Amsacrine-based induction therapy in AML patients with cardiac comorbidities: a retrospective single-center analysis

Intensive chemotherapy is the backbone of induction treatment in patients with acute myeloid leukemia (AML). However, AML patients with concomitant cardiac disease may not be eligible for anthracycline-based therapies. In a small cohort of patients, we have previously shown that anthracycline-free, amsacrine-based chemotherapy TAA (thioguanine, cytarabine, amsacrine) may be as effective as cytarabine/daunorubicin for induction therapy in these patients. In this systematic retrospective single-center analysis, we documented the outcome of 31 patients with significant cardiac comorbidities including coronary heart disease or cardiomyopathy receiving TAA as induction chemotherapy. Median (range) ejection fraction (EF) was 48% (30–67%) in this cohort. Patients with EF below 30% were considered unfit for intensive induction therapy. Event-free survival (EFS), overall survival (OS), and relapse-free survival (RFS) were 1.61, 5.46, and 13.6 months respectively. Poor outcome was primarily related to a high early mortality rate within the first 30 days of therapy, mainly caused by infectious complications. TAA cannot be recommended as a substitute of standard induction for AML patients with significant concomitant cardiac disease. In the era of novel agents, alternative strategies (e.g., hypomethylating agents plus venetoclax) should be considered when anthracycline-based regimens are not suitable.


Introduction
Acute myeloid leukemia (AML) is a disease of the elderly population with a median age of 72 years at diagnosis [1,2]. Outcome in older patients is poor compared to younger patients, mainly due to higher early mortality and lower remission rates [3]. In addition to adverse genetic and other biological disease characteristics with impact on prognosis in elderly AML, advanced age often entails a higher likelihood for comorbidities that may preclude standard induction chemotherapy [4]. Anthracycline-based chemotherapy is the backbone of induction therapy in AML with the intention to achieve first complete remission before subsequent consolidation treatment with either conventional chemotherapy or hematopoietic stem cell transplantation (HSCT). Due to the high prevalence of concomitant cardiac disease in elderly patients, it can be challenging for clinicians to identify an appropriate induction regimen [5,6].
Several recommendations have been made for the treatment of AML patients with significant cardiac comorbidities. As long as the ejection fraction (EF) has not fallen below 45%, patients may be treated with standard induction therapy [7]. The risk of clinical heart failure can be decreased by reducing the anthracycline dose or prolonging the anthracycline infusion time without significantly compromising disease control [8,9]. The use of cardioprotective drugs and optimized supportive therapy may further reduce the risk of complications during intensive induction therapy [10,11]. However, patients with coronary heart disease or patients after myocardial infarction do not necessarily suffer from reduced EF and even after optimization of cardiac therapy, they remain at high risk for complications [8,9].
In the past, intensive anthracycline-free chemotherapy regimens have been proposed for patients who are considered fit for intensive therapy in principle but show significant cardiac comorbidities [7,23,24]. In this context, others and we have reported that replacement of anthracyclines by amsacrine may provide an alternative intensive induction regimen with less cardiac toxicity and without compromising complete remission rates in AML [25][26][27][28][29][30][31]. Thus, amsacrine-based induction has been repeatedly used in the past decades in AML patients with cardiac comorbidities but otherwise preserved performance status.
In this retrospective single-center analysis, we re-evaluated outcomes of patients receiving the TAA regimen as induction chemotherapy in a comparatively large cohort of intensively treated AML patients with cardiac comorbidities between 2009 and 2020.

Data and statistical analysis
Patients with newly diagnosed de novo, secondary, or therapy-related AML (other than acute promyelocytic leukemia) were included. Patients with cardiac comorbidities, including cardiomyopathy, cardiovascular disease, or heart failure, who were not eligible for treatment with cardiotoxic anthracyclines were considered for treatment with amsacrine-based induction chemotherapy TAA (200 mg/m 2 thioguanine on days 3-9, 210 mg/m 2 amsacrine on days 3-5 and 200 mg/m 2 cytarabine over 24 h on days 1-8). Decision for induction therapy was made in the hematological tumor board of our center.
Cardiac function of each patient was assessed via transthoracic echocardiography before start of therapy. If only qualitative results were available, numeric equivalents were chosen according to the guidelines of the British Society of Echocardiography [33].
Survival times were plotted using Kaplan-Meier curves. Relapse-free survival (RFS) was calculated from the time of CR/CRi until relapse or death. Event-free survival (EFS) was calculated from start of induction therapy until induction failure, relapse after CR/CRi or death, whichever occurred first. We performed multivariable Cox regression analysis to evaluate the effects of different variables on OS. Likewise, multi-variable logistic regression models were computed to assess the influence of sex, age, EF, WBC, LDH, blast count, AML type, and genetic risk according to the ELN2010 classification on early death events. p values < 0.05 were considered to indicate significant differences. Calculations and plots were performed using the statistical environment R (version 4.2.0).

