Abstract
Background/Aim: Over the past several decades, new anti-cancer drugs have been developed for the treatment of epithelial ovarian cancer. The development of drugs has led to changes in improving the prognosis of ovarian cancer patients. One of these drugs, bevacizumab, is used for advanced or recurrent ovarian cancer. In this study, we aimed to evaluate survival improvement in patients with platinum-resistant relapsed epithelial ovarian cancer (PR-ROC) after introduction of bevacizumab in real world experience. Patients and Methods: We retrospectively divided patients with PR-ROC into two groups: bevacizumab plus chemotherapy (BEV-CT group) and chemotherapy alone (CT group). Progression-free survival (PFS), the primary endpoint, between two groups was compared to evaluate whether survival outcomes were improved. In addition, overall survival (OS) was also compared. Results: A total of 154 patients were included in the study: 57 and 97 patients in the BEV-CT and CT groups, respectively. OS was significantly longer in the BEV-CT group than in the CT group. The use of bevacizumab was identified as a favorable prognostic factor for OS. In a subgroup analysis confined to second-line chemotherapy, PFS and OS were statistically different between groups. More patients in the CT group suffered hematologic adverse events of grade 3 or above than patients in the BEV-CT group. Conclusion: In a real-world clinical setting, introduction of bevacizumab led to improvement of OS in patients with PR-ROC with a tolerable toxicity.
Epithelial ovarian cancer (EOC) is the most fatal gynecological cancer and ranks among the top 10 leading causes of female cancer deaths worldwide (1). Most patients with EOC are diagnosed with advanced disease and undergo platinum-based chemotherapy after surgical treatment (2). Patients with platinum-resistant relapsed epithelial ovarian cancer (PR-ROC) have a median survival of only 9 to 12 months, and fewer than 15% of these respond to chemotherapy after relapse (3). Since the prognosis of patients with PR-ROC is very poor, several studies have been conducted on various types of drugs that have been developed to increase the survival outcomes. To overcome the poor prognosis of PR-ROC, bevacizumab was developed as a putative therapy.
Bevacizumab is a humanized immunoglobulin G1 monoclonal antibody that targets angiogenesis by inhibiting vascular endothelial growth factor (4). It is the first anti-angiogenic drug to demonstrate survival benefits in patients with cancer and was shown to improve survival by 30% when administered with standard chemotherapeutic regimen to treat patients with metastatic colorectal cancer (5). The improvement in survival was also studied in EOC. The Avastin Use in Platinum-Resistant Epithelial Ovarian Cancer (AURELIA) trial was the first open-label randomized phase III trial to study the effect of bevacizumab combined with cytotoxic chemotherapy in patients with PR-ROC (6). The patients treated with bevacizumab showed improved progression-free survival (PFS) (6).
In Korea, the use of bevacizumab as a treatment for patients with PR-ROC was covered by the National Health Insurance in August 2015. Since that time, the number of patients receiving bevacizumab treatment for PR-ROC has increased, and a study about bevacizumab treatment in Korea has been conducted. The Korean Gynecologic Oncology Group study (KGOG 3041) compared the efficacy of three chemotherapy regimens containing bevacizumab and reported differences in survival outcomes (7).
We conducted a single center retrospective study in a real-world clinical setting to evaluate the improvement of survival outcomes both before and after bevacizumab treatment for patients with PR-ROC. The aim of this study was to identify changes in survival outcomes of patients with PR-ROC after introduction of bevacizumab and evaluate adverse events for safety comparison.
Patients and Methods
Study population. This retrospective cohort study was conducted after approval from the Institutional Review Board (IRB) of Seoul National University Hospital (No.2110-073-1261).
From the Institution’s ovarian cancer cohort, we identified patients with PR-ROC who received second- or third-line chemotherapy between September 2007 and March 2020. Patients with the following conditions were included: 1) those with EOC; 2) those who had received platinum-based chemotherapy as a first- and/or second-line chemotherapy; and 3) those who underwent more than two cycles of second- or third-line chemotherapy after confirmation of PR-ROC. Meanwhile, we excluded patients who received poly (ADP-ribose) polymerase (PARP) inhibitors, immunotherapies, or other medications related to clinical trials. The code for EOC, according to the International Statistical Classification of Disease, 10th revision (ICD-10), is C56.
Data collection. Through a review of medical records, we collected data on chemotherapy treatment, survival information, and clinicopathological data, such as patient age, histologic type, histologic grade, International Federation of Gynecology and Obstetrics (FIGO) stage, results of debulking surgery, and serum Cancer Antigen-125 (CA-125) levels, chemotherapy regimen.
