Longitudinal strain analysis for assessment of early cardiotoxicity during anthracycline treatment in childhood sarcoma: A single center experience

Abstract Background The growing population of long‐term childhood cancer survivors encounter a substantial burden of cardiovascular complications. The highest risk of cardiovascular complications is associated with exposure to anthracyclines and chest radiation. Longitudinal cardiovascular surveillance is recommended for childhood cancer patients; however, the optimal methods and timing are yet to be elucidated. Aims We aimed to investigate the feasibility of different echocardiographic methods to evaluate left ventricular systolic function in retrospective datasets, including left ventricular ejection fraction (LVEF), fractional shortening (FS), global longitudinal strain (GLS) and longitudinal strain (LS) as well as the incidence and timing of subclinical left ventricular dysfunction detected by these methods. Methods and Results A retrospective longitudinal study was performed with re‐analysis of longitudinal echocardiographic data, acquired during treatment and early follow‐up, including 41 pediatric sarcoma patients, aged 2.1–17.8 years at diagnosis, treated at Astrid Lindgren Children's Hospital, Stockholm, Sweden, during the period 2010–2021. All patients had received treatment according to protocols including high cumulative doxorubicin equivalent doses (≥250 mg/m2). In 68% of all 366 echocardiograms, LS analysis was feasible. Impaired LS values (<17%) was demonstrated in >40%, with concomitant impairment of either LVEF or FS in 20% and combined impairment of both LVEF and FS in <10%. Importantly, there were no cases of abnormal LVEF and FS without concomitant LS impairment. Conclusion Our findings demonstrate feasibility of LS in a majority of echocardiograms and a high incidence of impaired LS during anthracycline treatment for childhood sarcoma. We propose inclusion of LS in pediatric echocardiographic surveillance protocols.


| INTRODUCTION
In developed countries, the 5-year overall survival after childhood cancer is approaching 85%, and the majority of children diagnosed with cancer today will be long-term survivors. 1 Long-term childhood cancer survivors are at risk of treatment-related cardiovascular complications, associated with increased morbidity and mortality. [2][3][4] After high anthracycline doses during childhood (cumulative doxorubicinequivalent doses 5 of ≥250 mg/m 2 ), the 30-year prevalence of symptomatic heart failure is estimated to be almost 10%, which is comparable to the prevalence of heart failure in patients aged 75-84 years in the general population. 3,6,7 If radiotherapy involving the heart is added, the cumulative prevalence of heart failure rises to over 25%. 3 The overall risk of cardiovascular complications after anthracycline treatment is dose-dependent, with the highest risk in children treated with cumulative doxorubicin-equivalent 5

doses of
≥250 mg/m 2 , isolated chest radiotherapy ≥35 Gy, or with cumulative doxorubicin-equivalent doses ≥100 mg/m2 and chest radiotherapy ≥15 Gy, if combined together. 5,[8][9][10] Cardiovascular surveillance including echocardiography is recommended after anthracycline therapy and the frequency of surveillance is stratified by cumulative anthracycline doses. [8][9][10][11] However, there is a substantial variation in individual vulnerability 5,12,13 with no clear cut-off regarding safe doses of anthracyclines. 14,15 The aim of cardiac surveillance with echocardiography, during and after anthracycline therapy, is early detection of cardiac dysfunction since the prognosis is poor once clinical symptoms have evolved. 16 International cardiovascular surveillance guidelines address long-term follow-up, but there is no consensus regarding surveillance during and short after anthracycline treatment. [8][9][10] In children, standard methods to evaluate cardiac function by echocardiography include left ventricular ejection fraction (LVEF) and fractional shortening (FS). 17,18 Both LVEF and FS exhibit methodological limitations, including poor sensitivity for detection of early changes in cardiac function. 19,20 A significant decrease in LVEF, defined as a decrease from baseline of more than 10% to LVEF <53%, adds prognostic information regarding subsequent incidence of chronic heart failure both in children and adults during and following chemotherapy. [8][9][10]21,22 However, cardiotoxicity is commonly defined by abnormal FS (<0.28) in pediatric cancer treatment protocols. 8 Analysis of myocardial deformation that is, longitudinal strain (LS) or preferably global longitudinal strain (GLS), detect early signs of systolic myocardial dysfunction and are of prognostic value in adults during and after anthracycline treatment. 11,[22][23][24][25] Impairment in GLS of >15% from baseline or deformation measured by GLS less than 17% is defined as significant deterioration. 9,23,26 Previous studies indicate that deterioration in GLS, as well as in LS, during and early after anthracycline treatment may identify individuals at higher risk of subsequent cardiac dysfunction. 14,27,28 In children, the prognostic importance of GLS and LS warrants further clarification and these methods are not included in standard pediatric cancer surveillance protocols. 8,29 More sensitive and predictive markers of cardiovascular dysfunction would enable identification of patients who could benefit from more frequent surveillance, to enable timely interventions during and after completing cancer treatment with impact on prognosis. 21

