Correlation of circulating pro-angiogenic miRNAs with cardiotoxicity induced by epirubicin/cyclophosphamide followed by docetaxel in patients with breast cancer

Abstract

OBJECTIVE:

This study aimed to evaluate predictive value of 14 pro-angiogenic miRNAs for cardiotoxicity induced by epirubicin/cyclophosphamide follow by docetaxel (EC-D) in breast cancer (BC) patients.

METHODS:

Three hundred and sixty-three BC patients receiving EC-D neoadjuvant chemotherapy were consecutively enrolled in this prospective cohort study. Peripheral blood sample was obtained from each patient, and plasma was separated. The expressions of 14 pro-angiogenic miRNAs, cardiac troponin I (cTnI) and N-terminal pro brain natriuretic peptide (NT-proBNP) were evaluated. Left ventricular ejection fraction (LVEF) level at C0, the end of 4 cycles of EC chemotherapy (C4), the end of 4 cycles of docetaxel treatment (C8), 3rd months (M3), 6th months (M6), 9th months (M9) and 12th months (M12) after surgery were assessed.

RESULTS:

LVEF decreased at C4, C8, M3, M6, M9 and M12 compared with C0, and the total cardiotoxicity incidence was 5.2%. Additionally, the levels of let-7f, miR-17-5p, miR-20a, miR-126, miR-210 and miR-378 were reduced in cardiotoxicity patients. Multivariate logistic regression revealed that miR-17-5p and miR-20a were independently predictive factors for less cardiotoxicity. Receiver operating characteristics (ROC) curve displayed a satisfactory predictive value for lower cardiotoxicity risk with area under curve (AUC) of 0.842 of the combination of the miR-17-5p and miR-20a expressions. In addition, let-7f,miR-126, miR-210 and miR-378 levels negatively correlated with cTnI expression, and let-7f and miR-130a expressions reversely correlated with NT-proBNP level.

CONLUSIONS:

miR-17-5p and miR-20a could be served as biomarkers for lower cardiotoxicity induced by EC-D neoadjuvant chemotherapy in BC patients.

Keywords

Pro-angiogenic miRNAs cardiotoxicity EC-D neoadjuvant chemotherapy breast cancer

1. Introduction

Breast cancer (BC), the most common diagnosed cancer among females with roughly 1,700,000 new cases and 520,000 deaths in 2012 worldwide, accounts for 15% new female cancer patients and displays an increasing incidence in China in 2015, demonstrating a correlation with lifestyle change that results in the elevated prevalence of obesity and inactivity [1, 2, 3, 4]. For the purpose of down staging BC for surgeries or to achieve breast conserving, neoadjuvant therapy rises as a great importance in clinical practice [5, 6]. Unfortunately, cardiotoxicity induced by chemotherapeutics, such as the anthracycline-based drugs, largely limits the use of neoadjuvant chemotherapy in BC patients [7, 8].

MicroRNA (miRNA), a category of small non-coding RNAs consists of roughly 19–23 nucleotides, regulates cells differentiation, proliferation, migration and apoptosis through mediating target genes and proteins [9, 10, 11, 12]. Moreover, the modulating function of miRNAs in several cardiac pathological processes have been validated to some extent, such as angiogenesis, myocardial fibrosis and cardiac remodeling [13, 14, 15, 16]. Angiogenesis has long been considered as a crucial process in cardiac diseases and is also a therapeutic target through promoting angiogenesis and improving cardiac tissue perfusion [17, 18, 19]. Accumulating studies report that there are miRNAs proven to be pro-angiogenic, however, the correlations of those pro-angiogenic miRNAs with cardiotoxicity induced by neoadjuvant chemotherapy in BC patients are still obscure [20].

Therefore, the aim of our study was to evaluate predictive value of 14 pro-angiogenic miRNAs for cardiotoxicity induced by epirubicin/cyclophosphamide follow by docetaxel (EC-D) in BC patients.

2. Materials and methods

2.1 Patients

In this prospective cohort study, we consecutively enrolled 365 BC patients who received EC-D neoadjuvant chemotherapy between Jan 2014 and Dec 2016 in The Second People’s Hospital of Liaocheng. Inclusion criteria were: 1) Diagnosed as primary BC confirmed by clinical, imaging and pathology examination. 2) Aged 18–75 years old. 3) Left ventricular ejection fraction (LVEF) $\geqslant$ 55%. 4) ECOG performance $\leqslant$ 2. 5) With TNM stage I–III and about to undergo surgery. 6) About to receive EC-D neoadjuvant chemotherapy. Patients who 1) diagnosed as metastatic breast cancer or bilateral breast cancer, 2) complicated with bone marrow abnormal, severe hepatic dysfunction or renal dysfunction and unmanageable hypertension, 3) had a history of coronary heart disease, myocardial infarction, heart failure, severe infection and other malignant solid tumors or hematologic malignancy, 4) were unlikely to be followed up regularly, 5) were pregnant or breast-feeding were excluded from this study.

This study was performed under Institutional Review Board approvals from The Second People’s Hospital of Liaocheng and conducted according to Declaration of Helsinki. Written informed consents were obtained from all patients.

2.2 Treatment

In this present study, treatments of all BC patients were not intervened. Based on disease condition ands patients’ willing, patients received EC-D neoadjuvant chemotherapy. Conventional neoadjuvant chemotherapy regimen was as follows: epirubicin, 100 mg/m ${}^{2}$ d1, cyclophosphamide, 600 mg/m ${}^{2}$ d1, 21 days per cycle for 4 cycles, then followed by docetaxel, 75–100 mg/m ${}^{2}$ d1, 21 days per cycle for 4 cycles. Among all the 77 patients with human epidermal growth factor receptor-2 (HER2) positive, 66 of them received trastuzumab treatment on demand (6 mg/kg, 21 days per cycle after docetaxel treatment). Due to financial problem, the remaining 11 patients did not receive trastuzumab treatment. All patients underwent surgical treatments after neoadjuvant chemotherapy. And corresponding adjuvant therapy was conducted post the operation in terms of the disease condition and clinical requirements.

2.3 Candidate miRNAs selection

Through analyzing the previous articles, the 14 pro-angiogenic miRNAs were selected from two studies, including a review and a research article [20, 21].

2.4 Samples collection

Ten ml peripheral blood sample was obtained from each patient before neoadjuvant chemotherapy (Cycle 0, C0), and plasma was separated by centrifugation from each sample. Plasma pro-angiogenic microRNAs expressions were determined by quantitative polymerase chain reaction (qPCR), the level of cardiac troponin I (cTnI) was assessed by immunofluorescence and the level of N-terminal pro brain natriuretic peptide (NT-proBNP) was evaluated by electrochemiluminescence immunoassay.

