- Email: [email protected]
Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis
European Journal of Cancer (2015) xxx, xxx– xxx
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis Norihiro Yamaguchi a,1, Takeo Fujii b,c,1, Shunsuke Aoi d, Peter S. Kozuch a,⇑, Gabriel N. Hortobagyi e, Ronald H. Blum a a
Department of Medicine, Division of Hematology/Oncology, Mount Sinai Beth Israel, 16th Street and 1st Avenue, New York, NY 10003, USA Department of Breast Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA c School of Public Health, The University of Texas Health Science Center at Houston, 1200 Herman-Pressler Street, Houston, TX 77030, USA d Department of Medicine, Mount Sinai Beth Israel, 16th Street and 1st Avenue, New York, NY 10003, USA e Department of Breast Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, 1155 Pressler, Suite CPB5.3466, Houston, TX 77030, USA b
Received 10 April 2015; received in revised form 26 May 2015; accepted 22 July 2015
KEYWORDS Network meta-analysis Breast cancer Anthracycline Cardiotoxicity Cardiac events Cardiomyopathy
Abstract Background: Anthracyclines play a broad and important role in the care of patients with either operable or metastatic breast cancer. However cardiotoxicity narrows the therapeutic index of this drug class leading to potentially clinically meaningful treatment delays or discontinuations. We conducted a Bayesian network meta-analysis, a validated statistical methodology, allowing direct and indirect comparison of cardiotoxicity of different anthracycline and non-anthracycline regimens. Methods: We conducted a systematic review of prospective randomised controlled trials through MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials and Google Scholar comparing non-anthracycline based regimens (NON), doxorubicin (DOX), epirubicin (EPI) and liposomal doxorubicin (LD). We included studies published up to 1st January 2014 in both adjuvant and metastatic contexts. Notably, HER2/neu-targeted regimens were excluded. We assessed the studies’ eligibility criteria and data collection with consensus of two independent authors. Our primary outcome measure was cardiac events grade 3 or greater (CE3) in accordance with Common Terminology Criteria for Adverse Events (CTCAE) Version 4.0. A Bayesian pairwise and network meta-analysis was conducted to estimate pooled Odds Ratio (OR).
⇑ Corresponding author at: Department of Medicine, Division of Hematology/Oncology, Mount Sinai Beth Israel, 16th Street and 1st Avenue, New York, NY 10003, USA. Tel.: +1 212 844 8070. E-mail address: [email protected] (P.S. Kozuch). 1 Those authors equally contributed to this work.
http://dx.doi.org/10.1016/j.ejca.2015.07.031 0959-8049/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
2
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
Findings: Nineteen randomised controlled trials met eligibility criteria and were included in this analysis. We found a trend showing that LD is less cardiotoxic than DOX with an OR of 0.60 (95% confidence interval (CI) 0.34–1.07) There was no difference between Epi and LD with an OR of 0.95 (95%CI 0.39–2.33). DOX is more cardiotoxic than Non with an OR of 1.57 (95%CI 0.90–2.72). Interpretation: DOX has higher CE3 rates than NON does. LD statistically trended to lower cardiac event rates than DOX. Non-statistical significance among EPI, LD and DOX with regard to cardiac toxicity indicates that avoidance of CE3 should not motivate selection of a particular anthracycline in otherwise healthy women in whom total lifetime anthracycline exposure will likely be limited. Overall low incidence of CE3 with any type of anthracycline indicates that we can safely use any anthracycline if cumulative dose limits are not exceeded. While CE3 does not limit our choice of anthracycline LD appears to be the least cardiotoxic. Funding: Takeo Fujii is supported by the grant named Young Investigator Award for Study Abroad in Clinical Epidemiology from St. Luke’s Life Science Institution. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Breast cancer is the most commonly diagnosed cancer among women worldwide and anthracyclines are key therapeutic agents associated with improved survival [1,2]. Despite being a significant part of standard treatments in adjuvant, neoadjuvant and metastatic contexts anthracycline associated cardiac toxicity is a significant concern and may offset anti-neoplastic benefit [3]. Compared with non-anthracycline regimens, anthracyclines have been shown to cause five times more compromised left ventricle ejection fraction (LVEF) and chronic heart failure. The risk of cardiotoxicity correlates with lifetime cumulative anthracycline exposure [4,5]. Cumulative ceiling doses were defined since CE3 rates dramatically increase when doxorubicin is given at 400–450 mg/m2 and higher and epirubicin at 900–1000 mg/m2 and higher [6,7]. Molecular pathogenesis of anthracycline-induced cardiotoxicity is yet unclear, though, the most promising hypothesis is the interaction with topoisomerase II (Top2) [8]. Top2 has two isozymes which are Top2alpha and Top2beta [9]. Among the isozymes, anthracycline and Top2alpha complex lead to cell death. Mammalian cardiomyocyte only has Top2beta and successful prevention of cardiac toxicity with cardiomyocyte specific deletion of Top2beta in a mice study identified Top2beta/anthracycline complex role in the development of cardiotoxicity [10]. The search for anthracycline analogues with less cardiotoxicity led to the development of agents such as epirubicin and liposomal doxorubicin [11]. A comparative study has shown that epirubicin appears to be equally effective and less cardiotoxic compared to doxorubicin for metastatic breast cancer [12]. Liposomal doxorubicin, developed by encapsulating nanotechnology which allows targeting tumour cells without excess exposure of normal cells, has also been shown to have comparable efficacy with significantly reduced cardiac toxicity [13,14].
