Prevalence of anthracycline-related cardiac dysfunction in long-term survivors of adult-onset lymphoma

Abstract

Background

Anthracycline-related heart failure is a leading cause of morbidity in survivors of adult-onset lymphoma. There is a paucity of information on screening for late-occurring preclinical disease, which in turn has limited guidelines for early detection and intervention. Our objective was to examine the prevalence and risk of cardiac dysfunction, as measured by echocardiography (abnormal left ventricular systolic/diastolic function, or strain) in lymphoma survivors treated with anthracyclines; to evaluate the diagnostic yield of blood biomarkers in the asymptomatic setting.

Methods

Lymphoma survivors treated with hematopoietic cell transplantation ([HCT], n=78) or with conventional therapy (non-HCT, n=77) were compared to each other as well as to matched controls (n=51); the study was limited to lymphoma survivors who were >5 years from diagnosis.

Results

At a median follow-up of 9.4 years from diagnosis, one in five (20.6 %) lymphoma survivors had cardiac dysfunction; the odds of having cardiac dysfunction were 6.6-fold (Odds Ratio [OR]: 6.6, p=0.01) greater among lymphoma survivors when compared with matched controls. There was a dose-dependent risk of cardiac dysfunction by cumulative anthracycline dose (controls [referent]; 1–249 mg/m², OR=4.7, p=0.05; ≥250 mg/m², OR=7.6, p<0.01), but no difference in the prevalence of cardiac dysfunction between conventionally-treated and HCT survivors. The diagnostic accuracy of cardiac blood biomarkers in the asymptomatic setting was quite poor.

Conclusions

In these long-term survivors, there was a high rate of cardiac dysfunction that was independent of HCT status. The growing number of lymphoma survivors makes it imperative to identify reliable and cost-effective strategies to decrease the burden of heart failure in this population.

INTRODUCTION

Anthracyclines form the backbone of therapy for lymphoma. However, the use of anthracyclines is limited by a dose-dependent cardiotoxicity that can result in heart failure either during treatment or years after completion of therapy.¹ Anthracycline cardiotoxicity is thought to be related to direct injury to the cardiomyocytes, due to a combination of free radicals and intracardiac metabolic derangements.² The characteristically delayed manifestation of symptomatic heart failure in cancer survivors is likely related to the eventual failure of a reduced number of unhealthy cardiomyocytes in response to the myocardial injury.³ In this context, anthracycline-related heart failure can be characterized as a progressive disorder, with a variable period of asymptomatic cardiac dysfunction that precedes clinically overt disease.¹ As such, a number of clinical guidelines recommend routine screening with echocardiography for asymptomatic cardiac dysfunction, allowing for the implementation of appropriate interventions (e.g. angiotensin-converting enzyme [ACE] inhibitors, Beta-blockers) to prevent or delay the onset of symptomatic disease.^4–6 These recommendations emphasize the importance of screening for systolic as well as diastolic dysfunction during anthracycline therapy, and for the need to incorporate novel echocardiographic (e.g. global longitudinal strain) and blood biomarkers (e.g. natriuretic peptides) of cardiac dysfunction, when feasible.^4–6 The recommendations regarding long-term surveillance (beyond the immediate treatment period) are less clear.

In fact, the recent American Society of Clinical Oncology (ASCO) Guideline for Prevention and Monitoring of Cardiac Dysfunction⁵ could not provide guidance on screening in long-term (>5 years from diagnosis) cancer survivors, due to the paucity of information on the prevalence and risk factors for cardiac dysfunction in this population. In addition, while blood biomarkers are routinely used to diagnose acute heart failure in the general non-oncology population,^{7, 8} little is known about their diagnostic utility in asymptomatic long-term cancer survivors. As such, no recommendations could be made in the ASCO Guidelines regarding their use during routine surveillance.⁵ Keeping these knowledge gaps in mind, the current study sought to compare the prevalence and risk of cardiac dysfunction in long-term lymphoma survivors treated with anthracyclines with age- and sex-matched non-cancer controls. Secondary objectives were to compare the prevalence of cardiac dysfunction in conventionally-treated survivors to those who had undergone hematopoietic cell transplantation (HCT) - a population thought to be at higher risk of cardiac dysfunction,^{9, 10} and to evaluate the diagnostic accuracy of candidate blood biomarkers for detection of asymptomatic disease in lymphoma survivors.