Bevacizumab-containing regimens for the treatment of patients with PR-ROC included paclitaxel-bevacizumab, topotecan-bevacizumab, and pegylated liposomal doxorubicin (PLD)-bevacizumab. Regimens containing bevacizumab were as follows: paclitaxel 80 mg/m2 intravenously (IV) on days 1,8,15 and 22 every 4 weeks; PLD 40mg/m2 IV on day 1 every 4 weeks; or topotecan 4mg/m2 IV on days 1,8, and 15 every 4 weeks or 1.25 mg/m2 on days 1 to 5 every 3 weeks. Chemotherapy regimens were decided by the gynecologic oncologist after considering previous treatment regimen, Korean National Health Insurance coverage, and previous adverse events.
Patients were divided into two groups according to the use of bevacizumab for their treatment of PR-ROC: bevacizumab plus chemotherapy group (BEV-CT group) and chemotherapy alone group (CT group). Thereafter, survival outcomes were calculated, and treatment-related adverse events were also noted. Survival outcomes and chemotherapy-induced toxicity were compared between the two groups.
Endpoints. The primary endpoint was progression-free survival (PFS), from the first day of chemotherapy treatment to recurrence or progression. Secondary endpoints were overall survival (OS), from the first day of chemotherapy treatment to death and chemotherapy induced toxicities. For patients who were transferred to a hospital for terminal palliative therapy, 4 weeks after the last follow-up was defined as the date of death. Tumor assessment were performed according to the Response Evaluation Criteria in Solid Tumors (RECIST, version 1.1) criteria using computed tomography (8). Tumor markers were also used as references to evaluate treatment response. Toxicity were evaluated by Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 (9).
Statistical analysis. Survival analysis was performed using the Kaplan-Meier method, and results were compared using the log-rank test. Statistical analysis was performed using SPSS (version 25.0; IBM, Armonk, NY, USA). Categorical data were described as frequencies and percentages, and numerical data were described as median and ranges or as means and standard deviations. Hazard ratios (HRs) and 95% confidence intervals (CIs) for survival analysis were estimated using the Cox proportional hazard model. p-Values less than 0.05 were considered statistically significant. In accordance with the journal’s guidelines, we will provide our date for the reproducibility of this study in other centers if such is requested.
Results
Patient characteristics. A total of 154 patients with PR-ROC were evaluated over the study period: 57 patients in the BEV-CT group and 97 patients in the CT group. Of the total study population, 93 patients became resistant to platinum after first-line chemotherapy, and 61 patients became resistant to platinum after second-line chemotherapy. Patient characteristics, such as age at diagnosis, histologic type, histologic grade, FIGO stage, and baseline CA-125 at recurrence were not significantly differ between two groups. Baseline characteristics are summarized in Table I.
Clinicopathological characteristics of the study population.
Chemotherapy regimens containing bevacizumab included PLD-bevacizumab, topotecan-bevacizumab, and paclitaxel-bevacizumab. Thirty patients in the BEV-CT group were treated with PLD-bevacizumab and six patients were treated with topotecan-bevacizumab. Only two patients were treated with paclitaxel-bevacizumab. Second- or third-line chemotherapy regimen in the CT group were belotecan alone regimen, topotecan alone regimen, topotecan-cisplatin, and belotecan-cisplatin, etc. Detailed chemotherapy regimen is described in Table II.
Detailed chemotherapy regimen of the study population.
Comparison of survival outcomes. The cut-off date for the analysis of survival outcomes was September 30, 2021. The median duration of follow-up was 14.0 months in the BEV-CT group versus 11.4 months in the CT group.
The median PFS were 3.9 (3.0-4.8) months and 5.5 (4.2-6.8) months in the CT and BEV-CT groups, respectively (Figure 1A). There were not significantly differences between the two groups (p=0.216). The median OS was 16.0 (12.6-19.4) months and 29.0 (8.5-49.5) months in the CT and BEV-CT groups, respectively (Figure 1B). The OS of BEV-CT group was significantly longer than that of the CT group (p=0.007).
Kaplan-Meier curves for progression-free survival and overall-survival in the BEV-CT group and CT group. BEV-CT: Bevacizumab plus chemotherapy; CT: chemotherapy alone.
As for recurrence, the results from the multivariate analysis showed that the BEV-CT group did not have significantly better PFS than the CT group (HR=1.199; 95% CI=0.817-1.758; p=0.354), but OS was significantly better in the BEV-CT group than in the CT group (HR=2.059; 95% CI=1.196-3.545; p=0.009) (Table III and Table IV).
Univariate and multivariate analyses for progression-free survival in the study population (n=154).
Univariate and multivariate analyses for overall survival in the study population (n=154).
Subgroup analysis. A subgroup analysis of patients with PR-ROC who had prior first-line chemotherapy was conducted, and 93 patients were evaluated: 32 patients in the BEV-CT group and 61 patients in the CT group. Patient characteristics, such as age at diagnosis, age at recurrence, histological grade, FIGO stage, CA-125 at baseline and at recurrence did not significantly differ between the two groups. However, the platinum-free-interval (from the last day of platinum-based chemotherapy treatment to recurrence or progression) of BEV-CT group were significantly longer than those of CT group (p=0.008). Baseline subgroup characteristics are shown in Table V.
