In-Depth Insight Into the Mechanisms of Cardiac Dysfunction in Patients With Childhood Cancer After Anthracycline Treatment Using Layer-Specific Strain Analysis

Abstract

Background: Anthracycline cardiotoxicity affects clinical outcomes, and its early detection using methods that rely on conventional echocardiography, such as left ventricular ejection fraction (LVEF) is difficult. This study aimed to evaluate the characteristics and the differences in cardiac dysfunction among childhood cancer survivors in 3 age groups using layer-specific strain analysis in a wide age range.

Methods and Results: The 56 patients (median age: 15 [range: 6.8–40.2] years) who had been treated with anthracycline for childhood cancer were divided into 3 age groups (C1: 6–12 years, C2: 13–19 years, C3: 20–40 years) after anthracycline treatment, and 72 controls of similar ages were divided into 3 corresponding groups (N1, N2, and N3). Layer-specific longitudinal strain (LS) and circumferential strain (CS) of 3 myocardial layers (endocardium, midmyocardium, and epicardium) were determined using echocardiography. Myocardial damage had not occurred yet in C1. Endocardial CS at the basal level was less in C2 than in N2. Endocardial CS at all levels and midmyocardial CS at the basal and papillary levels were lower in C3 than in N3. LVEF and LS were not significantly different between patients and controls.

Conclusions: Among survivors of childhood cancer, impaired myocardial deformation starts in adolescence and extends from the endocardium towards the epicardium and from the base towards the apex with age. These findings are a novel insight into the time course of anthracycline cardiotoxicity.

The widespread use of anthracycline agents in chemotherapy for childhood cancer has significantly improved survival rates.^1,2 However, improved prognosis is accompanied by long-term health problems resulting from chemotherapy. Clinically, dose-dependent cardiotoxicity is the most prominent adverse effect of anthracycline chemotherapy^3,4 and has become the leading cause of long-term morbidity and early mortality.^3,5–7 The risk of congestive heart failure because of anthracycline treatment increases over time. A previous study estimated a morbidity rate of approximately 5% after 15 years of anthracycline treatment.⁵ Once anthracycline-induced heart failure occurs, the 2-year mortality rate is 60%.

Editorial p 648

Therefore, early detection of cardiotoxicity followed by prompt treatment is important. However, detection using methods that rely on conventional echocardiography, such as left ventricular ejection fraction (LVEF), is difficult. The parameters may remain relatively normal until there is advanced myocardial damage.^1,8,9 Following a decrease in LVEF, cardiac function hardly responds to the usual treatment for heart failure, such as β-blockers, and cardiovascular events increase in number.^10,11 Negishi et al reported that global longitudinal strain (LS) significantly improves in patients administered β-blockers, but decreases after chemotherapy.¹² Several recent studies have shown that assessment of myocardial deformation and rotation may allow for detection of early subclinical ventricular dysfunction in children after anthracycline therapy despite normal LVEF.^1,13,14

Layer-specific myocardial deformation may occur in the subendocardial layer of the ventricle in childhood cancer survivors because it is particularly sensitive to anthracycline-induced damage.^1,15 The recently introduced speckle-tracking echocardiographic technology has allowed non-invasive bedside assessment of layer-specific myocardial deformation.^16,17 This new sensitive indicator has been reported to be highly effective in detecting cardiac dysfunction in various cardiac diseases.^1,18–20 Yu et al showed that impairment of subendocardial circumferential deformation occurs in anthracycline-treated childhood cancer survivors.¹ However, the pathomechanisms and the time course of the impairment of LV myocardial deformation in childhood cancer survivors remain unclear and are hence important to clarify because they can be potential novel indicators for detecting cardiac dysfunction with high sensitivity. In this study, we aimed to evaluate the characteristics and the differences of cardiac dysfunction among 3 age groups of childhood cancer survivors treated with anthracycline using layer-specific strain analysis over a wide age range.

Methods

Study Population

In this prospective study, we recruited 56 patients (age, 6–40 years) with childhood cancer from Juntendo University. The patients had received chemotherapy consisting of anthracycline for at least 1 year before the study. The following data were collected from the medical records: diagnosis, date of completion of chemotherapy and duration of follow-up, cumulative anthracycline dose, cardiac and/or abdominal irradiation, symptoms and signs of heart failure, and cardiac medications. Doses of anthracycline derivative, such as daunorubicin, were converted into a doxorubicin dose. LV strain analysis was performed in all patients. Age-matched healthy individuals were recruited from Juntendo University and Shizuoka Children’s Hospital as normal controls for LV myocardial mechanical analysis; these were either healthy volunteers or children undergoing echocardiography for the evaluation of innocent murmurs. They had no history of cardiovascular disease and showed normal sinus rhythm on ECG and normal findings on echocardiography. All participants or their guardians provided written informed consent, as established by the Institutional Review Board of Juntendo University and Shizuoka Children’s Hospital. In children, cardiac function dramatically changes with age. Therefore, to more clearly identify the characteristics of myocardial deformation in childhood cancer, patients were divided into 3 age groups: C1, <13 years old; C2, 13–19 years old; and C3, 20–40 years old. Normal controls were also divided into the same 3 age groups (N1, N2, and N3).