MATERIALS AND METHODS

Study participants

A cross-sectional study design was used. Patients were identified from an existing electronic database at a single institution that included information on cancer diagnosis, age at diagnosis, and vital status. Eligibility criteria included: 1) diagnosis of non-Hodgkin lymphoma or Hodgkin lymphoma between January, 1995 and December, 2009; 2) ≥21 years old at the time of diagnosis; 3) ≥5y years from diagnosis; 4) known to be alive at the time of recruitment (September, 2013 to July, 2015); 5) off cancer therapy for ≥2 years; 6) received anthracycline-based treatment for lymphoma. We excluded survivors who were actively being treated for cardiomyopathy or heart failure, as echocardiographic indices or blood biomarkers in these individuals could have been altered by pharmacologic therapy. A total of 678 survivors met the eligibility criteria, and 200 (30%) were randomly selected for recruitment, stratified on age at enrollment (+/− 5 years), sex, HCT status (1:1), and time from diagnosis (+/−2 years); there were no significant demographic and treatment-related differences between the subset selected for recruitment and the remaining cohort (Supplementary Table 1).

Healthy controls without a history of cancer or cardiomyopathy were recruited from the general population (friends, acquaintances of study participants), frequency-matched (1:3) to lymphoma survivors on sex and age (+/− 5 years) at study participation. The study was approved by the City of Hope institutional review board, and all study participants provided written informed consent.

Cardiac evaluation

Echocardiograms were performed by a designated study technician, and consisted of complete two-dimensional (2D), M-mode, and Doppler evaluations, per the American Society of Echocardiography and the European Association of Cardiovascular Imaging practice guidelines.⁴ General Electric Vivid-7 echocardiography machine (General Electric; Chicago, IL) was used for all study-related echocardiographic evaluations. LV end diastolic/ systolic diameters (LVEDD, LVESD), inter-ventricular septal thickness, and LV posterior wall thickness in diastole and systole (LVPWD, LVPWS) were measured on M-mode recordings obtained from a standard LV parasternal view. LV EF was calculated from the apical 4- and 2-chamber views using a modified Simpson biplane method.¹¹ LV mass was calculated from the recommended AHA/ ACC formula¹¹ that takes into consideration LV dimension, posterior wall and intraventricular septal thickness, and correction factors derived from regression analysis, divided by body surface area. Measurements of LV diastolic function included pulsed Doppler measurement of mitral valve inflow, peak velocities of early filling (E-wave) and filling during atrial systole (A-wave), mitral annular myocardial e′ and a′ velocities, mitral E/A and E/e′ ratios. LV global longitudinal strain (GLS) was measured manually by tracing the endocardial border in end systole; commercially available analysis software (EchoPAC vBT08; GE Healthcare, USA) was used to automatically trace regions of interest as well as the entire myocardium. Study echocardiograms were first interpreted by a designated cardiologist at our institution, to ensure there was not a clinically actionable finding. Then, an anonymized copy of the digitized echocardiogram was sent to the study core cardiology laboratory (University of Toronto, Toronto, ON) where all study endpoints were re-measured by a single sonographer who was blinded to the lymphoma survivor or control status of the study participant. All measurements were based on the average of 3 cardiac cycles, in accordance with established criteria of the American Society of Echocardiography.¹² Intra- and inter-observer comparisons were made in fifteen randomly selected study participants (Supplementary Table 2: correlation coefficient range 0.81 – 0.99).

