Oncologists and cardiologists are increasingly working together to better understand the mechanisms of cardiotoxicity, monitor cardiac function in cancer patients, treat heart failure when it occurs, and develop prevention strategies. At this year's American College of Cardiology Annual Meeting, cardiologists from leading cancer centers discussed what can be done to minimize cardiac damage from cancer treatments.
Monitoring Chemotherapy-Induced Heart Failure
Cancer and heart failure are both common, especially in older age groups, so it is not surprising that they would coexist, noted Daniel Lenihan, MD, Professor of Cardiology and Director of Clinical Research at the University of Texas M. D. Anderson Cancer Center.
“Cancer therapy is more effective than ever before at treating cancer, but it has a price.” For example, use of anthracyclines and rastuzumab has led to significant improvements in the survival of patients with HER2-positive breast cancer, but trastuzumab was almost taken off the market because of its potential for cardiac adverse effects. “It would have been a tremendous shame if that had occurred,” he said.
In discussing ways to monitor for cardiovascular problems with cancer therapy in general, he predicted that new cardiac problems may arise as vascular endothelial growth factor (VEGF) antagonists such as bevacizumab and small molecule TKIs are increasingly used. Such problems may be hypertensive remodeling, destruction of capillary blood flow, cardiomyopathies, and renal dysfunction from microangiopathy.
The current guidelines of the American Heart Association, the American College of Cardiology, the Heart Failure Society of America, and the American Society of Clinical Oncology are not very specific for monitoring the development of heart failure, and mainly recommend obtaining a baseline LVEF measurement and retesting sometime later, he explained.
However, more than one-third of patients with heart failure have normal LVEFs, but their prognosis is still poor. “We diagnose heart failure primarily based on symptoms, and even the utility of that is in question,” he said, noting that in one study (Fromme EK et al: JCO 2004;22: 3485–3490), physicians missed identifying dyspnea 77% of the time.
“Furthermore, generally, other than in a Phase I trial, they only have to report Grade 3 [or greater] toxicity,” so in a trial, only if a patient became symptomatic for heart failure with an LVEF less than 40% would it show up.
Decreased LVEF can be detected by multi-gated acquisition scan (MUGA), radionuclide angiography, or echocardiography (echo). Small changes in echo readings may not be readily detectable. But in other research, when Edith Perez, MD, and colleagues looked carefully for LVEF changes in the NCCTG N9831 trial, cardiac toxicity was detected in a significant proportion of patients by both MUGA and echo after only four cycles of anthracycline and cyclophosphamide even though the cumulative dose of anthracycline was rather low (Perez EA et al: JCO 2004;22:3700–3704).
Biomarkers such as Troponin T and brain natriuretic peptide (BNP) may be useful indicators of heart failure but are not currently recommended in any guidelines for following cancer patients. Dr. Lenihan said that experience at M. D. Anderson has shown that besides the usual risk factors of hyperlipidemia, diabetes, family history, obesity, and smoking, that the strongest risk factors for chemotherapy-related heart failure were prior anthracycline use and radiation to the chest.
“Fifty-five percent of patients improved after optimal heart failure therapy, so we feel like we can make a big impact if we do recognize the problem,” he said.
In a pilot study of biomarkers involving 109 patients receiving an anthracycline but with no history of heart disease, 11 patients developed cardiac events, mostly symptomatic heart failure.
“The BNP [rise] preceded the event by a number of days,” Dr. Lenihan said. “If we look at the predictive value of a BNP of over 200 [pg/mL], it's very powerful and confers a greater than 44 times risk of having a cardiac event. But as I would also point out, ejection fraction was a poor predictor of cardiac toxicity.”
Reminding cardiologists in his talk that one of the objectives of the American College of Cardiology, stated in 1951, was “to cooperate with other organizations of practitioners and scientists dealing with the same or related specialties,” Dr. Lenihan urged participants to reach out to their oncology colleagues with their accumulating knowledge about detecting chemotherapy-related cardiac toxicity.
Steven Colan, MD, Professor and Chief of Noninvasive Cardiology at Children's Hospital Boston, rounded out the session with a talk on the treatment of doxorubicin (Adriamycin)-related CHF, using it as a prototype for other chemotherapy-related heart failure. He distinguished between early and late-onset heart failure, the early occurring usually within two to six weeks of dosing but up to a year after the completion of therapy, and the late onset occurring after a year. While the problem is very dose dependent, he said, the incidence has fallen over time as protocols have lowered the cumulative dose.
Studying the condition in children provides the advantage of a general lack of comorbidities, Dr. Colan said.
While there are many hypotheses to explain the problem at the cellular level, “in acute or early Adriamycin congestive heart failure, we almost always end up with dilated, hypocontractile ventricles that in many respects correspond to the typical findings we would see in other forms of dilated cardiomyopathy.”
