Abstract
Background/Aim: Brain metastases from ovarian cancer remain rare and the appropriate treatment is unknown. We investigated survival outcomes following salvage chemotherapy before and after bevacizumab approval to evaluate the efficacy of bevacizumab in patients with brain metastasis from ovarian cancer. Patients and Methods: We investigated 23 consecutive patients with brain metastasis from ovarian cancer at our hospital between 2001 and 2020. Bevacizumab was administered for treating ovarian cancer after approval in Japan in November 2013. Survival after brain metastasis was compared between 9 patients treated before bevacizumab approval (2000-2013) and 14 patients treated after approval (2014-2020). Seven patients treated in the latter period received bevacizumab-salvage chemotherapy for brain metastasis. Results: Median survival in all patients was 9.1 months [95% confidence interval (CI)=4.2-33.5]. In addition, patients treated during the latter period presented better survival outcomes than those treated in the former period (former, 2.9 months vs latter, 33.5 months, log-rank test, p=0.015; Wilcoxon test, p=0.009). Multivariate analysis revealed that bevacizumab addition (p=0.020), interval to brain metastasis (p=0.005), number of brain lesions (p=0.001), number of recurrences (p=0.001), and platinum sensitivity (p=0.028) were independently associated with survival in all cohorts. Conclusion: Bevacizumab-based salvage chemotherapy may improve survival outcomes in patients with brain metastasis.
Brain metastasis is a serious manifestation observed during the treatment of solid tumors and indicates poor prognosis, consistently determining the fatality of these cancers (1-3). Reportedly, the incidence of brain metastasis was highest in patients with lung cancer (19.9%), followed by melanoma (6.9%), renal (6.5%), breast (5.1%), and colorectal (1.8%) cancers (4); brain metastases from ovarian cancer remain rare, with an incidence of approximately 1% (5). However, the incidence of brain metastases from ovarian cancer has increased in recent years following long-term disease stabilization, as the introduction of novel target agents for ovarian cancer has improved survival outcomes. Although the incidence of brain metastases is recognized as a late manifestation, some researchers have reported that platinum-sensitive recurrence even after brain metastases is a better prognostic factor for survival (2).
Various treatments are available for patients with brain metastases, including craniotomy, chemotherapy, and whole-brain radiotherapy, as well as stereotactic radiosurgery (6). However, the standard treatment for brain metastasis from ovarian cancer has not been established as brain metastases are rare in this type of cancer. Reportedly, chemotherapy combined with whole-brain radiotherapy affords a superior prognosis than radiotherapy alone for treating brain metastasis from ovarian cancer (1). Typically, craniotomy is the standard treatment for patients with larger and symptomatic lesions, whereas systemic chemotherapy is often combined with radiation therapy. As whole-brain radiotherapy has a high incidence of cognitive toxicity, stereotactic radiosurgery is performed when less than 4 lesions are present. Systemic chemotherapy is considered to afford limited benefits, given the presence of the blood-brain barrier (BBB). In contrast, it has been speculated that the BBB is disrupted in the presence of intracranial metastases and that systemic chemotherapy is effective for brain metastasis.
Bevacizumab, a monoclonal antibody targeting vascular endothelial growth factor (VEGF), has demonstrated significant clinical benefits in diverse carcinomas. Angiogenesis is essential for the development of brain metastases, as explained by the “seed and soil” hypothesis (7), and antiangiogenic treatments may be efficacious for treating brain metastases (8). Bevacizumab inhibits VEGF-induced neovascularization, as well as normalizes the abnormal VEGF-mediated hyper-permeability of tumor vessels. This normalizing effect decreases interstitial fluid pressure, alleviates tumor hypoxia, and improves drug delivery into tumor tissues, thereby increasing the efficacy of conventional therapies (9). This mechanism has been investigated in metastatic lung cancer and glioblastoma. Although bevacizumab was approved for recurrent and advanced ovarian cancer in Japan in 2014 (10), it remains unclear whether bevacizumab has clinical benefits for treating patients with brain metastasis from ovarian cancer. In the present study, we retrospectively investigated the efficacy of bevacizumab in patients with brain metastasis and compared survival outcomes before and after bevacizumab approval at our hospital.
