conclusion: Risk factors for brain metastasis were deep location, trunk development and histologically high-grade tumor, or specific histological subtypes.
- Soft tissue sarcoma
- brain metastasis
- mortality
- secondary surgery
- soft tissue sarcoma
- SEER program
- neoplasm grading
- retrospective study
- treatment outcome
Soft tissue sarcomas (STSs) are malignant tumors that can arise in nearly any soft tissue in the body and comprise a variety of histological subtypes. Pulmonary metastasis is a common manifestation for STS patients and requires special caution (1, 2). Computed tomography (CT) of the chest is thus mandatory for staging at initial presentation (3-5). In addition, abdominal CT is required for limb myxoid liposarcoma (3).
Brain metastasis is relatively common among patients with epithelial cancers, showing an incidence of 10-20% (6). In contrast, brain metastasis is known to be rare among patients with soft tissue sarcoma, and the precise incidence remains unclear. Brain metastasis definitely exerts deleterious impacts on survival and needs to be detected using an appropriate imaging modality on initial presentation.
Among the epithelial neoplasms, lung and breast cancers tend to show brain metastases, so imaging screening for brain metastasis is commonly encouraged at initial presentation (7). In contrast, in non-epithelial cancers, brain screening may be added for alveolar soft part sarcoma, clear cell sarcoma, angiosarcoma (3). Routine screening for brain metastasis among STS patients is not realistic from the perspective of medical economics, due to the rarity of these metastases. Clarification of risk factors for brain metastasis in patients with STS at initial presentation is needed.
Considering the rarity of brain metastasis among patients with STS, information from a large number of patients is required. We, therefore, hypothesized that a nationwide oncology registry might clarify the situation regarding brain metastasis in STS patients. The purpose of the current study was: i) to clarify the incidence of brain metastasis at initial presentation among patients with STS; and ii) to investigate which patients should be screened for brain metastasis at initial presentation.
Patients and Methods
Patient population. The Surveillance, Epidemiology, and End Results (SEER) database is the largest public cancer database without identified patient data. The database covered approximately 28% of the total population in the United States. We included STS patients diagnosed from 2010 to 2015, according to the International Classification of Diseases for Oncology (ICD-O), using the following SEER histology codes: 8800-8806, 8810-8811, 8813-8815, 8821-8822, 8825, 8830, 8832, 8835-8836, 8840-8842, 8850-8855, 8857-8858, 8860, 8890-8891, 8893-8897, 8901-8902, 8910, 8912, 8920-8921, 8983, 8990-8991, 9040-9044, 9120, 9124-9125, 9130, 9133, 9140, 9150, 9170, 9251-9252, 9540, 9560-9561, 9571, and 9580-9581. SEER employed an original histological grading system and we defined SEER grades 1 and 2 as low-grade tumors, and grades 3 and 4 as high-grade tumors. SEER *Stat software was used to obtain the information (version 8.3.8; National Cancer Institute, Bethesda, MD), and a total of 31519 patients with STSs were identified. After excluding patients without information about age (2 cases), pathological confirmation (715 cases) or brain metastasis (4198 cases), the remaining 26,677 patients were enrolled in this study (Figure 1).
Flowchart showing the process of patient inclusion in this cohort.
Statistical analysis. The primary aims of the current study were to investigate the incidence of brain metastasis in patients with STS at initial presentation and survival. Cancer-specific survival (CSS) and overall survival (OS) were calculated by Kaplan-Meier survival analysis. The secondary aim was to seek risk factors for the development of brain metastasis in patients with STS at initial presentation. Multiple logistic regression analysis was employed to seek risk factor for brain metastasis. All data were analyzed using JMP Pro version 16.0.0 statistical software (SAS Institute, Cary, NC).
Ethical approval. All de-identified data used in this study were obtained from the SEER database. The individual SEER ID (12909-Nov2019) was used to support our analysis. Ethical review and approval were not required for the study on human participants in accordance with the local legislation and institutional requirements. All procedures performed in this study involving human participants were conducted in accordance with the ethical standards with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.