Table 1
Primers information in this study

Gene	Forward primer (5’- $>$ 3’)	Reverse primer (5’- $>$ 3’)
let-7b	ACACTCCAGCTGGGTGAGGTAGTAGGTTGTGT	TGTCGTGGAGTCGGCAATTC
let-7f	ACACTCCAGCTGGGTGAGGTAGTAGATTGTAT	TGTCGTGGAGTCGGCAATTC
miR-17-5p	ACACTCCAGCTGGGCAAAGTGCTTACAGTGCA	TGTCGTGGAGTCGGCAATTC
miR-17-3p	ACACTCCAGCTGGGACTGCAGTGAAGGCACTT	TGTCGTGGAGTCGGCAATTC
miR-18a	ACACTCCAGCTGGGTAAGGTGCATCTAGTGCA	TGTCGTGGAGTCGGCAATTC
miR-19a	ACACTCCAGCTGGGAGTTTTGCATAGTTGCAC	TGTCGTGGAGTCGGCAATTC
miR-19b-1	ACACTCCAGCTGGGAGTTTTGCAGGTTTGCAT	TGTCGTGGAGTCGGCAATTC
miR-20a	ACACTCCAGCTGGGTAAAGTGCTTATAGTGCA	TGTCGTGGAGTCGGCAATTC
miR-92a	ACACTCCAGCTGGGTATTGCACTTGTCCCGGC	TGTCGTGGAGTCGGCAATTC
miR-126	ACACTCCAGCTGGGCATTATTACTTTTGGTAC	TGTCGTGGAGTCGGCAATTC
miR-130a	ACACTCCAGCTGGGTTCACATTGTGCTACTGT	TGTCGTGGAGTCGGCAATTC
miR-210	ACACTCCAGCTGGGAGCCCCTGCCCACCGCAC	TGTCGTGGAGTCGGCAATTC
miR-296	ACACTCCAGCTGGGAGGGCCCCCCCTCAATCC	TGTCGTGGAGTCGGCAATTC
miR-378	ACACTCCAGCTGGGCTCCTGACTCCAGGTCCT	TGTCGTGGAGTCGGCAATTC
U6	CTCGCTTCGGCAGCACA	TGTCGTGGAGTCGGCAATTC

2.5 Cardiotoxicity assessment

LVEF was assessed by echocardiography at C0, the end of 4 cycles of EC chemotherapy (C4), the end of 4 cycles of docetaxel treatment (C8) and 3rd months (M3), 6th months (M6), 9th months (M9) and 12th months (M12) after surgery according to modified Simpson method [22]. Cardiotoxicity was defined as the LVEF declined by 10% from baseline to below 53% (the lower limit of normal) [23], heart failure, acute coronary artery syndrome or fatal arrhythmia [24].

Figure 1.

Study flow.

Table 2

Baseline characteristics of BC patients (C0)

	BC patients ( $N=$ 363)
Age (years)	45.38 $\pm$ 6.05
BMI (kg/m ${}^{2}$ )	23.27 $\pm$ 2.10
Obesity (BMI $\geqslant$ 25 kg/m ${}^{2}$ ) (n/%)	74 (20.4)
Smoke (n/%)	91 (25.1)
Hypertension (n/%)	73 (20.1)
Diabetes mellitus (n/%)	17 (4.7)
Dyslipidemia (n/%)	67 (18.5)
Hyperuricemia (n/%)	68 (18.7)
Chromic kidney disease (n/%)	12 (3.3)
ECOG performance (n/%)
0	288 (79.3)
1	71 (19.6)
2	4 (1.1)
LVEF (%)	67.15 $\pm$ 4.42
cTnI (ng/ml)	0.022 (0.010–0.057)
NT-proBNP (ng/ml)	0.077 (0.060–0.106)
Combined with trastuzumab (n/%)	66 (18.2)

Data were presented as mean $\pm$ standard deviation, count (%) or median (1/4–3/4 quantiles). BC, breast cancer; BMI, body mass index; ECOG, Eastern Cooperative Oncology Group; LVEF, left ventricular ejection fraction; cTnI, cardiac troponin I; NT-proBNP, N-terminal pro Brain Natriuretic Peptide.

2.6 qPCR assay

The expression of 14 pro-angiogenic miRNAs were detected by qPCR assay. Firstly, total RNA from 200 $\mu$ l plasma sample was extracted by the TRIzol solution (Invitrogen, USA) following the instruction of the manufacture. Then the total RNA was reversely transcript by PrimeScript™ RT reagent Kit (Perfect Real Time) (TaKaRa, Japan) with catalog number RR037A and labelled by KAPA SYBR FAST qPCR kit (KAPA, USA). In addition, sequences of the reverse primers were listed in Supplementary Table S1. The reaction volume, detection system and number of replicates were 20 $\mu$ l, ABI7500 (ABI, USA) and three, respectively. Using U6 as internal reference, the expressions of miRNAs were subsequently calculated by 2 ${}^{-\Delta\Delta\text{Ct}}$ formula. The primer sequences used for qPCR were shown in Table 1.

2.7 Detection of cTnI and NT-proBNP

The cTnI level was evaluated by One Step Test for CK-MB/cTnI/Myo (Colloidal Gold) (Getein Biotech, Inc., China) with catalog number Getein1100. The expression of NT-proBNP was assessed by the Cobas ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ e 411analyzer (Roche Ltd, Switzerland).

Table 3
Baseline relative expressions of pro-angiogenic miRNAs (C0)

miRNAs	Relative expression
let-7b	0.998 (0.485–1.460)
let-7f	1.226 (0.602–2.192)
miR-17-5p	0.884 (0.542–1.345)
miR-17-3p	1.054 (0.544–1.812)
miR-18a	1.340 (0.646–2.298)
miR-19a	1.871 (0.901–3.295)
miR-19b-1	1.309 (0.578–2.038)
miR-20a	1.415 (0.643–2.399)
miR-92a	1.316 (0.679–1.992)
miR-126	1.457 (0.712–2.392)
miR-130a	2.789 (1.383–4.678)
miR-210	1.175 (0.550–1.955)
miR-296	0.736 (0.384–1.378)
miR-378	0.878 (0.443–1.456)

Data were presented as median (1/4–3/4 quantiles).

2.8 Statistics

The SPSS 22.0 software (IBM Corp., Ltd, USA) and Graph Pad 6.0 (GraphPad Software Inc, USA) were used for statistical analysis in our study. Data was presented as mean $\pm$ standard deviation, count (%) or median (1/4–3/4 quartile). Comparison between two groups was determined by Wilcoxon rank sum test. Spearman test was used to assess the correlations of the expressions of 14 pro-angiogenic miRNAs with cTnI and NT-proBNP levels. Univariate and multivariate logistic regression analyses were performed to assess the predictive factors for cardiotoxicity risk, then receiver operating characteristics (ROC) curve was performed to evaluate the predicting values of independent pro-angiogenic miRNAs for cardiotoxicity risk. $p<$ 0.05 was considered significant.