Patients and Methods
Patients. This retrospective, case-control study included 23 patients with ovarian cancer diagnosed with brain metastasis who were treated at Chiba University Hospital between January 2000 and December 2020. Written informed consent was obtained from all patients before treatment for brain metastasis. Computed tomography (CT) and magnetic resonance imaging (MRI) were performed to diagnose brain metastases. The medical records were systematically reviewed to evaluate demographic and clinical characteristics, including initial stage [International Federation of Gynecology and Obstetrics (FIGO 2014) classification (11)], histological subtype, age at the onset of brain metastasis, number of recurrences, interval period to brain metastasis, and number of lesions at diagnosis of brain metastasis. In addition, the type of treatment received, administration of bevacizumab, number of bevacizumab cycles administered, bevacizumab induced adverse effects, and classification of patients according to platinum sensitivity were documented. Patients were considered platinum-sensitive if the period from completion of the first or previous platinum-based chemotherapy regimen to disease recurrence/progression was >6 months; patients were regarded as platinum-resistant if this period was ≤6 months. Bevacizumab was administered after approval in Japan in November 2013 for the initial treatment of FIGO III/IV and recurrent ovarian cancer. Survival from brain metastasis was compared between 9 patients treated before bevacizumab approval (2000-2013, former period) and 14 patients treated after bevacizumab approval (2014-2020, latter period) (Figure 1). The present study was approved by the Institutional Review Board of Chiba University (#M10088).
Study protocol comparing treatment periods for brain metastasis. In total, 21 patients received whole-brain radiation therapy or gamma knife radiosurgery. Five patients received craniotomy due to larger and symptomatic metastasis. #Two patients received palliative therapy for brain metastasis due to poor performance status.
Chemotherapy and bevacizumab. Weekly paclitaxel (80 mg/m2/week injected intravenously) and carboplatin (AUC 2-3/week injected intravenously) were administered for platinum-sensitive recurrence. Pegylated liposomal doxorubicin (40 mg/m2 for 4 weeks) or gemcitabine and irinotecan combination chemotherapy (gemcitabine 500 mg/m2 and irinotecan 50 mg/m2 for 3 weeks) was administered for platinum-resistant recurrence (12). Bevacizumab (15 mg/kg every 3 weeks) was administered to patients without any contraindication to its use after approval in November 2013 in Japan for treating ovarian cancer. The Common Terminology Criteria for Adverse Events scale, version 4.0, published by the National Cancer Institute, was used to grade toxicity.
Statistical analysis. The primary endpoint was survival after brain metastasis. Survival was defined as the time interval between treatment initiation for brain metastasis and the date of death or the last follow-up. The Kaplan–Meier method was used to estimate survival. Log-rank and Wilcoxon tests were used to compare statistically significant differences. In addition, patient characteristics were compared between former and latter periods using Fisher’s exact test or the Chi-square test. The variables included in the multivariate analysis were accepted using backward stepwise selection based on the corrected Akaike’s information criterion. Cox proportional hazards regression analysis was performed using five variables to analyze prognostic factors associated with survival after brain metastasis. All statistical analyses were two-sided, and a p-value <0.05 was considered statistically significant. Data analyses were performed using JMP statistical software (version 15.0; SAS, Cary, NC, USA).
Results
Patient characteristics. As shown in Table I, no statistically significant intergroup difference was observed in patient characteristics, except for platinum sensitivity and the addition of bevacizumab. During initial treatment, an advanced stage (III/IV) was diagnosed in 5 (56%) and 12 (86%) patients in the former and latter periods (p=0.253), respectively. Serous carcinoma was detected in 5 (56%) and 8 (57%) patients in the former and latter periods (p=1.000), respectively. Platinum-resistant disease was documented in 7 (78%) and 4 (29%) patients in the former and latter periods (p=0.036), respectively. The median number of brain lesions at brain metastasis was one for all included patients, and no significant differences were observed between groups (p=0.417). The median interval period to brain metastasis was 33.7 months [interquartile range (IQR)=10.7-44.2], and no significant differences were observed between groups (p=0.900). The median number of recurrences was 2 (IQR=1-3) in all patients, and no significant differences were observed between the groups (p=0.060). While 16 patients had extracranial diseases at the diagnosis of brain metastasis, the remaining 7 had isolated brain metastasis. In this study, 3 patients had BRCA1 mutation and 3 patients had BRCA wild type. These 6 patients did not have the event of death. BRCA status was unknown for the remaining patients.
Patient characteristics.
Treatment of brain metastasis. The treatment of brain metastasis in the present study is shown in Figure 1 and Table I. Overall, 21 patients received whole-brain radiation therapy or gamma knife radiosurgery. Two patients received palliative therapy for brain metastasis due to poor performance status: one received only steroid and glycerol for brain edema during the former period and one underwent salvage surgery with ventricular drainage for hydrocephalus during the latter period. In both periods, five patients received only whole-brain radiation therapy and five patients underwent craniotomy because of large and symptomatic metastasis. Sixteen patients received radiotherapy and surgery followed by additional chemotherapy based on the disease burden and performance status. Sixteen patients received gamma knife radiotherapy followed by additional chemotherapy. In the former period, four of nine patients received whole-brain radiation therapy or gamma knife radiotherapy only. In the latter period, one of 14 received gamma knife radiotherapy only. In the former period, four patients had received chemotherapy and three of them had received prior craniotomy. In the latter period, 12 patients had received chemotherapy, two of them had received prior craniotomy, and seven of them received chemotherapy with bevacizumab. No significant differences were observed in the number of patients who underwent craniotomy (p=0.343), radiation (p=1.000), or chemotherapy (p=0.066) between the two periods (Table I). Additionally, no significant differences were documented in the number of patients who received gamma knife radiotherapy followed by additional chemotherapy between the two periods (p=0.102).