Blood biomarkers

Blood samples were collected on the day of the echocardiographic assessment. B-type natriuretic peptide (BNP) was measured using the Triage BNP test (Beckman Coulter, Fullerton, California); measurable range: 2.0 pg/mL to 5000 pg/mL. NT proBrain Natriuretic Peptide (NT-proBNP) was measured using the Elecsys proBNP II test (Roche Diagnostics, Indianapolis, Indiana); lower limit of detection: 5 pg/mL. Protein ST2 was measured using the Presage ST2 ELISA (Critical Diagnostics, New York, NY) test; lower limit of detection: 2.0 ng/mL.

Clinical data collection and definitions

Self-reported questionnaires were used to obtain baseline data on sociodemographic variables (age, sex, race/ethnicity, highest level of education, employment status, and marital status) and cardiovascular risk factors. Individuals were noted to have hypertension, diabetes, or dyslipidemia if they were receiving medications for their management at the time of study enrollment. Overweight/obese was defined by body mass index (BMI) ≥25 kg/m². Physical inactivity was defined as not meeting the national guidelines for regular exercise for adults (e.g. three to five sessions of moderate- or vigorous-intensity exercise per week of ≥20 minutes in duration).¹³ Medical records provided the following information: date of diagnosis, type of cancer, cumulative dose of anthracycline exposure, receipt and dose of chest-directed radiation therapy, and HCT status. Chest-directed radiation therapy included radiation to the following fields: mantle, mediastinal, or lung.¹⁰ High-dose (HD) anthracycline was defined as cumulative doxorubicin dose ≥250 mg/m^2,14 and HD-chest radiation therapy was defined as receipt of ≥30 Gy.¹⁴

Statistical analysis

The primary endpoint for this study was cardiac dysfunction, defined a priori as having either LV systolic dysfunction (LV EF <50%),¹⁵ diastolic dysfunction (mitral E/e′ >14),¹⁶ or abnormal strain (GLS >−16).¹⁷ Prevalence of cardiac dysfunction was projected to be 20–25% for survivors and 5% for controls. Assuming a Type I error of 0.05, enrolling approximately 150 survivors and 50 controls provided a minimum of 80% power to detect a significant difference in the prevalence of cardiac dysfunction between the groups.

Descriptive statistics for clinical variables, echocardiographic indices, and blood biomarkers were generated for study participants. Categorical variables were compared using X² tests. Continuous variables were compared using independent two-sample t-tests or analysis of variance (ANOVA). Sensitivity, specificity, positive predictive value, and negative predictive value were calculated to assess the validity of blood biomarkers for detection of cardiac dysfunction among lymphoma survivors. Cutoffs for blood biomarkers (abnormal: BNP ≥100 pg/dL, NT-proBNP ≥300 pg/dL, ST-2 >35 ng/mL) were based on thresholds used to diagnose cardiac dysfunction in the general population.^18–20

Multivariable unconditional logistic regression was used to identify variables that were associated with cardiac dysfunction. The dependent variable was cardiac dysfunction. The independent variables included clinical and treatment-related factors that differed between the three groups (p<0.1), as well as variables thought to impact the risk of cardiac dysfunction, and included: age at examination, BMI; a separate regression model was created to evaluate the association between cardiac dysfunction and cumulative anthracycline dose (0, 1–249 mg/m², ≥250 mg/m²). Data were analyzed using SPSS Version 18.0 (IBM, Armonk, NY). All statistical tests were 2-sided, and P<0.05 were considered statistically significant.

RESULTS

Patient characteristics

Two hundred survivors were approached to participate in the study. Twenty-nine individuals refused participation, and 16 others did not complete the full battery of echocardiographic and blood biomarker testing or had incomplete echocardiographic measurements. The current report includes results from 155 study participants who completed all study measurements and had evaluable echocardiograms (participation rate: 78%). Table 1 summarizes the participants’ characteristics. Due to previously established matching criteria, there were no statistically significant differences by sex and age at study enrollment between lymphoma survivors and controls. Additionally, there were no differences in race/ethnicity, sociodemographics (education, marital status, employment), or prevalence of cardiovascular risk factors (overweight/obese, hypertension, diabetes, dyslipidemia, smoking history, physical activity level) between survivors and controls.

