Cancer Therapy-Related Cardiac Dysfunction and Heart Failure Part 2: Prevention, Treatment, Guidelines, and Future Directions

Abstract

Success with oncologic treatment has allowed cancer patients to experience longer cancer-free survival gains. Unfortunately, this success has been tempered by unintended and often devastating cardiac complications affecting overall patient outcomes. Cardiac toxicity, specifically the association of several cancer therapy agents with the development of left ventricular dysfunction and cardiomyopathy is an issue of growing concern. Though the pathophysiologic mechanisms behind cardiac toxicity have been characterized, there is currently no evidence-based approach for monitoring and management of these patients. In the first of a 2-part review, we discuss the epidemiologic, pathophysiologic, risk factors, and imaging aspects of cancer therapy related cardiac dysfunction and heart failure. In this second part, we discuss the prevention and treatment aspects in these patients and conclude with highlighting the evidence gaps and future directions for research in this area.

Mortality from cancer has decreased tremendously over the past few decades, in part through earlier diagnosis and novel treatments. Unfortunately, while cancer-free survival has increased, complications from cancer therapy, particularly effects of cardiac function have limited patient outcomes, impacting the overall morbidity and mortality adversely.¹ Heart failure (HF) as a result of cancer therapy has been linked to a 3.5 fold increased mortality risk compared with idiopathic cardiomyopathy². An integrative approach between the oncologist and cardiologist can aid in minimizing these detrimental effects. In the second part of this 2-part review, we discuss evaluation, surveillance, prevention, and treatment in this patient population. We highlight a proposed algorithm for approaching these patients before, during, and following cancer therapy. We conclude highlight challenges within the field and areas that need further research.

PREVENTION STRATEGIES

Beta-Blockers

There is growing evidence suggesting a cardioprotective role of beta-blockers in prevention of anthracycline-induced cardiotoxicity. Carvedilol, which also is an antioxidant and has the ability to chelate iron, prevented cardiac histopathology caused by doxorubicin.³ Carvedilol may prevent strain abnormalities after anthracycline use.⁴ In studies with carvedilol⁵ and nebivolol⁶ at initiation of anthracycline use, both agents resulted in higher degree of LVEF preservation. The use of beta-blockers during treatment with trastuzumab and anthracyclines was associated with a lower incidence of HF over a 5-year period.⁷ While carvedilol and nebivolol are beneficial, nonselective beta blockers such as propranolol may in fact be cardiotoxic⁸ and the effect of metoprolol is neutral.⁹

Renin-Angiotensin Inhibitors

Animal studies suggest that ACE-inhibitors (ACEI) may be cardioprotective in anthracycline toxicity.¹⁰ Enalapril treatment one week prior to doxorubicin and continued for three weeks after the last dose, preserved mitochondrial function and down-regulated free radical generation.¹¹ Beneficial mechanisms include attenuation of fibrosis and oxidative stress, and decreased angiotensin-induced blockade of the neuregulin (NRG)/ERb system.¹² Some data on ACEI benefit in reducing chemotherapy related HF have been disappointing,^{9, 13} though criticism of these trials included delay in initiation and broad enrollment of patients on multiple chemotherapy regimens. The benefit of a prophylactic ACEI strategy may be enhanced with risk stratification. Treatment with enalapril using troponin I elevation to identify and stratify high-risk anthracycline patients, prevented reduction in LVEF and the development of cardiotoxicity.¹⁴ The role of ACE-I’s in trastuzumab or radiation treated patients has not yet been extensively studied. Telmisartan, when administered immediately prior to epirubicin, may reduce the formation of reactive oxygen species and attenuate the development of myocardial dysfunction in patients receiving higher doses of epirubicin.¹⁵

Aldosterone Antagonist Therapy and Other Heart Failure Therapies

Aldosterone antagonists have been speculated to attenuate trastuzumab induced myocardial dysfunction through inhibition of the EGFR receptor, though further study is warranted.¹⁶ Spironolactone, used simultaneously with anthracyclines in breast cancer patients with preserved LV function, may attenuate LVD suggesting a role in prevention of both systolic and diastolic dysfunction.¹⁷ The role of other agents used in HF management such as hydralazine/nitrates and digoxin have not been studied in the cancer cardiotoxicity population, however guideline derived medical treatment in general should be employed in all patients with LVD.

