The Prevention of Cardiovascular Disease in Cancer Survivors

Abstract

The number of cancer survivors in the United States has exceeded 12 million and is increasing. After secondary malignancies, cardiovascular disease is the leading cause of late morbidity and death among cancer survivors. The cardiovascular needs of cancer survivors have not been described.

We describe the clinical characteristics of 53 patients seen during the first year of our Cardiovascular Prevention in Cancer Survivors clinic. The mean age of the patients was 40.1 ± 13.7 years. The mean survival since cancer diagnosis was 13.9 years. A history of chemotherapy-induced cardiomyopathy was present in 21%, and 5.7% had known atherosclerotic disease. One fourth had hypertension; 32.1%, dyslipidemia; and 13%, diabetes mellitus. Three quarters had received anthracycline chemotherapy, while half had received radiation. Half had an abnormal echocardiogram (55%), and 11 of 18 had an abnormal carotid ultrasonogram. The mean Framingham risk score for patients older than 30 years (n=37) was 8.4, yielding a 10-year risk of cardiovascular disease of 7.6%. The mean vascular age was 54.3 years, and the mean chronological age was 46.3 years. The mean follow-up duration was 566 ± 213 days. There were significant improvements in serum triglycerides and high-density lipoprotein levels, as well as trends toward improved blood pressure control.

Cardiovascular risk factors are prevalent in cancer survivors. There is an immediate need for the widespread availability of cardiovascular preventive services to reduce the late adverse effects of chemotherapy and radiation. Early intervention might help to improve the cardiovascular risk profile.

The fight against cancer is ongoing and achieving better results than ever before. Adjusted survival rates of cancer patients have gone up significantly: from 50% in 1975 through 1977, to 68% in 1999 through 2005¹; today, 2 out of every 3 adults and 4 out of every 5 children diagnosed with cancer are expected to survive at least 5 years.² This has created a new population subset in need of specialized healthcare programs, specific for their condition and for the late effects of their treatment.³ This population of cancer survivors has already exceeded 12 million in the United States⁴ and is expected to continue increasing in the coming years, mostly due to use of screening tests, targeted and effective cancer treatment, and the general aging of the population.^3,5

After second malignancies, cardiovascular (CV) disease is the leading cause of long-term morbidity and death among cancer survivors. A study by Mertens and colleagues⁶ concluded that CV events are the leading nonmalignant cause of death among survivors of childhood cancers, responsible for a 7-fold higher risk of death among such patients when compared with their age-matched peers. The underlying cause is thought to be late effects of cardiotoxic cancer therapy. The CV side effects of radiation, as well as of some specific chemotherapeutic/anticancer agents, are well described.⁷ Other agents are suspected of causing CV damage, on the basis of clinical or experimental observations.⁸ Reducing the impact of late effects requires early detection of subclinical disease and timely administration of effective therapy. Hence, there might be an advantage in screening patients regularly in the course of their treatment, and again after they have completed cancer therapy.^7,9 On the other hand, Enright and Krzyzanowska¹⁰ have shown the suboptimal control afforded by the more traditional cardiac risk-factors measures, such as blood pressure and cholesterol monitoring, among cancer survivors. This has reinforced an awareness of the need for a specific individualized CV disease prevention program for cancer survivors.

To meet these preventive needs, the Cardiovascular Prevention in Cancer Survivors (CPCS) clinic was established by the Department of Cardiology at The University of Texas MD Anderson Cancer Center, in January 2008. This study describes the clinical characteristics of patients seen in the CPCS clinic as representative of the overall cancer survivor population, which is an inherently heterogeneous group.

We hypothesize that, as a group, cancer survivors will have poorer CV health as evidenced by the presence of risk factors and subclinical disease and by a higher vascular than chronological age.