There are numerous trials comparing the cardiac toxicity of chemotherapeutic agents, however, there is no level 1 evidence showing their relative cardiac toxicity with quantitative analysis. Network meta-analysis (NMA) is a useful methodology that offers direct and indirect comparisons in order to estimate the relative efficacy between all interventions even though some have not been compared head to head [15–17]. To investigate the cardiac toxicity among anthracyclines, we conducted a Bayesian NMA. 2. Methods 2.1. Study search and selection criteria We conducted a systematic review of prospective randomised controlled trials through MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials and Google Scholar comparing non-anthracycline based regimens, doxorubicin, epirubicin and liposomal doxorubicin according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [18]. We included randomised controlled trials evaluating either regimens in the metastatic or adjuvant setting published up to 1st January 2014. Trials of HER2/neu-inhibitor containing regimens were excluded as were studies comparing various doses of the same cytotoxic agents and trials which didn’t provide cardiac adverse event data. 2.2. Study selection and data collection Two independent investigators (NY and TF) selected the articles and extracted information. Collected data presented in Table 1 included the total number of patients in each trial, the number of CEs, treatment setting (metastatic or adjuvant), the name of regimens, median patient age, history of prior anthracycline treatment, cumulative dose of anthracycline and observation
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
ID
Author
Published year
Setting
Regimen* (N)
Age median
Prior anthracycline based chemotherapy
Cumulative dose of anthracycline
Observation period
1
Bachelot
2011
Metastatic
Batist
2001
Metastatic
3
Botenbal
1998
Metastatic
4
Chan
1999
Metastatic
5
Chan
2004
Metastatic
6
Feher
2005
Metastatic
7
Fisher
1990
Locally advanced
8
Harris
2002
Metastatic
9
Hochster
2001
10
Hutchins
2005
Locally Advanced Local
11
Lopez
1989
Metastatic
12
Martin
2003
Local
12
O’Brien
2004
Metastatic
14
Perez
1991
Metastatic
15
Piccart
2001
16
Sledge
2003
Locally Advanced Metastatic
17
Sparano
2009
Metastatic
18
Vici
2011
Metastatic
19
Wils
1999
Locally Advanced
57 59 55 54 56 56 52 52 54 54 69 68 NE NE 58 58 56 59.5 48 48 52 55 N/A N/A 59 58 57 53 49 49 58 56 51.8 52.5 63 61 59 60
No No Yes Yes Yes Yes Yes Yes N/A N/A No No No No Yes Yes Yes No No No Yes Yes No No Yes Yes No No No No Yes Yes Yes Yes Yes Yes No No
N/A 600 mg/m2(median) N/A N/A 568 mg/m2(median) 800 mg/m2(median) N/A 450 mg/m2(median) N/A N/A N/A N/A N/A 240 mg/m2(median) 785 mg/m2(median) 570 mg/m2(median) N/A N/A N/A N/A N/A N/A N/A 264 mg/m2(median) N/A N/A N/A N/A N/A 640 mg/m2(median) N/A N/A N/A 403 mg/m2(median) 540 mg/m2(median) 240 mg/m2(median) 280 mg/m2(median) N/A
54 month
2
XT(33) ET(34) LDC(142) AC(145) A(118) E(114) T(161) A(165) LD(80) E(80) GEM(198) E(199) CMF(776) AC(781) LD(108) A(116) FV(56) AV(62) CMF(676) CAF(669) FAC(46) FEC(48) CMF(505) FAC(480) LD(254) A(255) E(72) A(69) CMF(255) EC(267) A(224) PTX(229) T(373) LD + T(378) EV(54) LDV(50) E + TAM(303) TAM(301)
36 month 60 month 36 month 24 month 24 month 36 month 24 month 60 month 120 month 30 month 160 month 36 month 50 month 80 month 80 month 48 month
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
72 month 96 month
PTX, Paclitaxel; T, Docetaxel; TAM, Tamoxifen; V, Vinorelbine; X, Capecitabine; N/A, Not available. * A, Doxorubicin; C, Cyclophosphamide; E, Epirubicin; F, 5-FU; GEM, Gemcitabine; LD, Liposomal doxorubicin; M, Methotrexate.
3
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
Table 1 Characteristics of included articles. The characteristics of 19 included articles are shown below. 14 articles were in the metastatic setting and 5 of them were in locally advanced setting.
4
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
period. We used the Cochrane risk of bias method to assess the risk of bias of each trial. Our outcome measure was CE3 as defined by Common Terminology Criteria for Adverse Events (CTCAE) Version 4.0. 2.3. Statistical analyses A Bayesian network meta-analysis was fit to analyse pooled Odds Ratio (OR) from direct and indirect trials [19]. We used a random-effect model and specified the OR association across trials. The natural logarithm of HR and standard error of it were calculated and used in the analysis. Predictive interval plot was made to compare pooled ORs of each treatment regimen. We also drew cumulative ranking plot with adjustment for small study effects model and then calculated the surface under the cumulative ranking curve (SUCRA) to compare the effects on CEs between each regimen. All data were analysed by using STATA version 13. (STATA Corp, College Station, TX). 2.4. Role of the funding source Takeo Fujii is supported by the grant named Young investigator award for study abroad in clinical epidemiology from St. Luke’s Life Science Institution. 3. Results We identified 966 unduplicated studies. Initial review of titles and abstracts identified 25 potentially eligible trials, most screened studies were excluded because they were not randomised controlled trials. Among the twenty five trials identified by initial screening, six did not meet our inclusion criteria (Fig. 1) and as a result nineteen studies were included in our analysis (Table 1) [20–31,13,32–37]. Fourteen trials were conducted in the metastatic setting and five trials were conducted in the adjuvant setting. Fig. 2 is a network plot of comparisons made between doxorubicin (DOX), liposomal
Fig. 1. Literature search and selection. We identified 966 articles for review and selected in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.
doxorubicin (LD), epirubicin (Epi) and nonanthracycline based regimens (Non). The size of each circle represents the number of patients included in the arm. The width of connecting bars reflects the number of comparisons made between the circles. DOX arm has the largest N = 3823 and the comparison between DOX and NON was most frequently made (=6 times). The pooled odds ratios are shown in predictive interval plot (Fig. 3). In that plot, circled chemotherapeutic agents were considered as a reference and comparisons were made in the order listed in the figure: Non versus DOX, Non versus Epi, Non versus LD, DOX versus Epi, DOX versus LD and EPI versus LD. It showed that LD has less CE3s than DOX (OR 0.60, 95% confidence interval (CI); 0.34–1.07) or Epi (OR 0.95, 95%CI; 0.39–2.33) but those differences were not statistically significant. In addition, there was a trend that DOX was more toxic than Non. The OR of DOX versus Non was 157 (95%CI; 0.90–2.72). We conducted cumulative ranking plot analysis (Fig. 4) to display hierarchy of these agents in terms of their relative cardiac toxicity. Surface under the cumulative ranking (SUCRA) showed the hierarchy. The larger an area under the curve becomes, the greater CE rates the agent has. The SUCRA revealed that DOX was associated with the highest rates of CE3 and LD with the least. 4. Discussion This is the first network meta-analysis comparing cardiac toxicity of different anthracyclines. Our study showed that doxorubicin has more cardiac toxicity than non-anthracycline based regimens. Also, liposomal doxorubicin trended towards less cardiac toxicity than doxorubicin.
Fig. 2. Network plot of included studies. The circles represent pooled N of included studies and the width of bridging bars reflects the number of both direct and indirect comparisons.
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
Fig. 3. Predictive interval plot for the network of LD, Epi, Dox and Non LD trends less cardiotoxic than Dox. Dox trends more cardiotoxic than Non. No difference between Epi, LD and Non.