Table 1.

Patient and treatment characteristics

Characteristics	Lymphoma (N=155)	Controls (N=51)*	P-Value
Sex, No. (%)
Male	79 (51.0)	25 (49.0)	-
Age at examination, Years
Median, range	57.9 (28.6–83.8)	56.9 (29.3–73.1)	-
Race/Ethnicity, No. (%)
Non-Hispanic White	95 (61.3)	35 (68.6)	0.47
Hispanic	42 (27.1)	11 (21.6)
Other	18 (11.6)	5 (9.8)
Education, No. (%)
<=High school	19 (12.5)	5 (9.8)	0.65
Some college	49 (32.2)	17 (33.3)
>=College	87 (55.3)	29 (56.8)
Employed Full/Part, No. (%)
Yes	81 (52.3)	25 (49.0)	0.68
Marital Status, No. (%)
Married	105 (68.6)	33 (64.7)	0.60
Cardiovascular risk factors, No. (%)
Overweight/obese‡	98 (63.4)	26 (51.0)	0.12
Hypertension	33 (21.3)	14 (27.5)	0.36
Diabetes	14 (9.0)	5 (9.8)	0.87
Dyslipidemia	49 (31.6)	14 (27.5)	0.57
Ever smoked tobacco	59 (38.6)	16 (31.4)	0.36
Physically inactive‡‡	97 (62.6)	29 (56.9)	0.47
Diagnosis, No. (%)
Non-Hodgkin Lymphoma	118 (76.1)	-
Hodgkin Lymphoma	37 (23.9)	-
Age at diagnosis, Years
Median, range	47.9 (19.5–77.3)	-
Time since diagnosis, Years
Median, Range	9.4 (5.0–19.9)	-
Lifetime doxorubicin dose, No. (%)
<150 mg/m2	20 (12.9)	-
150–249 mg/m2	26 (16.8)	-
250–349 mg/m2	96 (61.9)	-
≥350 mg/m2	13 (8.4)	-
Chest radiation, Gy; No. (%)
None	130 (83.9)	-
1–29.9	9 (5.8)	-
≥30	16 (10.3)	-
Underwent autologous HCT, No. (%)
Yes	78 (50.3)	-

^*Matched to lymphoma survivors on: Sex, Age at examination;

^‡Body mass index ≥25 kg/m²;

^‡‡Not meeting the national guidelines for regular exercise for adults (e.g. three to five sessions of moderate- or vigorous-intensity exercise per week of ≥20 minutes in duration)

Among lymphoma survivors, median time from diagnosis to study participation was 9.4 years (range 5.0 to 19.9 years). One hundred and nine (70.3%) individuals had been treated with HD-anthracycline, and 16 (10.3%) had received HD-chest RT. Because of study design, half (50.3%) had undergone autologous HCT for management of their lymphoma. HCT survivors were significantly less likely to have been treated with HD-chest RT (6.4% vs. 14.3%, p=0.02) when compared to conventionally treated patients (Supplementary Table 3). There was no difference in cancer diagnosis, age at diagnosis, time from diagnosis, or cumulative anthracycline dose between the HCT and conventionally treated survivors. Importantly, none of the survivors had clinical signs or symptoms of heart failure⁷ at the time of study evaluation.