Echocardiography

Echocardiography was performed using a Vivid E9 ultrasound system (GE Healthcare, Milwaukee, WI, USA) with an M5S or 6S probe as appropriate for patient size. Images were optimized for gain, compression, depth, and sector width and acquired at frame rates of 70–125 frames/s. Apical 4- and 2-chamber views and parasternal short-axis views at the basal, papillary, and apical ventricular levels were acquired. In each plane, images from 3 consecutive cardiac cycles were acquired during a breath hold at end-expiration, if possible. For younger children, we selected 3 cardiac cycles at end-expiration on the respiratory trace.

The mitral inflow E-wave, A-wave, E/A ratio, and Tei index were measured. LV end-diastolic volume (LVEDV) and LV end-systolic volume (LVESV) were calculated from the apical 4- and 2-chamber views using the modified Simpson rule. LVEF was calculated as (LVEDV−LVESV)/LVEDV. The LV diastolic function was quantified using the ratio between the E-wave velocity of the pulsed-wave Doppler mitral flow and the early diastolic velocity of the septum and LV free wall at the mitral annulus level (e′ wave) on tissue Doppler imaging.

LV Deformation Analysis

Analysis was performed offline with the aid of a commercially available software package (EchoPAC 113 1.0; GE Vingmed Ultrasound AS, Horton, Norway). Strain analysis was performed by 2 observers (K.Y. and T.I.) who were unaware of the clinical data. Strain was measured using 2D speckle-tracking echocardiography. The LV endocardium was manually traced, and the region of interest was manually adjusted to the LV wall thickness. The software tracks myocardial motion through the cardiac cycle, calculating strain from the echogenic speckles in the B-mode image. From the basal, papillary, and apical short-axis views and the apical 4-chamber view, 1 cardiac cycle was selected for subsequent analysis of circumferential strain (CS) and LS. The system used for this study allows calculation of mean strain values for the total wall thickness and for 3 separate myocardial layers (endocardium, midmyocardium, and epicardium), as described previously (Figure 1).²¹ All data were measured at least 3 times, and the averages are reported.

Figure 1.

Examples of typical LV layer-specific strain curves for papillary endocardial, midmyocardial, and epicardial circumferential strain in normal controls (A) and patients with childhood cancer after anthracycline treatment (B). Negative end-systolic strain represents myocardial shortening during systole. Progressive decrease in systolic strain from the endocardial to epicardial layer was observed in both patients and controls; maximal difference was found at the time of peak strain. The basic form of the strain curves does not differ significantly between the controls and patients, but the peak strain shows a significant difference. The timing of peak strain in the controls and patients is near closure of the aortic valve (AoV), decreasing rapidly during early diastole in each layer. LV, left ventricle; MV, mitral valve.

Statistical Analysis

Normally distributed continuous variables are expressed as mean±standard deviation (SD); non-normally distributed variables, as median (range). All group differences were assessed using a one-factor analysis of variance with a post-hoc comparison using the Tukey-Kramer method for normally distributed data or the Steel-Dwass test for non-normally distributed data. After evaluating the data for normality, the correlations between age at initiation and completion of anthracycline treatment, duration after completion of anthracycline treatment, cumulative anthracycline dose, and radiation dose and each peak strain were evaluated using either the Spearman’s correlation coefficient for data with non-normal distributions or Pearson’s correlation coefficient for data with normal distributions, which were expressed as ρ or r, respectively.

Intra- and interobserver agreements for the LV layer-specific strain were calculated using the Bland-Altman approach, including the calculation of mean bias (average difference between measurements), and the lower and upper limits of agreement (95% limits of agreement of mean bias) in 6 randomly selected patients and 6 controls at more than 2 months apart. The coefficient of variation was also determined (i.e., the SD of the difference of paired samples divided by the average of the paired samples). Statistical analyses were performed using JMP software (version 9.0.2; SAS Institute Inc., Cary, NC, USA). A P-value <0.05 was considered statistically significant.