NMA is a relatively new statistical method to compare existing data sets. The analysis allows both indirect and direct comparison by bridging multiple competitors with a shared competitor. For instance, if there are reports comparing treatment A versus B and A versus C and we do not have a head to head comparison between B and C, then NMA can provide the relative efficacy of B and C by establishing A as a reference [38,39]. This new method will give us an opportunity to reassess the existing data with a mock head to head comparison, especially in situations where a head to head comparison is difficult. For example, NMA allows us to compare a current standard regimen and a former standard regimen. A previous meta-analysis by Smith et al. [4] revealed that anthracycline-based regimens have significantly
5
higher cardiac adverse event rates than nonanthracycline regimens (OR 5.43, 95%CI: 2.34–12.62). In the same study liposomal doxorubicin was associated with significantly less CEs compared with non-liposomal doxorubicin (OR 0.18, 95% CI: 0.13–0.33). Notably that study included grade 1 and 2 cardiac adverse events. By contrast, when only CE3s are assessed, as in our study, the small number of events, while more common with anthracycline-based regimens do not reach statistical significance. These findings indicate that clinically significant cardiac toxicity can be avoided by limiting cumulative lifetime anthracycline exposure and therefore does not need to be factored into treatment selection. Our study has a couple of limitations. First, our study did not consider reasons other than cardiac events for stopping chemotherapy. If chemotherapy was postponed or discontinued early due to reasons other than CEs, the incidence of CEs would not be observed due to the right censoring. This fact may have contributed to the non-statistical significance of our results. Second, the dose of chemotherapeutic agents we included was heterogeneous. It is difficult to assess per unit cardiac toxicity of these agents since these agents have a unique therapeutic concentration. Yet studies we included used a standard dosage of each agent. EPI was mostly given at the median cumulative dosage from 500 to 800 mg/m2. The median cumulative dosage of DOX ranged from 264 to 570 mg/m2 (Table 1) We think the contribution of this heterogeneity is clinically minimal. Third, our study did not take molecular predictive factors of anthracycline-induced cardiotoxicity such as L461L within the SLC28A3 gene [40] into account. However this SNP theoretically should distribute equally to each arm during randomisation within each trial and should not alter our conclusion.
Fig. 4. Cumulative ranking plot for the network. 1: Most 4: Least cardiotoxic SUCRA: The surface under the cumulative ranking. Solid line corresponds to the unadjusted model. Red dashed lines are adjusted for small study effects. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
6
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
Finally, we could not perform sensitivity analysis with NMA meaning that we could not statistically evaluate the contribution of risk factors which might affect the incidence of CEs. We performed a descriptive review of these risk factors. In terms of observation time, we have two outliers whose observation times were 120 and 160 months. The reported rate of CE3 was 0.8% and 0.6% which was not higher than others. We did not find heterogeneity in terms of patients’ age among the included studies. Regarding the setting of treatment, CE3 rate for local or locally advanced disease ranged from 0.7% to 6.5%, in turn, those for metastatic disease ranged from 0.8% to 8.5% if we exclude one outlier. That study [21] reported a CE rate in the DOX arm of 23%. Notably 34% of patients in this arm had at least one out of four cardiac risk factors (age >65 years old, prior DOX >240 mg/m2, radiation to chest area and preexisting cardiac disease). The patients included in this study likely had more underlying cardiac risk factors than subjects in other trials although most other studies did not report cardiac risk factors. In terms of prior exposure to anthracyclines, CE3 rate for pre-treated groups ranged from 0.8% to 7.6% and that for anthracycline naive group ranged from 0.8% to 8.5%, again excluding the same outlier which showed a 23% CE rate. In conclusion, our systematic review and NMA revealed that DOX appears to be the most cardiotoxic anthracycline while LD is the least cardiotoxic member of the group although we could not demonstrate this to a degree of statistical significance due to the small number of CE3s. The low CE3 incidence was likely due to adherence of established lifetime anthracycline exposure limits. Among the three anthracyclines, LD is relatively safe in terms of cardiac events. Overall, we can say that potential cardiac toxicity need not be a major factor in selecting a particular anthracycline based regimen for the treatment of breast cancer patients without underlying heart disease. Contributors NY, TF and RB contributed to study concept and design. NY, TF, SA contributed to collection and assembly of data. NY, TF, SA, PK, GH and RB contributed to data analysis and interpretation. NY, TF and SA wrote the manuscript. All the authors approved the final version of the manuscript. Conflict of interest statement None declared. References [1] Kalam K, Marwick TH. Role of cardioprotective therapy for prevention of cardiotoxicity with chemotherapy: a systematic review and meta-analysis. Eur J Cancer 2013;49(13):2900–9.
[2] Sant M, Allemani C, Santaquilani M, Knijn A, Marchesi F, Capocaccia R, et al. EUROCARE-4. Survival of cancer patients diagnosed in 1995–1999. Results and commentary. Eur J Cancer 2009;45(6):931–91. [3] Pai VB, Nahata MC. Cardiotoxicity of chemotherapeutic agents: incidence, treatment and prevention. Drug Saf 2000;22(4): 263–302. [4] Smith LA, Cornelius VR, Plummer CJ, et al. Cardiotoxicity of anthracycline agents for the treatment of cancer: systematic review and meta-analysis of randomised controlled trials. BMC Cancer 2010;10:337. [5] Swain SM, Whaley FS, Ewer MS. Congestive heart failure in patients treated with doxorubicin: a retrospective analysis of three trials. Cancer 2003;97:2869–79. [6] Von Hoff DD, Layard MW, Basa P, et al. Risk factors for doxorubicin-induced congestive heart failure. Ann Intern Med 1979;91:710–7. [7] Ryberg M, Nielsen D, Skovsgaard T, et al. Epirubicin cardiotoxicity: an analysis of 469 patients with metastatic breast cancer. J Clin Oncol 1998;16:3502–8. [8] Lyu YL, Kerrigan JE, Lin CP, et al. Topoisomerase IIbeta mediated DNA double-strand breaks: implications in doxorubicin cardiotoxicity and prevention by dexrazoxane. Cancer Res 2007;67(18):8839–46. [9] Capranico G, Tinelli S, Austin CA, et al. Different patterns of gene expression of topoisomerase II isoforms in differentiated tissues during murine