Echocardiographic assessment

A comparison of echocardiographic measurements of LV systolic function, diastolic function, and GLS between HCT survivors and conventionally-treated survivors as well as controls is presented in Table 2. Overall, survivors were significantly more likely to have a larger LV diameter in systole, decreased LV posterior wall thickness in systole, lower EF, increased mitral E/e′, and decreased GLS when compared to controls. On the other hand, there was no difference in echocardiographic indices by HCT status. In lymphoma survivors, the prevalence of abnormal EF, diastolic dysfunction, and abnormal GLS was 8.4%, 5.2%, and 14.2%. The overall prevalence of cardiac dysfunction was significantly higher in lymphoma survivors when compared to controls (20.6% vs. 3.9%, p<0.01). Among survivors with cardiac dysfunction (N=32) the vast majority (75%) had an abnormality in only one parameter, and there was concordance (abnormal in all three indices) in 6.3% of individuals. In multivariable logistic regression analysis, the odds of having cardiac dysfunction were 6.6-fold (Odds Ratio [OR]: 6.6, 95% confidence interval [95% CI] 2.5–21.1; p=0.01) greater among lymphoma survivors when compared with matched controls. When examining this association by anthracycline dose, there was a dose-dependent association with cumulative exposure: controls (referent); 1–249 mg/m², OR=4.7 (95% CI 1.0–17.4), p=0.05; ≥250 mg/m², OR=7.6 (95% CI 2.7–24.3), p<0.01.

Table 2.

Cardiac echocardiographic and blood biomarkers

Biomarkers	HCT (N=78)	Non-HCT (N=77)	P-Value*	Controls (N=51)	P-Value**
Echocardiographic indices
LV end-systolic diameter, (cm)
Median, range	3.3 (2.3–5.9)	3.2 (2.1–5.2	0.19	2.9 (2.1–4.5)	0.01
LV end-diastolic diameter (cm)
Median, range	5.0 (3.8–7.9)	5.1 (3.7–6.56)	0.98	5.0 (4.2–6.8)	0.86
Inter-ventricular septal thickness (cm)
Median, range	0.81 (0.50–1.49)	0.86 (0.60–1.55)	0.48	0.80 (0.51–1.50)	0.91
LV posterior wall thickness, diastole (cm)
Median, range	0.74 (0.4–1.2)	0.8 (0.5–1.2)	0.25	0.8 (0.5–1.1)	0.68
LV posterior wall thickness, systole (cm)
Median, range	1.3 (0.9–2.1)	1.4 (1.0–2.2)	<0.01	1.5 (1.1–2.1)	<0.01
Ejection fraction (%)
Median, range	64.0 (30.0–81.0)	67.0 (40.0–87.0)	0.12	73.0 (52.0–88.0)	<0.01
LV mass index (g/m2)
Median, range	74.0 (36.9–162.0)	75.0 (36.5 –146.0)	0.42	82.6 (39.8–151.7)	0.10
Mitral E/A ratio
Median, range	1.0 (0.4–2.9)	1.1 (0.5–2.5)	0.75	1.1 (0.6–2.4)	0.40
Mitral E/e′ ratio
Median, range	7.0 (2.9–20.2)	7.7 (0.6–21.4)	0.78	6.8 (3.7–12.9)	0.05
Global longitudinal strain (%)
Median, range	−18.6 (−26.0, −10.3)	−18.9 (−24.0, −9.3)	0.84	−20.7 (−29.2, −15.5)	<0.01
Blood biomarkers
BNP (pg/dL)
Median, range	45 (13–887)	41 (7–145)	0.06	29 (13–79)	<0.01
NT-proBNP (pg/dL)
Median, range	132 (27–6,962)	107 (27–747)	0.10	44 (27–311)	<0.01
ST-2 (U/mL)
Median, range	23.2 (7.2–87.7)	22.8 (5.5–44.0)	0.08	23.0 (7.6–46.6)	0.62

^*HCT vs. Non-HCT;

^**All lymphoma survivors (N=155) vs. controls.