3. Results

3.1 Study flow

The study flow was displayed in Fig. 1, which presented that there were 848 patients who were invited at the initiation of our study and 256 patients were excluded due to missed invitation ( $N=$ 47) and disagreed to participate ( $N=$ 209). Subsequently, the remaining 592 patients were screened for eligibility and 227 patients were excluded, which consisted of 159 exclusions and 68 disagreed with informed consent, so leaving 365 patients who were included in our study. In the 365 patients, 37 patients didn’t complete the study as a result of there were 35 patients lost follow up and 2 patients withdrew the informed consents, so remaining 328 patients completed the whole study. The statistical analysis in our study was performed on the basis of intention to treat analysis ( $N=$ 363). The data of the patients who withdrew the informed consents ( $N=$ 2) were not analyzed, and the data of those patients who didn’t complete the study ( $N=$ 37) was recorded according to the last follow up.

3.2 Baseline characteristics of BC patients

As listed in Table 2, the mean age of total BC patients was 45.38 $\pm$ 6.05 years and the mean value of body mass index (BMI) was 23.27 $\pm$ 2.10 kg/m ${}^{2}$ . Seventy four (20.4%) patients had obesity and 91 (25.1%) patients had smoke history. The number of patients with hypertension, diabetes mellitus, dyslipidemia, hyperuricemia and chronic kidney disease were 17 (4.7%), 67 (18.5%), 68 (18.7%) and 12 (3.3%), respectively. Two hundred and eighty-eight (79.3%) patients were allocated to Eastern Cooperative Oncology Group (ECOG) performance score 0, and the number of patients had ECOG score 1 and 2 were 71 (19.6%) and 4 (1.1%), respectively. The mean level of LVEF at baseline was 67.15 $\pm$ 4.42 %. In addition, the median levels of cTnI and NT-proBNP were 0.022 (0.010–0.057) ng/ml and 0.077 (0.060–0.106) ng/ml. And 66 (18.2%) patients received treatments with the combination of trastuzumab.

Figure 2.

Correlation of pro-angiogenic miRNAs with cTnI level. let-7f (B), miR-126 (J), miR-210 (L) and miR-378 (N) expressions were negatively correlated with cTnI level, however, other pro-angiogenic miRNAs were not correlated with cTnI expression (A, C–I, K and M). Spearman test was used to assess the correlations of the expressions of 14 pro-angiogenic miRNAs with cTnI level. $p<$ 0.05 was considered significant. cTnI, cardiac troponin I.

Figure 3.

Correlation of pro-angiogenic miRNAs with NT-proBNP level. The levels of let-7f (B) and miR-130a (K) were reversely correlated with the level of NT-proBNP, while no association of the other pro-angiogenic miRNAs expressions with NT-proBNP level was discovered (A, C–J and L–N). Spearman test was used to assess the correlations of the expressions of 14 pro-angiogenic miRNAs wiht NT-proBNP level. $p<$ 0.05 was considered significant. NT-proBNP, N-terminal pro brain natriuretic peptide.

3.3 Expressions of 14 pro-angiogenic miRNAs at baseline

The expressions of 14 pro-angiogenic miRNAs at baseline detected by qPCR were presented in Table 3, which displayed that the expressions of let-7b, let-7f, miR-17-5p, miR-17-3p, miR-18a, miR-19a, miR-19b-1, miR-20a, miR-92a, miR-126, miR-130a, miR-210, miR-296, miR-378 were 0.998 (0.485–1.460), 1.226 (0.602–2.192), 0.884 (0.542–1.345), 1.054 (0.544–1.812), 1.340 (0.646–2.298), 1.871 (0.901–3.295), 1.309 (0.578–2.038), 1.415 (0.643–2.399), 1.316 (0.679–1.992), 1.457 (0.712–2.392), 2.789 (1.383–4.678), 1.175 (0.550–1.955), 0.736 (0.384–1.378) and 0.878 (0.443–1.456), respectively.

3.4 Correlations of 14 pro-angiogenic miRNAs with the levels of cardiotoxicity biomarkers

The levels of let-7f ( $r=$ $-$ 0.287, $p<$ 0.001, Fig. 2B), miR-126 ( $r=$ $-$ 0.361, $p<$ 0.001, Fig. 2J), miR-210 ( $r=$ $-$ 0.322, $p<$ 0.001, Fig. 2L) and miR-378 ( $r=$ $-$ 0.128, $p=$ 0.015, Fig. 2N) were negatively correlated with cTnI level. However, no association of other pro-angiogenic miRNAs with cTnI expression was discovered (Fig. 2A, C–I, K and M).

As presented in Fig. 3, the expression of let-7f ( $r=$ $-$ 0.104, $p=$ 0.047) (Fig. 3B) and miR-130a ( $r=$ $-$ 0.111, $p=$ 0.035, Fig. 3K) were reversely correlated with the level of NT-proBNP, while, the other pro-angiogenic miRNAs expressions were not associated with NT-proBNP level (Fig. 3A, C–J and L–N).

3.5 Cardiotoxicity incidence among BC patients received EC-D neoadjuvant chemotherapy

As shown in Fig. 4, LVEF level obviously decreased at C4 ( $p<$ 0.001) and C8 ( $p<$ 0.001) compared with C0, although the LVEF level presented an increasing trend from C8 to M9 and a decreasing trend from M9 to M12, the LVEF levels at M3 ( $p<$ 0.001), M6 ( $p<$ 0.001), M9 ( $p<$ 0.001) and M12 ( $p<$ 0.001) were still notably declined compared with C0. The median values of LVEF at all visits were presented in Supplementary Table S2.

There were 19 (5.2%) patients presented as LVEF level declined by 10% from baseline to below 53% throughout the whole follow up duration, while, no patient had heart failure, acute coronary syndrome or life threatening arrhythmias (Table 4). Therefore, the total cardiotoxicity incidence was 5.2%.

Figure 4.

LVEF levels post EC-D neoadjuvant chemotherapy. LVEF level was declined at C4 and C8 compared with C0, although it presented an increasing trend from C8 to M9 and a decreasing trend from M9 to M12, the LVEF levels at M3, M6, M9 and M12 were still decreased compared with C0. Comparison between two groups was determined by t test. $p<$ 0.05 was considered significant. LVEF, left ventricular ejection fraction; EC-D, epirubicin/cyclophosphamide follow by docetaxel.

Figure 5.

14 pro-angiogenic miRNAs expressions in cardiotoxicity and non-cardiotoxicity patients. let-7f (B), miR-17-5p (C), miR-20a (H), miR-126 (J), miR-210 (L) and miR-378 (N), cTnI (O) and NT-proBNP (P) expressions decreased in cardiotoxicity patients, but the expressions of other pro-angiogenic miRNAs between non-cardiotoxicity patients and cardiotoxicity patients were of no difference (A, D–G, I, K and M). Comparison between two groups was determined by Wilcoxon rank sum test. $p<$ 0.05 was considered significant.

