Cardiotoxicity in childhood cancer survivors: a problem with long-term consequences in need of early detection and prevention.

I n this issue of Pediatric Blood & Cancer, Orgel et al. [1] report the prevalence of early post-treatment cardiac dysfunction in a multi-center, retrospective cohort of childhood acute myeloid leukemia (AML) survivors. After a median follow-up of 13 months from the end of treatment, patients demonstrated an 8.7% decline in left ventricular shortening fraction (LVSF) from baseline to the most recent post-treatment echocardiogram. Further, declines in LVSF greater than 10% were found in more than half of the cohort [1]. These declines were significantly associated with cumulative anthracycline dose, body-mass index percentile, and ethnicity. The authors conclude that children treated with an anthracycline-intensive regimen are at risk of late-onset cardiomyopathy, should be monitored with echocardiograms annually, and should be considered for cardioprotective strategies [1]. Cumulative anthracycline dose, age at diagnosis, gender, and length of follow-up, are all associated with late anthracyclineassociated cardiac dysfunction [2]. Although the anthracyclines used in this study were daunorubicin or mitoxantrone (mg/m), the authors used doxorubicin-equivalent doses to calculate anthracycline exposure, resulting in a mean cumulative anthracycline dose of 339 14mg/m. The findings of this study support that anthracycline-induced cardiotoxicity is dose-dependent, such that every 100mg/m increase in cumulative anthracycline dose was accompanied by a percent decrease of 7.4% (95% CI 11.1 to 3.7) in LVSF. Further, both this study and others [3] report declines in cardiac function even in children who received lower cumulative doses, suggesting that there is no risk-free dose of anthracycline. Shortening fraction is a measure of overall left ventricular (LV) systolic performance, which is influenced by heart rate, preload (initial stretching of the cardiomyocytes prior to contraction), afterload (the ventricular wall stress during ejection), and myocardial contractility. A decrease in LVSF at the end of anthracycline therapy is not uncommon in childhood cancer survivors, and is associated with decreased LVSF 11 years after diagnosis [3]. However, a more thorough evaluation of cardiac structure and function might include additional echocardiographic measures, such as LV mass and end-diastolic posterior wall thickness, which are also affected by anthracycline therapy. For example, LV mass and wall thickness are significantly reduced at 3, 6, and 8 years after diagnosis [4]. Altogether, these findings suggest there is a progressive and persistent deterioration of heart structure and function with time. While researchers and clinicians agree that there is a need for established guidelines to monitor cardiac function in these children, there is still a lack of evidence-based research determining the most effective mode and frequency of serial monitoring. Echocardiography, a minimally invasive method of measuring cardiac structure and function, is a useful and common modality used to monitor these children, but has limited sensitivity and specificity. The Children’s Oncology Group’s Long-Term Follow-Up Guidelines [5], referred to by Orgel et al., are based on recommendations from their core committee and is intended to standardize follow-up care in survivors of childhood cancers. They recommend echocardiograms or MUGA scans in varying frequencies depending on a child’s age at treatment, radiation exposure to the heart, and anthracycline dose. Further evidence-based research is required to validate these guidelines. Although they do not include gender, all of the other previously mentioned variables are generally the most common risk factors of cardiotoxicity. Given the varying frequency and severity of cardiotoxicity even among children receiving the same dose, researchers suspect there are still unknown risk factors that may influence cardiotoxicity, such as genetics, and as Orgel et al. [1] showed, body-mass index and ethnicity. Although echocardiographic measures help assess heart structure and function during therapy, it lacks the sensitivity and specificity to detect early cardiac injury and the ability to distinguish between short-term depressions in LV function and myocardial injury with late cardiac implications. Biochemical markers of cardiac injury and cardiomyopathy, such as cardiac troponin T (cTnT) and N-terminal pro-brain natriuretic peptide (NT-proBNP), respectively, may also be useful in monitoring cardiotoxicity. Although not yet validated for clinical utility, elevations of these biomarkers during treatment are associated with abnormal echocardiographic findings 4 years after diagnosis [6]. As noted by Orgel et al., further efforts are needed to determine the validity and cost-effectiveness of using cardiac biomarkers and other screening models to predict cardiotoxicity [7]. The challenge lies in the ability to prospectively follow these children long enough to where sufficient vital status end points can be reached to reliably quantify the relationship between markers of cardiac injury and outcome.