Results

Feasibility

Table 1 and Table 2 show the characteristics of the study participants. The median age of the 56 patients (25 males) at echocardiography was 15 years (range, 6.8–40.2 years). The diagnoses were acute lymphoblastic leukemia in 24 patients, acute myeloid leukemia in 3, Hodgkin’s lymphoma in 1, non-Hodgkin’s lymphoma in 11, Wilms tumor in 8, hepatoblastoma in 2, neuroblastoma in 2, Ewing’s sarcoma in 3, and rhabdomyosarcoma in 2. All the patients had received anthracycline chemotherapy between March 1984 and December 2014. The median cumulative anthracycline dose was 126 mg/m² (range, 18–659 mg/m²), with only 5 patients receiving more than 300 mg/m². The median age of the patients was 4 years (range, 0.2–23.2 years) at initiation of anthracycline treatment and 7 years (range, 0.7–24.8 years) at completion of anthracycline treatment. The patients were studied at a median of 7.8 years (range, 1.1–26.2 years) after completion of anthracycline treatment. As shown in Table 2, the median duration after completion of anthracycline treatment was significantly longer in the C3 group than in the C2 group, and in C2 group than in the C1 group. Significant differences between patient groups were not found for the radiation dose. Only 2 patients were treated with ≥35 Gy. None had overt clinical heart failure at the time of the study. The median age of the 72 controls (36 males) at echocardiography was 14 years (range, 6.3–41.1 years, P=0.90). Body weight (45.6±14.0 vs. 46.4±16.2 kg, P=0.79) and body mass index (19.7±3.8 vs. 19.5±3.4 kg/m², P=0.70) were similar between patients and controls.

Table 1. Baseline Characteristics of Patients Surviving Childhood Cancer After Anthracycline Treatment

	C1	C2	C3	N1	N2	N3
Number (males)	21 (7)	22 (9)	13 (9)	26 (10)	26 (13)	20 (13)
Age (years)	10.5±1.6††,‡‡	16.3±2.3	28.0±6.3††	9.2±2.1††,‡‡	15.0±1.8	28.3±5.6††
HR (beats/min)	70.3±10.1	67.6±10.5	66.4±12.4	72.5±6.1††,‡	62.2±6.1	62.7±3.8
Height (m)	1.39±0.12††,‡‡	1.55±0.10	1.62±0.09	1.32±0.14††,‡‡	1.58±0.08	1.68±0.08†
Weight (kg)	35.1±10.0††,‡‡	47.7±11.3	59.1±10.2†	31.3±10.8††,‡‡	49.1±9.3	62.5±10.8††
BMI (kg/m2)	17.9±2.9‡‡	19.7±3.4	22.7±4.1	17.4±3.0‡‡	19.6±3.1	22.0±2.3
SBP (mmHg)	103±11	102±10	113±14	97±10†,‡‡	107±10	123±15††
DBP (mmHg)	56±6	57±5	64±14	52±8‡‡	57±9	71±8††
LVEDV/BSA (mL/m2)	45.8±6.6	46.4±8.3	47.2±8.5	45.6±12.6	53.6±15.6	47.7±10.6
LVESV/BSA (mL/m2)	16.5±2.8	16.1±3.6	16.5±3.8	16.1±5.8	18.9±6.6	18.1±5.4
LVEF (%)	63.4±4.6	62.0±7.1	63.7±4.8	65.2±4.3	65.2±4.2	63.3±6.0
LVEF <55% (n)	1	4	1	0	0	1
E-wave velocity (cm/s)	102±11‡‡	94±19	78±10†	105±15‡‡	100±18	88±18
E/A ratio	2.24±0.65	1.78±0.85*	1.85±0.53	2.44±0.59‡	2.33±0.69	1.80±0.41
Tei index	0.29±0.08†	0.38±0.10*	0.36±0.11	0.30±0.07	0.30±0.08	0.34±0.12
Septal e′ (cm/s)	12.6±1.5**	13.5±3.1	11.3±1.5	15.1±2.5	14.9±2.6	13.5±2.1
LV FW e′ (cm/s)	18.3±3.7	17.5±3.5	15.4±3.1	19.5±3.1‡	19.0±2.5	16.8±2.7
Septal E/e′	8.22±0.91	7.15±1.35	6.96±1.16	7.15±1.52	6.94±1.43	6.67±1.65
LV FW E/e′	5.81±1.25	5.49±1.00	5.21±1.09	5.50±1.23	5.35±1.05	5.31±1.02

*Patients after anthracycline treatment vs. controls in corresponding age groups, P<0.05. **Patients after anthracycline treatment vs. controls in corresponding age groups, P<0.01. ^†Significant difference between next to age groups, P<0.05. ^††Significant difference between next to age groups, P<0.01. ^‡Significant difference between C1 and C3 or N1 and N3, P<0.05. ^‡‡Significant difference between C1 and C3 or N1 and N3, P<0.01. C1: 6–12 years old, C2: 13–19 years old, C3: 20–40 years old, N1–N3: control groups of similar ages. BMI, body mass index; BSA, body surface area; DBP, diastolic blood pressure; FW, free wall; HR, heart rate; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LVESV, left ventricular end-systolic volume; SBP, systolic blood pressure.