Combination Therapy

The OVERCOME trial (Prevention of Left Ventricular Dysfunction with Enalapril and Carvedilol in Patients Submitted to Intensive Chemotherapy for the Treatment of Malignant Hemopathies) used a combination of carvedilol and enalapril in patients with leukemia or those planned for stem cell transplantation. In patients who received combination therapy, there was no reduction in LVEF compared those who received placebo. In addition, patients on combination therapy had a lower incidence of death or HF.¹⁸ Preliminary results of The Prevention of Cardiac Dysfunction during Adjuvant Breast Cancer Therapy (PRADA) trial revealed that candesartan, but not metoprolol tartrate, prevented a modest short-term decline in LVEF during anthracycline inclusive breast cancer therapy as measured by cardiac MRI. Longer-term follow up will further establish a preventative role of angiotensin receptor blockade.^{19, 20} MANTICORE-101 (Multidisciplinary Approach to Novel Therapies in Cardiology Oncology Research) is examining the use of perindopril versus bisoprolol in patients with HER2+ breast cancer undergoing treatment with trastuzumab in the prevention of LVD as assessed by cardiac MRI.²¹

Dexrazoxane

Dexrazoxane, a derivative of the metal-chelating agent ethyl enediamine tetra acetic acid (EDTA), is thought to attenuate anthracycline cardiac toxicity through iron chelation and decrease in production of free radicals.²² In addition, dexrazoxane binds to topoisomerase 2, preventing the formation of anthracycline-mediated DNA-Top2 complexes.²³ In patients treated with anthracyclines, dexrazoxane decreased HF risk and increased cardiac event-free survival.²⁴ However, the mechanism by which dexrazoxane provides cardioprotection has raised concerns that this agent may attenuate doxorubicin anti-tumor activity, through binding to both Top2-alpha and Top2-beta.²⁵ Clinical trial data are inconclusive but a Cochrane review demonstrated no difference in efficacy of anthracyclines against the primary malignancy with the addition of dexrazoxane.²³ The American Society of Clinical Oncology (ASCO) recommends that use of dexrazoxane should be limited to adult patients in the metastatic breast cancer setting and other malignancies who have received >300 mg/m2 and who may benefit from use of additional anthracyclines.²⁶ Recent data demonstrated long-term efficacy in reducing HF in cancer survivors from the pediatric cohort,²⁷ and therefore new studies are required to determine safety and efficacy in adult cancer populations. A meta-analysis on the prophylactic use of dexrazoxane in patients receiving anthracyclines revealed a decrease in cardiac events.²⁸

Statins

HMG-CoA reductase inhibitors, in addition to their lipid lowering, exert cardioprotective effects through pleotrophic mechanisms. Studies suggest the benefit of statins in reducing anthracycline mediated cardiomyocyte death.²⁹ In a retrospective cohort study of over 600 cancer patients, uninterrupted statin use resulted in reduced HF.³⁰ In patients without preexisting cardiovascular abnormalities, prophylactic atorvastatin led to higher preservation of LVEF.³¹ To date, no prospective trials have addressed the role of statins in the prevention of cancer therapy related cardiotoxicity. (Table 1)

Table 1.

Prevention of Left Ventricular Dysfunction in Patients with Cardiotoxic Chemotherapy