Bevacizumab administration and adverse effects. During the former period, none of the included patients received bevacizumab prior to approval. In the latter period, we administered bevacizumab to patients with no known contraindications to the drug. Seven patients received chemotherapy with bevacizumab during the latter period. The median cycle of bevacizumab administration was 10 (IQR=4-13). During the latter period, five patients did not receive bevacizumab, as more than four chemotherapy regimens had been administered (n=2) and due to adverse effects of bevacizumab during the previous regimen (n=2). Seven patients received chemotherapy with bevacizumab. No patient exhibited intracranial bleeding during bevacizumab administration. However, one patient experienced hypertension (grade 3) and proteinuria (grade 3).
One patient developed brain metastasis during maintenance therapy with olaparib after platinum-sensitive recurrence. Despite progression during maintenance therapy with olaparib, the patients requested maintenance therapy with olaparib after gamma knife treatment. Notably, this patient experienced the next recurrence of multiple lung metastases 13 months later and was detected to possess a BRCA1 mutation.
Survival analysis. Figure 2 presents the survival after brain metastasis. The median follow-up period was 8.6 months (IQR=3.8-15.7). For all included patients, the median survival was 9.1 months [95% confidence interval (CI)=4.2-33.4]. The median survival during the former and latter periods was 2.9 months (95% CI=0.5-11.9) and 33.5 months (95% CI=4.2- not reached, log-rank test p=0.015, Wilcoxon test p=0.009), respectively (Figure 3).
Survival after brain metastasis. The median survival was 9.1 months in this study.
Survival after brain metastasis between the former and latter periods. The median survival during the former and latter periods was 2.9 and 33.5 months, respectively.
Multivariate analysis of prognostic factors associated with survival after brain metastasis. The backward stepwise selection model accepted the following variables: addition of bevacizumab, interval to brain metastasis, number of metastatic brain lesions, number of recurrences to brain metastases and platinum sensitivity. The treatment periods (former and latter periods) were not accepted. Cox proportional hazards regression analysis was performed to analyze prognostic factors associated with survival after brain metastasis (Table II). The addition of bevacizumab [hazard ratio (HR)=0.04, 95% CI=0.00-0.62, p=0.02], interval to brain metastasis (HR=0.06, 95% CI=0.01-0.43, p=0.005), number of brain lesions (HR=0.01, 95% CI=0.00-0.12, p=0.001), number of recurrences to brain metastasis (HR=0.03, 95% CI=0.01-0.29, p=0.001), and platinum sensitivity (HR=0.14, 95% CI=0.02-0.81, p=0.028) were independently associated with survival.
Univariate and multivariate cox proportional analysis of risk factors for survival after brain metastasis.
Discussion
In patients with ovarian cancer, the prognosis after detection of brain metastasis is poor. The MITO-19 study has shown a median survival period of 12 months (3). In the present study, treatment in the latter period, during which a higher number of patients received bevacizumab in 2014 or later, significantly improved survival when compared with those treated during the former period (before 2013), when bevacizumab was not administered. Although multidisciplinary treatment strategies might have improved the prognosis after brain metastasis since 2014, univariate and multivariate analyses revealed that the addition of bevacizumab, along with previously indicated clinical prognostic factors for brain metastases, was an independent favorable prognostic factor for survival after brain metastasis. Accordingly, the backward stepwise selection did not accept the treatment period. No previous reports have documented the efficacy of bevacizumab for patients with brain metastases from ovarian cancer; however, the present study suggests that the bevacizumab administration may benefit patients with ovarian cancer presenting brain metastases. Specific features of the BBB have been reported to affect the systemic treatment of metastatic tumors in the brain (13, 14), as hydrophilicity and the large molecular size of an antibody are considered disadvantageous for BBB penetration. However, the efficacy of bevacizumab in metastatic brain tumors has been previously demonstrated in other carcinomas. In brain metastasis in severe combined immunodeficiency mice inoculated with non-small-cell lung cancer (NSCLC) cells (15), the activity of NSCLC cells in the brain was significantly lower in bevacizumab-treated mice than in human immunoglobulin-treated mice. Additionally, the brain bevacizumab concentration was higher in mice with brain metastasis than in normal mice. Therefore, the anti-proliferative effect of bevacizumab on brain metastasis could be attributed to the anti-angiogenic bevacizumab concentration achieved, given its penetration into brain metastases (15). Reportedly, the expression of VEGF-A was significantly higher in metastatic brain samples than in primary site in gynecologic malignancies and anti-VEGF therapy should be considered in treatment of brain metastases for gynecologic malignancies (16). Several trials have reported the efficacy of bevacizumab-based chemotherapy in patients with asymptomatic and untreated brain metastases from NSCLC (8, 17, 18). These results suggest that bevacizumab may cross the BBB.