Results
Between 2009 and 2020, a total of 31 patients were treated with TAA induction at our department. Patient characteristics are summarized in Table 1. Median (range) age of all patients was 63 (38 -77) years. The most common concomitant cardiac disease was coronary artery disease, which was found in 15 (48%) of these patients, followed by cardiomyopathy in 6 patients (19%) and chronic heart failure, which occurred in 4 patients (13%). The remaining 6 patients suffered from various concomitant cardiac diseases such as atrial fibrillation or severe aortic valve stenosis. If patients had more than one heart condition, they were classified according to the most severe or clinically leading condition. Median (range) EF was 48% (30-67%). Reviewing the medical records of all patients and identified 9 patients     (Fig. 1a). Median OS was 5.46 months (95% CI 0.82-17.7) (Fig. 1b). Early mortality within the first 30 days of induction was high, accounting for 12 fatalities (39%). Main cause of death were infectious complications ( Table 2). In the 16 patients achieving a CR/CRi, median RFS was 13.6 months (95% CI 11.3-NA) (Fig. 1c).
In multivariable Cox regression analyses, age and genetic risk were significantly associated with OS (HR 1.1 CI 1.02-1.18, p = 0.012; HR 0.09 CI 0.01 -0.85, p = 0.035; Table 3). A higher blast count at diagnosis was significantly associated with increased risk of early death in a binary logistic regression analysis (p = 0.038).

Discussion
In this systematic retrospective single-center analysis, median OS (5.46 months) and 5-year survival (< 10%) following induction therapy with TAA were dismal, mainly because of the high early-death rate related to infectious complications. Our findings are based on a patient cohort treated between 2009 and 2020 and contrast with a former analysis from our department where the TAA induction therapy applied to patients between 1997 and 2003 has been shown to be equally tolerated and efficacious compared to 7 + 3 induction [25]. This discrepancy may result from a smaller sample size of the previous study and the continuous improvement of supportive care strategies since 2008, leading to more patients being referred to intensive treatment despite of comparatively higher comorbidities. Consequently, risk of early death may have increased in our vulnerable cohorts of AML patients over time.
The high early death rate in our analysis exceeds most published rates for intensive AML induction regimens in a cohort of this age [2,[34][35][36]. Previous studies indicated that amsacrine-based and other common intensive induction regimens are equivalent in AML therapy in terms of toxicity and early death. However, these studies did not necessarily include patients with severe comorbidities [26][27][28][29][30][31]. Our data suggests unacceptably high treatment-related toxicity of the TAA regimen in AML patients with significant cardiac diseases who were otherwise considered eligible for intensive treatment.
EFS was reduced in our analysis compared with published data, whereas RFS was comparable to published results after induction therapy, underlining the antileukemic efficacy of the TAA regimen [37,38].
Although our analysis suffers from the limitations of a single-center retrospective study, our results strongly imply that patients with significant concomitant cardiac disease should not be treated with amsacrine-based induction chemotherapy. We believe that whenever application of an anthracycline-based induction regimen is contraindicated, alternative substances should be preferred, e.g., HMA or other targeted therapies. It should be noted that also venetoclax-based therapies may be associated with cardiac complications, even in the absence of pre-existing cardiac disease [39]. However, in terms of response rates, these alternative therapies have been shown to be very effective in AML and do not preclude consolidation with allogeneic HSCT with curative intent [40][41][42][43].

Data availability
The original data are available by e-mail from the corresponding author on reasonable request and provision of data must be in compliance with the relevant legal regulations.

Declarations
Ethics approval This retrospective analysis involving patient data was in accordance with the ethical standards of the institutional and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. It was conducted retrospectively from data obtained for clinical purposes. It was approved by the joint ethics committee of the Physicians` Chamber Westphalia-Lippe and the University of Münster, Germany (AZ 2021-490-f-S).

Competing interests The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.