Abbreviations: BNP, B-type natriuretic peptide; NT-pro-BNP, N-terminal prohormone of brain natriuretic peptide

Blood biomarkers

Candidate biomarkers of cardiac remodeling were compared across all three groups (Table 2). Lymphoma survivors had significantly higher BNP (42 pg/dL) and NT-proBNP (124 pg/dL) levels when compared to controls (BNP 29 pg/dL, p<0.01; NT-proBNP 44 pg/dL). There was no statistically significant difference in BNP and NT-proBNP levels between conventionally-treated and HCT survivors, or in ST-2 levels across all three groups. The sensitivity of blood biomarkers ranged from 6% [ST-2] to 28% [BNP]; specificity ranged from 86% [NT-proBNP] to 89% [BNP, ST-2]; positive predictive value ranged from 15% [ST-2] to 41% [BNP]; and negative predictive value ranged from 78% [ST-2] to 83% [BNP]; Table 3. There was no improvement in diagnostic accuracy of blood biomarkers according to type of cardiac dysfunction (systolic, diastolic, abnormal GLS, or by different cutoffs [BNP ≥50 pg/dL,²¹ NT-proBNP ≥125 pg/dL²²]; Supplementary Table 4).

Table 3.

Diagnostic yield of blood biomarkers for detection of cardiac dysfunction^* in lymphoma survivors

	BNP (>100 pg/dL)	NT-proBNP (>300 pg/dL)	ST-2 (>40 U/mL)
Sensitivity	28%	22%	6%
Specificity	89%	86%	89%
Positive Predictive Value	41%	29%	15%
Negative Predictive Value	83%	81%	78%

^*Defined as having either LV systolic dysfunction (LV EF <50%), diastolic dysfunction (mitral E/e′ >14), or abnormal strain (GLS >−16).

DISCUSSION

The principal findings from the current study were that one in every five long-term lymphoma survivors treated with anthracyclines had asymptomatic cardiac dysfunction; the odds of having cardiac dysfunction were seven-fold greater among lymphoma survivors when compared with age- and sex-matched controls without a history of cancer. We confirmed a dose-dependent increased risk of cardiac dysfunction by cumulative anthracycline dose, but found no difference in the prevalence or severity of cardiac dysfunction between conventionally-treated and HCT survivors. Lastly, while values of cardiac blood biomarkers were generally higher in survivors who had cardiac dysfunction compared to those with normal function, the diagnostic accuracy of these biomarkers in the asymptomatic setting was limited. The information from the current study can be used for the development of targeted long-term screening strategies in at risk survivors, and for development of interventional studies in at risk survivors aimed at improvement of cardiac function and prevention of symptomatic heart failure, with its attendant morbidity.

Large cohort studies have reported a 2- to 5-fold increased risk of symptomatic heart failure in lymphoma survivors when compared with the general population.^{10, 23–25} The latency of disease in these studies has varied (months to years) according to study design, differences in definitions used for the primary outcome, and treatment characteristics. Until now, very little was known about the prevalence of asymptomatic cardiac dysfunction in these long-term survivors. Recent studies have suggested that lymphoma survivors who have undergone autologous HCT may be at an especially high risk of developing symptomatic heart failure,^{9, 10} attributed to additional HCT-related conditioning chemotherapy and/or radiation exposures in this population. To our knowledge, no study has compared the prevalence of asymptomatic cardiac dysfunction in HCT survivors to conventionally-treated patients, and few have compared the prevalence to that in the general population.^{5, 26} The survivors in the current study were representative of the larger contemporary cohort of lymphoma patients treated with anthracyclines at our institution. Cumulative anthracycline exposure was equivalent between conventionally-treated and HCT survivors, and we found no difference in echocardiographic measures of cardiac function between the two groups. Not surprisingly, odds of having cardiac dysfunction in the current study (OR=6.6) was comparable to that reported in a recent study of long-term lymphoma survivors who had undergone HCT (OR=6.6),⁹ suggesting that our findings may be broadly applicable to other populations of anthracycline-exposed lymphoma survivors, regardless of HCT status.