| Echocardiographic image analysis
Echocardiographic datasets, previously acquired during cancer treatment and clinical follow-up and stored in the archiving system Siemens Syngo Dynamics at the Pediatric Cardiology Unit, Astrid Lindgren Children's Hospital, were re-analyzed. Left ventricular peak systolic deformation analysis by LS and GLS were performed in accordance with guidelines 33 using vendor-neutral software (TOMTEC Corporation). In summary, LS was measured in apical four-chamber views whereas GLS was automatically calculated from longitudinal strain measurements in apical four-chamber, apical three-chamber and apical two-chamber views. Re-analysis of echocardiographic data was performed by two senior echocardiographers, supervised by one pediatric cardiologist. Cardiac chamber quantification and functional measurements were performed according to recommendations in current guidelines 17,18,34 and z-scores were calculated for chamber dimensions. 35 When LVEF analysis according to the biplane method of disk summation was not available, visual LVEF (vLVEF) was performed.
Visual LVEF was preferred over estimation of LVEF by using the Teicholz formula, as the latter is reported inadequate in children. 17,18 To avoid confusion, the values of GLS and LS were presented as positive numbers. 36 Cut-off values for defining ventricular dysfunction were in accordance with cardio-oncological guidelines as follows: LVEF <53%, FS <0.28, LS <17%, and GLS <17%. 8-10

| Statistical analysis
The association between abnormal LS (<17%) at any time during treatment and vLVEF and FS after the end of treatment was evaluated at follow-up within 6-months and during the period of 1-2 years after the last anthracycline treatment.

| Patient characteristics
The patient cohort (Table 1)  This patient with symptomatic heart failure was the only patient in the cohort who received heart failure medication.

| Feasibility of retrospective analysis of echocardiograms
In total, 366 echocardiograms were available for re-analysis ( Figure 1).  18 GLS assessment was restricted by the lack of all three necessary echocardiographic views. 33 The main reason for LS ineligibility in our retrospective material was restricted image sectors from the apical four chamber view (the myocardium was not visualized throughout the cardiac cycle).

| Echocardiographic results
Of the 249 echocardiograms eligible for LS analysis (  In a sub-analysis, the cohort was dichotomised into patients with at least one echocardiogram with abnormal LS (<17%) and patients with normal LS (≥17%) throughout treatment ( Figure 3A (Table S1). We acknowledge that the Wilcoxon tests were used on groups that are highly unbalanced, which affected the power of the tests.

| Image quality
The relationship between frame rate and heart rate (FR/HR-ratio) as a measure for image quality may interfere with strain analysis. 38 We compared the FR/HR-ratio in echocardiograms analyzed for LS ( Figures S1A,B). The FR/HR-ratio was ≥0.7 in 155/249 echocardiograms (62%). In the 105 echocardiograms with LS <17% the FR/HR-ratio was ≥0.7 in 59/105 (56%), whereas in the 144 eligible echocardiograms with unimpaired LS, the FR/HR-ratio was ≥0.7 in 96/144 (67%).

| DISCUSSION
Cardiac dysfunction before, during and after cancer treatment has important impact on both cancer and overall outcome, as well as longterm quality of life, hence cardiotoxic cancer treatment regimens require serial evaluation of cardiac function by echocardiography. [8][9][10]39 Early detection of cardiac dysfunction, prior to overt symptomatic heart failure, is considered vital to enable timely interventions. 7,16,39 The standard echocardiographic methods for longitudinal surveillance of cardiac function in children, LVEF, and FS, 17,18 exhibit limitations regarding reproducibility as well as sensitivity to detect small changes in left ventricular systolic function. 19,40 The limited sensitivity is problematic, since significant reduction in LVEF as well as in and FS, often occur late in the process of cancer therapy induced cardiac dysfunction. 25 Limited reproducibility poses a challenge already during cancer treatment since reduction of LVEF or FS may motivate treatment alterations that may interfere with long-term cancer outcome. In conclusion, more sensitive and reproducible surveillance methods are warranted.
In adults, myocardial deformation, measured as GLS by 2D speckle-tracking echocardiography, has become the preferred parameter for early detection of asymptomatic cancer therapy-related cardiac dysfunction due to its high reproducibility and sensitivity, [9][10][11] even if the prognostic impact is yet to be proven. The availability of GLS measurements in children, especially during cancer treatment, might be hampered by the need of a high-quality images from three apical echocardiographic views. Including analysis of LS has therefore T A B L E 2 Re-analysis of FS and LVEF in echocardiograms eligible for LS-analysis. practice. In our retrospective material, LS demonstrated a higher feasibility (68%) than GLS (13%) as well as LVEF measured by the recommended biplane method (31%). 8,18 This may, in part, reflect the limitations of a retrospective dataset including previously acquired echocardiographic images. However, it may also be ascribed to the differences in feasibility between the echocardiographic methods that may be of relevance in clinical practice as well as in research.
A good correlation between GLS and LS measurements has been demonstrated by others, 43  The results of FS and LVEF measurements were contradictory in