Clinicopathological characteristics of second-line chemotherapy patients.
The median PFS was 3.2 (2.3-4.0) months and 5.8 (4.0-7.6) months in the CT and BEV-CT groups, respectively (Figure 2A). The p-value for the stratified log-rank test was 0.028. Results of the multivariate subgroup analysis revealed that the BEV-CT group had significantly better PFS than the CT group (HR=1.706; 95% CI=1.019-2.858, p=0.042). The median OS of the CT group was 13.7 (10.4-17.0) months, but the OS of BEV-CT group has not reached. The 2-year survival rate of the CT group was 23.6% and that of BEV-CT group was 52.3% (Figure 2B). The OS of the BEV-CT group was significantly longer than that of the CT group (p=0.021). Additionally, results of the multivariate subgroup analysis revealed that the BEV-CT group had significantly better OS than the CT group (HR=2.323; 95% CI=1.136-4.751, p=0.021). These results are shown in Table VI and Table VII.
Kaplan-Meier curves for progression-free survival and overall-survival in the BEV-CT group and CT group for subgroup analysis. BEV-CT: Bevacizumab plus chemotherapy; CT: chemotherapy alone.
Univariate and multivariate analyses for progression-free survival in the second-line chemotherapy group (n=93).
Univariate and multivariate analyses for overall survival in second line chemotherapy group (n=93).
Other subgroup analysis of patients with PR-ROC after previous second-line chemotherapy were performed. Baseline subgroup characteristics are shown in Table VIII. There were no significant differences in PFS and OS between the CT and BEV-CT groups (Figure 3A and B). Results of the multivariate analysis also revealed that there were no significant differences in PFS and OS between the two groups, as shown in Table IX and Table X.
Clinicopathological characteristics of third line chemotherapy patients.
Comparisons of progression free survival and overall survival in the BEV-CT group and CT group after third line chemotherapy. BEV-CT: Bevacizumab plus chemotherapy; CT: chemotherapy alone.
Univariate and multivariate analyses for progression free survival in third line chemotherapy group (n=61).
Univariate and multivariate analyses for overall survival in third line chemotherapy group (n=61).
Adverse events. The incidence of non-hematologic events, grade 3 or above, was not significantly different between the two groups (p=0.143). Three patients (5.3%) in the BEV-CT group and only one (1.0%) patient in the CT group experienced non-hematologic events of grade3 or above. For hematologic toxicities of grade 3 or above, a higher proportion of patients in the CT group suffered adverse events than those in the BEV-CT group (p=0.001): 55 patients (56.7%) in the CT group experienced hematologic toxicities during their second- or third-line chemotherapy and 16 patients (28.1%) in the BEV-CT group experienced hematologic toxicities. Details of these adverse effects are described in Table XI.
Comparisons of toxicities above grade 3 in the CT and BEV-CT groups (n=154).
Discussion
In this study, the median OS in the BEV-CT and CT groups after platinum resistant recurrence were 29.0 and 16.0 months, respectively. As bevacizumab was introduced, improvement in OS, but not in PFS, of patients with PR-ROC was observed. However, in patients who became platinum-resistant during the first regimen, second-line treatment with BEV-CT led to improvements in both PFS and OS.
Several studies have investigated the efficacy of regimens including bevacizumab for the treatment of PR-ROC. In the first studies investigating bevacizumab as a monotherapy for patients with recurrent EOC (10, 11), the median PFS was 4.4-4.7 months and median OS was 10.7-16.9 months (10, 11). Several phase II studies investigated different bevacizumab-cytotoxic chemotherapy regimens, including PLD-bevacizumab, topotecan-bevacizumab, paclitaxel-bevacizumab, and cyclophosphamide-bevacizumab (12-15). In these studies, the median PFS was 7.8-8.08 months, and the median OS was 16.6-33.2 months.
Based on these results, the AURELIA trial was conducted. The median PFS of the CT group was 3.4 months and the median PFS of the BEV-CT group was 6.7 months (p=0.001) (6). There were no significant differences in OS between the two groups (p=0.174). In our study, among 57 patients who were treated with bevacizumab, death occurred in only 19 patients during follow-up period, and there was a difference in follow-up periods between the two groups. The median OS in our study was longer than that in the AURELIA trial, but longer follow-up period is needed.