Results
Of the 26,676 patients, 162 patients (0.6%) showed brain metastasis at initial presentation (brain-mets), whereas 26,514 patients (99.4%) did not (non-brain-mets). Median ages at diagnosis of brain-mets and non-brain-mets were 55 and 57 years, respectively (Table I). One-year CSS rates for brain-mets and non-brain-mets were 29% and 86%, respectively (Figure 2A). In addition, 1-year OS rates for brain-mets and non-brain-mets were 22% and 80%, respectively (Figure 2B).
Characteristics of patients in the included cohort.
Kaplan-Meier curves for cancer-specific survival (A) and overall survival (B) according to whether brain metastasis is present. Brain-Mets: STS patients with brain metastasis at initial presentation, Non-Brain-Mets: STS patients without brain metastasis at initial presentation.
Next, we examined whether specific histological subtypes were associated with a higher incidence of brain metastasis. The histological subtypes showing the highest incidence of brain metastasis was alveolar soft part sarcoma (6.3%), followed by malignant hemangioendothelioma (3.1%), malignant schwannoma (2.9%), alveolar rhabdomyosarcoma (2.6%) and round cell sarcoma (2.0%) (Table II).
Histological subtypes according to International Classification of Diseases for Oncology 3 (ICD-O3).
Risk factors for brain metastasis at initial presentation were investigated using multiple logistic regression analysis with factors of age, sex, tumor size, histological type (non-round cell or round cell), laterality (left or right or not a paired site), tumor depth (superficial or deep-rooted), tumor location (limb or trunk) and histological grade (high or low) (Table III). Among these factors, deep-rooted [hazard ratio (HR)=7.025, 95% confidence interval (CI)=0.919-53.714], trunk location (HR=2.415, 95%CI=1.012-5.764) and high histological grade (HR=13.638, 95%CI=1.847-100.720) were associated with higher incidences of brain metastasis. Taking these results from multiple logistic regression, we next asked whether these factors have synergetic effects for brain metastasis. After excluding patients who did not have data (unknown status) regarding tumor depth, tumor location or histological grade, 12,868 patients were divided into 8 groups according to these factors. Interestingly, incidences of brain metastasis increased as patients obtained factors of trunk location, deep-rooted tumor and high grade (Table IV).
Multiple logistic regression analysis with factors of age, sex, tumor size, histological type, laterality, tumor depth, tumor location and histological grade.
Incidence of brain metastasis in patients with brain metastasis according to risk factors.
Discussion
Data from the current study clearly indicated that the overall incidence of brain metastasis in patients with STS at initial presentation was 0.6% and the 1-year OS in patients with brain metastasis was 22%. In addition, risk factors for brain metastasis were deep location, development in the trunk and histologically high-grade tumor, and specific histological subtypes such as alveolar soft part sarcoma, malignant hemangioendothelioma, and malignant schwannoma.
Previous reports regarding the incidence of brain metastasis in patients with STS are limited. Yoshida et al. (8) found brain metastasis in 7.2% of 279 patients at the Niigata Cancer Center Hospital. In a retrospective evaluation of 3829 STS patients at the Sloan-Kettering Cancer Center, Espat et al. (9) found brain metastasis in 21 patients at initial presentation and subsequent development in another 19 patients, accounting for 1%. According to their report, common types of STS with brain metastasis were leiomyosarcoma, liposarcoma, rhabdomyosarcoma, and malignant fibrous histiocytoma. Takemori et al. (10) investigated 509 patients with STS, with 5 patients (1%) showing brain metastasis. Median survival after brain metastasis was 1.5 months. Those three reports were retrospective, single-center studies. A retrospective multicenter study by the French Sarcoma Group (11) identified 246 STS patients with brain metastasis. The most frequent histopathological subtype among patients with brain metastasis was leiomyosarcoma (11). In the current study, we conducted a retrospective study using the SEER database, as the largest public cancer database, and found an incidence of brain metastasis of 0.6% at initial presentation. This incidence of brain metastasis was relatively low compared to previous reports. One reason for this was that we evaluated brain metastasis only at initial presentation and did not evaluate subsequent clinical courses. In addition, we investigated a total of 26,676 patients in this study. This large-scale analysis might lead to more precise results compared to those in previous reports, considering the rarity of the pathology in question.