Traditionally, detection of anthracycline-related cardiotoxicity has relied upon echocardiographic screening of LV systolic function such as EF.²⁷ However, LV EF has been increasingly recognized as a late-occurring measure of cardiac dysfunction,^{28, 29} and a poor predictor of future symptomatic disease.³⁰ Importantly, reliance on LV EF alone ignores the possibility of diastolic dysfunction in the setting of preserved EF. Newer echocardiographic parameters such as GLS may complement traditional measures of systolic and/or diastolic function, allowing more accurate prediction of future symptomatic disease.^{4, 6} For example, in breast cancer patients, screening strategies that incorporate EF + GLS have nearly twice the predictive power for future symptomatic disease when compared EF alone.^{31, 32} The current study is the first to utilize a comprehensive screening approach (systolic, diastolic, GLS) to assess cardiac function in anthracycline-exposed lymphoma survivors treated as adults, utilizing readily available echocardiographic screening modalities. The thresholds used to define systolic dysfunction (LV EF <50%),¹⁵ diastolic dysfunction (mitral E/e′ >14),¹⁶ or abnormal strain (GLS >−16)¹⁷ have been well-established in the general non-oncology community, and further supported by the low abnormal prevalence (range: 0% to 4%) seen in our control population. Surprisingly, we found that survivors treated with relatively modest doses of anthracyclines (1–249 mg/m²) had a nearly five-fold risk of cardiac dysfunction when compared to controls, suggesting that screening should not be limited to survivors treated with HD-anthracyclines alone. Importantly, there was little concordance between echocardiographic parameters (LV EF, mitral E/e′, GLS) in survivors with cardiac dysfunction, emphasizing the need for multi-faceted and comprehensive approaches to screening in this population. Larger studies are needed to examine the association between treatment exposures, comorbidities, and specific echocardiographic parameters (e.g. LV EF, mitral E/e′, GLS) of cardiac dysfunction.

Natriuretic peptides (BNP and NT-proBNP) are standard biomarkers used for the diagnosis and management of symptomatic heart failure in the non-oncology community.^{7, 8} However, due to the lack of large epidemiologic studies, current cardiology guidelines do not recommend the use of these biomarkers for primary screening in the asymptomatic setting.^{7, 8} Protein ST2, a newer biomarker and a member of the interleukin (IL)-1 receptor family, is secreted by cultured monocytes subjected to mechanical strain.³³ The ligand for this receptor is IL-33 which, like natriuretic peptides, is induced and released by stretched myocytes.³³ For cancer survivors treated with anthracyclines, the long latency of heart failure coupled with the cost and burden of echocardiographic screening make primary screening with these blood biomarkers an attractive option. In the current study, the diagnostic utility of these blood biomarkers was poor, confirming their lack of predictive value seen in other populations such as long-term childhood cancer survivors.^34–37

Unlike previous studies,^{23, 38, 39} we did not find a significant association between chest RT and cardiac dysfunction. These findings may be attributable to improvement in radiation techniques over time (e.g. better shielding of the myocardium, more precise delivery to area of disease involvement), the relatively low rate (16.1%) of chest RT use in our population, and even lower rate (9.6%) of HD-chest RT exposure in this contemporary cohort of lymphoma survivors. It is possible that differences in chest RT exposure between the HCT and non-HCT survivors may have impacted the interpretation of our findings; the small number of survivors (N=3) treated with chest RT who developed cardiac dysfunction limited our ability to do more detailed dose-specific analyses. We acknowledge that the cross-sectional design of our study and exclusion of survivors being actively treated for heart failure limited our ability to comment on the latency and incidence of cardiac dysfunction in this cohort. That said, more than 20% of survivors had significant asymptomatic cardiac dysfunction nearly 10 years after diagnosis, which may be an underestimation of the overall burden of cardiovascular disease in the nearly 600,000 lymphoma survivors living in the U.S.⁴⁰

The current study’s innovation lies in its ability to leverage existing information from non-oncology populations to develop a comprehensive assessment of cardiac function, setting the stage for novel screening strategies and interventions (pharmacologic, behavioral) in high risk survivors. In these long-term survivors, we found a high rate of cardiac dysfunction that was independent of HCT status, and a low yield of traditional cardiac blood biomarkers for primary screening of asymptomatic disease. The growing number of lymphoma survivors makes it imperative to identify reliable and cost-effective strategies to decrease the long-term burden of heart failure in this population.