Real‑world effect of bevacizumab and eribulin on metastatic breast cancer using a propensity score matching analysis
- Authors:
- Published online on: January 13, 2023 https://doi.org/10.3892/mco.2023.2608
- Article Number: 12
-
Copyright: © Matsui et al. This is an open access article distributed under the terms of Creative Commons Attribution License.
Abstract
Introduction
Bevacizumab and eribulin are novel agents for the treatment of HER2-negative metastatic breast cancer. Eribulin improved overall survival in the EMBRACE trials (1). Eribulin has been shown to improve survival in patients with advanced recurrent breast cancer previously treated with anthracycline or taxane chemotherapy. It may be effective in the early treatment of recurrent breast cancer after surgery, but this has not been reported at this time. On the other hand, bevacizumab has improved progression-free survival but not overall survival in several clinical studies (2,3,4). However, there are many reports that have denied the effectiveness of bevacizumab for OS, even though it has been suggested to be effective in improving outcomes in metastatic disease from several tumor types (5-7). Recently, real-world data have been reported from ESME in France, and the efficacy of BEV in improving overall survival was shown (8). Its excellent tumor reduction effect is expected to improve symptoms and complications associated with recurrent tumors, and the lack of a difference in the incidence of serious adverse events suggests that bevacizumab is a drug that preserves patients' quality of life and inhibits rapid tumor progression.
At present, there is no standard for treatment with these drugs. Making a clinical decision about which treatment option to choose can be difficult. Clinical data on this question are lacking. Although there are some reports about an improved prognosis and fewer adverse events in other cancers by changes in the sequence of treatment (9,10), there have been no reports of differences in the effectiveness of BEV+PTX and eribulin according to their order of administration.
The purpose of this study was to compare two treatment strategies, eribulin followed by BEV + PTX versus BEV + PTX followed by eribulin, to determine whether the order of administration affects patient outcomes in the real world.
Patients and methods
Patients
The database of the TBCRG, a multicenter study group, was established to centralize real-world data on metastatic breast cancer from 14 institutions in Toyama. We used this database to evaluate the survival benefit in patients who were treated with bevacizumab + paclitaxel (BEV + PTX) and eribulin. All patients who started BEV + PTX and eribulin treatment for MBC from August 2011 to June 2018 were selected. Among the 264 patients recorded in the TBCRG database, 180 patients who started BEV + PTX and eribulin treatment for HER2-negative MBC were selected. Of these, 84 patients were treated with both BEV + PTX and eribulin sequentially regardless of the treatment line. The following data were collected from each institution: age, PS, estrogen receptor (ER) status, progesterone receptor (PgR) status, adjuvant chemotherapy, most common metastatic sites, number of metastatic sites, and treatment line.
We retrospectively reviewed the medical records of 60 cases within the 2nd line and analyzed the following items: overall survival, time to failure of the strategy, efficacy of treatment, and adverse events. Computed tomography (CT) was performed after 2 or 3 months of treatment with eribulin or BEV + PTX to assess the efficacy. Disease status was assessed according to the response evaluation criteria in solid tumors (RECIST), and adverse events were assessed by CTCAE version 4.0. TFS was calculated as the duration of BEV+PTX and eribulin administration.
In all institutions, BEV+PTX and eribulin were continued until progression or unacceptable toxicity. Bevacizumab (10 mg/kg) was administered biweekly, and paclitaxel (80 mg/m²) was administered 3 weeks on/1 week off. Eribulin (1.4 mg/m²) was administered 2 weeks on/1 week off.
To evaluate the influence of the order of treatment, we compared the efficacy of eribulin followed by BEV + PTX (arm E-B) with the efficacy of treatment with the reverse treatment sequence (arm B-E).
Statistical analysis
Arms E-B and B-E were compared with adjustment for imbalances in patient background factors using a propensity score matching analysis (PSMA) (11,12). The PSMA method was used to examine the consistency between the analysis results, thereby making the clinical findings as robust as possible. The propensity scores were estimated using a logistic regression model with treatment line, age, performance status, number of metastatic sites, recurrence, liver metastasis, and triple negative status. OS and TFS were analyzed using the Kaplan-Meier method and compared using the log-rank test. In addition, univariate Cox regression analysis with the treatment arm as a covariate was used to estimate the hazard ratio and its confidence interval. When survival curves crossed over during the follow-up time, resulting in violation of the proportional hazards assumption, the two-stage test proposed by Qiu and Sheng (13) was used to compare survival curves between treatment arms.