Several prognostic factors for pulmonary metastasis have been reported, including histological high grade, size, development in the lower extremities, and non-round cell sarcoma (12). In contrast, prognostic factors for brain metastasis have not been addressed in the literature, due to the rarity of brain metastasis in patients with STSs. Herein, we indicated histological high-grade tumor, development in the trunk and deep location as risk factors for brain metastasis in patients with STS. Further, combinations of these factors increased the incidence of brain metastasis. In addition, histological subtypes such as alveolar soft part sarcoma, malignant hemangioendothelioma and malignant schwannoma showed higher incidence of brain metastasis.
The blood-brain barrier (BBB) elegantly organizes homeostasis of the central nervous system (13). The microenvironment possesses an original anatomical structure, cell types and metabolic pathways to inhibit metastasis to the brain when cancer cells penetrate the central nervous system (14-16). Cells metastasizing to the brain must acquire specific genetic changes to penetrate the BBB and survive in this unique environment (17, 18). Considering several types of epithelial cancers tend to acquire such changes (7), certain histological subtypes of non-epithelial cancer might be more likely to acquire such changes in the same manner.
Several limitations should be considered when interpreting our results. First, given the nature of the SEER database, some degree of selection bias was unavoidable due to the lack of information regarding general condition or details of comorbidities. In addition, the SEER database includes high rate of unknown status regarding to tumor location, tumor depth, histological grade, and tumor size. On the other hand, considering that soft tissue sarcoma comprises a variety of histological subtypes with limited numbers of patients in each histological subtype, a large, population-based cohort is useful to address the risk factors for brain metastasis in patients with soft tissue sarcoma. Our results must, therefore, be confirmed in future studies. Second, the development of brain metastasis was investigated only at initial presentation, not during the subsequent clinical course. Local treatment or systemic therapy might affect the development of brain metastasis during treatment. Finally, the definition of brain metastasis in the SEER database represents a clinicopathological judgement at initial presentation. Not all patients received imaging screening for the brain. Our results regarding the incidence of brain metastasis might thus be underestimates, considering the possibility of asymptomatic brain metastases.
In conclusion, our study provides new insight into brain metastasis among patients with STS at initial presentation. To the best of our knowledge, this represents the first report that shows the risk factors for brain metastasis in patients with soft tissue sarcoma using SEER database which is largest public cancer database. Although the next step will be a database study using other registry or prospective studies to confirm our results, oncologists should be careful if STS patients present with risk factors for brain metastasis at initial presentation.
Footnotes
Author’s Contributions
MM was involved in the design of the study, performed the clinical assessments, analysis, and interpretation of data, and drafted and revised the manuscript. RI, TO and IY assisted with data interpretation and revised the manuscript for important intellectual content. KI, SM, RH, YS, AI, EK, and NI were involved in data acquisition and revised the manuscript critically for important intellectual content. All Authors have read and approved the final manuscript.
Conflicts of Interest
IY reports grants from KAKENHI, AMED, and Health, Labour and Welfare Policy Research Grants, research fund by Nihon Medi-Physics, and lecture fees from Chugai Pharmaceutical Co, AstraZeneca plt, Japan Tabacco Pharamaceutical Division, and Nippon Shinyaku Co, outside the submitted work.
- Received August 14, 2021.
- Revision received September 10, 2021.
- Accepted September 29, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