Table 2. Characteristics of Study Patients Surviving Childhood Cancer After Anthracycline Treatment

Characteristic	C1	C2	C3
Age at initiation of anthracycline treatment (years)	3.7±2.4††	5.6±4.6	10.0±7.3*
Age at completion of anthracycline treatment (years)	5.0±2.5††	7.1±4.3	12.6±6.3**
Duration after completion of anthracycline treatment (years)	5.4±3.0††,*	9.2±3.6	15.4±8.2**
Cumulative anthracycline dose (mg/m2)	125±86	145±90	215±174
<120 (n)	11	10	5
120–300 (n)	9	11	5
>300 (n)	1	1	3
Radiation (n)	5	5	3

Data are expressed as mean±SD. *Significant difference between next to age groups, P<0.05. **Significant difference between next to age groups, P<0.01. ^††Significant difference between C1 and C3, P<0.01.

No significant differences were found between patients and controls for LVEF, LVEDV/body surface area (BSA), LVESV/BSA, E-wave velocity, e′ at the LV free wall, and E/e′ at the septum and LV free walls (Table 1). Only 6 patients showed a decreased LVEF (range, 50.5–54.5%). All endocardial CS values, excluding the basal CS value of 1 patient, were lower than the mean value of the corresponding control age group in 5 of these 6 patients. e′ at the septum was significantly lower in the C1 group than in the N1 group. E/A was significantly lower in the C2 group than in the N2 group. Tei index was significantly higher in the C2 group than in the N2 group.

LV Strain Patterns

Basal CS Peak endocardial basal CS was decreased in the C2 and C3 groups compared with the N2 and N3 groups (Figure 2, Table S1). Peak midmyocardial basal CS was decreased in the C3 group compared with the N3 group. Peak epicardial basal CS was not significantly different between the groups. Significant differences were not found in the basal CS in the C1 group compared with the N1 group.

Figure 2.

Comparison of the LV layer-specific strain analysis of (A) basal circumferential strain (CS), (B) papillary CS, (C) apical CS, and (D) longitudinal strain (LS). Blue, normal controls (N1–N3); red, patients (C1–C3) who survived childhood cancer after anthracycline treatment. *P<0.05. **P<0.01. C1: 6–12 years old, C2: 13–19 years old, C3: 20–40 years old; N1–N3, normal control groups of similar age.

Papillary CS Peak endocardial and midmyocardial papillary CS values were decreased in the C3 group compared with the N3 group (Figure 2, Table S1). Peak epicardial papillary CS was not significantly different between the groups. No significant differences were noted in the papillary CS in the C1 and C2 groups compared with the N1 and N2 groups.

Apical CS Peak endocardial apical CS was decreased in the C3 group compared with the N3 group (Figure 2, Table S1). Peak midmyocardial and epicardial apical CS values were not significantly different between the groups. Significant differences were not found in the apical CS in the C1 and C2 groups compared with the N1 and N2 groups.

LS Peak LS in all 3 layers was not significantly different between patients and controls (Figure 2, Table S1).

Relationship With Cumulative Anthracycline Dose

The LS value in all 3 layers correlated significantly with the cumulative anthracycline dose (endocardial, ρ=0.35, P=0.008; midmyocardial, ρ=0.34, P=0.010; epicardial, ρ=0.34, P=0.011). Endocardial and midmyocardial papillary CS values also correlated significantly with the cumulative anthracycline dose (ρ=0.36, P=0.007, and ρ=0.27, P=0.047, respectively). No correlations were noted for age at initiation and completion of anthracycline treatment, duration after completion of anthracycline treatment, and radiation dose.

Reproducibility

Table 3 presents the results for intra- and interobserver variability. Important differences were not observed in the variability scores of the endocardial, midmyocardial, and epicardial CS values at the basal, papillary, and apical levels or for LS.

Table 3. Intraobserver and Interobserver Variability of Layer-Specific Strain

Variable	Bias	LLA	ULA	CV
Intraobserver
Endocardial basal CS	−0.025	−1.319	1.269	2.90
Midmyocardial basal CS	0.344	−0.535	1.223	2.52
Epicardial basal CS	−0.082	−2.074	1.910	7.82
Endocardial papillary CS	−0.064	−1.349	1.221	2.77
Midmyocardial papillary CS	0.355	−0.935	1.644	3.75
Epicardial papillary CS	−0.346	−2.218	1.526	9.03
Endocardial apical CS	−0.319	−1.946	1.308	3.30
Midmyocardial apical CS	0.357	−2.986	3.700	8.88
Epicardial apical CS	−0.256	−2.303	1.791	8.79
Endocardial LS	−0.022	−1.417	1.373	3.62
Midmyocardial LS	0.307	−0.330	0.944	1.93
Epicardial LS	−0.183	−1.331	0.964	3.65
Interobserver
Endocardial basal CS	0.011	−1.595	1.617	3.61
Midmyocardial basal CS	0.004	−0.945	0.952	2.69
Epicardial basal CS	−0.168	−1.840	1.505	6.54
Endocardial papillary CS	−0.222	−1.514	1.070	2.77
Midmyocardial papillary CS	−0.107	−1.111	0.896	2.88
Epicardial papillary CS	−0.075	−1.078	0.928	4.90
Endocardial apical CS	−0.081	−1.683	1.522	3.26
Midmyocardial apical CS	−0.060	−1.733	1.613	4.40
Epicardial apical CS	0.180	−1.150	1.511	5.82
Endocardial LS	0.453	−0.620	1.526	2.82
Midmyocardial LS	0.038	−0.670	0.746	2.13
Epicardial LS	0.236	−0.522	0.994	2.44

CS, circumferential strain; CV, coefficient of variation; LLA, 95% lower limit of agreement; LS, longitudinal strain; ULA, 95% upper limit of agreement.