Reference	Type of Study	Patient Population and Cancer Therapy	N	Cardiac Treatment Modality	Timing of Initiation of Treatment	Mean Follow up	Results
Beta Blockers
Kalay, et al. 2006 5	RCT	Patients planned to receive anthracyclines planned (breast, lymphoma, other) Adriamycin or epirubicin	50	Carvedilol 12.5 mg once daily vs. placebo	Initiated prior to chemo	6 months	Mean EF 68.9 vs. 52.3; (p< 0.001) at 6 month follow up Both systolic and diastolic diameters were significantly increased in control, and diastolic parameters (E vel and E/A ratios sig reduced in control group) Lower mortality (but not statistically significant) in carvedilol group
Seicean, et al 2013 7	Propensity matched control Competing risk framework	Breast cancer pts. with nl EF before between 2005 – 2010 Anthracyclines or anthracyclines followed by trastuzumab with or without radiotherapy	920 (318 used )	Continuou s BB (n=106) vs. not on continuous BB (n=212)	BB initiated prior to cancer rx	3.2 years	Coincidental continuous use of BB associated with lower incidence of new HF in breast cancer pts. with normal EF during median follow of 5.2 years (hazard ratio, 0.2; CI 0.1–0.5, p=0.003) HF events 5 vs. 27 (p=0.008) in cont. BB vs. no BB group
Kaya et al. 2013 6	RCT	Breast cancer and planned chemotherapy	45	Nebivolol 5 mg daily vs. placebo		Echocardiogram and NT pro-BNP at baseline and 6 months of chemo	At 6 months, LVESD and LVEDD increased in placebo group but remained unchanged in nebivolol group (p=0.01 for both) Placebo group had lower LVEF than nebivolol at 6 months NT BNP increased in placebo group
Elitok, et al. 2014 4	RCT	Breast cancer patients planned for anthracyclines	80	Carvedilol vs. placebo	Carvedilol 12.5 mg daily for 6 months	Echo with strain at baseline and 6 month post anthracycline	Significant decrease in strain parameters in control group at 6 months
RAAS inhibitors
Nakamae, et al. 2005 32	RCT	Patients scheduled to undergo standard chemo with CHOP (cyclophosphamide, doxorubicin, vincristine, prednisolone)	40	Valsartan 80 mg daily vs. none	ARB simultaneous on day 1 of CHOP	Neurohormonal, echo, and ecg parameters measured before, days 3,5,7 after initiation of CHOP 7 days	Valsartan significantly prevented transient increases in LVEDd, QTc dispersion and BNP elevation (p<0.05) with no significant change in BP or HR Ang II may play a role in acute chemo-induced cardiotoxicity and ARB may prevent acute CHOP toxicity
Cardinale, et al. 2006 14	RCT	High risk high dose chemo patients (defined by early increased troponin I level) from 2002 to 2004 Included primary resistant BC, AML, relapsed or refractory Hodgkins lymphoma, Ewing’s sarcoma	114	Enalapril 20 mg daily vs. none	1 month after last high dose chemo, continued for 1 year	Cardiac eval including: Late TnI (at randomization, b/f enalapril, 2,3,6,12 mo later) at baseline, 1,2,6, and 12 months after high dose chemo	Absolute decrease in LVEF >10% to decline below normal value (LVEF 50%) (43 vs. 0%) and increase in end diastolic and end-systolic volumes only in untreated pts. (p<0.001)
Cadeddu, et al. 2010 15	RCT, placebo cont.	Patients planned to undergo chemo with epirubicin (cumulative dose 400 mg/m2) based chemo with baseline EF >55% and no history of heart disease, HTN, DM	49	Telmisarta n 40 mg daily vs. placebo	1 week prior to chemo	Echo, TD, strain/ strain rate (SR) and plasma levels of inflammatory and oxidative stress markers at baseline and at 7 days after every new epirubicin dose of 100 mg/m2 3 months	Impairment in strain rate peak at epi dose of 200 mg/m2 (no sig diff b/w groups) but at 300 and 400 mg/m2 SR normalized only in telmisartan group (p<0.001) Significant increase in ROS and IL-6 in placebo
Combination Neurohormonal Blockade
Bosch et al. 2013 18	RCT	Acute leukemia or malignancies planned for HSCT Without LVSD	90	Enalapril 10 mg BID and carvedilol 25 mg BID vs. control	Started simultaneousl y at least 24 hours before 1st chemo cycle	Echo and CMR before and at 6 months after randomization, trop I and BNP at baseline and 12 hours after each cycle 6 month	LVEF significantly decreased in controls Compared to controls patients in treatment group had lower incidence of combined end point of death or HF (6.7% vs. 22%, p=0.036) and of death, HF, or LVEF <45% (6.7% vs. 24.4%, p= 0.02) Marked differences in the acute leukemia group
Georgakop oulos, et al. 2010 9	RCT	Lymphoma patients (HL and NHL)	125	Metoprolol vs. Enalapril vs. placebo		Echo at baseline and 12 months 1 year follow up	Non-signficant reduction in HF, primarily in metoprolol group 16% early cardiotoxicity 7.3% late cardiotoxicity (more in elderly) Overall negative study: Metoprolol and enalapril do not reduce risk of cardiotooxicity in pts. with doxorubicin
Statins
Acar, et al. 2011 31	RCT placebo cont.	Patients undergoing anthracycline chemotherapy without previous cardiac history and regardless of lipid levels Adriamycin or idarubicin	40	Atorvastati n 40 mg qd vs. placebo	Prior to chemotherapy	Echo at baseline and 6 mo after chemotherapy 6 months	Statin vs. control at 6 months: LVEF: +1.3% vs. −7.9% in statin vs. control (p <0.001) LVEDD (mm): 0.15 vs. + 2.0 (p 0.021) LVESD -1.35 vs. 2.1 (p<0.001) Note: primary endpoint of LVEF <50% did not reach statistical sig
Seicean, et al 2012 30	Observation al Clinical Cohort Study	628 women with newly diagnosed breast cancer Anthracyclines +/- trastuzumab	628 (201 mat ched )	Uninterrup ted statin therapy	Statin therapy initiated incidentally prior to cancer therapy		New onset HF in 67 of 201 matched pts. HR sig lower 0.3 (p=0.02) for statin pts. Cardiotoxicity RFs: Baseline EF <55% Trastuzumab use