3.6 Correlations of 14 pro-angiogenic miRNAs with cardiotoxicity risk

The levels of let-7f ( $p<$ 0.001, Fig. 5B), miR-17-5p ( $p=$ 0.003, Fig. 5C), miR-20a ( $p<$ 0.001, Fig. 5H), miR-126 ( $p<$ 0.001, Fig. 5J), miR-210 ( $p=$ 0.021, Fig. 5L) and miR-378 ( $p=$ 0.002, Fig. 5N) were declined in cardiotoxicity patients. Nonetheless, the expressions of other pro-angiogenic miRNAs between non-cardiotoxicity patients and cardiotoxicity patients were similar (Fig. 5A, D–G, K and M). In addition, the expressions of cTnI ( $p<$ 0.001, Fig. 5O) and NT-proBNP ( $p=$ 0.001, Fig. 5P) were both down regulated in non-cardiotoxicity patients compared with cardiotoxicity patients.

As listed in Table 5, univariate logistic regression analysis revealed that let-7f ( $p=$ 0.001), miR-17-5p ( $p=$ 0.006), miR-20a ( $p=$ 0.001), miR-126 ( $p=$ 0.002), miR-210 ( $p=$ 0.026) and miR-378 ( $p=$ 0.011) correlated with lower cardiotoxicity. Additionally, cTnI ( $P<$ 0.001) and NT-proBNP ( $P<$ 0.001) were associated with higher risk of cardiotoxicity. All factors with $p$ value below 0.1 were included in the multivariate logistic regression, which displayed that miR-17-5p ( $p=$ 0.014) and miR-20a ( $p=$ 0.024) were independently predictive for less cardiotoxicity. And cTnI ( $p=$ 0.023) was an independent predictive factor for cardiotoxicity.

3.7 Predicting value of the combination of miR-17-5p and miR-20a for cardiotoxicity risk

For the purpose of further evaluating the value of independent predictive pro-angiogenic miRNAs for predicting the cardiotoxicity risk in BC patients receiving EC-D neoadjuvant chemotherapy, ROC curve was performed. As shown in Fig. 6, combining the expressions of miR-17-5p and miR-20a had a satisfying predictive value for low cardiotoxicity risk with area under curve (AUC) of 0.842 (95% CI: 0.778–0.906) and a sensitivity of 78.9% as well as a specificity of 80.8% at best cut off. Additionally, the AUC for cTnI and NT-proBNP were 0.940 (95% CI: 0.908–0.972) and 0.718 (95% CI: 0.602–0.833), respectively.

Table 4
Assessment of cardiotoxicity

Items	$n$ (%)
LVEF declined by 10% from baseline to below 53%	19 (5.2)
(the lower limit of normal)
Heart failure	0 (0.0)
Acute coronary syndrome	0 (0.0)
Life threatening arrhythmias	0 (0.0)
Cardiotoxicity	19 (5.2)

Data was presented as count (%). LVEF, left ventricular ejection fraction.

Figure 6.

ROC curve analysis of independent pro-angiogenic miRNAs for predicting cardiotoxicity risk. ROC curve analysis showed that combining miR-17-5p and miR-20a expressions had a good predictive value for cardiotoxicity risk in BC patients receiving EC-D neoadjuvant chemotherapy. The formula for calculating the joint prediction probability was $[Y]=B_{0}+B_{1}X_{1}+B_{2}X_{2}$ . $Y$ referred to prediction probability, $B_{0}$ , $B_{1}$ and $B_{2}$ were all constants, $X_{1}$ referred to the miR-17-5p relative expression and $X_{2}$ referred to the miR-20a relative expression. Then the values of $B_{0}$ , $B_{1}$ and $B_{2}$ were calculated by the binary logistic regression, afterward, the values of $B_{0}$ , $B_{1}$ and $B_{2}$ ( $B_{0}=$ $-$ 0.195, $B_{1}=$ $-$ 1.621 and $B_{2}=$ $-$ 1.421) were put into the formula. Therefore, the final formula that was used for the classifier was Logistic $[Y]=-0.195-1.621X_{1}-1.421X_{2}$ . ROC, receiver operating characteristic; BC, breast cancer; EC-D, epirubicin/cyclophosphamide follow by docetaxel.

Table 5

Logistic regression analysis of pro-angiogenic miRNAs in predicting cardiotoxicity risk

	Univariate logistic regression				Multivariate logistic regression
	$p$ value	OR	95% CI		$p$ value	OR	95% CI
			Lower	Higher			Lower	Higher
let-7b	0.945	0.977	0.503	1.896	0.386	4.598	0.146	144.550
let-7f	0.001	0.228	0.095	0.548	0.073	0.000	0.000	2.047
miR-17-5p	0.006	0.213	0.071	0.642	0.014	0.000	0.000	0.134
miR-17-3p	0.302	1.289	0.796	2.089	0.576	1.964	0.184	20.972
miR-18a	0.261	0.773	0.494	1.211	0.349	0.369	0.046	2.970
miR-19a	0.103	0.741	0.517	1.062	0.384	0.452	0.075	2.704
miR-19b-1	0.245	0.731	0.430	1.240	0.982	1.031	0.076	14.051
miR-20a	0.001	0.264	0.119	0.586	0.024	0.005	0.000	0.484
miR-92a	0.896	0.965	0.563	1.652	0.066	0.075	0.005	1.184
miR-126	0.002	0.358	0.185	0.693	0.074	0.125	0.013	1.228
miR-130a	0.522	0.931	0.747	1.160	0.088	3.050	0.848	10.969
miR-210	0.026	0.475	0.247	0.913	0.834	0.752	0.052	10.772
miR-296	0.311	0.681	0.324	1.432	0.247	0.093	0.002	5.191
miR-378	0.011	0.278	0.104	0.742	0.054	0.002	0.000	1.120
cTnI	$<$ 0.001	3.3 $\pm$ 10 ${}^{\text{17}}$	5.0 $\pm$ 10 ${}^{\text{11}}$	2.2 $\pm$ 10 ${}^{\text{23}}$	0.023	1.4 $\pm$ 10 ${}^{\text{77}}$	6.2 $\pm$ 10 ${}^{\text{10}}$	2.9 $\pm$ 10 ${}^{\text{143}}$
NT-proBNP	$<$ 0.001	1.5 $\pm$ 10 ${}^{\text{7}}$	3550.499	6.7 $\pm$ 10 ${}^{\text{10}}$	0.266	0.000	0.000	5.6 $\pm$ 10 ${}^{\text{10}}$

Data were presented as $p$ value, OR (odds ratio) and 95% CI. Significance was determined by univariate and multivariate logistic regression analysis. $p$ value $<$ 0.05 was considered significant. cTnI, cardiac troponin I; NT-proBNP, N-terminal pro Brain Natriuretic Peptide.

3.8 Correlations of baseline characteristics with the occurrence of cardiotoxicity

As listed in Supplementary Table S3, complicating with dyslipidemia ( $p=$ 0.001), complicating with hyperuricemia ( $p=$ 0.007), HER2 positive ( $p=$ 0.021), and treatment combined with trastuzumab ( $p=$ 0.029) correlated with higher incidence of cardiotoxicity in BC patients. And cTnI level ( $p<$ 0.001) as well as NT-proBNP level ( $p=$ 0.001) were elevated in BC patients with cardiotoxicity compared with patients with non-cardiotoxicity.