Discussion

To our knowledge, the present study is the first to use layer-specific strain analysis to examine the characteristics and the differences in cardiac dysfunction among 3 age groups of survivors of childhood cancer after anthracycline treatment. There were 3 main findings of this study (Figure 3). First, despite the wide age range at treatment and varying doses of anthracycline, abnormal myocardial deformation did not occur in childhood and impairment of the endocardial CS at the basal level in adolescent patients was the initial abnormality. Second, when focused on the same plane, impairment of LV peak strain began with the endocardial CS and subsequently involved the epicardial CS in young adult patients. Third, when focused on each of the 3 short-axis planes, impairment of CS began at the basal level in adolescent patients and extended towards the apex in young adult patients.

Figure 3.

Time course of impairment of LV myocardial deformation in patients who survived childhood cancer after anthracycline treatment. Impaired myocardial deformation starts in the C2 group and extends from the endocardial CS to the epicardial CS and from the base towards the apex in the C3 group. C1: Impairment of myocardial deformation has not occurred yet. C2: Only the endocardial basal CS (red circle) is decreased. C3: Endocardial CS at all levels (red circle) and midmyocardial CS at the basal and papillary levels (purple circle) is decreased.

Conventional Parameters

LVEF was used as the conventional criteria of cardiotoxicity; a decline in LVEF of at least 5% to less than 55% with accompanying signs or symptoms of congestive heart failure, or a decline in LVEF of at least 10% to less than 55% without accompanying signs or symptoms.²² However, in this study, only 6 (10%) patients showed decreased LVEF at the time of the study and all endocardial CS values, excluding the basal CS of 1 patient, decreased in 5 of the 6 patients, which suggested that LVEF was not a sensitive indicator. Furthermore, no significant difference was found between patients and controls. The present findings suggested that impaired myocardial deformation increases with age from adolescent to young adult, despite LVEF being within the normal limits, in patients who survive childhood cancer after anthracycline treatment.

Impairment of CS Before LS

Armstrong et al reported that LS is a more sensitive indicator of cardiac dysfunction than EF after anthracycline treatment.²³ Furthermore, LS is recommended for the detection of cardiac dysfunction in cancer survivors per the established guideline.²⁴ However, in our study the endocardial CS in adolescent patients decreased compared with normal controls despite LS values within the normal limits and normal conventional diastolic parameters.

Yu et al suggested that in contrast to LS, endocardial CS is more sensitive for early detection of LV systolic dysfunction in these long-term cancer survivors.¹ In our study, a significant difference was not noted in layer-specific LS between patients and controls, consistent with previous studies.^13,25

The cardiac wall in humans has a well-defined distribution of fibers, with their angles varying from approximately 60° (in the circumferential direction) at the inner surface to approximately −60° at the outer surface.²⁶ Longitudinally oriented fibers exist in both the endocardial and epicardial regions of the wall. The angle of fibers at 15% inside the LV endocardium is only 20°, which suggests that damage to the inner layer may affect both the longitudinally and circumferentially oriented fibers.²⁶ When myocardial damage occurs only in the endocardial layers, LS remains stable until the damage approaches the epicardial layers,²⁰ because deformation of each myocardial layer is dependent on active function within the layer and passive motion from adjacent layers.²⁷ These findings suggest that contraction impairment will affect the CS more than it does LS, particularly in younger patients when myocardial damage is not as severe. Therefore, both CS and LS could decrease in older patients who have a longer duration since completion of anthracycline treatment, in whom myocardial damage would become severe over time.⁵

Impairment of Endocardial CS Before Epicardial CS

Several studies have suggested a vulnerability of the endocardial layer to anthracycline cardiotoxicity from the view point of pathology. Progressive vacuolization of the myocardial fibers, leading to severe myocytolysis, involving mainly the subendocardium of the ventricular walls and the interventricular septum has been demonstrated in animal models of anthracycline cardiotoxicity.²⁸ Furthermore, a rabbit model of doxorubicin cardiotoxicity demonstrated that microscopic tissue damage occurs mainly in the subendocardial and intramural areas.¹⁵ Clinically, subendocardial late gadolinium enhancement suggestive of myocardial fibrosis has been reported in patients with anthracycline cardiotoxicity following treatment for Ewing’s sarcoma.²⁹

Yu et al showed that subendocardial circumferential deformation is the most sensitive indicator of cardiac dysfunction in anthracycline-treated survivors of childhood cancer.¹ Sensitivity of the subendocardial layer to anthracycline damage probably accounts for previous and current findings of preferentially impaired endocardial CS. The findings of our study further suggest that the impairment of endocardial CS does not occur in childhood, but rather begins in adolescence and gradually affects the epicardial CS over time. In a rabbit model of diabetes mellitus, Qiao et al showed that CS and LS decrease from the endocardium to the midmyocardium to the epicardium over time.³⁰ In patients with ischemia, not only the endocardial CS but also epicardial CS decreases as the disease progresses.^19,27 Furthermore, we previously reported on the time course of LV dysfunction in patients with repaired tetralogy of Fallot: impaired myocardial deformation increases with age, extending from the endocardium to the epicardium.¹⁸ These findings corroborate the extension of myocardial deformation towards the epicardium and the apex in the present study.