Patient background factors were compared between arms using Pearson's χ2 test or unpaired Student's t-test, where appropriate. P<0.05 was considered to indicate statistical significance. Statistical analyses were performed with JMP software version 14.0 and SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Baseline patient demographics and tumor characteristics
The baseline patient characteristics are summarized in Table I. Thirty-nine patients were treated with the B-E sequence, and 21 patients were treated with the E-B sequence; the median ages were 54.4 (range, 30-77) and 54 (range, 34-73) years, respectively. The ER- or PgR-positive rates were 84.6 and 52.4% for the B-E and E-B groups, respectively. Five patients in the B-E group and seven patients in the E-B group had triple-negative breast cancer. Significantly more patients in the B-E arm than in the E-B arm received it as first line treatment (32 vs. 9, P=0.0019). Twelve patients in the E-B arm had received capecitabine or S-1 previously.
Safety
Neutropenia was the most frequent grade ≥3 adverse event and had a similar incidence in both arms. There was one case of grade 3/4 hypertension as a result of treatment with BEV + PTX in each of the arms. There was only one case of grade 3/4 proteinuria in the B-E arm (Table IIA and B). There were no differences in adverse events between the two arms, and there were no deaths.
Efficacy
The overall response rates (ORRs) to eribulin treatment were 28.9 and 33.3% in the B-E and E-B groups, respectively, and the difference was not statistically significant (Table III). In contrast, the ORRs for BEV + PTX treatment were 66.6 and 23.9% in the B-E and E-B groups, respectively, and the difference was statistically significant (P=0.0147) (Table III). In the entire cohort, the median TFS was 16.8 and 9.9 months in the B-E and E-B arms, respectively (HR=0.515, 95% CI 0.298-0.889, P=0.017) (Fig. 1). Using PSMA with the caliper of 1.0, 19 pairs (38 patients) were identified, and their patient characteristics are summarized in Table IV. TFS was 16.9 and 9.9 months in the B-E and E-B arms, respectively (HR=0.491, 95% CI 0.253-0.952, P=0.031) (Fig. 2). The results (12 pairs, 24 patients) of PSMA with the caliper of 0.2, which was conducted as a sensitivity analysis (Fig. S1). Despite matching for treatment line, age, PS, triple-negative status, liver metastasis, and number of metastatic sites, the E-B and B-E arms showed significant differences in TFS.
Discussion
Patients who received BEV + PTX before eribulin (the B-E arm) had a significantly longer TFS than patients who received eribulin before BEV + PTX (the E-B arm) (16.8 vs. 9.0 months). Overall survival was also longer with the B-E treatment than with the E-B treatment (28.0 vs. 17.2 months). With regard to progression-free survival, while our result in the B-E arm was in line with that reported in E2100, RIBBON-1, AVADO, and JO19901, which were phase 2 and 3 trials of first-line therapies, our patients in the E-B arm had a shorter progression-free survival than obtained in those trials (2,3,4,14). On the other hand, progression-free survival with eribulin treatment was 3.7 and 4.1 months in the E-B and B-E arms, respectively, and the difference was not statistically significant. Thus, a significant difference appeared during the time to second progression. With regard to progression-free survival with eribulin, our results in both the E-B and B-E arms were in line with those reported in the EMBRACE trial and the 301 trials, which were phase 3 trials (1,15). When a regimen of BEV + PTX is administered late, progression-free survival is significantly decreased compared with initial eribulin treatment. The BEV + PTX regimen was the second-line treatment in the RIBBON-2 trial (16), in which the median progression-free survival was 9.1 months. In the present study, progression-free survival in the B-E arm was better than in the RIBBON-2 trial (11.5 months), a phase 3 trial designed to assess second-line bevacizumab-containing therapy, but it was significantly decreased in the E-B arm compared to the RIBBON-2 trial (6.2 months). It is supposed that BEV + PTX was administered after 2 regimens in 57.1% of the E-B arm. Therefore, we performed PSMA to minimize any imbalances in the background factors between the two arms. The PSMA analysis yielded results consistent with those in the overall analysis. Therefore, eribulin treatment is recommended after BEV + PTX treatment. A similar tendency for TFS was seen as a result of OS analysis by PSMA. However, a further follow-up survey is necessary because there were few observation events.