Patients and Methods

This study was approved by our institutional review board. All patients visiting the CPCS clinic in its first year (calendar year 2008) were included. Data were collected retrospectively from electronic medical records. The following variables were included: age, sex, age at diagnosis of cancer, cancer diagnosis, type of chemotherapy, radiation therapy, and CV risk factors like hypertension, diabetes mellitus, atherosclerosis, smoking history, lipid profiles, and C-reactive protein. Echocardiographic data and carotid ultrasonographic findings for all patients were also collected, as were data (when available) on the left ventricular ejection fraction (LVEF), right ventricular systolic pressure, diastolic dysfunction, presence of carotid atherosclerotic plaque, and carotid intima–medial thickness. Echocardiograms were classified on the basis of abnormal findings into normal, mildly abnormal, and significantly abnormal overall.

In addition, we calculated the Framingham risk score, the 10-year general CV disease risk, and vascular age, as well as the difference between chronological and vascular age using validated calculators.¹¹

Statistical Analysis

Descriptive statistics were generated. Continuous variables were represented as mean ± SD, and categorical variables were represented as frequencies. Comparison of clinical variables at initial visit and last follow-up visit were performed using a paired t test. The significance level was set at P <0.05. All analyses were performed using NCSS 2000 software (Kaysville, Utah).

Results

Demographics

Included in the study were the 53 patients seen at the CPCS clinic at the MD Anderson Cancer Center during its first year. Their mean age was 40.1 ± 13.7 years. The mean age at cancer diagnosis was 26.1 ± 18.3 years. The mean survival time since cancer diagnosis was 13.9 years (range, 1–35 yr). Almost 3 of every 4 patients (73.6%) were women. Most patients (68.5%) were new, while the others (31.5%) were undergoing the transition to survivorship care.

Clinical Characteristics

A history of hematologic malignancy was found in half of the survivors (27/53). Hodgkin lymphoma and breast cancer were the most represented single cancers (26.4% each) (Table I).

TABLE I. Demographic and Clinical Characteristics of the 53 Patients

A history of chemotherapy-induced cardiomyopathy was present in 21% of the patients, and 5.7% had known atherosclerotic disease. Cardiovascular risk factors were prevalent in the study group. One fourth of the patients were hypertensive (26.4%), 32.1% had dyslipidemia, and 13% had diabetes mellitus. More than three quarters of the patients (75.5%) had received anthracycline-based chemotherapy, and half (51%) had received either left-breast or mediastinal radiation.

More than half of the patients had an abnormal echocardiogram (55%), and 11 out of 18 had an abnormal carotid ultrasonogram. The mean LVEF was 0.57 ± 0.06, and more than a quarter (27.5%) of survivors had abnormal diastolic function. Cardiac medications were frequently prescribed: nearly one third received aspirin (32.7%); another third, a β-blocker (30.8%); and more than a quarter, an angiotensin-converting enzyme (ACE) inhibitor (28.8%) or an HMG-CoA reductase inhibitor (statin, 26.9%).

A transthoracic echocardiogram was obtained in 51 patients, lipid profiles in 39, C-reactive protein levels in 15, and carotid ultrasonograms in 18 patients. Most of these patients had overall abnormal echocardiographic ratings (28/51 or 55%) and abnormal carotid ultrasonograms (11/18 or 61%), as shown in Tables II and III.

TABLE II. Cardiovascular Risk Factors, Biomarkers, and Lipid Profiles of the 53 Patients

TABLE III. Echocardiographic and Carotid Ultrasonographic Findings

Framingham Risk Scores, 10-Year General Cardiovascular Risk, and Vascular Age

The mean Framingham risk score for patients older than 30 years (n=37) was 8.4. This confers a mean 10-year risk of general CV disease at 7.6%. The mean vascular age for this group was 54.3 years, as opposed to the mean chronological age of 46.3 years. The mean difference between vascular and chronological age was 8 years (Table IV).