EF, ejection fraction; GLS, global longitudinal strain; HF, heart failure; HFrEF, heart failure reduced ejection fraction; HL, Hodgkin’s lymphoma; LVSD, left ventricular systolic dysfunction; NHL, non-Hodgkin’s lymphoma; RF, risk factor; yoa, years of age

Primordial versus Primary versus Secondary Prevention

Prevention may be primary, extended to all patients already treated with potentially cardiotoxic therapies, or secondary in selected high-risk patients showing preclinical signs of cardiotoxicity as in the form of biomarker increase, strain decrease, etc. More recently, the concept of primordial prevention with cardioprotective therapies has been described to address the population of patients immediately following initial cancer diagnosis but preceding cancer therapy.³³ A tailored prevention strategy based on the cardiac risk stratification according to patient-related (including genetic predisposition) and to therapy-related risk factors bears further investigation.

TREATMENT OF LEFT VENTRICULAR DYSFUNCTION AND HEART FAILURE

Conventional therapy with beta-blockers and ACEI in LVD and HF is extrapolated to patients with chemotherapy-induced cardiotoxicity, though randomized trial data are lacking. Enalapril and carvedilol given 1–2 months post anthracycline therapy in patients with LVD led to LVEF recovery in most cases, however when treatment was delayed, recovery was partial or absent.³⁴ In another study patients with LVD, regardless of the presence of symptoms, were given either enalapril (before 1999) or enalapril and beta-blocker (carvedilol or bisoprolol after 1999), and showed recovery of LVEF with greater improvement with combined therapy.³⁵ The role of HF treatment in trastuzumab-induced LVD has not been established. (Table 2)

Table 2.

Treatment of Asymptomatic Left Ventricular Dysfunction in Adult Patients with Cardiotoxic Chemotherapy

