Anthracycline-Induced Cardiotoxicity in Acute Myeloid Leukemia Patients Who Undergo Allogeneic Hematopoietic Stem Cell Transplantation

Oren Pasvolsky,1,2 Olga Morelli,2,3 Uri Rozovski,1,2 Mordehay Vaturi,2,3 Ofir Wolach,1,2 Irina Amitai,1,2 Iuliana Vaxman,1,2 Roy Ratzon,1,2,4 Moshe Yeshurun,1,2 Ran Kornowski,1,2 Zaza Iakobishvilli,1,2 Pia Raanani1,2

Abstract

Impairment of cardiac function adversely affects the outcome of hematopoietic stem cell transplantation (HSCT). Therefore, we studied the acute cardiotoxic potential of anthracycline-containing induction regimens in 78 patients with acute myeloid leukemia who underwent HSCT. Fourteen patients (18%) developed dosedependent systolic dysfunction, and deterioration of systolic function was associated with inferior survival outcome.
Introduction: There is paucity of data regarding the cardiotoxic effects of anthracycline treatment in the context of acute myeloid leukemia (AML) patients who undergo allogeneic hematopoietic stem cell transplantation (HSCT). Even a transient decrease in cardiac function might affect transplantation outcome. Patients and Methods: We reviewed the clinical records and echocardiography examinations of 78 patients with AML who received induction therapy and underwent HSCT. Results: Twenty-two patients (28%) received daunorubicin at a dose of 90 mg/m2 per day and 53 patients (68%) received 60 mg/m2 per day or an equivalent dose of idarubicin. In 14 patients (18%) the postinduction ejection fraction declined by at least 10%. This change was temporary in 6 patients and longstanding in the remainder. Patients who developed systolic dysfunction had inferior overall survival (13 months compared with 27 months; P ¼ .013). Patients whose diastolic function deteriorated had improved survival outcome (38 months compared with 17 months; P ¼ .048). Conclusion: Although even transient reduction in systolic function might compromise survival outcome, diastolic dysfunction predicts improved survival in patients with AML who undergo HSCT.

Keywords: Bone marrow transplant, Daunorubicin, Ejection fraction, Induction, Systolic function

Introduction

The backbone of induction therapy for acute myeloid leukemia (AML), comprised of cytarabine and an anthracycline, has remained unchanged for more than 4 decades.1 The original induction pro- Acute anthracycline toxicity is of concern in patients who are candidates for hematopoietic stem cell transplantation (HSCT). In these patients, even transient decrease in cardiac function might postpone transplantation and adversely affect transplantation outcomes.10 However, it is currently unknown whether anthracyclines given before transplantation might lead to increased acute cardiac toxicity, and whether anthracycline-associated decrease in cardiac function might compromise transplantation outcome. Because even short exposure to anthracyclines has been shown to induce some degree of cardiotoxicity,11 we hypothesized that we would find anthracyclineassociated deterioration in cardiac function in our cohort.

Patients and Methods

This was a single-center retrospective study. We collected data from medical records of patients with AML who received induction therapy and underwent HSCT at the Davidoff Cancer Center, Rabin Medical Center in Israel between the years 2007 and 2016. We excluded patientswho didnot receiveanthracyclinesduring induction and patients for whom we did not have at least 2 available echocardiography studies (pre- and postinduction). AML European LeukemiaNet 2017 guidelines were used for risk stratification.1 The regimen used for induction therapy in most of our AML patients included daunorubicin at a dose of 60 mg/m2 per day or 90 mg/m2 per day on days 1 through 3, and continuous cytarabine infusion at a dose of 100 mg/m2 per dayon days 1 through 7. The cumulative dose of anthracyclines was calculated for all treatments administered before transplantation. When nondaunorubicin anthracyclines were given, we converted the dose given to the equivalent dose of daunorubicin.12 Echocardiography studies provided estimators of systolic left ventricular function including eyeballing estimation of ejection fraction (EF). The diastolic function was estimated using the transmitral flow E/A, and the left ventricular mitral annular lateral wall E/e’ ratio according to accepted international guidelines.13 Diastolic function was then categorized as either normal (grade 0), or mild, moderate, or severe dysfunction (grades 1, 2, and 3, respectively).Acutecardiotoxicitywasdefinedasa10%reductionfrom baseline to below 50% within a time frame of 1 year.14 The study was approved by the institutional review board.