4. Discussion

In our study, we discovered that: (1) the LVEF kept declining in BC patients receiving EC-D neoadjuvant chemotherapy and the cardiotoxicity incidence was 5.2%; (2) let-7f, miR-126, miR-210 and miR-378 expressions negatively associated with cTnI level, and let-7f as well as miR-130a were negatively correlated with the level of NT-proBNP; (3) multivariate logistic regression revealed that let-17-5p and miR-20a were independently associated with less cardiotoxicity, and ROC curve displayed good predictive value of combining miR-17-5p and miR-20a for lower cardiotoxicity risk.

Anthracyclines are basically used for the adjuvant chemotherapy of BC patients, nonetheless, they can induce cardiotoxicity through multiple approaches. It is reported that anthracyclines increases the oxygen radical that leads to lipid peroxidation, which inhibits mitochondrial respiration, promotes autophagy and apoptosis and destroys sarcomere structures of the cardia muscle cells [25, 26, 27]. Meanwhile, experiments also show that anthracyclines interacts with topoisomerase 2 (Top2)- $\beta$ to activate the fracture of deoxyribonucleic acid double bond, eventually resulting in cells death [25, 26, 27]. In addition, the anthracyclines also inhibits the receptor tyrosine-protein kinase erbB-2 (Erbb2) pathway and subsequently damages fiber structure of myocardial cells [28]. Thus, to eliminate the cardiotoxicity induced by the anthracyclines and improve treatment responses, sequential use of paclitaxel is becoming increasingly popular in clinical practice [29, 30, 31, 32, 33]. However, in view of the evaluation method of cardiotoxicity, previous studies mainly focus on recording the cardiac events related to cardiotoxicity but not LVEF, which is a more sensitive index [34, 35]. In our study, we found that the EC-D neoadjuvant chemotherapy for BC patients demonstrated a cardiotoxicity incidence of 5.2%, among whom all patients had a decline of LVEF by 10% to below 53% with the majority of the patients treated with combination of trastuzumab, and no patient presented with heart failure, acute coronary syndrome or life threatening arrhythmias. In a prior international, multicenter phase II clinical trial, an incidence of LVEF decrease of 4% in locally advanced BC patients or patients in pathological stage III receiving EC-D as neoadjuvant chemotherapy is reported [36]. While in another clinical study conducted in the US that evaluated the cardiotoxicity in BC patients treated by doxorubicin and cyclophosphamide followed by paclitaxel, trastuzumab, and pertuzumab (AC-THP), they find that 3.5% patients exhibit symptomatic heart failure and 12.3% patients had LVEF decline of 10% to below 53%, which leads to a total cardiotoxicity rate of 15.8% [37]. In Asia, a pilot, prospective cohort study from Japan elucidates that the cardiotoxicity rate is 10% in BC patients treated with anthracycline or trastuzumab [24]. The difference of cardiotoxicity incidence in our study compared with other studies might be resulted from the variant definition of cardiotoxicity, treatment duration and follow up time.

In order to better eliminate cardiotoxicity of anthracycline based neoadjuvant chemotherapy for BC patients, increasing studies aim to explore the biomarker for predicting the cardiotoxicity, such as cTnI and NT-proBNP. cTnI is a sensitive biomarker indicating cardiac damage, and is suggestive for numerous diseases such as acute coronary syndrome, heart failure and myocarditis [38, 39]. And NT-proBNP, a prohormone for brain natriuretic peptide representing increased ventricular filling pressure and myocardium stress, is also a sensitive biomarker for cardiac disease and is especially unique for heart failure [40, 41, 42, 43]. miRNAs, as emerging candidates serving as biomarker for cardiotoxicity, are drawing increasing attention among clinicians and researchers. A previous study reports that miR-140-5p exacerbates cardiotoxicity caused by the use of doxorubicin through enhancing myocardial oxidative stress via targeting Nrf2 and Sirt2 [44]. A recent study elucidates that miR-34a-5p promotes cardiotoxicity induced by Adriamycin through targeting Sirt1/p66shc pathway [45]. Another study illustrates that miR-320a enhances cardiotoxicity in patients receiving Adriamycin via targeting vascular endothelial growth factor (VEGF) signaling pathway [46]. Additionally, a clinical study discovers that circulating miR-1 expression could be served as a biomarker for cardiotoxicity induced by Adriamycin with a higher predictive value compared with cTn1 [47]. However, to our best knowledge, there is still no report about the predictive value of pro-angiogenic miRNAs for cardiotoxicity induced by EC-D neoadjuvant chemotherapy, and the effect of those miRNAs in the treatment of EC-D for BC patients is not clear either. Our study discovered for the first time that let-7f, miR-126, miR-210 and miR-378 expressions negatively associated with cTnI level, and let-7f as well as miR-130a were negatively correlated with NT-proBNP level. In addition, multivariate logistic regression displayed that miR-17-5p and miR-20a expressions were independently associated with less cardiotoxicity, and the combination of miR-17-5p and miR-20a disclosed a good predictive value for lower cardiotoxicity risk according to ROC curve analysis. The probable explanations of our results might be: (1) pro-angiogenic miRNAs diminish cardiotoxicity by promoting angiogenesis, regulating myocardial fibrosis, improving tissue perfusion and accelerating tissue recovering [48, 49, 50, 51]; (2) those pro-angiogenic miRNAs act similarly to miR-140-5p, miR-34a-5p and miR-320a in regulating the injury caused by Adriamycin, however this theory remains to be clarified by experimental studies in the future [44, 45, 46].

There were still several limitations in our study: (1) there was selection bias due to this was a single center study with the enrolled patients barely from the north China; (2) we only evaluated the expressions of 14 pro-angiogenic miRNAs before neoadjuvant chemotherapy but not at other periods; (3) the detail mechanism of those independent predictive pro-angiogenic miRNAs in regulating the process of cardiotoxicity induced by EC-D neoadjuvant chemotherapy still need to be investigated by further experimental studies; (4) our results showed that treatment combined with trastuzumab positively correlated with cardiotoxicity incidence, which caused bias in our study.

In conclusion, miR-17-5p and miR-20a could be served as biomarkers for lower cardiotoxicity induced by EC-D neoadjuvant chemotherapy in BC patients.

Footnotes

Conflict of interest

The authors have no financial conflicts of interest.

Supplementary data

Supplementary Table S1

Reverse transcript primers information in this study

Gene	RT primer (5’- $>$ 3’)
let-7b	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGAACCACAC
let-7f	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGAACTATAC
miR-17-5p	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGCTACCTGC
miR-17-3p	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGCTACAAGT
miR-18a	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGCTATCTGC
miR-19a	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGTGTAGTGC
miR-19b-1	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGGCTGGATG
miR-20a	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGCTACCTGC
miR-92a	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGAGCATTGC
miR-126	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGCGCGTACC
miR-130a	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGAGTAGCAC
miR-210	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGCAGTGTGC
miR-296	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGACAGGATT
miR-378	CTCAACTGGTGTCGTGGAGTCGGCAATTCAG
	TTGAGACACAGGA

RT, reverse transcript.