development. Biochim Biophys Acta 1992;1132(1):43–8. [10] Zhang S, Liu X, Bawa-Khalfe T, et al. Identification of the molecular basis of doxorubicin-induced cardiotoxicity. Nat Med 2012;18(11):1639–42 [Epub 2012 Oct 28]. [11] Weiss RB. The anthracyclines: will we ever find a better doxorubicin? Semin Oncol 1992;19:670–86. [12] Hortobagyi GN, Yap HY, Kau SW, et al. A comparative study of doxorubicin and epirubicin in patients with metastatic breast cancer. Am J Clin Oncol 1989;12:57–62. [13] O’Brien ME, Wigler N, Inbar M, et al. Reduced cardiotoxicity and comparable efficacy in a phase III trial of pegylated liposomal doxorubicin HCl (CAELYX/Doxil) versus conventional doxorubicin for first-line treatment of metastatic breast cancer. Ann Oncol 2004;15:440–9. [14] Rahman AM, Yusuf SW, Ewer MS. Anthracycline-induced cardiotoxicity and the cardiac-sparing effect of liposomal formulation. Int J Nanomedicine 2007;2(4):567–83. [15] Mills EJ, Thorlund K, Ioannidis JP. Demystifying trial networks and network meta-analysis. BMJ 2013;14(346):f2914. [16] Jansen JP, Fleurence R, Devine B, Itzler R, Barrett A, Hawkins N, et al. Interpreting indirect treatment comparisons and network meta-analysis for health-care decision making: report of the ISPORTask Force on Indirect Treatment Comparisons Good Research Practices: part 1. Value Health 2011;14(4):417–28. [17] Hoaglin DC, Hawkins N, Jansen JP, Scott DA, Itzler R, Cappelleri JC, et al. Conducting indirect-treatment-comparison and network-meta-analysis studies: report of the ISPOR Task Force on Indirect Treatment Comparisons Good Research Practices: part2. Value Health 2011;14(4):429–37. [18] Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement. Int J Surg 2010;8:336–41. [19] Chaimani A, Higgins JP, Mavridis D, Spyridonos P, Salanti G. Graphical tools for network meta-analysis in STATA. PLoS One 2013;8:e76654. [20] Bachelot T, Bajard A, Ray-Coquard I, et al. Final results of ERASME-4: a randomized trial of first-line docetaxel plus either capecitabine or epirubicin for metastatic breast cancer. Oncology 2011;80:262–8. [21] Batist G, Ramakrishnan G, Rao CS, et al. Reduced cardiotoxicity and preserved antitumor efficacy of liposome-encapsulated dox-
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
orubicin and cyclophosphamide compared with conventional doxorubicin and cyclophosphamide in a randomized, multicenter trial of metastatic breast cancer. J Clin Oncol 2001;19:1444–54. Bontenbal M, Andersson M, Wildiers J, et al. Doxorubicin vs epirubicin, report of a second-line randomized phase II/III study in advanced breast cancer. EORTC Breast Cancer Cooperative Group. Br J Cancer 1998;77:2257–63. Chan S, Davidson N, Juozaityte E, et al. Phase III trial of liposomal doxorubicin and cyclophosphamide compared with epirubicin and cyclophosphamide as first-line therapy for metastatic breast cancer. Ann Oncol 2004;15:1527–34. Chan S, Friedrichs K, Noel D, et al. Prospective randomized trial of docetaxel versus doxorubicin in patients with metastatic breast cancer. J Clin Oncol 1999;17:2341–54. Feher O, Vodvarka P, Jassem J, et al. First-line gemcitabine versus epirubicin in postmenopausal women aged 60 or older with metastatic breast cancer: a multicenter, randomized, phase III study. Ann Oncol 2005;16:899–908. Fisher B, Brown AM, Dimitrov NV, et al. Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifen-nonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project B-15. J Clin Oncol 1990;8:1483–96. Harris L, Batist G, Belt R, et al. Liposome-encapsulated doxorubicin compared with conventional doxorubicin in a randomized multicenter trial as first-line therapy of metastatic breast carcinoma. Cancer 2002;94:25–36. Hochster HS, Vogel CL, Burman SL, White R. Activity and safety of vinorelbine combined with doxorubicin or fluorouracil as first-line therapy in advanced breast cancer: a stratified phase II study. Oncologist 2001;6:269–77. Hutchins LF, Green SJ, Ravdin PM, et al. Randomized, controlled trial of cyclophosphamide, methotrexate, and fluorouracil versus cyclophosphamide, doxorubicin, and fluorouracil with and without tamoxifen for high-risk, node-negative breast cancer: treatment results of Intergroup Protocol INT-0102. J Clin Oncol 2005;23:8313–21. Lopez M, Papaldo P, Di Lauro L, Vici P, Carpano S, Conti EM. 5-Fluorouracil, adriamycin, cyclophosphamide (FAC) vs. 5-fluorouracil, epirubicin, cyclophosphamide (FEC) in metastatic breast cancer. Oncology 1989;46:1–5.
7
[31] Martin M, Villar A, Sole-Calvo A, et al. Doxorubicin in combination with fluorouracil and cyclophosphamide (i.v. FAC regimen, day 1, 21) versus methotrexate in combination with fluorouracil and cyclophosphamide (i.v. CMF regimen, day 1, 21) as adjuvant chemotherapy for operable breast cancer: a study by the GEICAM group. Ann Oncol 2003;14:833–42. [32] Perez DJ, Harvey VJ, Robinson BA, et al. A randomized comparison of single-agent doxorubicin and epirubicin as firstline cytotoxic therapy in advanced breast cancer. J Clin Oncol 1991;9:2148–52. [33] Piccart MJ, Di Leo A, Beauduin M, et al. Phase III trial comparing two dose levels of epirubicin combined with cyclophosphamide with cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer. J Clin Oncol 2001;19:3103–10. [34] Sledge GW, Neuberg D, Bernardo P, et al. Phase III trial of doxorubicin, paclitaxel, and the combination of doxorubicin and paclitaxel as front-line chemotherapy for metastatic breast cancer: an intergroup trial (E1193). J Clin Oncol 2003;21:588–92. [35] Sparano JA, Makhson AN, Semiglazov VF, et al. Pegylated liposomal doxorubicin plus docetaxel significantly improves time to progression without additive cardiotoxicity compared with docetaxel monotherapy in patients with advanced breast cancer previously treated with neoadjuvant-adjuvant anthracycline therapy: results from a randomized phase III study. J Clin Oncol 2009;27:4522–9. [36] Vici P, Colucci G, Giotta F, et al. A multicenter prospective phase II randomized trial of epirubicin/vinorelbine versus pegylated liposomal doxorubicin/vinorelbine as first-line treatment in advanced breast cancer. A GOIM study. J Exp Clin Cancer Res 2011;30:39. [37] Wils JA, Bliss JM, Marty M, et al. Epirubicin plus tamoxifen versus tamoxifen alone in node-positive postmenopausal patients with breast cancer: a randomized trial of the International Collaborative Cancer Group. J Clin Oncol 1999;17:1988–98. [38] Lu G, Ades AE. Combination of direct and indirect evidence in mixed treatment comparisons. Stat Med 2004;23(20):3105–24. [39] Caldwell DM, Ades AE, Higgins JP. Simultaneous comparison of multiple treatments: combining direct and indirect evidence. BMJ 2005;331(7521):897–900. [40] Visscher H, Ross CJ, Rassekh SR, et al. Pharmacogenomic prediction of anthracycline-induced cardiotoxicity in children. J Clin Oncol 2012;30(13):1422–8. http://dx.doi.org/10.1200/ JCO.2010.34.3467 [Epub 2011 Sep 6].