An exploratory analysis was conducted during the AURELIA trial: Some patients who developed progression disease (PD) after chemotherapy alone were selected to receive a bevacizumab combination therapy. Patients in both the new bevacizumab after PD group and the initial bevacizumab combination group had lower risk of death than patients in the chemotherapy alone group (HR=0.66; 95% CI=0.51-0.85, p=0.001) (16). Although differences in OS in exploratory analysis of AURELIA trial were consistent with the results of our study, more patients who enrolled in the exploratory analysis during the AURELIA trial developed primary platinum resistance than in our study. For this reason, patients in the AURELIA trial may have a worse survival prognosis than patients in our study.
Compared to the results of the KGOG 3041 study, and because most patients in the BEV-CT group of our study received PLD-bevacizumab, the PLD-bevacizumab group in KGOG 3041 study was comparable to ours (7). In the KGOG 3041 study, the median PFS and median OS of the PLD cohort were 5.4 months and 21 months, respectively. The median PFS in our study was consistent with that of the KGOG 3041 study, but the OS was longer in our study than in the KGOG 3041 study. Because more patients in the PLD cohort in the KGOG 3041 study received the second-line bevacizumab regimen than in our study, a difference in OS was evident.
In the survival outcomes of our study, there was a prolongation of OS in the BEV-CT group, but not PFS. This discrepancy is often observed in studies involving biologic/targeted agents, despite the high quality of the study as indicated by the Physiotherapy Evidence Database (PEDro) scale (17). The causes of these results, as mentioned, include the drug’s mechanism of action, the specific disease being treated, the population under study, and an imbalance in post-study treatment (17).
In the population of our study, there is time difference between the two groups before and after the introduction of bevacizumab, so there could be imbalance of post-study treatment. It is also possible that the action mechanism of bevacizumab, which was not fully understood, could have influenced post-study treatment and survival outcomes.
Our study analyzed the change in survival prognosis of patients with PR-ROC by comparing survival both before and after introduction of bevacizumab. After the introduction of bevacizumab, out study showed an increase in OS in the BEV-CT group despite no differences in baseline patient characteristics. Since we performed the analysis after excluding patients who had received a PARP inhibitor, an immunotherapy, or a clinical trial medication, we were able to evaluate the effects of bevacizumab alone. Therefore, our study has significant strengths in elucidating the trajectory of the prognosis of PR-ROC over time, particularly with the development of novel therapeutic agents. This exploration of changes in treatment approaches over time to track the impact of newly developed drugs can be directly used in the education and care of patients with PR-ROC.
Although our study yielded meaningful results, it had certain limitations. First, there was a difference in the follow-up period between the BEV-CT and CT groups due to the study design. It is necessary to follow-up for a longer period until equivalent numbers of deaths are reached in all groups. The second limitation was the small sample size due to a single-center study design. However, the small sample size was sufficient for the comparison of survival results between the two groups, which showed no significant differences in patient characteristics. Third, we did not analyze breast cancer susceptibility gene (BRCA) status because the CT group rarely underwent BRCA testing because BRCA testing recently been covered by the national health insurance. Our study may not fully reflect differences in tumor biology between the two groups, as there were missing data as a result of the retrospective nature of the study.
Since the introduction of bevacizumab, the number of indications that can be used in clinical settings has increased. Bevacizumab can be used as first-line chemotherapy for advanced ovarian cancer and can be applied to both platinum-sensitive recurrence and platinum-resistant recurrence when recurrence occurs after platinum-based chemotherapy. Recently, there has been a study on the effectiveness in these various indications, and it has been revealed that bevacizumab was effective in real world clinical setting (18). In addition, the effectiveness of bevacizumab with platinum-based chemotherapy followed by bevacizumab maintenance therapy was proven during PARP inhibitor therapy (19). And the effectiveness of combination therapy with immunotherapy was also proven in other cancer types (20). Bevacizumab is an important agent in several cancer types and various clinical settings.
In conclusion, we revealed that bevacizumab was a favorable prognostic factor for OS, something especially evident in second-line chemotherapy for PR-ROC patients. Moreover, the BEV-CT group experienced fewer severe hematologic adverse events than the CT group, suggesting the potential clinical benefits and safety of incorporating bevacizumab in the treatment of PR-ROC.
Acknowledgements
This work was supported by grants from the National Research Foundation of Korea funded by the Ministry of Science and ICT (grant number: 2020R1G1A1005711) and the Cooperative Research Program of Basic Medical Science and Clinical Science of Seoul National University College of Medicine (grant number: 800-20210297).
Footnotes
Authors’ Contributions
ML, SIK, HY, and AS designed the study and collected the data. AS and SIK analyzed and interpreted the data. AR and SIK wrote the manuscript. HSK, HHC, NHP, YYS revised the manuscript. All Authors contributed to the article and approved the submitted version.
Conflicts of Interest
The Authors have no competing interests to disclose in relation to this study.
- Received December 26, 2023.
- Revision received January 27, 2024.
- Accepted February 2, 2024.
- Copyright © 2024, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.