Impairment of Basal CS Before Apical CS

Streeter et al reported that the rate of circumferentially to longitudinally oriented fibers is 10:1, with this ratio increasing towards the base and decreasing towards the apex.²⁶ This suggests that the effect of impaired contraction in the circumferential direction is more prominent at the basal level than at the apex, as the base constitutes a considerable fraction of the LV wall. Furthermore, Laplace’s law supports the idea that the larger cavity radius at the base than the apex of the LV is more affected by interventricular pressure³¹ in the heart with subclinical myocardial damage. These findings support our results that impairment in the basal and papillary CS occurs before that of the apical CS.

Correlation Between Cardiotoxicity and Cumulative Anthracycline Dose and Age-Associated Change

A previous study suggested that patients treated with a cumulative anthracycline dose higher than 300 mg/m² are at highest risk for clinical heart failure and that the risk increases with cumulative dose.³ However, van der Pal et al⁵ reported an exponential relationship between the cumulative anthracycline dose and the hazard ratio of developing cardiac events, even when the cumulative dose was less than 300 mg/m². Among the patients who were treated with both anthracycline agents and cardiac irradiation, congestive heart failure had already occurred at a cumulative anthracycline dose of only 120 mg/m². Furthermore, other studies report that patients exhibited morphologic changes on cardiac biopsy with a cumulative dose of anthracycline as low as 200 mg/m².^32,33 Ganame et al reported a reduction in both the LS and the strain rate of the LV myocardium in children even after low-dose anthracycline therapy.³⁴ The median cumulative anthracycline dose of all patients in our study was less than 200 mg/m², and only 5 patients were treated with more than 300 mg/m². We found correlations between papillary CS and LS values in all 3 layers, excluding the epicardial papillary CS, and the anthracycline dose in the study patients. This finding suggested greater sensitivity of the present layer-specific approach in detecting subtle myocardial damage even after low-dose anthracycline therapy. However, the correlations may be attributable to the heterogeneity of the study population and the limited sample size. Increasing the number of subjects may result in correlation of other parameters with LV deformation parameters.

An adequate time course of cardiac events is also an important factor in following up patients. van der Pal et al’s long follow-up showed that the cumulative incidence of cardiovascular events continued to increase for 30 years and they suggested that 1 in 8 developed severe heart disease after 30 years.⁵ In our study, the median duration from completion of chemotherapy to the day of the echocardiography was longer in young adult group compared to children group. Therefore, we showed that impaired myocardial deformation progresses with age from children through adolescence to the young adult patient group, which suggests the timing of very early cardiac dysfunction in patients with childhood cancer after anthracycline treatment.

Clinical Implications

The present study showed the first evidence of preferential impairment of endocardial CS before epicardial CS and of basal CS before apical CS in patients with childhood cancer after anthracycline treatment. Endocardial CS at the basal level is considered to be a useful and sensitive indicator of cardiac dysfunction caused by anthracycline compared with conventional parameters. Future studies with larger number of patients may be able to evaluate a more accurate time course of the cardiac dysfunction and to detect it earlier in patients with childhood cancer after anthracycline treatment. These findings have important clinical implications and may allow for early therapeutic intervention, such as β-blockers,¹² for subclinical LV dysfunction to prevent LV dysfunction, and angiotensin-converting enzyme inhibitors³⁵ and angiotensin-receptor blockers to reduce wall stress and potentially improve the impairment of myocardial deformation. Furthermore, this parameter could allow better evaluation of the therapeutic effects of these medications.

Study Limitations

First, it included a small number of patients. Future studies with a larger sample size are necessary to provide a robust conclusion with regard to the time course of cardiac dysfunction in patients with childhood cancer after anthracycline treatment. Second, the design of the study was cross-sectional. A longitudinal follow-up study would have yielded a more accurate time course of cardiac dysfunction. Third, the variation in types of cancers, the wide age range during chemotherapy, the varying duration after completion of chemotherapy, and the cumulative anthracycline dose caused difficulties in analyzing the mechanisms of potential myocardial damage. Finally, the use of other cytotoxic drugs and radiotherapy also affected the results. In the future, by increasing the number of patients and classifying them by age at initiation of anthracycline therapy, whether or not treatment with other cytotoxic drugs and radiotherapy was given, and by type of disease, the mechanism of cardiotoxicity by anthracycline could be elucidated.