Reference	Type of Study	Patient Population an Cancer Therapy	N	Cardiac Treatment Modality	Timing of Initiation of Treatment	Mean Follow up	Results
Cardinale et al. 2010 34	Prospective	201 pts. with LVEF </= 45% due to anthracylcines	201	Enalapril up to 20 mg/day and coreg up to 50 mg / day of note: mean dose enalapril 11 mg/day and coreg 14 mg /d	Rx initiated immediately after detection of ASLVD	Echo at baseline, every month for 3 months, and every 3 months during following 3 years, every 6 months thereafter Mean follow-up 36 months	Primary EP: LVEF response to therapy Responders 42%, Partial responders 13%, Non-responders 45% Responders showed lower rate of cum cardiac events than partial and non- responders (5%, 31%, 29% p < 0.001) *responders had a significantly shorter time to initiation of therapy
Cardinale, et al. 2015 35	Prospective	Mix of cancer, non-Hodgkin’s lymphoma Chemotherapy naïve patients scheduled for anthracyclines (note excluded high dose anthracycline or trastuzumab)	2625	Enalapril alone (before 1999) enalapril and beta blockers carvedilol or bisoprolol (after 1999)	Therapy promptly administered and uptitrated to maximal tolerated doses	Echo at baseline, every 3 months during chemotherapy, at end of treatment (within one month, every 3 months during first-year following chemotherapy, every 6 months during the following 40 years, yearly afterward) Median follow-up 5.2 years	Anthracycline-induced cardiotoxicity occurred in 9% of adult treated patient’s (Dose dependent; Highest incidence in first-year after completion of chemotherapy) Median time between last dose of anthracycline and development of cardiotoxicity was 3.5 months, 98% of cases within the first-year follow-up 82% of patients recovered from cardiotoxicity (11% full recovery; 71% partial recovery)

ASLVD, asymptomatic left ventricular dysfunction; EP, end point; LVEF, left ventricular ejection fraction

SURVEILLANCE: PRE, DURING, and POST CANCER TREATMENT

Surveillance strategies are currently based on expert consensus. One approach includes screening for high-risk cardiac and oncologic factors to stratify patients. Optimization of preexisting conditions prior to cancer therapy should be attempted. Baseline detailed echocardiogram, including strain imaging when available, should be performed.^{36, 37} The presence of high-risk features should prompt cardio-oncology consultation. Incorporation of biomarkers is recommended, though their role in routine monitoring is not fully established.^{37, 38} The frequency and intervals with which to image patients varies with risk factors, type of chemotherapy, and planned dosing,³⁸ and is currently center-dependent. Radiation therapy requires a different surveillance algorithm,^{38, 39} due to the potential for delayed cardiac dysfunction. (Figure 1)

Figure 1.

Proposed Algorithm for the Surveillance and Treatment of Cardiotoxicity in Patients Receiving Potentially Cardiotoxic Chemotherapy Prior to (A), During (B), and Following (C) Therapy.

^*Continuous variables of risk

¹ MUGA may be considered if echocardiography or CMR not available

² For high-risk patients or when available, strain-imaging. Use of Echo contrast when indicated

³ 65–74 may represent an intermediate risk group

⁴ Troponin, BNP

⁵ Consider earlier imaging if higher baseline risk

Not all ASLVD and HF that develops during or after cancer therapy can be assumed to be the direct result of treatment, and thus all new or worsening cardiac dysfunction should be evaluated according to guideline recommendations. This should include evaluation for other reversible causes of myocardial dysfunction and ischemic evaluation in those patients in whom suspicion is high.⁴⁰ Higher suspicion for alternate causes is especially important in both older individuals and those with multiple comorbidities.³⁸

ALTERATIONS IN CANCER THERAPY IN PATIENTS WITH VENTRICULAR DYSFUNCTION AND HEART FAILURE

The Food and Drug Administration recommends withholding trastuzumab for at least 4 weeks if LVEF drops ≥16% from pre-treatment values or if it falls below normal and ≥10% absolute decrease in LVEF from pretreatment values. The agent can be resumed if the LVEF returns to normal and the absolute decrease from baseline is ≤15% within 4–8 weeks.⁴¹ It is also recommended to discontinue doxorubicin in patients who develop HF.⁴² The indications for withdrawal or withholding of therapy in patients with asymptomatic LVEF decline are not as clear. The Canadian Trastuzumab working group recommendations for “stopping/restarting” directly reflect those utilized in the largest adjuvant trastuzumab trials, based on baseline systolic function and degree of LVEF % decline.⁴³ Decisions on alteration in dosing or discontinuation of cancer treatment need to be weighed against oncologic risk, and requires collaboration between cardiologists and oncologists, personalizing the strategy to the individual patient and their prognosis. Predictors of LV recovery are poorly understood. Retrospective data has suggested that LV recovery can be expected in roughly 50% of patients who develop cancer therapy related cardiotoxicity, and that younger age, smaller left atrial volume, and lower BNP levels may be multivariate predictors of LV recovery.⁴⁴