In regards to the mechanism, there is a difference in tumor vessel remodelling and reoxygenation between BEV + PTX and eribulin. Eribulin increases the density of tiny blood vessels and the supply of oxygenated blood to breast cancer tissue (17). Eribulin may improve the state of hypoxia relative to bevacizumab treatment. Furthermore, eribulin stabilizes the microenvironment and may improve the treatment effect. From these results, it is suggested that the administration of eribulin after BEV + PTX is most effective.
In the present study, BEV + PTX treatment and eribulin treatment were generally well tolerated. The incidence of adverse events with BEV + PTX was similar to that in previous clinical trials (2-4,14,16). Eribulin had a manageable profile of adverse events, consistent with those in previous clinical trials; neutropenia, alopecia, leukopenia, and peripheral neuropathy were the most common (1,18-22). The incidence of grade ≥3 neutropenia was 17.9% in the B-E arm, which was higher than that in the E-B arm. Therefore, considering the adverse events of subsequent treatments, we must administer BEV + PTX carefully. It is considered that the prognosis can be effectively improved by reducing adverse events by dose reduction, withdrawal or maintenance therapies. Indeed, recent data suggest that maintenance therapy has a positive effect on overall survival (23,24).
There are more ER+ in the B-E group. If TFS (and OS) is longer in the ER+ group, then the B-E group would be better for this reason (if ER were a confounding factor). Therefore, we examined whether ER is associated with TFS (and OS) using multivariate Cox regression. We found that ER was not associated with TFS. For reference, TFS group comparisons were performed separately for ER+ and ER-, and it was confirmed that TFS was longer in groups B-E in both subgroups. As we examined the proportional hazards assumption using the two-stage test proposed by Qiu and Sheng (13), the violation of the assumption was not indicated (P =0.337 for OS and 0.766 for TFS) (Figs. S2 and S3; Table SI). Based on the above discussion, we believe that the B-E arm can be recommended regardless of ER status in TFS prolongation.
To our knowledge, this is the first report of the impact of the sequential treatment of HER2-negative metastatic breast cancer with BEV + PTX and eribulin. This study has some limitations. An important limitation is that although it is a multicenter database, the study design is retrospective and observational, and the number of cases is small. Although consistent results were obtained using the PSMA method to reduce selection bias, it is necessary to confirm the findings in a large, prospective study.
In conclusion, despite the retrospective nature of the present analysis and its inherent limitations, the data presented show that when BEV+PTX and eribulin are administered sequentially, the prognosis is better if BEV+PTX is administered first.
Supplementary Material
Kaplan-Meier curves for TFS after PMSA (12 pairs, 24 patients). PMSA, propensity score matching analysis; TFS, time to failure of strategy.
Kaplan-Meier curves for TFS after PMSA (ER+). PMSA, propensity score matching analysis; TFS, time to failure of strategy.
Kaplan-Meier curves for OS after PMSA (ER+). PMSA, propensity score matching analysis; OS, overall survival.
Prognostic analysis of ER-positive patients.
Acknowledgements
The authors would like to thank Dr Takuya Nagata (Department of Surgery, Toho University Ohashi Medical Center), Dr Yasuko Tanada (Department of Surgery, Toyama Nishi General Hospital), Dr Katsuo Shimada (Department of Surgery, Imizu Municipal Hospital), Dr Kaoru Kiyohara (Department of Surgery, Tonami General Hospital), Dr Tetsuro Shimizu (Department of Surgery, Saiseikai Toyama Hospital), and Dr Keiko Iwata (Department of Surgery, Kurobe City Hospital) for providing patient data. This research was presented at The San Antonio Breast Cancer Symposium in 2020.
Funding
Funding: No funding was received.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Authors' contributions
KM and TF designed the study concepts and confirm the authenticity of all the raw data. KM, ME, AY, WF, ZN, KO, KK and KM collected the data. SM and AN performed statistical analysis. All authors read and approved the final manuscript.
Ethics approval and consent to participate
This study was conducted according to the guidelines of The Declaration of Helsinki. However, ethics approval was not applicable because this was a retrospective study that did not include procedures outside of common and correct clinical practice.
Patient consent for publication
Not applicable.
Competing interests
SM received honoraria from Chugai Pharmaceutical Co., Ltd., and Eisai Co., Ltd., and received research funding (institution) from Eisai Co., Ltd., outside the submitted work.