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis Norihiro Yamaguchi a,1, Takeo Fujii b,c,1, Shunsuke Aoi d, Peter S. Kozuch a,⇑, Gabriel N. Hortobagyi e, Ronald H. Blum a a
Department of Medicine, Division of Hematology/Oncology, Mount Sinai Beth Israel, 16th Street and 1st Avenue, New York, NY 10003, USA Department of Breast Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA c School of Public Health, The University of Texas Health Science Center at Houston, 1200 Herman-Pressler Street, Houston, TX 77030, USA d Department of Medicine, Mount Sinai Beth Israel, 16th Street and 1st Avenue, New York, NY 10003, USA e Department of Breast Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, 1155 Pressler, Suite CPB5.3466, Houston, TX 77030, USA b
Received 10 April 2015; received in revised form 26 May 2015; accepted 22 July 2015
KEYWORDS Network meta-analysis Breast cancer Anthracycline Cardiotoxicity Cardiac events Cardiomyopathy
Abstract Background: Anthracyclines play a broad and important role in the care of patients with either operable or metastatic breast cancer. However cardiotoxicity narrows the therapeutic index of this drug class leading to potentially clinically meaningful treatment delays or discontinuations. We conducted a Bayesian network meta-analysis, a validated statistical methodology, allowing direct and indirect comparison of cardiotoxicity of different anthracycline and non-anthracycline regimens. Methods: We conducted a systematic review of prospective randomised controlled trials through MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials and Google Scholar comparing non-anthracycline based regimens (NON), doxorubicin (DOX), epirubicin (EPI) and liposomal doxorubicin (LD). We included studies published up to 1st January 2014 in both adjuvant and metastatic contexts. Notably, HER2/neu-targeted regimens were excluded. We assessed the studies’ eligibility criteria and data collection with consensus of two independent authors. Our primary outcome measure was cardiac events grade 3 or greater (CE3) in accordance with Common Terminology Criteria for Adverse Events (CTCAE) Version 4.0. A Bayesian pairwise and network meta-analysis was conducted to estimate pooled Odds Ratio (OR).
⇑ Corresponding author at: Department of Medicine, Division of Hematology/Oncology, Mount Sinai Beth Israel, 16th Street and 1st Avenue, New York, NY 10003, USA. Tel.: +1 212 844 8070. E-mail address: [email protected] (P.S. Kozuch). 1 Those authors equally contributed to this work.
http://dx.doi.org/10.1016/j.ejca.2015.07.031 0959-8049/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
2
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
Findings: Nineteen randomised controlled trials met eligibility criteria and were included in this analysis. We found a trend showing that LD is less cardiotoxic than DOX with an OR of 0.60 (95% confidence interval (CI) 0.34–1.07) There was no difference between Epi and LD with an OR of 0.95 (95%CI 0.39–2.33). DOX is more cardiotoxic than Non with an OR of 1.57 (95%CI 0.90–2.72). Interpretation: DOX has higher CE3 rates than NON does. LD statistically trended to lower cardiac event rates than DOX. Non-statistical significance among EPI, LD and DOX with regard to cardiac toxicity indicates that avoidance of CE3 should not motivate selection of a particular anthracycline in otherwise healthy women in whom total lifetime anthracycline exposure will likely be limited. Overall low incidence of CE3 with any type of anthracycline indicates that we can safely use any anthracycline if cumulative dose limits are not exceeded. While CE3 does not limit our choice of anthracycline LD appears to be the least cardiotoxic. Funding: Takeo Fujii is supported by the grant named Young Investigator Award for Study Abroad in Clinical Epidemiology from St. Luke’s Life Science Institution. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Breast cancer is the most commonly diagnosed cancer among women worldwide and anthracyclines are key therapeutic agents associated with improved survival [1,2]. Despite being a significant part of standard treatments in adjuvant, neoadjuvant and metastatic contexts anthracycline associated cardiac toxicity is a significant concern and may offset anti-neoplastic benefit [3]. Compared with non-anthracycline regimens, anthracyclines have been shown to cause five times more compromised left ventricle ejection fraction (LVEF) and chronic heart failure. The risk of cardiotoxicity correlates with lifetime cumulative anthracycline exposure [4,5]. Cumulative ceiling doses were defined since CE3 rates dramatically increase when doxorubicin is given at 400–450 mg/m2 and higher and epirubicin at 900–1000 mg/m2 and higher [6,7]. Molecular pathogenesis of anthracycline-induced cardiotoxicity is yet unclear, though, the most promising hypothesis is the interaction with topoisomerase II (Top2) [8]. Top2 has two isozymes which are Top2alpha and Top2beta [9]. Among the isozymes, anthracycline and Top2alpha complex lead to cell death. Mammalian cardiomyocyte only has Top2beta and successful prevention of cardiac toxicity with cardiomyocyte specific deletion of Top2beta in a mice study identified Top2beta/anthracycline complex role in the development of cardiotoxicity [10]. The search for anthracycline analogues with less cardiotoxicity led to the development of agents such as epirubicin and liposomal doxorubicin [11]. A comparative study has shown that epirubicin appears to be equally effective and less cardiotoxic compared to doxorubicin for metastatic breast cancer [12]. Liposomal doxorubicin, developed by encapsulating nanotechnology which allows targeting tumour cells without excess exposure of normal cells, has also been shown to have comparable efficacy with significantly reduced cardiac toxicity [13,14].