TABLE IV. Framingham Risk Scores, 10-Year General Cardiovascular Disease Risk, Vascular Age, and Variance of Chronological and Vascular Age in the Subgroup of Patients Older than 30 Years (n=37)

Follow-Up Duration and Changes in Clinical Variables over Time

The mean follow-up duration was 566 ± 213 days. Reductions in body weight, systolic and diastolic blood pressure, and levels of serum triglycerides, total cholesterol, and low-density-lipoprotein cholesterol were noted. On average, an increase in high-density-lipoprotein cholesterol and a marginal improvement in LVEF were also seen (Table V). Over the follow-up duration, there were 2 admissions for heart failure, and one for myocardial infarction. No strokes or deaths were reported.

TABLE V. Follow-up Duration and Changes in Clinical Values over Time

Discussion

The study shows the relatively high prevalence of CV comorbidities, as well as the risk factors for adverse CV events in cancer survivors. In addition to the risk factors shared with CV disease (like smoking, obesity, and increasing age¹⁰), CV risk factors associated with cancer therapy are prevalent as well. This emphasizes the need for CV health maintenance and a well-defined CV prevention program for all at-risk cancer survivors.

Although the mean LVEF was 0.57 ± 0.06, more than a quarter of survivors (27.5%) had abnormal diastolic function. This usually subclinical dysfunction might indicate an early stage in the overall disease process and, as a consequence, present an opportunity to prevent or delay overt heart failure. The precise role of routine noninvasive cardiac testing in cancer survivors is not clear,⁹ but it is an important avenue for future research.

The unusually high Framingham risk scores, the 10-year general CV risk (7.6%), and the mean vascular age of patients older than 30 years (8 years higher than chronological age) suggest that the underlying risk for CV disease in the cancer survivors is elevated.

Patients exposed to chemotherapeutic agents known to predispose to heart failure—like anthracyclines, trastuzumab, sunitinib, and sorafenib^12–14—should be screened for the stage of their heart failure on the basis of the American College of Cardiology/American Heart Association (ACC/AHA) guidelines.¹⁵ Because all patients receiving these drugs are at risk for the development of heart failure, they should be considered to have stage A heart failure, defined as LVEF ≥0.50, a normal electrocardiogram (ECG), and no evidence of LV hypertrophy by echocardiography. If they remain asymptomatic but experience a fall in LVEF below 0.50, an abnormal ECG, or echocardiographic evidence of abnormal diastolic function or LV hypertrophy (any of these by itself or in combination), they would be considered to have stage B heart failure. Progression to symptomatic (stage C) and refractory (stage D) heart failure might follow.

Jarfelt and colleagues¹⁶ have shown echocardiographic evidence of cardiac systolic dysfunction in asymptomatic survivors of cancers at a mean age of 27 years. However, these patients had been diagnosed in childhood and the average follow-up time was 20 years. In addition, data from studies conducted on pediatric cancer survivors suggest that earlier age at diagnosis of cancer is a risk factor for CV dysfunction later in life.⁹ In our study, a similar lag time from the mean age at cancer diagnosis of 26.1 years to the mean age (at CPCS clinic presentation) of 40.1 years could support the need for much earlier preventive and screening services.

More than three quarters (75.5%) of our patients had received anthracycline-based chemotherapy, while half (51%) had received either left-breast or mediastinal radiation. These subgroups would be at the highest risk for CV damage during long-term follow-up and might need early intervention at the first signs of cardiac dysfunction. This would be especially true for patients who received radiotherapy before 1980 or received primarily left-chest radiation therapy without adequate shielding of the heart.^17–22

A history of chemotherapy-induced cardiomyopathy was present in 21% of patients visiting our clinic. This is a substantial proportion—a subgroup that will need close attention and an aggressive approach to prevent further progression to overt heart failure.

Although some preventive strategies during anthracycline administration—such as the administration of ACE inhibitors,²³ liposomal anthracycline,^24,25 or dexrazoxane²⁶—have shown promise in reducing acute cardiotoxicity, their ameliorative effects on late cardiotoxicity are not known and surveillance of exposed patients is still needed.