Statistical Analysis

Patient characteristics are summarized using frequencies (percentages) for categorical and median and range for continuous variables. OS is defined as the time from AML diagnosis to death. To compare patients according to categorical variables, we used the c2 and the Student t test with Levene test statistics to test for equality of variance and applied the Welch correction when needed. Medians were compared using the ManneWhitney test. To predict changes in systolic or diastolic function with time after anthracycline treatment we applied logistic regression models with Exp(B) and confidence interval around it as an estimator of odds ratio. The probability of OS was estimated using the KaplaneMeier method. The log rank test was used to compare survival distributions. The statistical analysis was performed using SPSS software (version 24, SPSS Inc), and GraphPad Prism software (version 7.0).

Results

Between 2007 and 2017, we treated 78 patients with AML who underwent HSCT and had adequate echocardiographic studies. The median age at diagnosis was 59 years (range, 18-73) and 43 patients (55%)weremale.Twenty-two (28%)received daunorubicinat adose of 90 mg/m2 per day for 3 days and 53 (68%) received 60 mg/m2 per day for 3 days or the equivalent of 12 mg/m2 per day of idarubicin (n ¼ 3). In 3 patients anthracycline dose was not available. Thirty-six patients (46%) received salvage chemotherapy, most (n ¼ 31; 40%) with anthracycline-containing regimens. Demographic and leukemia-related parameters are presented in Table 1.1 hypertension; IHD ¼ ischemic heart disease.
In this highly selected transplantation-eligible patient cohort, the baseline systolic function was relatively preserved, with a median left ventricular EF at baseline of 61% (range, 45%-70%). The diastolic function, however, was within normal range in only 18% of patients (n ¼ 14), and most patients had either mild (53%; n ¼ 41), moderate (28%; n ¼ 22) or severe (1%; n ¼ 1) diastolic dysfunction.
Most patients had at least 1 cardiovascular risk factor (68%; n ¼ 53; Figure 1A), and 43 patients (55%) received at least 1 cardioprotective drug (Figure 1B), however, only 11 patients (14%) had previous vascular events (Figure 1C). The presence of cardiovascular risk factors or previous vascular events was not correlated with the dose of anthracyclines given at induction.
In 14 patients (18%) the postinduction EF declined by at least 10% within the first year. This change was observed after a median of 135 days (range, 37 to 364), was temporary in 43% (n ¼ 6) and longstanding at least for 1 year in the remainder. Diastolic function was available in 71 patients, and in 22 (28%) the diastolic function deteriorated after induction. Only in 3 patients (4%) both systolic and diastolic functions worsened. The postinduction change in EF was not associated with sex, age, the presence of cardiovascular risk factors, or the use of cardioprotective drugs (angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, b-blockers, and statins). However, women were more likely to develop postinduction diastolic dysfunction. Age, the presence of cardiovascular risk factors, or the use of cardioprotective drugs were not associated with a decrease in diastolic function.
Compared with patients with preserved diastolic function, patients whose diastolic function deteriorated had improved survival outcome (38 months compared with 17 months; P ¼ .048). In contrast, patients who developed systolic dysfunction had inferior survival outcome (13 months compared with 27 months; P ¼ .013; Figure 2). The cumulative dose of anthracyclines given at induction and postinduction treatments was not associated with reduced systolic or diastolic function. While 24% (n ¼ 5) of patients who 2 received anthracyclines at 90 mg/m developed systolic dysfunction, only 15% (n ¼ 8) of patients who received lower doses developed systolic dysfunction. However, these differenced were not statistically significant.