There are numerous trials comparing the cardiac toxicity of chemotherapeutic agents, however, there is no level 1 evidence showing their relative cardiac toxicity with quantitative analysis. Network meta-analysis (NMA) is a useful methodology that offers direct and indirect comparisons in order to estimate the relative efficacy between all interventions even though some have not been compared head to head [15–17]. To investigate the cardiac toxicity among anthracyclines, we conducted a Bayesian NMA. 2. Methods 2.1. Study search and selection criteria We conducted a systematic review of prospective randomised controlled trials through MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials and Google Scholar comparing non-anthracycline based regimens, doxorubicin, epirubicin and liposomal doxorubicin according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [18]. We included randomised controlled trials evaluating either regimens in the metastatic or adjuvant setting published up to 1st January 2014. Trials of HER2/neu-inhibitor containing regimens were excluded as were studies comparing various doses of the same cytotoxic agents and trials which didn’t provide cardiac adverse event data. 2.2. Study selection and data collection Two independent investigators (NY and TF) selected the articles and extracted information. Collected data presented in Table 1 included the total number of patients in each trial, the number of CEs, treatment setting (metastatic or adjuvant), the name of regimens, median patient age, history of prior anthracycline treatment, cumulative dose of anthracycline and observation
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
ID
Author
Published year
Setting
Regimen* (N)
Age median
Prior anthracycline based chemotherapy
Cumulative dose of anthracycline
Observation period
1
Bachelot
2011
Metastatic
Batist
2001
Metastatic
3
Botenbal
1998
Metastatic
4
Chan
1999
Metastatic
5
Chan
2004
Metastatic
6
Feher
2005
Metastatic
7
Fisher
1990
Locally advanced
8
Harris
2002
Metastatic
9
Hochster
2001
10
Hutchins
2005
Locally Advanced Local
11
Lopez
1989
Metastatic
12
Martin
2003
Local
12
O’Brien
2004
Metastatic
14
Perez
1991
Metastatic
15
Piccart
2001
16
Sledge
2003
Locally Advanced Metastatic
17
Sparano
2009
Metastatic
18
Vici
2011
Metastatic
19
Wils
1999
Locally Advanced
57 59 55 54 56 56 52 52 54 54 69 68 NE NE 58 58 56 59.5 48 48 52 55 N/A N/A 59 58 57 53 49 49 58 56 51.8 52.5 63 61 59 60
No No Yes Yes Yes Yes Yes Yes N/A N/A No No No No Yes Yes Yes No No No Yes Yes No No Yes Yes No No No No Yes Yes Yes Yes Yes Yes No No
N/A 600 mg/m2(median) N/A N/A 568 mg/m2(median) 800 mg/m2(median) N/A 450 mg/m2(median) N/A N/A N/A N/A N/A 240 mg/m2(median) 785 mg/m2(median) 570 mg/m2(median) N/A N/A N/A N/A N/A N/A N/A 264 mg/m2(median) N/A N/A N/A N/A N/A 640 mg/m2(median) N/A N/A N/A 403 mg/m2(median) 540 mg/m2(median) 240 mg/m2(median) 280 mg/m2(median) N/A
54 month
2
XT(33) ET(34) LDC(142) AC(145) A(118) E(114) T(161) A(165) LD(80) E(80) GEM(198) E(199) CMF(776) AC(781) LD(108) A(116) FV(56) AV(62) CMF(676) CAF(669) FAC(46) FEC(48) CMF(505) FAC(480) LD(254) A(255) E(72) A(69) CMF(255) EC(267) A(224) PTX(229) T(373) LD + T(378) EV(54) LDV(50) E + TAM(303) TAM(301)
36 month 60 month 36 month 24 month 24 month 36 month 24 month 60 month 120 month 30 month 160 month 36 month 50 month 80 month 80 month 48 month
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
72 month 96 month
PTX, Paclitaxel; T, Docetaxel; TAM, Tamoxifen; V, Vinorelbine; X, Capecitabine; N/A, Not available. * A, Doxorubicin; C, Cyclophosphamide; E, Epirubicin; F, 5-FU; GEM, Gemcitabine; LD, Liposomal doxorubicin; M, Methotrexate.
3
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
Table 1 Characteristics of included articles. The characteristics of 19 included articles are shown below. 14 articles were in the metastatic setting and 5 of them were in locally advanced setting.
4
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
period. We used the Cochrane risk of bias method to assess the risk of bias of each trial. Our outcome measure was CE3 as defined by Common Terminology Criteria for Adverse Events (CTCAE) Version 4.0. 2.3. Statistical analyses A Bayesian network meta-analysis was fit to analyse pooled Odds Ratio (OR) from direct and indirect trials [19]. We used a random-effect model and specified the OR association across trials. The natural logarithm of HR and standard error of it were calculated and used in the analysis. Predictive interval plot was made to compare pooled ORs of each treatment regimen. We also drew cumulative ranking plot with adjustment for small study effects model and then calculated the surface under the cumulative ranking curve (SUCRA) to compare the effects on CEs between each regimen. All data were analysed by using STATA version 13. (STATA Corp, College Station, TX). 2.4. Role of the funding source Takeo Fujii is supported by the grant named Young investigator award for study abroad in clinical epidemiology from St. Luke’s Life Science Institution. 3. Results We identified 966 unduplicated studies. Initial review of titles and abstracts identified 25 potentially eligible trials, most screened studies were excluded because they were not randomised controlled trials. Among the twenty five trials identified by initial screening, six did not meet our inclusion criteria (Fig. 1) and as a result nineteen studies were included in our analysis (Table 1) [20–31,13,32–37]. Fourteen trials were conducted in the metastatic setting and five trials were conducted in the adjuvant setting. Fig. 2 is a network plot of comparisons made between doxorubicin (DOX), liposomal
Fig. 1. Literature search and selection. We identified 966 articles for review and selected in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.
doxorubicin (LD), epirubicin (Epi) and nonanthracycline based regimens (Non). The size of each circle represents the number of patients included in the arm. The width of connecting bars reflects the number of comparisons made between the circles. DOX arm has the largest N = 3823 and the comparison between DOX and NON was most frequently made (=6 times). The pooled odds ratios are shown in predictive interval plot (Fig. 3). In that plot, circled chemotherapeutic agents were considered as a reference and comparisons were made in the order listed in the figure: Non versus DOX, Non versus Epi, Non versus LD, DOX versus Epi, DOX versus LD and EPI versus LD. It showed that LD has less CE3s than DOX (OR 0.60, 95% confidence interval (CI); 0.34–1.07) or Epi (OR 0.95, 95%CI; 0.39–2.33) but those differences were not statistically significant. In addition, there was a trend that DOX was more toxic than Non. The OR of DOX versus Non was 157 (95%CI; 0.90–2.72). We conducted cumulative ranking plot analysis (Fig. 4) to display hierarchy of these agents in terms of their relative cardiac toxicity. Surface under the cumulative ranking (SUCRA) showed the hierarchy. The larger an area under the curve becomes, the greater CE rates the agent has. The SUCRA revealed that DOX was associated with the highest rates of CE3 and LD with the least. 4. Discussion This is the first network meta-analysis comparing cardiac toxicity of different anthracyclines. Our study showed that doxorubicin has more cardiac toxicity than non-anthracycline based regimens. Also, liposomal doxorubicin trended towards less cardiac toxicity than doxorubicin.
Fig. 2. Network plot of included studies. The circles represent pooled N of included studies and the width of bridging bars reflects the number of both direct and indirect comparisons.
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
Fig. 3. Predictive interval plot for the network of LD, Epi, Dox and Non LD trends less cardiotoxic than Dox. Dox trends more cardiotoxic than Non. No difference between Epi, LD and Non.