References
|
Cortes J, O'Shaughnessy J, Loesch D, Blum JL, Vahdat LT, Petrakova K, Chollet P, Manikas A, Diéras V, Delozier T, et al: Eribulin monotherapy versus treatment of physician's choice in patients with metastatic breast cancer (EMBRACE): A phase 3 open-label randomised study. Lancet. 377:914–923. 2011.PubMed/NCBI View Article : Google Scholar | |
|
Miller K, Wang M, Gralow J, Dickler M, Cobleigh M, Perez EA, Shenkier T, Cella D and Davidson NE: Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med. 357:2666–2676. 2007.PubMed/NCBI View Article : Google Scholar | |
|
Robert NJ, Diéras V, Glaspy J, Brufsky AM, Bondarenko I, Lipatov ON, Perez EA, Yardley DA, Chan SYT, Zhou X, et al: RIBBON-1: Randomized, double-blind, placebo-controlled, phase III trial of chemotherapy with or without bevacizumab for first-line treatment of human epidermal growth factor receptor 2-negative, locally recurrent or metastatic breast cancer. J Clin Oncol. 29:1252–1260. 2011.PubMed/NCBI View Article : Google Scholar | |
|
Miles DW, Chan A, Dirix LY, Cortés J, Pivot X, Tomczak P, Delozier T, Sohn JH, Provencher L, Puglisi F, et al: Phase III study of bevacizumab plus docetaxel compared with placebo plus docetaxel for the first-line treatment of human epidermal growth factor receptor 2-negative metastatic breast cancer. J Clin Oncol. 28:3239–3247. 2010.PubMed/NCBI View Article : Google Scholar | |
|
Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, Berlin J, Baron A, Griffing S, Holmgren E, et al: Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med. 350:2335–2342. 2004.PubMed/NCBI View Article : Google Scholar | |
|
Sandler A, Grey R, Perry MC, Brahmer J, Schiller JH, Dowlati A, Lilenbaum R and Johnson DH: Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med. 355:2542–2550. 2006.PubMed/NCBI View Article : Google Scholar | |
|
Escudier B, Pluzanska A, Koralewski P, Ravaud A, Bracarda S, Szczylik C, Chevreau C, Filipek M, Melichar B, Bajetta E, et al: AVOREN Trial investigators. Bevacizumab plus interferon alfa-2a for treatment of metastatic renal cell carcinoma: A randomised, double-blind phase III trial. Lancet. 370:2103–2111. 2007.PubMed/NCBI View Article : Google Scholar | |
|
Delaloge S, Pérol D, Courtinard C, Brain E, Asselain B, Bachelot T, Debled M, Dieras V, Campone M, Levy C, et al: Paclitaxel plus bevacizumab or paclitaxel as first-line treatment for HER2-negative metastatic breast cancer in a multicenter national observational study. Ann Oncol. 27:1725–1732. 2016.PubMed/NCBI View Article : Google Scholar | |
|
LeCaer H, Barlesi F, Corre R, Jullian H, Bota S, Falchero L, Vergnenegre A, Dujon C, Delhoume JY and Chouaid C: GFPC 0504 Team. A multicentre phase II randomised trial of weekly docetaxel/gemcitabine followed by erlotinib on progression, vs the reverse sequence, in elderly patients with advanced non small-cell lung cancer selected with a comprehensive geriatric assessment (the GFPC 0504 study). Br J Cancer. 105:1123–1130. 2011.PubMed/NCBI View Article : Google Scholar | |
|
Fiala O, Pesek M, Finek J, Benesova L, Bortlicek Z and Minarik M: Sequential treatment of advanced-stage lung adenocarcinoma harboring wild-type EGFR gene: Second-line pemetrexed followed by third-line erlotinib versus the reverse sequence. Anticancer Res. 33:3397–3402. 2013.PubMed/NCBI | |
|
Rosenbaum PR and Rubin DB: The central role of the propensity score in observational studies for causal effects. Biometrika. 70:41–55. 1983. | |
|
Sturmer T, Wyss R, Glynn RJ and Brookhart MA: Propensity scoresfor confounder adjustment when assessing the effects of medicalinterventions using nonexperimental study designs. J Int Med. 275:570–580. 2014.PubMed/NCBI View Article : Google Scholar | |
|
Peihua Q and Jun S: A two-stage procedure for comparing hazard rate functions. J.R Statist Soc B. 70:191–208. 2008. | |
|