Discussion

Approximately one-fifth of our patients with AML who received anthracycline-based induction treatment and proceeded to allogeneic HSCT developed transient or long-lasting decrease in systolic cardiac function. The deterioration in cardiac systolic function translated to a reduction of 14 months in OS.
The risk for anthracycline-induced cardiotoxicity in our study was higher than the overall incidence of 2.2% shown by Von Hoff et al,9 approaching that reported by Swain et al of 26%.15 However, these figures cannot be directly compared, because of differences in study population and methodology. Experience from >4000 patients has shown that there is a continuum of increasing cumulative probability of developing heart failure after anthracycline therapy, rather than an absolute cutoff.9
With a median age of 59 years and most patients having at least 1 cardiovascular risk factor, our cohort comprised a population at risk for cardiac injury. Therefore, although the study was not powered to stratify patients according to cardiovascular risk factors, it is not surprising that early deterioration in cardiac function after anthracycline administration was detected in one-fifth of our patients. Furthermore, some degree of diastolic dysfunction was observed in most patients at baseline.
Historically, female sex has been considered a risk factor for chronic anthracycline-induced cardiotoxicity.8 However, historical data are limited, and in one of the largest studies by Swain et al,15 which included mainly patients with breast cancer, only 12% of the study population was male. A large prospective study by Cardinale et al showed female sex to be a risk factor for anthracycline-induced cardiotoxicity. However, this risk was obviated upon crossvalidation analysis,16 and most patients included in this study were female as well (74%). Remarkably, in our cohort women were more likely to develop diastolic rather than systolic dysfunction.
Recent studies have shown that daunorubicin at a dose of 90 mg/m2 per day or 60 mg/m2 per day used for induction treatment in AML might have comparable efficacy.5,7 Therefore, as the therapeutic efficacy of increased doses of anthracyclines remains questionable in AML patients; the deleterious effects of dose escalation should be taken into consideration as well. Because even mild cardiac dysfunction might increase transplantation-related morbidity and mortality,10 anthracycline-free induction regimens could provide improved outcomes for AML patients in the future, perhaps partially through mitigation of the reduction in systolic cardiac function.
The yield of baseline cardiac evaluation before induction therapy in AML patients has been questioned. Bryant et al reported a low incidence of abnormal baseline left ventricular EF in a single-center study (4 of 76 patients), and speculated whether unnecessary cardiac evaluation could be unnecessarily delaying chemotherapy.17 Clinical practice guidelines vary widely in their recommendations for cardiac monitoring before, during, and after chemotherapy.18 Our data strengthen the necessity of cardiac evaluation, at least in AML patients who are transplantation-eligible.
Unexpectedly, patients whose diastolic function deteriorated after treatment with anthracyclines had a lesser need for salvage chemotherapy, and had a 21-month longer OS expectancy. This raises the possibility that an increase in diastolic cardiotoxicity is associated with improved efficacy of anthracycline treatment. We speculate that efficient killing of tumor cells triggers systemic inflammation, which in turn reduces diastolic function, either by direct infiltration of the cardiac wall by inflammatory cells, or because of increased collagen production. In this way the systemic inflammatory response is associated with improved efficacy but also with deterioration of diastolic function. Previous research has shown that anthracyclines can promote a systemic inflammatory response mediated by increased levels of interleukin 1-b.19 Systemic inflammation has been linked to the development of diastolic dysfunction.20 Diastolic function decline rate was higher among women in our cohort. This finding is in line with the higher levels of markers of systemic inflammation found in women compared with men.21 The notion that anthracycline-induced adverse effects correlate with clinical efficacy has previously been explored. Elis et al showed that among patients with Hodgkin lymphoma treated with anthracycline-containing regimens, those who developed alopecia had higher response rates.22
Our study has several limitations. First, being a nonplanned retrospective analysis, we cannot rule out undocumented confounders that affect cardiac function. For example, patients’ characteristics at baseline have an influence on the dose of anthracyclines. Furthermore, because of the retrospective nature of this study, we could not preplan the timing of the echocardiographic examinations performed on each patient. Another limitation is the relatively small sample size, which might affect the generalizability of our conclusions and precluded us from performing subgroup analyses.

Conclusion

We describe an increased mortality rate among HSCT transplanted AML patients whose systolic cardiac function deteriorated in the first post induction year. Our data support the common practice of obtaining an echocardiography study at least at baseline, before anthracycline-containing induction regimens for AML, because most patients had significant risk factors and baseline diastolic dysfunction. The finding of increased survival outcome in patients with decreased diastolic function is intriguing and warrants further validation and mechanistic research.
The chronic cardiotoxic effect of anthracycline is well defined. We show herein that in the context of patients with AML who are candidates for HSCT, acute and often transient reduction in systolic function might affect outcome. Although assessment of cardiac function before transplantation is standard, our data show that even a small reduction in EF had an effect on OS. Whether echocardiography should be obtained before induction therapy is still debatable,17 but our data show that preinduction assessment of cardiac function should be obtained, and adds important baseline information with prognostic significance.

Clinical Practice Points

Although chronic anthracycline-associated cardiotoxicity is well described, little is known regarding the immediate cardiotoxic potential of anthracyclines, especially when given as part of induction treatment in AML. In patients with AML who undergo HSCT, impaired cardiac function can affect the timing and outcome of HSCT. Therefore, we sought to investigate whether acute anthracycline-induced cardiotoxicity affects outcome. In this single-center retrospective study, one-fifth of patients who received induction therapy and HSCT for AML had a significant reduction in systolic function post induction. In approximately half these cases the systolic function deterioration was long-lasting. The median survival outcome of patients with deteriorating systolic function was 13 months, compared with 27 months in patients with preserved systolic function (P ¼ .013). Unexpectedly, we found that deterioration in diastolic cardiac function was associated with improved survival outcome.Our data show that preinduction assessment of cardiac function should be obtained and adds important baseline information with prognostic significance.