One fourth of the patients seen in the clinic were hypertensive (26.4%). This is roughly comparable to the current prevalence of hypertension in the U.S. adult population (age, >18 yr), which stands at 29%.²⁷ However, these patients need special attention in the preventive service, because the effects of hypertension on cardiac function are aggravated by simultaneous or previous exposure to cardiotoxic therapy.²⁸

As we have said, Hodgkin lymphoma and breast cancer were the most represented single cancers (14/53 each) (Table I). Although these malignancies have a high prevalence, they also have high survival rates (in excess of 90%).² Therapy for both involves the administration of cardiotoxic drugs (anthracyclines) and radiation therapy.^18,22,29 Hence, Hodgkin lymphoma and breast cancer survivors form a large subgroup of cancer survivors in need of CV disease preventive services. Oncologists and primary care physicians treating this subgroup should ensure that these patients follow up with a preventive cardiology service early.

The prevalence of other shared CV risk factors of patients seen in our clinic was 32.1% for dyslipidemia, 13% for diabetes mellitus, and 5.7% for known atherosclerotic disease. These subgroups of patients would benefit most from CV disease preventive services, because their risk of developing adverse CV events is both high and manageable. The evaluation of global CV risk in this population is important because risk is increased by accelerated atherosclerosis and other vascular diseases, metabolic syndrome, obesity, physical inactivity, dyslipidemia, insulin resistance, hypertension, diabetes mellitus, smoking, and drug or alcohol use. Radiation-treated childhood cancer survivors have dose-related, incrementally elevated risks of developing diabetes as adults, regardless of whether they exercise regularly or maintain normal weight.³⁰

More than half of the patients in our study group had an abnormal echocardiogram (55%), and 11 out of 18 had an abnormal carotid ultrasonogram. Carotid ultrasonography was carried out in 18 patients who had a history of neck radiation. These results reinforced the fact that neck radiation is a risk factor for the development of carotid atherosclerosis; all patients receiving this therapy should be monitored regularly for timely prevention of such adverse sequelae as stroke or transient ischemic attack.³¹ Carmody and colleagues³² found a 21.7% prevalence of advanced carotid disease in a subgroup of patients who had received high-dose radiation to the neck about 6.5 years earlier. They recommended yearly Doppler screenings for carotid disease in these patients.

Most (74%) of the patients seen in our clinic were women, which is consistent with Surveillance Epidemiology & End Results (SEER) data.^2,3 This might be due to the higher female cancer survival rate or to referral patterns specific to our clinic. Although women in general are at lower CV risk than men, surviving a cancer significantly increases their risk of CV disease, and female sex has been a suspected risk factor for cancer-treatment-induced CV disease.^33–35

In patients at high risk of developing heart failure,^36–39 the ACC/AHA guidelines¹⁵ for the management and diagnosis of chronic heart failure recommend first treating hypertension with ACE inhibitors or angiotensin receptor blockers. Cancer survivors with a history of cardiotoxic chemotherapy are potentially eligible for this regimen, particularly if asymptomatic LV dysfunction is noted.

The ACC/AHA guidelines further suggest monitoring brain natriuretic peptide (BNP)⁴⁰ levels as a marker, followed if needed by echocardiography^41,42 for the noninvasive screening of patients in stage A heart failure. However, there is no recommended specific time interval for repeat testing. From our experience in the CPCS clinic, we propose that patients in stage A heart failure be monitored yearly for BNP (Fig. 1). Elevated BNP levels (>100 pg/mL, in accordance with our recommendation) should prompt echocardiography to detect early signs of structural heart disease. Progressive heart failure in increasing stages should trigger more intensive diagnostic testing and medical therapy.^43–54 Patients with refractory heart failure should receive specialized care.^55,56 Similarly, patients who received radiation to CV structures are at risk of late effects.^17–22,32 The primary effects of radiation in this group are on the vessels (accelerated or worsened atherosclerosis), the pericardium (pericarditis), and the valves.^57,58