Supplementary Table S2

LVEF at each visit

Time point	LVEF (%)
C0	67.15 $\pm$ 4.42
C4	64.24 $\pm$ 5.53
C8	62.74 $\pm$ 6.32
M3	62.99 $\pm$ 6.59
M6	64.44 $\pm$ 6.72
M9	65.25 $\pm$ 7.29
M12	64.61 $\pm$ 7.53

Data were presented as mean $\pm$ standard deviation. LVEF, left ventricular ejection fraction.

Supplementary Table S3

Correlation of baseline characteristics with the occurrence of cardiotoxicity

Characteristics	Cardiotoxicity	Non-cardiotoxicity	$p$
	( $N=$ 19)	( $N=$ 346)	value
Age			0.780
$\geqslant$ 45 years	11 (5.5)	189 (94.5)
$<$ 45 years	8 (4.8)	157 (95.2)
BMI			0.208
$\geqslant$ 25 kg/m ${}^{2}$	6 (8.1)	68 (91.9)
$<$ 25 kg/m ${}^{2}$	13 (4.5)	278 (95.5)
Smoke			0.492
Yes	6 (6.6)	85 (93.4)
No	13 (4.7)	261 (95.3)
Hypertension			0.208
Yes	6 (8.1)	68 (91.9)
No	13 (4.5)	278 (95.5)
Diabetes mellitus			0.898
Yes	1 (5.9)	16 (94.1)
No	18 (5.2)	330 (94.8)
Dyslipidemia			0.001
Yes	9 (13.4)	58 (86.6)
No	10 (3.4)	288 (96.6)
Hyperuricemia			0.007
Yes	8 (11.8)	60 (88.2)
No	11 (3.7)	286 (96.3)
Chromic kidney disease			0.069
Yes	2 (16.7)	10 (83.3)
No	17 (4.8)	336 (95.2)
Higher ECOG performance			0.888
0	15 (5.2)	274 (94.8)
1	4 (5.6)	68 (94.4)
2	0 (0.0)	4 (100.0)
HER2			0.021
Positive	8 (10.4)	69 (89.6)
Negative	11 (3.8)	277 (96.2)
LVEF (%)	65.0 (63.0–68.0)	67 (64.0–71.0)	0.169
cTnI (ng/ml)	0.122 (0.086–0.138)	0.021 (0.010–0.047)	$<$ 0.001
NT-proBNP	0.094 (0.077–0.184)	0.077 (0.060–0.102)	0.001
(ng/ml)
Combined with trastuzumab			0.029
Yes	7 (10.6)	59 (89.4)
No	12 (4.0)	287 (96.0)

Data were presented as count (%) or median (25 ${}^{\text{th}}$ –75 ${}^{\text{th}}$ quantiles). Comparison was determined by Chi-square test or Wilcoxon rank sum test. $p$ value $<$ 0.05 was considered significant. BMI, body mass index; ECOG, Eastern Cooperative Oncology Group; LVEF, left ventricular ejection fraction; cTnI, cardiac troponin I; NT-proBNP, N-terminal pro Brain Natriuretic Peptide.

References

Harbeck

and Gnant

, Breast cancer, Lancet 389 (2017), 1134–1150.

Torre

L.A.

Bray

Siegel

R.L.

Ferlay

Lortet-Tieulent

and Jemal

, Global cancer statistics, 2012, CA Cancer J Clin 65 (2015), 87–108.

Goss

P.E.

Strasser-Weippl

Lee-Bychkovsky

B.L.

Fan

Chavarri-Guerra

Liedke

P.E.

Pramesh

C.S.

Badovinac-Crnjevic

Sheikine

Chen

Qiao

Y.L.

Shao

Y.L.

Fan

Chow

L.W.

Wang

Zhang

Shen

Purushotham

Sullivan

Badwe

Banavali

S.D.

Nair

Kumar

Parikh

Subramanian

Chaturvedi

Iyer

Shastri

S.S.

Digumarti

Soto-Perez-de-Celis

Adilbay

Semiglazov

Orlov

Kaidarova

Tsimafeyeu

Tatishchev

Danishevskiy

K.D.

Hurlbert

Vail

St Louis

and Chan

, Challenges to effective cancer control in China, India, and Russia, Lancet Oncol 15 (2014), 489–538.

Varghese

and Shin

H.R.

, Strengthening cancer control in China, Lancet Oncol 15 (2014), 484–5.

Volders

J.H.

Negenborn

V.L.

Spronk

P.E.

Krekel

N.M.A.

Schoonmade

L.J.

Meijer

Rubio

I.T.

and van den Tol

M.P.

Breast-conserving surgery following neoadjuvant therapy-a systematic review on surgical outcomes, Breast Cancer Res Treat (2017).

Sun

Liao

and Zhu

, Comparison of breast-conserving surgery with mastectomy in locally advanced breast cancer after good response to neoadjuvant chemotherapy: A PRISMA-compliant systematic review and meta-analysis, Medicine (Baltimore) 96 (2017), e8367.

Perez

E.A.

Romond

E.H.

Suman

V.J.

Jeong

J.H.

Sledge

Geyer

C.E.

, Jr. Martino

Rastogi

Gralow

Swain

S.M.

Winer

E.P.

Colon-Otero

Davidson

N.E.

Mamounas

Zujewski

J.A.

and Wolmark

, Trastuzumab plus adjuvant chemotherapy for human epidermal growth factor receptor 2-positive breast cancer: Planned joint analysis of overall survival from NSABP B-31 and NCCTG N9831, J Clin Oncol 32 (2014), 3744–52.

Moja

Tagliabue

Balduzzi

Parmelli

Pistotti

Guarneri

and D’Amico

, Trastuzumab containing regimens for early breast cancer, Cochrane Database Syst Rev (2012), CD006243.

Shi

Yuan

Zhang

Yuan

and Wang

, MicroRNA-379 suppresses cervical cancer cell proliferation and invasion by directly targeting V-crk avian sarcoma virus CT10 oncogene homolog-like, Oncol Res (2018).

10.

Boon

R.A.

Iekushi

Lechner

Seeger

Fischer

Heydt

Kaluza

Treguer

Carmona

Bonauer

Horrevoets

A.J.

Didier

Girmatsion

Biliczki

Ehrlich

J.R.

Katus

H.A.

Muller

O.J.

Potente

Zeiher

A.M.

Hermeking

and Dimmeler

, MicroRNA-34a regulates cardiac ageing and function, Nature 495 (2013), 107–10.

11.

Zhang

Dai

Fan

and Wei

, MicroRNA-16 inhibits cell proliferation and migration by targeting heat shock protein 70 in heat-denatured dermal fibroblasts, Korean J Intern Med (2018).

12.

Kurkewich

J.L.

Boucher

Klopfenstein

Baskar

Kapur

and Dahl

, The mirn23a and mirn23b microrna clusters are necessary for proper hematopoietic progenitor cell production and differentiation, Exp Hematol (2017).

13.

Liang

Zhao

Zan

Zhu

Ren

and Zhao

, MiR-93-5p enhances growth and angiogenesis capacity of HUVECs by down-regulating EPLIN, Oncotarget 8 (2017), 107033–107043.