References

1. Dohner H, Estey E, Grimwade D, et al. Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood 2017; 129:424-47.
2. Yates J, Glidewell O, Wiernik P, et al. Cytosine arabinoside with daunorubicin or adriamycin for therapy of acute myelocytic leukemia: a CALGB study. Blood 1982; 60:454-62.
3. Fernandez HF, Sun Z, Yao X, et al. Anthracycline dose intensification in acute myeloid leukemia. N Engl J Med 2009; 361:1249-59.
4. Lowenberg B, Ossenkoppele GJ, van Putten W, et al. High-dose daunorubicin in older patients with acute myeloid leukemia. N Engl J Med 2009; 361:1235-48.
5. Devillier R, Bertoli S, Prebet T, et al. Comparison of 60 or 90 mg/m(2) of daunorubicin in induction therapy for acute myeloid leukemia with intermediate or unfavorable cytogenetics. Am J Hematol 2015; 90:E29-30.
6. Lee JH, Joo YD, Kim H, et al. A randomized trial comparing standard versus highdose daunorubicin induction in patients with acute myeloid leukemia. Blood 2011; 118:3832-41.
7. Burnett AK, Russell NH, Hills RK, et al. A randomized comparison of Idarubicin daunorubicin 90 mg/m2 vs 60 mg/m2 in AML induction: results from the UK NCRI AML17 trial in 1206 patients. Blood 2015; 125:3878-85.
8. Henriksen PA. Anthracycline cardiotoxicity: an update on mechanisms, monitoring and prevention. Heart 2018; 104:971-7.
9. Von Hoff DD, Layard MW, Basa P, et al. Risk factors for doxorubicin-induced congestive heart failure. Ann Intern Med 1979; 91:710-7.
10. Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation (HCT)specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood 2005; 106:2912-9.
11. Cardinale D, Sandri MT, Colombo A, et al. Prognostic value of troponin I in cardiac risk stratification of cancer patients undergoing high-dose chemotherapy. Circulation 2004; 109:2749-54.
12. Shankar SM, Marina N, Hudson MM, et al. Monitoring for cardiovascular disease in survivors of childhood cancer: report from the Cardiovascular Disease Task Force of the Children’s Oncology Group. Pediatrics 2008; 121:e387-96.
13. Nagueh SF, Smiseth OA, Appleton CP, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2016; 29:277-314.
14. Zamorano JL, Lancellotti P, Rodriguez Munoz D, et al. 2016 ESC position paper on cancer treatments and cardiovascular toxicity developed under the auspices of the ESC Committee for Practice Guidelines: the Task Force for cancer treatments and cardiovascular toxicity of the European Society of Cardiology (ESC). Eur Heart J 2016; 37:2768-801.
15. Swain SM, Whaley FS, Ewer MS. Congestive heart failure in patients treated with doxorubicin: a retrospective analysis of three trials. Cancer 2003; 97:2869-79.
16. Cardinale D, Colombo A, Bacchiani G, et al. Early detection of anthracycline cardiotoxicityandimprovementwithheartfailuretherapy.Circulation2015;131:1981-8.
17. Bryant A, Sheppard D, Sabloff M, et al. A single-institution analysis of the utility of pre-induction ejection fraction measurement in patients newly diagnosed with acute myeloid leukemia. Leuk Lymphoma 2015; 56:135-40.
18. Levis BE, Binkley PF, Shapiro CL. Cardiotoxic effects of anthracycline-based therapy: what is the evidence and what are the potential harms? Lancet Oncol 2017; 18:e445-56.
19. Sauter KA, Wood LJ, Wong J, Iordanov M, Magun BE. Doxorubicin and daunorubicin induce processing and release of interleukin-1beta through activation of the NLRP3 inflammasome. Cancer Biol Ther 2011; 11:1008-16.
20. Mann DL. Inflammatory mediators and the failing heart: past, present, and the foreseeable future. Circ Res 2002; 91:988-98.
21. Khera A, McGuire DK, Murphy SA, et al. Race and gender differences in Creactive protein levels. J Am Coll Cardiol 2005; 46:464-9.
22. Elis A, Blickstein D, ManorY, Lishner M. Association between alopecia and response to chemotherapy in patients with Hodgkin lymphoma. Ther Drug Monit 2005; 27:287-9.