Fig. 1 Cardiovascular Prevention in Cancer Survivors protocol for the monitoring and treatment of patients exposed to chemotherapy associated with left ventricular dysfunction. ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin receptor blocker; BNP = brain natriuretic peptide; CHF = congestive heart failure; ECG = electrocardiogram; echo = transthoracic echocardiogram; F/U = follow up; HTN = hypertension; LV = left ventricular; LVEF = left ventricular ejection fraction; LVH = left ventricular hypertrophy

When monitoring patients for accelerated or worsened atherosclerosis, important findings are atherosclerotic manifestations already in place, high intima–medial thickness of the carotid artery, and a high coronary calcium score. All these have been shown to be markers of atherosclerotic disease.^59–62 When any of these is present (especially in diabetic patients), monitoring and interventions are more stringent: CV risk-factor modification should be as aggressive as in patients who are at risk of heart failure. The low-density-lipoprotein (LDL) cholesterol level should be targeted at <70 mg/dL, according to the ACC/AHA guidelines.^63,64 In these patients, ACE inhibitors and β-blockers should be beneficial and are indicated for control of hypertension.^65,66 We recommend echocardiographic screening every 2 to 5 years, for the late effects of radiation therapy. A high suspicion of coronary artery disease should be maintained, and these patients should undergo stress testing if they are symptomatic. Angiography (whether invasive or by computed tomography) should be contemplated in view of the clinical picture and the stress-test results.

Patients with a history of pericarditis or tamponade should initially be monitored with echocardiography at 3 monthly intervals (Fig. 2). After 2 normal studies, we recommend yearly echocardiograms and ECGs for the rest of the patient's life. However, patients without a history of pericardial disease can be screened with echocardiography and ECG only once every 5 years. Existing valvular disease needs aggressive treatment of hypertension and the modification of other risk factors. Echocardiography should be performed at least every 2 to 5 years in asymptomatic individuals.

Fig. 2 Cardiovascular Prevention in Cancer Survivors protocol for monitoring and treatment of patients with high-risk radiation therapy exposure. ACEI = angiotensin converting enzyme inhibitor; ASA = aspirin; BB = β-blockers; CIMT = carotid intima–medial thickness; ECG = electrocardiogram; echo = transthoracic echocardiogram; LDL = low-density-lipoprotein cholesterol; RT = radiation therapy

Limitations

Because no specific and universally accepted recommendations exist for the screening and treatment of CV diseases in cancer survivors, we have extrapolated our CPCS clinic protocol from the general population guidelines, on the basis of our experience with these patients. While we consider such recommendations to be warranted on the basis of our daily clinical experience, we note the following limitations: a relatively small study group, a relatively short follow-up time, and the impossibility of subdividing patients in accordance with cancer type. Instead, we divided patients by type of exposure (chemotherapy or radiation therapy).

This study was conducted at a single center and the results might not be universally applicable. The sample was relatively small, because we have presented only those data from the first year of our CV disease preventive service. Our study describes a subset of the population of cancer survivors and their cardiovascular studies. Inherently, this population is heterogeneous, as is reflected in our findings. It is our hope that future, large-scale studies will answer more detailed questions and will define the best screening practices for cancer survivors.

Although pediatric malignancies, breast cancer, and lymphoma might contribute the largest numbers of survivors with treatment-related CV needs, cardiologists in the community are nowadays seeing an increasing number of survivors of all types of cancers. This is why we think that our findings present useful insight into an important and ever-growing clinical problem that is no longer the province of specialized centers but is applicable, rather, to cardiology practices in the community at large.

Summary

Cardiovascular comorbidities and risk factors are prevalent in cancer survivors. There is an immediate need for widespread availability of CV disease preventive services for this patient group, because cancer survivors are particularly susceptible to the late adverse effects of chemotherapy and radiation. Their CV health can be maintained and improved with appropriate interventions at the earliest evidence of CV damage. Therefore, these patients would benefit tremendously from preventive services offered by cardiologists who specialize in their care and are highly familiar with the late CV effects of cancer therapy.^9,36,67