Conclusions

This study showed that impairment of endocardial circumferential deformation in adolescent patients was the initial cardiac abnormality in those who had survived childhood cancer treated with anthracycline, and this impairment then extended from the endocardium towards the epicardium and from the base towards the apex with age. The present findings are a novel insight into the characteristics and the time course of cardiotoxicity in these patients.

Acknowledgments

We thank the staff of Shizuoka Children’s Hospital for collecting the echocardiographic data of normal children and adults.

Names of Grants

None.

Supplementary Files

Supplementary File 1

Table S1. Correlation of layer-specific strain between patients (C1–C3) and controls (N1–N3)

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-17-0874

References

• 1.   Yu W, Li S, Chan GCF, Ha S, Wong SJ, Cheung YF. Transmural strain and rotation gradient in survivors of childhood cancers. Eur Heart J Card Img 2013; 14: 175–182.
• 2.   Gatta G, Capocaccia R, Coleman MP, Ries LA, Berrino F. Childhood cancer survival in Europe and the United States. Cancer 2002; 95: 1767–1772.
• 3.   Kremer LC, van Dalen EC, Offringa M, Ottenkamp J, Voûte PA. Anthracycline-induced clinical heart failure in a cohort of 607 children: Long-term follow-up study. J Clin Oncol 2001; 19: 191–196.
• 4.   Lefrak EA, Pitha J, Rosenheim S, Gottlieb JA. A clinicopathologic analysis of adriamycin cardiotoxicity. Cancer 1973; 32: 302–314.
• 5.   van der Pal HJ, van Dalen EC, van Delden E, van Dijk IW, Kok WE, Greskus RB, et al. High risk of symptomatic cardiac events in childhood cancer survivors. J Clin Oncol 2012; 30: 1429–1437.
• 6.   Reulen RC, Winter DL, Frobisher C, Lancashire ER, Stiller CA, Jenney ME, et al. Long-term cause-specific mortality among survivors of childhood cancer. JAMA 2010; 304: 172–179.
• 7.   van Dalen EC, van der Pal HJ, Kok WE, Caron HN, Kremer LC. Clinical heart failure in a cohort of children treated with anthracyclines: A long-term follow-up study. Eur J Cancer 2006; 42: 3191–3198.
• 8.   Felker GM, Thompson RE, Hare JM, Hruban RH, Clemetson DE, Howard DL, et al. Underlying causes and long-term survival in patients with initially unexplained cardiomyopathy. N Engl J Med 2000; 342: 1077–1084.
• 9.   Ewer MS, Lenihan DJ. Left ventricular ejection fraction and cardiotoxicity: Is our ear really to the ground? J Clin Oncol 2008; 26: 1201–1203.
• 10.   Cardinale D, Colombo A, Lamantia G, Colombo N, Civelli M, De Giacomi G, et al. Anthracycline-induced cardiomyopathy: Clinical relevance and response to pharmacologic therapy. J Am Coll Cardiol 2010; 55: 213–220.
• 11.   Cardinale D, Colombo A, Bacchiani G, Tedeschi I, Meroni CA, Veglia F, et al. Early detection of anthracycline cardiotoxicity and improvement with heart failure therapy. Circulation 2015; 131: 1981–1988.
• 12.   Negishi K, Negishi T, Haluska BA, Hare JL, Plana JC, Marwick TH. Use of speckle strain to assess left ventricular responses to cardiotoxic chemotherapy and cardioprotection. Eur Heart J Cardiovasc Imaging 2014; 15: 324–331.
• 13.   Cheung YF, Hong WJ, Chan GC, Wong SJ, Ha SY. Left ventricular myocardial deformation and mechanical dyssynchrony in children with normal ventricular shortening fraction after anthracycline therapy. Heart 2010; 96: 1137–1141.
• 14.   Ganame J, Claus P, Uyttebroeck A, Renard M, D’hooge J, Bijnens B, et al. Myocardial dysfunction late after low-dose anthracycline treatment in asymptomatic pediatric patients. J Am Soc Echocardiogr 2007; 20: 1351–1358.
• 15.   Llesuy SF, Milei J, Gonzalez Flecha BS, Boveris A. Myocardial damage induced by doxorubicins: Hydroperoxide-initiated chemiluminescence and morphology. Free Radic Biol Med 1990; 8: 259–264.
• 16.   Ishizu T, Seo Y, Enomoto Y, Sugimori H, Yamamoto M, Machino T, et al. Experimental validation of left ventricular transmural strain gradient with echocardiographic two-dimensional speckle tracking imaging. Eur J Echocardiogr 2010; 11: 377–385.
• 17.   Adamu U, Schmitz F, Becker M, Kelm M, Hoffmann R. Advanced speckle tracking echocardiography allowing a three-myocardial layer-specific analysis of deformation parameters. Eur J Echocardiogr 2009; 10: 303–308.