Aogi K, Masuda N, Ohno S, Oda T, Iwata H, Kashiwaba M, Fujiwara Y, Kamigaki S, Ito Y, Ueno T and Takashima S: First-line bevacizumab in combination with weekly paclitaxel for metastatic breast cancer: Efficacy and safety results from a large, open-label, single-arm Japanese study. Breast Cancer Res Treat. 129:829–838. 2011.PubMed/NCBI View Article : Google Scholar | |
|
Kaufman PA, Awada A, Twelves C, Yelle L, Perez EA, Velikova G, Olivo MS, He Y, Dutcus CE and Cortes J: Phase III open-label randomized study of eribulin mesylate versus capecitabine in patients with locally advanced or metastatic breast cancer previously treated with an anthracycline and a taxane. J Clin Oncol. 33:594–601. 2015.PubMed/NCBI View Article : Google Scholar | |
|
Brufsky AM, Hurvitz S, Perez E, Swamy R, Valero V, O'Neill V and Rugo HS: RIBBON-2: A randomized, double-blind, placebo-controlled, phase III trial evaluating the efficacy and safety of bevacizumab in combination with chemotherapy for second-line treatment of human epidermal growth factor receptor 2-negative metastatic breast cancer. J Clin Oncol. 29:4286–4293. 2011.PubMed/NCBI View Article : Google Scholar | |
|
Ueda S, Saeki T, Takeuchi H, Shigekawa T, Yamane T, Kuji I and Osaki A: . In vivo imaging of eribulin-induced reoxygenation in advanced breast cancer patients: A comparison to bevacizumab. Br J Cancer. 114:1212–1218. 2016.PubMed/NCBI View Article : Google Scholar | |
|
Goel S, Mita AC, Mita M, Rowinsky EK, Chu QS, Wong N, Desjardins C, Fang F, Jansen M, Shuster DE, et al: A phase I study of eribulin mesylate (E7389), a mechanistically novel inhibitor of microtubule dynamics, in patients with advanced solid malignancies. Clin Cancer Res. 15:4207–4212. 2009.PubMed/NCBI View Article : Google Scholar | |
|
Tan AR, Rubin EH, Walton DC, Shuster DE, Wong YN, Fang F, Ashworth S and Rosen LS: Phase I study of eribulin mesylate administered once every 21 days in patients with advanced solid tumors. Clin Cancer Res. 15:4213–4219. 2009.PubMed/NCBI View Article : Google Scholar | |
|
Vahdat LT, Pruitt B, Fabian CJ, Rivera RR, Smith DA, Tan-Chiu E, Wright J, Tan AR, Dacosta NA, Chuang E, et al: Phase II study of eribulin mesylate, a halichondrin B analog, in patients with metastatic breast cancer previously treated with an anthracycline and a taxane. J Clin Oncol. 27:2954–2961. 2009.PubMed/NCBI View Article : Google Scholar | |
|
Cortes J, Vahdat L, Blum JL, Twelves C, Campone M, Roché H, Bachelot T, Awada A, Paridaens R, Goncalves A, et al: Phase II study of the halichondrin B analog eribulin mesylate in patients with locally advanced or metastatic breast cancer previously treated with an anthracycline, a taxane, and capecitabine. J Clin Oncol. 28:3922–3928. 2010.PubMed/NCBI View Article : Google Scholar | |
|
Morgan RJ, Synold TW, Longmate JA, Quinn DI, Gandara D, Lenz HJ, Ruel C, Xi B, Lewis MD, Colevas AD, et al: Pharmacodynamics (PD) and pharmacokinetics (PK) of E7389 (eribulin, halichondrin B analog) during a phase I trial in patients with advanced solid tumors: A California cancer consortium trial. Cancer Chemother Pharmacol. 76:897–907. 2015.PubMed/NCBI View Article : Google Scholar | |
|
Gligorov J, Doval D, Bines J, Alba E, Cortes P, Pierga JY, Gupta V, Costa R, Srock S, de Ducla S, et al: Maintenance capecitabine and bevacizumab versus bevacizumab alone after initial first-line bevacizumab and docetaxel for patients with HER2-negative metastatic breast cancer (IMELDA): A randomised, open-label, phase 3 trial. Lancet Oncol. 15:1351–1360. 2014.PubMed/NCBI View Article : Google Scholar | |
|
Park YH, Jung KH, Im SA, Sohn JH, Ro J, Ahn JH, Kim SB, Nam BH, Oh DY, Han SW, et al: Phase III, multicenter, randomized trial of maintenance chemotherapy versus observation in patients with metastatic breast cancer after achieving disease control with six cycles of gemcitabine plus paclitaxel as first-line chemotherapy: KCSG-BR07-02. J Clin Oncol. 31:1732–1739. 2013.PubMed/NCBI View Article : Google Scholar |