NMA is a relatively new statistical method to compare existing data sets. The analysis allows both indirect and direct comparison by bridging multiple competitors with a shared competitor. For instance, if there are reports comparing treatment A versus B and A versus C and we do not have a head to head comparison between B and C, then NMA can provide the relative efficacy of B and C by establishing A as a reference [38,39]. This new method will give us an opportunity to reassess the existing data with a mock head to head comparison, especially in situations where a head to head comparison is difficult. For example, NMA allows us to compare a current standard regimen and a former standard regimen. A previous meta-analysis by Smith et al. [4] revealed that anthracycline-based regimens have significantly
5
higher cardiac adverse event rates than nonanthracycline regimens (OR 5.43, 95%CI: 2.34–12.62). In the same study liposomal doxorubicin was associated with significantly less CEs compared with non-liposomal doxorubicin (OR 0.18, 95% CI: 0.13–0.33). Notably that study included grade 1 and 2 cardiac adverse events. By contrast, when only CE3s are assessed, as in our study, the small number of events, while more common with anthracycline-based regimens do not reach statistical significance. These findings indicate that clinically significant cardiac toxicity can be avoided by limiting cumulative lifetime anthracycline exposure and therefore does not need to be factored into treatment selection. Our study has a couple of limitations. First, our study did not consider reasons other than cardiac events for stopping chemotherapy. If chemotherapy was postponed or discontinued early due to reasons other than CEs, the incidence of CEs would not be observed due to the right censoring. This fact may have contributed to the non-statistical significance of our results. Second, the dose of chemotherapeutic agents we included was heterogeneous. It is difficult to assess per unit cardiac toxicity of these agents since these agents have a unique therapeutic concentration. Yet studies we included used a standard dosage of each agent. EPI was mostly given at the median cumulative dosage from 500 to 800 mg/m2. The median cumulative dosage of DOX ranged from 264 to 570 mg/m2 (Table 1) We think the contribution of this heterogeneity is clinically minimal. Third, our study did not take molecular predictive factors of anthracycline-induced cardiotoxicity such as L461L within the SLC28A3 gene [40] into account. However this SNP theoretically should distribute equally to each arm during randomisation within each trial and should not alter our conclusion.
Fig. 4. Cumulative ranking plot for the network. 1: Most 4: Least cardiotoxic SUCRA: The surface under the cumulative ranking. Solid line corresponds to the unadjusted model. Red dashed lines are adjusted for small study effects. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
6
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
Finally, we could not perform sensitivity analysis with NMA meaning that we could not statistically evaluate the contribution of risk factors which might affect the incidence of CEs. We performed a descriptive review of these risk factors. In terms of observation time, we have two outliers whose observation times were 120 and 160 months. The reported rate of CE3 was 0.8% and 0.6% which was not higher than others. We did not find heterogeneity in terms of patients’ age among the included studies. Regarding the setting of treatment, CE3 rate for local or locally advanced disease ranged from 0.7% to 6.5%, in turn, those for metastatic disease ranged from 0.8% to 8.5% if we exclude one outlier. That study [21] reported a CE rate in the DOX arm of 23%. Notably 34% of patients in this arm had at least one out of four cardiac risk factors (age >65 years old, prior DOX >240 mg/m2, radiation to chest area and preexisting cardiac disease). The patients included in this study likely had more underlying cardiac risk factors than subjects in other trials although most other studies did not report cardiac risk factors. In terms of prior exposure to anthracyclines, CE3 rate for pre-treated groups ranged from 0.8% to 7.6% and that for anthracycline naive group ranged from 0.8% to 8.5%, again excluding the same outlier which showed a 23% CE rate. In conclusion, our systematic review and NMA revealed that DOX appears to be the most cardiotoxic anthracycline while LD is the least cardiotoxic member of the group although we could not demonstrate this to a degree of statistical significance due to the small number of CE3s. The low CE3 incidence was likely due to adherence of established lifetime anthracycline exposure limits. Among the three anthracyclines, LD is relatively safe in terms of cardiac events. Overall, we can say that potential cardiac toxicity need not be a major factor in selecting a particular anthracycline based regimen for the treatment of breast cancer patients without underlying heart disease. Contributors NY, TF and RB contributed to study concept and design. NY, TF, SA contributed to collection and assembly of data. NY, TF, SA, PK, GH and RB contributed to data analysis and interpretation. NY, TF and SA wrote the manuscript. All the authors approved the final version of the manuscript. Conflict of interest statement None declared. References [1] Kalam K, Marwick TH. Role of cardioprotective therapy for prevention of cardiotoxicity with chemotherapy: a systematic review and meta-analysis. Eur J Cancer 2013;49(13):2900–9.
[2] Sant M, Allemani C, Santaquilani M, Knijn A, Marchesi F, Capocaccia R, et al. EUROCARE-4. Survival of cancer patients diagnosed in 1995–1999. Results and commentary. Eur J Cancer 2009;45(6):931–91. [3] Pai VB, Nahata MC. Cardiotoxicity of chemotherapeutic agents: incidence, treatment and prevention. Drug Saf 2000;22(4): 263–302. [4] Smith LA, Cornelius VR, Plummer CJ, et al. Cardiotoxicity of anthracycline agents for the treatment of cancer: systematic review and meta-analysis of randomised controlled trials. BMC Cancer 2010;10:337. [5] Swain SM, Whaley FS, Ewer MS. Congestive heart failure in patients treated with doxorubicin: a retrospective analysis of three trials. Cancer 2003;97:2869–79. [6] Von Hoff DD, Layard MW, Basa P, et al. Risk factors for doxorubicin-induced congestive heart failure. Ann Intern Med 1979;91:710–7. [7] Ryberg M, Nielsen D, Skovsgaard T, et al. Epirubicin cardiotoxicity: an analysis of 469 patients with metastatic breast cancer. J Clin Oncol 1998;16:3502–8. [8] Lyu YL, Kerrigan JE, Lin CP, et al. Topoisomerase IIbeta mediated DNA double-strand breaks: implications in doxorubicin cardiotoxicity and prevention by dexrazoxane. Cancer Res 2007;67(18):8839–46. [9] Capranico G, Tinelli S, Austin CA, et al. Different patterns of gene expression of topoisomerase II isoforms in differentiated tissues during murine development. Biochim Biophys Acta 1992;1132(1):43–8. [10] Zhang S, Liu X, Bawa-Khalfe T, et al. Identification of the molecular basis of doxorubicin-induced cardiotoxicity. Nat Med 2012;18(11):1639–42 [Epub 2012 Oct 28]. [11] Weiss RB. The anthracyclines: will we ever find a better doxorubicin? Semin Oncol 1992;19:670–86. [12] Hortobagyi GN, Yap HY, Kau SW, et al. A comparative study of doxorubicin and epirubicin in patients with metastatic breast cancer. Am J Clin Oncol 1989;12:57–62. [13] O’Brien ME, Wigler N, Inbar M, et al. Reduced cardiotoxicity and comparable efficacy in a phase III trial of pegylated liposomal doxorubicin HCl (CAELYX/Doxil) versus conventional doxorubicin for first-line treatment of metastatic breast cancer. Ann Oncol 2004;15:440–9. [14] Rahman AM, Yusuf SW, Ewer MS. Anthracycline-induced cardiotoxicity and the cardiac-sparing effect of liposomal formulation. Int J Nanomedicine 2007;2(4):567–83. [15] Mills EJ, Thorlund K, Ioannidis JP. Demystifying trial networks and network meta-analysis. BMJ 2013;14(346):f2914. [16] Jansen JP, Fleurence R, Devine B, Itzler R, Barrett A, Hawkins N, et al. Interpreting indirect treatment comparisons and network meta-analysis for health-care decision making: report of the ISPORTask Force on Indirect Treatment Comparisons Good Research Practices: part 1. Value Health 2011;14(4):417–28. [17] Hoaglin DC, Hawkins N, Jansen JP, Scott DA, Itzler R, Cappelleri JC, et al. Conducting indirect-treatment-comparison and network-meta-analysis studies: report of the ISPOR Task Force on Indirect Treatment Comparisons Good Research Practices: part2. Value Health 2011;14(4):429–37. [18] Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement. Int J Surg 2010;8:336–41. [19] Chaimani A, Higgins JP, Mavridis D, Spyridonos P, Salanti G. Graphical tools for network meta-analysis in STATA. PLoS One 2013;8:e76654. [20] Bachelot T, Bajard A, Ray-Coquard I, et al. Final results of ERASME-4: a randomized trial of first-line docetaxel plus either capecitabine or epirubicin for metastatic breast cancer. Oncology 2011;80:262–8. [21] Batist G, Ramakrishnan G, Rao CS, et al. Reduced cardiotoxicity and preserved antitumor efficacy of liposome-encapsulated dox-
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031
N. Yamaguchi et al. / European Journal of Cancer xxx (2015) xxx–xxx
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
orubicin and cyclophosphamide compared with conventional doxorubicin and cyclophosphamide in a randomized, multicenter trial of metastatic breast cancer. J Clin Oncol 2001;19:1444–54. Bontenbal M, Andersson M, Wildiers J, et al. Doxorubicin vs epirubicin, report of a second-line randomized phase II/III study in advanced breast cancer. EORTC Breast Cancer Cooperative Group. Br J Cancer 1998;77:2257–63. Chan S, Davidson N, Juozaityte E, et al. Phase III trial of liposomal doxorubicin and cyclophosphamide compared with epirubicin and cyclophosphamide as first-line therapy for metastatic breast cancer. Ann Oncol 2004;15:1527–34. Chan S, Friedrichs K, Noel D, et al. Prospective randomized trial of docetaxel versus doxorubicin in patients with metastatic breast cancer. J Clin Oncol 1999;17:2341–54. Feher O, Vodvarka P, Jassem J, et al. First-line gemcitabine versus epirubicin in postmenopausal women aged 60 or older with metastatic breast cancer: a multicenter, randomized, phase III study. Ann Oncol 2005;16:899–908. Fisher B, Brown AM, Dimitrov NV, et al. Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifen-nonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project B-15. J Clin Oncol 1990;8:1483–96. Harris L, Batist G, Belt R, et al. Liposome-encapsulated doxorubicin compared with conventional doxorubicin in a randomized multicenter trial as first-line therapy of metastatic breast carcinoma. Cancer 2002;94:25–36. Hochster HS, Vogel CL, Burman SL, White R. Activity and safety of vinorelbine combined with doxorubicin or fluorouracil as first-line therapy in advanced breast cancer: a stratified phase II study. Oncologist 2001;6:269–77. Hutchins LF, Green SJ, Ravdin PM, et al. Randomized, controlled trial of cyclophosphamide, methotrexate, and fluorouracil versus cyclophosphamide, doxorubicin, and fluorouracil with and without tamoxifen for high-risk, node-negative breast cancer: treatment results of Intergroup Protocol INT-0102. J Clin Oncol 2005;23:8313–21. Lopez M, Papaldo P, Di Lauro L, Vici P, Carpano S, Conti EM. 5-Fluorouracil, adriamycin, cyclophosphamide (FAC) vs. 5-fluorouracil, epirubicin, cyclophosphamide (FEC) in metastatic breast cancer. Oncology 1989;46:1–5.
7
[31] Martin M, Villar A, Sole-Calvo A, et al. Doxorubicin in combination with fluorouracil and cyclophosphamide (i.v. FAC regimen, day 1, 21) versus methotrexate in combination with fluorouracil and cyclophosphamide (i.v. CMF regimen, day 1, 21) as adjuvant chemotherapy for operable breast cancer: a study by the GEICAM group. Ann Oncol 2003;14:833–42. [32] Perez DJ, Harvey VJ, Robinson BA, et al. A randomized comparison of single-agent doxorubicin and epirubicin as firstline cytotoxic therapy in advanced breast cancer. J Clin Oncol 1991;9:2148–52. [33] Piccart MJ, Di Leo A, Beauduin M, et al. Phase III trial comparing two dose levels of epirubicin combined with cyclophosphamide with cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer. J Clin Oncol 2001;19:3103–10. [34] Sledge GW, Neuberg D, Bernardo P, et al. Phase III trial of doxorubicin, paclitaxel, and the combination of doxorubicin and paclitaxel as front-line chemotherapy for metastatic breast cancer: an intergroup trial (E1193). J Clin Oncol 2003;21:588–92. [35] Sparano JA, Makhson AN, Semiglazov VF, et al. Pegylated liposomal doxorubicin plus docetaxel significantly improves time to progression without additive cardiotoxicity compared with docetaxel monotherapy in patients with advanced breast cancer previously treated with neoadjuvant-adjuvant anthracycline therapy: results from a randomized phase III study. J Clin Oncol 2009;27:4522–9. [36] Vici P, Colucci G, Giotta F, et al. A multicenter prospective phase II randomized trial of epirubicin/vinorelbine versus pegylated liposomal doxorubicin/vinorelbine as first-line treatment in advanced breast cancer. A GOIM study. J Exp Clin Cancer Res 2011;30:39. [37] Wils JA, Bliss JM, Marty M, et al. Epirubicin plus tamoxifen versus tamoxifen alone in node-positive postmenopausal patients with breast cancer: a randomized trial of the International Collaborative Cancer Group. J Clin Oncol 1999;17:1988–98. [38] Lu G, Ades AE. Combination of direct and indirect evidence in mixed treatment comparisons. Stat Med 2004;23(20):3105–24. [39] Caldwell DM, Ades AE, Higgins JP. Simultaneous comparison of multiple treatments: combining direct and indirect evidence. BMJ 2005;331(7521):897–900. [40] Visscher H, Ross CJ, Rassekh SR, et al. Pharmacogenomic prediction of anthracycline-induced cardiotoxicity in children. J Clin Oncol 2012;30(13):1422–8. http://dx.doi.org/10.1200/ JCO.2010.34.3467 [Epub 2011 Sep 6].
Please cite this article in press as: Yamaguchi N. et al., Comparison of cardiac events associated with liposomal doxorubicin, epirubicin and doxorubicin in breast cancer: a Bayesian network meta-analysis, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.07.031