14.

Verjans

Peters

Beaumont

F.J.

van Leeuwen

van Herwaarden

Verhesen

Munts

Bijnen

Henkens

Diez

de Windt

L.J.

van Nieuwenhoven

F.A.

van Bilsen

Goumans

M.J.

Heymans

Gonzalez

and Schroen

, MicroRNA-221/222 family counteracts myocardial fibrosis in pressure overload-induced heart failure, Hypertension 71 (2018), 280–288.

15.

Luo

Chen

Shi

and Su

, Rescuing infusion of miRNA-1 prevents cardiac remodeling in a heart-selective miRNA deficient mouse, Biochem Biophys Res Commun 495 (2018), 607–613.

16.

Yan

Wang

Zhang

Liu

Zhou

and Hua

, Contrary microRNA expression pattern between fetal and adult cardiac remodeling: Therapeutic value for heart failure, Cardiovasc Toxicol 17 (2017), 267–276.

17.

Zachary

and Morgan

R.D.

, Therapeutic angiogenesis for cardiovascular disease: Biological context, challenges, prospects, Heart 97 (2011), 181–9.

18.

Yang

Jin

Huang

Shi

G.P.

Wang

and Xiang

, EphrinB2/EphB4 pathway in postnatal angiogenesis: A potential therapeutic target for ischemic cardiovascular disease, Angiogenesis 19 (2016), 297–309.

19.

Clayton

Z.E.

Sadeghipour

and Patel

, Generating induced pluripotent stem cell derived endothelial cells and induced endothelial cells for cardiovascular disease modelling and therapeutic angiogenesis, Int J Cardiol 197 (2015), 116–22.

20.

Yang

and Li

, Role of specific microRNAs for endothelial function and angiogenesis, Biochem Biophys Res Commun 386 (2009), 549–53.

21.

Jin

and Xing

, Circulating pro-angiogenic and anti-angiogenic microRNA expressions in patients with acute ischemic stroke and their association with disease severity, Neurol Sci 38 (2017), 2015–2023.

22.

Lang

R.M.

Bierig

Devereux

R.B.

Flachskampf

F.A.

Foster

Pellikka

P.A.

Picard

M.H.

Roman

M.J.

Seward

Shanewise

J.S.

Solomon

S.D.

Spencer

K.T.

Sutton

M.S.

Stewart

W.J.

, Recommendations for chamber quantification: A report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology, J Am Soc Echocardiogr 18 (2005), 1440–63.

23.

Lang

R.M.

Badano

L.P.

Mor-Avi

Afilalo

Armstrong

Ernande

Flachskampf

F.A.

Foster

Goldstein

S.A.

Kuznetsova

Lancellotti

Muraru

Picard

M.H.

Rietzschel

E.R.

Rudski

Spencer

K.T.

Tsang

and Voigt

J.U.

, Recommendations for cardiac chamber quantification by echocardiography in adults: An update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging, J Am Soc Echocardiogr 28 (2015), 1–39 e14.

24.

Kitayama

Kondo

Sugiyama

Kurimoto

Nishino

Kawada

Hirayama

and Tsuji

, High-sensitive troponin T assay can predict anthracycline- and trastuzumab-induced cardiotoxicity in breast cancer patients, Breast Cancer 24 (2017), 774–782.

25.

Huang

Zhou

Wei

B.Q.

Zhang

R.C.

Yin

S.J.

Zou

C.H.

Zhang

Y.H.

and Zhang

, Neuregulin-1 attenuates doxorubicin-induced autophagy in neonatal rat cardiomyocytes, J Cardiovasc Pharmacol 62 (2013), 130–7.

26.

Barry

Alvarez

J.A.

Scully

R.E.

Miller

T.L.

and Lipshultz

S.E.

, Anthracycline-induced cardiotoxicity: Course, pathophysiology, prevention and management, Expert Opin Pharmacother 8 (2007), 1039–58.

27.

Ito

Miller

S.C.

Billingham

M.E.

Akimoto

Torti

S.V.

Wade

Gahlmann

Lyons

Kedes

and Torti

F.M.

, Doxorubicin selectively inhibits muscle gene expression in cardiac muscle cells in vivo and in vitro , Proc Natl Acad Sci USA 87 (1990), 4275–9.

28.

Sawyer

D.B.

Peng

Chen

Pentassuglia

and Lim

C.C.

, Mechanisms of anthracycline cardiac injury: Can we identify strategies for cardioprotection? Prog Cardiovasc Dis 53 (2010), 105–13.

29.

Limwattananon

Maoleekulpairoj

and Soparatanapaisal

, Cost-effectiveness analysis of sequential paclitaxel adjuvant chemotherapy for patients with node positive primary breast cancer, J Med Assoc Thai 89 (2006), 690–8.

30.

Piccart

M.J.

Lohrisch

Duchateau

and Buyse

, Taxanes in the adjuvant treatment of breast cancer: Why not yet? J Natl Cancer Inst Monogr (2001), 88–95.

31.

Andre

Broglio

Roche

Martin

Mackey

J.R.

Penault-Llorca

Hortobagyi

G.N.

and Pusztai

, Estrogen receptor expression and efficacy of docetaxel-containing adjuvant chemotherapy in patients with node-positive breast cancer: Results from a pooled analysis, J Clin Oncol 26 (2008), 2636–43.

32.

Martin

Segui

M.A.

Anton

Ruiz

Ramos

Adrover

Aranda

Rodriguez-Lescure

Grosse

Calvo

Barnadas

Isla

Martinez del Prado

Ruiz Borrego

Zaluski

Arcusa

Munoz

Lopez Vega

J.M.

Mel

J.R.

Munarriz

Llorca

Jara

Alba

Florian

Lopez Garcia-Asenjo

J.A.

Saez

Rios

M.J.

Almenar

Peiro

Lluch

and Investigators

, Adjuvant docetaxel for high-risk, node-negative breast cancer, N Engl J Med 363 (2010), 2200–10.

33.

Coudert

Asselain

Campone

Spielmann

Machiels

J.P.

Penault-Llorca

Serin

Levy

Romieu

Canon

J.L.

Orfeuvre

Piot

Petit

Jerusalem

Audhuy

Veyret

Beauduin

Eymard

J.C.

Martin

A.L.

Roche

and Group

U.B.

, Extended benefit from sequential administration of docetaxel after standard fluorouracil, epirubicin, and cyclophosphamide regimen for node-positive breast cancer: The 8-year follow-up results of the UNICANCER-PACS01 trial, Oncologist 17 (2012), 900–9.

34.

Pierga

J.Y.

Delaloge

Espie

Brain

Sigal-Zafrani

Mathieu

M.C.

Bertheau

Guinebretiere

J.M.

Spielmann

Savignoni

and Marty

, A multicenter randomized phase II study of sequential epirubicin/cyclophosphamide followed by docetaxel with or without celecoxib or trastuzumab according to HER2 status, as primary chemotherapy for localized invasive breast cancer patients, Breast Cancer Res Treat 122 (2010), 429–37.