ADVANCED HEART FAILURE THERAPIES IN THE CANCER POPULATION

Similar to the non-cancer heart failure population, option for therapies in the chemotherapy-induced cardiomyopathy population include implantable cardioverter defibrillator (ICD) and chronic resynchronization therapy (CRT), left ventricular assist devices, and orthotopic heart transplantation (OHT). In a recent study, Adriamycin cardiomyopathy patients derived a similar echocardiographic and clinical benefit with CRT compared with other non-ischemic patients.⁴⁵ About 2.5% of non-ischemic cardiomyopathy patients undergoing transplantation have chemotherapy related cardiotoxicity and survival is comparable to other non-ischemic etiologies. ⁴⁶ Not surprisingly, cancer patients with cardiac toxicity are more likely to undergo mechanical circulatory support (MCS) as destination therapy rather than as bridge to transplant, given the obligatory five-year cancer-free survival as a requisite for transplant candidacy. Overall survival after MCS in these patients is 73% at 1-year and 63% at 5 years, also similar to the general population.⁴⁷ Importantly, chemotherapy-related cardiomyopathy patients have higher rates of RV dysfunction, and thus are more likely to require RV mechanical support. Nevertheless, cancer patients requiring MCS and OHT should be evaluated and considered appropriately for these advanced therapies.

CURRENT GUIDELINES

Several groups have published recommendations and consensus statements in the adult cancer population,^{36–38, 48} but formal guidelines for prevention, surveillance, and treatment of cancer therapy related cardiac toxicity are not yet available. Currently, the American College of Cardiology/American Heart Association HF guidelines report that agents such as anthracyclines, trastuzumab, cyclophosphamide, taxanes, 5-fluorouracil and interferons may cause cardiotoxicity, and dexrazoxane is cardioprotective against anthracycline-induced cardiotoxicity.⁴⁹ However there are no specific monitoring recommendations. The European Society of Cardiology HF guidelines are similar and also recommend pre- and post-therapy LVEF evaluation and discontinuation of chemotherapy along with initiation of standard HF therapy once LVD ensues.⁵⁰ The Canadian Trastuzumab Working group recommends baseline and 3-month interval imaging assessment for all trastuzumab patients, advocating for more frequent/stringent assessments in those patients at higher risk.⁵¹

The American Society of Echocardiography suggests echocardiography for baseline and follow-up monitoring, but there are no specific recommendations regarding frequency or duration of follow up.⁵² ASCO recommends dexrazoxane only in metastatic patients who have received more than 300mg/m² doxorubicin and would benefit from additional anthracycline administration.²⁶ ASCO reports that there is currently no standard strategy for long-term monitoring through biomarkers or imaging and no direct evidence regarding the treatment of LVD in asymptomatic patients.¹ (Table 3)

Table 3.

