Anthracycline-Induced Cardiothoxicity in Children Cancer Patients: An Imaging Study by Two-Dimensional Global Longitudinal Strain Using Automated Function Imaging Technique

It has been estimated that more than 14% of the children who were alive and adults with histories of cancer, were prone to cancer treatment-induced cardio toxicity in USA in 2014 (1). Anthracycline is an important drug for the treatment of pediatric and adult cancers. Despite its high effectiveness, some adverse and even life-threatening drug-induced side effects are potentially concerned. In this regard, cardiothoxicity is the most notorious and well-studied complication (2). The main pathophysiology of symptomatic heart failure is subclinical myocardial cell injury, which leads to an early asymptomatic decrement in the left ventricle ejection fraction (LVEF). This complication has been reported in numerous patients in one year following the Anthracycline treatment completion, however, late LVEF reduction has been revealed in only 2% of them; on the other hand, in about 82% of patients, the recovery from this event was achieved, especially by administrating cardio protective agents such as ACE inhibitors and beta-blockers over a mean period of 3 to 13 months. In this regard, cardiac monitoring of affected patients as well as those symptomatic or even asymptomatic patients under treated with cardio protective drugs is generally acknowledged (3-5). Left ventricular ejection fraction (LVEF) is used for cardiac function monitoring as routine (6), however, it has limited sensitivity (7, 8). It has been demonstrated that despite immediate diagnosis of LVEF decrement following the Anthracycline treatment, 36% could not recover normal EF in spite of therapeutic interventions (9). Some studies have shown that subtle and early changes of LV systolic dysfunction are detectable by myocardial deformation indices (strain and strain rate) before LVEF become impaired (10).

This study aims to investigate the value of using different methods (LVEF, E, E’, A and Tei index versus global longitudinal strain) for detection of early cardiac changes by measuring them before and one month after the Anthracycline administration in a cross-sectional cohort study.

This cross-sectional study was performed on 22 new diagnosed cancer patients aged 18 years or younger who referred to the children Medical center in Tehran in 2016 as the cases and also 22 age-matched healthy children who referred to the center due to any non-cardiac reasons as the control. The exclusion criteria included lack of consent to participate in research, using anthracycline in previous therapeutic regimen, inappropriate images in echocardiography, history of congenital heart defects, having left ventricular dysfunction of any reason before the start of treatment with anthracycline, and using cardiothoxic drugs before assessment. The baseline characteristics and medical history were collected by interviewing the parents or from the recorded files. All participants were assessed to determine the left ventricular functional indices before and also one month after starting the therapeutic regimen containing anthracycline. Echocardiography was conducted in a lateral decubitus position using a vivid E9 general electric ultrasound machine and a probe 5 seconds in 2, 3, and 4 chamber views. The frame rate was considered similar in both patients and healthy groups. The end of systole was considered as the time of aortic valve closure at apical long axis view. The region of interest was determined manually at the end of systole by determining the endocardial margin at apical and parasternal views. Manual balancing was performed in the presence of inappropriate automated tracking. Segmental and mean strain rates were determined automatically in all acoustic markers (six segments in each view). In total, the global longitudinal strain (averaging all segmental strain rates) was calculated. Diastolic function status was assessed through measurement of E’ parameter (early myocardial velocity) in lateral mitral annulus, TDI, and E velocity (early diastolic mitral inflow velocity). Left ventricular ejection fraction was assessed by M mode technique in four chamber views.

The SPSS version 22.0 for windows (SPSS Inc., Chicago, IL) was used for the statistical analysis, and P values of 0.05 or less were considered statistically significant.

In total, 4 patients diagnosed with cancer and 7 healthy children were excluded based on the exclusion criteria. In addition, 18 cancer patients who suffered from different types of hematologic cancers and 15 healthy individuals, as the controls, were included into the study. The mean age of participants in the case and control group was 7.55 ± 3.29 years and 5.73 ± 3.16 years respectively with no difference (P = 0.11). Also, 61.1% in the case group and 53.3% in control group were male, respectively (P = 0.65). Regarding types of cancers, the most common type was acute lymphoblastic leukemia (57.5%) followed by hutchins lymphoma (30%), acute myelocytic lymphoma (6.5%), and Burkitt lymphoma (6%). As shown in Table 1, among all parameters assessed by echocardiography, only the mitral annular peak systolic velocity (Sm) significantly reduced after administrating anthracycline compared to before that, however, the change in other parameters remained statistically insignificant. All echocardiographic data related to the case group after first dose administration of Anthracycline, showed no statistical meaningful differences from those before the therapy, except for Sm. Data for patients and the control group showed no statistical significant differences before the therapy.