35.

Jacot

Bibeau

Gourgou-Bourgade

Gutowski

Colombo

P.E.

Bleuse

J.P.

Kramar

and Romieu

, Phase II trial of dose dense docetaxel followed by FEC100 as neoadjuvant chemotherapy in patients with operable breast cancer, Am J Clin Oncol 33 (2010), 544–9.

36.

Smith

J.W.

, 2nd Buyse

M.E.

Rastogi

Geyer

C.E.

, Jr. Jacobs

S.A.

Patocskai

E.J.

Robidoux

Conlin

A.K.

Ansari

Keogh

G.P.

Stella

P.J.

Gross

H.M.

Lord

R.S.

Polikoff

J.A.

Mauquoi

Mamounas

E.P.

Swain

S.M.

and Wolmark

, Epirubicin With cyclophosphamide followed by docetaxel with trastuzumab and bevacizumab as neoadjuvant therapy for HER2-positive locally advanced breast cancer or as adjuvant therapy for HER2-Positive Pathologic Stage III breast cancer: A phase II trial of the NSABP foundation research group, FB-5, Clin Breast Cancer 17 (2017), 48–54 e3.

37.

A.F.

Singh

J.C.

Wang

Liu

J.E.

Eaton

Oeffinger

K.C.

Steingart

R.M.

Hudis

C.A.

and Dang

C.T.

, Cardiac safety of dual anti-HER2 therapy in the neoadjuvant setting for treatment of HER2-positive breast cancer, Oncologist 22 (2017), 642–647.

38.

Newby

L.K.

Jesse

R.L.

Babb

J.D.

Christenson

R.H.

De Fer

T.M.

Diamond

G.A.

Fesmire

F.M.

Geraci

S.A.

Gersh

B.J.

Larsen

G.C.

Kaul

McKay

C.R.

Philippides

G.J.

and Weintraub

W.S.

, ACCF 2012 expert consensus document on practical clinical considerations in the interpretation of troponin elevations: A report of the American College of Cardiology Foundation task force on Clinical Expert Consensus Documents, J Am Coll Cardiol 60 (2012), 2427–63.

39.

Carlberg

D.J.

Tsuchitani

Barlotta

K.S.

and Brady

W.J.

, Serum troponin testing in patients with paroxysmal supraventricular tachycardia: Outcome after ED care, Am J Emerg Med 29 (2011), 545–8.

40.

Kim

H.N.

and Januzzi

J.L.

, Jr., Natriuretic peptide testing in heart failure, Circulation 123 (2011), 2015–9.

41.

Braunwald

, Biomarkers in heart failure, N Engl J Med 358 (2008), 2148–59.

42.

Bettencourt

Azevedo

Pimenta

Frioes

Ferreira

and Ferreira

, N-terminal-pro-brain natriuretic peptide predicts outcome after hospital discharge in heart failure patients, Circulation 110 (2004), 2168–74.

43.

Hartmann

Packer

Coats

A.J.

Fowler

M.B.

Krum

Mohacsi

Rouleau

J.L.

Tendera

Castaigne

Anker

S.D.

Amann-Zalan

Hoersch

and Katus

H.A.

, Prognostic impact of plasma N-terminal pro-brain natriuretic peptide in severe chronic congestive heart failure: A substudy of the Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) trial, Circulation 110 (2004), 1780–6.

44.

Zhao

Tao

Han

Yin

and Peng

, MicroRNA-140-5p aggravates doxorubicin-induced cardiotoxicity by promoting myocardial oxidative stress via targeting Nrf2 and Sirt2, Redox Biol 15 (2017), 284–296.

45.

Zhu

J.N.

Y.H.

Z.Q.

W.Y.

Tang

C.M.

Fei

H.W.

Yang

Lin

Q.X.

Gou

D.M.

S.L.

and Shan

Z.X.

, Activation of miR-34a-5p/Sirt1/p66shc pathway contributes to doxorubicin-induced cardiotoxicity, Sci Rep 7 (2017), 11879.

46.

Yin

Zhao

Yan

Zhou

Chen

and Wang

D.W.

, miR-320a mediates doxorubicin-induced cardiotoxicity by targeting VEGF signal pathway, Aging (Albany NY) 8 (2016), 192–207.

47.

Rigaud

V.O.

Ferreira

L.R.

Ayub-Ferreira

S.M.

Avila

M.S.

Brandao

S.M.

Cruz

F.D.

Santos

M.H.

Cruz

C.B.

Alves

M.S.

Issa

V.S.

Guimaraes

G.V.

Cunha-Neto

and Bocchi

E.A.

, Circulating miR-1 as a potential biomarker of doxorubicin-induced cardiotoxicity in breast cancer patients, Oncotarget 8 (2017), 6994–7002.

48.

Sassi

Avramopoulos

Ramanujam

Gruter

Werfel

Giosele

Brunner

A.D.

Esfandyari

Papadopoulou

A.S.

De Strooper

Hubner

Kumarswamy

Thum

Yin

Mayr

Laggerbauer

and Engelhardt

, Cardiac myocyte miR-29 promotes pathological remodeling of the heart by activating Wnt signaling, Nat Commun 8 (2017), 1614.

49.

Zhu

Liu

and Yu

, MiR-200a mediates protection of thymosin beta 4 in cardiac microvascular endothelial cells as a novel mechanism under hypoxia-reoxygenation injury, J Cell Biochem (2017).

50.

Liao

Y.F.

Feng

Chen

L.L.

Zeng

T.S.

and Hu

L.J.

, Coronary heart disease risk equivalence in diabetes and arterial diseases characterized by endothelial function and endothelial progenitor cell, J Diabetes Complications 28 (2014), 214–8.

51.

Yla-Herttuala

, Cardiovascular gene therapy with vascular endothelial growth factors, Gene 525 (2013), 217–9.

Correlation of circulating pro-angiogenic miRNAs with cardiotoxicity induced by epirubicin/cyclophosphamide followed by docetaxel in patients with breast cancer

Abstract

OBJECTIVE:

METHODS:

RESULTS:

CONLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Patients

2.2 Treatment

2.3 Candidate miRNAs selection

2.4 Samples collection

Table 1 Primers information in this study

2.7 Detection of cTnI and NT-proBNP

Table 3 Baseline relative expressions of pro-angiogenic miRNAs (C0)

3. Results

3.1 Study flow

3.2 Baseline characteristics of BC patients

3.4 Correlations of 14 pro-angiogenic miRNAs with the levels of cardiotoxicity biomarkers

3.5 Cardiotoxicity incidence among BC patients received EC-D neoadjuvant chemotherapy

3.7 Predicting value of the combination of miR-17-5p and miR-20a for cardiotoxicity risk

Table 4 Assessment of cardiotoxicity

4. Discussion

Footnotes

Conflict of interest

Supplementary data

References

Table 1
Primers information in this study

Table 3
Baseline relative expressions of pro-angiogenic miRNAs (C0)

Table 4
Assessment of cardiotoxicity