Guideline Recommendations for Cardiac Monitoring

Guideline	Year	Recommendation	Level of Evidence
Cardiology
American College of Cardiology/ American Heart Association Management of Heart Failure 49	2013	The incidence and reversibility of chemotherapy-related cardiotoxicity are not well documented, and meaningful interventions to prevent injury have not yet been elucidated	Not Stated
European Society of Cardiology 50	2012	Pre- and post-evaluation of EF is essential in patients receiving cardiotoxic chemotherapy Patients developing LVSD should not receive further chemotherapy and should receive standard treatment for HFrEF	Not Stated
American Society of Echocardiography 52	2003	Baseline and re-evaluation examinations in patients receiving cardiotoxic chemotherapeutic agents	Class I
Oncology
Canadian Trastuzumab Working Group 43	2008	Cardiac imaging (echo or MUGA) at baseline and 3-month intervals until completion of therapy at minimum with more frequent/ stringent monitoring for higher risk patients	Not Stated
American Society of Clinical Oncology: Cardiac and Pulmonary Late Effects 1	2007	The optimal duration, frequency, and method of cardiac monitoring during trastuzumab and anthracycline treatment remains unknown	Not Stated
Position Statements
American Society of Echocardiography/ European Association of Cardiovascular Imaging: Multimodality Imaging Evaluation 36	2014	Treatment with anthracycline → baseline LVEF assessment with 3D or 2D Echo, GLS, and Troponin I measurement. If abnormal, cardiology consultation. Follow-up at completion of therapy and 6 months later for doses <240mg/m2 Treatment with trastuzumab → baseline LVEF assessment with 3D or 2D Echo, GLS, and Troponin I measurement. If abnormal, cardiology consultation. Follow-up every 3 months and 6 months later.	Not Stated
ESMO Clinical Practice Guidelines 37	2012	In patients receiving anthracyclines +/− trastuzumab→Serial monitoring of cardiac function at baseline, 3, 6, and 9 months during treatment, 12 and 18 months after start of treatment	Level I, Grade A
		In patients with metastatic disease→monitor EF at baseline and infrequently in absence of symptoms	Level II, Grade A
		Measurement of Troponin, BNP at baseline and periodically during therapy	Level III, Grade B
		Cardiac function assessment 4–10 years after anthracycline in patients treated at <15 yoa or >15 with cumulative dose doxorubicin >240mg/m2	Level II, Grade B
		LVEF drop <50% during anthracycline-containing →reassess in 3 weeks. If confirmed, hold chemotherapy and consider therapy for LVSD	Level II, Grade B
		LVEF drop <50% during trastuzumab therapy →reassess in 3 weeks. If confirmed, continue trastuzumab and consider therapy for LVSD	Level II, Grade B
Heart Failure Association of the European Society of Cardiology: Cardiovascular Side Effects of Cancer Therapies 48	2011	Regular cardiovascular evaluation should be part of routine care in patients receiving treatment regimens known to be associated with cardiotoxicity Follow-up beyond completion of therapy should be considered, particularly in those receiving high doses of anthracyclines Use of Troponin and BNP should be strongly considered	Not Stated

EF, ejection fraction; GLS, global longitudinal strain; HFrEF, heart failure reduced ejection fraction; LVSD, left ventricular systolic dysfunction; yoa, years of age

KNOWLEDGE GAPS AND FUTURE DIRECTIONS

The specialty of cardio-oncology has gained significant momentum, with increasing awareness and interest in advancing the field. This parallels the larger armamentarium of therapies now available to cancer patients, many of which have redefined life expectancy. There are, however, multiple gaps in the field, which bear addressing. (Table 4) At present, there are no internationally published guidelines to address this specific patient population, and no standardized classification system to define cancer related cardiac toxicity, LVD, and HF. Cardio-oncology guidelines will need to account for different subpopulations such as those with metastatic and non-metastatic disease. Furthermore, defining cardiac dysfunction through LVEF alone is insufficient. LVEF estimation may predict development of later cardiotoxicity but may not be sensitive enough to assess early pre-clinical changes, which might impact on management decisions. Currently, incidence of cancer therapy related LVD and HF are likely underestimated, representing the typically younger and healthier population in most larger cancer trials. Standardization of cardiac toxicity definitions will allow for prospective study of epidemiology.

Table 4.