Intracranial hemorrhage remains a major concern during bevacizumab therapy. Patients with intracranial metastases were previously excluded from bevacizumab trials based on an early case report of a 29-year-old man with hepatocellular carcinoma who experienced a fatal intracranial hemorrhage in a phase I study (19). However, some retrospective studies of brain metastases from other cancers have revealed that patients with brain metastases had a similar risk of developing intracranial hemorrhage as those who did and did not receive bevacizumab therapy (20, 21). In addition, the incidence of intracranial hemorrhage was similarly lower in prospective trials assessing patients with brain metastases from lung cancer (22-26). A review article that included 7096 patients with ovarian cancer treated with bevacizumab has reported the incidence of 5 cases (0.07%) of intracranial hemorrhage (27). Therefore, in the absence of apparent intracranial hemorrhage, the administration of bevacizumab can afford a substantial clinical benefit in patients with brain metastases from ovarian cancer.
In addition to bevacizumab, poly ADP-ribose polymerase (PARP) inhibitors are employed as a maintenance drug for platinum-sensitive recurrent ovarian cancers; however, which treatment could afford superior survival remains unknown (28). In the future, the precise use of bevacizumab or PARP inhibitors may present challenges in maintenance therapy for brain metastasis, given the potential for serious damage. According to the “seed and soil” hypothesis (7), antiangiogenic therapy with bevacizumab could be efficacious for brain metastasis. Recently, a case report has documented a patient with serous carcinoma of the ovary who underwent stereotactic radiosurgery and chemotherapy, followed by treatment with PARP inhibitors (29). In addition, maintenance therapy with PARP inhibitors was similarly successful in patients with brain metastasis from serous carcinoma of the uterine corpus (30). Both cases involved BRCA mutations. Therefore, BRCA mutation may be a useful biomarker for determining the use of bevacizumab or PARP inhibitors. In contrast, histology may be another biomarker of bevacizumab. We have reported the efficacy of bevacizumab for clear cell carcinoma during initial treatment in the previous study (31). The patient with clear cell carcinoma had the low incidence of BRCA mutations, reportedly (32). Therefore, we believe that the chemotherapy with bevacizumab is efficacious in patients with clear cell carcinoma and brain metastases.
There are no previous reports on the efficacy of bevacizumab for patients with brain metastases from ovarian cancer; however, the present study suggests that the administration of bevacizumab may be beneficial in patients with brain metastases from ovarian cancer. There are some limitations to our study. First, as a retrospective study over 20 years, the primary treatment policy for brain metastasis from ovarian cancer, surgery, or radiotherapy, was inconsistent. Second, the patient population was extremely small to analyze the efficacy of bevacizumab, and patients with brain metastasis from ovarian cancer are extremely rare. Third, differences in second-line therapy after brain metastases in the intergroup periods may also affect overall survival. However, only three and four patients in the former and latter periods, respectively, received bevacizumab-containing regimens as second-line therapy, suggesting that second-line therapy is unlikely to affect overall survival.
In conclusion, consistent with previous analyses for other cancers, the present study demonstrated the encouraging efficacy and tolerable safety profile of combination stereotactic radiosurgery and systemic chemotherapy with bevacizumab. In addition, bevacizumab improved the control of brain metastases and clinical outcomes in patients with brain metastases from ovarian cancer. As the incidence of brain metastasis from ovarian cancer is low, it is unclear whether the safety in patients with lung cancer receiving bevacizumab can be adapted to ovarian cancer patients with brain metastasis. The safety and efficacy of bevacizumab for brain metastasis reported in the present study will be clarified in further studies.
Footnotes
Authors’ Contributions
Conceptualization, S.T. and M.S.; methodology, S.T.; formal analysis, S.T.; investigation, S.T.; data curation, S.T. and K.N.; writing – original draft preparation, S.T.; writing – review and editing, S.T., K.N., A.M., and S.O.; visualization, S.T.; supervision, M.S.; project administration, S.T. All Authors have read and agreed to the published version of the manuscript.
Conflicts of Interest
None.
- Received February 20, 2022.
- Revision received March 18, 2022.
- Accepted March 23, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.