Anthracycline is considered as the main treatment of the hematologic cancers (11-13). Symptomatic heart failure caused by administration of Anthracycline is occurred in 2% - 5% of patients and is more common in hematologic malignancies (14-17). According to these findings we chose children with hematologic malignancies administrated by Anthracycline for the chemotherapy regimen as our study group.

Studies assessing the influence of Anthracycline on myocardial cellular changes in experimental models have demonstrated several forms of cardiac injury leading to myocyte death. The evidences of this cellular damage can be detected by some echocardiographic and laboratory changes such as elevations in serum troponin. In total, development of cellular “sarcopenia” characterized by disruption of normal sarcomere structure is considered as a mechanism of myocyte injury, other than myocyte cell death, which seems to be the probable mechanism for anthracycline-induced cardiac injury (18). When the Anthracycline-induced heart failure becomes symptomatic, the survival will be less than 50% in the following year (19). It has been estimated that death, due to cardiac events, is more probable in childhood cancer survivors compared to normal population with the relative risk of 8.2 (7). Regarding the high mortality of heart failure caused by Anthracycline, early detection of cardiac changes and using preventable therapeutic strategies for decelerating this process becomes much more important.

Potential risk factors for Anthracycline-induced cardiac damages are: female sex, earlier age at the time of treatment, higher body fat content, concurrent radiation therapy, and higher cumulative dose of Anthracycline (20, 21).

Evaluation of left ventricular ejection fraction (LVEF), fractional shortening (FS), E’ parameter (early diastolic velocity), E parameter (trans mitral early diastolic filling velocity), A parameter (trans mitral late diastolic filling velocity) and mitral annular peak systolic velocity (Sm) by 2 dimensional (2D), and M-mode echocardiography, are the conventional methods for cardiac changes assessment. Recently, utilizing of cardiac strain and strain rate, for regional and global systolic dysfunction evaluation, has been considered. Studies suggest that LVEF and FS lack enough sensitivity for early cardiac changes detection (7, 8). On the other hand, recent studies show that cardiac strain decreases in both early and late cardiac changes caused by Anthracycline (22, 23). Some studies have shown that subtle and early changes of LV systolic dysfunction are detectable by myocardial deformation indices (strain and strain rate) before LVEF decreases (10, 24). Comparing the conventional methods with recent ones, our study aimed to determine the relative efficiency of these indices.

In a study done by Ali et al., 6% of treated patients suffered from cardiac adverse events after the use of anthracyclines that manifested by lowering LVEF and GLS (25). In a similar study by Pignatelli et al., abnormal global longitudinal peak systolic strain was detected in 60% of patients and 76% had abnormal peak circumferential strain compared to age-matched controls (26). As indicated by Corapcioglu et al., 48% of the patients had cardiac dysfunction following the anthracycline chemotherapy. Found by echocardiography, all dysfunctions were systolic, whereas 29% of the patients had diastolic and 38% of the patients had systolic dysfunction in a study by multigated radionuclide angiography (27). In another study by Ishii et al., the patients who received a low dose and those who received a moderate to high dose of Anthracycline had a significant difference in the Tei index (28). In another study by Agha et al., the LV developed a change in its longitudinal systolic function in the form of a reduction in MAPSE, TDI-derived systolic velocity of lateral mitral annulus, 2D-STE-derived global longitudinal strain, and 2D-STE-derived global longitudinal strain rate, following the chemotherapy (29).

Children affected by different types of cancer may experience significant adverse changes in the left ventricular functional and also structural parameters. These changes can be manifested by reducing the left ventricular ejection fraction, increasing left ventricular Tei index, and also impaired segmental wall motion as impaired left ventricular strain. As mentioned above, it is crucial to detect the early stages of the Anthracycline-induced cardiac changes. Our aim was to determine the immediate effect of the Anthracycline on cardiac function by using conventional and newer methods of echocardiography. Based on our findings, only the mitral annular peak systolic velocity (Sm) was significantly reduced after administrating Anthracycline and the changes in other parameters, including global longitudinal strain and global longitudinal strain rate, remained statistically insignificant. Our findings show that evaluating Sm is most useful for early detection of cardiac changes and recent indices have no privilege compared to more conventional ones. This may be due to our limitation of small cohort studies and also the short interval after the drug administration, although, this was related to the main goal of our study.