• 18.   Yamada M, Takahashi K, Kobayashi M, Yazaki K, Takayasu H, Akimoto K, et al. Mechanisms of left ventricular dysfunction assessed by layer-specific strain analysis in patients with repaired tetralogy of Fallot. Circ J 2017; 81: 846–854.
• 19.   Becker M, Ocklenburg C, Altiok E, Füting A, Balzer J, Krombach G, et al. Impact of infarct transmurality on layer-specific impairment of myocardial function: A myocardial deformation imaging study. Eur Heart J 2009; 30: 1467–1476.
• 20.   Hamada S, Schroeder J, Hoffmann R, Altiok E, Keszei A, Almalla M, et al. Prediction of outcomes in patients with chronic ischemic cardiomyopathy by layer-specific strain echocardiography: A proof of concept. J Am Soc Echocardiogr 2016; 29: 412–420.
• 21.   Leitman M, Lysyansky P, Sidenko S, Shir V, Peleg E, Binenbaum M, et al. Two-dimensional strain-a novel software for real-time quantitative echocardiographic assessment of myocardial function. J Am Soc Echocardiogr 2004; 17: 1021–1029.
• 22.   Seidman A, Hudis C, Pierri MK, Shak S, Paton V, Ashby M, et al. Cardiac dysfunction in the trastuzumab clinical trials experience. J Clin Oncol 2002; 20: 1215–1221.
• 23.   Armstrong GT, Joshi VM, Ness KK, Marwick TH, Zhang N, Srivastava D, et al. Comprehensive echocardiographic detection of treatment-related cardiac dysfunction in adult survivors of childhood cancer: Results from the St. Jude Lifetime Cohort Study. J Am Coll Cardiol 2015; 65: 2511–2522.
• 24.   Cheng KH, Handschumacher MD, Assuncao BMBL, Sebag IA, Halpern EF, Scherrer-Crosbie M. Contraction timing patterns in patients treated for breast cancer before and after anthracyclines therapy. J Am Soc Echocardiogr 2017; 30: 454–460.
• 25.   Dillenburg RF, Nathan P, Friedberg MK, Mertens L. Regional myocardial function in pediatric patients after anthracycline exposure: The effect of afterload on strain and strain rate measurements. Circulation 2010; 122: A20202.
• 26.   Streeter DD Jr, Spotnitz HM, Patel DP, Ross J Jr, Sonnenblick EH. Fiber orientation in the canine left ventricle during diastole and systole. Circ Res 1969; 24: 339–347.
• 27.   Sarvari SI, Haugaa KH, Zahid W, Bendz B, Aakhus S, Aaberge L, et al. Layer-specific quantification of myocardial deformation by strain echocardiography may reveal significant CAD in patients with non-ST-segment elevation acute coronary syndrome. J Am Coll Cardiol Imaging 2013; 6: 535–544.
• 28.   Milei J, Boveris A, Llesuy S, Molina HA, Storino R, Ortega D, et al. Amelioration of adriamycin-induced cardiotoxicity in rabbits by prenylamine and vitamins A and E. Am Heart J 1986; 111: 95–102.
• 29.   Perel RD, Slaughter RE, Strugnell WE. Subendocardial late gadolinium enhancement in two patients with anthracycline cardiotoxicity following treatment for Ewing’s sarcoma. J Cardiovasc Magn Reson 2006; 8: 789–791.
• 30.   Qiao YY, Zeng M, Li RJ, Leng ZT, Yang J, Yang Y. Layer-specific myocardial strain analysis: Investigation of regional deformation in a rabbit model of diabetes mellitus during different stages. Med Ultrason 2016; 18: 339–344.
• 31.   Attias D, Macron L, Dreyfus J, Monin JL, Brochet E, Lepage L, et al. Relationship between longitudinal strain and symptomatic status in aortic stenosis. J Am Soc Echocardiogr 2013; 26: 868–874.
• 32.   Vejpongsa P, Yeh ET. Prevention of anthracycline-induced cardiotoxicity: Challenges and opportunities. J Am Coll Cardiol 2014; 64: 938–945.
• 33.   Ewer MS, Ali MK, Mackay B, Wallace S, Valdivieso M, Legha SS, et al. A comparison of cardiac biopsy grades and ejection fraction estimations in patients receiving Adriamycin. J Clin Oncol 1984; 2: 112–117.
• 34.   Ganame J, Claus P, Uyttebroeck A, Renard M, D’hooge J, Bijnens B, et al. Myocardial dysfunction late after low-dose anthracycline treatment in asymptomatic pediatric patients. J Am Soc Echocardiogr 2007; 20: 1351–1358.
• 35.   Silber JH, Cnaan A, Clark BJ, Paridon SM, Chin AJ, Rychik J, et al. Enalapril to prevent cardiac function decline in long-term survivors of pediatric cancer exposed to anthracyclines. J Clin Oncol 2004; 22: 820–828.