Future Directions in the Field of Cardio-Oncology

Domain	Problem	Potential Solution
Definition	Lack of universally accepted definition of cardiac toxicity Definition of cardiac dysfunction through LVEF alone is insufficient	Consensus-driven definition for use in prospective clinical trials and practice
Disease Entity	Limited understanding of the pathophysiologic mechanisms of various chemotherapy agents	More basic science research to better define pathophysiology of cancer therapeutics
Guidelines	No internationally published guidelines to account for different subsets of patient populations	More large prospective clinical trials to influence evidence-based guideline development
Epidemiology	Probable underestimation of incidence as most trials include younger, healthier population Focus on systolic dysfunction	Inclusion of older and/or higher risk populations in clinical trials Study HF with both reduced and preserved LVEF
Risk Stratification	No universal validated risk prediction mod	Development of validated model based on different clinical profiles
Screening and Surveillance	Lack of universal agreement on best practice for screening and surveillance interval	Development of prospective monitoring paradigm
Prevention	Lack of data to support prevention of cardiotoxicity and the role for prophylactic medical therapy	More trials for prevention of cardiotoxicity with attention to primary versus secondary prevention strategies
Surrogate Markers	Focus on primarily troponin and natriuretic peptide	Further studies on novel biomarker approaches
Treatment	No established treatment in this population; lack of data on safety and feasibility of re-challenging with chemotherapy once LVD ensues	More clinical trials to address role of single agent versus combination therapy
Novel Approaches	Limited data on novel approaches for management, e.g. the role of cardiac progenitor cells in treatment or the role of genetic polymorphisms in cardiotoxicity	Expansion of data from animal studies to human subjects Better definition of the role for genetic testing

LVEF, left ventricular ejection fraction; LVD, left ventricular dysfunction

There are various limitations in our understanding of optimal clinical management of cardiac disease in the cancer population. At present, the largest focus has centered on heart failure reduced ejection fraction (HFrEF), but the incidence and management of heart failure preserved ejection fraction (HFpEF) is largely unknown. Clinical trials addressing prevention, prophylactic medical therapy, length and types of therapy once cardiotoxicity develops, and the safety of rechallenging with cancer therapy, all remain critical unaddressed issues. Furthermore, prospectively validated risk predictive models would help clinicians to individualize care, tailor biomarker and imaging surveillance strategies, and initiate early or prophylactic medical therapy for those patients in highest risk categories. Another issue is cardioprotection during reinitiation of the culprit chemotherapy in the context of optimizing cancer outcomes for patients with previous cancer therapy-related cardiotoxicity.

There is a lack of universal agreement on prospective cardiovascular screening, especially for newer chemotherapeutic agents and for treatment strategies that require dual or additive therapy. Further understanding of drug mechanisms would allow for more targeted prevention and treatment. Echocardiographic imaging remains critical to surveillance. The addition of strain/speckle tracking shows promise but the additive role of this technique in routine practice requires further definition.

There is limited understanding on the long-term effect of modest LVEF declines during therapy, the implications of dose interruptions and treatment delays in cancer therapy, and how cardiac interventions impact on long term cardiac and cancer survival. Altering cancer treatment without strong supportive data may put patients at risk of under-treatment or decreased efficacy of therapy. While all patients receiving anthracyclines should be considered to have some degree of cardiac toxicity, choosing which patients command dose alterations or treatment discontinuation remains arbitrary, with potentially far reaching implications on survival. The term “actionable cardiotoxicity” has been put forth to address the “when” and “if” of treatment alterations.⁵³

Cardiac progenitor cells may play a role in the treatment and prevention of anthracycline-induced cardiotoxicity, and may be considered as a potential translational therapy in the future, helping to promote cardiac repair. Autologous cardiac progenitor cells can be obtained before antineoplastic drugs are given to cancer patients and subsequently administered to individuals who are particularly predisposed to cardiotoxicity.⁵⁴

Significant individual variability in tolerance to cumulative anthracycline dose has suggested a role for genetic susceptibility. The clinical significance of this remains unknown, but with further study may allow for more personalized oncologic therapy.

CONCLUSION

With survival gains in cancer therapy, attention and recognition of cardiac toxicities in cancer patients has become increasingly critical. Cancer patients, either with preexisting cardiac disease or increased cardiac risk, require individualized risk stratification strategies. Patients who develop myocardial dysfunction during therapy often require modifications or withdrawal of life-saving cancer therapies, with profound implications on clinical outcome. Preclinical identification of cardiac toxicities may allow oncologists to continue cancer therapy without interruption. An evidence-based approach would allow enhanced delivery of care to this patient population. This can only be accomplished by further investigation and through a partnership between cardiologist and oncologist, increasingly delivered via formal cardio-oncology services with access to a multidisciplinary team effort to ensure optimal patient outcomes.