Therapy, therefore, is typically the standard for CHF, including cessation of Adriamycin and introduction of the usual medications. Late onset tends to exhibit less severe ventricular dilation and systolic dysfunction, more severe diastolic dysfunction, and more marked endocardial and myocardial fibrosis, he said.
Data from the 1970s to 1990s showed a one-year survival rate of less than 50% percent for the early-onset condition. Data on outcomes with newer treatments are generally lacking, Dr. Colan said, and most reports do not distinguish between the early and late forms. A newer study in adults (Tallaj JA et al: J Heart Lung Transplant 2005;24:2196–2201) showed that most of the cases were late onset (average of 6.4 years after starting doxorubicin), that there was a five-year survival rate of 80%, and that 50% of patients regained normal left ventricular function.
The report described a comparison of patients who received just ACE inhibitors alone with those receiving ACE inhibitors plus beta-blockers. Those receiving the combination therapy appeared to have a better outcome, with normalization of LVEF.
Dr. Colan mentioned carvedilol, a nonselective beta-blocker with potent antioxidant properties, because it received a lot of press earlier this decade. When given to rats, carvedilol reduced doxorubicin toxicity, and there were case reports of recovery from doxorubicin congestive heart failure in adult patients. But since recovery is now “not considered so rare,” he said it could not necessarily be ascribed to carvedilol and is still an open question.
Dr. Colan said that cardiologists are increasingly being faced with the issue of subclinical doxorubicin-related CHF, since the incidence is far higher than that of congestive heart failure itself—“and almost certainly accounts for the instances of late cardiac failure.”
The manifestations are different from dilated cardiomyopathy in that patients typically have nondilated ventricles with reduced wall thickness, elevated wall stress, normal to moderately impaired systolic function, and almost always, diastolic dysfunction.
“Normalization of function after completion of therapy can be followed years later by deterioration, and therefore, just because it's subclinical doesn't mean it's not important,” Dr. Colan cautioned. Again, the severity of injury is dose-related but with sensitive methods, can be detected at cumulative doses as low as 200 mg/m2.
In a study of children treated with a median cumulative dose of doxorubicin of 352 mg/m2 and followed for a median of 11.8 years (Lipschultz SE et al: JCO 2005;23: 2629–2636), “the primary observation is that as a group, these patients are subject to slow but nevertheless relentless progression,” Dr. Colan said.
Cardiotoxicity can persist for years without symptoms, and some reports suggest that when it becomes overt doxorubicin-related CHF, the prognosis is poor and perhaps worse than other forms of dilated cardiomyopathy. Predicting who will end up with late deterioration is difficult, so Dr. Colan's advice and practice is to monitor everyone at risk.
“It is unlikely we are going to prevent the toxicity,” he predicted. “Most likely we are going to find various ways to attenuate the toxicity.” Because anthracyclines are such effective agents, he speculated that the incidence of cardiomyopathy will not decrease. Instead, he said oncologists will probably be able to increase their doses to effect higher cancer cure rates, with no change in the rates of cardiotoxicity.
Preventing progression of chronic cardiomyopathy to CHF depends on aggressive monitoring to detect early signs of cardiac dysfunction. “Most patients with cumulative doses greater than 200 [mg/m2] get lifetime surveillance,” Dr Colan said. “Unfortunately, it is unproven at this point that we can prevent progression.”
Studies of ACE inhibitors in children with asymptomatic doxorubicin cardiomyopathy showed early improvements but no long-term benefits in clinical status or clinical course. With little data to go on, standard therapy for CHF is now the norm.
Dr. Colan noted that based on accumulating evidence of mitochondrial dysfunction, some people have recommended supplementation with Coenzyme Q10, which has a rather expensive, out-of-pocket cost based on insufficient data to recommend it.
Some data support using beta-blockers sooner rather than later, he said. But late doxorubicin-related CHF patients may be different from typical CHF patients. The former have a predominantly restrictive diastolic dysfunction, wherein bradycardia would cause reduced cardiac filling and therefore reduced cardiac output, so beta-blockers may be harmful.
“I don't think it's possible to just smudge these together and assume that the response in one group will predict the response in the other,” Dr. Colan warned.
Some reports have shown that beta-blockers have improved outcomes in doxorubicin-related CHF, but these drugs may be very poorly tolerated in patients with restrictive physiology. A trial of beta-blockers may be worthwhile, withdrawing them for poor responses, he said.
Finally, heart transplantation is a viable option if treatment of the cancer was successful. Two obstacles to transplantation are recurrent malignancy and pulmonary hypertension before transplant because of restrictive cardiomyopathy. If pulmonary hypertension is detected with careful monitoring, accelerated transplant listing is warranted, Dr. Colan said.
He concluded that the ability to prevent progression of Adriamycin cardiomyopathy is unproven and that what agent should be used has not yet been determined. Some evidence supports using combinations of ACE inhibitors and beta-blockers. Fortunately, recent reports indicate that the condition is no more refractory to therapy than are other cardiomyopathies.